industrial policy, its spatial aspects and cluster development in ...

INDUSTRIAL POLICY, ITS SPATIAL ASPECTS AND
CLUSTER DEVELOPMENT IN PAKISTAN
BY
ABID A. BURKI
KAMAL A. MUNIR
MUSHTAQ A. KHAN
M. USMAN KHAN
ADEEL FAHEEM
AYESHA KHALID
SYED TURAB HUSSAIN
LAHORE UNIVERSITY OF MANAGEMENT SCIENCES
Lahore University of Management Sciences
Opposite Sector U, DHA, Lahore Cantt.54792, Lahore, Pakistan.
http://www.lums.edu.pk

INDUSTRIAL POLICY, ITS SPATIAL ASPECTS
AND CLUSTER DEVELOPMENT IN PAKISTAN
BY
ABID A. BURKI
KAMAL A. MUNIR
MUSHTAQ A. KHAN
M. USMAN KHAN
ADEEL FAHEEM
AYESHA KHALID
SYED TURAB HUSSAIN
LAHORE UNIVERSITY OF MANAGEMENT SCIENCES
THIS VERSION: MAY 18, 2011
Lahore University of Management Sciences
Opposite Sector U, DHA, Lahore Cantt.54792, Lahore, Pakistan.
http://www.lums.edu.pk
ii

Table of Contents
List of Tables ............................................................................................................................................... vii
List of Figures ............................................................................................................................................... ix
List of Boxes ................................................................................................................................................ xii
Acronyms and Abbreviations ......................................................................................................................xiii
Acknowledgements ..................................................................................................................................... xv
Executive Summary ................................................................................................................................. xvi
Principal Recommendations ..................................................................................................................... i
Horizontal Recommendations ................................................................................................................... i
Sectoral Reforms ................................................................................................................................... xxiii
1 Introduction ...................................................................................................................................... 49
1.1 Raison d’être for an Industrial Policy .......................................................................................... 51
1.1.1 Brief Historical Review ........................................................................................................ 52
1.1.2 Towards an Industrial Policy ............................................................................................... 54
1.2 Methodology and Structure ........................................................................................................ 57
2 Industrial Sector of Pakistan: Structure, Performance & Problems ....................................... 59
2.1 Structure & Performance of the Industrial Sector in Pakistan ................................................... 60
2.1.1 Growth Trends .................................................................................................................... 61
2.1.2 Structural Rigidity & Inadequate Transformation ............................................................... 64
2.1.3 Productivity ......................................................................................................................... 69
2.1.4 Export Performance ............................................................................................................ 73
2.1.5 Summary Analysis ............................................................................................................... 80
2.2 Investment Trends in Pakistan .................................................................................................... 82
2.2.1 Aggregate Investment Trends ............................................................................................. 82
2.2.2 Regional Comparison and Pakistan’s Investment Gap ....................................................... 84
2.2.3 Widening Resource Gap between Investment & Domestic Savings ................................... 86
2.2.4 Actual & Potential Total Private Investment & Domestic Savings ...................................... 87
2.2.5 Sectoral Composition of Investment .................................................................................. 89
iii

2.2.6 Foreign Direct Investment .................................................................................................. 91
2.2.7 Sector‐wise Decomposition of FDI ...................................................................................... 93
2.3 Constraints to Investment & Manufacturing Growth in Pakistan .............................................. 95
2.3.1 Macroeconomic Instability & its Impact on Investment & manufacturing ........................ 96
2.3.2 Infrastructural Constraints on Industrial Growth ............................................................. 109
2.3.3 Factor Markets Constraints on Investment & Manufacturing .......................................... 124
2.3.4 Standards, Regulation & Governance Constraints ............................................................ 146
Competitiveness & Industry Stagnation ........................................................................................... 160
2.3.5 Other Horizontal Issues & Interventions .......................................................................... 167
3 Competitiveness of Key Industrial Sectors of Pakistan ........................................................... 173
3.1 Primary Industrial Sector .......................................................................................................... 174
3.1.1 Chemical Industry ............................................................................................................. 174
3.1.2 Steel Industry .................................................................................................................... 182
3.1.3 Fertilizer Industry .............................................................................................................. 185
3.2 Value Added Knowledge Based Industrial Sectors ................................................................... 190
3.2.1 Auto & Farm Machinery Industry ..................................................................................... 190
3.2.2 Electronics Industry ........................................................................................................... 193
3.2.3 Pharmaceutical Industry ................................................................................................... 200
3.3 Value Added Skill & Engineering Based Sectors ....................................................................... 208
3.3.1 The SME Sector ................................................................................................................. 208
3.3.2 Fan Sector ......................................................................................................................... 210
3.3.3 Cutlery, Utensils & Hunting Equipment Sector ................................................................. 218
3.3.4 Horticulture Processing Sector ......................................................................................... 227
3.3.5 Surgical Instruments Sector .............................................................................................. 231
3.3.6 Sports Goods Sector .......................................................................................................... 237
3.3.7 Ceramics Sector ................................................................................................................. 242
3.3.8 Furniture Sector ................................................................................................................ 246
3.3.9 Leather Sector ................................................................................................................... 254
3.3.10 Gems & Jewellery Sector .................................................................................................. 260
3.3.11 Marble & Granite Sector ................................................................................................... 264
3.3.12 Light Engineering Sector ................................................................................................... 271
iv

3.3.13 Fisheries Sector ................................................................................................................. 283
4 Spatial Concentration of Economic Activity .............................................................................. 285
4.1 Introduction .............................................................................................................................. 285
4.1 Introduction .............................................................................................................................. 285
4.2 New Economic Geography and Development .......................................................................... 287
4.3 Mapping Measures of Regional and Spatial Inequality ............................................................ 290
4.3.1 Regional Spatial Distribution of Population and Density .................................................. 290
4.3.2 Measuring Regional Poverty and Income Inequality ........................................................ 296
4.3.3 Evidence on Spatial Disparities in Human Capital Infrastructure ..................................... 310
4.3.4 Market Access and Spatial Inequality in Road Infrastructure ........................................... 322
4.3.5 Agglomeration of Manufacturing Industries in Pakistan .................................................. 327
4.3.6 Localization versus Urbanization Externalities ................................................................. 337
5 Poverty Impacts of Public Investments and Causes of Industry Agglomerations .............. 342
5.1 Introduction .............................................................................................................................. 342
5.2 Are Regional Infrastructure Disparities Influencing Incidence of Poverty? .............................. 343
5.2.1 Background ....................................................................................................................... 343
5.2.2 The causes of poverty in Pakistan ..................................................................................... 343
5.2.3 Methodology and data ...................................................................................................... 345
5.2.4 The basic regressions ........................................................................................................ 347
5.2.5 The effect of spatial inequality in social infrastructure on rural poverty ......................... 358
5.2.6 Spatial inequality in road infrastructure and rural poverty .............................................. 360
5.3 Nexus between Infrastructure, Income Inequality and Poverty ............................................... 364
5.3.1 Does high inequality cause high poverty in Pakistan? ...................................................... 365
5.3.2 Method of Analysis ........................................................................................................... 367
5.3.3 Social infrastructure and poverty: How inequality affects poverty? ................................ 368
5.3.4 Road infrastructure and poverty: Does inequality matter? .............................................. 370
5.4 What Factors Cause Agglomeration of Manufacturing Industries? .......................................... 372
5.4.1 Background ....................................................................................................................... 372
5.4.2 Explaining agglomeration .................................................................................................. 373
5.5 Nature of Scale Economies and Patterns of Industry Agglomeration ...................................... 378
5.5.1 Background ....................................................................................................................... 378
v

5.5.2 Empirical specification ...................................................................................................... 378
5.5.3 Empirical results ................................................................................................................ 379
5.6 Compendium of Policy Notes .................................................................................................... 382
5.6.1 Policy for pro poor infrastructure investment .................................................................. 382
5.6.2 Location policy for industrial development ...................................................................... 384
5.6.3 Agglomeration economies, scale externalities and growth of firms ................................ 386
References .............................................................................................................................................. 388
vi

List of Tables
Chapter 2
Table 2‐1: Distribution of workforce by formal/informal sectors and by gender in Pakistan, 1999‐2008
(%) ............................................................................................................................................................... 66
Table 2‐2: Average Product Shares in the Manufacturing Sector of Pakistan, 1970‐90 (percent) ............. 67
Table 2‐3: Growth Accounting in Pakistan by Decades, 1961‐2005 (percent) ........................................... 70
Table 2‐4: Growth Accounting in Asia, 1961‐2005 (percent) ..................................................................... 70
Table 2‐5: Annual Labor Productivity Growth, 1990‐2006 (percent) ......................................................... 72
Table 2‐6: Country Export Shares Relative to Total World Exports, 1970‐2008 (percent) ......................... 73
Table 2‐7: Revealed Comparative Advantage and PRODY in Textiles, 2008/09 ......................................... 74
Table 2‐8: Technological Level of Pakistani and World Exports, 1998‐2008 (percent) .............................. 79
Table 2‐9: Comparisons of Savings and Investments, 1994‐2009 (% of GDP) ............................................ 86
Table 2‐10: Inward FDI Performance Index, 2003‐08 ................................................................................. 92
Table 2‐11: Regional Comparison of FDI, 1990‐2000 ................................................................................. 92
Table 2‐12: Harbison‐Myer Skills Indicators ............................................................................................. 127
Table 2‐13: Data on Enforcing Contracts .................................................................................................. 153
Table 2‐14: individual Tax, Corporate Tax and Tax Revenue Comparisons, 2007‐2009 (% of GDP) ........ 156
Table 2‐15: Pakistan’s Global Competitiveness Index Ratings, 2003‐2010 .............................................. 162
Table 2‐16: Malmquist Index of Sector Means, 1998‐2007 ...................................................................... 164
Table 3‐1: Demand for Petrochemical Products ....................................................................................... 178
Table 3‐2: Import of Basic Chemicals in Pakistan (Million US$) ............................................................... 178
Table 3‐3: Foreign Direct Investment in the Electronics Industry, 2002‐2009 (US$ Million) .................. 196
Table 3‐4: Fan Industry Characteristics ..................................................................................................... 211
Table 3‐5: Cutlery and Steel Ornaments Industry Characteristics ............................................................ 219
Table 3‐6: Surgical Industry Characteristics .............................................................................................. 232
Table 3‐7: Sports Goods Industry Characteristics ..................................................................................... 237
Table 3‐8: Pakistan Ceramics Industry ...................................................................................................... 243
Table 3‐9: Leather Industry Characteristics .............................................................................................. 255
Table 3‐10: Marble and Granite Reserves in Pakistan .............................................................................. 264
Table 4‐1: Population Density by Area...................................................................................................... 291
Table 4‐2: List of poverty studies and their methods ......................................................................... 297
Table 4‐3: Poverty estimates for the 1990s reported by previous studies ....................................... 299
Table 4‐4: Inflation-adjusted poverty lines per adult equivalent per day used for poverty
estimates .................................................................................................................................................. 301
Table 4‐5: Regional poverty in Pakistan, 1990-91 to 2005-06 .......................................................... 302
Table 4‐6: Changes in Gini inequality in Pakistan and urban and rural areas .......................................... 307
vii

Table 4‐7: Changes in Gini inequality by provinces .................................................................................. 309
Table 4‐8: District level variables used in principal component analysis ................................................. 312
Table 4‐9: Principal component rotated factor loadings .......................................................................... 313
Table 4‐10: Most and least developed districts based on rankings from principal component post‐
primary school system and hospital index ................................................................................................ 316
Table 4‐11: Change in concentration by Gini index in Punajab ................................................................ 330
Table 4‐12: Agglomeration of 3‐digit manufacturing industries in Pakistan, 2005‐06 ............................. 333
Table 4‐13: Geographic concentration of 3‐digit industries in Punjab, 1995‐96 – 2005‐06 ..................... 336
Table 4‐14: Mean values of industrial concentration measures overtime ............................................... 337
Table 4‐15: Localisation versus urbanization externalities in Punjab, 1995‐96 to 2005‐06 ..................... 340
Table 5‐1: Incidence of poverty by household characteristics, 2005‐06 .................................................. 344
Table 5‐2: Distribution of Rural Sample by Household Survey ................................................................. 346
Table 5‐3: Definition of variables used in the Probit Regressions .......................................................... 349
Table 5‐4: Effects of household and individual characteristics on rural poverty in Pakistan, basic results
.................................................................................................................................................................. 356
Table 5‐5: Marginal effects of the impact of education and health infrastructure on rural poverty ....... 359
Table 5‐6: Summary Statistics of Punjab’s Data ....................................................................................... 361
Table 5‐7: Post‐primary school system and poverty under different inequality regimes ........................ 369
Table 5‐8: Road density and poverty under different inequality and polarization regimes ..................... 371
Table 5‐9: Descriptive statistics for agglomeration regression ................................................................ 374
Table 5‐10: Pearson’s correlation ............................................................................................................. 375
Table 5‐11: OLS specification for agglomeration regression .................................................................. 377
Table 5‐12; Scale Externalities and Productivity in Manufacturing Industries in Punjab, 1995‐96 – 2005‐
06 .............................................................................................................................................................. 380
viii

List of Figures
Chapter 2
Figure 2‐1: Association between Manufacturing Value Added (MVA) Growth & GDP ........................... 61
Figure 2‐2: Historic GDP growth Rate in Pakistan, 1954‐2006 (percent) ................................................... 61
Figure 2‐3: GDP Growth Rate in Pakistan, 2001‐09 (percent) .................................................................... 62
Figure 2‐4: Sectoral growth rates in Pakistan, 2001–09 (percent) ............................................................. 63
Figure 2‐5: Annualized Growth Rates of Key Sectors in Pakistan for 10 Year Periods, 1970‐2009 (percent)
.................................................................................................................................................................... 63
Figure 2‐6: Sectoral Shares in GDP of Pakistan, 1970‐2009 (percent) ........................................................ 64
Figure 2‐7: Shifts in Sectoral Shares in GDP of Pakistan, 2000‐09 (percent) .............................................. 65
Figure 2‐8: Employment Shares by Sector in Pakistan, 1980‐2009 (percent) ............................................ 66
Figure 2‐9: Performance of the Large Scale and Small Scale Manufacturing Sector in Pakistan, 1950‐2010
(%) ............................................................................................................................................................... 68
Figure 2‐10: Manufacturing Value Added in US$ (Billions), 2001 & 2007 .................................................. 69
Figure 2‐11: Total Factor Productivity by Firm Size & Firm Age, 2001‐2006 (log values) ........................... 71
Figure 2‐12: Sectoral Value Added/Worker in Pakistan, 1981‐2002 (Rs at contact prices of 1999‐2000) . 71
Figure 2‐13: Output per Worker Growth in Pakistan, 1980‐2006 (percent) .............................................. 72
Figure 2‐14: Output per Worker in Asian Economies, 1990‐2004 (assuming 1990 as base of 100) .......... 73
Figure 2‐15: Competitiveness and Performance of Pakistan’s Exports, 2000‐2007 (percent) ................... 76
Figure 2‐16: Product Concentration Index, 1974‐2008 .............................................................................. 76
Figure 2‐17: Market Diversification Index, 1971‐2008 ............................................................................... 77
Figure 2‐18: Elasticity of Pakistan’s Exports to GDP of Industrial Countries, 1985‐2007 (percent) ........... 79
Figure 2‐19: Total, Private and Public Gross Fixed Capital Formulation (GFCF) in Pakistan, 1980‐2009 (%
of GDP) ........................................................................................................................................................ 83
Figure 2‐20: Aggregate Investment Gap of Pakistan Relative to other Countries, 1980‐2008 (percent) ... 85
Figure 2‐21: Long Run Indexed Real/Potential Investment Decline in Trend Rate of Private Investment,
1980‐2009 ................................................................................................................................................... 88
Figure 2‐22: Growth Rate in Real Private GFCF & Total GFCF, 2000‐2010 (%) ........................................... 88
Figure 2‐23: Share of Manufacturing Investment in Total Fixed Investment, 2000‐10 (percent) .............. 91
Figure 2‐24: FDI Investment as % of Gross Fixed Capital Formation, 1990‐2009 (percent) ....................... 93
Figure 2‐25: Sectoral Decomposition of FDI, 2004 and 2008 9percent) ..................................................... 94
Figure 2‐26: Real GDP and Investment Growth in Pakistan, 1991‐2009 (percent) .................................... 98
Figure 2‐27: Exports, Net Factor income from Abroad and Imports, 1990-2008 (percents of GNI)
.................................................................................................................................................................. 101
Figure 2‐28: Inflation and Investment in Pakistan, 2001‐2010 (% p.a.) .................................................... 104
Figure 2‐29: Electricity Demand and Supply, 2003 – 10 (MW) ................................................................. 110
Figure 2‐30: Provision of Training to Workers .......................................................................................... 132
Figure 3‐1: Petrochemical Value Chain ..................................................................................................... 177
ix

Figure 3‐2: Demand and Supply Gap of Steel in Pakistan ......................................................................... 182
Figure 3‐3: Pakistan’s Fertilizer (N.P.K) Plant Capacity & Demand (000’ tons in nutrients) ..................... 186
Figure 3‐4: Pakistan’s Fertilizer Import ..................................................................................................... 186
Figure 3‐5: Comparison of Pakistan’s NPK Ratio with other Countries .................................................... 188
Figure 3‐6: Pakistan’s Gas Demand and Supply Balance .......................................................................... 188
Figure 3‐7: Motorization Levels per 1000 persons ................................................................................... 191
Figure 3‐8: Broad Comparison of Pakistan and Indian Pharmaceutical Value Chain ............................... 204
Figure 3‐9: Pakistan’s Fan Exports, 2004‐2009 (US$) ............................................................................... 212
Figure 3‐10: Typical Value Chain of Fan .................................................................................................... 213
Figure 3‐11: Capacity Utilization over typical 12 Month Period ............................................................... 214
Figure 3‐12: Average Export Price of Fans US$ ......................................................................................... 216
Figure 3‐13: Pakistan’s Cutlery and Steel Ornaments Sector Exports (US$) ............................................ 220
Figure 3‐14: Value Chain for a 12 Inch Dagger ......................................................................................... 223
Figure 3‐15: Average Export Price of Steel Used in Making Cutlery and Ornaments, US$/Kg ................. 224
Figure 3‐16: Product Split of World Exports & Pakistan Exports, 2005‐09 (US$ Million) ......................... 233
Figure 3‐17: Exports of Pakistan’s Surgical Instruments Sector, 2005‐09 (US$) ....................................... 233
Figure 3‐18: Value Chain Breakdown of a 5.5 Inch Forceps ..................................................................... 234
Figure 3‐19: Pakistani Exports of Sports Goods, 2003‐2009 (US$ Billion) ................................................ 238
Figure 3‐20: Comparison of World and Pakistan’s Sporting Goods Product Split .................................... 238
Figure 3‐21: Typical Value of Hand‐Stitched Export Quality Training Soccer Ball .................................... 240
Figure 3‐22: Ceramics Exports for Pakistan, 2005‐2009 (US$) ................................................................. 243
Figure 3‐23: Furniture Export of Pakistan, 2005‐2009 (US$) .................................................................... 248
Figure 3‐24: Pakistan Export Price Comparison, 2005‐2009 (US$/Kg of Wood) ...................................... 249
Figure 3‐25: Key Challenges in Pakistan’s Furniture Value Chain ............................................................. 250
Figure 3‐26: Comparison of Pakistan and World Product Split ................................................................ 256
Figure 3‐27: Marble & Granite Exports of Pakistan, 2005‐2009 (US$ Million) ......................................... 265
Figure 3‐28: Value Chain of Marble and Granite Industry ........................................................................ 266
Figure 3‐29: Traditional Quarrying Method .............................................................................................. 270
Figure 3‐30: Advanced Method of Quarrying ........................................................................................... 270
Figure 3‐31: Value Chain for Radiator Production .................................................................................... 277
Figure 4‐1: Post‐primary school system and current population, 2005‐06 .............................................. 293
Figure 4‐2: Hospital size and current population, 2005‐06 ...................................................................... 294
Figure 4‐3: Value of large‐scale manufacturing production and population growth ............................... 295
Figure 4‐4: Regional Poverty headcount in Pakistan, 1990-91 to 2005-06 ..................................... 302
Figure 4‐5: Poverty headcount in urban Pakistan, 1990-91 to 2005-06 .......................................... 304
Figure 4‐6: Poverty headcount in rural Pakistan, 1990-91 to 2005-06 ............................................ 304
Figure 4‐7: Changes in inequality in Pakistan and by regions ........................................................... 308
Figure 4‐8: Changes in inequality in the four provinces ..................................................................... 309
Figure 4‐9: Location of districts by education and hospital index .................................................... 319
Figure 4‐10: Industry clusters and development ranking of districts, 2005-06 ............................. 320
Figure 4‐11: Resource based clusters and development ranking of districts, 2005-06 ................. 320
x

Figure 4‐12: Spatial inequality in road density in Punjab, 2005‐06 .......................................................... 324
Figure 4‐13: Relative road density of the districts of Punjab with Lahore district, 1992‐93 vs. 2005‐06 . 324
Figure 4‐14: Spatial inequality in road density in Khyber Pakhtunkhwa, 2005‐06 ................................... 325
Figure 4‐15: Relative road density of the districts of KP with Peshawar district, 1993-94 vs. 2005-
06 .............................................................................................................................................................. 327
Figure 4‐16: District level employment shares in Pakistan’s manufacturing sector ....................... 329
Figure 4‐17: Distribution of 3-digit Ellison-Glaeser index ................................................................. 334
Figure 5‐1: Distribution of inequality and per capita income ............................................................ 365
Figure 5‐2: Distribution of Poverty and Per Capita Income ...................................................................... 366
Figure 5‐3: The effect of inequality on poverty in Pakistan, 1990-91 to 2005-06 ........................... 367
xi

List of Boxes
Chapter 2
Box 2-1: Inward Investment Incentives Regime in Pakistan ................................................... 45
Box 2-2: Energy in Comparison with China and India ............................................................ 63
Box 2-3: The Case of Gawadar .................................................................................................. 72
Box 2-4: Cartelization & Case Studies .................................................................................... 109
Chapter 3
Box 3-1: The Steel Industry of China ....................................................................................... 125
Box 3-2: The Automobile Market of Malaysia ......................................................................... 132
Box 3-3: Electronic Industry in China ..................................................................................... 139
Box 3-4: Singapore Electronics Industry ................................................................................. 141
Box 3-5: New Drug Filings of Some Successful Indian Pharmaceutical Companies ............ 144
Box 3-6: Past Government Intervention in the Furniture Sector .......................................... 189
Box 3-7: Depiction of Suha Bazar of Lahore ........................................................................... 196
Box 3-8: Italian Marble Industry ............................................................................................ 203
xii

Acronyms and Abbreviations
ADR Alternative Dispute Resolution
API Active Pharmaceutical Ingredients
BOP Balance of Payments
CRO Clinical Research Organisation
DAP Diammonium Phosphate
DFID Department for International Development
EDB Engineering Development Board
EPZ Export Processing Zones
EOBI Employee Old-Age Benefit
FDI Foreign Direct Investment
FPCCI Federation of Pakistan Chambers of Commerce and Industry
GDP Gross Domestic Product
GMP Good Manufacturing Practices
IAG Industry Advisory Group
IDB Industrial Development Board
IPR Intellectual Property Rights
ITP International Trade Price
KIBOR Karachi Interbank Offer Rate
MMF Man Made Fibre
NAVTEC National Vocational & Technical Education Commission
NEQS National Environmental Quality Standards
NPO National Productivity Organisation
OEM Original Equipment Manufacturer
PEFMA Pakistan Electric Fan Manufacturers Association
PEPCO Pakistan Electric Power Company
PHADC Pakistan Hunting Arms Development Company
PIB Pakistan Investment Bond
PLDC Pakistan Leather Development Council
PPP Public Private Partnership
PSDC Pakistan Stone Development Company
PSM Pakistan Steel Mill
PSQCA Pakistan Standards & Quality Compliance authority
PSI Pre-shipment Inspections
R&D Research & Development
RFID Radio Frequency Identification
SECP Securities & Exchange Commission of Pakistan
SEZ Special Economic Zones
SME Small & Medium Enterprises
SMEDA Small & Medium Enterprise Authority
SG Safeguards
SPS Sanitary & Phytosanitary
TBS Tariff Based System
TBT Technical Barriers to Trade
TDAP Trade Development Authority of Pakistan
TEVTA Technical Education & Vocational Training Authority
TVET Technical & Vocational Education Training
UNEP United Nation Environment Programme
WAPDA Water & Power Development Authority
WHO World Health Organisation
xiii

WTO World Trade Organisation
xiv

Acknowledgements
This study has been completed with the support of several individuals, departments and
organizations. The project would not have been possible without the facilitation and
overall guidance provided by Prof. Ijaz Nabi (Dean, LUMS). We would like to thank Ms.
Shaista Sohail, Joint Secretary, Ministry of Industries and Production for providing us
invaluable information and support throughout the project. We would also like to thank
officers from Engineering Development Board, Small & Medium Development
Enterprise (SMEDA), Technical and Vocational Training Authority, and Chambers of
Commerce & Industry across the country for their useful ideas on industrialisation and
its issues in Pakistan. We owe a debt of gratitude to Professor Ha Joon Chang
(University of Cambridge), Professor Alice Amsden (Massachusetts Institute of
Technology) and Dr. Faheem ul Islam (LUMS) for their insightful comments and ideas
in deriving the policy proposals. We thank the members of the staff of the World Bank
for providing the funding and over all support for the project. We would like to
acknowledge the research support provided by Kiran Javaid and Abubakar Memon and
the editorial input received from Shahbano Ijaz and Nadia Mukhtar. Finally we would
like to thank the numerous representatives of the private sector, especially the members
of the Task Force who took time out from their busy schedules to discuss in detail the
problems of Industry and the possible solutions.
Prof. Abid A. Burki
Prof. Kamal A. Munir
Dr. Mushtaq A. Khan
M. Usman Khan, CFA IMC
Adeel Faheem
Ayesha Khalid
Dr. Syed Turab Hussain
xv

Executive Summary
Introduction:
The Industrial Policy report
constitutes a comprehensive analysis
of Pakistan’s manufacturing sector
aimed at deriving policy
recommendations for Industrial
growth and development. The report is
structured to provide a three tiered
analysis of Pakistan’s industry.
The first tier is a macro level analysis
of industrial structure and
performance which helps identify the
major constraints hampering
structural change in the country. The
policy recommendations derived from
this section are the broader or
horizontal interventions required for
industrial development spanning a
wide sphere of economic activity in the
country.
The second tier of analysis presents
detailed look at the manufacturing
sector through a firm/sector level
analysis. The main obstacles in
increasing firm competitiveness and
sectoral growth are investigated
through both a value chain and a
stakeholder analysis. This
methodological approach helps divulge
the necessary vertical or sector level
policy interventions. There are a total
of twenty four sectors covered in this
analysis ranging from important
ancillary industries, knowledge based
industries to the major small and
xvi
medium scale export based industrial
sectors.
The third tier of analysis takes a more
telescopic view of the industrial sector
where the focus is on the spatial
aspects of industrialization - economic
geography, covering issues such as
inter and intra provincial inequalities
in infrastructural provision and its
resultant impact on income inequality,
poverty and industrial cluster
formation. The analysis in this section
uses the Census of Manufacturing
Industries (CMI) disaggregated data
set and rigorous econometric and
statistical techniques to come up with
both indexes measuring industrial
concentration and the factors which
significantly affect cluster formation or
the lack of it.
Industrial Sector of
Pakistan: Structure,
Performance & Problems
The first tier - macro level analysis
analyses the structure/performance of
the industrial sector using various
macro and industry specific indicators
across time and in comparison to other
countries. The primary focus in this
section is on macroeconomic
management and performance with
respect to industrial sector growth.
The section highlights the fact that
although over the last four decades
Pakistan has experienced over five

percent average GDP growth rate,
these episodes of high growth rates
were mostly consumption driven
leading to frequent ‘boom’ and ‘bust
cycles. The fact that industry has not
been the main engine of Pakistan’s
growth has led to on average a low and
variable growth rate.
The macro analysis also reveals that
the structural change in Pakistan has
been skewed towards a shift from
agriculture to services with
manufacturing growth remaining fairly
stagnant. It shows that Pakistan’s
manufacturing sector is largely
concentrated in a few industrial
products with relatively less value
addition compared to countries within
the region. With a narrow
manufacturing base and a
concentration in low technology and
low value-added products whose share
in the world market is decreasing, the
prospects of industrial growth remain
uncertain. Pakistan’s economy was
built around textile and its current
base is still concentrated in textiles.
Cotton textile production is the most
important of Pakistan's industries,
accounting for about 19 percent of
large-scale industrial employment and
about 60 percent of total exports.
After reviewing the characteristics
and performance of the manufacturing
sector the report looks at the over time
trend of investment in the country
relative to regional economies.
Investment in large scale
manufacturing sector is shown to be
volatile with a significant recent
xvii
decline as a consequence of both
periodic macroeconomic instability
and an investment climate which has
been persistently poor. As increased
investment or capital accumulation is a
necessary condition for growth,
substantiated by the experience of the
newly industrialized countries such as
Korea and Taiwan, the secular fall in
investment is shown to be the key
factor behind an overall sluggish
growth in GDP.
Following the discussion on
industrial performance and investment
trends, the broader impediments
resulting in the systemic stagnation of
industrial growth and investments are
highlighted in this section. Constraints
related to factor markets,
infrastructure provision, lack of
macroeconomic stability, regulatory
environment, and security/negative
perception of Pakistan, are brought to
the forefront. This is done by using the
repository of existing evidence on the
above mentioned constraints such as
the cost of doing business data
collected by World Bank and other
similar studies. The analysis is used to
inform policy which by removing
cross-cutting bottlenecks would
facilitate the overarching growth of the
industrial sector. These essentially are
the horizontal policy interventions
required to stimulate overall
manufacturing growth. Some of these
interventions are summarized below.
In terms of macroeconomic stability
the need for enhancing the fiscal space

– through expansion of the tax base –
is stressed. The importance of prudent
fiscal and monetary policy is
underscored albeit with a
recommendation to lower the nominal
interest rate which has been a major
factor in stifling investment and
manufacturing growth recently.
Emphasis is also put on the need to
coordinate trade policy with industrial
policy. Effective use of Non-Tariff
Barriers NTBs and creation of a
Science Park is also suggested as
mechanisms to protect and facilitate
the growth of value added industry in
the country.
Specific policy recommendations are
given to resolve the energy crisis –
which by far has proven to be the most
binding constraint to industrial growth
and development over the past couple
of years. Effective management of load
shedding, peak load pricing and the
need for preferential treatment of the
industrial sector over both consumer
and commercial sector is stressed.
Captive power generation in industrial
estates and special economic zones is
also recommended. It is suggested that
the long term energy mix should shift
towards more efficient/cost effective
sources, such as wind, hydel and solar
energy. The need for the development
of localized machinery for hydel,
thermal and coal based power plants is
also pointed out.
After highlighting the issues and
problems with the country’s logistical
infrastructure there are specific
xviii
recommendations given to upgrade it.
For example, major restructuring of
the railways and new investments in
freight services are suggested. To
reduce transportation and trucking
costs the building of Logistical Parks
near industrial estates is
recommended. The importance of an
Automated Customs Clearance system
is highlighted in Ports and it is
suggested to maintain and improve the
existing computerized clearance
system. Finally, the importance of
Gawadar as a potential vent for future
growth is emphasized with suggestions
to harness this potential through the
help of China.
To promote the growth of existing
and resource based industries policies
are put forward to create industrial
estates and agro-processing zones at
the identified ‘hot spots’ of economic
activity. The intervention of the
government is specific to technology
up-gradation, branding, improving
quality standards such as phytosanitary
measures and facilitating
access to international markets.
Provision of hard and soft
infrastructure and common effluent
treatment plants in the industrial
zones is also put forward.
The issues in factor markets, labor,
land and credit, are also dealt with and
appropriate polices are derived to
tackle some of these. Various
mechanisms are suggested to improve
Labor skills and training. For example,
specific programs such as a Skill

Development Fund and a Skill based
wage subsidy scheme are proposed to
provide incentives to businesses to
invest in worker training. In terms of
credit markets, the rationing of small
businesses is shown to be one of the
major constraints to their growth.
Policies such as credit guarantee
schemes and promotion of venture
capital funds are discussed as possible
mechanisms to address the credit
market failures arising from
asymmetry in information. In case of
land markets, the need for achieving
security of property rights by
expediting the computerization of land
records is underscored.
There is a detailed discussion on
strengthening governance to stimulate
investment and manufacturing growth
in the country. It is noted that
businesses spend too much time and
money dealing with bureaucratic red
tape and meeting the “informal” costs
of relations with the state. Also,
uncertainty about the security
situation acts as a deterrent to outside
investors from seeking business
partnerships in Pakistan. The judicial
system works slowly and the sanctity
of contracts and of land rights can be
opaque. Policy proposals addressing
these governance issues are also put
forward.
Finally, a strategic recommendation
given in this section is the creation of a
full scale Science Park to promote
knowledge based industries in the
country The Science Park will have
xix
formal operational links with
universities and research centers. It
will have the necessary infrastructure
to facilitate R&D, manufacturing,
marketing and branding – covering
thus the entire product cycle. The Park
would not only provide incubators for
scientific innovations it will also be the
production center for cutting edge
products. This initiative would attract
professionals, businessmen and
scientists which would to an extent
reverse brain drain facilitating the
technological up-gradation of industry.
The Science Park would also allow
government policy coordination and
targeted government interventions to
knowledge based and technology
intensive sectors falling under the
ambit of WTO rules and regulations.
Competitiveness of Key
Industrial Sectors of
Pakistan
The second section of the industrial
policy report is a micro level analysis
as it focuses on the issues of selected
firms and sectors within the country.
This section comprises analyses of
both large and small manufacturing
sectors. In both cases, the
competitiveness of each sector is
assessed and policy options are
proposed to enhance it. This is done by
conducting in depth consultative
sessions with sector representatives,
analyzing all existing literature on
sector’s in Pakistan and for some
sectors through a simple value-chain

analysis supplemented by a broader
industry assessment that
contextualized value-addition achieved
in a particular chain. This
methodology helps identify those
factors which inhibit value addition
and result in relegation of firms to the
lower rungs of the global value chain.
The reasons behind the lack of value
addition and competitiveness is
identified with particular emphasis on
factor conditions, demand conditions,
domestic rivalry, related industries
and the role of the government. This
broader analysis provides insights that
became the basis for vertical policy
interventions that are both strategic
and emergent.
We are able to cover all major large
scale sectors, key SME export sectors
and light engineering sectors. This
coverage includes primary/ancillary
industry such as Steel, Chemical and
Fertilizer. We include knowledge based
sectors such as Auto, Electronics and
Pharmaceutical including Bio-
Chemical. And finally, we look at
eighteen small and medium scale
enterprise sectors some of which are
critical for export and some feed in
directly into the local markets. The
sectors in this category among others
include, Surgical Instruments, Sports
Goods, Fans, Cutlery, Agro-food,
Leather, Furniture, Marble and
Granite, Gems & Jewllery and Light
Engineering Sector.
We argue that the inclusion of
primary sectors is necessary as almost
xx
all the value added industry depends
for their inputs on these sectors. For
example, leather tanning requires the
use of 101 chemicals, out of which
around 98 are currently being
imported. This not only adds more to
the costs but also results in shortages.
Similarly, the pharmaceutical sector is
entirely dependent on imported
chemicals. We recommend that if
Pakistan is to move more towards
value addition then government will
have to invest in building the chemical
industry in the country. One such
recommendation is to set up a Naphta
cracking facility. Moreover, steel is an
input into almost all the value added
industries. It is surprising to see that
even with a large local demand for
steel the Steel Mill of Pakistan is
running in significant deficits. We
recommend to the government to
improve governance structure of the
Steel Mill and triple its capacity over
the next five years and make
investments to enable the Steel Mill to
manufacture high grade steel. We also
propose that government should use
the existing iron ore deposits of
Pakistan to its full benefit.
In this section we also analyze few of
the sun-rise sectors. The selection of
the sun-rise sectors is based on where
we believe gaps exist in the global
value chain and there is room for
Pakistan to augment its strategic
comparative advantage to competitive
advantage. We argue that such
opportunities mainly exist in

knowledge based industries such as
Auto, Electronics and
Pharmaceuticals. While analyzing the
auto sector we find that the past
policies were heavily influenced by
MNCs and the interests of the local
industry were highly compromised.
We present evidence that in periods
the deletion policy was followed,
significant growth of local industry
resulted, however, since the
introduction of Trade Related
Intellectual Property Rights (TRIPs)
there has been no indigenization or
growth of the local auto industry. The
licensing regimes of international
collaborators especially in the tractor
industry are prohibitive. They limit the
permission of sales, especially in value
added export markets. We recommend
ways by which government can
promote outward FDI and assist the
tractor manufacturers, for example, to
acquire international brands. The
deletion program is also strongly
supported, in fact, deletion should
occur at a much higher pace.
At this point in time, no country can
afford not to develop competencies in
electronics. Electronics pervade almost
all industries and competitiveness in
any industry requires knowledge of
electronics both at the level of product
and process. Moreover, with increasing
affluence a large demand has
developed for electronic items such as
Televisions, Computers, Washing
machines, Air conditioners etc.
Similarly, with the Tata Nano, India
xxi
has demonstrated that producing
competitive automobiles is not the sole
preserve of rich, developed nations.
Developing countries are just as
capable of producing products that will
appeal to the global segment which
cannot yet afford automobiles but will
settle for slightly lower quality.
However, realizing the domestic
constraints and limitations we
recommend that Pakistan should only
focus on product manufacturing and
not component manufacturing. For
this, product design and development
institutes will have to be established
with strong linkages in national and
international engineering universities.
Similarly, Pharmaceutical, in fact
Bio-Chemical is a sector with great
future potential. Both India and China
are fast developing capabilities to
improve their market shares in the biochemical
industry. Pakistan cannot
afford not to develop this industry as
there is going to be enhanced demand
both locally and in international
markets. For development of such
knowledge intensive industries we
recommend that the government
should establish Science Parks and
provide attractive fiscal incentives to
attract investment into the biotechnology
industry.
Finally, in the economic activities of
most developing regions, small and
medium enterprises (SMEs) play a
major role and Pakistan in this case is
no exception. Its economy is
dominated by SMEs, which produce

most of its output and employ most of
its workforce. 1 Specifically, SMEs
constitute 90 percent of the economic
establishments and contribute 30
percent of GDP and 25 percent of
export earnings, and employ 78
percent of the non-agricultural labour
force. Hence, we have broadly looked
at around eighteen critical SME sectors
in Pakistan.
Our findings reveal that most of these
sectors face similar issues and
impeding factors to growth. One of the
major issues faced is the inadequate
availability of appropriate human
resource and skills. The existing
government entities such as TEVTA
and NAVTEC are providing class based
skills that are not necessarily
alleviating the problems faced by the
industry. During our field visits we
found that almost all industrial units
in Sialkot Small Industrial Estate had
vacancies for only skilled workers.
Another major impediment is low
levels of productivity, resulting from
both inadequate up-gradation of
technology and poor production and
floor management techniques. Due to
high levels of illiteracy and lack of
training, most of the entrepreneurs’
struggle with basic costing techniques
and lean production management
methods. The floor designs are
suboptimal resulting is high wastages
1 The data for all Pakistan indicate that SMEs
account for about 90 percent of all enterprises,
employ 80 percent of the non-agricultural
labor force, and account for approximately 40
percent of the GDP
xxii
of labor time and inputs and makes it
impossible to capture the actual costs
of production. Similarly, in most
industries, such as leather, sports,
surgical, fans, cutlery the technology
being used is outdated. When
combined with low scale of production
all this results in significant
inefficiencies and loss of productivity.
Hardly, any of the SME sectors employ
professional management or formal
systems of record keeping. We
recommend that the National
Productivity Organization of Pakistan
should play a lead role in
benchmarking the SME industrial
sectors and assist them in lean
management tools so as to extract the
maximum results from existing
capacities.
Almost all the SME sectors that were
considered face issues with innovation
and developing original designs and
products. This is indicative of the lack
of collaboration and link between
sectors and university engineering
departments and product development
centers. In fact there is hardly any
evidence of formal linkages between
research & development centers and
industry. Any innovation or product
design that takes place is a result of
reverse engineering done at the
initiative of the factory owner.
Government has funded some
common facility and product
development centers, however, their
capacity to deliver to the needs of
specific sectors is fairly limited.

Therefore, we recommend specific type
of linkages that need to be created
between SME sectors and research and
development centers. We also indicate
priority sectors where government
needs to immediately establish design
and research centers.
Finally, another issue facing the
sectors is lack of branding and
marketing. Pakistani products are
increasingly fetching lower price in
international markets. Moreover, it is
also becoming exceeding difficult for
Pakistani product to find access into
newer international markets. Given
the lack of innovation, most of the
industry acts as an Original Equipment
Manufacturers OEM hence branding
has always been limited. In addition,
marketing has been a consistently
weak area limiting the growth
potential of firms. Finally, all of the
above factors have made it extremely
difficult for the sectors to meet
international quality standards and
compliance requirements. The
international standards on quality,
health & safety and environment are
becoming increasingly stricter. With
their current characteristics Pakistani
products are being refused access to
many markets based on non
compliance to these standards. It is
imperative that the government not
only provides the hard infrastructure
such as laboratories and testing
centers but also builds the capacity of
Pakistan Standards Quality
Compliance Authority (PSQCA) to
enforce domestic quality standards.
This will ensure mandatory
compliance to some minimum
standards which will then build
xxiii
capacity of the sectors to meet similar
standards in international markets.
Spatial Concentration of
Economic Activity
The most striking feature of
economic activity in Pakistan is the
geographic concentration (clustering)
and location of factors of production in
few cities, including unequal spatial
distribution of income, poverty,
education, health and physical
infrastructure, among others. The
concentration of economic activity in
few metropolitan areas symbolizes the
coexistence of development and
underdevelopment within and between
regions. We examine the economic
geography of Pakistan by exploring
regional and spatial inequality and
document various mapping measures
to highlight spatial distribution of
population, regional poverty and
income inequality, spatial disparities
in social infrastructure, road
infrastructure, and agglomeration of
manufacturing industries. We employ
different measures of concentration to
gain insight into spatial inequality (in
poverty, incomes, road infrastructure
and industrial concentration) as a
function of different push and pull
factors that explain concentration
versus dispersion of economic
activities.
Our results suggest that investment
in social infrastructure and physical
infrastructure is highly concentrated in
metropolitan cities, big cities and their
surrounding districts while districts
located away from these urban
demand centers are lagging behind.
While we have evidence of
convergence in the leading districts,

the gap between the leading and the
lagging districts is increasing with the
passage of time, which should worry
policy makers who are interested in
bringing about spatial equality.
Poverty Impacts of Public
Investments and Causes of
Industry Agglomerations
The econometric evidence shows that
poverty coexists with illiteracy of
household heads and lack of household
asset ownership. Therefore, the
government emphasis on human
capital accumulation and social and
physical infrastructure development as
tools for poverty reduction makes
sense. Policies that seek universal
access to education by increasing the
quantity and quality of schools and
colleges seem to have a strong power
to reduce poverty. While current
attention to investment in social
infrastructure is clearly appropriate,
policy inattention to spatial inequality
in income is very costly, especially in
more deprived districts. Our empirical
analysis of the evolution of poverty on
the basis of high quality district level
data on post-primary education and
hospital infrastructure index indicates
that increased public sector
investments on education and health
infrastructure lead to a sharp decline
in poverty in all but highly unequal
districts where these investments are
not associated with a decline in
poverty. Policies that encourage
investment in social infrastructure by
also promoting more spatial equality
in the distribution of income may yield
higher returns.
xxiv
The econometric results further show
a strong negative association between
road density and poverty incidence.
The magnitude of fall in poverty due to
road density is even higher than the
effect of investment on post-primary
school system and hospitals. We also
find that the long run effects of road
density on poverty are far greater than
the short run effects. We conclude that
income equality matters as far as
public investment on infrastructural
projects is concerned. The long run
poverty alleviation potential of
investment on roads almost doubles
when we move from high income
inequality districts to low income
inequality districts. Therefore, we
reiterate that pubic policies that seek
more regional equality in incomes are
far more desirable for pro poor growth
policies.
This study has also examined the
nature of geographic concentration of
manufacturing industries in Pakistan
by using the Ellison and Glaeser
concentration index. While there is
little doubt that increasing returns to
scale associated with agglomeration
externalities do exist at a wider scale in
Pakistan, it is much more difficult to
identify factors that cause industry
agglomeration. We have also explored
how geographic concentration of
manufacturing industries emerges
from the dynamic process overtime
and what is the nature of
agglomeration economies. Our
findings show that agglomeration of

manufacturing industries is
widespread in Pakistan where only
27% of the industries are not
agglomerated, 35% are highly
agglomerated and 38% are moderately
agglomerated. However, industry
concentration remained roughly
constant from 1995-96 to 2000-01, but
it drastically fell by about 33% in the
next five years.
Exploring the causes of
agglomeration, we find that district
population, road density, and the pool
of technically trained workers all help
promote agglomeration of
manufacturing industries. The
determinants of industry
agglomeration guide us on the causes
of dense economic activity across
spatial units, and the difficulties faced
in attracting manufacturing activities
in remote districts. However, a range
of policy instruments have already
been tried in Pakistan, e.g., tax
holidays, building infrastructure in
industrial estates, free trade zones,
export processing zones, etc. There is
no evidence on their systematic
success or failure. We, therefore,
advise that more empirical research
need to be conducted to evaluate the
effectiveness of past policies to
conclude under what circumstances
these programs and policies are likely
to succeed.
Exploring the relative strengths of
localization versus urbanization
economies our results suggest that
localization economies (within-
xxv
industry externalities) are much more
important than urbanization
economies (inter-industry spillovers),
which implies that industry-specific
subsidies and infrastructural
investments are likely to yield much
higher pay offs. However, productivity
growth in the manufacturing sector as
a whole is almost stagnant since 1995-
96. Therefore, policies aimed at
rebuilding the industrial sector of
Pakistan by adopting coherent
strategies are long overdue to which
we turn to in the next chapter titled
principal recommendations.

Principal Recommendations
Horizontal Recommendations
Achieving Macro­economic Stability
Pakistan’s economy frequently suffers from macroeconomic instability, which increases
risk and uncertainty, distorts resource allocation and adversely affects total factor
productivity growth. The root cause of this problem is low tax to GDP ratio (at around
10%), which gives rise to low national savings rate, large fiscal deficits, increased money
creation and borrowing, which in turn leads to higher inflation rates, higher current
account deficit and increased debt to GDP ratio. Higher inflation rates are often
followed by higher nominal interest rates, both of which not only hurt the welfare of the
people but also increase the cost of doing business, making domestic firms less
competitive while simultaneously discouraging new investment. Moreover, repayments
become increasingly difficult for operating firms, forcing them out of business.
Therefore, a strategy to ensure macroeconomic stability must be based on an increase in
tax to GDP ratio. Moreover, the recurring boom and bust cycles in the country are
symptomatic of a weak and uncompetitive manufacturing and export base. A diversified
and competitive manufacturing sector by generating enough foreign exchange through
export earnings would act as buffer against any short term shock which otherwise would
create external and internal imbalances.
In this context, the Ministry of Industries & Production should strongly advocate:
• An increase in tax to GDP ratio from the existing 10% to 15% in the next five
years, i.e., an increase of one percentage point per annum.
• Raise taxes against the provision of ‘service efficiencies including public goods’
that are required by the businesses 2 .
2 Businesses (especially large ones) benefit directly from enhanced government efficiency and better provision of necessary public
goods. Therefore if industry sees their tax receipts being translated into improvements in service efficiency and in the provision of
public goods, the incentive to contribute to the national exchequer would increase. Public policy in designing systems of taxation
should ensure that the link between taxes and public goods and service provision is strengthened. In this regard, pilot projects
addressing urgent needs of the industry can be initiated to raise tax revenue.
i

• Expansion of the tax base by including those sectors in the tax net that are
currently not being taxed.
• Prudent monetary and fiscal policy so that the rate of inflation and the nominal
interest rate do not increase to double digit levels.
Bridging the Development Gap between Regions
There is a widening infrastructure gap across districts. The evidence shows that not only
public education, health and road infrastructure is highly skewed across districts, the
gap between most-developed and least-developed districts is also widening over time.
These infrastructure gaps serve as major impediments in (1) optimal resource allocation
in deprived districts on the basis of comparative advantage, and (2) market access to
domestic producers, traders and service providers in least-developed districts.
Moreover, these infrastructural gaps increase poverty, promote unbalanced industrial
development and uneven location of urban population. This pattern of development is
politically unsustainable in the long run.
The Ministry of Industries & Production should strongly advocate to:
• Introduce non-income based poverty mapping of districts to identify most
deprived and least deprived districts in the country.
• To reverse the trend of widening infrastructure gaps across districts, use
education and health based mapping to disproportionately allocate development
funds to least developed districts for the next 10-years.
• Allocate more funds to develop quality road network in deprived districts. For
this purpose, road density based mapping of districts needs to be employed to
allocate more funds to deprived districts for the next 10-years.
Meeting the Energy Challenge
The ongoing energy crisis has crippled Pakistan’s industry and economy. A conservative
estimate puts the loss to industry at 13% of total manufacturing sales or nearly Rs. 130
billion per annum. The energy crisis is hitting the industry at multiple levels: energy
tariff increases are forcing businesses with low margins and those who are unable to
ii

generate their own power (e.g., SMEs) to close down; unannounced load-shedding and
voltage fluctuations damage machinery worth millions of dollars; unavailability of
electricity harms productivity of workforce. With an abundance of two of the cheapest
sources of energy in the country, there is no reason why Pakistan should not be able to
overcome this crisis soon.
To overcome this crisis The Ministry of Industries & Production should strongly
advocate the following:
• As long as power shortfall remains, load needs to be managed carefully. Areas
with a heavy presence of industry, both large-scale and SMEs, should be given the
status of industrial corridors. These corridors could be given separate feeders,
where load shedding will only occur when absolutely necessary3. This needs to be
finalized in consultation with PEPCO so that the present suboptimal matching of
grid stations and 11kv feeders can also be addressed. Load-shedding schedules for
such units need to be announced at least 2 months in advance and load shedding
days should be clustered for both electricity and gas. No such priority should be
given to industry that is based in residential areas.
• Conduct sector-wide energy audits, initiating the programme from heavy load
industries. Based on the recommendations of the audit, provide incentives to the
industry to shift toward more energy efficient production methods and
technology.
• In the industrial corridors, peak-load pricing schedules should be announced.
• In the medium to long-run, we strongly propose a shift in energy mix. This may
be helped by developing localized, cheaper machinery for hydel, thermal and
coal-based power plants. For this purpose, the interaction of manufacturers and
power producers, including WAPDA needs to be facilitated to develop machinery
best suited to Pakistan’s needs.
3 This point is also addressed in the section on Industrial Estates and other special economic zones.
iii

• In special economic zones, science park and industrial estates, captive power
generation should be allowed.
• At the same time, steps should be taken to facilitate the local development of
wind turbines and solar energy technology. For this purpose, two pilot research
projects should be initiated bringing together universities, industry, foreign and
local experts and relevant government departments.
Upgrading Logistical Infrastructure
An efficient and well integrated transport system is integral to the growth and
development of the industry as it creates access to markets, reduces the costs of
production and increases the competitiveness of firms. In Pakistan, however, the
transport systems suffer from insufficient investments, poor management and a lack of
essential maintenance. The low reliability and high costs of the transport sector hinders
Pakistan’s economic growth by reducing the productivity of its industry and
competitiveness of its exports. The cost of transport system inefficiencies has been
estimated at about 4%-6% of the GDP.
In order to increase the efficiency of logistics in the country, The Ministry of Industries
& Production should strongly advocate the following measures:
ROADS & TRUCKING
• In partnership with the private sector international standard Logistical Parks
near industrial areas, agricultural hubs, and ports need to be setup. For this
purpose, land needs to be provided, and government should invite local/foreign
investors to develop facilities for storage, loading-unloading facilities, and rest
areas. Dry ports should be established within or adjacent to these logistical parks.
• Vehicle quality testing stations and Radio Frequency Identification (RFID) based
tracking facility should be established within the logistical parks.
• A roads commission should be set up to develop the overall strategy for roads
giving priority to areas with relatively low connectivity. It would include in its
iv

purview connecting roads to major industrial estates, railways and border points
as well as connectivity with Baluchistan and tribal areas.
• The government should standardize and provide technical advice, licensing and
registration of trailers.
• The government should ensure the creation of cold storage facilities near
agricultural hubs and expedite the creation of cold chains from agro-based
clusters to Karachi.
RAILWAYS
• The condition and performance of Pakistan Railways needs to be improved
significantly. New investment is required in the Railway Freight Service and new
tracks need to be laid in order to support transportation within the country.
• It is proposed to increase the share of national railway in national freight from
the current 4% to 22% by 2030 as proposed in Vision 2030.
PORTS
• Pakistan Automated Customs Clearance System should be maintained and the
software issues need to be solved immediately.
• The government needs to facilitate the private sector to invest in bulk handling
port facility. This could be done through public-private partnerships by following
the ‘landlord concept’ employed successfully in the Karachi Port Terminal.
• The government should facilitate the establishment of silos at Karachi port for
the storage of commodities such as coal and cement.
The Case of Gawadar:
Gawadar is located in Baluchistan whose strategic location needs to be exploited. It is
east of Iran, south of Afghanistan, and boasts a sea port almost at the mouth of the
Strait of Hormuz. Gawadar is an essential node in the Iran-Pakistan-India pipeline,
v

which could cross from Iranian to Pakistani Balochistan. At the same time, Gawadar is
of enormous strategic importance to China, Afghanistan, Iran and Central Asia.
However, the development and usage of the Gawadar Port has stalled as a consequence
of a myriad of issues ranging from lack of infrastructure investment by the government,
prohibitively high cost of transportation, security issues, absence of any significant
industry in Baluchistan, and foreign interests which feel threatened by this
development. Accordingly, the following policy recommendations are given:
• Scrap the deal with PSA and seek assistance from China in developing the port.
• Urgently complete road works linking Gawadar to upcountry. Currently,
transporters have to go through Sukkur.
• Incentivize China and/or UAE to build oil refineries near the port.
• Seek Chinese assistance in building a rail link between Gawadar and northern
parts of the country.
Promotion of Knowledge­based Industries through a Full­Scale Science Park
A cornerstone of this industrial policy is the establishment of a science park in Pakistan.
The proposed science park will have formal operational links with one or more
universities, research centers, or other institutions of higher education. It will be
designed to encourage the formation and growth of knowledge- based industries. It will
have all the infrastructure required to attract professionals, consultants and managers
and will house organizations ranging from R&D to manufacturing, marketing and
branding. In short, the objectives of the establishment of a science & technology park in
Pakistan are the promotion of technology commercialization, transfer and diffusion by
fostering links between Industry and universities, R&D labs, and promoting the
formation and growth of knowledge-based companies.
A science park fulfils several purposes that are critical to the success of the 2010
industrial policy. Most importantly, it enjoys the status of an R&D establishment, which
means it sidesteps the same WTO regulations that prohibit any subsidies to local
vi

organizations. Local organizations can be encouraged to indigenize through collective
government sponsored R&D and build brands within a science park.
Second, a science park captures the entire product cycle, rather than just focusing on
manufacturing. It provides incubators for scientific innovations as well as a production
center for cutting-edge products. Universities will be playing a very strong role in the
emergence of the proposed science park and will be required to supply trained human
resource and a knowledge base for industrial innovation.
Third, a science park provides economies of scale and scope with respect to common
infrastructure facilities (such as transport, power, information and communication
technology connectivity, office and production space, and waste treatment) and
technical services (such as recruitment, training, mentoring, financing, networking, and
legal and IPR consulting).
Finally, through delegation of powers to the science park level, various ministries and
government bodies can achieve coordination, a goal that otherwise remains difficult to
achieve.
The aim is to create an environment where knowledge-based industries can thrive and
the required professionals can reside with their families. We envisage the park project to
be 6-year long with approximately US $500 million required in startup costs. The total
area for the park will be about 200 hectares. The project will be executed in 3 phases.
Phase 1 will include the survey of different sites in Pakistan for the park and the
selection of the most suitable site. The 2nd phase will include building the infrastructure
as well as building and recruiting park clients. The last phase will include facilitating the
investment environment for the clients and building bridges between different
universities of Pakistan, R&D institutes and park clients.
The government of Pakistan will incentivize the investment environment in the science
park through the following measures:
• Absorb initial startup costs for the science park
vii

• Provide infrastructure of the park including amenities
• Establish venture capital funding for startups
• Provide five year tax holiday
• Ensure industrial laws and regulations simplification
• Ensure delegation of power from several other government agencies to park
administration for ‘one stop operation’
• Pass special commerce and tax laws for science park operation
• Ensure cooperation programs among universities, research institutes and park
clients
• Provide transparent, competitive R&D grants
Development of Industrial Estates, Special Economic Zones, Export Promotion Zones
The following is proposed in this respect:
To promote the growth of existing and resource-based industries the government should
create industrial estates and agro processing zones at the identified ‘hot spots’ of
economic activity. This should be done while preserving agricultural lands and
observing municipal zoning laws.
The government should develop these estates through the provision of necessary soft
and hard infrastructure and world class logistic facilities.
The government should facilitate estates and zones in:
• Accessing appropriate technology.
• Obtaining information about modern methods of production.
• Improving quality standards such as phyto-sanitary measures for agriculture
processing.
viii

• Product branding and up gradation.
• Accessing international markets.
• This should be done by creating linkages with universities, research centres
available locally and internationally, and by creation of centres of excellence.
• Captive power generation should be permitted at the estate level.
• Effluent Treatment Plants and Solid Waste dumping sites should be established
in all industrial estates. Incentives such as tax break should be provided for local
construction/fabrication of effluent treatment plants.
• With falling duties, the attraction of EPZs (where imports are duty free) has fallen
dramatically. EPZ such as the one in Gujranwala should be turned into an
industrial zone.
• The government should rationalize its existing portfolio of industrial estates. The
failed estates or SEZs (e.g., Gadoon) need to be liquidated and lessons learned
from their failures. Sukkur Small Industrial Estate should be developed.
• Business incubators for SMEs should be set up to reduce costs for business startups.
Strengthen Governance
There are several issues in governance that can easily compromise any industrial policy.
The following is a list of recommended measures:
Smuggling (Afghan Transit Trade): The Afghan Transit Trade Agreement is a serious
issue for industry. A range of goods which are officially imported for the Afghan market
make their way into Pakistan through well established and strong smuggling networks
operating at both sides of the Afghan-Pakistan border. Prices of these smuggled items in
the local market are much below those of local manufactures resulting in an erosion of
their market share and a loss in revenue. One way to control smuggling is to increase the
ix

transaction cost of smuggling and reduce the wedge between importable prices of legal
and smuggled goods.
The following are some suggested policies in this regard:
• The government should increase controls at Karachi port and Afghan border to
curb smuggling.
• Efforts should be made for tariff harmonization between Afghanistan and
Pakistan. A bilateral agreement could be made to adjust tariff from both sides
with the objective to minimise tariff deferential so that it reduces the premium on
smuggling.
• Afghan custom clearance should be at Karachi.
• Letter of Credit for import of these products should only be issued in
Afghanistan. This has already been implemented.
• The government should negotiate some degree of quantity control on products
meant for Afghanistan based on the market size or demand estimates in
Afghanistan.
• Rationalization of sales tax regime in Pakistan would also help in reducing the
premium on smuggled goods that are manufactured domestically; e.g. Tyres.
• Another serious issue concerns under-invoicing. To tackle this problem, the
government should:
• Use the International Trade Price (ITPs) lists to reduce the incidence of underinvoicing.
• Cost of Doing Business: Costs of doing business effect both domestic and foreign
investment in the country. These costs relate to bureaucratic inefficiency or red
tape, endemic corruption, complicated systems of taxation and business registry
x

to weak contract enforcement systems stemming from dysfunctional provincial
judicial systems. We propose the following:
• All labour related issues to be handled at a single window.
• All labour related tax dues (EOBI, WWF, Social Security, etc.) to be lumped as a
single payment.
• Online system for registration of new businesses should be developed.
• All unnecessary procedures of starting a business, such as the requirement to
establish a company seal, should be abolished.
• Refunds of duty draw backs and sales tax should be completed in one month.
• Businesses should be allowed to keep the 3 months security deposit with Sui Gas
in National Saving Certificates.
• Contract enforcement needs to be improved through the following measures:
• Setting up of specialised courts or commercial divisions in existing courts
• Design and implement case management systems in courts.
• Introduce a maximum time limit of 3 months on all commercial cases.
• Strengthen the Alternative Dispute Resolution (ADR) System. Encourage and
support the provinces to replicate the ADR system piloted in Karachi.
Building World Class Human Capital
Due to years of neglect the human resource capacity in Pakistan has deteriorated to
alarming levels. In order to develop a competitive advantage across particular
industries, the level of human resource needs to be go up significantly. For existing and
resource-based industries generic and specific skills need to be developed. For
knowledge-based industries, an additional requirement is to attract Pakistani expatriate
professionals currently working in the West.
xi

Accordingly, the government should take the following measures on a priority basis:
Primary and Secondary Education: Universal Primary and Secondary Education is
a Pre requisite for Skills Development. Primary and Secondary education is the right of
every individual in the country; therefore its free and universal provision is an integral
responsibility of the government. Through the widespread provision of basic education
Pakistan can successfully turn its demographic challenge into a formidable strength.
Along with many other positive externalities generated by education, it significantly
increases the trainability of workers and thus has a direct effect on labour productivity
which in turn leads to industrial growth. Therefore it is essential for the government to
give the highest priority to primary and secondary education. It is proposed that the
following steps should be taken to increase both the coverage and quality of basic
education.
• The government should drastically increase the budgetary allocation for
education. This increase needs to be maintained for successive years. Currently
Pakistan spends 2 percent of its GDP on education; this has to increase to at least
4 percent in the next 5 years. The allocation should also cater in providing
essential facilities such as infrastructure, sanitation and water, and improving the
accessibility of schools.
• The curricula at the primary and secondary level should undergo periodic
revisions and over a phased period aim at reducing the mismatch between skill
concentration and education and market demand.
• The government should also consider a phased shift towards uniform curricula
across private and public schools in Pakistan at the primary and secondary level.
• Teacher training programs should be established to strengthen the capacity of the
teachers to effectively deliver the national curricula.
• The recruiting criteria for teachers should be strictly on merit and should be done
in a transparent manner.
xii

• Public schools administration should have the flexibility to reward teachers who
produce better educational results.
• Together with increased incentives for teachers there must be increased
accountability on performance. The government should explore different
methods of accomplishing this, for example, by providing a framework for
parents to be more involved in the education of their children.
Technical Education and Vocational Training: Although the government has come up
with various initiatives to improve skills and technical education there are some serious
gaps in the existing system. The provision of skills and vocational training is not in sync
with the demands of the industry. The curriculum and class room teaching methods are
outdated and have not developed the capacity or mechanism to respond to the needs of
the industry. In 2009 the National Vocational and Technical Education Commission
(NAVTEC) developed a strategy to overcome some of these issues. The vision of the
2009-2013 NAVTEC skills strategy stresses a shift from supply led training to demand
driven skills development by promoting the role of industry in both the design and
delivery of TVET. In light of the NAVTEC’s skills strategy the following are the main
policy proposals:
• The government should map out all the existing provincial federal and private
skills and vocational training institutes operating across the country.
• The government should evaluate the condition, available infrastructure, quality
of trainers and relevance in terms of meeting demands of the industrial sectors.
• The government should upgrade the existing training institutes and only
establish new training institutes – Centres of Excellence in those areas/sectors
where they do not exist. Under the Nation Skills Strategy the government
proposes to promote and facilitate the establishment of sector specific training
institutes and Centres of Excellence. These institutes are to be located in
proximity to relevant industry in order for them to benefit from new
technologies, and to facilitate industry placement of trainees and trainers and
xiii

information flows between the training institute and industry. The existing
institutes should be upgraded based on the following strategy:
• Upgrade the entire range of curricula and make it more demand driven and
industry specific.
• Improve the quality of the faculty and the master trainers.
• Replace outdated equipment with more relevant equipment.
• Assist the training institutes to develop international linkages to offer certified
courses such as, City and Guilds UK. The NAVTEC’s strategy also proposed to
help Pakistani Centres of Excellence establish partnerships and links with
counterpart reputed industry specific training facilities abroad.
• Develop a framework for continuous feedback from industry to assist the training
institutes in maintaining their usefulness and relevance. The institutes should
also develop job exchange system whereby they can track the placement of their
trainees.
In order to facilitate on floor training a skill based wage subsidy program should be
piloted in a few sectors. The program is designed to incentivize the factory owners to
train workers on their factory floor and also register them with locally available training
institute for classroom based general training. The firms will pay seventy percent of the
minimum wage to the worker, whereas the remaining thirty percent would be the
government’s contribution payable through the training institute directly to the worker.
To link the industry and government, sector specific Industry Advisory Groups (IAG)
may be established. Each IAG will be represented by members of large, medium and
small industry, including all sub-industries that fall within the category, international
employers, and employees. Their responsibilities will be to carryout periodic sector
surveys, identify skill needs in their sectors, indicate new and emerging areas and
occupations and determine and update competency standards for workers. IAGs should
xiv

e established gradually over time, with the aim to establish two each year until all the
major sectors are represented.
Leading firms should sponsor a training unit within a public sector institute, where
trainees can be trained specifically on their machines and according to their
requirements and standards. Such a model may also be useful for clusters of industries
which do not have their own training establishments and have difficulty financing
training individually.
Higher education – University- Industry Linkages: Simply increasing the number of
engineering universities will not achieve much. Industry needs to be the driver of higher
education in engineering and science. Engineering and science degrees need to be
developed around industry needs. Furthermore, there is a dire need for an effective
linkage between industry and academia. Following are some proposals to effectively
enhance the University-Industry linkage:
• Science and engineering universities need to introduce new courses and degrees
in view of local industry needs.
• With collaboration from industry state-of-the-art laboratories should be
established in universities for R&D needs of business.
• Government will sanction $10 million for joint research projects involving
universities and industry based around actual industry problems.
• Provide incentives to attract Pakistani and other professionals from abroad. The
Science Park is proposed as an initiative which can resolve this issue by providing
them with all the amenities they require to relocate.
Securing Land Markets
The current land registration laws present certain elements that make the acquisition
process lengthy and often require high informal costs. In addition to the formal land
registration and land tenure system, there are strong and widely used informal land
transfer systems, such as oral commitments, collateral assurances, powers of attorneys
xv

etc., that operate outside the formal land registrations. The existence of these parallel
systems is said to have created serious complications by providing shelters to land
mafias, registration fee evaders, and public sector rent seekers. An even more critical
issue in terms of land availability is the "title of record" dimension of land parcels.
Under current legislation, the records of rights in land primarily reflect a fiscal
responsibility. The person mentioned in the records is liable for land revenue or
property tax; that he is also the rightful owner is only incidental. The land registration
documents only have persuasive, not conclusive, evidentiary value.
We propose that the government should:
• Reform the Land Registration ACT to provide security of titles.
• Ensure that the Provinces complete computerization of all land records
Effectively Using Trade Policy
Trade policy reforms in the 1990s and 2000s in line with the WTO agreement have
reversed the protectionist, inward-oriented import substitution policies of the past
decades. Since 1996, substantial trade liberalization and tariffication has been realized
through tariff cuts, tariff rationalization, removal of quota regimes and import
surcharges. Pakistan has also signed regional trading agreements with neighbouring
countries.
Keeping in mind the limited policy space available to the government the following
proposals are put forward:
• Local knowledge-based industry must be supported within the ambit of a Science
Park (proposed above). A Science Park allows for support while avoiding conflict
with the WTO regime.
• Trade policy is an instrument of industrial policy, therefore it should be made in
conjunction with industrial policy and not independently. It is proposed that the
Ministry of Commerce in the future should frame its trade policy in accordance
with the broader objectives of the industrial policy.
xvi

• To protect the value added industry, the government should rationalize tariffs
according to the following escalating schedule: (1) raw material: 0 – 5%; (2)
intermediate goods: 5% – 10%; and (3) final goods: 15% to maximum bound
tariff.
• Given limited policy space in tariff based measures, the government should
effectively use non-tariff measures to its advantage in line with the WTO
agreement. In this regard, government should more effectively use border control
measures such as Sanitary and Phytosanitary (SPS) measures, quality standards
such as technical barriers to trade (TBT), quantity and quality measures such as
Pre-Shipment inspection (PSI), temporary measures to slow imports or to impose
import restrictions such as Safeguards (SG) and Balance of Payment Measures
(BOP) to protect local industry.
Facilitating SME Growth
Strengthening Institutional Support to SMEs
According to the most recent study conducted on SMEDA, Pakistan’s premier agency for
supporting small and medium sized enterprises: “Pakistan’s economy is dominated by
SMEs with more than 90% of enterprises belonging to this category. The SME sector is
suffering from many constraints including lack of access to finance4, limited access to
markets, lack of infrastructure, hostile business environment, corruption and red tape,
weak management and lack of access to skilled labour. Also, many of the government
policies are devised from the perspective of large firms and not SMEs. The
implementation of SME policies in Pakistan is fragmented and limited and needs to be
more effective in light of the SME sector’s importance and contribution”.
4 Small and Micro enterprises are often rationed out of the credit market due to information asymmetries such as
opaque knowledge of firms and sectors on the part of commercial banks. Information asymmetry results in relatively
high collateral requirements for SMEs which along with the degree of documentation required for loan application
deter them from resorting to the formal sector credit market – commercial banks. Lack of credit availability thus
severely impedes the growth potential of small scale and micro enterprises. Data suggests that over 90 percent of small
enterprises finance their business and working capital requirements through retained earnings.
xvii

In view of the credit problems faced by SMEs, the State Bank of Pakistan has already
initiated a few measures. Below, we supplement these with additional programs to
increase accessibility of credit to the SMEs.
The first program launched under the Financial Inclusion Plan of the SBP is the Credit
Guarantee Scheme for Small and Rural Enterprises in which the SBP will share 60% of
the losses from short term and medium term loans (up to a maximum of Rs 5 million
per loan) extended by banks to SMEs. Each loan will have a tenor of up to three years
and will have an interest rate equivalent to the 3-month Karachi Inter Bank Offer Rate
(KIBOR) plus 300 basis points (bps). The initial capital for this scheme is to be provided
by DFID. The government should help in disseminating information about this
particular scheme through the various business/industry associations representing
SMEs around the country.
The second program is the Refinance Scheme for SMEs in KP, FATA and Gilgit
Baltistan, with a stated purpose of increasing the supply of credit to SMEs. The interest
rates for this program will be floating based on a formula linked to the rates of 6-month
Treasury Bills and Pakistan Investment Bonds (PIBs).
The procedural requirements such as excessive documentation of loan application by
commercial banks are perceived as a major hurdle by SMEs. The government (MOI)
should advocate with the State Bank to facilitate the development of a uniform
simplified and efficient procedure for loan disbursement to small and micro level
enterprises. This will substantially reduce transaction costs for SMEs and increase their
incentive to apply for a loan.
Information asymmetry is the principle factor behind lack of access of SMEs to credit, it
is imperative to have a Credit Rating Agency specialising in the SME sector. The State
Bank has successfully launched the Pakistan Credit Rating Agency, however it does not
cover the SME sector. An agency similar in scope but targeted towards the SME sector
should be supported by the government
xviii

The government should recommend to the State Bank to promote Venture Capital
Funds in Pakistan in order to stimulate small scale entrepreneurship in the country and
also rectify credit market failures present in the formal commercial banking system.
Venture Capital Funds do not only provide capital but also support nascent enterprises
and entrepreneurs with technical and managerial support. Such support is crucial for
the success and survival of a new business.
The State Bank along with the Ministry of Industries should provide information on
investment opportunities in the country and facilitate investment by expatriate
Pakistani’s both directly and through such Venture Capital funds.
Invoice based financing, where the borrower gives accepted invoices (or receivables) of
its business customers or downstream buyers, as collateral to the commercial bank
should be promoted by the State Bank as an alternative to collateral based lending.
Other measures that should be taken with respect to SMEs include:
• A cabinet committee for SMEs should be established to fast track decisions on
SME policy matters.
• Exemption limit for excise duty should be raised from Rs. 5 million to Rs. 10
million.
• A capital subsidy for investment in technology should be provided.
• Incubation centres should set up in sunrise industries. Some of the sunrise
industries include:
• Bio-technology
• Information Technology
• Electronics and telecommunication equipment
• Non-conventional energy sources
xix

• Establishment of testing and certification facilities should be financed.
• A vendor development programme should be initiated under which ‘buyer –
seller meets’ exhibitions are organised at regular intervals and at dispersed
locations.
• A fresh census of small scale industries should be conducted with special focus on
gathering information on the incidence of sickness of industries and its causes.
• Skill development initiatives specific to SMEs should be initiated. Schemes
covering the areas highlighted below should be developed:
• Training should be provided in quality management systems
• Training support for entrepreneurial and managerial development of SMEs
• Training and benchmarking for designing lean manufacturing techniques
• SMEDA should be institutionally restructured to enable it to design sectoral
programs for the SMEs that involve tailor-made investment projects in various
sectors. The new organization structure should also enable SMEDA to interact
with the government (federal and provincial) and its entities for developing
favourable policy environment for SMEs. SMEDA’s new organization structure
should also allow it to network domestically and globally to bring the maximum
benefits for the SMEs of Pakistan.
• Adequate human resource with greater financial flexibility should be made
available to run the revised organisations structure.
• The joint equity participation fund should be revitalised and should be allowed to
provide loans up to 25% of the project value with a maximum cap of Rs. 10
million. The mark up charged on these loans should be 5%.
Assisting Sick Industrial Units
In any business not all firms survive due to a variety of reasons however in relatively
efficient markets the productive capacity of failed firms is acquired by others. In
xx

Pakistan due to difficulties in liquidation of businesses a huge amount of productive
capacity lies dormant at any point in time. The government should institute a more
transparent and efficient procedure/system to help distressed firms get out of business
quickly and at least cost.
For this purpose the government should:
• Recommend changes to the bankruptcy laws to make it easier for businesses to
acquire distressed firms. The transition and exchange of hands in business will be
made swifter under these legal revisions.
• Also fully support the Corporate Rehabilitation ACT being developed by SECP
and ensure that the ACT covers the possibility of providing technical support in
reviving the sick industry.
Implementing Cleaner Production Programs
Growth of the industrial sector cannot be visualized without strengthening cleaner
production programs in the country. As defined by the United Nations Environment
Program (UNEP), cleaner production is “the continuous application of an integrated
environmental strategy to processes, products and services to increase efficiency and
reduce risks to humans and the environment.” The implementation of cleaner
production programs is an integral part of the policies that aim at increased
competitiveness and increased efficiency of firms because they help firms save energy,
conserve water, control pollution, ensure safety of machines and equipment, improve
health and safety of workers, improve environmental conditions and the image of the
firm at the local and international levels.
To achieve the goals of implementation of cleaner production programs and to promote
investment in cleaner productions, the Ministry of Industries & Production should
advocate the following measures:
• Encourage and promote investment in local manufacturing of cleaner production
equipment along with other emerging environmental technologies. However, in
xxi

the interim period the Ministry should advocate policies that ensure low prices
for imported cleaner production equipment.
• Pilot projects that help demonstrate costs and benefits of private investment in
cleaner production technologies, e.g., CO2 recovery, should be initiated as part of
awareness raising campaigns. Similarly, awareness raising seminars and training
programs should also be used to harness full benefits of cleaner production
programs.
• A policy for cleaner production needs to be devised that offers tax incentives and
tax rebates to compliant firms and to firms who want to establish Effluent
Treatment Plants. These incentives should also apply to those who get ISO 14001
certifications.
• Offer tax incentives to encourage private sector to set up quality testing and
certification laboratory facilities in all sectors, especially those relating to textiles,
sugar, leather and pulp and paper sectors.
• Allocate funds for next ten years to set up well designed landfill sites for disposal
of hazardous and non-hazardous industrial solid waste and installation of Hi-tech
incinerators.
• The existing National Environmental Quality Standards (NEQS) were developed
in isolation without consulting all the stakeholders. That is the reason why these
Standards are non-productive and hard to implement in their present form.
There is a need to revise the NEQS in consultation with all the stakeholders. The
new NEQS should be industry/sector specific and should be based on the ground
realities.
xxii

SECTORAL REFORMS
Development of Steel Industry
It is next to impossible for Pakistan’s industry to become competitive without a reliable
steel production base. Unfortunately, while Pakistan has an abundance of iron ore, local
steel production has not kept pace with increasing demand. For this reason, Pakistan
has become increasingly reliant on steel imports, exposing the economy to exogenous
shocks. The problem of an increasing production deficit is further compounded by
power shortages, poor infrastructure, outdated equipment, and shortage of skilled
human resource. Pakistan’s policy on steel is to decrease reliance on imports and
develop indigenous steel industry.
For this purpose, it is proposed that the government should:
• Triple the capacity of the PSM over the next 5 years to make it economically
viable. The government will take steps to attract investment into enhancing
PSM’s capacity to 3 million tonnes.
• Incentivize further steel production with a goal of producing 8 million tonnes of
steel indigenously by 2015 by establishing an industrial park. This park will be
given priority in terms of power, gas and financing. The park will also contain
facilities for foreign experts, whose visits will be sponsored by the government.
• Invest in the enhancement of existing knowledge resources. For this a steel and
metallurgy institute will be set up in conjunction with the PSM, universities and
downstream industries.
• Re-evaluate the impact on overall industry and the basis on which FATA was
allowed tax exemptions on running steel furnaces.
Chemical Industry
Chemicals are one of the primary raw materials for manufacturing, contributing to all
manufacturing sub-sectors, including pharmaceuticals, automobiles, textiles, furniture,
paint, paper, electronics, construction and appliances. The unavailability of an
integrated Petrochemical complex with a Naphtha cracking facility has constrained the
xxiii

growth of downstream industries especially synthetic textiles, paints, pharmaceutical,
and packaging, making them dependent on costly imports. The annual import cost of
raw material and finished chemical products comes to nearly $4billion/year.
In light of this situation, the government should take the following immediate steps:
• Begin groundwork for the establishment of a petrochemical complex in Hub. This
complex should include a refinery, a naphtha cracking facility, and facilities for
manufacturing of certain petrochemicals for which demand has been ascertained.
Hub is a suitable location because Byco (formerly Bosicor) refinery is already
located in the area and another refinery project is proposed at that location. After
expansion Byco will be the largest refinery in the country while National Refinery
and Pakistan Refinery are located nearby in Karachi. This will provide easy access
to Naphtha and reduce transportation costs. This will have to be a public-private
partnership given the magnitude of investment involved. The textile industry will
be a major beneficiary of this initiative. Despite extensive BMR, the industry
appears to have shifted down the value-chain in the past few years, with many
value-added units closing down. Also, while the world has moved to MMF-based
textiles, Pakistan has remained hugely dependent on cotton-based textiles. The
establishment of a Naphtha cracking facility should greatly boost the downstream
development of MMF-based textiles, technical textiles, and facilitate the
processing sector which is hugely dependent on imported chemicals.
• In order to proceed with the above initiatives, we recommend that the chemical
industry be brought under the purview of MoIP rather than the Ministry of
Textiles, where it is currently housed.
• The Petrochemical plant should be a government led, economically viable project
with equity from private investors. In order to make it feasible, maximum duties
will have to be imposed on all chemical-related end products, as soon as the
facility starts production.
xxiv

Fertilizer Industry
While the fertilizer industry has done well to meet the extensive domestic demand, the
current NPK mix is now seen to be leading to lower average crop yield as compared to
other countries. Furthermore, with an increasing gap in supply and demand for gas,
future viability of our primarily gas-based plants is threatened.
In view of this, the government should:
• Incentivize industry to add capacity in production of DAP. This will correct the
current NPK mix.
• Conduct energy audits of fertilizer manufacturing plants to assess their energy
efficiency, and vary their gas subsidy based on their efficiency.
• Incentivize fertilizer plants to invest in alternate modes of energy production.
• Explore the feasibility of coal as alternate feedstock.
Auto Sector & Farm Machinery
Until 2006 a deletion program was followed in the Auto industry. Due to some
structural reasons as well as poor governance leading to smuggling and under-invoicing,
the deletion program could not be as successful as one would have hoped. Indeed, by
many accounts after many years, with a few exceptions, major OEMs have achieved less
than 40% local content. This too is in components that are inherently difficult to import
(e.g., seats, batteries, tyres etc). Moreover, transfer pricing where it occurs acts as a
disincentive to produce locally. Where the auto industry has localized, it has yielded
highly favourable results generating almost 200,000 jobs and an extensive vendor
industry. Several parts are now produced competitively within the country, and in some
areas we are ideally poised to even export. However, a few barriers continue to prevent
this industry from growing further.
In view of these, we propose the following:
xxv

• Strictly enforce the Tariff–based System (TBS) which is officially in use, but
effectively in abeyance. This is absolutely crucial. In the absence of effective
monitoring, the progress of local auto industry will be severely impeded.
• Revise the indices that are still quoted for parts deletion. These are not based on
international prices and are misleading. New indices need to be based on
international prices of respective components, and the TBS needs to be brought
in line with these.
• Transfer pricing severely impedes the development of a local market, and
benefits parent companies of OEMs. While Pakistan officially has an ‘arm’s
length’ rule in effect, it is very weak. Application of the arm’s length principle by
FBR needs to be considerably strengthened.
• Granting of Pakistan-specific licenses for assembly restrict the export potential of
OEMs in parts of the industry where high levels of deletion have been achieved.
The government needs to take remedial steps to eliminate this constraint. In this
regard, priority will be given to three sectors: two-wheelers, three-wheelers and
tractors. These are the sectors where most deletion or indigenization has been
achieved, and these are poised for international growth. Starting with these
sectors, the government will a) complete deletion fully, b) facilitate the
acquisition of foreign brands and c) facilitate the establishment of joint ventures
where such restrictions are not in force.
• For the sectors identified above (2-wheelers, 3-wheelers and tractors) the
government should engage immediately in outward FDI. This will involve
purchasing small firms abroad with requisite R&D capabilities that can be shared
with Pakistani manufacturers, as well as those with known brand names in these
markets.
• Any firms completing full deletion and operating under locally owned brands
(including that purchased through outward FDI) should get a subsidy on markup.
xxvi

• Establishment of Clusters: Three different clusters should be organized within
the Auto and Farm Machinery sectors. Two automobile clusters should be located
in Lahore (Sheikhupura Road, near Motorway) and Karachi (near Port Qasim).
These will cater to cars, motor cycles and three wheelers. Furthermore, a Tractor
and Farm machinery cluster should be located in Daska (near Gujranwala).
• In order to facilitate deletion, and develop local parts based on global standards,
an Auto Design Institute will be established in conjunction with NED University.
All OEMs will be required to collectively sponsor this Institute, and transfer
technology through this initiative to vendors. The government should sponsor
foreign consultants to come and stay at the Institute for the next two years,
within which the Institute should complete full localization.
• Indigenization and development of local industry has suffered in the past
whenever fully made up cars (whether as cabs or otherwise) have been imported.
This practice should not be allowed in the future.
Electronics
Despite being one of the most critical sectors, electronics industry has not been able to
develop in Pakistan. There is hardly any indigenization and most activity is based
around assembly, or simply retailing. Since electronics are now part of most industries,
including textile machinery, auto, home appliances, computers, mobile phones, etc., it is
imperative that Pakistan take significant strides in this industry. Given our
infrastructure, however, we believe that it would be very difficult for us to go directly
into manufacturing of components. It would be more prudent for Pakistan to develop
capabilities in system design instead. This is the model India has followed. System
design involves innovation in architecture and function. Using mostly off-the-shelf
components, entrepreneurs innovate to create new designs. In due course of time,
manufacturing of Printed Circuit Boards and other components follows.
We propose the following in this regard:
xxvii

• New courses in systems design must be initiated across engineering universities
around the country. These course need to be designed jointly by faculty and
industry experts.
• In order to kick start this process, the government should initiate a series of
workshops open to related entrepreneurs, faculty and business people. Foreign
experts should be invited to conduct these hands-on design workshops.
• The government should also identify other mechanisms to transfer design skills
to Pakistan.
• The government should assist electronics firms in purchasing undervalued
international firms that have R&D and branding assets and capabilities.
• Where there is demand of a particular electronics item, e.g., energy savers, the
government should first look to local parties. If a part can be manufactured at
home, local vendors should be preferred.
• As a matter of priority, the government should facilitate anyone who wishes to
initiate manufacturing of Printed Circuit Boards in the country.
• EME College should be given a grant to establish an internationally accredited
design, testing and certification institute.
• The government will also match any investment geared at acquisition of modern
and emerging electronic technologies.
Pharmaceutical
The size of the local pharmaceutical market is around $ 1.5 billion. The local
manufactures comprise 60% of this share and the multinationals account for the
remaining. Over the recent years, the local industry has experienced impressive growth
(double digit) as compared to multinationals in the local market. This said the potential
in this market is huge. The market size can easily increase to US$2.5 billion in Pakistan
as incomes increase and the health expenditure as percentage of GDP increases
(currently 1% - US$7/8 per capita spending). The industry is currently struggling due to
xxviii

lack of chemical industry in the country, poor governance, lack of compliance with
standards and electricity shortages. In addition, the industry also complains that strict
regulatory control on prices of about 900 active ingredients creates distortion in the
market and impedes effective supply to end consumers.
For further growth of the sector the following policies are recommended:
• Prices should not be regulated for drugs whose markets are competitive or
monopolistically competitive. Price regulation should only be there in the case of
a monopoly or collusive oligopolistic behaviour.
• Allow duty free imports of all APIs (active pharmaceuticals ingredients) and
machinery for both domestic and export markets – However, this benefit should
be lifted as the domestic industry picks up and protection should be provided for
WHO pre-qualified APIs manufactured locally.
• All locally purchased items like packaging material, etc. should be exempted from
sales tax.
• The government should share mark-up cost on loans and allow investment
adjustment against future taxes for putting up international standard
manufacturing facilities targeting the pharmaceutical markets of the US, UK,
Australia, Japan and also those in developing countries.
• Provide incentives such as tax breaks for carrying out R&D on molecules and all
ancillary activities like setting up CROs (Contract / Clinical Research
Organisations) subject to meeting certain targets.
• Incentives should be provided for new start-ups in Bio-technology.
• The government should provide tax breaks for setting-up internationally certified
bioequivalence / bioavailability labs, local manufacturing / fabrication of
machines and other hardware.
xxix

• Allow long-term subcontracting with drug manufacturers as opposed to 2 years
as per the current rules.
• Ensure internationally acceptable manufacturing quality standards / (Good
Manufacturing Practises) CGMP compliance in both the local and multinational
manufacturing companies in the country
• Ensure continuous power supply to the pharmaceutical manufacturing to ensure
international GMP compliance of maintaining the required levels of temperature
control. This will not only ensure high quality of the locally produced medicines
but will also make the industry competitive internationally.
Fan Industry
Gujrat and Gujranwala, two important SME clusters, are home to 98% of the country’s
fan industry. The cluster houses around 500 SMEs, of which 300 are based in Gujrat
and the rest in Gujranwala. The sector employs over fifteen thousand workers. However,
employment is seasonal as most factories operate only for five to six months during the
year. The industry produces on average 8 million fans a year with an estimated value of
Rs 17 billion. Over the last four years exports from the sector have increased by about
120% to around US$30 million. This number is expected to rise further. The sector
currently faces several issues such as low levels of productivity due to small scale of
enterprises and lack of mechanisation as a consequence of years of little or no
investment in the sector. There is difficulty in getting adequately trained manpower due
to seasonal demand for fans and absence of training programs. Lack of R&D and design
innovation and the inability to meet changed market demand, e.g., for energy efficient
fans, has greatly impeded the extent of current penetration in high income markets.
Finally, the inconsistent quality has made it difficult for the sector to comply with
international quality standards, resulting in an inability to export to high-income
markets.
To address the issues highlighted above the government should:
xxx

• Create awareness and facilitate common branding and consortium-building of
companies to benefit from scale economies. National Productivity Organisation
(NPO) will support the cluster actors to develop standardised parts and move
towards common production and sourcing.
• Strongly advocate to State Bank to ensure effective implementation of initiatives
provided for SME financing at reduced mark ups.
• Strongly advocate with State Bank and commercial banks to ensure availability
and utilisation of long term project financing and technology upgradation. Share
mark up costs for technology upgradation.
• Provide funding to Fan Development Institute to upgrade machinery at the
common facility centre.
• The government should assist fan manufacturers in purchasing undervalued
international firms that have R&D and branding assets and capabilities.
• Establish training facility by supporting TEVTA and FDI under PPP
arrangements to provide labour training on modern machines.
• Establish a centre of fan engineering and designing at the University of Gujrat.
• Work with Pakistan Electric Fan Manufacturing Association (PEFMA) to develop
domestic standards for fan manufacturing. The standards will be enforced and
monitored by PSQCA.
• Share costs to meet compliance with international requirements such as CE
Marking, UL Marking, etc.
• The government should facilitate joint ventures with foreign companies and
ensure technology transfer.
Cutlery, Utensils and Hunting Equipment
Wazirabad is one of the oldest SME clusters producing fine-quality handmade swords,
knives, cutlery, utensils and hunting equipment. The other cluster is based around Dir
xxxi

in KP. The total production of the sector is around 6.5 million pieces. The sector is
suffering heavily from the increased cost of production due to non-availability of
electricity. An average-sized factory obtaining electricity from WAPDA will consume
electricity worth approximately Rs100,000/month. However, the reliance on self
generation can as much as triple the monthly cost. The products are also losing
international market share due to lack of research and development. There has not only
been a continuous decline in the level of skills of workers but also a gradual fall in the
number of workers entering the sector. There is an absence of a Cutlery/Sword specific
training centre in the cluster. Although the products are of relatively high quality they
are unable to fetch high prices due to a lack of branding and marketing. Finally, lack of
investment has caused the industry to stagnate and has left it with outdated production
processes and lower productivity. The government realising the potential of this sector
has already established a national representative body namely the Pakistan Hunting
Arms Development Company (PHADC).
To enhance the competitiveness of the sector the government should:
• Continue to support the initiatives of the PHADC and ensure that the initiatives
carried out by PHADC are balanced in regional representation.
• Regularly monitor the performance of the PHADC against established targets and
goals to ensure that the company delivers to the needs of the sector.
• Finance joint projects between University of Gujrat and Pakistan Cutlery
Association to work on research and development of newer metals required for
the industry.
• Advocate with provincial government to ensure that TEVTA’s service centre in
Wazirabad also offers skill training specifically for the cutlery and swords
industry.
• Share mark up costs of technology upgradation conditional on companies
meeting set export requirements.
xxxii

• Provide support to PSQCA to work with the sector associations to develop a set of
core standards, especially relating to the use of materials and their quality.
• The government should assist firms from this sector in purchasing undervalued
international firms that have R&D and branding assets and capabilities.
• Advocate with State Bank and commercial banks to ensure availability and
utilisation of long term financing.
• Advocate with TEVTA Punjab to fully operationalise their cutlery common facility
centre.
• Engage Pakistan International Freight Forwarders Association (PIFFA) to run
export training programmes in Wazirabad and work with the sector association
to run training programmes on business processes, business English, etc.
Horticulture­Processing
Pakistan is endowed with a diverse range of fruits and vegetables. Citrus, mango and
dates from Punjab, mango and dates from Sindh, cherries, strawberries, peaches, plums
and apricots from the Khyber Pakhtunkhwa, and, dates, apricots, peaches and plums
from Balochistan, have a large domestic market and also a growing export market. The
country produces 2.2 million tonnes of citrus, 1.7 million tonnes of mango, 0.7 million
tonnes of dates and around 0.6 million tonnes of apricots, peaches and plums. The agroprocessing
industry is affected by seasonal products and fruit quality. The industry shies
away from value added business in horticulture due to coarse market links between the
growers/fruit marketers and industry. Due to product seasonality the price of the fruit
varies significantly and so does the quality. Owing to the risk of non- availability of raw
fruit (fresh or frozen), the value addition industry feels hampered in making
investments. In addition, even after several efforts made by the government and the
private sector, quality and compliance infrastructure is still inadequate. Furthermore, a
significant amount of fruit in Balochistan, Sindh and KP is going to waste due to adverse
conditions with processing plants lying idle and capital stock depleting.
In order to assist the sector the government should:
xxxiii

• Keep on supporting PHDEC to improve awareness on quality and compliance
(establish PAKGAP Standards) and up-gradation of cool chain infrastructure and
also to advocate with the provincial governments to strengthen their strategic
support to upgrade the horticulture sector in meeting quality, compliance and
certification requirements. The provincial government departments responsible
for markets are to be requested to further facilitate linkage between the valueadded
sector and fresh fruit markets.
• Incentivise local manufacturing of juicing and pulping equipment by providing
technical information and support from research centres such as Ayub Reserach
Institute, Faisalabad.
• Strengthen PSQCA’s enforcement capacity to ban / fine all illegal / counterfeit
juice suppliers in the market.
• Provide special time bound incentives to initiate a support plan to restart
business activity in war stricken areas. Special horticulture processing and value
added zones should be developed in interior Sindh, Balochistan, KP and
Southern Punjab. Land should be provided on attractive terms. The zones should
be declared exempt from all taxes and duties, however a target employment for
each zone must be set to qualify for this exemption.
• Develop seven clusters in the agro-sector including dairy cluster in Sahiwal, Pak
Pattan and Okara; Citrus cluster in Sargodha; Mango cluster in Multan and
Mirpur Khas; exotic fruit cluster in Swat; dates cluster in Sukkur/Khairpur and
exotic fruit and apricot drying cluster in Gilgit.
• The government should assist horticulture firms in purchasing undervalued
international firms that have branding assets and capabilities.
Surgical Instruments Sector
Surgical instrument sector is one of the critical export clusters based predominantly in
Sialkot. There are over 2,400 units involved in production of surgical instruments. The
major impediments faced by the sector include declining levels of skill and workforce,
xxxiv

limited designing and innovation capacity, extremely limited branding in international
markets and lack of proper testing and certification infrastructure.
In order to improve the growth performance of the surgical sector the government
should:
• Establish a Surgical training school in Sialkot. This will train workers not only in
the basic skills but will create an educated workforce which understands the
nature of the product. This is imperative given the fact that as quality standards
and compliance regimes get tougher worldwide the workforce will have to
understand the requirements of these standards.
• Upgrade / Establish laboratory in Sialkot to provide surgical instrument specific
testing and conformity assessment certifications. Under new WHO regulations
the industry will soon have to make a change in their polishing and cleaning
techniques. The industry will have to shift from chemical to water based cleaning
and polishing systems. This shift will increase costs and also change the testing
and certification requirements. A laboratory is required that can test for
sterilisation, chemical presence, etc. The government working with the sector will
either upgrade the Surgical Association Laboratory or will extend more support
to the laboratory being established by the Punjab Government.
• Facilitate via foreign offices for local companies to acquire international brands.
• Support development of steel industry which is the primary input into surgical
industry.
• Establish an excellence centre for instrument design and innovation in medical
device industry.
• The government should assist surgical manufacturers in purchasing undervalued
international brands in this sector.
• Launch a sector competitiveness project executed by NPO.
xxxv

Sports Goods
Sports Goods industry is another historic cluster based around the city of Sialkot. The
sector, although extremely critical for Pakistan’s exports, is characterised by low capital
per worker and a poor ratio of skilled to unskilled workers. The key weakness of this
sector is chronically insufficient investment, which makes it difficult for the industry to
adapt technologically to keep pace with changes in the international demand for sports
goods. The largest single export item from Sialkot is inflatable soccer balls. Over the last
two years, Pakistan's exports of this item have dropped from $226 million to $160
million, while world exports have increased from $984 million to $1.15 billion. The
reason for this is a shift in technology, as the world's demand for soccer balls has moved
to thermo-bonded and machine-stitched balls. If this trend continues, the one-time
prime export of Sialkot (soccer balls) faces extinction. This, in fact, was the fate that
befell the rackets industry. Other issues include inadequate availability of newer
materials, deficiency in product development and design, testing and certification
including social compliance and inadequate work force development. The government
realising the potential of this sector has already funded a competitiveness exercise for
the sector.
The government should:
• Continue to support the ongoing competitiveness initiative being managed by
SMEDA and agree to fund the initiatives that will be identified as a result of this
exercise.
• Establish a sports good sector company with broad base representation of the
private sector. This company will implement the sector competitiveness strategy
developed with support from SMEDA.
• Operationalise the Sports Industries Development Centre being established in
Sialkot.
• Support development of a material bank in partnership with the private sector to
facilitate development of newer products.
xxxvi

• Facilitate establishment of strong linkages of the sector with NED Karachi, Gujrat
University and Loughborough University to work on product and design
development.
• Assist sporting goods manufacturers in purchasing undervalued international
firms that have R&D and branding assets and capabilities.
• Advocate with the provincial government to complete its Sports Goods testing
laboratory project on a priority basis.
• Provide targeted tax benefits for companies diversifying into more competitive
products such as golf equipment. The credits will be conditional development and
exports of these products.
Ceramics
The ceramics industry which is primarily located in Gujranwala and Gujrat has made
significant progress over the last few years. The industry has been able to penetrate
export markets and also supply to an increasing local demand. The critical issue faced by
the industry at the moment is the uncertain electricity load shedding which severely
impacts productivity. The heating kilns in the industry need continuous electricity for
maximum efficiency. Moreover, the sector lacks availability of skilled workers and
formal training facilities. The export market growth is also hampered by limited design
and product development capacity. Finally, the sector has limited marketing and
branding capacity and weak links in export markets.
To invigorate the potential of this sector the government should:
• Strongly recommend that the Ceramics Industry be provided continuous
electricity for 10 months during the year in lieu of complete shut down for two
months
xxxvii

• Ensure that the ceramics centre established in Gujranwala focuses on product
design and development and offers training courses specifically for the ceramics
sector.
• Support Identification of international buyer clusters of ceramics and facilitate
international market access through TDAP.
• Assist ceramics firms in purchasing international firms that have marketing and
branding assets and capabilities.
Furniture
Furniture is a critical industry serving both the domestic and international consumers.
The industry is mostly located around Lahore, Karachi, Peshawar, Chiniot and Gujrat.
The government realising the importance of the sector conducted a detailed
competitiveness analysis of the sector. Issues identified during this exercise included: (i)
shortage of Sheesham and other wood resources; (ii) low productivity and low value
added; (iii) issues with quality of the wood processing especially drying of wood and; (iv)
inadequate availability of furniture accessories. To address these issues the government
established a Section 42 company “Furniture Pakistan”. This company has now
established its office is Lahore and is working on the implementation of strategic
recommendations developed by the sector.
The government should:
• Keep supporting the ‘Furniture Pakistan’ initiative and provide funding for
implementation of strategic actions developed by the furniture sector.
• On priority basis ensure provision of solar kilns for wood drying in Karachi,
Chiniot, Peshawar, Lahore and Gujrat.
• Ensure early completion of common facility centers for product development and
training facilities in Karachi, Chiniot and Peshawar.
• Allow duty free import of low cost woods and alternative material for furniture
manufacturing.
xxxviii

• Support establishment and implementation of domestic quality standards for
furniture.
• Assist furniture firms in purchasing international firms that have marketing and
branding assets and capabilities.
Leather & Leather Products
The leather sector of Pakistan is one of the oldest sectors with an established share in
world market for tanned leather, leather garments and leather gloves. In 2008-09 the
industry exports exceeded the US$ 1 billion mark. Realising the importance of this
sector, the government initiated a leather sector competitiveness project to address the
issues of worker training, product development, quality compliance, environmental
compliance, marketing and branding. A detailed strategy was developed by taking input
from all four sub-sectors of the leather industry: (i) tanning; (ii) garments; (iii) footwear
& accessories; and (iv) gloves.
The government should:
• Establish the Pakistan Leather Development Council (PLDC). This should be a
section 42 company and should act as the representative body of the entire
sector.
• Ensure appropriate funding is provided for the implementation of the PC-1
developed for implementation of strategic initiatives under PLDC.
• Work with University of Veterinary and Agricultural Sciences to improve quality
and quantity of livestock.
• Incentivise establishment of modern abattoirs and allow duty free import of
flaying machines.
• Reduce duty on Bulk Chemicals to 5%.
xxxix

• Allow free import at 0–5% for 3 years to acquire tanning plants and machinery
from international locations where tanning industry is closing due to higher costs
or environmental compliance.
• Provide matching grant for establishing Effluent treatment plants.
• Take immediate steps to stop smuggling of live animals.
• Establish design centres and glove development institute.
• Assist in creating linkages with international design centres and also incentivise
development of local centres of design excellence.
• Develop a footwear development institute and link it with design centre at
Pakistan Institute of Fashion Designs.
• Promote Charsadda Chappal by registering it as Geographical Indicator.
• Ensure that announced tanning zones in Karachi and Sialkot are operationalised.
• Ensure footwear parks are established at Muridke and Charsadda. It will be
ensured that the land is available at viable rates.
• Assist leather apparel manufacturers in purchasing international firms that have
marketing and branding assets and capabilities.
Gems & Jewellery:
Despite its abundant reserves of precious and semi-precious gemstones and rich history
of jewellery manufacturing, Pakistan has been unable to develop an internationally
competitive gems and jewellery industry. Pakistan can capitalize on its vast natural
resources, low labour costs, skilled craftsmen and a growing national and international
demand. Seeing this opportunity, the Government initiated a sector level
competitiveness project to upgrade the gems and jewellery sector of Pakistan. As a result
of this initiative the government established the Pakistan Gems and Jewellery
Development Company.
xl

The government should:
• Continue to support the initiatives launched by the Pakistan Gems and Jewellery
Company. Funding should also be provided for agreed projects.
• Regularly monitor the performance of the company against established targets
and goals to ensure that the company delivers to the needs of the sector.
• Assist local brands in internationalization.
Marble & Granite
Marble and Granite industry of Pakistan is still in its infancy stage. The industry
suffered heavily in the past due to poor blasting techniques and outdated quarrying
technologies of marble and stone extraction. The blasting techniques of extraction
wasted around 85% of the raw marble and also severely damaged the quality of marble.
Realising the potential of the industry the Government initiated a marble sector
competitiveness project. The work done under the project identified the poor quality
and techniques of quarrying as the major impediment for the development of the sector.
In order to assist the sector move up the value chain the government established the
Pakistan Stone Development Company (PSDC) with the short term goal of upgrading
and establishing new quarries for extraction of square blocks. PSDC has turned out to be
one of the success stories of government intervention and now square blocks are being
extracted which has improved the quality, productivity and value of the marble.
The government should:
• Keep on supporting PSDC and ensure funding is provided for all planned
strategic interventions.
• Establish seven more model quarries in identified areas of Balochistan and KP.
• Support up-gradation of existing quarries.
• Enhance the common machinery pool for extraction of square blocks.
xli

• Operationlaise the planned marble cities at Risalpur and Karachi and provide
requisite infrastructure in partnership with the provincial governments.
• Fund establishment of common facility and training centres at Gadani, Risalpur,
FATA and Karachi.
• Regularly monitor the performance of the company against established targets
and goals to ensure the company delivers to the needs of the sector.
• Support the internationalization of local brands by purchasing Italian firms that
have marketing and branding assets and capabilities and/or assisting with the
development of local brands.
Agriculture Implements
The agriculture implements (excluding tractors and tractor parts) industry of Pakistan
consists of a large number of micro and small scale manufacturers throughout the
country. Most of these entities operate out of their back yards and small workshops with
outsourcing for components to other small scale operators. The manufacturers are
clustered in and around Daska, Faisalabad, Okara and Mian Channu. Due to years of
operating in the informal sector and lack of investment has resulted in outdated
production assets and technology. The workforce is largely without any formal training,
however, they are rich in innate abilities and raw skills. As peculiar in the informal
sector, the management skills are weak. The manufacturing processes are mostly based
on reverse engineering and hit and trial methods. This limits the design and engineering
capacity and also results in high wastage. The production is not standardized which on
one hand offers the benefit of flexible production but on the other hand limits the
advantages of scale economies. Finance is also a critical impediment impeding growth of
the sector. Against all these odds, the industry has done well in meeting a large local
demand and in accessing international markets. The equipment manufactured in
Pakistan is currently being exported to Afghanistan and in small quantities to Africa.
To ignite the potential of the sector the government should:
xlii

• Establish a design & innovation and training centre in Daska to provide technical
skills and to help the sector move towards product standardization. The centre
should also be provided with funding to develop a common facility centre.
• SMEDA and PSQCA should facilitate businesses putting up new production lines
to shift towards standardized common parts on large scale required by the
agriculture implements industry.
• Ensure that NPO conducts a productivity benchmarking for the sector and impart
training on lean production methods.
• Support equitable business partnerships through the design & innovation
enabling the developer of a new product to partner with investors for putting up
large commercial units.
• Assist leading firms in this sector with development of domestic and
international brands.
• Establish and strengthen a national association for agriculture implements and
help them establish national sales centres especially in remote areas.
• Advocate with the State Bank of Pakistan to design both short term and long term
credit schemes based on the requirements of the sector.
Home Appliances
The domestic home appliance engineering industry consists of medium to large size
units in organized sector with enough managerial capacity. The main products include
refrigerators, freezers, air conditioners and washing machines. Most of the
manufacturing is done through assembly relying on imported CKD/SKD kits. The
critical issue facing the sector is the inability to add greater value during assembly.
To help the sector add more value the government should:
• Establish CAD-CAM designing facility to develop new product designs, improve
quality and bring standardization in production parts.
xliii

• Support provision of market information and linkages with international design
houses to assist the sector add value through design innovation.
• Explore possibility of increasing the potential of intra-industry trade within
South Asia region.
• Share cost of compliance with international standards such as ISO, CE, UL and
other such marks conditional on companies meeting their export targets.
• Ensure that NPO conducts a productivity benchmarking for the sector and impart
training on lean production methods.
• Assist home appliance manufacturers in purchasing international firms that have
R&D, marketing and branding assets and capabilities.
Iron & Steel Pipes and Tubes
Iron & Steel pipe and tube industry in Pakistan produces a large range of products from
cast iron, galvanised iron pipes, cold rolled tubes, square Cathode Ray tubes,
rectangular CR tubes, elliptical CR tubes, API line pipes, black pipes, structural pipes
and spiral welded pipes. In seamless tube category the industry produces heat exchange
tubes, pressure tubes, boiler tubes, low temperature service tubes, precision shaft tubes,
linked tubes, etc. as per international standards. The sector is sharply fragmented with a
few large scale corporate units having sufficient managerial, technological and financial
capability and several medium and small scale manufacturers. The industry needs to
expand its production capacities to produce enough exportable surpluses for growth in
exports. This will require small and medium scale firms upgrading their product range,
increasing their scale and becoming compliant with international quality standards.
Moreover, the current strength of the industry is in producing large diameter pipes,
which attract prohibitive cost of freights and cannot be exported to far distance markets
as they cannot remain price competitive. The government will have to facilitate the
sector in moving into innovative products that are more competitive globally.
The government should:
xliv

• Share the mark up costs of small scale producers expanding their production
facilities to enhance their product mix.
• Provide technical support to companies through PSQCA in improving quality,
standardisation and compliance requirements.
• Provide support to PSQCA to establish domestic standards of quality and
standardize product range.
• Improve the logistical facilities and advocate for cost efficient freight services on
long routes.
• Establish a product design and development centre to facilitate innovation in the
sector. This centre will also provide common facilities for manufacturers.
• Ensure that NPO conducts a productivity benchmarking for the sector and impart
training on lean production methods.
Pumps
The pump manufacturing is dominated by an unorganized SME sector. The industry is
concentrated and clustered in and around Karachi, Lahore, Gujranwala and Faisalabad.
There are about 70 registered and relatively organized units. The number of
unregistered and unorganized units is estimated to be over 500. This number also
includes indirectly involved businesses of vendors supplying to manufacturers operating
at medium, small and micro level. The industry has been experiencing falling
competitiveness in spite of the rise in product prices. This is attributed to be rising
production costs, high prices of raw materials and a cut throat competition through
price cutting in the domestic market. The other issues impeding the growth of the sector
include unskilled labour, outdated technology and lack of coordination with support
agencies. Lack of standardization and quality control has also halted the growth
potential of the sector, especially in the export markets. Limited investments have
generally resulted in low scales of production, a weak product mix and a general
dependence on obsolete designs and quality.
xlv

In order to improve the competitiveness of the sector the government should:
• Develop local quality and product standards to shift production towards better
quality and standardized products. This will be done in consultation with the
sector.
• Share cost of complying with established quality standards.
• Establish a centre of excellence in Gujranwala to work on design development,
quality improvement and standardization of products.
• Ensure that NPO conducts a productivity benchmarking for the sector and impart
training on lean production methods.
Electrical Fittings
The electrical fittings manufacturers are concentrated in Sargodha (70%), Lahore (20%)
and Karachi (10%). There are around 20 small scale units, around 200 small workshops
and around 1,000 manufacturers in cottage sector. The total estimated monthly
production is nearly 10 million pieces of various fittings per month. The sector faces
acute competition from import of similar goods from China. The critical impediment to
growth in the sector is its inability to meet the quality and standards requirements. The
quality of these fittings is extremely poor and in most cases the durability is not beyond
few weeks. In addition, the design and shapes vary significantly making most fittings not
suitable for use with majority of electric equipment.
To improve product quality in the sector the government should:
• Establish a product design centre and testing facility in Sargodha that should also
be responsible for defining national standards on quality, design, shape and
durability. The design centre should over time refine domestic standards to
become in line with international standards such as CE mark.
• Ensure strict compliance with these domestic standards and ban sale of noncompliant
goods in the domestic market.
xlvi

• SMEDA and PSQCA should support companies upgrading their products to
comply with defined standards.
Steel Structures/Towers
Pakistan steel structures industry produces towers for electricity transmission and
towers to support telecom operations (fixed line as well as cellular). A number of
companies are manufacturing standardized towers and other structures for this
purpose. There are around 10 units operating in the organized sector. They manufacture
steel structures using design drawings and use the required quality of steel as per
specifications. Exporting steel towers to African countries offer significant potential.
However, steel towers cannot be exported in assembled form and need to be assembled
on site. Therefore, the prospective exporters need to establish a business of installation
of structures in the country of import or establish partnership with an installation
company which can market their products as well as undertake installation. Unless, this
relationship is established, export may not become possible.
The government should:
• Share cost and facilitate through the foreign office the companies establishing
installation facilities in export markets and should facilitate/coordinate these
efforts through foreign office support.
• Conduct detailed market studies to provide a marketing strategy for the sector.
Prefabricated Buildings
Prefabricated buildings industry has recently picked up momentum in Pakistan. These
buildings are increasingly being used as grain storage silos, sheds, industrial
applications, commercial activities & warehousing and hospitals/ schools/ emergency
housing in disaster hit areas. It is a critical support sector to develop low cost industrial
sheds and housing. The critical challenge faced by the sector is lack of awareness in
using these prefabricated buildings in the market.
To further build on the recent success the government should:
xlvii

• Advocate the use of pre-fabricated buildings in all government schemes of low
cost housing and infrastructure projects where applicable.
Fisheries
The Pakistan fish and seafood industry is worth $1.2 billion with exports of fish and fish
products from Pakistan reaching $213 million per annum. There are around 800,000
people relying for their livelihood on the sector. The seafood industry is currently facing
two critical issues (i) depleting fish stock and (ii) poor quality control and inadequate
hygiene measure which result in the value of the catch not being maximized and also
results in high levels of wastage.
In order to facilitate the sector the government should:
• Create awareness on SPS, health and safety standards that are required to comply
with international requirements.
• Invest in expansion of the existing laboratory infrastructure to provide testing
and certification facilities acceptable to destination markets
• Provide financial support to revamp fish jetties at Gadani, Dam, Pasni and
Jewani.
• Ensure provision of facilities such as landing stations/jetties/port along Makran
coast in partnership with the provincial government of Balochistan.
• Support Korangi Seafood processing companies in developing traceability system
for smoked, canned, fresh and frozen fish products.
xlviii

1 Introduction
The process of structural change has been a central feature in the economic growth and
development of both the Western Economies and the Newly Industrialized Countries of
the East. As an economy develops the share of agriculture in GDP inevitably declines
while that of manufactures and services increases. In other words, structural change is a
gradual shift from low productivity to high productivity activities. Along with this
observable structural transformation there is a large body of empirical literature which
suggests that there is a “U” shaped relationship between a country’s income level and its
degree of product specialization or sectoral concentration [see, Klinger and Lederman
(2004)]. At low income levels specialization is high and is primarily determined by
resource based comparative advantage. As the country becomes richer the
manufacturing base diversifies with firms producing and exporting a wider range of
relatively more sophisticated products. However, at higher levels of income, the process
reverses; specialization again increases but in high value added and technologically
advanced products. Therefore increased product diversification is an intermediate stage
in the process of structural transformation and economic development of a country.
In the context of the Pakistan economy this structural transformation has been skewed
in favour of the services sector. Since the 1970s the growth in services has outstripped
that in agriculture and industry resulting in its current 50% share in GDP. Agriculture
and industry over the past decade have contributed around 25% each to GDP with the
share of agriculture declining over time and that of industry remaining fairly stagnant.
In fact, between 2008-09, in the wake of both internal security issues and the
increasingly binding energy constraint, output in the manufacturing sector contracted
by 3.3 percent with large-scale manufacturing registering a substantial decrease in
output of 7.7 percent [GoP (2009a)]. Notwithstanding the recent contraction of overall
industrial productivity and output, the industrial structure of the country has not
experienced any significant change in the course of the past thirty years. The
manufacturing base remains rigidly narrow leading to a lack of product and export
49

diversification which has been a major impediment to sustained economic growth and
development of the country.
The short term revival and the long term growth of Pakistan’s economy thus hinges on
both the performance and the structural transformation of its industry. As mentioned
above, in the past year and a half the country has seen a dramatic retardation of
economic activity characterized in particular by a stagnating manufacturing sector. The
current energy crisis has further eroded the competitiveness of manufacturing resulting
in a tremendous loss of income, employment and export revenues. Given the fact that
the potential of growth and development of a country is inextricably linked to the extent
of investment and industrialization the continued dismal performance in industrial
growth in Pakistan does not augur well for the future. Therefore it is imperative to
develop an industrial policy which is implementable and has the ingredients to provide
the much needed impetus to industrial growth and diversification. While focusing on
the revival and restructuring of the industry the strategy would have to be guided by the
over arching objective of achieving efficient, sustainable and equitable development.
Pakistan today has the highest population growth rate5 in the South Asian region with
hordes of unskilled entrants into the labour force every year. These adverse
demographics pose a serious challenge to effective policy making. If the industrial base
of the country does not expand to absorb this surplus labour, the bourgeoning
unemployment in both urban and rural areas is likely to have serious socio-economic
and political ramifications. 6 On the other hand, a growing population has the potential
to become a significant economic asset, if adequate policies are in place to facilitate the
development of a large, healthy and skilled labor force. Therefore a central motivation
and aim of the industrial policy should be to generate widespread employment and raise
income levels across the country, with the longer term aspiration of achieving
convergence in living standards in rural and urban areas. This would subsequently
5 1.8% in 2007-08, [see GoP (2009a)].
6 The manufacturing sector in 2007 -08 absorbed only 13% of the country’s labour force compared to
11.5% in 1999, a meager 1.5% point increase over the course of almost a decade, [GoP (2009a)].
50

educe the incidence of poverty and lessen the widening inter and intra regional income
inequality. An industrial policy, which emphasizes domestic as well as international
linkages, promotes and facilitates entrepreneurial activity, focuses on the development
of small and medium scale industries, and provides an impetus to services, trade,
transport and other ancillary sectors, can achieve the objective of inclusive and broad
based growth.
The formulation of the industrial policy is in accordance with the overall growth and
development priorities of the Government of Pakistan (GOP) as broadly identified by
the “9-point” Plan put forward by the Planning Commission (PC) and the Prime
Minister’s Economic Advisory Council (EAC) [GOP(2009a)]. In fact most of the priority
areas highlighted by the Plan where ‘deep, broad-ranging, and sustained’ policy
intervention is required are in sync with the objectives of the industrial policy being
proposed here. For example, Industrial Competitiveness, Human Capital Development,
Energy, Capital Markets, Public-Private Partnership for Infrastructure and
Institutional/Administrative reform are some of the relevant areas which would be dealt
with in considerable detail in the industrial policy report.
The outcome of the industrial policy project is in the form this comprehensive report.
This particular report, constitutes the analysis on the basis of which the industrial policy
has been derived. Moreover, this report also provides a rigorous and detailed analysis of
the spatial aspects of industrialization from which specific policy briefs are drawn
addressing issues of disparities in infrastructural provision, regional inequalities and
factors influencing spatial distribution and concentration of industries. Therefore this
report provides the analysis and evidence for the policies that have been developed for
industrialization in Pakistan presented as summary form in earlier chapter titled
“Principal Recommendations”.
1.1 Raison d’être for an Industrial Policy
In order to articulate the rationale of government intervention and hence outline the
broader principles on which an industrial policy is to be based it is important to
contextualize this discussion in a historical perspective. Therefore the following section
51

gives a brief synopsis of the competing ideas on industrial policy in the twentieth
century.
1.1.1 Brief Historical Review
In the course of the economic development of the West and the Newly Industrialized
Countries (NICs) of Asia, there is no example of a country following a completely laissez
faire policy with regard to industrial development. The nature and the extent of
involvement and intervention of the state varied across countries, with each having its
particular recipe for industrial development, but in none of these was industrialization
achieved through the unfettered workings of the market. The relative success of the
Asian Tigers and more recently of China and India in sustaining high growth rates has
been on the back of an activist industrial policy. The extent of state intervention in these
economies ranged from a variety of input subsidies; tax exemptions; tariff protection to
direct public sector investments in large scale projects e.g., steel manufacturing plant in
Korea and Japan.
The post 1945 Structuralist conception of development considered capital accumulation
or industrialization as the engine of economic growth and development. For the
Stucturalists an interventionist industrial policy in conjunction with trade protectionism
was imperative for the transformation of an economy from primary low value added
production to high value added manufacturing, i.e., Import Substitution
Industrialization. 7 The logic of state intervention was based on the prevalence of
economies of scale (agglomeration economies) in manufactures and the inherent
coordination failure present in an under developed economy. In the absence of
complimentary investments (backward and forward linkages) and the lack of necessary
industrial infrastructure the costs of entry for a pioneering firm into a sector were
considered exorbitant i.e., coordination failure. Given these structural constraints,
industrial development or breaking away from resource and factor based comparative
advantage was viewed as next to impossible in a free trade environment, giving credence
7 Hans Singer (1950) and Raul Prebisch (1950) showed a secular decline in terms of trade of poor
countries. This finding served as one of the critical arguments for structural change.
52

to the infant industry argument for protection and the big push model of large scale
state planning and intervention for industrial development. 8
State intervention in its various manifestations was not always successful in achieving
efficient industrialization and sustained economic growth. The experience of Latin
America stands in sharp contrast to that of East Asia. Although both regions pursued an
activist industrial and trade policy, the outcomes achieved were significantly different.
In countries like Korea and Taiwan protection in the form of subsidies and tariffs was
afforded to firms with a clear state objective of achieving economies of scale in
production through explicit export requirements, forced mergers and investment
licensing [Chang (2003)]. Latin America on the other hand did not have such a clear
focus in terms of achieving economies of scale or creating incentives for firms to be
outward oriented. In fact in most countries the industrial policy was essentially inward
looking, focusing more on achieving domestic self sufficiency in manufactures rather
then creating future export vents. The outward oriented industrial policy of the East
Asian countries gave an opportunity to the domestic firms to target regional and
international markets allowing economies of scale and efficiency in production. The
inward orientation of Latin America, and, the shift of resources from the larger
agricultural sector, limited and stifled the source of demand for the nascent
manufacturing sector retarding its growth, efficiency and competitiveness, e.g., the
Argentinean car industry [see, Bruton (1998)].
Notwithstanding some notable exceptions, such as the aircraft, steel and shoe industry
of Brazil, generally the industrial transformation of Latin America was not as successful
as in East Asia. According to Rodrik (2004), “the difference between East Asia and Latin
America is not that industrial transformation has been state driven in one and market
driven in the other. It is that industrial policy has not been as concerted and coherent in
Latin America as it has been in East Asia, with the consequence that the transformation
has been less deeply rooted in the former than it is in the latter.”
8 Within the economics profession the proponents of a development strategy led by the state were many
such as Ragnar Nurkse (1953), Arthur Lewis (1955), Paul Baran (1957), Rosentein-Rodan (1943). In fact
with the exception of a few like Albert Hirschman, who viewed development as a more spontaneous –
chaotic process, the consensus was towards state led development strategy.
53

The 1980s brought a major shift in development thinking. The debt crisis and the
ensuing macro economic instability which ravaged Latin America for more than a
decade was blamed on the ‘villainous’ policies of state intervention and trade protection
pursued by the region [Taylor (1996)]. The fact that East Asia remained unscathed from
the vagaries of the debt crisis gave support to the Neo-Liberal view which was cantered
on the concept of market supremacy and trade liberalization. Export led growth was
highlighted as the main factor behind the successful transformation of East Asia while
its activist industrial policy and initial years of trade protection were essentially
overlooked. The alternative put forward was articulated by what came to be known as
the Washington Consensus. Under the neo-liberal development paradigm the role of the
state in industrial development was limited to correcting market failure and to the
provision of public goods. The rampant rent seeking, corruption and inefficiency of the
governments in developing countries brought further scepticism on the effectiveness of
the role of the state in industrial and economic development. Although, market failure
or externalities gave a theoretical justification of state intervention, economists and
policy makers of the mainstream became more circumspect of government intervention
stressing the greater propensity and prevalence of government failure as opposed to
market failure in developing countries.
1.1.2 Towards an Industrial Policy
As is evident from the above historical overview of industrial policy demarcating the
domain of state intervention in an economy has been the source of continued
controversy and debate. However, there is an emerging consensus between the
heterodox and the mainstream economists on the broader role of Industrial Policy. The
importance of structural transformation is realized by all, the differences which remain
are on how much or to what extent can a country deviate from its comparative
advantage. The notion of comparative advantage in neo-classical trade theory is
essentially static; it does not allow the possibility of industrial transformation given
current prices and factor endowments present in an economy. As stated earlier the
central objective of industrial policy is to diversify the industrial base and move into
higher value added activity. While the mainstream economists stress on a ‘step by step’
– gradual transition towards higher value added production in conformity with available
54

technology and factor resources, the heterodox stress the importance of defying
comparative advantage by taking ‘leaps’ in certain areas which are considered strategic
[see, Lin and Chang (2009)].
Given that the fundamental principle of an industrial policy is to facilitate the process of
structural transformation a possible bridge between the two perspectives on Industrial
Policy could be to focus intervention to facilitate both ‘new activity’ into the industrial
sector and also help existing firms upgrade their products and production methods. In
other words the policies should consolidate and strengthen existing comparative
advantage and at the same time harness potential or dynamic comparative advantage
resulting in a diversified and internationally competitive manufacturing and export
base.
The ‘new activity’ entrepreneurs are those who either introduce a new product in the
domestic market or develop a more efficient production technology for an existing
product. In fact the formation of a potential industrial cluster is to a great extent
dependent on the incentive structures present for the first entrant or the pioneer firm in
a new industry. Although the ability of the state to ‘pick winners’ is in any case limited
on account of imperfect information on future returns on new investment opportunities,
however, through strategic policy intervention the state can increase the probability of a
new entrant becoming a ‘winner’[see, Hausmann and Rodrik (2003)]. Such strategic
intervention can lead to the formation of an economically viable and vibrant cluster. The
policies which facilitate the emergence of new clusters, especially in economically
depressed areas, would result in a more equitable and inclusive growth – a broader
objective of the industrial policy.
The facilitation a state can provide the entrepreneur is by addressing both coordination
and market failures which would otherwise impede the germination and growth of the
new product or production technique. Coordination failures might arise due to spatial
impediments such as the absence of necessary infrastructure (road network, electricity
etc.) or the non-existence of critical complimentary investments in the area. Therefore
55

the costs of entry for a pioneer firm would generally be far more than that of later
entrants. In fact these costs would tend to decline with an increase in the number of
firms in the cluster – agglomeration economies. 9 Also, greater competition through
entry of firms in the future would reduce the profits of the pioneer or ‘new activity’ firm,
impeding incentives to ‘enter’ in the first instance and hence jeopardizing the formation
of a potential cluster. This problem might also require a competition policy which could
ensure a certain degree of profits for the pioneering firm so as not to inhibit entry
incentives.
Similarly, a market failure hindering a new activity or investment could be in the credit
market, either because of a lack of collateral or asymmetry in information due to the
absence of a credit history. In either of the two cases (coordination and market failure)
the state can intervene through the provision of necessary infrastructure and by devising
mechanisms such as a venture capital fund which would enhance the prospects of a new
entrant or an innovator to qualify for financing. Furthermore, industrial policy can be
employed to gradually alter the existing factor resources and thus comparative
advantages. For example, investments in education and vocational training would
increase the endowment of skilled labour. This would not only help existing firms move
up the value chain but also encourage new investments which are skilled labour
intensive.
The key lesson derived from East Asia is that industrial policy is not a ‘one off’ but a
process where firms and the government learn about the major impediments as well as
opportunities and ‘engage in strategic coordination’. Therefore the broader method or
approach towards formulating industrial policy should be through the continued
interaction between the private sector and the government. This engagement should be
effectively institutionalized so that it can form the basis of a dynamic, flexible and
evolving process of industrial policy making. The crucial point about this public-private
9 An investment by a firm creates positive externalities (benefits) for other firms. Therefore while there
are diminishing returns to investment at the level of the firm there exists increasing returns at the level of
the industry (cluster). As individual firms do no take into account the positive externality of their
investment decisions they tend to under invest. Hence private investment is socially sub-optimal – thus
creating the rationale for government intervention.
56

engagement is a balanced position which is somewhere between bureaucratic ad hocism
and complete embeddedness of the private sector in industrial policy and planning.
While the former extreme (ad hoc-ism) would increase the likelihood of inappropriate
policy formulation the latter (complete embeddedness) would spawn a state patronized
inefficient industrial sector [see, Rodrik (2004)].
An important constraint in industrial policy making which is relevant today is the more
restrictive ‘policy space’ available to countries. Leaving aside the debate on the
effectiveness of trade policy, developing countries under the WTO world trading system
cannot readily use trade policy as an instrument for industrial development as was done
by most industrialized countries in the past. For example, export subsidies, local content
requirements and quantitative restrictions on imports are all illegal under the WTO.
Moreover, the WTO’s TRIPS Agreement ensures that there is no reverse engineering
and copying of goods which has rendered technology transfer through learning and
imitating difficult. The positive aspect of WTO is the institutional mechanism which
allows developing countries to negotiate for increased ‘policy space’. The failure of
recent rounds of negotiations which attempted to extend multilateral disciplines to
national competition and investment policies is evidence of developing countries
effectively protecting their policy space [Rodrik (2004)].
Finally, given the tight fiscal constraints facing most developing countries the extent and
ability of state intervention through either public sector investments or subsidy
provision is severely limited. In many instances these limitations are part of the
conditionalities imposed by multi lateral agencies. Prudent macroeconomic policy and
an increase in domestic public revenue generation would facilitate a more proactive role
of the government in industrial policy. In the absence of fiscal space, the feasibility issue
of industrial policy is of paramount importance and it underscores the need for
strategizing and prioritizing government intervention.
1.2 Methodology and Structure
The particular methodology adopted in this report is to devise specific programs and
policy interventions from a three tiered analysis of the industrial sector. The first tier is a
57

macro level investigation of industrial performance which is supplemented by a second
tier firm-level analysis of competitiveness. The third tier is a detailed study on the
spatial aspects of industrialization or industrial cluster formation. Therefore, the report
has been divided into three sections in accordance with the employed methodology.
The first section covers the first tier of analysis that is macro level analysis of Pakistan’s
manufacturing sector. This section gives detailed description and analysis of structure
and performance of Pakistan’s industrial sector and investment trends in the country. It
also highlights major impediments to both investments and increased productivity and
product diversity within the manufacturing sector. These constraints range from
macroeconomic instability, energy shortage, infrastructure bottlenecks, factor market
imperfections to governance and security issues.
The analysis in this section is based on the repository of existing evidence on the above
mentioned constraints such as the cost of doing business data collated by World Bank
and other similar studies. From the description and analysis of the major constraints
hampering industrialisation we derive the broader policy prescriptions or horizontal
interventions. These policies are essentially aimed at creating the necessary conditions
for industrialisation in the country.
In section two of the report the first tier macro level analysis is supplemented by a
second tier micro level analysis of firm/sectoral competitiveness. This is to help identify
both specific problems affecting the competitiveness of a particular industry and issues
that cut across all sectors under study. The analysis covers both large and small
manufacturing sectors. In both cases, the competitiveness of each sector is assessed and
policy options proposed to enhance it. This has been done by conducting in depth
consultative sessions with sector representatives, analysing all existing literature on
sector’s in Pakistan and for some sectors through a simple value-chain analysis
supplemented by a broader industry assessment that contextualized value-addition
achieved in a particular chain. The outcome of this analysis is both general as well as
specific policy programs which corroborate and enrich the policies outlined in the macro
level analysis.
58

The third part of the report – tier three- takes a ‘bird’s eye view’ of industry by focusing
on the spatial aspects of industrialization. The analysis in this part provides a
compendium of policy notes based on an in depth investigation of the formation of
industrial clusters and their spatial distribution. The mapping measures reveal the main
determinants of cluster formation and identify the determinants of existing clusters.
This analysis has helped in identifying the degree to which cluster formation is
dependent on ‘historical accidents’, government policy or regional economic
characteristics or comparative advantages. This section has also looked at the impact of
disparate infrastructural provision on regional poverty and income inequality. Hence
the critical nexus between public policy and inter and intra regional inequality has been
drawn through the analysis of industrial cluster formation. Hence the three tiered
approach in this report has resulted in a set of policy programs which are not only
consistent but are founded on a rigorous and comprehensive analysis of the industrial
sector of Pakistan.
2 Industrial Sector of Pakistan: Structure, Performance & Problems
Pakistan’s economy faces a critical period. The growth rate of the country’s population is
the fastest in the region while the growth of its Gross Domestic Product (GDP) has
slowed substantially in the past couple of years. This declining economic performance
received another major setback at the hands of the destruction caused by floods in
August 2010. Moreover, excessive inflation, energy and food shortage, rising insecurity
about life & property and alarming levels of mistrust and cynicism in citizens towards
the state is rapidly weakening the state structure and threatens the stability of the
country. If these trends continue, the country will be faced with diminishing
expectations, a demoralized citizenry, and large numbers of unemployed youths fuelling
extremism not only within Pakistan but outside as well. These trends must be reversed.
A key element for reversing this economic degradation in Pakistan is to achieve rapid
industrialization. Only this can bring the much needed economy-wide structural change
for sustained economic growth and productive job creation to absorb the country’s
enormous young labour force. However, the required industrialization will not happen
59

on its own. The Government of Pakistan will have to take bold steps to create the
confidence in the private sector to rebuild the industrial base of the country.
In this section of the report we present an account of the historic performance and
structural change of the industrial sector in Pakistan, the investment trends and their
link to industrialization and the current depth of cross-cutting problems and issues
faced by the industry. Finally, based on this economy wide diagnostic of the industrial
sector, in this section we also present ‘horizontal’ policy prescriptions and programmes
to be adopted by the Government of Pakistan, to enable industrial growth in the
country.
2.1 Structure & Performance of the Industrial Sector in Pakistan
The main ingredient in the growth and development experience of the West and the
Newly Industrialized Countries (NIC) of the East was the rapid capital accumulation, or
growth, in the manufacturing sector. The increasing significance of industry and a rapid
rise in the proportion of manufacturing output traded internationally has been a central
feature of these economies. The high positive correlation between manufacturing value
added and GDP growth across countries is evident from Figure 2-1 below. 10 In the
developed countries, in tandem with rising per capita income, there has been a gradual
movement of resources from manufacturing towards the services sector, in particular
financial services. However, in Pakistan, over the past three decades the contribution to
GDP of the manufacturing sector has been stagnant. The structural change in recent
years has been skewed primarily towards the services sector. Industry, particularly
manufacturing has hardly shown any dynamism in contributing to GDP growth and
employment in the country. This section reviews the performance of the industrial
sector in detail, highlighting its characteristic and the reasons behind its lack of
competitiveness.
10 This is a scatter plot for 131 developing countries for the period 2000 -2005 to show the relationship
between growth in manufacturing value added (MVA) and the growth in gross domestic product (GDP).
The scatter points represent individual countries. The regression line shows that the higher the growth in
manufacturing, the more rapid is the GDP growth.
60

Figure 2‐1: Association between Manufacturing Value Added (MVA) Growth & GDP
Growth
Source: UNIDO Database
2.1.1 Growth Trends
The average annual GDP growth rate in South Asian countries, in the 1990s, ranged
between 4% in Pakistan to about 5.5% in India [World Bank (2010b)]. The growth rates,
other than that of Pakistan, were considerably higher than the respective countries’
historical rates. The growth rate in Pakistan was fairly volatile in the initial decades
(50s-60s) with a noticeable sharp increase in the late 50s as a consequence of large scale
industrialization in Ayub’s time followed by short periods of negative growth as a result
of the two wars with India. The growth performance showed relative stability in the
1970s with a spike in the early 1980s reaching as high as 9%. However in the 1990s,
GDP growth exhibited a declining trend particularly in the years 1993 and 1997 (see
Figure 2-2). The growth started to recuperate after the nuclear explosions in 1998 but
then again decreased considerably in 2001. From 2001-2006 the growth rates of almost
all South Asian countries including Pakistan rose. The growth in Bangladesh increased
by about one percentage point whereas the growth in India jumped by 2%. Pakistan too
overcame its poor growth performance of the 1990’s. The surge in remittances and
foreign aid during this period led to a predominantly consumption fueled boom
resulting in a growth rate of over 7% in 2006 [Government of Pakistan (2007b)]
Figure 2‐2: Historic GDP growth Rate in Pakistan, 1954‐2006 (percent)
61

Source: IMF, 2008
However, in the past two years due to a number of factors, Pakistan experienced a
dramatic retardation in economic activity. The growth rate fell from 4% in 2008 to 2%
in the fiscal year 2009 (see Figure 2-3 below). During the same time period the growth
rates of other countries in South Asia also fell, owing to global recession, but not to such
low levels as that of Pakistan. For example, India’s GDP growth rate fell to 6.7% in 2008
but has risen since then to 7.2% in 2009. Similarly Bangladesh has maintained its
growth rates at an average of 6.2% [Asian Development Bank (2010)].
Figure 2‐3: GDP Growth Rate in Pakistan, 2001‐09 (percent)
10.0
8.0
6.0
4.0
2.0
0.0
2001 2002 2003 2004 2005 2006 2007 2008 2009
Source: Federal Bureau of Statistics
62

The collapse of Pakistan’s growth in 2008, after the high growth in the period 2002-
2007, is a testament to the fact that without addressing the structural weaknesses of the
economy, GDP growth cannot be sustained on the back of capital inflows and
consumption spending alone. With the exception of 2003-04 the declining
manufacturing production and exports during the period of rapid growth was indicative
of the balance of payments crisis that Pakistan faced in 2008. In 2008-09, output in the
manufacturing sector shrank by 3.3% in face of both the increasingly binding energy
constraint and internal security issues [Asian Development Bank (2010)]. The
performance of the three key sectors over the last ten years and over longer periods is
depicted in Figure 2-4 and 2-5 below.
Figure 2‐4: Sectoral growth rates in Pakistan, 2001–09 (percent)
Source: Government of Pakistan (2009a)
Figure 2‐5: Annualized Growth Rates of Key Sectors in Pakistan for 10 Year Periods, 1970‐2009 (percent)
‐
9
8
7
6
5
4
3
2
1
1970s 1980s 1990s 2000‐09
Total Agriculture Industry Services
Source: Government of Pakistan (2009a) and Bureau of Statistics data on GDP
63

2.1.2 Structural Rigidity & Inadequate Transformation
As mentioned before, structural change has been a fundamental characteristic in the
growth and development of western economies and the NIC’s. Structural change entails
a gradual shift from low productivity to high productivity activities. As an economy
develops, the agriculture share in GDP declines whereas that of manufactures and
services increases. In addition, empirical evidence suggests a ‘U’ shape relationship
between a country’s income level and its degree of product specialization [Klinger and
Daniel (2004), and Cadot, O. et al. (2007)]. Specialization is high at low levels of income
per capita but as the country becomes richer it tends to diversify and produce and export
a wider range of relatively more sophisticated goods. However, as income level further
increases, specialization increases again but now in technologically advanced and high
value added goods. This implies that increased product diversification is a middle stage
in the process of structural change and essential for sustained growth of a country.
Figure 2‐6: Sectoral Shares in GDP of Pakistan, 1970‐2009 (percent)
60
50
40
30
20
10
0
1970s 1980s 1990s 2000‐09
Agriculture Industry Services
Source: Government of Pakistan (2004a) and (2009a), and Bureau of Statistics data on GDP.
In the case of Pakistan (see Figure 2-6 above), the structural change has been tilted
towards the services sector. Pakistan, from being a largely agrarian economy in terms of
contribution to GDP, has become a dominantly services led economy accounting for
more than 50% of the GDP. Manufacturing share, on the other hand, has grown more
slowly. Both agriculture and industry have about 25% share in the GDP with the share of
agriculture falling and the share of industry remaining fairly constant over time11 . This
11 http://www.statpak.gov.pk/depts/fbs/statistics/national_accounts/table13.pdf
64

means that structural transformation in Pakistan has been from agriculture to services,
circumventing to a large extent the manufacturing sector.
Figure 2‐7: Shifts in Sectoral Shares in GDP of Pakistan, 2000‐09 (percent)
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Services Sector Industrial Sector Agriculture
Source: Government of Pakistan (2004a) and (2009a), and Bureau of Statistics data on GDP.
Even though the services sector has become the main growth driver in Pakistan, its
potential contribution to employment is limited as compared to the industrial sector.
Therefore, the agriculture sector continues to provide employment to a significant
proportion of the country’s labor force. The share of employment in Pakistan’s
agriculture sector has declined over time but it still employs more than 40% of the labor
force (Figure 2-8). This shows that a major part of the labor force is stuck in the low
productivity sector. The service sector employs above 30% of the labor force while the
industrial sector employs only 20%. Given the deteriorating performance of the
manufacturing sector in the past years, most of the workers move from the agricultural
sector to the services sector implying that the sector with highest value addition has the
lowest share in terms of employment. This is a matter of concern since movement of
labor from low productivity sectors (agriculture) to relatively high productivity sectors
(such as manufacturing) generates the surplus that spurs growth. Even during the
recent economic growth spurts in Pakistan, non-agricultural formal sectors, particularly
manufacturing failed to generate employment. In fact, both male and female worker
65

share in formal non-agriculture employment fell between 1999-2000 and 2007-08 as
shown in Table 2-1.
Figure 2‐8: Employment Shares by Sector in Pakistan, 1980‐2009 (percent)
Source: Government of Pakistan (2004a) and (2009a), and Bureau of Statistics data on GDP
Table 2‐1: Distribution of workforce by formal/informal sectors and by gender in Pakistan, 1999‐2008 (%)
Male
Female
1999-2000
2007-08
66
1999-2000
2007-08
Agriculture 44.3 36.87 72.93 74.98
Non-agriculture 55.56 63.12 27.14 25.07
Formal 19.01 17.15 9.29 6.86
Informal 36.55 45.97 17.79 18.22
Source: Government of Pakistan (2003) and (2008)
Felipe (2007) provides a comparison of Pakistan with other Asian countries, which
shows that Pakistan, along with India and Philippines, are the only exceptions where
industry has not been the main engine of growth. Whereas the share of the
manufacturing sector in GDP in countries like Malaysia, Indonesia and Thailand has
increased considerably over the past three decades, this share has remained stagnant in
the case of Pakistan. In 2007, the manufacturing output share was only 19% of GDP in
Pakistan compared to the Peoples Republic of China (35%), the Republic of Korea
(24.7%), Indonesia (28%), and Malaysia (29.8%) [Asian Development Bank (2008)].
This lack of structural transformation in Pakistan is indicative of the manufacturing
sector’s unsatisfactory growth performance. The manufacturing sector grew by an

average 10.61% during the period 1998-2007[Raheman, A., et al (2008)].The growth
rate in the manufacturing sector deteriorated from 14.0% in 2004 to 8.2% in 2007 and
further to 5.4% in 2008 [Asian Development Bank (2008)]. This decline in growth can
be attributed to the fact that this sector continues to be heavily concentrated in low
value added consumer products such as food, beverages and textile, as depicted in Table
2-2 below. The industrial sector in Pakistan has failed to move into more sophisticated
products such as capital goods and continues to be dominated by resource based and
low technology activities. The lack of production of capital goods and an absence of
upstream ancillary industries such as chemicals and engineering limits the growth
potential of the industry.
Moreover, the manufacturing sector is stuck in production of items that are losing share
in the world market. For example, in 2008, the share of textiles, garments and footwear
in Pakistan’s exports was more than 60% whereas the share of these items in world
exports was merely 5.8% [United Nations Commodity Trade Statistics Database]. The
manufacturing sector, due to the lack of innovation or improving competitiveness, is
reliant on outdated production techniques and a labor force with low technical skills.
Table 2‐2: Average Product Shares in the Manufacturing Sector of Pakistan, 1970‐90 (percent)
Product 1970s 1980s 1990s
Food & Beverage 30.45 30.94 22.89
Textiles 27.78 18.14 25.06
Apparel, leather &
textile
2.04 2.37 2.80
Industrial
Chemicals
11.20 14.29 15.50
Petroleum & Coal 5.27 6.01 3.26
Rubber & Plastic 1.80 1.80 1.42
Metals & Nonmetals
9.1 14.2 13.2
Nonelectrical
Machinery
1.84 2.14 2.09
Electrical
Machinery
3.31 3.26 5.43
Transport
Equipment
Source: Felipe (2007)
2.99 2.89 3.05
67

Breaking down the analysis of manufacturing performance into large and small scale, it
is apparent that the growth of small scale industries has remained almost constant since
2000 as shown in Figure 2-9 below. The growth of large scale industries on the other
hand has been quite volatile. Large scale manufacturing saw a rapid increase in growth
rate of (from around 4% in 2002 to 20% in 2004-05) which was followed by a sudden
retardation of growth. In 2007-08, the increase in inflation, a bourgeoning fiscal and
trade deficit and an outflow of capital forced Pakistan to resort to the IMF program in
2008. The overall economic instability, the hike in interest rates on account of the
macro stabilization program coupled with the deteriorating security situation and power
crisis contracted the growth of large scale manufacturing significantly.
Figure 2‐9: Performance of the Large Scale and Small Scale Manufacturing Sector in Pakistan, 1950‐2010 (%)
25
20
15
10
5
0
‐5
‐10
1950s1960s1970s1980s1990s 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Large Scale Small & HH Linear (Large Scale)
Source: Government of Pakistan (2007b) and (2010a)
The following figure is a comparison of manufacturing value added across different
countries. During the period 1968 and 2001, MVA in Thailand, Malaysia and Korea
increased by a factor of 14, 27 and 40 respectively [World Bank (2010b)]. In Pakistan,
however, manufacturing value added in the same period increased merely by a factor of
7 [World Bank (2010b)]. Even in more recent years i.e., 2001-2007, there has not been
any marked increase in the manufacturing value added in Pakistan as illustrated in the
figure 2-10 below.
68

Figure 2‐10: Manufacturing Value Added in US$ (Billions), 2001 & 2007
Source: World Bank (2010b)
2.1.3 Productivity
The competitiveness of an economy is closely associated with the productivity of its
industry, particularly at the firm level. Strong growth in productivity is essential for
maintaining export share in an increasingly competitive world market. Pakistan,
however, faces difficulty competing with its competitors owing to either lower factor
productivity or higher worker wages. Pakistan has about the same factor productivity as
Bangladesh but its wages are almost 50% higher [The International Monetary Fund
(2010a)]. On the other hand, wages in Pakistan are similar to China and India but factor
productivity is much lower.
The contribution of factor productivity in the growth process of Pakistan has been
trivial. During the period 1960-2005, about 80% of the GDP growth rate in Pakistan was
explained by capital accumulation and labor expansion whereas total factor productivity
(TFP) contributed the remaining 20% of the overall growth (Table 2-3). However, the
contribution of TFP in GDP growth has fluctuated considerably during this period. TFP
explained almost 38% of GDP growth rate in the 1980s but this fell to merely 18% in the
1990s (Table 2-3). TFP growth then recovered to some extent during 2001-2005 and
contributed to 22% of the GDP growth rate [The International Monetary Fund (2010a)].
69

Moreover, during the 1990s, total factor productivity growth in the manufacturing
sector was 2.4% [Chaudhry (2009)]. And over the period 1998-2007, total factor
productivity in the overall manufacturing sector, increased only by 0.9% [Raheman, A.
et al (2008)]. This implies that the growth of the manufacturing sector has been mainly
driven by growth in inputs i.e., labor and capital and the contribution of total factor
productivity has been fairly low.
Table 2‐3: Growth Accounting in Pakistan by Decades, 1961‐2005 (percent)
GDP Growth Capital Labor Total Factor
Productivity
1961-1970 6.97 4.48 1.63 0.86
1971-1980 4.58 1.80 2.30 0.48
1881-1990 6.09 1.90 1.90 2.30
1991-2000 3.86 1.45 1.71 0.71
2001-2005 4.55 1.58 1.92 1.04
Source: World Bank (2006)
Even though labour expansion is an important source of growth, it does not explain the
variation in growth across countries. Despite almost similar labor expansion in
Pakistan, India, Bangladesh, and East Asia over the period 1961-2005, the countries
experienced different growth rates (Table 2-4). This means that the difference in growth
across countries can be attributed to capital accumulation and productivity
improvements. Available evidence [World Bank (2006)] shows that for the period 2001-
2006, the growth differential between East Asia and South Asia can be explained mainly
by total factor productivity.
Table 2‐4: Growth Accounting in Asia, 1961‐2005 (percent)
GDP
Growth
Capital Labour Total Factor
Productivity
Pakistan 5.28 2.31 1.89 1.08
India 4.57 1.77 1.50 1.30
Bangladesh 3.38 1.16 1.64 0.57
East Asia & Pacific 6.46 3.15 1.74 1.71
Source: World Bank (2006)
Looking at inter firm productivity differentials, the following figure depicts that larger
firms have a higher total factor productivity relative to medium and small sized firms
suggesting the greater accessibility and use of more efficient production techniques.
70

Figure 2‐11: Total Factor Productivity by Firm Size & Firm Age, 2001‐2006 (log values)
2.5
2.0
1.5
1.0
0.5
0.0
Small
(=25 & =100
employees)
Young
(=5 years)
2.1.3.1 Labor Productivity
In Pakistan, overall labor productivity calculated as total value added (GDP) divided by
the total number of workers, increased between 1981 and 2002. 12 However, there is
significant variation in labor productivity over the period across the three sectors as
evident from Figure 2-12 below.
Figure 2‐12: Sectoral Value Added/Worker in Pakistan, 1981‐2002 (Rs at contact prices of 1999‐2000)
Source: Bjornestad, Liv (2008)
In Pakistan, worker productivity in industry was higher than for services in the 1980’s
and the 1990’s but has fallen recently. Worker productivity in agriculture, which was
robust in the past, fell drastically in 2000-06 (Figure 2-13).
12 This only provides an approximate measure as it does not take into account any differences in the
number of hours worked in the two periods
71

Figure 2‐13: Output per Worker Growth in Pakistan, 1980‐2006 (percent)
Source: World Bank (2010b)
During the period 1990-2006, overall labor productivity in Pakistan grew by only 1.29%,
which was quite low compared to labor productivity growth in other developing
countries such as India, Bangladesh, Indonesia, Malaysia and Thailand (Table 2-5).
Table 2‐5: Annual Labor Productivity Growth, 1990‐2006 (percent)
Overall Manufacturing
Pakistan 1.29 2.08
India 3.38 -1.1
Bangladesh 2.92 4.05
Indonesia 2.99 3.09
Korea 3.93 8.76
Sri Lanka 2.84 1.74
Malaysia 2.69 5.11
China 9.07 1.97
Thailand 3.61 2.6
Source: Asian Productivity Organization (2009)
Relative to other countries in the Asian region, marginal improvements in labor
productivity over time have reduced Pakistan's competitiveness to a great extent. Figure
2-14 sourced from the ILO shows that based on output per worker, Pakistan has fallen
considerably behind many other regional economies over the past two decades, despite
a relatively similar starting point.
72

Figure 2‐14: Output per Worker in Asian Economies, 1990‐2004 (assuming 1990 as base of 100)
Source: The Conference Board and Groningen Growth and Development Centre, Total Economy
Database, January 2006
2.1.4 Export Performance
A major change in the global economy has been the relentless increase in the proportion
of internationally traded manufacturing products. The exports of manufactures have
been growing rapidly and many developing countries are gaining a global market share.
Table 2‐6: Country Export Shares Relative to Total World Exports, 1970‐2008 (percent)
1974 1980 1990 2000 2008
Pakistan 0.14 0.15 0.18 0.15 0.15
India 0.56 0.43 0.57 0.70 1.32
Malaysia 0.55 0.74 0.94 1.61 1.43
Thailand 0.32 0.37 0.74 1.13 1.25
Source: United Nations Commodity Trade Statistics Database
Despite such overwhelming manufacturing success in developing countries, many
countries including Pakistan have still not been able to establish a vibrant industrial
sector. During the period 1990-2008, exports of India, Malaysia and Thailand as a share
73

of world exports increased on average by 185% but Pakistan’s share declined by 17%
over the same period (see Table 2-6).
Moreover, Pakistan exports a narrow range of manufactured products that are
essentially low value added. To illustrate the lack of competitiveness in the industrial
sector in Pakistan, two measures of competitiveness for textile products, which
comprises of a significant proportion of total exports, are shown in the table below. The
first measure, PRODY13 , is an index that associates a certain income level to each
product. A high value of PRODY means that the product is produced by high income
countries; the implication is that higher values of PRODY imply more sophisticated
products. The second measure is revealed comparative advantage14 , where a value
greater than unity implies that the country has a comparative advantage in that product.
The values of these two measures shown in the table 2-7 below indicate that despite
being a major producer of raw cotton, Pakistan does not show a comparative advantage
in those products of the
Table 2‐7: Revealed Comparative Advantage and PRODY in Textiles, 2008/09
Product PRODY ($) Revealed Comparative
Advantage (RCA)
Raw Cotton 2,036 5.89
Cotton Yarn 5,631 86.5
Cotton Fabrics 4,541 114.98
Men & Boys Readymade garments 6,777 13.18
Women & Girls Readymade garments 5.16 5.16
13 To construct PRODY, the methodology in Hausmann et al. (2005) and data from the UN Commodity
Trade Statistics Database (COMTRADE) at the 5-Digit level were used. PRODY is constructed as the
weighted average of the per capita GDPs of the countries exporting a specific product, and thus represents
the income (and productivity) level associated with that product. The weights are the revealed
comparative advantage of each country in each product (normalized to one). If most high-income
countries have revealed comparative advantage in the export of a product, PRODY would be high.
14 To construct RCA, the methodology in Balassa (1965) and data from COMTRADE were used. The RCA
is constructed by finding the ratio of the share of product A in Pakistan’s total exports to the share of
product A in total world exports. If RCA is greater than 1, then it implies that the country has a
comparative advantage in that product.
74

Clothing Accessories Knit Wear 9,419 22.51
Articles of Felt (Technical Textile) 22,486 0.1
Bonded Fibre Fabric 31,250 0.18
Source: UN Commodity Trade Statistics, and Asian Development Bank
textile sector that fetch higher prices. The low values of PRODY suggest that Pakistan’s
exports are concentrated in those cotton products (raw cotton, cotton yarn, and cotton
fabrics) that are produced by low-income countries. Apparently Pakistan’s strong
comparative advantage in these products is mainly due to policies of the government
that backed low value-added elements. Government policy has not yet fostered the
industry to move into more sophisticated products, such as articles of felt and bonded
fibre. The low value added and narrow range of exports is therefore a direct
consequence of a stagnant manufacturing base [Department for International
Development (2010)].
Pakistan’s export positioning in the world markets is also not very encouraging. In the
figure below, the vertical axis measures the worldwide growth in the export of product x,
as a share of total world exports, while, the horizontal axis measures the growth in the
export of product x for a country as a share of total world export of product x. If the
country’s product is in the right hand upper quadrant (competitive quadrant), it means
that the product is internationally competitive. Hence, the more products a country has
in the ‘competitive quadrant’ the stronger is its international competitiveness [World
Bank (2010b)].
As evident from the figure 2-15 below, Pakistan has a minute proportion of its exports in
the competitive quadrant, i.e., chemicals at 2.6%. Manufactures exports are dominated
by textiles whose world demand is falling. This clearly illustrates lack of product
diversification. In comparison, Indian manufactured exports are beginning to be more
diversified the exports of India and quite a few products namely pharmaceuticals,
chemicals, iron and steel and automotive parts are located in the ‘competitive quadrant’
[World Bank (2010b)].
75

More so, the product concentration index of Pakistan’s export has historically remained
above 0.4015 as depicted in figure 2-16 below. This index reached its highest value for
Pakistan in the year 1997 after which it has shown a declining trend, and is now at an all
time low value of 0.43 [Ahmad, H et al. (2010)]. This clearly illustrates the need for
focusing on product diversification.
Figure 2‐15: Competitiveness and Performance of Pakistan’s Exports, 2000‐2007 (percent)
Annual growth of product’s share in world’s total exports in 2000‐2007
Positioning of India's Major Manufactured Exports in
2007
10.0
Pharmaceutical
s, 3.1%
8.0
6.0
4.0
2.0
0.0
‐4.0
Clothing, 6.7%
‐6.0
‐8.0
Chemicals,
11.4%
Iron and steel,
5.7%
Textiles, 6.6% Machinery and
transport
equipment,
11.6%
Automotive
products, 2.4%
‐5% 0% 5% 10% 15% 20%
‐2.0
Annual growth of Country’s share in Product’s global exports in
2000‐2007
Source: World Bank (2010b)
Figure 2‐16: Product Concentration Index, 1974‐2008
Annual growth of product’s share in world’s total exports in 2000‐2007
‐20% ‐10% 0% 10% 20% 30% 40% 50%
‐2.0
Telecommunicat
Clothing, 21.4%
‐4.0
Machinery and
transport
equipment,
4.5%
ions equipment,
0.4%
Office and
telecom
Textiles, 41.4%
‐6.0
equipment,
0.5%
76
Positioning of Pakistan's Major Manufactured Exports in
2007
10.0
8.0
Pharmaceuticals
, 0.6% 6.0
4.0
2.0
0.0
Chemicals, 2.6%
‐8.0
Annual growth of Country’s share in Product’s global exports in
2000‐2007
15 21/2
The product concentration level is measured using the index, Gt= ( ∑ W
it
) , where k is the number of
k
products that account for more than 90% of Pakistan’s exports and Wi is the share of commodity i in total
export earnings. The index can take a value between 0 and 1; the closer it is to 1, the greater the degree of
concentration.

Source: Ahmad, H et al. (2010)
Pakistan’s exports are quite concentrated geographically, with the Hirschman
concentration index estimated at 0.2516 in 2008 (Figure 2-17 below). This shows that
Pakistan is doing fairly well in terms of a diversified market for exports. This is also
evident from the fact that 90% of the country’s exports were going to 45 countries in
2008.
Figure 2‐17: Market Diversification Index, 1971‐2008
Source: Ahmad, H et al. (2010)
16 The Hirschman index measures the geographical concentration of exports i.e. it shows the degree to
which a country’s exports are dispersed across different destinations. The Index can take a value between
0 and 1; higher values indicate that exports are concentrated in fewer markets. A value of 1 indicates that
all exports go to a single destination. Hence high concentration levels can be interpreted as an indication
of vulnerability to economic changes in a small number of export markets.
77

The underlying reason for South Asia’s inability to achieve dynamic competitive is the
failure to climb up the technology ladder in production. This has been shown by Sanjaya
Lall (2003) in his analysis of evolving technological intensity in East and South Asian
economies. He groups exports in four technology categories: resource based exports
(industrial raw materials), low technology embodying exports, and medium and high
technology embodying exports. The technological change is assessed for the period
1981-2000 for Pakistan, India, China, Malaysia and Thailand. The increase in the share
of medium and particularly high technology exports has been remarkable in China,
Malaysia and Thailand. On the contrary, Pakistan increased its share only in low
technology exports with almost no change in the share of medium and high technology
exports.
Over the past two decades, the share of low technology manufactured products
increased from 54% to 76% of total exports where as that of medium technology
products increased from 7.8% to 8.4%. However, high technology products remained an
insignificant 0.6% of total exports [Government of Pakistan (2010b)]. This shows that
the technological sophistication in manufacturing in Pakistan continues to be low
resulting in the country’s exports being dominated by low technology manufactures.
In order to improve export competitiveness, a country must upgrade technologies in all
activities, build new capabilities, and find new markets [Lall and Weiss (2004)].
Unfortunately, Pakistan has not been able to move up the technology ladder, i.e., from
low to medium and from medium to high technology exports. Felipe (2007) calculates
EXPY for Pakistan, which is a proxy for a country’s level of export sophistication or an
indicator of its competiveness. The results show that Pakistan’s index in 2004 (4628) is
almost the same as that in 1986 (4,664), which is about the same that Indonesia,
Malaysia, Philippines, and Thailand had in the late 1970s, and well below the levels of
China and India in the 1980s. This indicates lack of export sophistication in Pakistan,
which can be attributed to slow technology adoption and up gradation.
To date, Pakistan has not been able to upgrade its export structure and exports continue
to be dominated by low-technology manufactures whose share is declining in the world
78

market. In fact, in 2008, when the share of medium and high technology products in
world exports was 57.6%, these products accounted for merely 9.5% of Pakistan’s
exports (see Table 2-8).
Table 2‐8: Technological Level of Pakistani and World Exports, 1998‐2008 (percent)
Pakistan Exports World Exports*
Growth Share Share Growth Share
Sector 1998-2008 1998-2000 2006-2008 1998-2008 2008
Total 9.6 10.1
Primary 10.1 12.3 12.7 11.2 11.4
Resource-based 23.9 3.5 10.9 10.5 14.8
Low-technology 8.2 74.7 66.7 9.2 16.2
Medium technology 8.7 8.6 8.1 10.1 35.5
High technology 17.5 0.8 1.4 9.3 22.1
*Exclusive of oil exports
Source: United Nations Commodity Trade Statistics Database
More so, the performance of medium technology exports has not been encouraging with
their growth rate being below both the world growth rate of this category and Pakistan’s
total export’s growth rate. The growth of high technology exports, on the contrary, has
been remarkable at 17.5% per year; however, the share of high technology products in
total exports is negligible to make any significant impact. Moreover, the share of high
technology exports of manufactured goods in Pakistan is abysmally low in comparison
to other countries.
The fact that Pakistan has been unable to upgrade technology in export products is also
illustrated by the figure 2-18 below. The figure shows the elasticity of Pakistan’s export
with respect to the GDP of industrial countries17 since 1985. This elasticity is a proxy for
the non-price competitive (i.e. the quality attribute) of Pakistan’s exports [Felipe and
Lim (2008)]. As evident from the figure, this elasticity is low, taking an average value of
0.87 for the period 1985-2007. This highlights the need for upgrading the export
package of Pakistan.
Figure 2‐18: Elasticity of Pakistan’s Exports to GDP of Industrial Countries, 1985‐2007 (percent)
17 This shows by what percentage Pakistan’s exports change when the GDP of these countries increases by
one percentage point.
79

Source: Felipe and Lim, 2008
2.1.5 Summary Analysis
Starting from a dominantly agricultural economy, the structural transformation of
Pakistan required an increasing share of the industrial and manufacturing sector in the
GDP, which would have accelerated growth by creating more employment opportunities
and higher value of exports. However, the structural transformation in Pakistan has
been from agriculture to services circumventing the manufacturing sector to a large
extent. The current growth pattern in Pakistan, with more than 50% of the growth
coming from the services sector, is not sustainable as the services sector cannot continue
to grow without growth and development in the industrial and manufacturing sector
since the demand for services largely comes from these sectors.
The East Asian experience shows that the export sector has the potential to serve as a
growth engine of an economy. But for that it is essential for the country to move up the
technology ladder i.e. from low to medium and from medium to high technology
exports. This, however, has not been the case for Pakistan. The performance of the
80

manufacturing sector in Pakistan, particularly in terms of positioning of manufactured
exports in the world market has not been very encouraging. This is because Pakistan
continues to produce low value added and less technology intensive goods, whose shares
are decreasing in the world market. The stagnant growth of the manufacturing sector
along with a narrow base of manufactured goods and lack of product diversification has
been an obstacle to sustained economic growth and generation of employment in the
country. Pakistan has the highest population growth rate in South Asian with hordes of
entrants in the labor force every year. If the industrial base of the country does not
expand to accommodate this surplus labor, the mounting unemployment may lead to
grave socio-economic and political consequences.
Thus, for sustained economic growth and employment creation in Pakistan, it is vital to
accelerate growth and development of the industrial sector with particular emphasis on
boosting export competitiveness. Pakistan has had a relatively stable export
composition over the years with only two sectors, textiles and rice, accounting for more
than half of total exports. Furthermore, the share of about 40% of Pakistan’s export
goods particularly textile items, in the world market, is declining. This shows that the
Pakistan’s export basket mainly comprises of products that are not aligned with global
trends. Thus, in order to enhance export competitiveness, there is an urgent need to
diversify manufacturing export base and to shift focus from low to medium and high
value manufacturing. Moreover, the range of markets to which Pakistan exports is
relatively small. So, efforts must be made to target world markets such as Russia, United
Arab Emirates and Saudi Arabia in which Pakistan is currently an underachiever so that
the country can realize the growth potential offered by these dynamic world markets.
The revival of the Pakistan economy hinges on the performance of its industry. Even
though, increases in total factor productivity at the firm level are essential for sustained
growth, factor productivity in the manufacturing sector, increased only by 0.9% during
1998-2007 [Raheman, A. et al. (2008)]. Thus, it is important for Pakistan’s
manufacturing sector to increase total factor productivity, which can be done by
improving both technical efficiency and technological progress. Improvement in
technical efficiency requires quality enhancement of inputs like labor and capital. In
81

order to bring about technological progress, efforts should be made to increase the
research and development activities in the manufacturing sector.
2.2 Investment Trends in Pakistan
There is a vast body of empirical literature which confirms that investment in capital
goods is a necessary condition for economic growth [Levine and Renelt (1992); Durlauf and
Quah (1999)]. The economic miracle of Korea, Taiwan and lately that of China was based
on sustained high rates of capital accumulation which transformed the economic
structure of these countries from agriculture to industry in a relatively short period of
time. High rate of investment not only stimulated domestic economic activity and
increased employment and income in these countries but also allowed them to export
value added manufactured goods to the world [Rodrik (1995)].
Although GDP growth in Pakistan over the last three decades is comparable with other
developing countries, investment in the manufacturing sector has been sluggish.
Consequentially the share of manufacturing in total GDP over the same period has been
stagnant in Pakistan and there has been a gradual erosion of the country’s export
competitiveness. There are various factors which explain low investment in Pakistan
ranging from macroeconomic instability, imperfections in factor markets to
infrastructural constraints which have reduced the perceived return to the private
investor. The factors behind low investment would be discussed in detail in the next
section. Here we focus on describing the key trends and patterns in public and private
fixed investment in Pakistan.
2.2.1 Aggregate Investment Trends
Total fixed investment as a percentage of GDP in Pakistan has remained within the
range of 15% to 20% in the last three decades. Between the years 2003 to 2006 the share
of investment rose from 15% to more than 20% (see Figure 2.19), due to improvement in
the macroeconomic environment. However, in the last three years, the investment trend
reversed and investment’s share in GDP decreased substantially. The worsening internal
security situation and increased macroeconomic instability brought about a sharp fall in
82

private investment pulling down the share of total investment in GDP by almost four to
five percentage points (see Figure 2-19).
Figure 2‐19: Total, Private and Public Gross Fixed Capital Formulation (GFCF) in Pakistan, 1980‐2009 (% of GDP)
Note: Total GFCG includes private sector, public sector and general government.
Source: GoP (various years)
Decomposing total investment into private and public investment, in the early 1980s the
fall in public investment was more than compensated by an increase in private
investment, resulting in rising overall investment rates in the 1980s. Public investment
declined from 9.2% of GDP in 80s to below 5% in 2000s. On the other hand private
investment increased from 7.8% of GDP in the 1980s to 12.7% of GDP in 2000s (see
Figure 2.19).
The fall in public investment came as a result of economic reforms introduced in the late
1980s which were aimed at achieving fiscal consolidation. 18 Given the large and
increasing debt servicing burden and chronic fiscal deficits, it was public investment and
development expenditures which bore the brunt of these expenditure reductions. As is
apparent in the figure below, the fall in public investment coincided with a significant
increase in private investment. Financial reforms which entailed greater autonomy to
the State Bank and privatization of commercial banks gave an impetus to the private
18 As a result of fiscal and external deficits Pakistan sought help in the late 1980s from the World Bank
and IMF on an emergency basis for the revival of economy. The Structural Adjustment Program (SAP)
was introduced to address the twin deficits through reductions in expenditures and increases in tax
revenue along with conditions pertaining to the restructuring of the economy through liberalization,
privatization and deregulation.
83

sector. The role of the government in the economy was further reduced through
privatization of large public sector entities leading to the share of the private sector
increasing to 84% of GDP [ADB (2008)].
It has to be noted that the relationship between private and public sector investment to
a large extent depends on the nature of public investment and does not have to be
necessarily negative. Recent literature has discussed the crowding-in phenomena (as
opposed to the standard crowding out effect) by analyzing public and private investment
in Pakistan. Public investments in infrastructure and utilities (roads, transports, ports,
energy, gas, water and sewerage) enhance investments from the private sector whereas
poor infrastructure increases uncertainty, distorts investment returns and increases the
cost of doing business. In the next subsection of this report, we discuss the various
binding constraints impeding investments in Pakistan.
2.2.2 Regional Comparison and Pakistan’s Investment Gap
In order to allow a comparative analysis of Pakistan with the rest of the region, we
compute the decade wise investment gap: difference between the rate of growth of
investment in Pakistan and other developing countries within and outside the region for
each decade. On average Pakistan’s investment gap in the 1980s was narrower than the
investment gap in the 1990s and 2000s. Figure 2-20 below shows that in 1980s
investment in Pakistan grew on average 1.73 % less than Bangladesh’s investment.
Investment during this period was marginally below India (-0.27) and South Asia (-0.15)
but was 0.81% more than the low middle income countries. The relatively lower
investment gap in 1980s can be attributed to both, the revival of the private sector
through denationalization and the large amount of economic aid given to the country in
this period.
The investment gap was the widest in the 1990s primarily as a consequence of political
and economic instability in the country. However, during the years 2000-08, there was
a marginal improvement in Pakistan’s negative investment gap relative to the gap of
84

1990s. This narrowing of the gap was due to the increase in investment share as
percentage of GDP from 17.4 in 2000 to 22.40 in 2008. 19
Figure 2‐20: Aggregate Investment Gap of Pakistan Relative to other Countries, 1980‐2008 (percent)
Source: Authors’ calculations using World Development Indicators for various years.
In order to explain the sustained negative investment gap of Pakistan we now analyze
the disaggregated trends in investment across the region. The negative investment gap
of Pakistan compared to its regional competitors is explained by consistently low public
as well as private investment over the past three decades.
There is a systematic fall in public sector investment in the South Asian region post
1985-86. This might be attributed to reduction in the size of government or reduced
investment by the public sector in the wake of liberalization reforms adopted across the
region. However, the depth of the reform seems to have been much more in Pakistan as
is evident by the fall in Public investment over this period which is substantially more
than in both India and Bangladesh. The private investment trend in Pakistan also
remained sluggish compared to the sharp increase in private investment in both India
and Bangladesh. During the 1990s, Pakistan’s private investment remained
approximately close to 10% of GDP whereas private investment of both in India and
Bangladesh remained above 10% of GDP and followed an increasing trend (see Table
2.19). The reasons for Pakistan’s relatively stagnant private investment in the 1990s thus
range from political instability, to poor infrastructure coupled with deeper economic
19 GoP (2009).
85

eforms in particular trade liberalization (e.g. reducing trade barriers and duties) which
exposed a vulnerable manufacturing sector to greater international competition.
The short spurt in private investment in Pakistan post 2002 was not enough to bridge
the regional gap. Both public and private investments as percentage of GDP were less
than India and Bangladesh’s investment from 2002 to 2008. Public investment
remained stagnant at 10% of GDP while private investment growth could not be
sustained due to worsening macroeconomic instability, high nominal interest rate, and
low economic growth.
2.2.3 Widening Resource Gap between Investment & Domestic Savings
Pakistan’s share of manufacturing in GDP pales in comparison to the Newly
Industrialized Asian countries (Malaysia, South Korea, and Indonesia) which started
industrialization in the same time period (1960s) as Pakistan. 20 The share of investment
and saving as percentage of GDP were above 30% during 1965-90 in East Asian
economies in sharp contrast to the low investment saving trends, approximately 13-14%
of GDP in South Asia. Pakistan has had persistently low levels of savings and investment
compared to other countries with similar levels of income (Table 2-9 below). Savings in
the economy have been constrained by both high fiscal deficits (reduced public savings)
and an inflationary environment (increased consumption expenditures).
The low level of saving compared to investment has widened the resource gap in
Pakistan. The resource gap went up by 3.2% in 2008 due to the increase in the
investment rate in that year and narrowed down to 5.4% in 2009 due to a reduction in
investment as a percentage of GDP. Low investments in the economy slow down
economic growth whereas low savings exert pressure on external accounts to finance
domestic investments. The large and widening resource gap also increases the external
debt and debt servicing burden which can potentially lead to macroeconomic instability.
Table 2‐9: Comparisons of Savings and Investments, 1994‐2009 (% of GDP)
20 Investment Climate Report (2007)
86

1994-2001 Average 2007 2008 2009
World Saving 22.1 23.7 23.8 21.4
Investment 22.4 23.3 23.4 21.5
Resource Gap -0.3 0.4 0.4 -0.1
Developing Saving 24.1 33.0 33.5 31.1
Economies
Investment 24.8 28.8 29.3 29.2
Resource Gap -0.7 4.2 4.2 1.9
Developing
Asia
Saving 32.7 44.7 44.7 43.6
Investment 32.4 37.9 39.1 39.5
Resource Gap 0.3 6.8 5.6 4.1
Pakistan Saving 15.4 17.8 13.3 14.3
Investment 19.0 22.9 21.6 19.7
Resource Gap -3.6 -5.1 -8.3 -5.4
Source: SBP (2008) and ADB (2010)
2.2.4 Actual & Potential Total Private Investment & Domestic Savings
Major structural, governance, and economic reforms21 were introduced in early 2000s
to improve business environment and foster macroeconomic stability (ADB, 2008). In
spite of these reforms the investment response from the private sector was
unimpressive. The contribution of private fixed investment to total GDP showed only a
5% increase in the last ten years (from 10% of GDP in 1999 to 15% in 2008). In the last
two years, the share of private investment in total GDP has fallen from 15% in 2008 to
11% in 2010 [ADB (2008)].
As noted earlier, the private investment growth rate has been uneven and sluggish in the
last three decades. In figure 2-21 below the potential indexed investment line shows the
trajectory that Pakistan would have followed post 1980s if Pakistan had maintained an
annual 9% investment growth rate. On comparing the post-1980s potential private
investment trajectory with actual private investment growth, private investment grew
less than the potential growth rate except for two short time periods 1988-94 and 2004-
21 Implementation of the Privatization Act 2000, creation of the Ministry of Privatization and investment,
Board of investments (BOI), and Insurance Act 2001 were important steps in this direction. A new legal
structure was introduced to strengthen the financial system which includes introduction of recovery laws,
legal structure for non-bank financial institutions, monetary and fiscal board for better coordination
between monetary and fiscal policies, and the Fiscal Responsibility and Debt Limitation Act 2005
primarily aimed at reducing fiscal deficits.
87

07. In the last two years, the private investment growth rate shows increasing deviation
from the potential investment trajectory (see Figure 2.21).
Figure 2‐21: Long Run Indexed Real/Potential Investment Decline in Trend Rate of Private Investment, 1980‐2009
Source: GoP (various years)
On average private investment grew 5.1% in the 1980s, 8.8% in 1990s, and 18% in
2000s. Even though private investment showed impressive growth during 2000s, there
were sharp fluctuations in the annual growth rate of private investment (Figure 2-22
below). The volatility in private investment reflects the uncertainty in the economic
environment exemplified by poor macroeconomic management and low demand for
produced goods. The uneven growth rate in private investment also matches the
economic growth patterns achieved during the same time period (see Figures 2.22 and
2.2).
Figure 2‐22: Growth Rate in Real Private GFCF & Total GFCF, 2000‐2010 (%)
88

25.00
20.00
15.00
10.00
5.00
0.00
-5.00
-10.00
-15.00
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Source: GoP (various years)
Total GFCF Private GFCF
2.2.5 Sectoral Composition of Investment
Empirical evidence shows that the social and economic returns of manufacturing
investment are higher compared to investment in other areas. Investment in the
manufacturing sector results in value addition, employment generation and expansion
of the productive capacity of the economy. In the context of Pakistan there is a dearth of
empirical evidence on economic growth and differential and average returns of
investment in manufacturing sector versus non-manufacturing sectors. Wang (2009)
analyses the impact of foreign direct investment on GDP growth in Asian economies and
finds higher growth effects for economies receiving FDI in manufacturing sectors.
The sectoral shift of investment from industrial sector to services sector (finance,
transports and communications) has serious implications on the long term growth
prospects of the economy. The spurt in economic growth experienced in early 2000s
was fueled by growth in services diverting resources from the manufacturing sector. The
share of industrial (large scale manufacturing, mining and quarrying, construction and
electricity) investment in total investment has been declining since 2003 whereas the
share of transport and communication investment rose during the same time period.
The share of industrial investment in total investment has declined from 38% to 20%
from 2000-10. On the other hand the share of transport and communication went up
89

from 12% to 24% of total investment22 . Although, the recent decline in total investment
is matched with falling shares of both industrial investment and transport and
communication investment. Interestingly, the share of small scale manufacturing
industries has increased slightly over the same time period and flattened at 5% of total
investment 23 .
Within the industrial sector, the manufacturing sector commands the highest
investment share in total industrial investment. In the last 10 years, the share of
manufacturing investment has slowly increased from 63% in 2000 to 75% in 201024 . On
average the increase in the share of manufacturing in total industrial investment has
been approximately 1% in the last 10 years. The change in annual share of
manufacturing investment is unimpressive compared to the sizeable contribution of
manufacturing sector to the GDP growth rate. The declining manufacturing share,
which decreased from 30% in GDP growth in 2001 to 23% in 2010, is a direct
consequence of the slow fixed capital formation in the manufacturing sector.
The share of total manufacturing investment in total investment has declined from 23%
in 2001 to below 15% in 201025 . Within manufacturing investment, the share of large
scale investment has substantially decreased in the last 2 years in contrast to a slight
increase registered by the small scale manufacturing industries. The share of large scale
industry was around 20% in 2000 which has dramatically declined below 10% in 2010
(see Figure 2-23) 26 . The downward trend in large scale manufacturing investment
reflects both an increase in the cost of doing business in Pakistan and the worsening
macroeconomic and security environment.
22 Authors calculation using various Economic surveys.
23 ibid
24 ibid
25 ibid
26 ibid
90

Figure 2‐23: Share of Manufacturing Investment in Total Fixed Investment, 2000‐10 (percent)
25.00
20.00
15.00
10.00
5.00
0.00
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Source: GoP (various years)
Share Large scal Share Small scale
2.2.6 Foreign Direct Investment
Foreign Direct Investment (FDI) is a package of capital, transfer of technology and
managerial skills from developed countries to developing countries. The literature
identifies strong linkages between economic growth and FDI in recipient countries. The
growth benefits largely hinge upon the flow of FDI in specific sectors. The concentration
of FDI in the manufacturing sector brings more benefits than in non-manufacturing
sectors of the economy [Wang (2009)].
Pakistan experienced phenomenal growth in FDI inflow in the last four years (2006-
09). During this time period, the country received $17.8 billion in the form of FDI which
grew six times (from $0.4 billion in 2001 to $2.4 billion in 2009) [SBP (2008-09)]. FDI
reached a peak of $5.6 billion in 2007 and stayed high for another one year in 2008. On
comparing Pakistan’s FDI performance with other countries using IND (inward FDI
performance index), Pakistan performed better than even China in attracting FDI in
91

2005-07 controlling for the size of the economy. 27 Pakistan clearly outperformed in the
last five years in receiving FDI compared to its regional competitors (see Table 2-10).
Table 2‐10: Inward FDI Performance Index, 2003‐08
Countries 2003-05 2004-06 2005-07 2006-08
Bangladesh 0.46 0.48 0.4 0.33
Nepal 0.05 -0.01 0.00 0.00
India 0.45 0.62 0.63 0.71
Pakistan 0.75 1.03 1.12 1.06
Sri Lanka 0.66 0.61 0.54 0.57
China 2.02 1.32 0.99 1.08
Source: SBP (2009)
However over the long term Pakistan’s ability and performance in attracting FDI in
comparison to its regional competitors in South Asia has been below par. South Asia as
a region claimed on average less than 1% of the total world FDI in the last decade with a
small proportion going to Pakistan. Table 2-11 below reveals the long term trend of FDI
in Pakistan in the last two decades. Pakistan had attracted on average $0.5 billion FDI
annually (about 20% of the total FDI inflow in South Asia) during the 1990-00 time
period whereas India received almost two thirds of total FDI to the region in the same
period.
Since 2001, South Asia’s share in total FDI is following an upward trend, but the share
of Pakistan in the regional FDI exhibits cyclicality. The share of Pakistan’s FDI declined
from 9.4% in 2001 to 5.6% in 2009 of total FDI in the region. However, India managed
to keep a high share which was above 80 percent of total regional FDI to the region in
2009.
Table 2‐11: Regional Comparison of FDI, 1990‐2000
Inward FDI Flows (US $ Billions)
Pakistan China India South Asia and Developing World
Asia Oceania Economies
1990-00 (annual
averages)
0.5 30.1 1.7 2.6 76.7 130.7 495.4
2001 0.4 46.9 3.4 4.1 108.7 217.8 825.9
27 The index ranks countries on the basis of relative share of FDI compared to world FDI and the size of
the economy of recipient country compared to world economy. If the value of index is more than one it
means the country is receiving proportionally more FDI than the relative size of its economy.
92

2002 0.8 52.5 3.4 4.5 92.0 155.5 716.1
2003 0.5 53.5 4.3 5.3 101.4 166.3 632.6
2004 1.0 60.6 5.3 7.0 147.6 233.2 648.1
2005 2.2 72.4 7.6 12.1 210.6 316.4 958.7
2006 4.3 72.7 20.3 27.8 284.4 434.4 1459.1
2007 5.6 83.5 25.0 33.9 338.2 564.9 2100.0
2008 5.4 108.3 40.4 49.7 374.6 630.0 1770.9
2009 2.4 95.0 34.6 41.4 303.2 478.3 1114.2
Sources: UNTAD (various country fact sheets) available at http://www.unctad.org
In line with the above data, FDI as percentage of gross fixed capital formation (GFCF)
has been quite volatile in Pakistan since 2001. The share of FDI as percentage of GFCF
increased from below 5% to close to 25% in 2007 (see Figure 2.24) due to heavy inflow
of FDI (see Figure 2-24). This share has shown a secular decline since 2008 as a result
of the international liquidity constraints. In comparison with other countries it is
apparent that the FDI as percentage of GFCF is not as volatile as in Pakistan. The FDI as
percentage of GFCF has followed a gradual upward trend in India and South Asia since
2005. However, recently there has been a decline in the FDI as percentage of GFCF due
to international liquidity constraints.
Figure 2‐24: FDI Investment as % of Gross Fixed Capital Formation, 1990‐2009 (percent)
30.0
25.0
20.0
15.0
10.0
5.0
0.0
1990-
00
Source: UNCTAD (various years)
2001 2002 2003 2004 2005 2006 2007 2008 2009
Pakistan China India South Asia Developing Economies
2.2.7 Sector­wise Decomposition of FDI
Optimal sectoral composition of FDI is immensely important in terms of realizing the
real gains from technological spillover effects, managerial skills transformation and
economies of scale. The growth premium of FDI in the manufacturing sector is higher
than the nonmanufacturing sector. The FDI sectoral composition for Pakistan has been
93

skewed away from the manufacturing sector. Pakistan mostly received FDI in the last
two decades in the oil and gas sector, power, financial, and telecommunication sectors.
In 1990, the concentration of FDI was in the power sector following the government’s
policy of allowing foreign IPPs (independent power producers) to enter the market. In
the late 1990s, the focus and share of FDI shifted towards areas like financial services
and telecommunication.
The recent phenomenal growth in the inflow of FDI in Pakistan also confirms the above
findings that Pakistan’s manufacturing sectors receives a low and insignificant amount
of FDI. The Figure 2-25 below provides sectoral comparison of FDI for two years 2004
and 2008. Financial business, telecommunication, and oil & gas are the main recipients
of FDI in Pakistan whereas the manufacturing sector claims less than one fifth of total
FDI received in 2008. The share of the manufacturing sector’s FDI has further declined
in 2008 compared to 2004. The low share of the manufacturing sector in FDI reinforces
the existing state of low technology accumulated over the time period.
Figure 2‐25: Sectoral Decomposition of FDI, 2004 and 2008 9percent)
100%
80%
60%
40%
20%
0%
Source: SBP (2008)
2004 2008
Manufacturing Oil and Gas Telecommunications
Financial Business Other Services
In order to ensure competitiveness and to improve the returns of investment in
manufacturing sector, the Government of Pakistan has introduced various incentives for
foreign investors. For the last two decades, Pakistan has introduced incentives like
allowance of 100% foreign equity, low import/custom duties and tax relief for
manufacturing investments supported by liberalization, deregulation and privatization
reforms and opening of all major sectors for FDI (See Box 2-1). The main objective of
94

these reforms is to ensure a business friendly environment by unleashing control over
market forces and offering a more competitive environment. For example, the initiation
of a one window option for the foreign investors and avoidance of double taxation are
key steps to be noted here.
Box 2‐1: Inward Investment Incentives Regime in Pakistan
The incentives include:
1. All economic sectors open to Foreign Direct Investment
2. Foreign investment fully protected
3. Equal treatment to local and foreign investors.
4. 100 % foreign equity on repatriable basis allowed
5. No Government sanction required.
6. Attractive tax / tariff incentives package.
7. Remittance of Royalty, Technical and Franchise Fee, Capital, Profits, Dividends allowed
8. Zero rated sales tax on import of plant, machinery and equipment
9. Initial Depreciation Allowance at the rate of 50% is permissible on an “eligible depreciable asset”
placed into service in Pakistan for the first time in a tax year.
10. Amortization of pre-commencement expenses allowed at the rate of 20% annually.
11. Amortization of intangible assets allowed over a period of ten years.
12. Rationalization and lowering of corporate tax rates
13. Full repatriation of capital, capital gains, dividends and profits allowed.
14. No restriction on payment of royalty and/or technical service fees for the manufacturing sector.
15. Agreements on Avoidance of Double Taxation with 52 countries
16. Bilateral Agreements on promotion and protection of investment with 46 countries.
Source: Pakistan Board of Investment
2.3 Constraints to Investment & Manufacturing Growth in Pakistan
In this section the major constraints hampering Pakistan’s investment and
manufacturing performance are highlighted in order to derive broader industrial policy
interventions. As discussed in the previous section the main ingredient behind economic
growth is increased capital accumulation/investments and higher productivity and
diversity within the manufacturing sector. An understanding of the major obstacles
impeding investments and manufacturing growth is therefore imperative in the
formulation of an industrial policy which is centered on the objective of achieving
95

sustained growth and wide spread employment. There is rich literature on comparative
analysis of developing countries explaining their disparate historical growth
performances. These studies show that the institutional, policy and regulatory
environment of an economy effect the investment and production decisions of firms
thus playing a critical role in explaining the cross country variations in aggregate growth
performance [Dao, 2008; Dollar, David, Mary and Mengistae (2005)]. The empirical
evidence supports the fact that firms perform better in developing countries which have
a stable macroeconomic and political environment, are relatively less bureaucratic, have
efficient financial and regulatory institutions and are supported by well functioning
factor markets.
Pakistan’s economic performance is marred by an array of constraints hindering the
growth potential of the economy and more specifically, that of the manufacturing sector.
These include supply side problems of the factor markets, macroeconomic instability in
the country and issues pertaining to the economic and business environment
(infrastructure, regulatory and administrative, crime and insecurity). This section will
begin with an analysis of the impact of macro instability on investment and
manufacturing growth. This would be followed by problems in factor markets and of
physical infrastructure which inhibit both investments and manufacturing sector
growth. Finally, the regulatory and security environment would be discussed in the
context of investment and manufacturing performance.
2.3.1 Macroeconomic Instability & its Impact on Investment & manufacturing
There is an integral link between investment and growth in the manufacturing sector
and the macroeconomic environment of a country encapsulated by the GDP growth rate.
Increase in investment leads to growth which in turn acts as a positive stimulus to
investments. This bi directional causality between growth and investments has been
empirically verified for both developed and developing economies. Macroeconomic
variables such as inflation, interest rates and exchange rate also impact investment and
manufacturing performance and in turn can be affected by the investment rate. An
unstable macroeconomic environment characterized by high rates of inflation, large
fiscal deficits, increasing external debt, and volatile exchange rate is in most instances
96

accompanied by a fall in both investments and manufacturing growth causing an overall
decline in GDP growth. For investors, macro instability and uncertainty about future
economic outcomes, such as demand conditions and expected relative prices, act as a
large cost potentially delaying or reducing current investments. On the other hand a
stable macroeconomic environment enhances investment rate, allows efficient
allocation of resources and raises the level of productivity and growth of the
manufacturing sector. Therefore the management of key macroeconomic policy
instruments, that is, fiscal policy, monetary policy, debt management, and exchange rate
policy can significantly influence factors determining investment level and growth.
Macroeconomic stability is gauged through level and consistency in fiscal deficits,
external debt, inflation rate, and extent of exchange rate volatility. These important
factors determine degree of risk in the business environment. For example, to raise
more revenue in response to poor fiscal management, the governments in developing
countries tend to monetize the debt or levy inflation taxes which cause a hike in overall
prices and thus negatively impact investment and manufacturing. Volatility in exchange
rate distorts relative price levels in the economy whereas poor debt management may
create a possibility of credit rationing from international financial institutions. There is
a large body of literature on developing countries which offers empirical evidences on
gains in economic growth and investments associated with macroeconomic stability
[Bleaney, 1996; World Bank, 1993; Fischer, 1993; and World Bank, 2004]. On average,
developing countries experience 0.15% decrease in GDP growth and 0.35% in
investment ratio for a 1% increase in fiscal deficit [Bleaney, (1996)].
The East Asian economic miracle brought about by a phenomenal increase in
investments and exports would not have been possible had it not been for the relatively
sound macroeconomic environment in these countries28 over a time period of 30 years
(from 1965 to 1990). 29 The fiscal deficit during these 30 years remained at a manageable
level across these countries. The inflation rate hovered around 9 percent compared to 18
percent in low and middle income countries. Due to sustained low inflation, the real
28 Hong Kong, the Republic of Korea, Singapore, Taiwan, China; and the three newly industrializing
economies of Southeast Asia, Indonesia, Malaysia, and Thailand.
29 World Bank (1993).
97

interest rate remained stable in the same time period. As a result, investment on average
remained 20 percent of GDP from 1965 to 1990 with a high composition of private
investment. These developments translated into a long period of sustained high
economic growth transforming the East Asian Economies.
Macroeconomic stability has been a key challenge for the policy makers in Pakistan for
the last four decades dominating the economic management agenda of successive
governments. On average Pakistan has had relatively higher GDP growth rates
compared to other developing countries in the last four decades. In early 1970s and
1980s Pakistan experienced GDP growth rates reaching almost 7% in 1970s and 9% in
1980s. However, growth rates flattened in 1990s and remained below 5%. The growth
rates during 2001-10 remained slightly above 5% [GoP (2009-10)]. The performance of
other macroeconomic variables such as inflation rate, fiscal deficit, exchange rate, and
balance of payments were less resilient as compared to the GDP growth rates in the
same time period. Figure 2-26 shows the strong positive correlation between
investments and GDP growth in the country. The downward trend of GDP growth post
2005 is in tandem with the fall in investment rates.
Since the initiation of macroeconomic reforms in 1999, Pakistan experienced a mixed
trend in its macroeconomic variables. A relatively stable macroeconomic environment
was experienced during the period 2001- 2007. Fiscal deficit showed a downward trend
since 1998 and declined from 7.7% of GDP in 1998 to 2.4% in 2004. On average, fiscal
deficit remained close to 4% in this period. In addition, current account deficit reduced
to 0.7% of GDP in 2001 from 4.1% in 199830 .
Figure 2‐26: Real GDP and Investment Growth in Pakistan, 1991‐2009 (percent)
40.00
35.00
30.00
25.00
20.00
15.00
10.00
5.00
0.00
-5.00
1991 1992199319941995 1996199719981999200020012002200320042005200620072008 2009
-10.00
30
Authors’ calculations using various issues of economic surveys.
investment growth rates GDP growth rates
98

Source: GoP (various years)
In fact, in 2003 the current account deficit turned into surplus due to the double digit
growth rate experienced in exports which reached to 19.1%. Stable exchange rate, falling
interest rate and single digit inflation boosted investment rate to an unprecedented level
of 22% of GDP in 2007.
However, macroeconomic instability revisited Pakistan after five years in 2008. The
current account deficit was as high as 8.8% in 2008 due to increase in oil import bill.
Fiscal deficit increased to the level of 7.6% in 2008, 5.2% in 2009, and is believed to
remain above 5% in the years to come due to future debt repayment31 . Since 2008,
inflation and interest rates have gone up signalling deterioration in the macroeconomic
environment. The level of inflation reached a peak of 20% in 2009 and 13% in 2010. In
response to high inflation and as part of the macroeconomic reform agenda, the State
Bank resorted to a contractionary monetary policy resulting in a significant increase in
the interest rate. The worsening economic and security conditions in the country caused
an outflow of short term capital which consequently led to the depreciation of the
currency. The rupee exchange rate touched an historic low of Rs 85 per dollar (22
percent depreciation) in the last 3 years. Resultantly, the investment to GDP ratio also
followed a downward trend in the same time period. Figure 2-27 below shows the
dramatic fall in investment growth around the year 2007-08, which coincides with the
increase in the interest rates from single to double digits.
Figure 2‐27: Investment and Interest Rates, 2001‐2009 (percent)
31 ibid
99

45.00
40.00
35.00
30.00
25.00
20.00
15.00
10.00
5.00
0.00
Source: GoP (2010a)
2001 2002 2003 2004 2005 2006 2007 2008 2009
6 months T-bills lending rate
investment growth rate Private investment growth rates
Therefore the growth experience of Pakistan can be characterized as having spates of
high growth followed by periods of low growth rates. In recent years these ‘boom’ and
‘bust’ cycles have been brought about mainly by changes in consumption demand
triggered by short term capital flows, aid and a surge in remittances. The main reason
behind a lack of sustained GDP growth is the instability in investment rates and the
resultant sluggish growth in the manufacturing sector which in turn has impeded the
structural transformation of the economy. The failure in fiscal management, both in
terms of revenue generation and allocation and size of expenditure, has been the root
cause of the country’s macroeconomic problems.
2.3.1.1 The Twin deficits – Fiscal and Trade
Pakistan’s macroeconomic instability is largely explained by its persistent failure in
containing the fiscal deficit within 4-5% of GDP. 32 The chronic high fiscal deficit resulted
in increased borrowing from both internal and external sources to cover the resource
gap.
32 The large fiscal deficit of the 1970s is attributed to the expansionary fiscal policy as a result of
nationalization policy followed by Bhutto’s regime. Large increase in interest payments over the debt
accumulated in the 1970s worsened the fiscal deficit in subsequent periods, 1980s and 1990s. The fiscal
deficit grew approximately 7% until 1990s. The high fiscal deficit in 1990s is also attributed to the failure
of economic management in terms of reducing expenditure and increasing revenues along with
implementing key structural reforms in the economy. During this period the deficit was mostly financed
by extensive controls on financial markets, relatively strong monetary growth and external borrowing.
The deficit peaked at almost 9% of GDP in 1990-91. It was, however, subsequently managed down to 5.5%
by 1994-95. After the adoption of macroeconomic stability program in 1999, fiscal deficit dipped down to
the level of 5% during 2000-09
100

Since 2004, the internal gap (Government consumption and tax revenue gap) have
coincided with the external gap (export and net factor income i.e. remittances) (see
Figures 2-28 and 2-29). There was a large capital inflow into Pakistan in the form of
remittances and aid post 9/11 which led to an appreciation of the nominal and the real
exchange rate. This appreciation reduced the competitiveness of our exports (the Dutch
Disease effect) and increased import demand significantly leading to a widening current
account deficit. In order to bridge the trade and fiscal gap, the economy has to rely on
external sources of financing. Delays or non-availability of external finance jeopardize
the macroeconomic stability of Pakistan in the long term exerting pressure on foreign
reserves, exacerbating inflation and interest rates, and deteriorating currency stability.
Figure 2‐28: Tax Effort and Government Expenditure Gap, 1990‐2008 (percent)
Source: Felipe and Lim (2008).
Figure 2‐27: Exports, Net Factor income from Abroad and Imports, 1990-2008 (percents of GNI)
101

Source: Felipe and Lim (2008).
Recently, the external account has further deteriorated after showing some recovery
signs in early 2000s. With the slowdown of international economic growth and surge in
international oil prices, the current account surplus went into a deficit in 2005. In the
last two years as a result of an increased import bill due to the sharp depreciation of the
exchange rate accompanied with sluggish domestic manufacturing growth and a
resultant fall in exports, the current account deficit mounted to 8.4% of total GDP
[Felipe and Lim (2008)].
Since external capital inflows are quite unpredictable, external imbalance depletes
foreign reserve, exerts pressure on inflation and results in exchange rate volatility. In
2007, Pakistan’s foreign reserves declined by 27% due to current account deficits and
net outflows from portfolio investment. In 2008-09, the foreign exchange reserves
accumulated ($12 billion) on receiving inflows from IMF as part of macroeconomic
stabilization program and other agencies. In the current year, 2010, Pakistan’s foreign
reserve further strengthened ($15 billion) due to lower current account deficit and
increase in remittances [GoP (2009-10)].
2.3.1.2 Effect of Budgetary Rigidity and Reduced Fiscal Space on Investments
The inability of the government to finance its fiscal deficits through increased domestic
tax revenue has led to a severe restriction in fiscal space over time. The accumulated
debt over the years and interest payments on the principle require a large portion of the
budgetary allocation limiting the share for development expenditures. According to
2010 budget estimates, almost 79 and 21 percent of total expenditure is spent on current
102

and development expenditures, respectively. Key components of current expenditures
like debt servicing, subsidies and defense expenditure explain its rigid structure.
Interest payments alone, account for 27% of total expenditure.
Due to the rigid structure of current expenditures, it is development expenditure which
ends up being curtailed to reduce the fiscal deficit for e.g., cuts in spending on physical
infrastructure, despite its significance in enhancing both the quality and quantity of
private investments. The quality of investment is also affected by low development
expenditure on health and education. Low spending on education and vocational
training leads to a poorly trained or worse, unskilled labour force (discussed in the next
section), trapping the manufacturing sector in a ‘bad’ equilibrium of low value-added
and low technological intensive products.
The mode of budget deficit financing in developing countries can further add to
macroeconomic instability [Fischer and Easterly (1990); Easterly and Schmidt-Hebbel
(1993)]. Low tax revenue to GDP ratios leave few options for the Government to finance
the deficit, except borrowing from external sources, domestic banks, and from the State
Bank of Pakistan. Financing through domestic banks raises the interest rate and leaves
less credit for private sector investment – crowding out effect. Following a downward
trend till 2005, interest rates rose sharply (contractionary monetary policy) in response
to high inflation, causing a significant fall in investment.
With little improvement in the repayment capacity, debt continued to accumulate at an
alarming rate. Serving to reinforce each other, the volumes of both the fiscal deficit and
debt soared continuously. The bulging domestic debt profile was the result of “excessive
government expenditures, stagnant tax revenues, high returns on government securities
and inappropriate sequencing of fiscal reforms” [State Bank of Pakistan, (2001)]. On the
other hand, unwieldy current account deficits, sluggish export revenues and declining
worker remittances caused a rise in the debt burden of the country.
During 2000- 2001, the debt situation was dire. After 9/11 and the war on terror, huge
amounts of debt were either written off or rescheduled in return for Pakistan’s support.
In addition, the quota for Pakistani exports to the US and the European Union was
103

increased as a premium for being a part of the coalition in the war against terrorism. As
a result of foreign aid and foreign investment inflows, there was a substantial increase in
remittances as well, all of which helped the country with its debt situation. In 2006, the
debt to GDP ratio improved further and fell below 60% for the first time in two decades.
However this improvement was not long-lived as Pakistan’s debt position deteriorated
sharply in fiscal year 2008, reflecting the country’s large fiscal and current account
deficit.
2.3.1.3 Monetization, Inflation and Investments
Structural issues in the budget imply that financing budget deficits through printing
money increases inflation in the economy. In other words, fiscal imbalance and
expansionary monetary policy have a strong linkage with inflation. From 1991 to 1995,
inflation rates ranged between 9.3% and 12.9% as a direct result of monetary expansion
and low growth rates. Money growth rates had climbed up to a level of 12.6% in 1990
and had stayed in the range of 16% to 18%, except when they reached an exceptionally
high rate of 30% in 199233 . Pakistan had experienced sustained inflation, ranging from
10% to 13% till 1998. The persistence of double-digit inflation along with high fiscal
deficits has been one of the biggest issues for the Pakistani economy, as both were major
causes of the macroeconomic imbalances of the 90s. There has been general agreement
that excessive growth in money supply, adjustment in government–administered prices,
imported inflation (pass through of exchange rate adjustment), escalations in indirect
taxes, and inflationary expectations are the major factors responsible for the persistence
of double-digit inflation during most of the 1990s. Both food and non-food inflation
were at play, averaging 11.6% and 10.3% respectively during the eight years of
the1990s34 .
Figure 2‐28: Inflation and Investment in Pakistan, 2001‐2010 (% p.a.)
33 Authors calculation using various issues of Economic Surveys
34 ibid
104

40
35
30
25
20
15
10
5
0
Source: GoP (2010a)
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Grow th WPI Grow th GDPD Investment grow th rates
During 2000-2003, inflation decelerated to less than double figures. However this did
not last long as inflation rose again in fiscal year 2005-2006. This rise was mostly on
account of a sharp economic recovery that raised income levels and hence domestic
demand, coupled with the continued impact of mounting oil and commodity prices
internationally. Sharp increases in international prices of food and energy in 2008
again led to a noticeable spike in inflation rate (as measured by the Wholesale Price
Index and GDP deflator). The accompanying decline in investment during this period
created a situation of stagflation in the economy [see Figure 2-30].
As part of the fiscal reforms tied to the ‘Fund’ conditionalities and in a bid to curtail
subsidies for fiscal consolidation, the Government is obligated to transfer the hike in oil
and utilities prices to consumers (elaborated in detail in the energy section). Passing on
higher prices to consumers and industries leaves economic managers with even fewer
options for reducing the cost of doing business and improving the competitiveness of
the economy.
2.3.1.4 Policy Recommendations
2.3.1.4.1 Fiscal Policy
105

The government faces an immediate need to reduce the bourgeoning budget deficit and
increase its limited fiscal space. For that, the Federal Bureau of Revenue (FBR) has to
substantially increase its revenues by widening the tax base. The tax to GDP ratio of the
country is around 9 percent which is one of the lowest in the world. The tax base has to
expand to include sectors which are out of the tax net, such as agriculture. Currently, the
manufacturing sector bears most of the tax burden.
Moreover, instead of levying more taxes or increasing tax rates, FBR has to improve its
tax collection system. Viable incentives need to be provided to tax officials to minimise
rent-seeking and increase tax collection. At the same time, appropriate incentives have
to be generated to incorporate the large informal economy existing in manufacturing,
services and retail.
On the expenditure side, there has to be a significant reduction in wasteful current
expenditure and a slash in defense spending (without compromising the security of the
country). A reallocation of resources towards development expenditure, in particular
health, education, energy and physical infrastructure, is required to stimulate
investment in the economy and increase the vitality and productivity of the
manufacturing sector. The Ministry of Industries is an important stakeholder when it
comes to budgetary allocation decisions. It should play its role as an industry
representative in making a case for increased spending in areas which directly or
indirectly benefit industry. However, given the tight availability of fiscal space, the
ability of the state to invest in large-scale infrastructure development projects is
impaired. Also, any new or greater spending allocation carries with it a huge opportunity
cost. The government would therefore need to carefully prioritize its development
spending on the basis of maximum social and economic rate of returns.
Moreover, an increase in domestic revenue generation and the resultant reduction in
fiscal deficit would reduce the debt burden, both domestic and international, on the
economy, creating greater fiscal space in the future. A reduction in international
borrowing from multi-lateral agencies that impose conditionalities in the form of
prescribed policy rules and restrictions would also buy the government greater policy
autonomy or ‘policy space’.
106

2.3.1.4.2 Monetary Policy
In times of macro-instability characterized by high unemployment and inflation, the
central bank or monetary authority normally resorts to a contractionary monetary policy
to curb demand-pull inflation in the economy. However, the resultant higher interest
rates raise the cost of capital, negatively affecting investment in the economy and
causing stagflation. As investment in the manufacturing sector also contracts due to the
higher cost of borrowing, industrial growth tends to slow down during such
macroeconomic reforms. The first-best policy prescription is therefore to avoid any such
macro imbalance or instability which would then call for contractionary measures.
Developing countries in general and Pakistan in particular, have more often than not
financed fiscal deficits by printing money (monetisation) or borrowing commercially.
While both methods of financing are inflationary, the latter also results in crowding out
of private investment. Therefore using expansionary monetary policy to finance fiscal
deficits is not a viable policy option.
2.3.1.4.3 Trade and Industrial Policy under WTO
Pakistan has, over the past two decades, significantly liberalized its trade regime, both
on account of its membership of the WTO and the Structural Adjustment Programs of
the World Bank and the IMF. Liberalization has entailed a conversion of quota
restraints to tariff barriers with subsequent reductions in tariff rates and elimination of
export subsidies. Therefore liberalization reform to some extent makes government
usage of trade policy [as an instrument of industrial policy] a thing of the past.
International trade rules and multilateral agency conditionalities have an impact on the
policy space available to governments. The term ‘policy space’ has been defined as ‘the
scope for domestic policies, especially in the area of trade, investment and industrial
development which might be framed by international disciplines, commitments and
global market consideration’ [ODI, (2007)]. The major international commitments as
outlined in the Doha Round of WTO talks are in six areas: tariffs, agricultural policy,
107

services, TRIPs (Trade Related Intellectual Property Rights), investment and Aid for
Trade.
As far as tariffs are concerned, ‘bound’ tariffs on commodities as negotiated with the
WTO are normally much higher than ‘applied’ tariffs. This allows considerable policy
space to countries to increase tariffs as part of an industrial strategy to give temporary
protection to a strategic sector. In the case of Pakistan, for example, the bound tariff rate
on average is almost 100%, while the applied tariff is around 15%. This low applied tariff
is not due to the WTO rules, but because of liberalization reforms included in the IMF
and World Bank Structural Adjustment and Macro Stabilization programs. It should be
noted that bilateral and multilateral trading agreements may also result in tariff
reductions.
Flexibility under international rules allows government to use tariffs (between applied
and bound rates) for temporary protection of industries. Peak (higher than average) and
escalating (low on inputs and high on finished products) tariffs can be used to protect
strategic or ‘sunrise’ industries/sectors that can potentially exhibit economies of scale or
are value-added industries. However such protection has to be conditioned on clearlydefined
objectives and targets, such as export performance and product quality
improvement. Otherwise, a policy of arbitrary unconditional protection would either
breed inefficiency in production (eventually lowering the quality of the protected
product) or just give another lease of life to ‘sunset’ or declining sectors.
The use of subsidies such as export subsidies or subsidies on domestic inputs (to
discourage use of imported intermediate goods) is prohibited under the WTO
Agreement on Subsidies and Countervailing Measures (SCM). The subsidies which are
classified as non-actionable and hence permitted under SCM fall under the category of
Research & Development and Environment Initiatives for disadvantaged regions. This
category can be effectively employed to upgrade the technology of production and
quality standards within an industry.
Other available import-substituting policy instruments are anti-dumping duties and
safeguard measures, which are invoked on the premise of fairness in competition.
108

Availing these measures requires diligence and capacity on the part of the importcompeting
sector under threat as well as effective pursual of the case by the Ministry of
Commerce and Industries at the WTO.
The most controversial of the WTO rules deal with TRIPs, which introduces regulations
on patents and copyrights. The main implication of this is the imposition of costs on
countries which are net importers of technology, i.e., developing countries. Such rules
would limit the diffusion of technology through reverse engineering and copying. The
upside of TRIPs for developing countries is Article 66.2, which requires developed
countries to incentivize enterprises and institutions in their territories to promote and
encourage technology transfer to developing countries members of the WTO. Although
the TRIPs agreement also allows developing countries to subsidize R&D and then
protect the innovation through property rights, very few developing countries can be
classified as innovators, thereby restricting the benefit of TRIPs to the leaders in
technology.
2.3.2 Infrastructural Constraints on Industrial Growth
2.3.2.1 Energy
The acute energy shortage of the past two years is the most important constraint
adversely affecting the industrial sector today. The government of Pakistan has been
unsuccessful in meeting the growing demand of energy and has failed in maintaining a
consistent supply of this vital input to the industrial sector. The energy crises of Pakistan
manifested in the unreliability of power supply and frequent load-shedding places a
huge burden on the manufacturing sector. Moreover, the existing power subsidy
structure charges a higher tariff from the manufacturing sector in order to subsidize
power for commercial, agricultural and domestic consumption. The shortage in supply
and periodic upward revisions in the power tariff structure has substantially increased
the costs of production of industry, eroding its profit margins and international
competitiveness. The unscheduled and persistent outages affect the production
processes of a large number of industrial units, with electricity-intensive sectors bearing
109

most of the burden, for e.g., those producing textiles, basic metals, leather products,
rubber and plastic products, paper and paper products, etc. The gravity of the situation
is evident from data gathered by the Investment Climate Surveys (2006) which suggest
that the number of firms that consider electricity to be a major constraint has risen by
50% between 2002 and 2006. Since the problem is now worse than it was in 2006, the
number of affected firms is bound to be greater (See Figure 2-31).
The main reason behind this problem was the inability of successive governments to
increase energy supply in accordance with the projections of demand in the country.
The demand for electricity has been rising steeply over the years, with consumer
demand rising sharply in the current decade. In fact, the share of domestic consumer
demand in total consumption has increased from 23% in 1980-81 to 46% in 2007-08
[Institute of Public Policy (2009)]. Moreover, electricity demand is outstripping
supply by a significant margin, as is evident from Figure 2-31, and by 2010, demand is
expected to exceed supply by approximately 5,500 MW. The government was, however,
unable to anticipate the rising demand and therefore adequate measures were not made
in time. Proof of the government’s complacency on the energy issue is found in the
declining proportion of public sector expenditure on the power sector from 28% in the
1980s to less than 3% in 2000s [Institute of Public Policy (2009)].
Figure 2‐29: Electricity Demand and Supply, 2003 – 10 (MW)
Source: Private Power Infrastructure Board: http://www.ppib.gov.pk
Ad hoc government policymaking regarding privatization and delays in tariff reforms
did not allow IPPs to make appropriate investments in improvement and upgradation of
110

technology which could have absorbed some of the excess demand in electricity. Circular
debt was one of the consequences of such government indifference. The government
failed to make timely payment to power producers who were unable to pay the oil and
gas companies supplying them with the inputs, resulting in decreased production. The
consequence of these deferred payments was the accumulation of PKR370 billion in
debt [Institute of Public Policy (2009)]. Therefore mismanagement in the power sector and
lack of appropriate and timely policy intervention fuelled the problem. Other proximate
causes include downturn in the economy owing to the war on terror, fluctuating oil
prices, and collection losses due to the dilapidated transmission and distribution
systems.
A study conducted in 2008 by the Institute of Public Policy (IPP), Beacon House
University, estimated costs of outages to industry by surveying a sample of 65 industrial
units. The survey was conducted in four main industrial centers of Pakistan and
included both continuous and batch-making industries. The estimated cost of loadshedding
for firms with self-generation facility was found to PKR74 billion, and for
firms without self-generation, PKR83 billion. Therefore the total cost to the industrial
sector was estimated to be PKR157 billion, which was 9 percent of the total industrial
value added and constituted a 7 percent loss in production. The resultant employment
loss was estimated at 300,000 workers [Institute of Public Policy (2009)].
Slowdown in the manufacturing sector results in reduced economic activity in sectors
such as wholesale and retail trade, transport and communications, banking and
insurance etc. As a change in industrial value added has a multiplier effect on the rest of
the economy, the IPP report also extrapolated the national cost of load-shedding with
the aid of an estimated short run multiplier using their estimated industry costs. The
total cost of industrial load-shedding to the economy was estimated at PKR210 billion,
which comes to around 2 percent of GDP. The estimated loss of exports was found to be
PKR75 billion [Institute of Public Policy (2009)].
From an international standpoint, given that Pakistan has the greatest number of firms
suffering from maximum hours of power interruptions; it should come as no surprise
111

that it comes second only to Bangladesh in terms of losses to firms due to such
interruptions. According to the World Banks Enterprise Survey 2008 (Pakistan country
Profile 2007), in Pakistan the loss to firms due to power outages is approximately 10
percent of sales. This is slightly lower than the South Asian average of 11 percent but
much higher in comparison to that of low income countries which is around 7 percent.
Nationally, we observe that the situation was much worse in 2007 than it was in 2002,
and that amongst the provinces, Sindh suffered the most. Based on size, small-sized
firms were the largest victims of the energy problem.
Installing generators to supply electr icity when there is an outage, is however, not a
very feasible alternative since the cost per unit of electricity is significantly higher if
using generators. It must also be noted that only around 6 percent of domestic energy
production occurs using generators, suggesting that the problem of power outages is still
more or less unresolved.
2.3.2.1.1 Inefficient Distribution and Management System
In 2006, the rundown condition of the transmission and distribution system cost
Pakistan an estimated PKR4.3 billion for each percentage point line loss. According to
the World Development Indicators, following India, these losses are greatest in
Pakistan. This situation, though still fairly serious, is an improvement over 1998-99,
when these losses amounted to 31%. 35
Another indicator of the inefficiency of power production is the delay in getting an
electricity connection. The waiting period for a new connection in Pakistan is up to 92
days. This is more than three times that in India and about 50% higher than that in
Bangladesh [World Bank, (2009b)]. Moreover, the waiting time for a new connection in
Pakistan has deteriorated drastically from 2002, when it was about 32 days. On a
national level, the situation has actually worsened in 3 out of 4 provinces from 2002 to
35 http://socyberty.com/issues/energy-crunch-in-pakistan-cause-effect-solution
112

2007, with delays in NWFP now almost 3 times more than in 2002. As for different
sized industries, medium-sized firms suffer the most with a delay of around 140 days.
The loss to firms as a result of delays in installing a connection is compounded by the
informal payments (rents) made in order to get a connection. According to World Bank
(2009b), informal payments are highest in Pakistan if compared internationally. This
implies a high level of corruption another important constraint to be discussed later.
Power therefore remains an important issue that needs to be addressed by the
government, since it not only has direct costs for the industrial sector, but also a
multiplier effect in the economy as it slows down the activity in other sectors such as
transportation, banking and wholesale and retail trade. Thus it is now essential to
prepare a strategy for the development of the energy sector in Pakistan, with focus on
increasing electricity generation, improving the quality of electricity distribution, and
finding alternatives to depleting national gas resources.
113

Box 2‐2: Energy in Comparison with China and India
Energy is a vital ingredient for driving and sustaining the engine of growth in any country. This is
precisely why rapidly growing and industrializing economies like China and India have paid special
attention to energy production. According to a report by the United Nations, Global Trends in Green
Energy (2009), "Manufacturing leadership is shifting from Europe to Asia, as countries like China, India
and South Korea continue to increase their commitments to renewable energy.”
In China, coal production is the main energy source, providing 70-75% of energy presently. In 2007, the
amount of coal produced was equivalent to 2.37 billion tons, making China the second largest energy
user in the world. However, China intends to improve its energy mix by increasing the shares of
alternative energy and nuclear energy by 15% and 5%, respectively.
India is reported as the fifth largest wind and solar-water heating energy producer by the Global Trends
in Green Energy report (2009). The shares in the energy mix of the country in 2009 were: coal 60%,
hydropower 20%, oil and gas 11%, alternative 6% and nuclear 3%. Like China, India also plans to
increase the share of alternative and nuclear energy in its energy mix. The goal set by National Action
Plan on Climate Change (NAPCC) is to meet 10% of country energy needs through renewable sources by
2015. As for solar power, the target set by the National Solar Mission is 20GW by 2022. In order to meet
these targets, the government is offering generation-based incentives for the solar power industry.
Moreover, in collaboration with some advanced countries, India aims to set up power stations that are
powered by coal and have zero emissions in the long-run. The country also currently has around a
million solar systems operating in rural areas.
Pakistan also has a huge reservoir of renewable energy sources that are yet to be exploited. If proper
attention is paid to these energy sources, Pakistan would be able to overcome an energy deficit that is
estimated to rise to 64% by 2030. During the year 2005-2006, the government commenced many
projects on renewable energy including the development of wind and solar energy so that these could
generate at least 5% of power by 2030. The government has also devised a strategy to expand the
generation capacity of nuclear power to 8800mw by 2030. For this purpose, Pakistan Atomic Energy
Commission (PAEC) plans to set up multiple units on the same site. As for manufacturing nuclear fuel,
the PAEC intends to institute a Pakistan Nuclear Power Fuel Complex (PNPFC). However, a well thought
out approach and vigorous efforts are essential to make the aforementioned projects meet their desired
objectives.
114

2.3.2.2 Policy Recommendations
2.3.2.2.1 Securing Supply
An analysis of the energy mix of Pakistan reveals that out of the 5 most commonly used
sources of energy [namely oil, nuclear, natural gas, hydroelectric and coal], Pakistan
uses natural gas (50%) and oil (30%) the most, in contrast to the rest of South Asia,
high-income countries and the world at large, which rely on coal. Although coal is
abundant in Pakistan, it is of poor quality in most cases. The energy mix in Pakistan
needs to be altered so as to allow extensive power production through low cost means
which would make industry more competitive.
In order to bridge the growing energy deficit, investments in the power sector are
imperative. Expanding power generation capacity in the country would entail upgrading
existing generation facilities and investing in both thermal and hydel plants.
In the medium- to long-run, we strongly propose a shift in energy mix. This may be
attained by developing localized, cheaper machinery for hydel, thermal and coal-based
power plants. For this purpose, the interaction of manufacturers and power producers,
including WAPDA needs to be facilitated to develop machinery best suited to Pakistan’s
needs.
In special economic zones, science parks and industrial estates, captive power
generation should be allowed. At the same time, steps should be taken to facilitate the
local development of wind turbines and solar energy technology. For this purpose, pilot
research projects should be initiated bringing together universities, industry, foreign
and local experts and relevant government departments.
The following is a description of the merits and demerits of the various sources of power
generation.
115

Hydel or Hydropower
While the generation of electricity using hydropower saves fuel, greater investments are
required in hydropower than in thermal generation. Hydropower plants are also located
away from areas where demand for energy is high, raising transmission costs for
hydropower. But keeping these constraints in mind, estimates suggest that hydropower
is more attractive as its maintenance costs are lower and that it is more efficient when
fuel prices rise and discount rates decrease [(Ministry of Industries (2005)]. The hydel
projects which are in the pipeline should be completed as soon as possible. These are
Neelum-Jhelum (969MW), Tarbela 4th Extension (960MW), Suki Kinari (840MW),
Munda Dam (700MW), Khan Dubar (130MW), Allai (126 MW), and Jinnah Hydro
(96MW) [Institute of Public Policy (2009)].
Coal
Coal is an under-utilized resource and generates only 5 percent of Pakistan’s energy
needs and 1 percent of electricity needs. In comparison, India uses coal to meet 54%
and 47% of its energy and electricity needs respectively [Ministry of Industries (2005)].
In addition, coal can also be used for gasification, fuel and chemical extraction,
briquetting and in the processing industry. Substitution of coal for gas has taken place in
the cement industry, but coal is mostly imported. Thus in the long-run there needs to be
a development of coal fields (in Thar for example), and setting up of power plants. This
sort of exploration and development is a provincial matter and provincial governments
will need to be on board if coal is to become an alternative energy source.
Oil and Gas
It is estimated that at prices of PKR11000 per ton for oil and PKR172 per MCF for gas,
the substitution of gas for oil will reduce the fuel component of electricity generation by
PKR0.52 per kWh to PKR2.37 per kWh. This amounts to a reduction of 18% in fuel cost
and 13% fall from the average electricity tariff of PKR4.09 in 2003 (WAPDA, 2003).
Furthermore the World Bank (2003) estimates annual savings of 130,000 tons of fuel as
a 300MW plant is converted from oil to gas. Yet caution is advised, as conversion to gas
116

carries the risk of depleting resources in the next five years unless new reserves are
found [Ministry of Industries (2005)]
2.3.2.2.2 Achieving Efficiency in Distribution
A comprehensive program that focuses on reducing technical losses and improves the
reliability of the distribution system is required. Distribution companies should be
provided adequate resources to upgrade their overloaded transmission and distribution
systems. It is estimated that the required investment could potentially be recovered in
less than three years through savings in transmission and distribution losses [Institute
of Public Policy (2009)].
The circular debt problem needs to be resolved in the short-run. The payments to IPPs
(thermal) should be conditioned on the utilisation of at least 75 percent of their capacity.
Sector-wide energy audits should be conducted, starting with heavy load industries.
Based on the recommendations of the audit, incentives should be provided to the
industry to shift toward more energy-efficient production methods and technology.
2.3.2.2.3 Improvement in the Load Management System (LMS)
The schedule of outages should be in accordance to the needs of various industries.
Those sectors or clusters which are heavily reliant on electricity and incur large losses
due to unscheduled power interruption should be given priority and should be provided
continuous electricity for the scheduled period.
As long as power shortfalls remain, the load needs to be managed carefully. Areas with a
heavy presence of industry, both large-scale and SMEs, should be given the status of
industrial corridors. These corridors could be given separate feeders, where loadshedding
would only occur when absolutely necessary36 . This needs to be arranged after
consultation with PEPCO, so that the present suboptimal matching of grid stations and
11kv feeders can also be addressed. Load-shedding schedules for such units need to be
36 This point is also addressed in the section on industrial estates and other special economic
zones.
117

announced at least two months in advance and load-shedding days should be clustered
for both electricity and gas. No such priority should be given to industry that is based in
residential areas.
2.3.2.2.4 Pricing Policies
If subsidies are essential, they must be up-front budget allocations rather than crosssubsidization.
In 2003, fertilizer producers and household consumers were paid a
subsidy of PKR 14 billion and PKR 9 billion respectively [Ministry of Industries (2005)].
These were cross subsidies that resulted in a higher price for industrial consumers.
Cross-subsidies increase the cost of energy for industry other than fertilizer, and results
in supply shortages due to over- consumption in the household sector. Energy pricing
should be rationalised according to usage; however priority should be given to industry
in determining the extent of sector-specific electricity subsidies. In industrial corridors,
peak-load pricing schedules should be announced.
Finally, the industrial sector should be consulted and informed about any changes in
pricing policy well in advance so that these are factored into their costs of production.
Post-dated billing of industry electricity consumption should be avoided.
2.3.2.3 Transportation
An efficient and well-integrated transport system is integral to the growth and
development of industry as it reduces the cost of production and increases the
competitiveness of firms. In Pakistan, however, transport systems suffer from
insufficient investments, poor management and neglect of essential maintenance. The
low reliability and high costs of the transport sector hinder Pakistan’s economic growth
by reducing the productivity of its industry and the competitiveness of its exports. The
cost of inefficiencies in the country’s transport system has been estimated at about 4-6
percent of GDP [The International Monetary Fund (2010a)].
2.3.2.3.1 Roads
118

Road transport caters for 90% of Pakistan’s passenger traffic and 96% of its freight. The
road network in Pakistan not only lacks quality, but also suffers from undermaintenance.
In fact, 70% of the 7086km national road network is of poor to fair quality
[Energy and Infrastructure Sector Unit South Asia Region (2002)]
Despite the significance of the road transport system, it has received insufficient
investment particularly from the private sector. Even though public sector investments
have been made in the road sector led by the National Highway Authority (NHA), the
capacity increase has not been sufficient to match the increase in traffic flows. For this
purpose, NHA devised a policy framework in 1999, along with incentives to attract
private sector investment for national highway projects. But the efforts have been in
vain. This is because out of the two projects granted to the private sector by NHA, one
contract was cancelled whereas the other project developer was unable to raise the
finances needed for the project [Asian Development Bank (2008)].
Moreover, due to the poor condition of roads, Pakistan has a record of over 70 fatalities
per million of population per year [Energy and Infrastructure Sector Unit South Asia
Region (2002)]. This is estimated to cost USD1 billion to the economy [Energy and
Infrastructure Sector Unit South Asia Region (2002)]. The major entities responsible in
this regard are the Ministry of Industries and Production, Ministry of Commerce,
Ministry of Railways and the provincial governments.
Finally, trucking, a much-used form of transport, has not been developed as a formal
activity. The owners of these sole proprietorships and partnerships have little incentive
to register themselves as doing so does not promise a growth to business; it is an
additional regulatory burden. Further problems include the age (20 years on average)
and condition of trucks, overloading of trucks and high costs for less-than-containerload
shipment [World Bank (2006)].
Thus in order for the road sector to be able to aid sustained GDP growth, an in-depth
analysis of the road transport network should be conducted and followed up by a
carefully planned and implemented strategy for the expansion and upgradation of the road
119

network. Road linkages connecting markets and industry must be developed for efficient
logistics.
2.3.2.3.2 Railways
Pakistan Railways is a non-profitable and technically insolvent public sector unit. Over
the past thirty years there has been a continued diversion of resources towards the
expansion of the road network at the cost of the railways. During this period, as a
consequence of lack of resources, railway share of inland passenger traffic declined from
41% to 10%, and in freight traffic, from 73% to a meager 7% [World Bank (2007)].
There has hardly been any private sector activity in this sector and plans to privatize
railways have not been implemented. There has been no change in the Railways Act
since 1980. Although the Act allows the private sector to participate in rail travel, it
assigns the Government the role of both competitor and regulator [Asian Development
Bank (2008)]. Thus the Act needs to be revised to create effective Public-Private
Partnerships (PPPs) in the railway sector [Asian Development Bank (2005)]. A business
plan has been developed for 2005-2011 that stresses private sector participation in the
railways sector in order to enhance its efficiency and competitiveness. However, this
plan needs to be implemented speedily and effectively [Asian Development Bank
(2008)].
2.3.2.3.3 Ports
The total dry cargo handled at Pakistan ports in 2003-4 was 25.2 million tons which
represents a growth of over 8 percent from the 1999-2000 [Ministry of Industries
(2005)]. The two main ports, Karachi and Port Qasim, handle 56% and 44% of dry cargo
respectively, and 18 tons of liquid cargo. The main problem facing ports is congestion at
the terminals. Facilities have not expanded sufficiently to handle the increased traffic
that resulted from trade liberalization. The major inefficiencies associated with ports
include the following [Ministry of Industries (2005)].
120

Custom clearance is cumbersome due to time-consuming manual systems, particularly
the amount of paperwork and high number of physical interactions. Also, the period of
free storage at 7-9 days is longer than international norms. Cargo is often uncollected
for long periods and has to be auctioned.
There are limited rail services for quick transportation of cargo. This is because Pakistan
Railways has difficulty in organizing a competitive freight service owing to the priority it
gives to passengers. Moreover, the freight forwarder role is not institutionalized and
leads to high freight prices. Shipping agents representing foreign carriers quote higher
prices to Pakistani exporters, without intermediation by a freight forwarder that can
negotiate bulk rates for trade.
The Dock Labour Board at Karachi port is another drag on the competiveness of exports.
The overstaffing and unnecessary labour regulations at the Karachi Port Trust raise the
cost of service to users [Competitive Support Fund (2007)].
Thus, the working of ports must be improved if Pakistan is to become more competitive
in international trade.
2.3.2.4 Policy Recommendations:
2.3.2.4.1 Roads
Revamping of the reporting structures of the National Highway Authority (NHA) and
strengthening of audits and monitoring can help in improving road quality. As poor
quality of roads does not negatively affect the revenues earned by the NHA, the
incentives for road maintenance have to be revamped. One way could be to bring in
private contractors to maintain roads under contracts that allow returns to cover costs
and yield a reasonable profit.
A roads commission is needed to develop an overall strategy for roads. It would include
in its purview connecting roads to major industrial estates, railways and border points
as well as connectivity with Balochistan and tribal areas. Major stakeholders such as the
121

Ministry of Industries and Production, Ministry of Commerce, Ministry of Railways and
provincial governments must participate in decision making.
Truckers get away with low quality haulage services due to a lack of regulation. The
market offers low barriers to entry and trucks can be financed at low interest rates. This
causes cut-throat competition and a downward pressure on fares, often below operating
costs. Overloading and poor maintenance of trucks affects both road quality and
transport service. To correct this market failure, truckers need to register and the
government must impose quality standards and encourage branding, so that with
premium attached to brands, fares will increase. In case of overloading, the magnitude
of the fine should nullify the incentive to overload.
In partnership with the private sector, logistical parks of international standard need to
be setup near industrial areas, agricultural hubs, and ports. For this purpose, land needs
to be provided, and government should invite local/foreign investors to develop
facilities for storage, loading-unloading facilities, and rest areas. Dry ports should be
established within or adjacent to these logistical parks37 . Vehicle quality testing stations
and a Radio Frequency Identification (RFID) based tracking facility should be
established within the logistical parks. The government should also ensure the creation
of cold storage facilities near agricultural hubs and expedite the creation of cold chains
from agro-based clusters to Karachi.
2.3.2.4.2 Ports
Many of the inefficiencies of ports have to do with delays at the port, high charges,
labour problems and restrictive practices in on-shore cargo handling. This results from
the port authority being run as a government monopoly. The government needs to
transition to the role of setting standards and monitoring compliance, leaving daily port
operations to the private sector. This will promote competition and result in efficiencies
such as lower service tariffs from competitive operators. Privatized container terminals
show that significant efficiency gains can be achieved [Ministry of Industries (2005)].
37
Logistical parks provide trucking space, short residences, hotels, recreation etc.
122

Port entry and maritime service charges need to be reduced. Privatization needs to be
extended beyond container terminals to operations such as the engineering department
workshop and shore handling. Preparing regular statistical reports for port traffic and
shipping, waiting time and financial indictors can help to benchmark port efficiency and
take informed decisions.
Pakistan Automated Customs Clearance System should be maintained and the software
issues need to be resolved immediately.
The government needs to facilitate private sector investment in a bulk-handling port
facility. This could be done through public-private partnerships by following the
‘landlord concept’ employed successfully in the Karachi Port Terminal. The government
should facilitate the establishment of silos at the Karachi port for the storage of
commodities such as coal and cement.
Box 2‐3: The Case of Gawadar
Gawadar is located in Baluchistan and its strategic location needs to be exploited. It is east of Iran, south
of Afghanistan, and boasts a sea port almost at the mouth of the Strait of Hormuz. Gawadar is an
essential node in the Iran-Pakistan-India pipeline, which could cross from Iranian to Pakistani
Balochistan. At the same time, Gawadar is of enormous strategic importance to China, Afghanistan, Iran
and Central Asia. However, the development and usage of the Gawadar Port has stalled as a
consequence of a myriad of issues ranging from lack of infrastructure investment by the government,
prohibitively high cost of transportation, security issues, absence of any significant industry in
Baluchistan, and foreign interests which feel threatened by its development. Accordingly, the following
policy recommendations are offered:
1. Scrap the deal with PSA and seek assistance from China in developing the port.
2. Urgently complete roadworks linking Gawadar to up-country. Currently, transporters have to go
through Sukkur.
3. Incentivize China and/or UAE to build oil refineries near the port.
4. Seek Chinese assistance in building a rail link between Gawadar and northern parts of the country.
2.3.2.4.3 Railways
2.3.2.4.4 Railways
123

According to the World Bank (2004), Pakistan Railways has the assets and potential to
do much better, especially with regards to freight. Pakistan Railways will need to
address its structural deficiencies that span institutional weaknesses and suppression of
private sector participation.
The financial weaknesses stems from an ad-hoc and unsustainable budget process that
is a mixture of deficit financing and capital injections. There is no long-term budget
strategy. The government’s monopoly discourages private capital since they are not
subject to accountability and market discipline.
The business plan for 2005-2011 that stresses private sector participation in the railways
sector in order to enhance its efficiency and competitiveness, also needs to be
implemented. New investment is required in the Railway Freight Service and new tracks
need to be laid in order to support transportation within the country. An increase in the
share of national railway in national freight from the current 4% to 22% by 2030 as
proposed in Vision 2030 [GoP Document] also needs to be realized.
2.3.3 Factor Markets Constraints on Investment & Manufacturing
2.3.3.1 Labour Market, Skills and Human Capital Development
The main problems in the labour market in Pakistan that inhibit competition are the
lack of demand for educated and skilled human labour, and a legal regulatory
framework that does not support market flexibility. More so, weaknesses in labour
legislations for minimum wage, social benefits, employment contracts and layoff
regulations remain unaddressed. The out-of-date regulations concerning maximum
hours, labour taxes and temporary contracts cause real wages to stray from productive
levels causing inefficiencies and eroding the competitiveness of firms.
In comparison with export-oriented East Asian firms (in China and Thailand), the
employment conditions are much more rigid in South Asian firms including Pakistan.
Moreover, Pakistan’s highly inflexible labor market particularly with respect to hiring
124

and firing is manifested in its low rank of 136 on the doing business indicators for
employing workers [World Bank (2009b)]. The hiring conditions in Pakistan are strict
owing to laws governing minimum wage and temporary contracts. Social security
payments associated with hiring a worker are about 12% of the salary which is
comparatively higher than in the South Asian region. Firing conditions are relatively
easier; however, an employer still has to pay the equivalent of 90 weeks of salary to
dismiss a worker, which is again high compared to that in the region [Asian
Development Bank (2008)].
Such rigidities in the labour market have a negative impact on foreign direct investment
[Asian Development Bank (2008)]. Moreover, the need for more flexible arrangement in
the labour market arises as firms are now well-integrated in global markets, trading
systems and value chains, which makes them susceptible to international macro or
industry-specific shocks. And producers can best respond to these shocks if the
manufacturing process and input and output markets are flexible.
The high levels of compliance cost issues and enforcement requirements lead to reliance
on temporary and informal markets. This is why labour demand in Pakistan is
dominated by a large informal sector and participation in the formal sector is very small.
The informal sector is dominated by a relatively less educated and unskilled workforce.
Moreover, it operates with low technology and little institutional support from the
government.
The educational foundation of the work force is weak. In fact, enrollment trends show
that half the numbers employed in firms have less than three years of education. In
Pakistan, 51% of the working population has no proper schooling and on average had
3.88 years of schooling in 2000. Bangladesh stands at 50.1% and 2.58 years, with Sri
Lanka in much better condition at 14% without schooling and an average of 6.87 years
of schooling (Barro-Lee Indicators from Barro and Lee (2000)). Moreover, small firms
usually have less educated employees than medium and large industries in Pakistan
[World Bank (2009b)].
125

Absence or weakness of skills in the labor force greatly hampers the productivity and
competitiveness of firms in Pakistan. Workers have few opportunities to acquire skills as
government agencies responsible for skill enhancement function poorly. Employers, on
the other hand, do not have incentive to invest in the training of workers owing to the
fear of losing them to competitors. As a result, firms are stuck with workers operating at
low levels of productivity.
Hence, labour market factors such as issues in labor laws, hiring and firing difficulty,
and lack of skilled labor force result in sub-optimal outcomes that constrain production,
growth and competitiveness of firms. Thus, there is a dire need to reform the labor
market, particularly to increase flexibility in hiring and firing and to lower the costs of
compliance. Moreover, in order for Pakistan to be competitive in the international
market, human resources must be developed through focusing on education and
vocational and technical training.
2.3.3.1.1 Human Capital and Skills
Upgrading worker skills is a key to stimulating investment by dynamic firms and for
improving workers’ incomes and enhancing the competiveness of firms. To be
competitive in today’s integrated world requires an increasing supply of skilled
manpower to complement rapidly changing technologies. Moreover, countries with a
broad base of skilled human labor are more likely to enhance productivity and become
internationally competitive as is evident from the case of South Korea, Singapore and
Taiwan.
A poor skill base is one of the main factors contributing to the low technological
intensity of manufactured items in Pakistan. A study by the World Bank (2004) shows
that 70% of firms identify the supply of skilled labour as an important impediment to
business operations in Pakistan. The Harbison-Myer skills indicator is a classic
skill index based on school and university enrollment. The index places Pakistan below
all South Asian countries other than Bangladesh and Nepal. Moreover, this ranking has
fallen from 69 to 77 during the period 1985-1998 ( See Table 2-12).
126

Table 2‐12: Harbison‐Myer Skills Indicators
Country 1985 (Country Ranking) 1998 (Country Ranking)
South Korea 26.8 (6) 36.1 (10)
Taiwan 22.5 (21) 27.8 (23)
Singapore 14.8 (37) 23.1 (29)
Thailand 10.8 (48) 15.6 (45)
Malaysia 9.2 (51) 11.1 (55)
Sri Lanka 10.1 (58) 9.1 (53)
China 9.8 (59) 5.2 (67)
India 7.2 (60) 8.1 (69)
Nepal 6.4 (71) 5.4 (66)
Bangladesh 4.0 (72) 4.3 (76)
Pakistan 4.4(69) 4.2(77)
Source: Ministry of Industry (2005)
Moreover, labour productivity growth rates for Pakistan’s industrial sector remained at
1.48 percent from 1992 to 2001 which is far less than other South Asian countries such
as Bangladesh and India that stand at 1.52 and 5.05 percent respectively [Japan
International Cooperation Agency (2006)].
Since the inadequacy of skills is a major constraint to the growth and competitiveness of
many sectors in Pakistan (particularly those that manufacture exported goods), it is vital
to expand the skilled labour base.
2.3.3.1.2 Basic Education
Education augments the productivity of a worker by enhancing cognitive skills.
Literature shows that returns to education increase with the level of education.
According to Kingdon and Soderbom (2007), returns to education monotonically
increase with education so that an extra year of education is progressively more valuable
when attained at successively higher levels of education. Haque (1977) also finds
positive and higher returns to education at higher levels of education. According to a
study conducted by PIDE on the returns to education in Pakistan, each year of
additional schooling results in a 7 percent increase in returns for wage earners [Nasir
127

and Nazli (1998)]. In Balochistan, rewards are higher for tertiary education whereas in
Punjab they are higher for secondary education [Jamal et al. (2003), Shabbir and Khan
(1991)] find different inter-provincial returns to schooling by using data on male wage
earners. At the national level, the returns to schooling are 9.1% whereas for Punjab and
Baluchistan returns stand at 9.9% and 4.4% respectively. According to Nasir and Nazli
(2000), gender, regional, provincial, skilled/unskilled, public/private dimensions have
significant impact on earnings. Graduates of private schools earn 31 percent higher than
public schools counterparts. More so, workers with literacy and numeracy skills receive
wages that are roughly 15% higher than workers without these skills [Ministry of
Industries (2005)].
Despite the higher returns associated with education, an average Pakistani worker
receives about half the average years of schooling that is received in East Asia and
around 80% of the years of schooling provided to workers in other South Asian
countries [Barro and Lee (2000)]. In fact, Pakistan ranks last in both primary and
secondary education enrollment in South Asia. Secondary schools in Pakistan serve only
two fifths of the population (PIHS, 2001-02). In 2002, even though net enrollment
improved slightly at both the secondary and matriculation level, it declined for primary
education. According to official statistics, gross enrollment in 2005-06 was close to
90%, however there is a sizeable drop as we move from the primary level to the
secondary level [See Figure 2-32]. However, according to the World Bank (2009), only
42% of children in Pakistan are enrolled in primary school and of those only two-fifths
make a transition to secondary school, with secondary school enrollment standing at
16%. Enrollment at matriculation then is only 9 percent.
2.3.3.1.3 Basic Education
Education augments the productivity of a worker by enhancing cognitive skills.
Literature shows that returns to education increase with the level of education.
According to Kingdon and Soderbom (2007), returns to education monotonically
increase with education so that an extra year of education is progressively more valuable
when attained at successively higher levels of education. Haque (1977) also finds
128

positive and higher returns to education at higher levels of education. According to a
study conducted by PIDE on the returns to education in Pakistan, each year of
additional schooling results in a 7 percent increase in returns for wage earners (Nasir
and Nazli, 1998). In Balochistan, rewards are higher for tertiary education whereas in
Punjab they are higher for secondary education [Jamal et al. ,(2003). Shabbir and Khan
(1991)] find different inter-provincial returns to schooling by using data on male wage
earners. At the national level, the returns to schooling are 9.1% whereas for Punjab and
Baluchistan returns stand at 9.9% and 4.4% respectively. According to Nasir and Nazli
(2000), gender, regional, provincial, skilled/unskilled, public/private dimensions have
significant impact on earnings. Graduates of private schools earn 31 percent higher than
public schools counterparts. More so, workers with literacy and numeracy skills receive
wages that are roughly 15% higher than workers without these skills [Ministry of
Industries, (2005)].
Despite the higher returns associated with education, an average Pakistani worker
receives about half the average years of schooling that is received in East Asia and
around 80% of the years of schooling provided to workers in other South Asian
countries [Barro and Lee, (2000)]. In fact, Pakistan ranks last in both primary and
secondary education enrollment in South Asia. Secondary schools in Pakistan serve only
two fifths of the population (PIHS, 2001-02). In 2002, even though net enrollment
improved slightly at both the secondary and matriculation level, it declined for primary
education. According to official statistics, gross enrollment in 2005-06 was close to
90%, however there is a sizeable drop as we move from the primary level to the
secondary level [See Figure 2-32]. However, according to the World Bank (2009), only
42% of children in Pakistan are enrolled in primary school and of those only two-fifths
make a transition to secondary school, with secondary school enrollment standing at
16%. Enrollment at matriculation then is only 9 percent.
Figure 2‐32: Gross Enrolment Trends in Pakistan, 2001‐2006 (percent)
129

Source: PIHS (2001-02 and 2005-2006)
In the year 2003-2004, around 77% of the workers employed in the manufacturing
sector had education level below matriculation [Government of Pakistan (2004b)].
Across firm sizes, small scale industry employs the least educated in terms of number of
years of schooling followed by large scale industry. Interestingly, medium sized
industries employ workers with relatively greater number of years in school (see Figure
2-33). This highlights the need for expanding the educated workforce for raising the
productivity of firm in the manufacturing sector and hence to make them internationally
competitive.
Figure 2‐33: Average Education Attainment of Workers in Pakistan
100%
80%
60%
40%
20%
0%
Small Medium Large Avg. mfg. Avg. service
Source: PIHS (2001-02 and 2005-2006)
130
13+ yrs
7-12 yrs
4-6 yrs
0-3 yrs

2.3.3.1.4 Vocational Training
There are considerable benefits from investment in training, such as improvement in
worker skills, increase in entrepreneurial activity, facilitation of workers in adapting to
technological change and increase in the rate of movement of workers into more
productive sectors [Middleton et al. (1993)].
Formal training institutes have been set up in Pakistan with the purpose of broadening
the skilled labor base. Such training institutes have the capacity to train over 200,000
workers, which is about 1.2 percent of the 17 million population that drops out from
basic education [Ministry of Industries (2005)]. However, there is regional disparity in
the number of technical education and vocational training institutes across provinces;
Punjab has 287, Sindh has 156, NWFP with 55 and Baluchistan only 14 [Japan
International Cooperation Agency (2006)].
The technical and vocational education and training (TVET) systems in Pakistan remain
supply-driven and lacking a national direction. This leads to a shortage of skilled
manpower that decreases total factor productivity of the manufacturing sector. In order
to become competitive in the global market, it is necessary to reform the TVET system,
in particular, change it from a supply-driven system to one that is connected to market
demand. The National Vocational and Technical Education Commission, established in
2005, intends to transforms the existing supply-driven TVET system to a demanddriven
one. The Commission also aims to assist provincial governments in development
of skills, create Public Private Partnership (PPP’s) so that the private sector can clearly
indicate the type of skills they are seeking and coordinate national sectoral training
policies [The International Monetary Fund (2010a)]. There are also a number of privatepublic
initiatives. They include the Pak-Swiss Training Centre and the Textiles Institute
of Pakistan.
According to World Bank (2000), the only significant variable that increases labour
productivity is training programs. Moreover, Nasir and Hina (2000) identify the return
to training in Pakistan as ranging from 2.5 to 4 percent. Nonetheless, most firms in
131

Pakistan do not consider inadequate skill level to be an obstacle in growth even though
evidence shows that technical education and training increase worker earnings. Due to
this belief, very few Pakistani firms provide formal or technical training to their
employees as compared to other countries. And in manufacturing sector, on average,
barely 5 percent firms provide technical training (see Figure 2.34).
Figure 2‐30: Provision of Training to Workers
International comparison Type of firm
South Africa
Philippines
Brazil
Egypt
Sri Lanka
Turkey
India
Chile
Bangladesh
Pakistan
Source: World Bank (2009b
0% 10% 20% 30% 40% 50% 60% 70%
Moreover, training is subject to market failures. The costs of training an employee are
sunk costs. Because workers often leave their jobs for better ones, employees underinvest
in training. To add to this, a large proportion of Pakistani labour is hired on an
informal contractual basis. This further lessens the likelihood of firms recovering their
training costs.
Since technical training by enhancing skills of workers can make a potent contribution
to industrial productivity, there is utmost need of addressing the structural weakness in
the technical and vocational education and training systems.
132
Avg.
service
Avg. mfg.
Large
Medium
Small
temp workers
regular workers
0% 10% 20% 30% 40%

2.3.3.2 Policy Recommendations:
2.3.3.2.1 Increased Provision of Primary and Secondary Education across the country
Basic education is the right of every individual; therefore it is the obligation of the
government to ensure universal provision of quality education at the primary and
secondary level. The first broad policy intervention required by the government is to
increase the allocation of budgetary resources towards primary and secondary
education, with the long term aim of achieving universal literacy in the country. Along
with provision of basic education, the quality of education needs to improve significantly
through improvement and updating of curricula, teacher training programs, regulation
of educational and teaching standards in private schools (especially in smaller towns
and rural areas where these proliferate) and standardization of education across both
private and public schools.
The disparities across and within provinces in terms of both enrollment and literacy
rates have to be addressed through increased provision of functioning primary and
secondary schools by the government in areas which have a higher incidence of
illiteracy.
2.3.3.2.2 Upgrading Vocational Training and Skills
Many vocational training institutes have been established, yet their delivery mechanism
and policy goals need to be revised. The general strategy in the last few years has been to
encourage private sector partnership through government grants. The requirements of
the workplace can usually be best determined by the employers themselves. In Pakistan
employers play a negligible role in influencing what is taught in TVET institutes.
In 2009, the National Vocational and Technical Education Commission (NAVTEC) was
established to overcome the lack of standardization, skill gaps, non-availability of proper
133

curricula, poor quality of instructional staff, inadequate accreditation, poor
infrastructure and low private sector technical education and vocational training
capacity. NAVTEC is a regulatory body responsible for long-term planning in this
particular field. It will also be responsible for setting standards for formulating the
syllabus, accreditation, certification and trade testing, etc.
In 2009, NAVTEC gave PKR2000 per month to each trainee during its training
course38 . Presently, 1522 technical institutes with an enrollment of 314,188 are
operating in the country, providing technical skills to the labour force. The goal is to
produce one million skilled laborers per year [National Vocational and Technical
Education Commission (2009)].
In order to develop a skilled labour force along modern lines, the Labour and Manpower
Division has established five Skill Development Councils (SDCs) in Islamabad, Karachi,
Lahore, Peshawar and Quetta. The SDCs assess the training needs of their geographical
areas, prioritize them on the basis of market demand and facilitate training of workers
through training providers in the public and private sector. These Councils have met the
diversified training needs of the industrial and commercial sectors and have so far
trained 46,674 workers [Mumtaz, K. et al (2010)].
In 2006, TEVTA made changes in its loan policy so as to provide loans to its training
institutes that could only be used on books, equipment and student transport and not on
real estate or bank deposits. In June 2010, the government moved to make TEVTA
functional and proposed to give seed money of PKR15 million, with the rest of the
amount to be generated from TEVTA’s own resources39 .
38 http://www.molm.gov.pk/ -> Labor and Manpower Division -> Information and Services -> The steps
taken by the Government to curtail unemployment in the country
39http://archives.dawn.com/archives/79036
134

The 2009-2013 NAVTEC Skills Strategy is the first attempt at a formal policy direction
for TEVT activities. The vision is “Skills for Employability, Skills for All.” The strategy
aims at a shift from time-bound, curriculum-based training to flexible, competencybased
training as well as a shift from supply-led training to demand-driven skills
development by promoting the role of industry in both the design and delivery of TVET.
The main objectives are to provide relevant skill, improving access, equity and
employability while assuring quality [National Vocational and Technical Education
Commission (2009)].
The reform agenda includes establishing industry specific Centres of Excellence,
increasing the role of the private sector, reforming the apprenticeship system,
encouraging entrepreneurship, making training delivery flexible, integrating informal
economy workers, enhancing worker mobility, providing career guidance and placement
services, registering and accrediting institutes and reforming the management of
training institutes. These initiates when launched at the sector level will trigger
industrial competitiveness.
The action plan of the government to achieve these objectives is to strengthen
institutional linkages with industry, as training is designed around skills and knowledge
that are not necessarily relevant to the market. To link the industry and government, it
is proposed that sector specific Industry Advisory Groups (IAG) be established. Each
IAG will be represented by members of large, medium and small industry, including all
sub-industries that fall within the category, international employers, and employees.
Their responsibilities will be to carryout periodic sector surveys, identify skill needs in
their sectors, indicate new and emerging areas and occupations and determine and
update competency standards for workers. IAGs will be established gradually over time,
with the aim to establish two each year until all major sectors are represented.
Under the National Skills Strategy, the government proposes to promote and facilitate
the establishment of sector-specific training institutes and Centres of Excellence. These
institutes are to be located in proximity to relevant industry in order for them to benefit
from new technologies, and to facilitate industry placement of trainees and trainers and
135

also to enhance information flows between the training institute and industry. Where
possible, NAVTEC will help Pakistani Centres of Excellence establish partnerships and
links with reputed industry-specific training facilities abroad.
Training by firms is usually workplace-based and therefore, practical and relevant to
industry needs but it does not lead to any nationally recognized certification. Therefore,
though workers gain skills, their skills are not formally recognized beyond the
boundaries of the firm. The hope is that leading firms may consider sponsoring a
training unit within a public sector institute, where trainees can be trained specifically
on their machines and according to their requirements and standards. Such a model
may also be useful for clusters of industries which do not have their own training
establishments and have difficulty financing training individually.
The following are two specific programs which can be piloted by the government to
incentivize the firms to train unskilled workers.
Skills Development Fund
Upgrading worker skills requires more than just vocational and technical training.
Worker training that is led by firms themselves has an impact on firm-level productivity
like in several East Asian and Latin American economies. The problem is that there is a
market failure, that firms incur costs while training. These sunk costs cannot be
retrieved when workers leave their jobs. Thus firms tend to under-invest in training
employees.
The under-investment effect is magnified when there is imperfect information about
both the productivity value of training and the skill requirements associated with the use
of new technologies. To encourage firms to invest in upgrading worker skills, there
should be the creation of a Skills Development Fund financed though a payroll levy
grant system (Figure 2-35). A small compulsory levy of about 1% payroll outlay is
imposed on firms, and is matched by the government. Firms are issued vouchers
against the contribution they make to training institutes in exchange for the required
training. The training institutes cash their vouchers with the Skills Development Fund.
136

Figure 2‐34: Payroll Grant System
137
Source: Ministry of
Industries, 2005.
Implementing
such a system will
require extensive
consultation with
the private sector
in order to ensure ownership and decrease private sector perception of an increased tax
burden.
Such initiates have been successful in Malaysia and South Africa. Two successful
training centers in Pakistan include the Pak-Swiss Training Centre (PSTC) and the
Textile Institute of Pakistan (TIP). The PSTC provides training in Precision Mechanics
and Instrument technology and its mandate is extended to the private sector. TIP was
established under public funding as a private not-for-profit institution to meet the acute

shortage of qualified textile professionals. Its board comprises mostly private sector
professionals.
Skill based Wage Subsidy Scheme
In a skill based wage subsidy the government would pay a certain proportion of the
stipulated minimum wage of an unskilled worker (see Figure 2-36). For example, the
firms could pay sixty percent of the minimum wage to the worker, whereas the
remaining forty percent would be the government’s contribution payable through the
training institute directly to the worker. The size
Figure 2‐35: Skill Based wage Subsidy Scheme
Continuous
feedback &
monitoring
Factory
Employ Fresh
Workers to Train
‘on the Job’ (4 days
a week)
60% of
Wage Bill
Government
Factory registers
Fresh Workers at
the Institute
Worker
138
40% of
Wage Bill
40% of
Wage Bill
Sector Training
Institute
Provide class room
based training (2
days a week)
of the subsidy would be in accordance to the skill premium. This subsidy would be tied
to both, on the job training received by the worker, and, classroom based instruction at
the local training center. To claim the subsidy the firm or employer would have to
register the worker at the training center. With the subsidy in place the employer would
have an incentive to employ, register and train the worker. The training would be on the
job and hence would be relevant to the technology and skill set required for the specific
job.
2.3.3.2.3 Higher education – University‐ Industry Linkages

There is need for an effective linkage between industry and academia. Industry needs to
be the driver of higher education in engineering and science. Engineering and science
degrees need to be developed around industry needs. Following are some proposals to
effectively enhance the university-industry linkage:
1. Science and engineering universities need to introduce new courses and degrees
in view of local industry needs.
2. With collaboration from industry, state-of-the-art laboratories should be
established in universities to meet the R&D needs of business.
3. Government should sanction USD10 million for joint research projects involving
universities and industry, based around actual industry problems.
2.3.3.3 Capital Market
Capital markets help mobilize savings and raise investment funds in excess of the loans
banks are willing to provide to firms. Moreover, they help mitigate risk by spreading it
across a large number of share owners.
The major sources of credit in Pakistan are banks and trust funds at 67% of assets
[Japan International Cooperation Agency (2006)]. The Central Directorate of National
Savings comes second at 22%. Seven percent is provided by non-bank financial
institutions like development funds, investment banks, leasing, Modarba companies,
trust funds, venture capital etc. The remaining 4% comes for insurance [Japan
International Cooperation Agency (2006)].
A detailed breakdown of lending categories in Pakistan shows that the fastest growth in
credit is in the personal loan category. As evident from Figure 2-37 below, banks seem to
have devoted much of their resources in expanding the personal loans portfolio rather
than lending to private enterprises and industry.
139

Figure 2‐36: growth in lending by Borrower Category in Pakistan, 2001‐2007 (base Year 2001)
Source: World Bank (2010b)
The financial market in Pakistan is shallow as firms rely on retained earnings to finance
their working capital and investment needs (see Figure 2-37 below). Trade credit and
banks plays a smaller role in working capital financing; Pakistan’s firms rely more on
internal funds. The pattern of funding using retained earnings is robust; ranging from
88% in NWFP to 78% in Sindh [World Bank (2009b)].
Credit rationing in the formal credit market is a major constraint faced primarily by
small- and medium-sized firms (see Figure 2.38) that constitute a significant portion of
the industry. A study conducted by the Lahore University of Management Sciences
(LUMS) in 2006 also highlighted that access to finance is the most critical impediment
to the growth of the SME sector. A survey found that 57% of new investment by SMEs
and 67% of their working capital comes from retained earnings, whereas according to
studies conducted by international organizations, banks provide only 7–8% of the total
funding requirement of SMEs [The International Monetary Fund (2010a)]. This is due
to constraints on both the demand and supply side. On the supply side these are: weak
and poorly enforced creditor rights, high unit costs of SME lending, and the reputation
of SMEs as riskier borrowers in the financial system. On the demand side the
constraints consist of loan disbursement procedures, high interest rates and collateral
140

equirements that raise the cost of access to credit [Bari, Cheema and Haq (2005)]. As a
result of these financial constraints it is difficult for micro enterprises and SMEs to grow
in size and scale. The resultant diseconomies of scale and inaccessibility to efficient
technology of production are the main factors behind their low productivity, which
inhibits their growth potential.
Figure 2‐37: Access to Finance by Firm Size
By applying for loans/lines of credit
Large
Medium
Small
0.0 0.1 0.2 0.3
Source: World Bank (2009b)
By result of loan applications
2.0
1.6
1.2
0.8
0.4
0.0
Information asymmetries and the fear of adverse selection are factors that restrict
commercial bank lending to SMEs. This is because the information that SMEs can
provide to banks in the form of financial accounts, business plans, feasibility studies,
etc. often lack detail and rigor. This problem is aggravated by the low level of education
of small entrepreneurs. In a sample surveyed by Pasha, 35 percent were either illiterate
or educated to below the matriculation level and unable to adequately articulate their
case [Department for International Development (2010)].
In order to mitigate the problem of asymmetric information and adverse selection,
banks have adopted precautionary measures such as requiring that financing be
collateralized, with collateral often exceeding 100 percent of the loan. In Pasha’s sample,
141
No. of times applied No. of times rejected
Small Medium Large

the average value of the collateral was higher, the smaller the business. This, however,
leads to other issues as SMEs generally do not have many assets that can be
collateralized; for example, over 40 percent of the SMEs in Pasha’s survey admitted to
not receiving a bank loan because they lacked the necessary collateral. Moreover, it is
difficult for banks to take possession of the collateral though legal procedures given the
slow working of Pakistan’s judicial system. The World Bank team for the 2005 report
was told by banks that the average period for taking custody of the collateral, in
instances when they were able to do so, could be 4–5 years or more after default had
been declared, and the time and costs of pursuing the process of obtaining possession
kept mounting.
Moreover, most commercial banks regard SMEs as riskier than large firms due to many
reasons. Firstly, SMEs face a more uncertain competitive environment compared to
larger companies; they experience more variable rates of return and higher rates of
failure. Secondly, SMEs are less equipped in terms of both human and capital resources
to withstand economic adversities, particularly unanticipated economic shocks. Thirdly,
their inadequate accounting systems and lack of financial controls undermine the
accessibility and reliability of information on profitability and repayment capacity.
Finally, SMEs in Pakistan operate in a somewhat ambiguous environment regarding
governance, which reduces the security of transactions. Thus there is a greater risk that
banks will not get paid, or that assets such as property will not be properly registered.
In view of all these difficulties, banks frequently refuse to lend at all or severely restrict
the size of their loans to SMEs. It is, therefore, not surprising that 86 percent of the
SMEs surveyed by Pasha had not acquired a bank loan. Moreover, in the total private
sector credit disbursement of PKR1,669.740 billion in FY2007, the share of the SME
sector was a shocking 1.7% [Asian Development Bank (2008)].
40 State Bank of Pakistan. (2007). SBP Annual Report 2006-07, Vol I. Karachi
142

2.3.3.4 Policy Recommendations:
The State Bank of Pakistan, under its Financial Inclusion Program (FIP), recently
launched two programs to enhance the flow of credit to small and rural enterprises. The
Financial Inclusion Program was launched in 2009 in collaboration with UK’s
Department for International Development (DFID) in order to promote inclusive
economic growth through the provision of sustainable financial services to the poor,
small entrepreneurs, women and marginalized communities. The government should
help in disseminating information about this particular scheme through the various
business/industry associations representing SMEs around the country.
The first program launched under FIP is the Credit Guarantee Scheme (CGS) for Small
and Rural Enterprises in which the SBP will share 60% of the losses from short-term
and medium-term loans (up to a maximum of PKR5 million per loan) extended by
banks to SMEs. Each loan will have a tenure of up to three years and will have an
interest rate equivalent to the 3-month Karachi Inter Bank Offer Rate (KIBOR) rate plus
300 basis points (bps). The initial capital for this scheme is to be provided by DFID41 .
The main problem with the CGS scheme is ‘moral hazard’ on the part of both the
borrower and the lender. As the loss to the borrower from defaulting is mitigated by the
guarantee given by the State Bank, the incentive of paying back the loan is reduced. On
the other hand the lending commercial bank does not have the incentive to rigorously
scrutinise borrowers and vigorously follow up on the credit disbursed because of the
loan repayment guarantee provided by the State Bank.
The second program is the Refinance Scheme for SMEs in NWFP, FATA, Gilgit and
Baluchistan, with the stated purpose of increasing the supply of credit to SMEs. The
interest rates for this program will be floating based on a formula linked to the rates on
6-month Treasury Bills and Pakistan Investment Bonds (PIBs).
41 http://www.sbp.org.pk/press/2010/CreditSchemes-19-Mar-10.pdf
143

The procedural requirements such as excessive documentation of loan application by
commercial banks are perceived as a major hurdle by SMEs. The government (MOI)
should advocate the State Bank to facilitate the development of a uniform simplified and
efficient procedure for loan disbursement to small and micro level enterprises. This will
substantially reduce transaction costs for SMEs and increase their incentive to apply for
loans.
Information asymmetry is the principle factor behind lack of access of SMEs to credit, so
it is imperative to have a Credit Rating Agency specialising in the SME sector. The State
Bank has successfully launched the Pakistan Credit Rating Agency, however it does not
cover the SME sector. An agency similar in scope but targeted towards the SME sector
should be supported by the government
The government should recommend to the State Bank to promote Venture Capital
Funds in Pakistan in order to stimulate small-scale entrepreneurship in the country and
also rectify credit market failures present in the formal commercial banking system.
Venture Capital Funds do not only provide capital but also support nascent enterprises
and entrepreneurs with technical and managerial support. Such support is crucial for
the success and survival of a new business.
The State Bank along with the Ministry of Industries should provide information on
investment opportunities in the country and facilitate investment by expatriate
Pakistani’s both directly and through such Venture Capital funds.
Invoice-based financing, where the borrower gives accepted invoices (or receivables) of
its business customers or downstream buyers, as collateral to the commercial bank
should be promoted by the State Bank as an alternative to collateral based lending.
2.3.3.5 Land Market
Land represents a principal asset for manufacturing firms. In Pakistan, however,
inefficient and archaic land registration procedures and transfer laws have an adverse
effect on investment and growth in the manufacturing sector.
144

According to Ellis and Dowell (2009), present land markets in the province of Punjab
for example, are fairly dysfunctional. They highlight a list of problems faced in the land
and housing market. The list includes: excessive public ownership,
inadequate infrastructure services, weak property rights, pervasive public and private
sector rent-seeking, counter-productive urban planning policies and regulations, costly
construction regulations, limited financing for property development and acquisition,
rent controls and inadequate property tax based revenue-generating mechanisms.
Land acquisition is a cumbersome process that results in delays and high costs.
Moreover, it is extremely costly to impose penalties if a party fails to abide by the
contract. Also, difficult regulations relating to property registration encourage formal
titles to remain in place and do not facilitate land being put to best productive use. In
other words, high costs give incentive to entrepreneurs to undervalue their property or
evade registration altogether. According to the Doing Business in 2005, data of the
World Bank, it takes 49 days to register property and the cost of this is 4.2% of the
property.
Numerous studies have pointed out the deficiencies owing to which land property rights
are not clearly defined in Pakistan. These include multiple agencies involved in land
registration and transfer, complex record keeping, and sale transactions without valid
conveyance documents. These factors, by effectively removing land from the pool of
acceptable collateral in many cases, retard finance access to firms. As the World Bank
study (2009) pointed out “without clear property rights, lenders will not consider
collateral as loan security without an original sale deed in the bank's possession. The
resulting ‘dead capital’ [i.e., capital that cannot be used for investment, collateral, or in
some other productive purpose] . . . hinders leveraged investment, firm level entry, and
efficient resource allocation.” Also, banks prefer using land titles as collateral as they
cannot be moved or hidden [World Bank (2009b)]. Therefore the establishment of
secure property rights can provide incentives for asset owners to actively invest in their
property and can increase the effectiveness of land being used as collateral.
145

2.3.3.6 Policy Recommendations:
The current land registration laws present certain elements that make the acquisition
process lengthy and often require high informal costs. In addition to the formal land
registration and land tenure system, there are strong and widely used informal land
transfer systems, such as oral commitments, collateral assurances, powers of attorneys
etc., that operate outside the formal land registrations. The existence of these parallel
systems is said to have created serious complications by providing shelters to land
mafias, registration fee evaders, and public sector rent seekers.
An even more critical issue in terms of land availability is the "title of record" dimension
of land parcels. Under current legislation, the records of rights in land primarily reflect a
fiscal responsibility. The person mentioned in the records is liable for land revenue or
property tax; that he is also the rightful owner is only incidental. The land registration
documents only have persuasive, not conclusive, evidentiary value.
In light of the problems stated above, the following steps should be taken immediately:
1. Reform of the Land Registration Act is imperative in order to provide security of
titles.
2. There is an immediate need to computerize land records. Therefore the provinces
should complete computerization of all land records as soon as possible.
3. There is an immediate need to computerize document land records. Therefore the
Provinces should complete computerization of all land records as soon as
possible.
The provinces are on board with these recommendations, in fact some reform work has
already started in Punjab and Sindh.
2.3.4 Standards, Regulation & Governance Constraints
146

2.3.4.1 International Standards and Quality Compliance
The use of standards and technical regulations to measure quality has become an
inevitable need for international trading. The use of standards contributes to
technological progress by promoting quality and increasing competition. Under the
WTO, countries are allowed to decide on national standards in a way that their
enforcement does not discriminate between domestic and imported goods. Compliance
implies upgrading production methodologies to international standards and satisfying
conformity assessment procedures.
Institutions for quality control and standardization have existed in one form or the other
in Pakistan since 1950. For example, the Standards Institution (PSI) was established in
1951 under the Ministry of Industries, with the objective of formulating national
standards. A Testing Laboratory (CTL) was also established in 1951 under the Ministry
of Industries to undertake the testing of industrial raw material. However, it was in the
1980s that the need for a certified quality management system became apparent in
Pakistan. Thus the ISO-9000 Quality Management System was introduced and was a
good confidence-building measure.
A number of other certification agencies entered the country in the late 1990s. Around
10,000 firms are now certified. The Export Promotion Bureau and Ministry of Science
and Technology joined hands with the Pakistan Institute of Quality Control (PIQC) to
work towards better certification and international quality control. However, there are
concerns regarding its effectiveness [Shah and Shariff (2010)]. Since 2000 onwards,
sector-specific Quality Management Standards like TS-16949, ISO – 22000, AS9100B,
etc. have been implemented.
However, the pace of development of quality in the government itself is presently not
very strong. Very few original standards in line with the needs of the local industry were
developed and a greater part of the productive sector is not acquainted with these
standards. For this purpose, a national accreditation body- the Pakistan National
Accreditation Council (PNAC)- was formed to be responsible for qualifying and
147

approving the certification bodies, consultants and trainers. PNAC is a signatory with
International Laboratory Accreditation Cooperation (ILCA) and Asia Pacific Laboratory
Accreditation Cooperation. Along with the Pakistan Standard and Quality Control
Authority (PSQCA) and the Pakistan Council for Scientific and Industrial Research
(PCSIR), the PNAC deals with issues of technical barriers to trade (TBT).
PCSIR, established in 1953, provides ISO 9000 certificates. PCSIR labs offer testing for
exporters but the facilities are inadequate for exporter needs. Over the years the PCSIR
research budget has also shrunk, and no significant research is being conducted
[Ministry of Industries (2005)]. PCSIR also lacks a focus on industry and rarely attracts
contractual work from the private sector, which could potentially improve PCSIR’s
financial problems. In Pakistan compliance with quality standards is highest in the
exporting sector. With textile leading the manufacturing sector with 90% of the imports,
it is the most aware and compliant domestically [Ministry of Industries (2005)].
However, it falls short in international standards of working conditions, child labour,
and environmental standards. There is also a limited ability for measurement,
standards, testing and quality. Services in Pakistan that can ensure these are also costly
and difficult to access.
The private sector has found alternatives by developing in-house techniques by
approaching private testing institutions or going abroad to higher quality testing
initiations. Furthermore investment in testing and accreditation to avoid TBTs is
substantial and prevents the exporting sector expansion. There is a need to expand the
current network of labs and testing facilities. Public research institutes with testing and
certification potential should focus more on developing state of the art testing and
accreditation, rather than engaging in commercial research.
The other issue of standardization is that of environmental protection. Health costs for
Pakistan due to water pollution in 2003 were estimated between US$1.7 billion and
US$4.7 billion [Ministry of Industries (2005)]. Specific industries such as leather,
textile, paper and board, sugar and polyester cause more pollution. This is an area of
148

concern since cotton and leather together account for 70% of total exports from the
country [Government of Pakistan (2004a)].
Laws in place fall under the Environmental Protection Act 1997, which establishes
federal and provincial Environmental Protection Agencies. The act makes pollution
control a priority and focuses on industrial waste water discharges and air and car
emissions. However, regulation relating to effluents in water bodies have not been
enforced [Ministry of Industries (2005)]. There are other laws dealing with industrial
pollution under the Act but they are qualitative and so have limited potential for
implementation.
2.3.4.2 Governance and Corruption
One measure of good governance is the consistent and predictable interpretation of
rules by the government officials. However, this area is considered to be a problem in
Pakistan as only 46% of firms believes that rules and regulations are consistently
applied by the officials [see Figure 2-39] which is worse than that in comparator
countries which average at about 60% of firms. Nationally, this perception is weakest in
Baluchistan. Similarly small firms are more disillusioned with the interpretation of
rules.
Figure 2‐38: Comparison of Consistent and Predictable Interpretation of Rules
By firms size and province By international comparison
149

Balochistan
Sindh
Punjab
NWFP
Small
Medium
Large
0 20 40 60 80 100
Source: World Bank (2009b)
Corruption has been seen as another major impediment to private sector activity in
Pakistan. Bribe incidence in Pakistan has increased from 40% in 2002 to 48% in 2007
[World Bank (2009b)]. According to the Doing Business Indicators (2009), around 57%
of the firms consider it to be a severe problem in Pakistan and this percentage is second
only to Brazil and Bangladesh (60-70 %) while all the other comparator countries rank
better on this indicator. More so, large firms are more susceptible to corruption, where
the percentage rises to 78% [World Bank (2009b)].
2.3.4.2.1 Crime and Security
The poor law and order condition is another impediment to industrial activity in
Pakistan. The problem is impacting industry at two levels. The security concerns have
created a poor perception of Pakistan in international markets and the buyers in these
markets have become strongly skeptical about Pakistan’s ability to supply consistently.
This has resulted in loss of several large orders where buyers have refused to work with
Pakistani companies. Secondly, the domestic crime situation and ineffectual role of the
150
Brazil
Pakistan
Philippines
Turkey
S. Africa
India
Sri Lanka
Chile
Average
0 20 40 60 80 100

‘thana’ impedes the activity of the private sector42. The private sector is insecure about
life, property and assets. The failure of the police to inhibit crime has resulted in private
sector reducing their economic activity in the country. Based on an earlier World Bank
Survey, the percentage of firms considering law and order to be a major problem
increased from 22% in 2002 to 35% in 2007.
2.3.4.2.2 Smuggling & under invoicing
The Afghan Transit Trade Agreement is a serious issue for industry. A range of goods
which are officially imported for the Afghan market make their way into Pakistan
through well-established and strong smuggling networks operating on both sides of the
Afghan-Pakistan border. Prices of these smuggled items in the local market are much
lower than those of local manufactures, resulting in an erosion of local market shares
and loss in revenue43 . One way to control smuggling is to increase the transaction cost of
smuggling and reduce the wedge between importable prices of legal and smuggled
goods.
The same issue is being faced by significant amount of under-invoicing of imported
goods. The importers normally import products at very low prices to avoid paying high
amounts of duty. This evasion of duty leads to lower costs of acquiring the goods and
hence it becomes extremely difficult for the local industry to compete with these
imported products. In some cases the manufacturers in the local industry are able to
convince the government to set a fixed ‘international trade price’ (ITP), however, in
most cases the import lobby is able to influence the determination of ITPs in their favor.
For example, the team working on this report was told by a local tyre manufacturer that
the ITPs on some types of radial car tyres was set at a value lower than the material cost
42 Firms due to the general prevalence of crime devote large resources to security expenses, even though
not all of them suffer such losses. 50% of firms allocate resources to security, while only 10% actually
experience incidence of crime. Such needless expenditure on security is detrimental as it results in
inefficient allocation of resources which otherwise could have been put to more productive use. Therefore,
the environment needs to be made more secure so that firms will not devote large sums of money towards
security and instead increase their productivity and competitiveness.
43
Field interviews conducted by team under the current project.
151

of manufacturing the tyre. Such inadequate governance and corruption is damaging the
industrial activity of the country.
2.3.4.3 Regulatory Constraints
2.3.4.3.1 Functioning of Courts
A well functioning court system is essential for creating an environment conducive to
investment and business transactions. In many developing countries, a common
impediment in doing business is the absence of efficient courts. Consequently, 80% of
people opt for informal institutions in order to seek justice [Wojkowska (2006)].
Even in Pakistan, the functioning of courts is a significant problem for over one third of
the firms [World Bank (2009b)]. In fact, according to the World Bank’s Enterprise
Surveys, firms in Pakistan identified courts as one of the top ten constraints to
investment, as important as political instability and access to finance in 200744 . At the
provincial level, almost 66% firms in Sindh find the functioning of courts to be more of a
problem. Sindh is followed by Baluchistan, NWFP and then Punjab. With these poor
perceptions about courts, many firms avoid using the judicial system to settle disputes.
World Bank studies [The International Monetary Fund (2010a)] cite a backlog of more
than 100, 000 cases in the High Courts of Sindh and Punjab, and estimate that when
these cases were taken up by the courts, it took an average of 46 steps. A significant
proportion of such cases took more than 5–7 years to be decided, whereas appeals and
implementation of the judgment took even longer. This could be a reason why the use of
a third party to settle disputes has risen by 70% [World Bank (2009b)].
Hence, formal adjudication systems are often unable to settle disputes in a timely and
cost-effective manner in Pakistan that can act as a severe disincentive to investment
particularly foreign investment. Moreover such slow and costly judicial procedures
44 World Bank. Enterprise Surveys. Pakistan Country Profile. Available at http://www.enterprisesurveys.org
152

create unpredictability, a low trust environment, resulting in market segmentation and
wasteful litigation.
In June 2009, the National Judicial Policy came into effect in Pakistan to tackle the
shortcomings in the judicial system. The policy, in particular, aimed to ensure an
independent judiciary, clear the backlog of cases, reduce the time period for disposal of
cases and exterminate corruption. However, since the policy is at its inception stage it is
too early to measure its effectiveness across the nation [World Bank (2010a)].
Contract Enforcement
An efficient and effective legal system, particularly enforcement of contracts, is vital for
functioning of a market economy. This is because lack of procedures for effective
contract enforcement restricts contestability, restrains corporatization, increases project
risk, reduces technology transfers and may hamper the development of the financial
system [Department for International Development (2010)]. As Hobbes puts it in 1651
“He that performeth first has no assurance that the other will perform after, because the
bonds of words are too weak to bridle men’s ambitions, avarice, anger, and other
passions without the fear of some coercive power.”
Table 2-13 illustrates the ease of enforcing contracts in Pakistan in comparison with that
in South Asia. This is determined by tracking the time, cost, and number of procedures
involved from the moment a plaintiff files the lawsuit until actual payment. In Pakistan,
it takes on average 1261 days and 47 procedural steps to enforce a contract at a cost of
29.7% of the claim value (see Table 2-13). This performance is longer and more costly
than in the South Asia region, where it takes on average 1052.9 days and 43.5
procedural steps to enforce a contract at a cost of 27.2% of the claim value.
Table 2‐13: Data on Enforcing Contracts
Indicator Pakistan South Asia
Procedures (number) 47 43.5
Time (days) 1261 1,052.9
153

Cost (% of Claim)
Source: World Bank (2010a)
29.7 27.2
Moreover, a survey conducted by the World Bank in 2009 showed that the time and cost
of enforcing contracts greatly varied across cities in Pakistan. The time needed to
enforce a contract is on average about 1300 days. In Faisalabad it takes the least amount
time (730 days) to enforce a contract, whereas in Peshawar it takes 2190 days. And in
Punjab, for seven major cities, it took 1116 days on average to enforce a contract at an
average cost of 31.9% of the value of the claim.45 The costs comprise court, attorney,
expert and enforcement fees. Again the costs vary across cities in Pakistan. They are
cheapest in Sukkur (20.6% of the claim value) and most expensive in Lahore (42.8% of
the value of the claim) [World Bank (2010a)].
It is evident from recent research that, among comparable economies, a country’s
effective enforcement of contracts is a vital determinant of its comparative advantage in
the world market. This means that countries that are able to enforce contracts properly
tend to produce and export more sophisticated products compared to those with poor
contract enforcement [Nunn (2007)].
Hence an effective contract enforcement regime is required for Pakistani firms to
become internationally competitive. Moreover, it is crucial to develop an efficient
contract enforcement mechanism to reduce the cost of doing business in Pakistan.
Reforms are in progress to improve enforcement of contracts, which include setting up
of judicial tribunals to speedily resolve customs, labor, tax and banking disputes [Asian
Development Bank (2008)]. However, these reforms must be pursued with commitment
over the long-term to lead to expected results.
45 According to a study of the World Bank in 2008 of the cost of doing business in India, using Ludhiana
as the typical city for the Indian Punjab, it took 862 days on average to enforce a contract at a cost of 20%
of the value of the debt.
154

System of Taxation
Pakistan’s tax system is one of the major hindrances towards growth in the
manufacturing sector. According to The Investment Climate Report by the World Bank,
in 2009, 40% firms in Pakistan stated taxes as being a barrier in doing business; even
though this number decreased from 47% in 2002. Over all, taxes increase the cost of
doing business, take away the incentive to invest in the manufacturing sector, and create
obstacles for local producers to face competitive international markets, all of which are
hindrances towards growth of the manufacturing sector [World Bank (2009b)].
Procedures
Pakistani manufacturers have to go through a cumbersome and difficult procedure to
pay a large number of taxes to many agencies. According to World Bank (2008),
Pakistani manufacturers have to pay taxes to 47 agencies which, on average, takes about
560 hours annually. It is for this reason that Pakistan ranks 143 out of 183 countries in
the ease in paying taxes indicator, where China ranks 140th and Malaysia ranks 24th;
Maldives ranks the first in the paying taxes indicator where it takes less than an hour to
pay yearly tax. The time taken for Pakistani manufacturers in paying taxes is nearly
double the global and South Asian average, which are 286 and 285 hours respectively. It
takes a lot of time for producers to understand the complicated procedure that goes
through adjustments every now and then [World Bank (2010a)].
Between 1990 and 2007, 75 changes had been made to the Sales Tax Act by the Federal
Bureau of Revenue; this shifting set of rules and complicated tax-paying procedures
increase compliance costs and undermine the trust of the tax payers. It had been
announced that corporate tax rate would be further lowed to 30% by 2007, which has
not come to pass as the corporate tax rate still remains at 35%; such discrepancies also
weaken the taxpayer’s trust in the government [World Bank (2010a)].
155

Due to the tax rates and complicated tax collection system, tax evasion has become quite
significant leading to low tax-to-GDP ratio as compared to other countries in South Asia
[see Table 14]. In 2007, in a country of 156.7 million, only 2.3 million tax payers were
registered [World Bank (2010a)]; of course the reasons stated above cannot be
completely blamed for these statistics, though it is up to the tax administrative system to
correct this. Until recently corruption within the federal bureau of revenue was a major
problem in tax collection which was curbed to some extent by increasing employee
wages, but more needs to be done to avoid tax evasion and to solve problems in tax
administrating agencies.
Tax Structures
Where the number of taxes and collection procedures are a problem, so are the average
tax rates. Recent studies by the World Bank show that, in Pakistan, the corporate
income tax rates on private companies have dropped from over 55% in the 1990s to a
uniform 35%, apart from an adjusted smaller rate for small companies. However, even
this decrease in tax rate is not enough. Table 2-14 shows Pakistan’s Individual and
Corporate tax rates since 2007 compared to those of other South and East Asian
Countries. In addition to the already high tax rates, the introduction of the Value Added
Tax (VAT) or its variant form of GST will further burden manufacturers. VAT, which
was to replace the Sales Tax in July 2010, is defined as a multi-stage tax which will be
levied on the value-added items at each stage of chain of supply of goods and services.
This tax will cause pessimism among producers, making them hesitant about growth in
their production.
Table 2‐14: individual Tax, Corporate Tax and Tax Revenue Comparisons, 2007‐2009 (% of GDP)
Country Year Individual tax rate (%) Corporate tax rate Tax Revenue (% of
(%)
GDP)
2007 20 35 9.8
Pakistan
2008 20 35 9.8
2009 20 35 10.9*
2007 30 33.99 12.4
India
2008 30 33.99 12.9
2009 30 33.99 --
2007 45 33 --
156

China 2008 45 25 --
2009 45 25 --
2007 -- 30 8.0
Bangladesh 2008 -- 30 8.8
2009 -- 27.7 --
2007 35 35 14.22
Sri Lanka
2008 35 35 --
2009 35 35 --
2007 28 27 --
Malaysia
2008 28 26 --
2009 27 25 --
2007 20 20 13.9
Singapore
2008 20 18 14.6
2009 20 18 --
2007 35 27.5 16.6
Korea, Rep 2008 35 27.5 --
2009 35 24.2 --
Source: GoP (2010a).
Pakistan's tax base is also a major concern as the tax burden is unevenly distributed
among taxpayers; this needs attention if manufacturing is to grow. The income-tax
ordinance of 2001 contains 70 pages of exemptions which shelter many activities from
taxes and a majority of the weight of taxes falls on the manufacturing sector. According
to the recent Economic survey, industry contributed 63% of the collected taxes in 2009-
10 whereas agriculture and services contributed only 1% and 26% respectively. Even
within the 63% total tax collected from the industrial sector, most is collected from a few
large-scale manufacturers. The threshold for corporate income tax for small companies
is very low which would motivate firms to remain small. Also, this discriminatory
treatment with taxes creates bias, causing investors to invest in selected instruments
and sectors.
It is rightly argued that the trade of manufactured goods contributes to the growth of the
manufacturing sector; for this reason policies should be implemented to make trading
among manufacturers an easy process. Pakistan has recently taken steps to integrate
itself in the global economy. By reforming its tariff system it has become one of the more
open economies in South Asia. However, there is room for further improvement as there
are specific tariffs which, like discriminatory taxes, exempt certain sectors, leading to
misallocation of resources. There is also the issue of complex tariff and tax structures for
importers and exporters, both of whom have to pay a multitude of taxes; some of these
157

are: import duties, import surcharge, export duties, income tax on imports and export
development surcharge. Again, paying this many taxes after going through complicated
procedures disincentivizes manufacturers to import raw materials or to export their
goods. As local goods are not exported, local producers do not face competition in
international markets lowering growth in the manufacturing sector.
Efforts have been made to decrease the compliance cost of taxes as specific tax reforms
have been undertaken in order to improve tax filing and payment. Three large and
medium taxpayer units (LTUs) and thirteen Regional Tax Offices (RTOs) have been set
up in major cities to educate and facilitate taxpayers; a universal Self-Assessment
Scheme has been introduced to limit interactions between taxpayers and tax officials;
institutional reforms have been made in the FBR so it can function more practically
while an internal audit and internal affairs department has also been set up. Frequent
surveys are being undertaken for feedback from taxpayers. These efforts have had a
positive impact on tax collection, leading to a decrease in the number of firms that
considered taxes a major obstacle. However more reforms need to be made in order to
make tax collection smoother and less of an obstacle in the way of growth for
manufacturing firms in Pakistan.
Policy Recommendations:
The following are some proposed solutions in this regard:
1. The government should increase controls at the Karachi port and Afghan border to
curb smuggling
2. Efforts should be made for tariff harmonization between Afghanistan and Pakistan.
A bilateral agreement could be made to adjust tariffs on both sides, with the
objective of minimizing tariff differentials to reduce the premium on smuggling.
3. Afghan custom clearance should be conducted at Karachi
4. The Letter of Credit for the import of these products should only be issued in
Afghanistan. It should be noted that this has already been implemented.
158

5. The government should negotiate some degree of quantity control on products
meant for Afghanistan, based on the market size or demand estimates in
Afghanistan.
6. Rationalization of the sales tax regime in Pakistan would also help in reducing the
premium on smuggled goods that are manufactured domestically; e.g. tyres.
7. Another serious issue concerns under-invoicing. To tackle this problem, the
government should:
o Use International Trade Price (ITPs) lists to reduce the incidence of
under-invoicing.
8. Costs of Doing Business: Costs of doing business affect both domestic and foreign
investment in the country. These costs relate to bureaucratic inefficiency or red
tape, endemic corruption, complicated systems of taxation and business registry to
weak contract enforcement systems stemming from dysfunctional provincial judicial
systems. We propose the following:
o All labour-related issues to be handled at a single window.
o All labour-related tax dues (EOBI, WWF, Social Security, etc.) to be
lumped as a single payment.
o An online system for registration of new businesses should be
developed.
o All unnecessary procedures of starting a business, such as the
requirement to establish a company seal, should be abolished.
o Refunds of duty drawbacks and sales tax should be completed in
one month.
o Businesses should be allowed to keep their three month security
deposit with Sui Gas in National Saving Certificates.
o Contract enforcement needs to be improved through the following
measures:
I. Setting up of specialized courts or commercial divisions
in existing courts
II. Design and implement case management systems in
courts.
159

Competitiveness & Industry Stagnation
III. Introduce a maximum time limit of three months on all
commercial cases.
IV. Strengthen the Alternative Dispute Resolution (ADR)
System. Encourage and support provinces to replicate the
ADR system piloted in Karachi.
A competitive economy experiences high economic growth providing more
opportunities for industrial investment and expansion. Competitiveness is a function of
the physical, institutional and environmental factors of an economy which collectively
shape the business environment faced by manufacturers [State of Competition in
Pakistan, (2009)]. These factors not only determine future prospects but also explain
past performance of the manufacturing sector of an economy. After closely examining
the growth manufacturing sector, we conclude that Pakistan has underperformed in
providing a growth-conducive environment to its manufacturers.
The industrial sector of Pakistan has been rather stagnant and uncompetitive for the
past four decades. As the graph below shows (Figure 2-40), value-added in industry (as
% of GDP) has been ranging between 21% and 27% since 1970. From 1994 to 2003 it
was stagnant at about 23%, after which it rose to 27% and has been at that level to date.
Pakistan’s value-added has been the lowest amongst comparators, as can be seen in the
graph below. Even low income countries add more value Pakistan. And while world
value-added has also been falling since the 1970s, it started from a substantial 38% and
is still above that of Pakistan. The most important trend to notice is that of the lower
middle-income economies, in which, according to the World Bank (YEAR), Pakistan is
also included. The average value-added in lower middle-income countries is
significantly higher than that of Pakistan, with a large gap of 14% [see Figure 2.39].
Figure 2‐39: Industry Value Added, 1970 – 2008 (% of GDP)
160

Source: World Bank (2010c).
2.3.4.4 Pakistan and Global Competitiveness
The stagnant growth of Pakistan’s manufacturing sector is also in line with its
deteriorating ranking of competitiveness as measured by the Global Competitiveness
Index (GCI) on the parameter of intensity of local competition. Pakistan ranked 83 out
of 122 countries in 2008, and went further down the list of competitive economies to 101
out of 134 countries in 2010. For the last two consecutive years, Pakistan’s ranking is
unchanged at its ranking of 101 [see Table 2-15].
The GCI ranking of Pakistan is not very impressive and depicts a dismal picture of local
competition in the industrial sector. On comparison with its regional competitors,
Pakistan is above Bangladesh, but far below India’s ranking. India outperforms Pakistan
across all twelve pillars46 on which the GCI based its ranking. Though Bangladesh is
ranked at the 106th position on GCI, its macroeconomic stability is better than Pakistan.
However, on almost all twelve pillars, Pakistan’s performance is better than 42 other
countries having similar income per capita in 2010. Pakistan is scaled among other
46 Competitiveness is defined by 12 pillars (institutions, infrastructure, macroeconomic stability, health
and primary education, higher education and training, goods market efficiency, labor market efficiency,
financial market sophistication, technological readiness, market size, business sophistication, and
innovation) and divided into three categories (basic requirement, efficiency enhancers & innovation and
sophistication)
161

egional competitors like India and Bangladesh as a factor-driven economy; an economy
in an earlier stage of development which lacks sophistication and innovation
[Competitiveness Support Fund (2010)].
Table 2‐15: Pakistan’s Global Competitiveness Index Ratings, 2003‐2010
2003 2004 2005 2006 2007 2008 2009 2010
Rank 73 91 94 91 83 92 101 101
Out of 101 104 117 125 122 131 134 133
Hard score (out
of 7)
3.4 3.17 3.51 3.66 3.82 3.77 3.65 3.6
percentile 27.72 12.50 19.66 27.20 31.97 29.77 24.63 24.06
Source: Various Global Competitiveness Reports
2.3.4.5 Assessment of Competition in Pakistan’s Manufacturing Industry
Competition and competitiveness are two sides of the same coin. Competition plays a
vital role in economic development and growth through productivity and efficiency
improvement (Carlin et al., 2004). Firms growing in competitive markets charge low
prices, practice allocative efficiency, develop new products, innovate in distribution and
production, and adapt new technology. For example a country-level finding by
Sakakibara and Porter (2001) show that export competitiveness in Japan is attributed to
domestic competition and not to collusion or government intervention. Competition or
domestic rivalry measured by instability in market shares, has a strong and positive
relationship with trade performance in terms of enhancing export shares in world trade.
Stability of market share of firms in a particular industry marks an overt cooperation
among firms. In addition, unstable market shares signify competition among firms.
These results are robust for industries involved in research and development in
technology-intensive products.
A substantial portion of the industrial sector of Pakistan has been characterized by low
levels of competition ever since its inception in 1947. From being concentrated in the
hands of a few families to nationalization, the economy finally embarked on the path of
market-oriented private-sector-led growth in the late 1980s and 1990s. This
162

privatization was aided by a favourable world economy and led to an increase in
economic activity, bringing growth rates to around 7.5 percent in 2005. However,
operation of most of the basic utilities still remained in the hands of the public sector.
Even in the private sector, large-scale manufacturers continued to enjoy the attention
that the government bestowed on them.
Another important development over the years was the transition from having no smallscale
sector, to an indifference towards its existence in 1970s and 1980s and then to a
conscious attempt to promote SMEs in late 1990s, allowing the emergence of a
nontrivial small-scale manufacturing sector in the industry (Zaidi, 2005). Although the
size of an SME does not allow for scale advantages, it benefits from the higher level of
competition in the market which allow market forces to dictate outcomes. However, the
bias is still towards large-scale manufacturing with policies such as preferential access to
credit.
Even though high concentration ratios in some industries can lead to productivity
advantages due to economies of scale, power is concentrated in a few players resulting in
oligopolistic behaviour. While such anti-competitive practices are mostly carried out
without government knowledge, there are instances when they actually occur with
government acquiescence. At times price-setting along with other defensive measures
have been a form of political patronage carried out in the name of protecting industry.
Such cartelization behavior leads to barriers to entry and exit, dampening the spirit of
competition and therefore decreasing the incentive to innovate. This lowers productivity
and makes it difficult for the industry to be internationally competitive.
The empirical evidence on low total factor productivity (TFP) of Pakistan’s large-scale
manufacturing sector supports the outcome of low competition mentioned above.
Raheman et al. (2008) analysed most of the large-scale manufacturing industries in
terms of total factor productivity, which was decomposed into technical efficiency
change and technical change (technological adoption). During the time period 1998-
2007, the overall TFP growth rate was merely 0.9 percent. This comprised a positive
technical efficiency change of 1.2 percent but a negative technical change that reduced
163

the overall TFP growth rate to 0.9 percent. Out of the eleven industries in question,
seven have a technical change index of less than 1 (see Table 2-16).
Table 2‐16: Malmquist Index of Sector Means, 1998‐2007
No. Industry TFP Change TE Change Tech PE SE
Change Change Change
1 Automobile Assembler 1.013 1.000 1.013 1.000 1.000
2 Automobile Parts &
Accessories
1.002 1.011 0.992 1.000 1.011
3 Cement 1.023 1.030 0.993 1.027 1.002
4 Chemical 1.016 1.026 0.991 1.026 1.001
5 Engineering 1.007 1.014 0.994 1.006 1.008
6 Oil & Gas Exploration
Refinery
1.020 1.000 1.020 1.000 1.000
7 Oil & Gas Marketing 1.038 1.016 1.024 1.000 1.016
8 Pharmaceutical 1.005 1.000 1.005 1.000 1.000
9 Textile Composite 0.993 1.012 0.984 1.013 0.999
10 Textile Spinning 0.998 1.017 0.984 1.015 1.002
11 Textile Weaving 0.988 1.007 0.982 1.000 1.007
Mean All Industries 1.009 1.012 0.998 1.008 1.004
Source: Rehman et al., 200847.
Large-scale manufacturing industries are dominated by a few large players who prefer
not to invest in new technology and R&D due to lack of pressure from market
competition. The industrial sector stagnates as TFP growth rates are low. There is a
dearth of empirical evidence on large-scale Pakistani manufacturing firms for analyzing
industries on the basis of productivity growth and degree of competition. However, the
Competition Commission of Pakistan offers evidences of anti-competitive behaviour
found in key industries like sugar, cement, automobile, and fertilizer (see box 2-4).
Box 2‐4: Cartelization & Case Studies
47 Rehman, A., Afza, T., Qayyum, A. & Bodla, M. A. (2008). Estimating Total Factor Productivity and its
Components: Evidence from Major Manufacturing Industries of Pakistan. Pakistan Development Review,
47 (4).
164

Thus there is a good case for intervention in Pakistan’s manufacturing sector, which consists of many
industries that are dominated by large firms that have been accused of collusive behaviour.
The Cement Industry
One such example is of the cement industry. The sudden price hike of cement in February-March 2007
came after prices rose from July 2005 to April 2006. And while price declined till January 2007, it shot
up by PKR50 per bag in the following month. Low product differentiation and high market
concentration may give rise to collusive behaviour among cements manufacturers. Figure 1 below shows
that 54% of the market is shared by three manufacturers.
Figure1: Capacity Based Market Shares
Source: CCP internal research, based on data for the year 2006-07, CCP 2008.
It may also be argued that seasonality could be a reason for the price hike. Other factors could include
higher demand arising from the construction of mega-structures ( like Gawadar port), a real-estate
boom and earthquake reconstruction, yet the rise was still too much, too sudden and too unwarranted.
Therefore, the Association of Builders and Developers (ABAD) which buys around 30 percent of cement
asked the government to look into the matter, accusing the industry of cartelization. ABAD’s
involvement meant that the situation was serious, as bulk consumers like the government and ABAD can
not only influence production, but are also protected against sudden price hikes by their contracts. It
was small consumers in the end who were hardest hit. In addition, unfair business practices like absence
of expiration dates allowed dealers to create artificial shortages to raise prices. In this regard, the
Competition Commission of Pakistan (CCP) found a discrepancy and proof of collusive behavior: while
the All Pakistan Cement Manufacturers Association (APCMA) claimed that cement could not be stored
for more than 10 days, the committee found that it could be stored for around 60 days. This meant that
while the manufacturers could cause a temporary shortage, it would not last for long. The CCP’s
endeavors were further supported by a company which had left the APCMA and provided proof (by
supplying minutes of an APCMA meeting) that manufacturers were allotted supply/production targets,
which falls outside the jurisdiction of the APCMA (CCP, 2008).
The Sugar Industry
The sugar industry has also been very concentrated in Pakistan. Historical political patronage, aided by a
licensing system, has made it a rent-seeking industry par excellence. While sugar prices had shown a
gradual increase till 2007-08, it was the major sugar crisis in late 2009 and early 2010 that proved
harmful for consumers. The government intervened and fixed the price of sugar atPKR40/kg while
165

aiding hidden sugar stocks. This ill-considered policy intervention led to the creation of black markets
where sugar was sold at PKR60/kg, allowing traders to make windfall gains. The price did eventually fall
to PKR52/kg once the price limit was lifted. The massive sugar shortage depicts an abuse of free market
forces. According to the CCP report, the information obtained from the Pakistan Sugar Mills Association
(PSMA) clearly shows that its role was not limited to being a representative of the sugar producers-it
also allowed cartelization under its auspices. Not only did it intervene in terms of deciding output and
prices, but it also worked with the Trade Corporation of Pakistan to decide on the level of imports to be
allowed inro the country (CCP, 2010a).
Fertilizer Industry
The fertilizer industry was founded in the 1960s when two plants were set up, one of which – National
Fertilizer Corporation – was a public enterprise and the other Esso Pakistan Fertilizer (now Engro) was
in the private sector , resulting in a duopoly market structure. Even after another player, Dawood
Hercules Chemicals Limited, started urea production in 1971, the sector remained largely protected with
fixed prices, subsidies, and quotas. The public sector continued to enjoy dominance until complete
privatization in 2002. However, even after privatization, the level of concentration was high with Fauji
Fertilizer (FFC) and Engro having a majority of the market share. As depicted in Figure 1.XX, the
industry is dominated by 6 to 7 players, with FFC and Engro enjoying around 43 % of market power in
2010. This has given the main players substantive power in terms of setting prices. The level of
competition has fallen substantially and barriers to entry and exit have emerged.
Figure II: Market Power of Participating Firms
Collusive behavior was observed in early 2009, when a case of tie-in sales of urea with DAP, (an
expensive fertilizer) which had been in effect for quite some time, was brought to the attention of the
CCP. The subsequent investigation done by the CCP revealed that the firms were forcing dealers to buy
DAP with urea and had cut off the dealership of those who refused. The farmer class was hardest hit, as
these policies were inevitably passed onto them by the dealers. The farmers were placed in a difficult
position as they could not always afford to buy expensive fertilizer when they wanted to purchase just
urea (CCP, 2010b).
Source: NFDC data.
Anti-competitive behaviour in large-scale manufacturing industries hurts allocative
efficiency and productivity in the economy. Few dominant players in large-scale
166

manufacturing industries protect themselves from competition and resort to unfair
trading practices such as controlling prices, and restrict trading practices through
collusive behaviour. To raise the productivity of the manufacturing sector, competition
is inevitable. A dynamic manufacturing sector that can actively contribute to national
and international markets is not viable unless competition constraints are removed.
2.3.5 Other Horizontal Issues & Interventions
2.3.5.1 Knowledge­based Industries through a Full­Scale Science Park
The industry in Pakistan is suffering due to an extreme shortage of research and
development and innovation. Although the investment regime of Pakistan is one of the
most liberal in the region, it has failed to attract knowledge based industry investments
even from non-resident Pakistanis.
In light of this a cornerstone of the proposed industrial policy is the establishment of a
science park in Pakistan. The proposed science park will have formal operational links
with one or more universities, research centers, or other institutions of higher
education. It will be designed to encourage the formation and growth of knowledgebased
industries. It will have all the infrastructure required to attract professionals,
consultants and managers and will house organizations ranging from R&D to
manufacturing, marketing and branding. In short, the objectives of the establishment of
a science & technology park in Pakistan are the promotion of technology
commercialization, transfer and diffusion by fostering links between Industry and
universities, R&D labs, and promoting the formation and growth of knowledge-based
companies.
A science park fulfils several purposes that are critical to the success of the 2010
industrial policy. Most importantly, it enjoys the status of an R&D establishment, which
means it sidesteps the same WTO regulations that prohibit any subsidies to local
organizations. Local organizations can be encouraged to indigenize through collective
government sponsored R&D and build brands within a science park.
167

Second, a science park captures the entire product cycle, rather than just focusing on
manufacturing. It provides incubators for scientific innovations as well as a production
center for cutting-edge products. Universities will be playing a very strong role in the
emergence of the proposed science park and will be required to supply trained human
resource and a knowledge base for industrial innovation.
Third, a science park provides economies of scale and scope with respect to common
infrastructure facilities (such as transport, power, information and communication
technology connectivity, office and production space, and waste treatment) and
technical services (such as recruitment, training, mentoring, financing, networking, and
legal and IPR consulting).
Finally, through delegation of powers to the science park level, various ministries and
government bodies can achieve coordination, a goal that otherwise remains difficult to
achieve.
The aim is to create an environment where knowledge-based industries can thrive and
the required professionals can reside with their families. We envisage the park project to
be 6-year long with approximately US $500 million required in startup costs. The total
area for the park will be about 200 hectares. The project will be executed in 3 phases.
Phase 1 will include the survey of different sites in Pakistan for the park and the
selection of the most suitable site. The 2nd phase will include building the infrastructure
as well as building and recruiting park clients. The last phase will include facilitating the
investment environment for the clients and building bridges between different
universities of Pakistan, R&D institutes and park clients.
Policy Recommendations:
The government of Pakistan should incentivize the investment environment in the
science park through the following measures:
• Absorb initial startup costs for the science park
• Provide infrastructure of the park including amenities
168

• Establish venture capital funding for start-ups
• Provide five year tax holiday
• Ensure industrial laws and regulations simplification
• Ensure delegation of power from several other government agencies to park
administration for ‘one stop operation’
• Pass special commerce and tax laws for science park operation
• Ensure cooperation programs among universities, research institutes and park
clients
• Provide transparent, competitive R&D grants
2.3.5.2 Development of Industrial Estates, Special Economic Zones, Export Promotion Zones
Industrial zoning and infrastructure is significantly deficient in Pakistan. The Industrial
estates were developed several year ago and lack modern attire and facilities that are
required by modern age industries. Even in places where the government has
established new estates/zones the private sector has resisted moving to these states
either due to poor planning or due to exorbitant land prices. When designing most of
these estates the economic opportunity cost is hardy considered and in some cases
prime agriculture land has been reserved for ill-organized industrial activity. We have
considered Special Economic Zone and location choice in much more detail in the
section on spatial aspects; however, we have presented some policy guidelines to
address other issues:
Policy Recommendations:
The following is proposed in this respect:
1. To promote the growth of existing and resource-based industries the government
should create industrial estate and agro processing zones at the identified ‘hot
spots’ of economic activity. This mapping is provided in Chapter 5 of the report.
169

This should be done while preserving agricultural lands and observing municipal
zoning laws.
2. The government should develop these estates through the provision of necessary
soft and hard infrastructure and world class logistic facilities.
3. The government should facilitate estates and zones in:
• Accessing appropriate technology.
• Obtaining information about modern methods of production.
• Improving quality standards such as phyto-sanitary measures for agriculture
processing.
• Product branding and up gradation.
• Accessing international markets.
• This should be done by creating linkages with universities, research centres
available locally and internationally, and by creation of centres of excellence.
• Captive power generation should be permitted at the estate level.
• Effluent Treatment Plants and Solid Waste dumping sites should be
established in all industrial estates. Incentives such as tax break should be
provided for local construction/fabrication of effluent treatment plants.
• With falling duties, the attraction of EPZs (where imports are duty free) has
fallen dramatically. EPZ such as the one in Gujranwala should be turned into
an industrial zone.
• The government should rationalize its existing portfolio of industrial estates.
The failed estates or SEZs (e.g., Gadoon) need to be liquidated and lessons
learned from their failures. Sukkur Small Industrial Estate should be
developed.
• Business incubators for SMEs should be set up to reduce costs for business
start-ups.
2.3.5.3 Assisting Sick Industrial Units
In any business not all firms survive due to a variety of reasons however in relatively
efficient markets the productive capacity of failed firms is acquired by others. In
170

Pakistan due to difficulties in liquidation of businesses a huge amount of productive
capacity lies dormant at any point in time. The government should institute a more
transparent and efficient procedure/system to help distressed firms get out of business
quickly and at least cost.
Policy Recommendations:
• Recommend changes to the bankruptcy laws to make it easier for businesses to
acquire distressed firms. The transition and exchange of hands in business will be
made swifter under these legal revisions.
• Also fully support the Corporate Rehabilitation ACT being developed by SECP
and ensure that the ACT covers the possibility of providing technical support in
reviving the sick industry.
2.3.5.4 Implementing Cleaner Production Programs
Growth of the industrial sector cannot be visualized without strengthening cleaner
production programs in the country. As defined by the United Nations Environment
Program (UNEP), cleaner production is “the continuous application of an integrated
environmental strategy to processes, products and services to increase efficiency and
reduce risks to humans and the environment.” The implementation of cleaner
production programs is an integral part of the policies that aim at increased
competitiveness and increased efficiency of firms because they help firms save energy,
conserve water, control pollution, ensure safety of machines and equipment, improve
health and safety of workers, improve environmental conditions and the image of the
firm at the local and international levels.
Policy Recommendations:
• Encourage and promote investment in local manufacturing of cleaner production
equipment along with other emerging environmental technologies. However, in
the interim period the Ministry should advocate policies that ensure low prices
for imported cleaner production equipment.
171

• Pilot projects that help demonstrate costs and benefits of private investment in
cleaner production technologies, e.g., CO2 recovery, should be initiated as part of
awareness raising campaigns. Similarly, awareness raising seminars and training
programs should also be used to harness full benefits of cleaner production
programs.
• A policy for cleaner production needs to be devised that offers tax incentives and
tax rebates to compliant firms and to firms who want to establish Effluent
Treatment Plants. These incentives should also apply to those who get ISO 14001
certifications.
• Offer tax incentives to encourage private sector to set up quality testing and
certification laboratory facilities in all sectors, especially those relating to textiles,
sugar, leather and pulp and paper sectors.
• Allocate funds for next ten years to set up well designed landfill sites for disposal
of hazardous and non-hazardous industrial solid waste and installation of Hi-tech
incinerators.
• The existing National Environmental Quality Standards (NEQS) were developed
in isolation without consulting all the stakeholders. That is the reason why these
Standards are non-productive and hard to implement in their present form.
There is a need to revise the NEQS in consultation with all the stakeholders. The
new NEQS should be industry/sector specific and should be based on the ground
realities.
172

3 Competitiveness of Key Industrial Sectors of Pakistan
A key component of the 2010 industrial policy comprises increasing the global
competitiveness of various significant sectors in the country. The sectors are selected on
various bases, including whether the sector provides an input that is critical to
downstream industries, enjoys significant domestic demand, and has potential to
become internationally competitive to boost exports. In this report we present brief
analyses of the challenges facing such sectors along with strategic recommendations. We
should note at the outset that in this task we were helped by various studies that have
already been completed by organizations ranging from SMEDA to the EDB. Similarly,
agencies such as JICA and the World Bank had completed studies that proved to be
useful.
A sectoral approach adds greater specificity to policy recommendations. In the past, the
Government of Pakistan has generally relied on blanket policy interventions. Blanket
policy interventions, while useful in defining the broader contours of industrialization,
more often than not, tend to inadequately address the sector specific tradeoffs. For
example, whenever, APTMA is able to convince the government to devalue the Rupee, it
creates negative externalities for the pharmaceutical sector which is dependent on
imported raw material. Analyzing various sectors alerts us not only to the sector-specific
challenges but also to potential conflicts between them.
In formulating sector level policies, four considerations were foremost in our minds.
First, we aimed for dynamic comparative advantage, or competitive advantage, rather
than static comparative advantage. In other words, we are of the view that Pakistan
should not keep supporting only industries where it has a resource advantage (e.g.,
textiles) but also broaden its support to span sectors where it can develop a competitive
advantage based on capabilities. In industries with resource-advantages, value-addition
must be the goal. Second, Pakistan must invest in ramping up for industries where it is
likely to grab global market share. These are industries where either production has
173

ecome too costly for more developed nations, or industries where the knowledge gap is
not prohibitively large (e.g., industries where knowledge, technology or components are
no longer proprietary). Pakistan will have to quickly move into the sectors where gaps
are appearing in global value chains. Third, Pakistan should ensure that local
enterprises are able to exploit all potential local demand, in particular focusing on
industries where large demand exists domestically. 48 Finally, in order to boost export
competitiveness, a key strategy should be ‘outward foreign direct investment’. This will
involve plugging value chain gaps (e.g., absence of R&D or branding knowledge) by
buying assets abroad.
Below, we provide descriptions of three primary sectors (steel, chemical and fertilizer),
followed by those of several downstream sectors.
3.1 Primary Industrial Sector
This section of the report covers three of the primary industrial sectors of Pakistan;
Steel, Chemicals and Fertilizers. These sectors are important not only in terms
employment and revenue generation, but also as significant contributors in supporting
the competitiveness of the value added Industry. All value added industry, be it textile,
leather, engineering or food processing is dependent on inputs produced by these
industrial sectors. If Pakistan has to improve its industrial competitiveness it will have
to significantly upgrade and strengthen these primary/ancillary industries.
3.1.1 Chemical Industry
48 It is now established that securing the domestic market gives local firms the platform from where
export competitive products can be launched. Furthermore, research has shown that “ ‘national’
ownership to secure the home market (defined by majority stock ownership and directorships) has
become more critical over time for five reasons: Without national enterprise, the illegality of local content
regulations under WTO law is binding, outward foreign direct investment is implausible, out-sourcing for
developed countries is untenable, establishing brand names is impossible, and reversing brain drain of the
top national talent is less likely.”
174

In recent years, the global chemical industry has become one of the fastest-growing
manufacturing sectors despite soaring crude oil prices and stricter international
environmental protection standards. It is now well established that because of forward
and backward linkages, polymer consumption has a significant multiplier effect on GDP
growth. Moreover, the chemical industry is considered to be a major driver for
innovation, and, developments within it have strong positive effects across industries
such as auto, rubber, textiles, consumer products, agriculture, petroleum refining, pulp
and paper, health services, construction and metals. In India for instance, the chemical
industry has been a leading driver behind economic growth.
A key by-product of oil refineries is Naphtha, which when cracked further, produces
ethylene, butadiene, propylene, benzene, toluene, paraxylene etc. for a number of
downstream industries such as Textiles, Paints, Plastics, Pharmaceuticals, Agriculture,
Packaging and others (see Figure 3-1). Pakistan has no facility to produce basic olefins
like Ethylene, Propylene, Butadiene, Styrene as well as aromatics like Benzene, Toluene
and Xylene due to absence of Naphtha cracking facility. This has resulted in limited
growth of chemical and petrochemical industry in the country.
At present, the petrochemical industry of Pakistan is mainly downstream and includes
production of polyvinyl chloride (based on imported Ethylene), synthetic fibres (based
on imported MEG), Purified Terephthalic Acid (based on imported Paraxylene),
aromatics and carbon black. Current demand of the Petrochemical based Products is
around 2.0 million tons/year [BOI, (2010)] (Table 3-1). For manufacturing of these
products, a large quantity of basic raw material is imported e.g., Pakistan spent around
$1 billion on import of these polymers in 2009 (Table 3-2) [BOI, (2010)].
Similarly, the chemical industry is not fulfilling domestic requirements, as a result every
year a large amount of foreign exchange is also spent on the import of different
chemicals. This has caused a burgeoning chemical import bill of approximately US$ 4
billion49 which is approximately 12 percent of total imports. Given the high demand for
49 Federal Bureau of Statistics (2008-09)
175

inputs/raw material by the domestic downstream industry there is a vast potential to
manufacture petrochemicals within the country.
176

Figure 3‐1: Petrochemical Value Chain
Source: : ‘Pakistan Petrochemical Market”, Lotte Pakistan, 2008: 16
177

Table 3‐1: Demand for Petrochemical Products
Products Quantity Metric Tons
Polypropylene 15,000
Polyethylene 200,000
Polyvinyl chloride 90,000
Polystyrene 30,000
Monoethylene glycol 400,000
Synthetic fibres 730,000
Pure terephthalic acid 400,000
Total 2,000,000
Source: BOI, (2010)
Table 3‐2: Import of Basic Chemicals in Pakistan (Million US$)
Product HS Code Import value
(Million US $)
Polymers of ethylene, in primary
forms
3901 383
Polymers of propylene, other
3902 317
olefins in primary forms
Polymers of styrene, in primary
forms
3903 27
Polymers of vinyl chloride, other
halogenated olefins
3904 20
Styrene 290250 36
Ethylene glycol (ethane diol) 290531 162
PTA 291736 47
Total
Source: UN Commodity Trade Statistics Database
992
It is evident that the lack of integration in Petrochemical industry of Pakistan is
hampering the growth of downstream chemical industry. For the establishment of a
fully integrated industry, it is imperative to set up a naphtha cracker plant in the country
which is the missing link in the value chain. Pakistan has a crude oil refining capacity of
12.81 Million Tons/Annum50 and there are further refining projects in the pipe line
which will increase this capacity in the near future. Refineries produce around 900,000
Tons/Annum51 of Naphtha which is being exported at a nominal price because of
absence of Naphtha cracking facility in the country. Thus, a large-scale naphtha cracker
in Pakistan is the need of the hour. There have been many studies conducted recently
50 Ministry of Petroleum and Natural Resources
51 Pakistan Energy Yearbook 2008
178

which have stressed the need for establishing such a facility. 52 With this building block
in place a large number of chemicals can be produced which include plastics, synthetic
rubbers, fertilizers, explosives, solvents, dyes and pharmaceuticals. The estimated cost
of this project is US $ 1.5 Billion53 .
Keeping in view the importance of a naphtha cracker for the development of the
chemical industry and subsequently for spurring growth in other sectors, it is imperative
to revive the hydrocracker project either through the public sector or under publicprivate
partnership. There needs to be an expeditious decision on this as the costs of
installing a hydrocracker plant/complex have risen substantially. In 1992, its cost was
$350 million54 , whereas the current international estimates of constructing the project
are around $ 1.5 Billion. India and Singapore are some examples where the government
has played an active role in the establishment of cracker plants. In India majority of
cracker projects are either state owned or in public-private partnership. The largest
naphtha cracker in India is established by state owned Indian oil which started
production in March 2010.
Similarly, Petrochemical Corporation of Singapore (Pte) Ltd, a joint venture between the
Government of Singapore, The Development Bank of Singapore Limited and the Japan-
Singapore Petrochemicals Company Limited (JSPC), established a $2 Billion
petrochemical complex after setting up the country’s first cracker plant in 1984.
Investments in naphtha cracker worldwide are dependent on local availability of
feedstock and downstream demand55. China and India are the demand centers
therefore more investments have been made in naphtha crackers. Opportunity also
exists for Pakistan where naphtha feedstock is available and demand for chemical
52 Second Quarterly Report 08 (SBP); BOI; Technology Based Industrial Vision and Strategy for Socioeconomic
development (PIDE)
53 Oil & Gas Pakistan; BOI; p-20
54
http://www.dawn.com/wps/wcm/connect/dawn-content-library/dawn/in-paper-magazine/economicand-business/the-elusive-hydrocracker-project
55 Chemicals in the Middle East 2020, McKinsey & Company, 2008, pp.5
179

products is reasonable (import of ethylene polymers range between 200,000 tons to
300,000 tons per annum during 2005-200956 which is expected to grow further).
3.1.1.1 Policy Recommendations:
The chemical industry must be developed in order to steer the composition of industrial
output toward more capital-intensive and technology-intensive sectors and engineering
heavy products with the aim of upgrading our export profile and reducing our reliance
on costly imports. We believe that significant inter-industry externalities will be
generated by focusing on the chemical sector.
• Begin groundwork for the establishment of a petrochemical complex in Hub. This
complex should include a refinery, a naphtha cracking facility, and facilities for
manufacturing of certain petrochemicals for which demand has been ascertained.
Hub is a suitable location because Byco (formerly Bosicor) refinery is already
located in the area and another refinery project is proposed at that location. After
expansion Byco will be the largest refinery in the country while National Refinery
and Pakistan Refinery are located nearby in Karachi which produce more than
60%57 of Naphtha in the country. This will provide easy access to Naphtha and
reduce transportation costs. This will have to be a public-private partnership
given the magnitude of investment involved. The textile industry will be a major
beneficiary of this initiative. Despite extensive BMR, the industry appears to have
shifted down the value chain in the past few years, with many value-added units
closing down. Also, while the world has moved to MMF-based textiles, Pakistan
has remained hugely dependent on cotton-based textiles. The establishment of a
Naphtha cracking facility should greatly boost the downstream development of
MMF based textiles, technical textiles, and facilitate the processing sector which
is hugely dependent on imported chemicals.
• In order to proceed with the above initiatives, we recommend that the chemical
industry be brought under the purview of MoIP rather than the Ministry of
Textiles, where it is currently housed.
56 UN Commodity Trade statistics
57 Lotte Pakistan
180

• The Petrochemical plant should be a government led, economically viable project
with equity from private investors. In order to make it feasible, maximum duties
will have to be imposed on all chemical-related end products, as soon as the
facility starts production.
181

3.1.2 Steel Industry
Steel is a critical ingredient for most intermediate and final products in an industrial
economy, and it is difficult if not impossible for Pakistan’s industry to become
competitive without a reliable base for steel production. Pakistan has an abundance of
iron ore deposits (with the largest ones located near Kalabagh (an estimated 300 million
tons), Dilband (200 million tons) and Chiniot (100 million tons). Unfortunately, despite
this availability, local steel production has not kept pace with increasing demand (see
Figure 3-2), and Pakistan has become increasingly reliant on steel imports, thus
exposing the economy to exogenous shocks.
Figure 3‐2: Demand and Supply Gap of Steel in Pakistan
Source: ‘National Steel Policy’, Engineering Development Board, Pakistan, (2010: 13)
It is clear from the figure above that the gap between demand & supply of steel is
increasing dramatically. The present gap of 1.4 million tons is projected to increase to
more than 3.6 million tons by the year 2015.
Pakistan’s total installed manufacturing capacity of steel is estimated to be around 4
million tons with capacity utilization averaging around 80%. The following are the
major players in the industry:
182

• Pakistan Steel Mills (the country’s only integrated steel plant)
• Steel melters (re-melting imported & locally generated scrap)
• Steel re-rollers (using local manufacturing & imported & ship-demolition rollable
materials)
• Ship demolition industry
Box 3‐1: The Steel Industry of China
In 1949, steel in China recorded a production of a meagre 158 thousand tons. However, by 1996, the
Chinese steel industry had forged its way to the number one place in the world in terms of production
with over 100 million tons. This breakthrough was achieved with prudent industrial policy of China
towards its steel sector. Back in 1960s, China adopted a series of standardized designs in the steel
industry and set up billet mills, rod mills, cross-type section mills, seamless pipe mills and high plate
mills. Furthermore, with the development and spread of continuous casting techniques, the steel
industry underwent a thorough up-gradation in the 1980s.
After achieving dominance in terms of quantity, the industry is now focusing on improving the quality of
its products in order to supply a high percentage of the needs of its most demanding customers, such as
automobiles and electrical appliances. This is particularly true of steel sheets which have to be imported
in order to satisfy the demands of a number of industries. In order to achieve this, a vigorous program
has been set in motion to invest in equipment that will improve steel quality. This includes continuouscasting
equipment and improved rolling & finishing equipment. The hot-strip mill has become a focal
point, with five new units starting production in the 2000s.
Source: Hogan (1999) and industry reports
With an installed capacity of 1.1 million tons, Pakistan steel is the country’s only
integrated steel plant employing blast furnace and converter technology with total
dependence on imported ore and coking coal. Their product mix include continuous cast
slabs & billets, rolled billets, hot-rolled flat products, thin strip, pig iron and hard coke.
In the steel melting sub-sector, there are about 110 units with an installed capacity of
2.0 million tons. Most of these units are run in the informal sector and their product
range includes continuous cast billets, thin ingots, and a non-standard product called
‘runners’. On the other hand, the installed capacity in the steel re-rolling sub-sector is in
the vicinity of 3.2 million tons. Their product mix includes rebars for the housing &
construction sector and a wide range of structural steel sections for industrial and other
applications.
Until a few years back, the ship demolition sub-sector was a supply source of close to a
million tons of re-rollable & re-meltable scrap besides forged steel shafts, non-ferrous
183

scrap and serviceable engineering goods. However, due to a sharp increase in the
international market for demolition vessels, the output of this source has declined to a
negligible 0.2 million tons (National Management Consultants, 2007) .
All sub-sectors of Pakistan’s steel industry suffer glaring weaknesses, some of which are:
• General fragmentation leading to conflict of interests within the industry subsectors;
resultantly leaving little room for the government’s fiscal measures to
induce capacity enhancement.
• Aging of equipment installed at Pakistan Steel Mills resulting in a breakdown
frequency much above internationally acceptable levels; lack of preventive
maintenance and the inability of the Pakistan Steel to carry out major capacity
enhancement during the 25 years of its operating life.
• Old and obsolete technology being used by the steel re-rollers with an industryaverage
rolling capacity of plants less than 2 tons/hour. The same problem of
obsolete facilities is faced by the ship demolition industry.
Since Pakistan Steel Mills is the sole source of supply for more than 450 medium
capacity and small capacity foundries in the country, therefore, it is extremely important
to revive PSM. The Government of Pakistan has been under serious pressure to privatize
the steel mill as it is continually running in a deficit. The amount of money that the
government had to pour into the steel mill to keep it running exceeded Rs 25 Billion in
2009-10. However, we feel that privatization is not the solution to this problem.
Privatization will only result in asset stripping and loss of an asset for which the
government in future may not have enough recourse to re-establish. The steel mill
should stay in control of the public sector, however, governance and transparency in its
operations should be achieved. A possible way of addressing the problem of steel mill
management is to contract it out to a professional steel management company using a
transparent international tender. The management should be negotiated against strict
targets and the remunerations should be tied to the performance against these targets.
184

The private sector in the steel industry is mainly clustered in Lahore, Gujranwala and
more recently in Peshawar and Noshera. The Steel Melters Association of Pakistan
claims that around 80% of the countries steel producing furnaces are located in Lahore
and Gujranwala (EDB Report, 2010: 20). The major impediment to the private sector is
the limited availability of power. Steel production’s major input is electricity and given
the current shortage the industry is extremely handicapped.
In the backdrop of the above, Pakistan’s policy on steel is to decrease reliance on
imports and develop and expand the indigenous steel industry.
3.1.2.1 Policy Recommendations:
Based on the importance of steel as a basic industry, foreign currency constraints,
employment considerations and the ability of the steel industry to insulate Pakistani
economy from world economic cycles, we recommend the following:
• Triple the capacity of the PSM over the next 5 years to make it economically
viable. The government will take steps to attract investment into enhancing
PSM’s capacity to 3 million tones.
• Incentivize further steel production with a goal of producing 8 million tones of
steel indigenously by 2015 by establishing an industrial park. This park will be
given priority in terms of power, gas and financing. The park will also contain
facilities for foreign experts, whose visits will be sponsored by the government.
• Invest in the enhancement of existing knowledge resources. For this a steel and
metallurgy institute will be set up in conjunction with the PSM, universities and
downstream industries.
• Re-evaluate the impact on overall industry and the basis on which FATA was
allowed tax exemptions on running steel furnaces.
3.1.3 Fertilizer Industry
Agriculture is one of the major economic sectors of Pakistan contributing around 22% of
GDP. Fertilizer industry is one of the key suppliers of raw material to the agriculture
185

sector. Investments in improved fertilizer technology have made a vital contribution to
the economy by helping farmers improve crop yield and productivity over the last few
years. With population predicted to grow by 19 % by 2025, the country’s need for
enhanced food production will be critical. Combined with a predicted decrease in arable
land up to 25% per capita, the crop yield per acre must rise.
As it is, in 2008 imports of Urea and DAP reached approx Rs 82 bn. As depicted in
Figure-I, after 2018 if not before, the demand supply gap will begin to widen. If further
capacity is not added and Pakistan continues to import fertilizer, the burden on the
exchequer will be unsustainable. Figure 3-3 shows the Pakistan’s fertilizer import - DAP
has the highest share in imports as there is only one plant producing DAP in the
country.
Figure 3‐3: Pakistan’s Fertilizer (N.P.K) Plant Capacity & Demand (000’ tons in nutrients)
Increase in Capacity by
Source: ‘Development of a National Fertilizer Strategy for Pakistan’, Arthur D. Little, 2009: 15
Figure 3‐4: Pakistan’s Fertilizer Import
186

Source: ‘National Fertilizer Strategy, Pakistan’, Arthur D. Little, 2009: 45
In addition, a recent study conducted by Arthur D Little consultants has identified
Pakistan’s NPK mix (Figure 3-4) as one of the reason for below average crop yield
compared to other countries and suggested a mix of NPK- (Nitrogen, Phosphorus,
Potassium) ratio of 65:17:17. Keeping in view the demand and to stop ongoing imports
of fertilizers in the future, an additional 1,000 - 1,500 tons of urea and 1,000 - 1,500
tons of DAP are needed annually until 2025 to balance the NP ratio. Potassium fertilizer
(K) is to be imported because of a lack of any accessible potassium footprint in the
country.
Another critical issue facing the industry is availability of natural gas. For nitrogen
based urea fertilizer the primary input/raw material is natural gas. Government of
Pakistan is the major stakeholder in the gas utility companies providing gas to the
fertilizer plants. Fertilizer sector consume 16% of total gas supply of the country out of
which 80% is used as feedstock. However, natural gas demand has far exceeded the
available supply. Figure 3-5 shows that indigenous gas reserves and planned projects in
pipelines, which if materialized could give energy security for a shorter period. However
there still will be a supply gap post 2015. This situation calls for better resource
management to sustain the growth of the industry.
There is a dire need for both government and the fertilizer companies to thrash out an
alternate solution to work around limited gas availability without (1) substantially
cutting domestic urea production and hence raising reliance on imports and (2)
weakening contribution margin for urea players. A two pronged strategy is
recommended where on one side focus should be on managing the existing resources
efficiently and on the other hand efforts should be made to increase the gas supply (new
explorations and long term contracts with neighbouring countries).
187

Figure 3‐5: Comparison of Pakistan’s NPK Ratio with other Countries
Source: ‘Development of a National Fertilizer Strategy for Pakistan’, Arthur D. Little, 2009: 80
Figure 3‐6: Pakistan’s Gas Demand and Supply Balance
Source: ‘National Fertilizer Strategy, Pakistan’, Arthur D. Little, 2009: 63
To better utilize the existing resources, it is advisable to link the subsidy on gas with
efficient utilization of gas. This will force the manufacturers to improve energy efficiency
in their plants by upgrading these to reduce gas consumption. Fertilizer plants should
also be encouraged to invest in alternative feedstock for energy production (water, solar,
coal, fuel oil, bio fuel) which can reduce the burden on gas reserves.
Similarly, efforts should be made to use alternative feedstock and reduce the
consumption of Natural Gas. Energy sector which is so far the largest consumer of gas
needs to switch to alternatives like hydro, nuclear, solar and coal based technologies.
Demand in transport sector is growing at fastest pace and efforts should be made to use
188

alternate bio fuels (bio diesel, bio ethanol). Only the fertilizer sector is completely
depending on Natural Gas because it is by origin the key building block to produce
Ammonia. Alternative feedstock (naphtha, fuel oil) are limited in Pakistan and while
coal is expensive (because of quality of Pakistan’s coal reserves) and less efficient than
Natural Gas58 . With expected future increase in gas prices further studies should be
carried out to assess the possibility of using coal based plants. As China is the only
country which has coal based ammonia plants, their assistance could be sought to
develop alternate feedstock strategy for the fertilizer sector.
3.1.3.1 Policy Recommendations:
Being the backbone of agricultural productivity, the role of fertilizers will always remain
crucial and owing to its strategic importance, we recommend the following:
• Incentivize industry to add further capacity of Urea and DAP in line with the
suggestions of ADL. This will correct the current NPK mix.
• Conduct energy audits of fertilizer manufacturing plants to assess their energy
efficiency, and vary their gas subsidy based on their efficiency.
• Incentivize fertilizer plants to invest in alternate modes of energy production.
• Explore the feasibility of coal as alternate feedstock.
• Keeping in view the rising energy demand and supply gap, efforts in further
explorations of reserves needs to be expedited
58
‘National Fertilizer Strategy, Pakistan’, Arthur D. Little, 2009: 80
189

3.2 Value Added Knowledge Based Industrial Sectors
3.2.1 Auto & Farm Machinery Industry
The auto market is one of the largest segments in world trade. The annual size of
automotive export trade in the world has grown to a massive level of over US$ 600
billion, which accounts for about 10 per cent of the world export [‘The automobile
market of Pakistan’, UNESCAP, (2005: 1)]. Changing models, improving fuel efficiency,
cutting costs and enhancing user comfort without compromising quality are the most
important challenges of the auto industry in a fast globalizing world. Hence there is a
need for exploring the industrial complementarities in the South Asia region for better
quality, favorable costs, fuel efficiency and attractive designs. The automobile market of
Malaysia is a case in point (see Box 3-2).
Automobile sector is one of the fastest growing sectors in Pakistan. It contributes
towards the nation’s economy in the form of technology transfer, employment,
investment and much more. Automobile sector contributed over PKR23 billion to the
national exchequer in the year 2003-04 (www.pama.org.pk).
Box 3‐2: The Automobile Market of Malaysia
In late 1970s, the Malaysian automobile market had 11 assemblers of automobiles. The volume of
imported Completely Knocked Down (CKD) units increased rapidly during this period due to
government protection of local assembly. However, the excessive number of makes and models
produced by the 11 assemblers for the small local market made it difficult for local part makers to
achieve economies of scale, reflected in high prices and low local content., (averaging only 8% in
1979). The local content was largely limited to tires, batteries, paints, filters, seat-belts and glass items.
There was official dissatisfaction with the slow growth of and limited local participation in the
industry, encouraging the idea of a state-led development of a ‘national car.’ The first national car
company, Perusahaan Otomobil Nasional (Proton), was then launched in 1985 as a joint venture with
Mitsubishi. With strong support and protection from the state, Proton has managed to dominate the
passenger car market by offering the most economical price. By 1995, it managed to capture 77% of
the domestic market and also exported its cars to 28 countries, accounting for 23% of total sales.
Afterwards, the government initiated a 2nd national car project named Perodua, to produce smaller
passenger cars for the local market. Perodua also received preferential treatment from the
government and its Kancil model became the 2nd best selling car in Malaysia. The government has
also encouraged other automotive projects due to which the Malaysian auto industry consisted of final
assembly, parts and components producers, supporting industries (input suppliers of parts and
component producers), repair services, franchised distributors and financial agencies.
Source: Jomo & Sundaram (2007)
190

It is estimated that 32% of Pakistan’s 170 million population is urbanized with over 60%
of the population aged between 15-35 years. Additionally, Pakistan has one of the lowest
motorization levels (8 vehicles per 1000 persons) in Asia compared to India (11),
Indonesia (21), Sri Lanka (25), and Malaysia (641) [BMA Capital report, (2009)]. These
factors indicate that the domestic market for automobiles is large and has the potential
to grow tremendously in the future.
Figure 3‐7: Motorization Levels per 1000 persons
Source: United Nations World Statistics pocketbook and Statistical yearbook, BMA Research
The supplier industry is driven by the level of subcontracting provided by the
assemblers and by the auto replacement market. According to the Pakistan Association
of Automotive Parts & Accessories Manufacturers (PAAPAM, 1995), the supplier
industry consists of about 1900 firms. These firms fall into two quite distinct segments,
one supplying to the replacement and the other to the OEM market, i.e., to assemblers.
191

About 75% of the firms supply to the replacement market. These firms are small with an
average of 4 employees, belong to the informal sector, and supply generally low-quality
products. The remaining 25% of firms supply components to the major assemblers.
These Original Equipment Manufacturer (OEM) suppliers are relatively large (average
of 100 employees) and were mostly developed by Pakistan Automobile Corporation
(PACO) through technical assistance and other support.
Until 2006 a deletion program was followed in the Auto industry. Due to some
structural reasons as well as poor governance leading to smuggling and under-invoicing,
the deletion program could not be as successful as one would have hoped. Indeed, by
many accounts after many years, with a few exceptions, major OEMs have achieved less
than 40% local content. This too is in components that are inherently difficult to import
(e.g., seats, batteries, tyres etc). Moreover, transfer pricing where it occurs, acts as a
disincentive to produce locally. Where the auto industry has localized, it has yielded
highly favourable results generating almost 200,000 jobs and an extensive vendor
industry. Several parts are now produced competitively within the country, and in some
areas we are ideally poised to even export. However, a few barriers continue to prevent
this industry from growing further.
3.2.1.1 Policy Recommendations:
The automobile industry is essential for promoting related industries, enhancing
utilization of locally-made components, encouraging the upgrading of technology,
engineering and technical skills, therefore we recommend the following:
• Strictly enforce the Tariff–based System (TBS) which is officially in use, but
effectively in abeyance. This is absolutely crucial. In the absence of effective
monitoring, the progress of local auto industry will be severely impeded.
• Revise the indices that are still quoted for parts deletion. These are not based on
international prices and are misleading. New indices need to be based on
international prices of respective components, and the TBS needs to be brought
in line with these.
• Transfer pricing severely impedes the development of a local market, and
benefits parent companies of OEMs. While Pakistan officially has an ‘arm’s
192

length’ rule in effect, it is very weak. Application of the arm’s length principle by
FBR needs to be considerably strengthened.
• Granting of Pakistan-specific licenses for assembly restrict the export potential of
OEMs in parts of the industry where high levels of deletion have been achieved.
The government needs to take remedial steps to eliminate this constraint. In this
regard, priority will be given to three sectors: two-wheelers, three-wheelers and
tractors. These are the sectors where most deletion or indigenization has been
achieved, and these are poised for international growth. Starting with these
sectors, the government will a) complete deletion fully, b) facilitate the
acquisition of foreign brands and c) facilitate the establishment of joint ventures
where such restrictions are not in force.
• Any firms completing full deletion and operating under local brands should get a
subsidy on mark-up.
• Establishment of Clusters: Three different clusters should be organized within
the Auto and Farm Machinery sectors. Two automobile clusters should be located
in Lahore (Sheikhupura Road, near Motorway) and Karachi (near Port Qasim).
These will cater to cars, motor cycles and three wheelers. Furthermore, a Tractor
and Farm machinery cluster should be located in Daska (near Gujranwala).
• In order to facilitate deletion, and develop local parts based on global standards,
an Auto Design Institute will be established in conjunction with NED University.
All OEMs will be required to collectively sponsor this Institute, and transfer
technology through this initiative to vendors. The government should sponsor
foreign consultants to come and stay at the Institute for the next two years,
within which the Institute should complete full localization.
• Indigenization and development of local industry has suffered in the past
whenever fully made up cars (whether as cabs or otherwise) have been imported.
This practice should not be allowed in the future.
3.2.2 Electronics Industry
193

Electronics is one of the largest manufacturing sectors in the global economy. It is
considered as one of the world’s fastest growing industries with market size of $ 1659
billion in 2008, which is comparable to automobile approx $ 2500 billion59 . The
industry in search of low cost destinations offers huge potential to developing countries.
Singapore, Malaysia, Thailand, Taiwan and Philippines are some of the examples where
electronics is the largest export industry. Developments in this industry have an impact
on other sectors in the economy such as automobiles, telecommunication, engineering
etc.
Electronics industry in Pakistan is still in infancy and unfortunately never became a
major revenue generating industry despite the huge potential of electronics industry for
developing economies. The industry largely focused on activities confined to assembly
operations based on imported Complete Knocked Down (CKD) or Semi Knocked Down
(SKD) kits and materials. Nevertheless, industry assembles a variety of products ranging
from power supplies, inverters, converters to consumer electronics (TV, Cassette
Recorders and DVDs), telecommunication equipment to more sophisticated
manufacturing operations like Printed Circuit Board (PCB) and wafer fabrication. Most
companies are involved in consumer electronics while there are few companies involved
in sophisticated production/assembly but using imported components or subassemblies.
Therefore, contract manufacturing by domestic firms under the patronage
of foreign firms (OEMs) did not increase the technological know-how.
The electronic sector in Pakistan could not develop to the desired level due to a host of
issues which are mainly because of the absence of a long term vision for the
development of this vital sector. Unlike Malaysia, Korea, China and India where the
electronics industry attracted focused government attention which provided various
incentives and an enabling environment for making profitable investments. Foreign
investors found these countries attractive to establish Joint Ventures, Technical
collaboration and set up facilities to produce cost effective products.
59 World Electronics Industries 2008-13, ’, Decision Etudes Conceal, 2009
194

3.2.2.1 Critical Issues
Following are major issues that have impeded development and growth of this sector:
3.2.2.1.1 Poorly Articulated Value Chain
As stated earlier, assembler imports majority of components. There is no deletion policy
adopted in electronics industry as the case in automobile which hinders the transfer of
technology. This situation indicates that even large multinational companies assembling
TV sets are as a matter of fact engaged in import business, as they are actually selling the
imported parts just after giving them the shape of an assembled product.
In past Pakistan has an indigenization/deletion programme for different products of the
engineering industry which was formulated in 1987 to promote the engineering base of
the country. It aimed at facilitating the exploitation of available resources; the transfer
of technology and linkages between large, medium and small-scale industries. The
programme encompassed the engineering, electrical goods and automobiles industries.
Under this programme tariff exemptions were linked with local content requirements.
This deletion programme for electronics industry was abandoned in 2001 following the
World Trade Organization (WTO) Agreement on Trade-Related Investment Measures
(TRIMs) which came into force in 199560.
Later, Government introduced Emerging Electronic Products Assembly Scheme
(EEPAS) in Trade policy of 2001 which was very much focused on assembling. Under
EEPAS indigenization is not necessary. To facilitate investors, CKD kits @ 5% Customs
Duty are allowed for various items and the determination of local manufacturing status
is not required. These include television sets, cellular mobile phones, VCD & DVD
players, cassettes player, electronic calculators, radio, microwave ovens and CLI
apparatus.
60 Trade policy review’, WTO, 2006: 99
195

3.2.2.1.2 Lack of Technological Know How
Pakistani electronics manufacturing firms are dislocated from the main international
sources of technology and research & development. It is because industry is confined to
the assembling of imported components. It operates in relative isolation from the world
centres of science and innovation and is behind in engineering and technical skills.
Universities in Pakistan are not technologically advanced and other educational and
technical /vocational institutions are poorly equipped.
3.2.2.1.3 Low Foreign Direct Investment
The most effective way of enhancing competitiveness is collaboration with foreign
companies. Table 19 below compares the foreign direct investment in Pakistan’s
electronics industry with Philippines which eyes $1 billion investment in 2010. It shows
that Pakistan has very low level of FDI in electronics.
Table 3‐3: Foreign Direct Investment in the Electronics Industry, 2002‐2009 (US$ Million)
Country 2002 2003 2004 2005 2006 2007 2008 2009
Pakistan 26.4 17.6 17 16.5 21 22 51.9 39.9*
Philippine
s
*Till April
270 230 443 776 732 1400 400 484.1
Source: Economic Survey of Pakistan, ‘Electronics Industry’, SEIPI p.10
The low level of foreign investment indicates that the industry does not have significant
foreign support which is the most effective tool to improve quality and strengthen
competitiveness amid increasing competition with imported goods and rapid expansion
of markets. This in turn may hinder favorable development of the industry in future.
This situation can be contrasted with Philippines where electronics the largest export
industry. Leading component manufacturers and OEMs have established their plants
with billions of dollars of investment.
196

However, foreign collaboration must move beyond the assembly stage. In other words,
forward and backward linkages needs to be created and sector needs to transform from
consumer to the provider of inputs. Foreign investment should be under the
government control and foreign investors should be facilitated to go into joint ventures
with local firms. The only way that Pakistan can establish an innovative technological
base is by diffusing imported technologies to domestic firms. Therefore, the government
should import foreign technologies only in the form of technology licensing, technology
transfer agreements with foreign firms, capital goods imports and reverse engineering.
Box 3‐3: Electronic Industry in China
China started from focusing on developing R & D institutes and training centres in early 1950s. By 1959,
25 Chinese national universities were teaching classes in electronics fields, including radio technology,
vacuum electronics, semiconductor technology etc. it started developing this industry initially for military
application with soviet assistance but later shifted to other areas. Its impressive performance in this
industry over the years is mainly attributed to launching large projects of national importance aimed at
solving the common problems being faced. This includes ‘Golden series’ projects which include a
nationwide public economic information processing network (Golden Bridge Project), an electronic
monetary and modern payment system (Golden Card Project), a foreign trade information sources
network (Golden Customs Project), an electronic taxation system (Golden Taxation Project), the
industrial production and circulation information network (Golden Enterprises Project), an education
and research network (Golden Intellectual Project), an agricultural management and service network
(Golden Agriculture Project), and a national economic micro-policymaking support system (Golden
Policy Project). In addition, long-range national planning encompassing the electronics industry includes
a semiconductor project called the 909 semiconductor manufacturing projects, the air traffic control
system project, and the Three Gorges Dam project.
China has been luring outside investment with incentives like subsidized loans, tax exemptions, and a 50
percent discount on land. FDI has focused on attracting high-tech industry; upgrading foreign-invested
enterprises beyond simple processing and assembling; and inviting foreign purchasing, venturing, or
investing in state-owned enterprises. China’s policies for encouraging foreign trade include the following:
• The government has designated 5 SEZs, 14 open coastal cities empowered to exercise the same
policies as the SEZs, and a series of zones along the Chinese coast connecting the SEZs and open
cities to form a coastal opening belt; these offer attractive incentives for foreign investment and
trade.
• A government-sponsored export network includes several hundred factories nationwide,
producing a range of products; participants receive guaranteed supplies of electrical power, raw
materials, tax reductions on inputs, and attractive purchase prices.
• Special investment funds are made available by Ministry of Foreign Trade & Economic Cooperation
(MOFTEC) for the technological upgrading of selected enterprises.
• Preferential treatment is given to Sino-foreign joint ventures if they are categorized as either
export-oriented or technologically advanced projects; for example, enterprises that export 70
percent or more in value of their products may reduce their income tax liability by half at the end
of the tax reduction or exemption period.
• Local enterprises are exempted from import duties on raw materials provided by overseas
suppliers to meet export contracts or for use in manufacturing exports.
• The People’s Bank of China (PBOC) offers trade credits in domestic currency to exporting
enterprises (most of them foreign trade corporations (FTCs)) to finance exports.
• PBOC offers export seller’s credits to Chinese 197 enterprises selling electronic and machinery
equipment in the international market.
• Credits (in foreign currency) are extended to buyers of complete sets of Chinese-made machinery
and electronic equipment valued at a minimum of US$1M per transaction

3.2.2.1.4 Trade through Illegal Channels and Under Invoicing
Local industry is able to meet the 75% of the domestic demand. Rest is imported to fill
the gap. Current import duties on luxury consumer goods encourage people to invest in
illegal activities such as smuggling through the Afghan transit trade (ATT) and under
invoicing of imported products to cut duties. The local market prefers the latest product
and buys from smugglers or their comrades. 61 For example, one industry executive told
during the interview that their latest LCD is available in the local market which they
have yet to launch. This not only affects the industry but also a loss of revenue for
Government. Similarly, under invoicing is one of the issues which also need to be dealt
with.
3.2.2.1.5 Strategic Direction
Pakistan has lagged behind in the development of its electronics industry; therefore, a
coherent strategy is needed to be put in place to develop this sector with a view to
increasing the country’s growth potential as well as achieving self-sufficiency by
reducing dependence on foreign sources of products, materials, components and
equipment. China, Singapore and other Asian countries are examples to learn where
countries without strong local capabilities have become major exporters of electronics
products in international markets. They paid attention to integrated production
systems, starting by performing relatively simple assembly. Many countries have
managed to upgrade their role by moving into greater local content, design and
development, regional marketing and so on. China and Singapore is worth mentioning
for advanced electronics, with impressive design capabilities and growing local linkages
(Box 3-3 and 3-4). The strategy adopted by Singapore was based on FDI and the country
followed a strict focused criteria. If Pakistan adopts the same strategy FDI will definitely
61 ‘National Electronic Policy’ Engineering Development Board, p.20.
198

add to resources and capital formation. This will transfer production technology, skills,
innovative capacity and international marketing network.
Given our infrastructure, however, we believe that it would be very difficult for us to go
directly into manufacturing of components. It would be more prudent for Pakistan to
develop capabilities in system design instead. This is the model India has followed.
System design involves innovation in architecture and function. Using mostly off-the
shelf components, entrepreneurs innovate to create new designs. In due course of time,
manufacturing of Printed Circuit Boards and other components follows.
Box 3‐4: Singapore Electronics Industry
The Singapore's semiconductor industry has grown from humble beginnings as an assembly-
and test-subcontracting supplier to a fully integrated, cutting-edge technology wafer
fabrication hub. It was because Singapore Government focused on learning, technological
acquisition, rapid movement up the industrial ladder, and the skills of its working population.
On the other hand, their government provided capital, tax concessions, infrastructure,
education and skills training, and a stable and friendly business environment. Following are
the salient features of Singapore’s strategy employed over the years:
• Economic Development Board (EDB), a governmental body that had more autonomy
than normal government ministries established in 1961, acted as a one stop shop for
developing the new industries in Singapore.
• EDB was governed by its own Board of Directors. It has freedom to develop its own
salary structure and promotional grades and procedures, conduct commercial affairs
etc, without being bound by governmental rules and regulations.
• Reduced the tax rate for approved industries to 4 percent. The incentive could be
enjoyed for up to 15 years.
• Taxes were also reduced on royalties, license fees and R&D costs payable to overseas
enterprises.
• Export incentives were also given by way of duty-free inputs as well as enhanced tax
allowances for market development expenditures overseas.
• The Government also sought to encourage the establishment of even larger firms by
exempting firms with fixed assets in excess of S$150 million from income tax for up to
15 years.
• After 1970, Initial emphasis to develop low-tech, labor-intensive industries was
changed to high-tech industries.
• To facilitate the transformation, EDB initiated overseas industrial training schemes,
joint
Government-industry training centres and provided local industrial training grants for
developing necessary skills.
• To attract Multi National Companies (MNCs) they provided substantial cuts in taxes
and fees to reduce the cost of production. MNCs in electronics industry in 1970s and
early 1980s were a major thrust for Singapore.
• Since 1991, Singapore’s strategy revolved around industrial clusters. Ten areas are
singled out for special attention by the Board. These are: electronics, chemicals, life
sciences, engineering, education, healthcare, logistics, communications and media,
headquarters and promising local enterprises.
199

3.2.2.2 Policy Recommendations:
The competitive advantage of our electronics industry must be shifted from low wage
labor manufacturing organized by multinationals to low cost, rapid ramp-up, high
volume and increasingly automated manufacturing activities. In order to achieve this,
we propose the following:
• New courses in systems design must be initiated across engineering universities
around the country. These course need to be designed jointly by faculty and
industry experts.
• In order to kick start this process, the government should initiate a series of
workshops open to related entrepreneurs, faculty and business people. Foreign
experts should be invited to conduct these hands-on design workshops.
• The government should also identify other mechanisms to transfer design skills
to Pakistan.
• Where there is demand of a particular electronics item, e.g., energy savers, the
government should first look to local parties. If a part can be manufactured at
home, local vendors should be preferred.
• As a matter of priority, the government should facilitate anyone who wishes to
initiate manufacturing of Printed Circuit Boards in the country.
• EME College should be given a grant to establish an internationally accredited
design, testing and certification institute.
• The government will also match any investment geared at acquisition of modern
and emerging electronic technologies.
3.2.3 Pharmaceutical Industry
The size of the local pharmaceutical market is around US$1.5 billion. The local
manufactures comprise 60% of this share and the multinationals account for the
200

emaining. Over the recent years, the local industry has experienced impressive growth
(double digit) as compared to multinationals in the local market. Industry sources
report that the potential in this market is huge62. The market size can easily exceed
US$2.3 billion by 201262 in Pakistan as incomes increase and the health expenditure as
percentage of GDP increases (currently 1% - US$7/8 per capita spending). The industry
is currently struggling due to lack of chemical industry in the country, poor governance,
lack of compliance with standards and electricity shortages. In addition, the industry
also complains that strict regulatory control on prices of about 900 active ingredients
creates distortion in the market and impedes effective supply to end consumers.
Another area where the local industry has raised concerns is that the government policy
for pharmaceutical sector by design has always been pro-multinationals (MNCs). The
policy allows MNCs to act as price leaders in the market. As most of the MNCs are
driving their profits out of transfer pricing, they set drug prices at unreasonably low
prices. This creates a disincentive for the local industry to compete.
In comparison, India has always had a strong protectionist policy for its pharmaceutical
industry. During the protection period the Indian industry rode high on the bases of
large local demand. The Indian industry gained significant momentum and size over the
protection period. The difference in scale is evident with Pakistan having only 400
manufacturers and India having 25,000 units. The market share dispersion is also
extremely different in India as compared to Pakistan. In Pakistan the MNCs are the
largest and there is a significant difference in size of an MNC and the largest local
industry, on the other hand in India, whereas, the MNC are part of the top 10 the
shares/size are almost equal to the local companies. Under the protection period the
Indian plants were able to invest significantly in production capacity and using the
strong support of their ancillary industry such as chemical was able to emerge as a
global leader in the pharmaceutical industry.
The Indian industry felt threatened to the introduction of the TRIPs agreement as until
that time the industry was only producing drugs through reverse engineering. But with
62 Pakistan Pharmaceutical Manufacturers’ Association (PPMA)
201

new patent laws under the TRIPs which protected not just the process but also the
product made posed new challenges for the industry. However, with government
support and strong management in the Indian pharmaceutical industry the challenge
was seen as an opportunity. The pharmaceutical industry for the first time post TRIPs
realized the gains to me made through research and development. The industry started
to invest heavily in new drug formulations and started applying for patents. Box 3-5
below presents the recent developments of some of the pharmaceutical companies in
India. Indian ratified the TRIPs agreement in 2005, and in a little less than 5 years the
Indian industry has emerged a leader in original drug formulations as well. This policy
of enhanced expenditure on research and development aided by the policy of outward
FDI has resulted in Indian companies acquiring several pharmaceutical companies
around the globe. This strategy has helped India emerge even stronger post TRIPs
agreement.
Box 3‐5: New Drug Filings of Some Successful Indian Pharmaceutical Companies
1. Ranbaxy Laboratories is one of the most successful companies and have filed six Abbreviated New
Drug Applications (ANDAs) in 2008 showing extreme focus on research and development. It also has
the largest basket of products for the US market with 141 approved drugs and another 98 applications
waiting in pending. The company has also filed 271 Drug Master Files (DMFs) comprising 48 Active
Pharmaceutical Ingredients (APIs). The company has also managed to secure patents on 12 of the total
185 patents it has applied for.
2. Dr. Reddy’s has so far filed in 122 cumulative ANDAs and have already received 13 final approvals for
the USA and 4 from Canada. In addition 7 approvals are tentatively granted by the US. Moreover, Dr.
Reddy’s has the largest APIs pipeline in the Indian pharmaceutical industry and has filed 23 DMFs only
in 2008. In total they have filed for 281 DMFs, out of which 127 are in the US.
3. Aurobinda filed the highest number of ANDA’s and DMFs from India in 2007-2008 (47 were filed
and 17 were approved). It has approval on 67 ANDAs from the US FDA and has also filed in a total of
318 patents. Globally the company has 1017 DMF filings.
Source: Company websites
The Pharmaceutical industry is also suffering due to lack of compliance with the quality
and standards requirements of international markets. Majority of the plants/units in
Pakistan do not conform to the minimum quality standards, certified production
processes, health & safety measures and quarantine issues. This has been a result of
inadequate investment, lack of awareness and the local industry doing business for ‘life
202

style management’ and not processing the managerial capabilities of building
corporations. India, on the other hand, has over 800 WHO certified units; Pakistan has
yet to come up with one. Another important compliance requirement is the enforcement
of ‘Good Management Practises’ (GMP). GMP compliance is a necessary requirement of
the US FDA. Several local companies whereas, have managed to get GMP compliance
certificates but implementation of requirements are not adhered to (i.e.) compliance is
only limited to process and not to functional implementation.
The local pharmaceutical industry in Pakistan is still predominantly small scale. The
reason is that the local industry is mostly financed by equity and none of the bigger
groups in Pakistan have ventured into this industry due to the lack of technical
knowhow. Pharmaceutical industry in normally classified as a knowledge based industry
and Pakistan generally lacks in availability of trained human resource required by such
an industry. Moreover, the other industrial sectors, such as textiles, offer too high an
incentive in terms of handouts and subsidies that the opportunity cost for entrepreneurs
entering into pharmaceutical is too high.
Other areas of advantage for India include; (i) lower minimum wage (Indian Rs 3,600 as
compared to Pakistan where the wage is Rs. 7,000); (ii) Energy cost to industry is
Indian Rs 2.5/unit in India, in comparison the average cost of electricity in Pakistan
from the grid is on average Rs 12/unit and through diesel generators is around Rs 24-
25/unit. India has consistent energy for industry, whereas, industry here is working half
its shift on self generated energy. Such cost implications imply lower profits and hence
lower investments. Moreover, as discussed above, there is hardly any innovation in
Pakistan even at the stage of raw material. Whereas, India has developed indigenous
technologies, and the industry there is working on innovating low cost formulas. Small
manufacturers normally concentrate on ‘low cost drugs’, whereas the bigger players are
making more sophisticated drugs. Figure 3.-8 below compares the broad contours of a
typical pharmaceutical value chain in Pakistan and India.
203

Figure 3‐8: Broad Comparison of Pakistan and Indian Pharmaceutical Value Chain
APIs – raw
material
1. Pakistan: 95% imported APIs.
India: 5% imported API –
remaining are produced in
India.
2. Exchange rate devaluations
hurt Pakistan industry
significantly, no such adverse
impact on Indian industry.
3. Most of the API’s produced by
India are certified and hence
they are able to export. Little
incentive in Pakistan to make
APIs due to lack of quality
control and certifications and
manufacturing APIs require
economies of scale so exports
Formulation/
B lk
At the
formulation
stage no major
difference
exists between
Indian and
Pakistan –
Indian
industry
however, has
all the required
certifications.
Another area
where Indian’s
are adding
value is
Batch manufacturing,
packing &
The next generation success in the pharmaceutical industry is likely to come in the Biotechnology
industry. This is one of the sun rise industries where a lot of growth is
expected to happen. Moreover, Bio-technology industry for a country is also important
from a strategic point of view. Given this importance of the Bio-Technology industry
India has recently announced a comprehensive ‘Bio-Tech’ policy aimed at attracting
investment in the sector, training human resource and providing support for the
requisite hard and soft infrastructure. In addition, the Indian policy provides several
fiscal benefits such as tax breaks, research and development grants to businesses
venturing into the Biotechnology industry. On the other hand, Pakistan has not even
started the thought process on establishing a Bio-technology industry. The stakeholder
analysis conducted for this report revealed that current there is only one company in the
entire pharmaceutical sector of Pakistan that has moved into bio-tech. In their
experience, the government’s current policy stance has been more restrictive rather than
attractive towards bio-technology industry. It is necessary for the Pakistan’s policy
204
Packing &
distribution
no major
difference
Pharmacies,
hospitals, exports
Pakistan has developed
strong local network to
sell pharmaceutical
products through sales
agents. However, on the
export side the
branding and quality
compliance issues stay
a significant weakness.

makers to introduce significant incentives to motivate investors to invest in this
critically important sector of the future.
Other factors that are hampering the growth of the local pharmaceutical include the
following:
• Inadequate branding: The branding should focus on product differentiation and
the quality of the drug.
• Regulatory barriers: The list of Active Pharmaceutical Ingredients (APIs) that can
be imported keeps on changing at short intervals based on the interests of the
MNC lobbies.
• Weak enforcement: Inadequacy of enforcement of counterfeit drugs is also
retarding local investment.
• Capacity & behaviour of health inspectors: The health officers do not possess the
requisite skills and knowledge of the pharmaceutical industry to conduct health
and safety inspections. Inspectors normally, use petty excuses as a way of rent
seeking. Furthermore, the industry feels that under the 18th amendment the
transferral on inspection to provincial inspectors will be disastrous as it will
increase the ‘informal’ cost of doing business significantly.
• Inconsistency in import duty: The import duty on pharmaceutical raw materials
(not manufactured locally) was to be fixed at 5 percent (as proposed in the
Federal Budget for 2005/06). However, a number of items are still assessed at
10-15 percent as the budget policies have not been fully implemented.
• Exemption from Sales tax: The sales tax on local purchases of packing material
such as cartons, leaflets, bottles, ampoules, labels, corrugated boxes, aluminium
foil, rigid PVC, gelatine capsules, and other such items is a heavy burden. This tax
is inconsistent with the government's policies to keep the prices of
pharmaceutical products affordable for patients. In keeping with such policies,
the imports of pharmaceutical raw materials, and the local purchases of such
materials, are exempted from sales tax; the pharmaceutical industry argues that
it would be logical to extend the same treatment to the packaging for
pharmaceutical products.
205

• Income Tax: The advanced income tax on import letters of credit has been
increased to 3 percent from 1 percent in previous years. This has made a
significant increase to the cost of production, because the refunds of this tax (if
and when made by the tax authorities) take at least two to three years.
• Environmental Degradation: Another area which the federal and the provincial
authorities should promptly deal with, is the environmental degradation created
around pharmaceutical manufacturing plants by other industries. Such pollution
often creates health hazards and therefore impacts with particular severity on the
pharmaceutical industry. It also gives purchasing missions from overseas the
impression of dirty and unsafe conditions, and thereby affects the chances for
increasing exports.
3.2.3.1 Policy Recommendations:
It is important to pursue a particular policy on the pharmaceutical sector in order to
correct market failures (since social benefits exceed private ones), fostering innovation
and promoting competition. Therefore, we recommend the following:
• Prices should not be regulated for drugs whose markets are competitive or
monopolistically competitive. Price regulation should only be there in the case of
a monopoly or collusive oligopolistic behaviour.
• Allow duty free imports of all APIs (active pharmaceuticals ingredients) and
machinery for both domestic and export markets – However, this benefit should
be lifted as the domestic industry picks up and protection should be provided for
WHO pre-qualified APIs manufactured locally.
• All locally purchased items like packaging material, etc. should be exempted from
sales tax.
• The government should share mark-up cost on loans and allow investment
adjustment against future taxes for putting up international standard
manufacturing facilities targeting the pharmaceutical markets of the US, UK,
Australia, Japan and also those in developing countries.
206

• Provide incentives such as tax breaks for carrying out R&D on molecules and all
ancillary activities like setting up CROs (Contract / Clinical Research
Organisations) subject to meeting certain targets.
• Incentives , such as the carryover of net operating loss (NOL) for purposes of tax
deduction, should be provided for new start-ups in Bio-technology. Other such
incentive is to make it attractive for such high technology industry to locate
themselves in science parks proposed earlier in the report.
• The government should provide tax breaks for setting-up internationally certified
bioequivalence / bioavailability labs, local manufacturing / fabrication of
machines and other hardware.
• Allow long-term subcontracting with drug manufacturers as opposed to 2 years
as per the current rules.
• Ensure internationally acceptable manufacturing quality standards / (Current
Good Manufacturing Practises) CGMP compliance in both the local and
multinational manufacturing companies in the country
• Ensure continuous power supply to the pharmaceutical manufacturing to ensure
international (Good Manufacturing Practises) GMP compliance of maintaining
the required levels of temperature control. This will not only ensure high quality
of the locally produced medicines but will also make the industry competitive
internationally.
• The government should introduce technical courses relevant to the
pharmaceutical industry in universities or other institutions. Representatives of
the industry contended that, for example, it was next to impossible to find trained
mechanics to deal with the electronics that are in almost all the machinery used
by manufacturing units.
207

3.3 Value Added Skill & Engineering Based Sectors
The industrial sectors considered in this section fall in the category of SME business. At
the outset we have provided policy prescriptions for the SME sector as a whole. The
Sector specific policies that follow are designed in line with these general policies.
3.3.1 The SME Sector
According to the most recent study conducted on SMEDA [SME Policy (2006)],
Pakistan’s premier agency for supporting small and medium sized enterprises:
“Pakistan’s economy is dominated by SMEs with more than 90% of enterprises
belonging to this category. The SME sector is suffering from many constraints including
lack of access to finance 63 , limited access to markets, lack of infrastructure, hostile
business environment, corruption and red tape, weak management and lack of access to
skilled labour. Also, many of the government policies are devised from the perspective of
large firms and not SMEs. The implementation of SME policies in Pakistan is
fragmented and limited and needs to be more effective in light of the SME sector’s
importance and contribution”.
3.3.1.1 Policy Recommendations:
Clusters of SMEs can also meet the requirements of the new competitive global
marketplace very efficiently and effectively. Therefore, we can no longer treat SMEs as
unproductive and marginal, but these have to be characterized as key elements of
industrial development and growth. Therefore, we recommend the following:
63 Small and Micro enterprises are often rationed out of the credit market due to information
asymmetries such as opaque knowledge of firms and sectors on the part of commercial banks.
Information asymmetry results in relatively high collateral requirements for SMEs which along with the
degree of documentation required for loan application deter them from resorting to the formal sector
credit market – commercial banks. Lack of credit availability thus severely impedes the growth potential
of small scale and micro enterprises. Data suggests that over 90 percent of small enterprises finance their
business and working capital requirements through retained earnings.
208

• A cabinet committee for SMEs should be established to fast track decisions on
SME policy matters.
• Exemption limit for excise duty should be raised from Rs. 5 million to Rs. 10
million.
• A capital subsidy for investment in technology should be provided.
• Incubation centres should set up in sunrise industries. Some of the sunrise
industries include:
o Bio-technology
o Information Technology
o Electronics and telecommunication equipment
o Non-conventional energy sources
• Establishment of testing and certification facilities should be financed.
• A vendor development programme should be initiated under which ‘buyer –
seller meets’ exhibitions are organised at regular intervals and at dispersed
locations.
• A fresh census of small scale industries should be conducted with special focus on
gathering information on the incidence of sickness of industries and its causes.
• Skill development initiatives specific to SMEs should be initiated. Schemes
covering the areas highlighted below should be developed:
• Training should be provided in quality management systems
• Training support for entrepreneurial and managerial development of SMEs
• Training and benchmarking for designing lean manufacturing techniques
• SMEDA should be institutionally restructured to enable it to design sectoral
programs for the SMEs that involve tailor-made investment projects in various
sectors. The new organization structure should also enable SMEDA to interact
with the government (federal and provincial) and its entities for developing
favourable policy environment for SMEs. SMEDA’s new organization structure
should also allow it to network domestically and globally to bring the maximum
benefits for the SMEs of Pakistan.
209

• Adequate human resource with greater financial flexibility should be made
available to run the revised organisations structure.
• The joint equity participation fund should be revitalised and should be allowed to
provide loans up to 25% of the project value with a maximum cap of Rs. 10
million. The mark up charged on these loans should be 5%.
3.3.2 Fan Sector
Global fan trade is classified on the basis of energy consumption. Fans consuming less
than 125 watts of energy are generally referred to as domestic fans and fans consuming
over 125 watts are classified as industrial fans. Fan industry in Pakistan is mainly
dominated by production of domestic fans. Hence, by its very design is missing on the
opportunity offered by the high value added and the growing value added market of
industrial fans.
Pakistan’s fan industry is mainly clustered in four major cities namely, Gujrat,
Gujranwala, Lahore and Karachi. However, 98% of the countries production is centred
at Gujrat and Gujranwala. The sector comprises over 450 SMEs, of which 300 are based
in Gujrat and the rest in Gujranwala. The industry produces on average 8 million fans a
year with an estimated value of Rs 17 billion. Out of the total production, approximately
30% fans consist of pedestals, 7% brackets and the remaining 63% are ceiling fans64 .
The industry belongs to the light engineering industry category, and is one of the
industries that existed at the time of independence.
The sector, whereas, has shown high levels of growth in the recent years suffer from low
levels of productivity, inadequate technology upgrade and shortage of skilled staff.
Moreover, most of the companies operate under locally created brands with only a
couple moving to international branding of their products. Even the local branding is
64 ‘Pakistan Light Engineering Sector’, Board of Investment, 2009: 18
210

sharply fragmented with 4/5 large companies representing over 60% of the local market
share. The industry also requires testing and certifications of their products mainly
conducted or required for electrical safety. Certifications are normally required for
export markets, whereas, general performance and safety testing are conducted
regardless of the market. As a result of the field exercise we have identified that those
firms which have large local market demand are better at innovation and new designs as
compared to those firms which are solely focused on export markets. The reason for this
being that local market provides enough stability for companies to invest in research
and development. Moreover, the companies supplying in the local market are better
connected to their end consumers and hence are able to incorporate customer feedback
in a more diligent manner. Additionally, it was also found that current export markets
are not necessarily offering higher margins as compared to local markets.
The table 3-4 below presents the main characteristics of the industry:
Table 3‐4: Fan Industry Characteristics
Characteristic Value
Number of units 450
Total Installed Capacity 9.5 – 10 Million Fans
Current Production 8 Million Fans
Contribution to National Exports 0.20%
Contribution to GDP 0.27%
Sector Employment 25-30,000
Capital Labor Ratio 6 workers/Million (Rs)
Total Estimated Investment Rs 5.0 Billion
Source: CMI Data, PEFMA interviews and ‘Pakistan Light Engineering Sector’, BOI, 2009: 20
Out of the 450 companies only 5-6 companies can be categorized as large scale
integrated manufacturing units. These units have in house capacities to conduct most of
the production processes and are also characterized with higher levels of investment and
more modern technology. Tier 2 companies are numbered somewhere between 40-50
and have medium sized units and in house capacities to conduct major portion of the
production process, however, lacks in investment and modern technology. The
remainder can be categorized as Tier 3 companies with small operations, high degree of
outsourcing and outdated machinery. The average age of the machinery employed is
211

etween 5-10 years. Sales are also fairly concentrated with five large firms in Gujrat and
three in Gujranwala, accounting for 40% of total industry sales.
The fan sector contributes in multiple ways to the national economy (Table 20). It offers
pro-poor employment creation opportunities, income generation, foreign exchange and
social development by strengthening cluster development in and around Gujrat and
Gujranwala. It offers direct employment to around 25-30,000 people. However,
employment is seasonal as most factories operate for only five to six months during the
year. This scale of employment is far below its potential as the industry currently faces a
seasonal demand. The sector has recently picked up in exports as well. Figure 3-9 below
shows that foreign exchange earnings of the sector are consistently on the rise. For the
most recent year (2009) the exports crossed US$32 Million. This has increased
Pakistan’s share in world fan exports to over 1%. The sector contributes 0.2% to
Pakistan’s total exports. The sector has experienced double figure export growth over
the last five years and hence has continued to contribute more and more to national
economy.
Figure 3‐9: Pakistan’s Fan Exports, 2004‐2009 (US$)
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
12.6
15.5
18.4
212
23.4
27.5
32.0
2004 2005 2006 2007 2008 2009
Source: UN Commodity Trade Statistics
3.3.2.1 Value Chain Analysis
Pakistan Exports Mean Exports
Value Chain analysis tool is employed to explain the specific issues faced by the
producers of the fan industry. Where data is available analogies with reference to the

global value chains are also made to identify critical weaknesses. The analysis points out
that there are several supply side constraints as well as weaknesses in value addition. It
must be emphasized that the value chain presented below represents what is typical for
the industry. It is possible that some larger firms may be slightly better placed and some
small ones slightly adversely placed than the representation presented below. However,
the numbers provided below have been verified by industry representatives.
The overall structure of the value chain (Figure 3-10) suggests that industry typically
adds 23% in value addition of around Rs 450 per fan. The vendors are also adding
almost same amount of value addition, however the importers take the major bulk of the
value addition.
Figure 3‐10: Typical Value Chain of Fan
Vendors &
Suppliers
50% of the materials are
imported.
Component Cost (Rs)
Raw Material 1453
Casting 50
Assembly 67
Overheads 186
Marketing 20
Selling Price : Rs.2100
Raw
Material
81.8%
Casting
2.8%
95% of the costs are for electricity
consumption.
Major cost in this component is electricity
charges.
Assembly
Value Chain analysis has revealed the following issues facing the industry:
3.8%
213
Overheads
10.5%
Marketing
1.1 %
Importers

3.3.2.1.1 Low levels of productivity
The industry composition suggests that it is dominated by small firms and as such does
not benefit from economies of scale. This is a critical reason why the fan industry in
Pakistan is not able to compete on costs with is international competitors. The average
capacity of a typical firm is around 200-300 Fans/day which is extremely low as
compared to Chinese counterparts where firms average productivity is 45-50,000
Fans/day.
Another reason for low levels of productivity is capacity utilization. The numbers in the
table 20 above indicate that the capacity utilization for the industry is over 80%. This
number however is misleading as the industry currently operates seasonally. Figure 3-11
below suggests that industry only operates to full production capacity in the first five
months of the year. From July onwards the operations fall to a quarter of the total
installed capacity. This seasonal production is not only an impediment to investment
but is also draining out the skilled workers from the industry. The workers have little
incentives to train for an industry that only provides employment for 5 months during
the year. The seasonality is affecting the productivity and deterioration of skills in the
industry.
Figure 3‐11: Capacity Utilization over typical 12 Month Period
100
90
80
70
60
50
40
30
20
10
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Source: Board of Investment, Pakistan (2009: 23)
Lack of technological advancement is another reason for low productivity. Older
technology is being used as compared to its counterparts. For example, currently lathe
214

machines are used for body turning, if technology is upgraded to CNC machines the
productivity will increase by at least 4 times. Similarly, shifting drilling technology to
multiple drilling will increase production 3 times. For pedestal fans currently majority
of the industries use manual coiling techniques, moving to automatic coil inserters will
increase productivity by 5 times. Similarly, over 90% of the firms are using spray paint
technology with wet paint. This technique has a high wastage ratio. Around 40% of the
paint is wasted and furthermore, this technique is environmentally hazardous also
[PEFMA interviews]. The industry needs to move to more efficient techniques, such as
electrostatic powder coating paint. This technology is not expensive and provides much
better results in terms of quality and avoids wastage. The change in technology will not
increase productivity but will also increase quality and performance of the product.
When considering global value chains, China’s production processes are much more
efficient and highly productive. In addition to technological gaps, productivity is also
affected by weak production process flow management. The assembly line in factories
need to be rationalized so that the existing set ups achieve better technology.
3.3.2.2 Difficulty in getting adequately trained manpower
As discussed above seasonality of demand is an impediment to skilled manpower. Fan
industry offers employment opportunities for 5-6 months which does not attract
workers at first place. Secondly, there are also inadequate training facilities for the
industry. This is an impediment to achieving better technology i.e. low level of skills is
impeding investment into modern machinery.
3.3.2.3 Lack of R&D and design innovation
Value chain analysis shows that average value addition in the industry is around 25-
30%. This low value addition is due to high content of input and low price fetched by the
product in the market. The VCA above reflect that the average price of a Pakistani fan
(ceiling) is around US$25 65 . This number goes down to US$20 66 if we take the overall
average export price. In comparison, fans from countries such as Finland, France, Brazil
65 Based on interviews with industry
66 UN Commodity Trade Statistics Database
215

etc are fetching much higher export prices. Figure 3-12 below provides a comparison of
per fan export price of various countries.
Figure 3‐12: Average Export Price of Fans US$
17
16
16
16
11
20
31
30
45
50
76
216
122
0 10 20 30 40 50 60 70 80 90 100 110 120 130
Source: UN Commodity Trade Statistics Database
France
Finland
Netherlands
Brazil
Thailand
Italy
Germany
Spain
USA
India
China
Pakistan
This difference in price is mainly due to better designs, better quality, marketing and
branding. Some of the high value added ceiling fans sell for around US$500-600 in the
US retail market. This analysis presents an opportunity and a threat. The industry
should not only focus on producing and exporting more fans, in fact, it should focus
more on value addition and making better design and innovative fans that fetch higher
prices. Given the nature of the industry value addition seems only possible if Pakistan
begins to make fans for high income markets.
VCA also reveals that over 80% of the cost represents material and parts required to
produce fan. Out of this 50% of the materials are normally imported. Therefore,
compared with international competitors such as China the local industry is at a
disadvantage due to lack of research and development in materials. China has been able
to diversify its production of materials required for fan manufacturing moving into PVC,
composites of metals etc. Pakistan on the other hand is still relying on pure materials
which are not cost effective.

There is also lack of training institutes and research and development centres. The
industry has access to Fan Development Institute, however, the capacity of this institute
to do meaningful research and product development is fairly limited. The institute
currently works as a service centre, provides basic testing and certification facilities in
partnership with Intertek. The institute compares unfavourably to the research and
training facilities that are available in countries which compete with Pakistan in
International markets.
3.3.2.4 Inconsistent Quality and Certification
The industry generally feels that improving quality and standards is extremely
important. The Fans standards are normally two types; performance and safety. There
are no defined quality or safety standards for Fans in Pakistan. This practice has led to
inefficiencies in the system. First, lack of quality and standards result in customary
production techniques and a variety of different part being used by the industry. Quality
is difficult to maintain when there are non-standard parts being employed. Therefore,
lack of standards at local level results in unreliable quality in domestic market.
Furthermore, moving to higher value product in export markets requires complying
with certain quality standards and attaining certifications. Currently testing facilities are
inadequate for this purpose.
3.3.2.5 Policy Recommendations:
The light engineering sector is playing an important role in employment generation and
poverty alleviation through endogenous technology. It consists of small and mediumsized
light engineering industries, like the fan cluster, which have been producing
import-substitution products. Our recommendations are the following:
• Create awareness and facilitate common branding and consortium-building of
companies to benefit from scale economies. National Productivity Organization
(NPO) will support the cluster actors to develop standardized parts and move
towards common production and sourcing.
• Strongly advocate to State Bank to ensure effective implementation of initiatives
provided for SME financing at reduced mark ups.
217

• Strongly advocate with State Bank and commercial banks to ensure availability
and utilization of long term project financing and technology up gradation. Share
mark up costs for technology upgradation.
• Provide funding to Fan Development Institute to upgrade machinery at the
common facility centre.
• Establish training facility by supporting TEVTA and FDI under PPP
arrangements to provide labor training on modern machines.
• Establish a centre of fan engineering and designing at the University of Gujrat.
• Work with Pakistan Electric Fan Manufacturing Association (PEFMA) to develop
domestic standards for fan manufacturing. The standards will be enforced and
monitored by PSQCA.
• Share costs to meet compliance with international requirements such as CE
Marking, UL Marking, etc.
• The government should facilitate joint ventures with foreign companies and
ensure technology transfer.
3.3.3 Cutlery, Utensils & Hunting Equipment Sector
The cutlery and hunting equipment industry is mainly clustered in and around
Wazirabad with some production also taking place in Gujranwala, Sialkot and Dir. 95%
of the countries production is centred at Wazirabad. The sector comprises over 400
SMEs, of which 250 are involved in cutlery manufacturing, whilst the remaining is
involved in production of hunting knives, swords etc. The industry produces on average
4.3 million different types of products with the value estimated at over Rs 6 billion.
Global trade in cutlery is classified as; kitchenware cutlery (HS Code: 8215, 821591,
821510, 821520) and non-kitchenware cutlery (HS Code: 8208, 8211, 8212 and 8214)
where non-kitchenware includes swords and hunting equipment. Cutlery industry in
Pakistan is involved in both kitchenware and swords and hunting equipments.
The industry belongs to the light engineering industry category, and is one of the
industries that have existed prior to independence. This is one sector where Pakistan
218

over the years has developed capabilities to penetrate high value / high income markets
such as Germany, USA, France, Belgium etc. The average export price of goods made in
Wazirabad is around $25-27/Kg (Steel), which is much higher than the Chinese
products which on average fetch (US$3-5 (Composite material)). However, the price is
lower than some of the more sophisticated producers such as Germany and Switzerland.
The sector, whereas, have achieved reasonable export growth in the recent years have
suffered from intense competition from China especially in the cutlery sector. The major
impediments of the sector include low levels of productivity, inadequate technology
upgrade and shortage of skilled staff. Moreover, most of the companies operate without
any brands with only a couple moving to branding of their products. Furthermore, the
industry in the years to come will face higher compliance requirements, especially the
cutlery manufacturers, who would be required to meet standards on use of ‘food grade
materials’. Currently not much compliance or testing requirement exists and only a few
companies adhere to ISO standards.
Table 3‐5: Cutlery and Steel Ornaments Industry Characteristics
Characteristic Value
Number of units 400
Total Installed Capacity 7.5 – 8.0 Million pieces
Current Production 4.3-4.5 Million pieces
Contribution to National Exports 0.25%
Contribution to GDP 0.11%
Sector Employment 10-15,000
Capital Labor Ratio 6 workers/Million (Rs)
Source: ‘Pakistan light engineering sector’, BOI , 2009: 28, and Industry interviews
Out of the 400 companies only 10-15 companies can be categorized as large sized
manufacturing units within the sector (relative basis). These units have in house
capacities to conduct most of the production processes and are also characterized with
higher levels of investment and more modern technology. Tier 2 companies are
numbered somewhere between 40-50 and have medium sized units and in house
capacities to conduct major portion of the production process, however, lacks in
investment and modern technology. The remainder can be categorized as Tier 3
companies with small operations, high degree of outsourcing and outdated machinery.
The average age of the machinery employed is between 10-15 years.
219

The above composition suggests that the industry is dominated by small firms and as
such does not benefit from economies of scale. This is a critical reason why the cutlery
industry in Pakistan has not been able to compete on costs with is international
competitors – and China has been able to easily replace Pakistani products in traditional
markets. The average capacity of a typical firm is around 50-60 Cutlery/day and 25-30
swords, hunting equipment/day which is extremely low as compared to Chinese
counterparts where firms average productivity is in thousands per day.
Nevertheless, the sector contributes in multiple ways to the national economy. It offers
pro-poor employment creation opportunities, income generation, foreign exchange and
social development by strengthening cluster development in and around Wazirabad. It
offers direct employment to around 10-15,000 people. This scale of employment is far
below its potential as the industry currently faces a stiff competition from China.
Unhelpful government policies and energy crises have resulted in closure of over 300
units in last few years. Figure 3-13 below shows that foreign exchange earnings of the
sector are consistently on the rise. For the most recent year (2009) the exports were
almost at US$50 Million. This has increased Pakistan’s share in world exports to over
0.4%. The sector contributes 0.25% to Pakistan’s total exports.
The export data suggests that Pakistan has been able to develop capabilities to supply in
majority of products categories, however, the scale and share of markets remained small
or insignificant. This presents with both an opportunity and a threat. It is an
opportunity as the market is large enough to allow Pakistani exporters to expand their
shares and increase export earnings. It is a threat because, if Pakistan does not upgrade
its production and improve efficiency, it may lose even the existing market shares to
new entrants such as Viet Nam.
Figure 3‐13: Pakistan’s Cutlery and Steel Ornaments Sector Exports (US$)
220

52.00
50.00
48.00
46.00
44.00
42.00
40.00
38.00
2005 2006 2007 2008 2009
Source: UN Commodity Trade Statistics
3.3.3.1 Value Chain Analysis
Value Chain analysis tool is employed to explain the specific issues faced by the
producers of the industry. Where data is available analogies with reference to the global
value chains are also made to identify critical weaknesses. The analysis points out that
there are several supply side constraints as well as weaknesses in value addition. It must
be emphasized that the value chain presented below represents what is typical for the
industry. It is possible that some larger firms may be slightly better placed and some
small ones slightly adversely placed than the representation presented below. The
numbers provided below have been verified by industry representatives.
The overall structure of the value chain (Fig 62) suggests that industry typically adds
43% of ex-factory sale price in value addition in the tune of around Rs 145 per dagger.
The vendors are also adding around 15% in basic raw materials; however the importers
take the major bulk of the value addition.
3.3.3.1.1 Uncompetitive Cost Structure
The value chain analysis of daggers using imported steel (Figure 62) depicts that over
60% of the costs represents the material and parts required to produce the dagger knife.
221

Out of this 70% of the cost is for the steel used. Although the local steel is low in cost but
mostly it do not meet the quality standards required by foreign buyers. Product finish is
poor for local steel. It is mainly used by cutlery manufacturers selling in local markets.
However, local steel do not qualify as ‘food grade’ steel and cannot be used for export
market cutlery. Availability of steel is an issue. The stocks and prices are quite variable
resulting in pricing problems for the industry. Damascus steel is one of the success
stories of the sector. Damascus steel is the highest quality steel for hunting knives,
swords and similar equipment. The sector has developed capacity to produce its own
Damascus steel. Some of the larger units have developed the capability to produce high
quality Damascus steel and are able to enter global market at much favourable terms.
In comparison, China which is the leader in global trade of cutlery and similar products
has a competitive advantage due to availability of alternative low cost materials. For
instance, the average export price of Chinese cutlery per Kg is US$3.5, whereas of
Pakistan is US$ 20. The main difference is that Chinese have captured a major market
share by competing on low costs and high turnovers. Their strong research and
development capacity in alternative materials have given them a clear advantage in the
market.
Other issues concerning cost structure are use of imported machinery and high
electricity charges. Almost 95 % of Chinese manufacture use local machinery in the
production process. They are able to purchase machinery at a fraction of the cost of
European machinery. Firms in Pakistan have to rely on imported machines which are
quite expensive. Hence, lack of support from engineering industry leads to a significant
competitive disadvantage. Similarly, electricity is the major input in all of the
production processes. Therefore, shortage of electricity and high electricity charges are
critical impediment to competitiveness. Manufacturers use diesel fuelled generators for
continuing the production which is expensive and cost of electricity for a typical
manufacturer comes to around US$ 0.20/Kwh as compared to US$0.11/Kwh for a
Chinese manufacturer.
222

3.3.3.2 Overall Value Addition & Productivity
Value chain analysis shows that average value addition in the industry is around 43%.
Value addition could be increased with reducing the input costs as discussed above and
fetching the high prices in the market. Input costs also affected with level of
productivity. The industry as a whole suffers from productivity issue. This has resulted
both due to inadequate investment in technology and also a lack of availability of skilled
labor. Most of the processes employ less productive technology. Comparison with the
Chinese value chain reveals that:
Figure 3‐14: Value Chain for a 12 Inch Dagger
• Pakistani firms use one piece die press for cutting, in comparison the Chinese and
other competitors such as Germany, France etc. use 12 piece heavy die press.
Hence, their process is 12 times more productive.
• Pakistani firms use hand hammering for straightening purpose, the Chinese
counterparts use automatic straightening machine. The hand hammering
technique can process up to 250 pcs/day and in comparison the automatic
machine straightening process 5,000 pcs in an 8 hour shift.
223

• Similarly, grinding is done using an open grinder in Pakistan, in comparison,
Chinese use belt grinders. The local practice can only process 100-125 pcs/day as
compared to belt grinding that can process around 600 pcs/day.
• Finishing of the products is the weakest area in the local value chain. Some of the
firms have shifted to automatic cleaning machines which have improved the
quality of the cleaning and also the speed. However, generally the sector relies on
manual labor doing the cleaning work. This process is slow and also has some
environmentally hazardous effects. Similarly, the polishing process is manual as
compared to Chinese firms which use vibrator machines. With current practices
one worker is able to polish a maximum of 100 pcs/day and in comparison
vibrator machine technology can process 1000 pcs/day.
• On the other hand, in terms of price Pakistan operates much favorably to China
(Figure 3-15). This suggests that Pakistan is not directly competing with China,
due to China’s focus on low value high turnover products. Pakistan needs to
capitalize on more sophisticated producers such as Switzerland, Germany, Japan,
France etc. All these countries have higher costs of production than Pakistan,
especially labor costs. Pakistan can compete with these countries more fruitfully
rather than trying to compete with China. Pakistan will not be required to invest
significantly in scale, instead, investments can be made: in sophisticated
technologies, improving quality and finishing and marketing and branding. Weak
branding and marketing is one of the main reasons for low value addition. Almost
all the firms supply unbranded products to international buyers, who sell these
products under their brands and are able to add more value.
Figure 3‐15: Average Export Price of Steel Used in Making Cutlery and Ornaments, US$/Kg
224

4
12
14
21
30
30
35
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90
Source: UN Commodity Trade Statistics
44
225
83
China
Brazil
Belgium
Pakistan
Spain
France
Japan
Germany
Switzerland
3.3.3.2.1 Difficulty in getting trained manpower
Pakistan is far behind in technology and modern production processes. As discussed
above this lack of technology results in inadequate productivity and higher costs of
production. The impediment to achieving better technology is not only lack of capital
but also lack of knowledge and appropriate skills to use modern technologies. The level
of skills is impeding investment into modern machinery. It is because there is no specific
training facility for the industry. Another issue is the lower wages in the sector as
compared to others which does not attract the workers.
3.3.3.2.2 Lack of R&D and Design Innovation
According to general competitiveness framework the sophistication of local customers is
an integral reason why companies are forced to innovate and then succeed abroad. In
this aspect the sword and hunting equipment industry is 100% export focused, whereas
cutlery industry generates 85% of its sales in the local market. However, the segment of
the local market which is catered by the cutlery segment is highly price conscious. The
consumers in the local markets are not willing to pay high price and are always looking
for low price products. This is a critical reason why the industry has not been able to
jump up the value curve even with massive domestic demand.
This structure of the local demand has led to little product development. The industry
continues to produce (with some exceptions) traditional cutlery using low cost and
below standard material and hence is unable to compete in the international markets.

The industry seems to be doing reasonable well in the export market which mainly
comprise hunting knives, swords and daggers. Pakistani products normally fetch higher
prices as compared to majority of other countries. There is significant demand for high
end products in the market and if the industry can enhance its productivity without
compromising on quality then it will be able to gain a greater level of market share.
There is also lack of training institutes and research and development centers to spur
innovation. Although industry has access to Cutlery Institute, however, the capacity of
this institute to do meaningful research and product development is fairly limited. The
institute compares unfavorably to the research and training facilities that are available
in countries which compete with Pakistan.
The government realizing the potential of this sector has already established a national
representative body namely the Pakistan Hunting Arms Development Company
(PHADC).
3.3.3.3 Policy Recommendations:
The interventions in the cutlery sector are directed at the cluster level. This lowers
transaction costs and helps generate relationships between enterprises which improve
their efficiency through the development of cooperation and maximize the potential of
the cluster through the development of mutual learning. Specific recommendations are:
• Continue to support the initiatives of the PHADC and ensure that the initiatives
carried out by PHADC are balanced in regional representation.
• Regularly monitor the performance of the PHADC against established targets and
goals to ensure that the company delivers to the needs of the sector.
• Finance joint projects between University of Gujrat and Pakistan Cutlery
Association to work on research and development of newer metals required for
the industry.
• Advocate with provincial government to ensure that TEVTA’s service centre in
Wazirabad also offers skill training specifically for the cutlery and swords
industry.
226

• Share mark up costs of technology upgradation conditional on companies
meeting set export requirements.
• Provide support to PSQCA to work with the sector associations to develop a set of
core standards, especially relating to the use of materials and their quality.
• Advocate with State Bank and commercial banks to ensure availability and
utilization of long term financing.
• Advocate with TEVTA Punjab to fully operationalize their cutlery common facility
centre.
• Engage Pakistan International Freight Forwarders Association (PIFFA) to run
export training programmes in Wazirabad and work with the sector association
to run training programmes on business processes, business English, etc.
3.3.4 Horticulture Processing Sector
Pakistan is endowed with a diverse range of fruits and vegetables. Citrus, mango and
dates from Punjab, mango and dates from Sindh, cherries, strawberries, peaches, plums
and apricots from the Khyber Pakhtunkhwa, and, dates, apricots, peaches and plums
from Balochistan, have a large domestic market in addition to a growing export market.
The country produces around 2.2 million tones of citrus, 1.7 million tones of mango, 0.7
million tones of dates and around 0.6 million tones of apricots, peaches and plums
(Pakistan Horticulture Development and Export Board, 2008). Even with this extensive
availability of natural resource, the value added horticulture industry has not grown
accordingly.
One critical reason for this lack of growth has been coarse market links between the
growers/fruit marketers and industry. Due to product seasonality the price of the fruit
varies significantly and so does the quality. Owing to the risk of non- availability of raw
fruit (fresh or frozen), the value addition industry feels hampered in making large
investments. In addition, even after several efforts made by the government and the
private sector, quality and compliance infrastructure is still inadequate. Furthermore, a
significant amount of fruit in Balochistan, Sindh and KP is going to waste due to adverse
conditions with processing plants lying idle and capital stock depleting.
227

Another major area that has restricted the development of horticulture value added
industry is the inability of the growing sector to meet the international quality and SPS
standards. The health and safety standards for food products have become extremely
strict in international markets and require certain protocols and procedures to be met to
qualify for market access. Developing countries such as Pakistan has faced several
export restriction due to non-compliance with such requirements. The small size of
farms makes it even more difficult to expand the certification and good practices
knowledge compliance. PHDEC has been working in creating awareness on Gobal Good
Agricultural Practices (GAP) and other similar Sanitary & Phytosanitary (SPS)
requirements impacting the potential of the horticulture sector. United Nations
Industrial Development Organization (UNIDO) has also been very active in supporting
Pakistan to achieve compliance with international quality and SPS requirements. The
EU Trade Related Technical Assistance (TRTA) II programme managed by UNIDO is
closely looking at the citrus and mango regions of Pakistan in terms of building trade
capacity and develop value addition opportunities.
Until such time Pakistan is able to grow fruits that satisfies most of the protocols
required by international standards it will be difficult to attract significant investment in
the value added industry. This is because that even with a population of over 170 million
there is still a minimal demand for products such as dried exotic fruits, canned fruit,
packaged juices etc. The little local demand that exists is entirely captured by either
imported products or products sold by the MNCs. To build demand locally the industry
will have to initially rely on export markets. However, if the fruit is unable to satisfy the
compliance requirements it will be impossible for products made out of these fruits to
meet the quality and certification requirements. Hence, the efforts to improve grower
capacity to meet requirements need to be strengthened.
Another issue that has restricted local investment in the value added horticulture
industry is the lack of implementation of local quality and standards. The value added
horticulture industry, for example, the juice industry suffers extremely from counterfeit
products and smuggling from India. It is impossible for local fruit juice makers that are
complying with all the standards laid down by the PSQCA to compete with counterfeit
228

juice manufacturers that do not comply with any standards. As the high income markets
are captured by imported products or MNCs such as Nestle, the local manufacturers
have to compete in low income markets. As price is the main criteria in these markets,
the counterfeit producers are driving out the genuine producers. Additionally, a lot of
low price/poor quality juice is being smuggled from India into the country. Some of this
is being sold in original packaging while some is brought in bulk quantity or concentrate
form and repackaged in Pakistan. In presence of such elements it will be difficult to
attract new investors into the value addition industry.
Finally, horticulture is a critical pro-poor sector and ties in well with the agenda of
achieving balanced development/growth. As shown in the later part of this report,
development in all four provinces of Pakistan has been sharply asymmetric. The areas
where industry has developed have become the epicentre of government’s focus and
tended to receive proportionally high amount of government spending. Most of the
underdeveloped areas are those where little or no industrial activity has emerged.
However, these underdeveloped areas are rich in natural resources and horticulture
production is the main activity in most of these under developed areas. As part of the
cluster mapping exercise conducted for this report we have identified several ‘hot
spots 67 ’ throughout Pakistan. The hot spots identified in the underdeveloped areas
comprise mainly fruit and vegetable plantations.
Furthermore, at the time of writing this report Pakistan was faced with severe floods.
The severity of destruction caused by these floods has mostly hit the already
underdeveloped areas, further destroying the minimal infrastructure that existed. In
recognition of this fact, it is reasonable to assume that government spending is likely to
increase in these flood affected areas. Given most of the flood affected areas are also the
least developed it makes evident sense for the government to structure its expenditure
in such a way that redevelopment work also encourages private sector investment and
industrialization in these areas. One such policy measure could be to provide fiscal
incentives to set up horticulture value added industry in these regions. The raw material
67 Areas where some form of clustering has emerged
229

for such industry already exists in these regions and hence if government channels its
redevelopment expenditure in such a way to build industry friendly infrastructure
connecting the hot spots, then private sector can be lured in to set up industry and
produce employment. However, this will require certain measures including compliance
requirements and strong fiscal incentives to attract private sector incentives.
3.3.4.1 Policy Recommendations:
Fresh vegetables and fruits are a good example of a non-traditional agricultural export
crop, and they illustrate the potential for agricultural diversification and production of
high-value crops. Our recommendations are:
• Keep on supporting PHDEC to improve awareness on quality and compliance
(establish PAKGAP Standards) and up-gradation of cool chain infrastructure and
also to advocate with the provincial governments to strengthen their strategic
support to upgrade the horticulture sector in meeting quality, compliance and
certification requirements. The provincial government departments responsible
for markets are to be requested to further facilitate linkage between the valueadded
sector and fresh fruit markets.
• Incentivise local manufacturing of juicing and pulping equipment by providing
technical information and support from research centres such as Ayub Research
Institute, Faisalabad.
• Strengthen PSQCA’s enforcement capacity to ban / fine all illegal / counterfeit
juice suppliers in the market.
• Provide special time bound incentives to initiate a support plan to restart
business activity in flood and war stricken areas. Special horticulture processing
and value added zones should be developed in interior Sindh, Balochistan, KP
and Southern Punjab. Land should be provided on attractive terms. The zones
should be declared exempt from all taxes and duties; however a target
employment for each zone must be set to qualify for this exemption.
• Develop seven clusters in the agro-sector including dairy cluster in Sahiwal, Pak
Pattan and Okara; Citrus cluster in Sargodha; Mango cluster in Multan and
230

Mirpur Khas; exotic fruit cluster in Swat; dates cluster in Sukkur/Khairpur and
exotic fruit and apricot drying cluster in Gilgit.
3.3.5 Surgical Instruments Sector
In the early twentieth century, clusters of surgical instrument production were found in
Sheffield (UK), Nogent-sur-Marne (France), and Solingen and Tuttlingen (Germany).
With the exception of Tuttlingen, none of these locations survived as significant centres
for surgical instrument manufacturing. Instead, a number of new actors, such as
Pakistan, Malaysia, Poland and Hungary, emerged as important producers of traditional
surgical instruments. Pakistan has seen consistent export success in the surgical
instrument sector for several years now.
The cluster consisting mainly of small and medium enterprises (SMEs), has its main
markets in the United States and Western Europe. Over 99% of Pakistan’s production is
generated from Sialkot. The industry produces on average over 150 million pieces a year
with an estimated value of around Rs 22 billion. Out of the total production,
approximately over 95% is exported68 . The industry belongs to the light engineering
industry category, and is one that has specialized in skill and stable export market share.
For the purpose of trade; four broad categories can be defined where Pakistan is
supplying in the export markets. The categories include; (i) HS Code 9018 –
Instruments for medical, surgical and dental; (ii) HS Code 9021 – Orthopaedic
appliances; (iii) HS Code 9022 – Equipment using X-rays, alpha, beta, gamma rays. The
exports of Pakistan predominantly fall in the category 9018. Pakistani exports make up
only a small fraction of world trade in surgical and medical device industry, which
amounts to over $113 billion (just for above 4 HS Codes). This is one sector where
Pakistan has developed special capabilities to penetrate high income markets such as
Germany, USA, France, Belgium etc. The average export price of goods made in Sialkot
is around $1.5-2.5 (Note: some products sell for much higher prices – the price quoted is
the average trade price for disposable products), which is much higher than what
68 Authors’ Field Interviews
231

Chinese products fetch (US$0.35 – in disposable products). However, the price is lower
than some of the more sophisticated producers such as Germany and France.
Table 3‐6: Surgical Industry Characteristics
Characteristic Value
Number of units Over 2300
Total installed capacity 225 – 250 million pieces
Current production 150 million pieces
Contribution to national exports 1.21%
Contribution to GDP 0.41%
Sector Employment 100,000 – 150,000
Capital labor ratio 10 workers / million (Rs.)
Total estimated investment Rs 12 billion
Source: BOI reports, CMI data and PEFMA interviews
The sector comprises over 2300 companies, of which around 30 can be considered large
and the remainder can be split as 150 units of medium sized and remaining as small.
Besides small and medium units, a few units are large and have a 90% integrated
system. Most of the larger and medium sized firms are exporting, however, the
smaller/vendor units usually supply to commercial exporters/traders. The main raw
material used in the production is ‘steel’. Around 60% of this steel is manufactured
locally and the remaining 40% is mostly imported from Germany.
The categories of firms can be divided in to three tiers. The tier 1 firms are the ones that
are technically advanced and have exports over Rs.100 Million. These are technically
advanced firms with lot of mechanization. These are also the firms which have shifted
some of their production out of the disposable instruments and moved into more
sophisticated products such various medical devices. Tier 2 and Tier 3 firms are
predominantly involved in making basic surgical components.
Figure 3-16 below compare the product split of world trade in surgical instruments (this
is HS Code 9018) with that of Pakistan. It is quite clear the device and apparatus
component of the trade is growing over time relative to basic disposable instruments. In
addition, the disposable instruments are fetching lower prices over time, whereas the
medical device industry has been climbing up the value curve. Pakistan has only grown
232

slightly in basic instruments and the industry’s capacity to supply more sophisticated
equipments is fairly low.
Figure 3‐16: Product Split of World Exports & Pakistan Exports, 2005‐09 (US$ Million)
World ($ Bn) Pakistan ($ Mn)
60
50
40
30
20
10
0
Source: UN Commodity Trade Statistics
Figure 3‐17: Exports of Pakistan’s Surgical Instruments Sector, 2005‐09 (US$)
300.0
250.0
200.0
150.0
100.0
50.0
2005 2006 2007 2008 2009
0.0
2005 2006 2007 2008 2009
Source: UN Commodity Trade Statistics
High Value
Apparatus,
Syringes &
Needles
Dental
Instruments
Basic
Surgical
Instruments
3.3.5.1 Value Chain Analysis
The overall structure of the value chain suggests that industry typically adds 80% of exfactory
sale price in value addition in the tune of around Rs 160 per forceps. The
retailers in the foreign markets are adding the maximum value which is around 3 times
120.00
100.00
233
80.00
60.00
40.00
20.00
0.00
2005 2006 2007 2008 2009
High Value
Apparatus,
Syringes &
Needles
Dental
Instruments
Basic Surgical
Instruments

more than the producer. The mean reason for this is branding and direct contact with
the end user and market.
Figure 3‐18: Value Chain Breakdown of a 5.5 Inch Forceps
In these three processes high degree of
outsourcing to vendor industry exists. Large firms
outsource between 15-25%, while medium to
small firms outsource up to 70% of this work to
vendors.
3.3.5.1.1 Availability of Raw Material
The value chain analysis above depicts that over 35% of the costs represents the material
and other inputs required to manufacture a forceps. A major portion of this cost is the
steel used. Two different types of steel are currently being used. Locally manufactured
which costs Rs. 110/Kg and imported steel which costs Rs. 250/Kg. The above analysis
refers to imported steel forceps. Although the local steel is low in cost it mostly does not
meet the quality standards that are required by foreign buyers. The finish of the product
is poor if local steel is used and rusting is a common issue due to non-uniform
tampering. Some industry has raised concern about the quality of the local steel as not
conforming to ‘health’ grade steel.
234
Material = Rs 42
Forging & Shaping = Rs 5.5
Milling & Machining = Rs 7
Rough grinding & tampering
= Rs 18
Final grinding, fitting &
polishing = Rs 22
Overheads & Packing = Rs 24

Moreover, the availability of steel is also an issue. The stocks and prices are quite
variable resulting in pricing problems for the industry. Large firms normally import
large quantities whereas the small ones have to rely on supply from the vendor industry.
Especially in the case of reusable instruments which are made entirely from imported
stainless steel, the firms have to import steel from Japan, France and Germany. In most
of the cases, the lead time is two to three months, which results in the firms not being
able to fulfil their orders and eventually the customers turning away from them. Due to
this huge lead time, the price of raw material varies substantially and the Sialkot firms
are not able to quote any specific prices for a long period, due to which the firms are not
able to successfully do their costing.
3.3.5.1.2 Difficulty in Getting Adequately Trained Manpower
There is currently no formal training and research centre in Sialkot which caters to the
training needs of the industry. As the statistics show, the industry is currently involved
in the manufacture of conventional surgical instruments which are made out of locally
made steel; most of the processes are out sourced, and are performed by vendors sitting
in the small sweat-shops located in and around the city of Sialkot. These shops mostly
employ informally trained workforce or new job seekers who work as apprentices. The
industry is not ready to move into high-tech, higher value surgical goods because of the
absence of technical and research centres in Sialkot.
3.3.5.1.3 Limited Diversification of Products
One of the major weaknesses of the Sialkot industry is the limited diversification of
products. Almost all of the manufacturer’s are competing in the same product line,
which makes the supply excessively greater than the demand and reduces the profit
margins significantly.
3.3.5.1.4 Inconsistent Quality Producer Image
Currently there are no set of rules and regulations governing the standards of the
industry. There is no watch dog that can monitor the negative activities in the industry.
Moreover, the overall image of the Sialkot surgical instruments industry is that of a low
cost producer. The cluster produces a variety of standard instruments, relying heavily on
labor intensive techniques.
235

3.3.5.2 Policy Recommendations:
The surgical industry of Sialkot suggests that sector-specialized clusters of small and
medium sized producers can be well positioned to respond to rapidly changing,
demand-driven, global markets or the new competitive landscape. Since the
international success of Sialkot’s surgical instrument producers owes much to gains
obtained from clustering, we recommend the following:
• Establish a Surgical training school in Sialkot. This will train workers not only in
the basic skills but will create an educated workforce which understands the
nature of the product. This is imperative given the fact that as quality standards
and compliance regimes get tougher worldwide the workforce will have to
understand the requirements of these standards.
• Upgrade / Establish laboratory in Sialkot to provide surgical instrument specific
testing and conformity assessment certifications. Under new WHO regulations
the industry will soon have to make a change in their polishing and cleaning
techniques. The industry will have to shift from chemical to water based cleaning
and polishing systems. This shift will increase costs and also change the testing
and certification requirements. A laboratory is required that can test for
sterilization, chemical presence, etc. The government working with the sector will
either upgrade the Surgical Association Laboratory or will extend more support
to the laboratory being established by the Punjab Government.
• Facilitate via foreign offices for local companies to acquire international brands.
• Support development of steel industry which is the primary input into surgical
industry.
• Establish an excellence centre for instrument design and innovation in medical
device industry.
• The government itself is one of the largest procurers of surgical equipment for
public hospital. The procurement rules should be made more transparent so it
becomes fairer to local manufacturers and also the payment methods should be
such that rent seeking and any unnecessary delay is minimized.
• Launch a sector competitiveness project executed by NPO.
236

3.3.6 Sports Goods Sector
In most segments of the global sports goods market, industrial producers face a supply
side structure which is typically a ‘fringed’ global oligopoly; global because a handful of
big Transnational Corporations (including Nike, Adidas, Reebok etc.) covers a major
share of the world market for each sports good; and ‘fringed’ because, in each developed
country, a number of competing small and medium-sized enterprises take over the
remaining ‘fringe’ or margin of the local domestic market left by the big firms. The
global sports goods industry is expected to grow by $8-9 billion in the next 3 years and
is expected to reach total sales volume of about $78.6 billion by 2013. The total global
market of sports goods is segmented in different categories. The main categories include
Golf (22.40%), Fitness/Athletics (12.60%), Fishing (9.50%), Racquets (4.30%), and
others (51.10%).
Currently, the total exports from Pakistan amount to $600 million a year which are less
than 1% of the total world market share. Out of the total world market, Asia comprises
$22 billion annual market of sports goods. However, exports from Pakistan to Asia are
minimal and hardly register on the radar screen.
Table 3‐7: Sports Goods Industry Characteristics
Characteristic Value
Number of units Over 10,360
Contribution to national exports 3%
Contribution to GDP 0.51%
Sector Employment 300,000 – 350,000
Skilled to unskilled ratio 2:30
Capital labor ratio 12 workers / million (Rs.)
Total estimated investment Rs 34 billion
Source: BOI reports, CMI data and PEFMA interviews
The sports goods being manufactured for export from Sialkot include tennis balls,
rackets, hockey sticks, polo sticks, cricket bats & balls, footballs, badminton rackets and
other associated products. Pakistan exports more than half of its total exports in
sporting goods to the US, the UK and Germany. Figure 3-19 below shows that the
exports of sporting goods from Pakistan have stagnated and declined to a low of
237

US$0.26 billion in 2009 from a high of US$0.34 billion in 2006 (www.trademap.org).
The firms that were surveyed reported that not only they were unable to attract new
buyers but were also finding it exceedingly difficult to maintain their existing clientele.
Figure 3‐19: Pakistani Exports of Sports Goods, 2003‐2009 (US$ Billion)
Source: TRADEMAP
Figure 3-20 below compare the product diversity in Pakistan’s sport’s goods industry
with World market. The product mix of Pakistan is not in line with the trend of word
market demand. This pattern may explain why the exports from the sector has been
declining in the recent years.
Figure 3‐20: Comparison of World and Pakistan’s Sporting Goods Product Split
238

Source: TRADEMAP 2008 data
3.3.6.1 Value Chain Analysis
For illustrative purposes we have considered a representative value chain of an export
quality hand stitched soccer ball. As shown in Figure 3-21 the overall value added is
between 15-20%. We were told that this number was much higher a few years back,
however, the increased cost of local inputs and stiff competition from China has caused
a hard squeeze on this number. Below we have presented issues identified along the
value chain.
239

Figure 3‐21: Typical Value of Hand‐Stitched Export Quality Training Soccer Ball
3.3.6.2 Issues
3.3.6.2.1 Diversification of Market
Pakistan’s export of sporting goods to the US, the UK and Germany covers more than
half of Pakistan’s total export of sporting goods. Diversification of geographical export
markets is badly needed. France for example is Europe’s largest market for sporting
goods.
3.3.6.2.2 Variability in Demand
The production of footballs follows a roller coaster ride due to several reasons. The
prime reason for this fluctuation in output is due to the hike in demand for soccer balls
for the FIFA world cup, which is held after every 4 years. Similarly, the manufacturers
face a sudden surge in demand six to nine months before UEFA Euro cup. Immediately
after these major tournaments the manufacturers face a sudden dip in demand. This
variability in demand has a severe impact on the manufacturing systems of sports goods
firms. The impact of variation propagates from one manufacturing stage to the next with
increasingly adverse effects on production planning and control activities.
3.3.6.2.3 Fierce Competition in the International Market
Pakistan is now facing serious challenges from India, China, Taiwan, and South Korea in
international sports goods markets. India and China have an advantage of cheap labor
and raw material while countries like Taiwan, Japan and South Korea are experts in
mechanization and use of modern equipment. So Pakistan once the uncrowned king of
240
Raw Material = Rs 202
Assembly process & overheads = Rs
172
Marketing = Rs 5
Factory margin = Rs 30
Importers margin = Rs 481

sports industry is now facing serious competition from Asian competitors. A few years
back Pakistan was the leader in the exports of sports gloves. Now China has surpassed
Pakistan and the gap is growing with each passing year. Chinese sports industry is
growing at a rate of 8% every year. It holds 50% share in the sports market of cricket,
hockey and baseball, while Pakistan and India merely have a share of 8% and 6%.
Similarly our industry is lagging in composite (graphite) based sports goods. Sialkot
used to be the hub of wooden (mulberry) hockey sticks but with the shifting of world
demand towards composites; the local industry failed to cope up with the pace of the
world which resulted in the reduction of our share in the international market.
3.3.6.2.4 Issue of Child Labor
The soccer ball industry of Pakistan, principally concentrated in the city of Sialkot, has
been under scrutiny in recent years for employing child workers. Many reports describe
children stitching soccer balls in small rural workshops or in homes. The widespread
coverage of children stitching soccer balls galvanized consumers, labor and religious
groups and government officials to call for a stop to this practice. In June 1996, the U. S.
Secretary of Labor and members of Congress helped launch the Foul Ball campaign to
ensure that soccer balls used by American children are stitched by adults and not
children.
The public attention led U. S. importers to enter into an agreement with Pakistani
manufacturers, the ILO and UNICEF to stop using child labor. More recently, public
attention has focused on the use of child labor in India's soccer ball industry, leading to
negotiations between Indian soccer ball manufacturers, exporters, NGOs and
international organizations to develop a code of conduct prohibiting child labor. This
code of conduct has in the meantime been implemented by several key Sialkot
producers who became world class and are now major suppliers to the world’s biggest
brands: Nike, Adidas, Puma. In addition, smaller, individual companies have taken
steps to assure consumers that their products are not made by children. They have done
so by revising their production process or creating child labor labels. Some U. S.
importers, including Reebok and Baden Sports, have begun to label their soccer balls as
produced without child labor.
241

3.3.6.3 Policy Recommendations:
Since a collective capacity to compete, adapt and innovate has emerged in the sports
goods industry, therefore, we recommend the following:
• Continue to support the ongoing competitiveness initiative being managed by
SMEDA and agree to fund the initiatives that will be identified as a result of this
exercise.
• Establish a sports good sector company with broad base representation of the
private sector. This company will implement the sector competitiveness strategy
developed with support from SMEDA.
• Operationalize the Sports Industries Development Centre being established in
Sialkot.
• Support development of a material bank in partnership with the private sector to
facilitate development of newer products.
• Facilitate establishment of strong linkages of the sector with NED Karachi, Gujrat
University and Loughborough University to work on product and design
development.
• Advocate with the provincial government to complete its Sports Goods testing
laboratory project on a priority basis.
• Provide targeted tax benefits for companies diversifying into more competitive
products such as golf equipment. The credits will be conditional development and
exports of these products.
3.3.7 Ceramics Sector
Ceramics industry can be divided into ceramic tiles, Tableware, Refractory, Sanitary
ware and Electric Insulators. The ceramics industry in Pakistan consists of both
organized and unorganized segments. The production of insulators and tiles is entirely
concentrated in the organized sector, and are fully documented. However, Sanitary
ware, Tableware and refractory are primarily in the unorganized sector. The industry
has made significant progress over the last few years. The industry has been able to
penetrate export markets and also supply to an increasing local demand. Table 3-8
shows the ceramic industry structure in Pakistan. Industry is mainly located in
242

Gujranwala, Gujrat, Lahore, Karachi, and Peshawar. Figure 3-22 shows recent exports
of ceramic products from Pakistan which are consistently on decline. More than 60% of
the exports comprised of bathroom, kitchen and other sanitary fixtures (HS Code 6910)
and bulk of it goes to UAE, Saudi Arabia and Afghanistan. US, UK, Germany and France
are the leading importer in this category but Pakistan’s exports to these countries are
negligible. This shows that the current exports are not even a fraction of the total
potential offered by the sector. Ceramics and sanitary ware is a growing industry and as
the incomes, tastes and preferences of the world consumers change it offers diverse
opportunities. As the sector has a significant demand it can be classified as an important
growth sector for Pakistan, given, Pakistan’s share in world market is still very weak.
Table 3‐8: Pakistan Ceramics Industry
Tiles Insulators Sanitary
ware
Status Highly Highly Organized+
Organized Organized Un-Org
243
Tableware Refractories
Organized+
Un-Org
Un-
Organized
No. of Units 7 1 63 150 20
Total
Employment
2,435 465
6,000 7,500 207
Technology Capital Capital Labor Intensive Labor Labor
Level
Intensive Intensive
Intensive Intensive
Automation
Level
Highly
Automated
Source: SMEDA Report, 2007: 4
Auto, Semi
&
Manual
Figure 3‐22: Ceramics Exports for Pakistan, 2005‐2009 (US$)
Low
Automation
Automated Low
Automation

Source: UN Commodity Trade Statistics
The critical issue faced by the industry at the moment is the unannounced electricity
load shedding which severely impacts productivity and costs. The ceramics industry
runs on a continuous production process where most of the cost is incurred at the
starting phase of the process when the kiln is being heated. Once the kiln has reached its
optimal temperature the variable cost falls significantly. In most countries the costs are
controlled by not switching of kilns over several decades. However, given the uncertain
nature of electricity supply this is not possible for manufacturers in Pakistan. The costs
of production as a result of this frequent switching on and off of the kilns (sometime 3 to
4 times a day) increase the electricity usage and cost to such an extent that Pakistan’s
industry is unable to compete internationally.
Another issue faced by the industry is lack of design and product development capacity.
The basic technology of making ceramics and bath room fitting and accessories is
exactly the same around the world. What differs is the design and product innovation
capacity. The colour scheming and designing and size of tiles can add more than 150% to
the total value added. Similarly, consumers would pay exorbitant prices for high fashion
bathroom fittings and accessories. It is extremely simple to make intelligent
bathroom/wash room accessories by adding low cost electronic gadgets and sensors
made in China. Several Indian companies have recently launched a whole range of
244

electronically controlled bath room fittings and accessories. Simply adding these low
cost electronic parts significantly increase the value of the products. However, to do this
the sector will require strong linkage with the electronics industry and also with
academia working in the field of electronics and product design. For this purpose the
government did facilitate the establishment of a product development centre for
ceramics industry, however, the project was converted into in a marble factory by
TUSDEC. The reason provided was to make the facility commercially viable service
centre and to generate revenue for TUSDEC. This action may help in justifying a high
expense government created company, however, fails to achieve the purpose of assisting
an industry where significant value addition potential exists. This lack of vision, element
of corruption and poor governance has on several occasions illuminated the
effectiveness of extremely important initiatives.
The sector also lacks availability of skilled workers and formal training facilities. There
are no formal schools that offer training to workers of the industry at all levels. The
industry is unable to find basic skills, managerial skills, technical skills and design skills.
The sector has no linkage with design schools such as National College of Arts (NCA),
Pakistan School of Fashion Design and the Indus Valley. The students of these colleges
must by default attend 6-8 months on factory internship programme in the sector. This
will not only help the industry improve its design capacity but also contribute in training
our youth closer to the ground realities of the country and to come up with original
creative work in line with the needs of the local industry.
Finally, as with most of the other SME sector, this sector is also constrained by branding
and marketing. The small size of the enterprises makes it impossible for businesses to
spend money on branding or marketing their product. The packaging quality is poor and
often the products are damaged during transport. There are no display centres at places
where foreign buyers/tourists have easy access.
3.3.7.1 Policy Recommendations:
The objective of the interventions directed at the ceramics industry is to improve its
competence and technology among the ceramics firms in order to help them become
more innovative.
245

• Ensure that the Ceramics Industry be provided continuous electricity for 10
months during the year in lieu of complete shut down for two months
• Ensure that the ceramics centre established in Gujranwala focuses on product
design and development and offers training courses specifically for the ceramics
sector.
• Engage SMEDA to assist in creation of networks in the cluster and facilitate these
networks to acquire international brands that are currently up for sale due to
rising costs in the west and continued economic down turn. This outward FDI
should be facilitated by the government to assist the sector develop its
international branding.
• Facilitate HEC, TEVTA, NAVTEC and the sector association to work with NCA,
PIFD and Indus Valley to assist them in revising their courses to allow for 8
month to a yearlong internship on the factory floor to understand the designing
requirements and limitations. The factories should also contribute partially to
run this programme on a sustainable level; their contribution can be in the form
of providing material, space and machinery to student to work on.
• Support Identification of international buyer clusters of ceramics and facilitate
international market access through TDAP.
3.3.8 Furniture Sector
Furniture sector of Pakistan has historically grown on the basis of large local demand.
Traditionally, it has been a resource and labor intensive industry which included both
local craft-based firms and large volume producers. It was only recently that Pakistan
entered into the furniture export market. However, this entry has occurred on
competitive terms as countries such as China, Indonesia, Vietnam, Malaysia, and Brazil
have also over time expanded their furniture industry replacing market shares of USA
and EU firms. Top exporting countries in the world of furniture trade include China,
Italy, Germany, Canada, and USA, respectively 69 . Major importing countries are USA,
U.K, Germany, France, and Canada.
69 UN Commtrade statistics 2005-2009
246

Pakistan’s furniture industry has a long history of craftsmanship supported with strong
forest base. Pakistan has 2.3 million of hectares of wood forest, supplying Sheesham
wood to around 82% of entire furniture industry of Pakistan. Sheesham is well known
for its durability and quality and normally fetches a higher price both in the local and
international markets. The main wood reserves are the State owned forests in Khyber
Pakhtunkhwa, Northern areas and Azad Jammu Kashmir (AJK). The local furniture
industry also has to rely on import, especially furniture accessories, wood tops, plywood,
roundwood, veneer and packaging materials).
The sector predominantly comprises MSMEs 70 with about 6,000 firms or productive
units in Pakistan employing 80,000 workers71 . The cities of Karachi, Lahore, Peshawar,
Gujrat and Chiniot are considered as the main clusters of furniture manufacturing in
Pakistan with majority of manufacturing units concentrated in Chiniot and Karachi.
Industries in Karachi and Lahore are predominantly focusing on adding value to
furniture produced in other clusters.
The industry’s performance has been impacted by low managerial and technological
capabilities, inconsistent quality and insufficient production capacity to meet growing
domestic and international demand and product diversity. Most of its current
production volume is geared towards the domestic market, competing directly with lowpriced
imports which have been aggressively capturing Pakistan’s domestic market
share. Especially in low cost office furniture Chinese and Southeast Asian firms have
replaced local manufacturers significantly.
On the positive side the furniture industry of Pakistan in endowed with several favorable
environment and conditions with regards to craftsmanship, geographical location, low
labor cost and a young entrepreneurial base. However, unprecedented increases in
input prices, stiff competition from Asian exporters like China, India, Vietnam and
Malaysia are serious challenges faced by the industry. High utility costs, improper wood
seasoning, substandard finishing and packing, poor designs & quality, lack of R & D, low
70 Micro, Small, and Medium Enterprises
71
‘The furniture sector in Pakistan’, European Union TRTA program, 2007
247

level of quality control, inefficient production capacity & technological
practices/methods, access to international markets, and not meeting world standards
and requirements are the problems plaguing the Pakistani furniture industry
The performance of the sector in terms of export has been much below expected
potential. Over the last five years the exports have averaged at just a minimal US$10
million as compared to the total world market of $50 billion. The growth in exports over
this five year period has also been largely negative (see Figure 3-23). The major export
markets of Pakistan include; UAE (16%), USA 912%0, UK (11%) and Afghanistan
(10%) 72 .
Figure 3‐23: Furniture Export of Pakistan, 2005‐2009 (US$)
Source: UN Commodity Trade Statistics
On the other hand the domestic furniture market has expanded at much a rapid pace,
growing at around 25% per annum a year reaching over $160 million in value [European
Union TRTA report, (2007: 25)]. It is expected that the local demand will reach $1
billion by 2015 73 . This growth potential offers significant opportunity to develop the
industry on home grown demand and build capabilities that will ensure heightened
competitiveness in international markets.
72 UN Commtrade statistics 2005-2009
73 IFCA estimates (Dec. 2006), SMEDA Furniture Strategy (2007)
248

Figure 3‐24: Pakistan Export Price Comparison, 2005‐2009 (US$/Kg of Wood)
Source: UN Commodity Trade Statistics
3.3.8.1 Value Chain Issues
Figure 3-25 provides a snapshot of the furniture value chain and highlights the challenges which
are currently impeding the performance of the industry.
3.3.8.2 Shortage of Sheesham and Other Wood Resources
Although Pakistan possess some of the best quality wood varieties (e.g. Sheesham, Deodar, and
Walnut), this 2.3 Million hectares of wood is not managed well. There are claims of widespread
exports of unprocessed logs and sawn timber to the Gulf countries. The waste happens from
sawing to poor storing practice of final products, which can reach 50% on average based on the
value chain analysis provided below. Another issue is quick depletion of natural resources.
Estimates indicate that around 1.23 to 2.4 million cubic feet of Sheesham wood is cut annually
[Furniture Strategy paper, SMEDA (2007 :p.9)]. At such an accelerated pace of deforestation,
replenishing of resources will not be possible.
249

Figure 3‐25: Key Challenges in Pakistan’s Furniture Value Chain
Weak managerial capacity at firm and cluster level
Consumer
Consumer
Channel
Channel
Distributor
(National &
International)
Transport
Manufacturing
Inputs
Timber/board
Wood Supply,
Roundwood
PRoduction
Domestic
Bedroom Kitchen Other furniture
Non-Specialized Distribution
Domestic
manufacturers
Flooring, Joinery, Furniture and parts - firms
Intermediate & final
Products
Sawmills, veneer, plywood
(4.2 million sheets (8x4) particleboard / year)
Sheesham (82%)
Log bidding and auction
Reconstituted panel
plants
(Kiln Drying is Missing)
Other, i.e. deodar, poplar,
mulberry (18%)
250
Selling agents
Buying agents
International Market
Hypermarket Mail order Dept Franchise Chain Retail
DIY Stores housee Stores Stores Stores Stores
Bedroom Kitchen Other furniture
Specialized Distribution Non-Specialized Distribution
Local and imports: Woodworking items, seating/upholstery items, glass, marble,
metal, plastic (structural design), finishers, measuring, jig fixture
Raw material instability
Source: Furniture Strategy paper, SMEDA 2007
Imported sawn wood (USA,
Europe, Malaysia)
Large retailers Buying Groups
Poor market knowledge
key inputs are imported
Importers/
Wholesalers
Domestic
International market
Domestic & international
Key sector challenges
Inadequate handling of wood / lack of skilled labor
Sources: Peter Rayner (2006), ITC report (2007), SWOG analysis (2007)
3.3.8.2.1 Value‐Chain Productivity
In addition to the raw material replenishment gaps wood is also wasted due to lack of
proper handling and processing techniques. The quality of the wood is compromised

due to improper wood drying techniques. According to analysis above wood waste can
reach up to 50% across the value-chain.
Inconsistent quality and waste problems are result of widespread use of informal wood
drying practices (e.g. sun drying). Currently, less than 10% of Pakistan’s wood
production is kiln-dried, resulting not only in waste, but quality problems that are basic
requirements for export markets. Kiln drying eliminates moisture (to about 8-10%),
bugs, and other impurities. In order to address this problem the Government of
Pakistan has initiated pilot projects to provide appropriate wood drying kilns Chinyot.
Moreover, lack of proper cutting techniques, handling of wood and scarcity of inputs or
specialized suppliers (e.g. glue and dyes) affects the final product in both cost and
quality. Basic professional woodcraft techniques, such as cabinet making skills rather
than using nails are widely lacking. Moreover, adequate quality and finishing
procedures are lacking, and modern packaging is virtually non-existent in most
factories. Such techniques and resources are essential to modernize Pakistan’s firms
across the value-chain to make it competitive.
3.3.8.2.2 Volume and Price Strategy
The sector has historically employed a volume rather than value based production
strategy. The industry has always focused on way and means to increases the number of
unit sold rather squeeze more value out of existing production. This strategy will have to
be revised as the quantity of local wood available is declining. The industry is
increasingly becoming dependent on imported woods and accessories. This dependence
is a big factor contributing to cost hikes and rendering industry uncompetitive in
international markets. The industry will have to change its focus to a value based
strategy and compete on the basis of quality, design and product variety using more of
made up materials.
3.3.8.3 Changing Consumer Preferences
251

Another issue faced by the furniture industry in Pakistan is its inability to innovate.
Pakistan’s furniture historically was valued both in local and international markets due
to the use of extremely good quality of wood. However, the market demands and trends
have changed and consumers both in the local and international markets prefer
furniture that is cotemporary, easy to move and economical to replace as fashions and
designs change. The demand of furniture which used to be tied down to the quality of
wood and endurance has now shifted to economical, trendy and light looking well
designed furniture. IKEA which is one of the biggest furniture retailers in the world
work on a simple model based on reasonable cost, high fashion decor and easy to
replace furniture. The Pakistan industry will have to shift gears to stay competitive and
flexible enough to rapidly changing demands of consumers.
Box 3‐6: Past Government Intervention in the Furniture Sector
252

In order to address the issues being faced by Pakistan’s Furniture sector, United States Agency for
International Development (USAID) as part of Strategic Development and Competitiveness Project
formed a strategic working group (SWOG) of committed industry stakeholders in 2006. SWOG has
become a recognized private sector led platform on which all the key players in the value chain, including
the public sector have come together. SWOG has developed a vision shared and agreed by the private
sector and the Government of Pakistan:
In the short-term, it is focused on ensuring 100% usage of kiln dried solid wood for manufacturing. This is
a critical basic processing requirement both to gain domestic market share from imports, as well as an
international standard to export. Another complementary goal is to ensure credible supply of sustainable
timber immediately and throughout the next 10 years. In the long-term, the industry’s vision is to become
a globally competitive and sustainable furniture industry with improved productivity throughout the
value-chain and increased export levels.
To accomplish the above directions, while addressing the key sector challenges, SWOG identified
following initiatives).
Pakistan’s Furniture Sector Challenges and Proposed Initiatives
Challenge Proposed Initiative Benefits
Shortage of
Sheesham and other
Wood
Value Chain
Productivity
Weak coordination
mechanisms
Sustainable forestry system
Wood and component bank
Reconsider use of sheesham
in public buildings
Chain of custody
reforestration
Establish 250solar kilns
across Pakistan
Common Facility Training
and Manufacturing centres
Furniture Testing Labs
Marketing (exhibitions, show
rooms) and industry/market
information capturing
Establish a sector
development company
(Furniture Pakistan)
253
Replenishment of sheesham and other
wood resources
Stable prices and reliability of inputs
Greater availability of supply for value
added products
Improved supply by curbing illegal trade
Ensure availability of quality raw material
World class technical capabilities,
including design, CAD and manufacturing
techniques
Furniture accreditation for greater export
competitiveness
Informed industry and market choices for
timely strategic actions
Critical capacity requirement to
coordinate actions proposed by SWOG
In order to coordinate the activities and implement devised strategies a sector development company
Furniture Pakistan has been established in 2007. It is focused to be a lead organization in developing the
Pakistan furniture industry into a world leader in manufacturing and trading of furniture products. It is a
focal platform for sustained public-private sector dialogue and interaction. Till 2010, Furniture Pakistan
has implemented some initiatives and with these targeted interventions, Pakistani firms have regained
some share of the growing domestic market by curbing the current growth rate of imports. However,
exports have not picked up and are consistently declining.
Source: SMEDA Strategy on Furniture

3.3.8.4 Policy Recommendations:
It will be difficult to compete against the Asian furniture giants from the small workshop
culture that Pakistani industry is pursuing. In order to strengthen the industry, we
recommend the following:
• Keep supporting the ‘Furniture Pakistan’ initiative and provide funding for
implementation of strategic actions developed by the furniture sector.
• On priority basis ensure provision of solar kilns for wood drying in Karachi,
Chiniot, Peshawar, Lahore and Gujrat.
• Ensure early completion of common facility centres for product development and
training facilities in Karachi, Chiniot and Peshawar.
• Allow duty free import of low cost woods and alternative material for furniture
manufacturing.
• Develop capacity within universities and research centres to work on
development of alternative to wood materials that can be used for furniture
making. This type of material is required for trendy yet economical furniture.
IKEA is an example which is a leading furniture retailer in the world. Pakistan
will have to build capacity to move toward low cost materials but high value
added design and contemporary furniture.
• Support establishment and implementation of domestic quality standards for
furniture.
3.3.9 Leather Sector
The size of the leather industry, on a global level, stands at $100 billion. A significant
portion of this number is shared by the developing countries (around 50% belongs to
China, Italy and India), while developed markets of US & Europe are major consumers
of leather products. The global industry is largely ‘buyer-driven’ with producing
countries manufacturing in line with specifications, designs, fashion guidelines, quality
and social standards and technical advice provided by the buyer countries. China and
Italy are the leading producers and exporters of leather products / leather with exports
worth US$25 billion and US$19 billion respectively. After China and Italy, there is a
fairly large drop in the export market shares, with Brazil and India taking the 3rd and
254

the 4th positions. Pakistan’s trade size of US$1.2 billion is close to the formal trade data
for Turkey, Vietnam and Indonesia and Argentina 74 .
The Pakistan Leather Industry has a significant place in the world leather industry built
up primarily on the basis of a large raw material supply, which supports a strong
tanning industry, a considerable garment business and growing footwear, bag and glove
components. However the sector has failed to achieve its potential as t the latest figures
show exports are under US$ 1 billion mark. The Pakistan leather industry has been
unable to keep up with the changing demands of this extensively ‘buyer-led’ industry.
This has caused the overall industry to fall short competing with the competitors who
have moved ahead of them in terms of quality, productivity, completeness of the value
chain, as well as a favorable domestic enabling environment.
Table 3‐9: Leather Industry Characteristics
Characteristic Value
Number of units 675 tanneries
Current Production 30 million m2
Contribution to National Exports 6.5%
Contribution to GDP 5%
Sector Employment 0.5 million
Source: BOI Reports, CMI Data and PEFMA interviews
Pakistan's leather industry represents 6.5% of Pakistan’s total export base. During 2007-
08, the production of leather was about 30 million m275 and export of leather and
leather products amounted to US$ 1.25 billion. This number has however fallen below
the US$ 1 billion mark since then. About 90% of the leather and value addition goods
and products are exported in finished form (Pakistan Tannery Association). The
industry is predominantly clustered in and around the cities of Karachi, Sialkot, Lahore,
Kasur, Charsadda and Bannu with a much smaller cluster in Multan. There are about
74 Turkey’s informal trade data brings it to over US$3bn.
75 Leather Sector Strategy, SMEDA 2009
255

675 tanneries in the formal sector, 76 which are concentrated in a few clusters, of which,
Kasur (275), Sialkot (250) and Karachi (150) are the most important.
The history of the modern leather industry in Pakistan dates back to the beginning of
the 20th century, when hand-made leather was produced in Kasur by using flour and
vegetables 77 . Raw hides and skins were exported to many countries around the world.
The trend from cottage vegetable tanning to wet blue was initiated in the early ‘60s and
the industry started production of finished leather in the early ‘70s. Production of
leather products, such as garments and footwear, started in the mid ‘70s. The product
profile of the leather sector export is dominated by the garments sub-sector with the
largest share of 43%, the finished raw leather sub-sector (finished and un-finished) at
34%, the gloves sub-sector at 13%, footwear at 9% and other leather products including
saddles, belts, bags and other clothing accessories at 1%.
Figure 3‐26: Comparison of Pakistan and World Product Split
Source: TDAP, PLGMEA, PGMEA
The figures above re-iterate the anomaly that the Pakistan leather sector is divergent
from the global market trend as Pakistan’s two smallest sub-sectors are the world’s
largest, namely footwear and other goods. The comparison above again suggests that
there is considerable room for Pakistan’s leather sector to grow in footwear and other
76 Leather Sector Strategy, SMEDA 2009
77 The Sialkot leather cluster believes leather making in what is today Pakistan may date back 2,000 or
more years.
256

products. Developing capacities in these two products can assist the sector climb a rapid
growth trajectory.
More generally, Pakistan’s leather sector consists of the following key subsectors:
• Tanning and finishing
• Footwear and some of its components
• Garments
• Gloves
• Saddlery
• Bags
The industry structure is predominantly small to medium size proprietorships / family
run businesses (nearly 80% 78 ). The production centres are predominantly located in
Karachi, Sialkot, Lahore and its surroundings, Kasur, and to a much lesser extent in
Charsadda and Bannu.
3.3.9.1 Issue in Value Chain
3.3.9.1.1 Losses in the Tanning Process
• Analysis of the value chain highlights that in the tanning process alone there are
losses that can be brought down to add more value. In addition there are
environmental problems that must be tackled for the industry to remain
compliant with:
o 25% value loss due to animal skin diseases
o 15% value loss due to poor slaughtering techniques
o 2% value loss due to poor transpiration facilities
o 10% value loss due to poor collection processes and techniques
• Similarly, the value chain analysis for leather products suggests that there are
serious gaps in the quality of raw material and the capacity of the industry to
come up with value added designs and product innovation.
3.3.9.1.2 Lack of supporting Institutions
The review of the sector has highlighted that there is:
78 Industry Sources
257

• Limited availability of modern up to date abattoirs and collection centres
• Limited quality testing and certification facilities and laboratories
• No proper design and product development centres
• Limited and independent research and development capacities
• Limited or no interaction with academia
• Outdated capacity of training and vocational institutes
3.3.9.1.3 Inconsistence policies
A clearly articulated, consistent and predictable policy that enables and facilitates
private sector growth is essential for any area of business to develop. The leather sector
has raised its concern to have the debate on policy with the government as an ongoing
one that must reflect broad sector interests, and that the policies must be consistent and
predictable. The policies are counterproductive as they distort allocation of resources
and reflect that government is not a credible partner to the private sector. Policy
development will only be effective if it is evidence based and is developed in consensus
with the sector and general agreement between stakeholders on policy will ensure
voluntary enforcement.
During the stakeholder meetings it emerged that the industry has been impacted more
by sudden policy changes or lack of implementation of policies rather than those policies
that are viewed as unfavourable. Some examples of the inconsistent policy actions of the
government highlighted by the sector include:
• Discontinuance of R&D grant without notification or consultation with industry
• The export rebate increased to 2.8% from 1% but the policy is still not
implemented
• SNGPL disconnects gas without notification and no firm date is given for
reconnection
• EOBI system needs rationalizing to make it more SME friendly
258

3.3.9.2 Policy Recommendations:
The challenge of the future is to find ways to improve the quality of the product from
semi-processed to fully processed leather goods and to be able to compete
internationally. Our recommendations for doing this are the following:
• Establish the Pakistan Leather Development Council (PLDC). This should be a
section 42 company and should act as the representative body of the entire
sector.
• Ensure appropriate funding is provided for the implementation of the PC-1
developed for implementation of strategic initiatives under PLDC.
• Work with University of Veterinary and Agricultural Sciences to improve quality
and quantity of livestock.
• Incentivise establishment of modern abattoirs and allow duty free import of
flaying machines.
• Reduce duty on Bulk Chemicals to 5%.
• Allow free import at 0–5% for 3 years to acquire tanning plants and machinery
from international locations where tanning industry is closing due to higher costs
or environmental compliance.
• Provide matching grant for establishing Effluent treatment plants.
• Take immediate steps to stop smuggling of live animals.
• Establish design centres and glove development institute.
• Assist in creating linkages with international design centres and also incentivize
development of local centres of design excellence.
• Develop a footwear development institute and link it with design centre at
Pakistan Institute of Fashion Designs.
• Promote Charsadda Chappal by registering it as Geographical Indicator since its
special kind of hand made chappal is popular all over the world.
• Ensure that announced tanning zones in Karachi and Sialkot are operationalised.
• Ensure footwear parks are established at Muridke and Charsadda. It will be
ensured that the land is available at viable rates.
259

3.3.10 Gems & Jewellery Sector
Pakistan’s gems and jewellery sector comprise fragmented cottage jewellery
manufacturing having rich tradition of craftsmanship and an abundant reserves of
precious and semi precious stones. Jewellery segment is predominantly retail driven
due to a huge local market. Annual local demand for gold jewellery is estimated at
around US$1.2 billion79 . Demand is predominately driven by 22 karat wedding
jewellery. Accumulation of gold jewellery is also perceived as a long-term investment.
The total employment in the sector is estimated at around 800,00080 . Total exports for
the Gems and Jewellery sector were around US$449 million in 2008, representing 1.0
percent of Pakistan’s total exports for the year 81 . However, the hike in gold prices has
increased this number to US$ 479 million in 2009-10.
There are more than thirty major cities and nearly three hundred smaller cities/mandi
towns where jewellery manufacturing and trading clusters cater to domestic demand.
With a rich tradition of jewellery manufacturing, Pakistan’s skilled/semi-skilled labor
force is available at relatively lower rates, which offers a comparative advantage to the
country. The sector demonstrates strength based on local demand and the trend of local
and international branding is on the rise. Even with outdated equipment and
machinery, innovations in designs and product diversity have been happening to meet
the tastes of the local demand. One such innovation is the development of a metal called
‘Manchoos’. This metal in terms of its outlook looks exactly like gold, however, is much
lighter in weight. Due to increased gold prices Manchoos is increasingly being used to
make both contemporary fashion jewellery and also the tradition wedding jewellery. The
existence of the large local demand has provided sufficient depth to several micro
entrepreneurs operating as several clusters all over the country. An example of such a
cluster is the ‘Suha Bazar’ in Lahore. We have presented an overview of the bazaar
characteristics and value chain issues the Box 3-7 below:
79
Pakistan Gems and Jewelry Strategic Working group strategic plan, 2006: 6
80
Industry Resources
81 ‘Review on Pakistan’s exports, 2008-09, TDAP, 2009: 6
260

Box 3‐7: Depiction of Suha Bazar of Lahore
Suha Bazaar is one of the oldest clusters of small and micro enterprises in Lahore. The table below
shows that Suha Bazaar is home to over 25,000 micro enterprises (mostly informal) and employs
over 150,000 workers.
S.NO Processes No. Of Units
1
2
3
4
5
6
7
8
9
10
11
Total
Refining 200 8
Design & Pattern Making 170 3
Die Making-Rubber 2500 4
Die Making – Press 500 7
Jewellery Casting 3200 15
Filing and Polishing 9450 4
Chain Manufacturing 1100 5
Stone Setting 6000 4
Re-polishing 100 10
Beading 200 6
Sales Outlets 3500 6
26920 72
261
Average
Workers
The value chain analysis of the Suha Bazaar cluster is shown in the figure below:
Total
Workers
1600
510
10,000
3,500
48,000
37,800
5,500
24,000
1,000
1,200
21,000
154,110

Raw Mat erial
Waste
is 7%
Filing & Polishing
Chain
Manufacturing
5% -10% value
St o ne Set t ing
improvement
Refining the Raw
Material
Significant wastage can be avoided by better
techniques & technology
Jewellery casting
Sale O ut let s
3.3.10.1 Broader Issues
Jewellery manufacturing is a fragmented cottage industry with a rich tradition of
craftsmanship, however, limited design capabilities, lack of modern manufacturing
262
Value loss 5%
Value Improvement
15%
Re-polishing
Design & Pattern
Makers
Value
improve
ment
50%-100%
Die Makers
Value
Impro
vemen
t 10%
Based on the value chain analysis the following are the major constraints faced by the sector:
1. Inadequate design and pattern making
The current practice of the cluster is based on copying designs and patters from international
magazines. No original design work is done at the Suha Bazaar. This creates wastage and loss of
time as international designs are normally created through nigh powered computerized laser cutting
techniques which is impossible to copy by hand. The results are suboptimal designs.
2. Old machinery
Beading
The cluster has no access to the latest technology, such as CAD-CAM, accurate die making
machines, etc. This reduces the quality of the jewellery.
3. Obsolete skills
The art of jewellery making at Suha Bazaar is a traditional one and has continued as family
businesses. However, the sector has had no formal training in the art of jewellery making. Not
upgrading the skills has resulted in a decline in quality of the jewellery over time.
4. Marketing deficiencies
Suha Bazaar supplies jewellery to all four provinces of Pakistan and has a brand name within
Pakistan. However, no support has ever been provided to take the sector to the next level where it
can be marketed in international export markets.

technology and techniques, and poor international and branding have prevented the
industry from realizing its full potential. There are limited vocational training
opportunities, no investment in research and product development and weak linkages
between training institutes and the industry. Furthermore, due to lack of assaying and
hallmarking, there is no institutional mechanism for providing product guarantee in the
local and international market.
Even in the gems sector, despite abundant reserves of precious and semi-precious
gemstones, Pakistan has been unable to significantly penetrate the international market
for gemstones at favourable price points. Gem reserves are located in Northern Areas
and KP, with significant potential in Balochistan, however there is currently limited
knowledge of the specific quantity of deposits. Mining technology and processes are
rudimentary and unscientific resulting in significant wastage at the extraction stage.
Non-scientific blasting compared to modern mining practices, damages the gemstone
crystals and mineral specimen thus drastically reducing their value. Similarly, limited
understanding of gemology and lack of standardization and certification are other
factors hampering the development in this sector.
Developing the potential of the Gems and Jewellery sector will have a significant impact
on Pakistan’s economy in terms of increase in export revenues, employment and
entrepreneurship, income generation, and consequently poverty alleviation. Consisting
of mainly small and medium entities, growth of this sector will also have positive
externalities for social indicators such as health and education.
The government has already initiated sector level competitiveness project to upgrade the
gems and jewellery sector of Pakistan. As a result of this initiative the government has
established the Pakistan Gems and Jewellery Development Company (PGJDC). PGJDC
has initiated various initiatives which include establishment of Gems and Jewellery
Training centres in provincial capitals and Assaying and Hallmarking Centres in Karachi
and Lahore and has also assisted industry participation in exhibitions and trade fairs.
The results so far have been encouraging and exports have increased manifolds.
263

3.3.10.2 Policy Recommendations:
Capitalizing on its vast natural resources, low labor costs, and skilled craftsmen and
growing national and international demand, Pakistan has the potential to position itself
as a regional hub for precious stone cutting and jewellery manufacturing. In order to
achieve this objective, we recommend the following:
• Continue to support the initiatives launched by the Pakistan Gems and Jewellery
Company. Funding should also be provided for agreed projects.
• Regularly monitor the performance of the company against established targets
and goals to ensure that the company delivers to the needs of the sector.
• Create industry/cluster academia linkage. Pakistan Institute of Fashion Design
offer a full undergraduate course in gems and jewellery making. This course
should have mandatory training with jewellery makers and design makers.
3.3.11 Marble & Granite Sector
Marble and Granite is the fifth largest mineral extracted in the country after coal, rock
salt, lime stone and china clay. There are enormous reserves of marble and granite in
the country. However, little efforts have been made in the past to identify and estimate
marble and granite reserves. Rough estimates done by the industry and government
departments suggest that over 70 types of natural colour marble and granite with
reserves as high as 160 and 414 million tons respectively is available in Northern Areas
alone. Further reserves also spread across Khyber Pakhtunkhwa, Balochistan, and Sindh
(Table 3-10).
Table 3‐10: Marble and Granite Reserves in Pakistan
Province Reserves in Million Tons
Marble Granite
Khyber Pakhtunkhwa 157.9 N.A
Baluchistan 2.2 N.A
Sindh N.A N.A
Punjab - -
Northern Areas N.A 4140
FATA 0.1 -
Total 160.2 4140
Source: SMEDA Marble & Granite Sector Brief, 2006: 7
264

Relative to the potential available the Marble and Granite Industry in Pakistan is lagging
far behind. There are nearly 2000 quarries and more than 1500 processors located
across the country. Almost half of the quarries fall in the formal sector and the
remainder fall in the informal sector.
Although the industry currently represents a small fraction of the GDP and less than 1%
of current exports, the potential to raise exports and foreign exchange, and deliver value
to rural areas is immediate and significant. The marble and granite export market has
grown to over $22 billion in 2008 82 and is still growing further. Pakistan, to date has
only managed to capture a very small fraction, a mere $42 million 83 in exports, of the
total international market. The average exports from the sector over the last five years
are just over $35 million, with products going predominantly to the Middle East, USA,
China, Italy, UK and Germany. Moreover, local market is also a big buyer of marble and
granite. The construction boom in five years preceding 2009 resulted in a huge demand
spike in the local market.
Figure 3‐27: Marble & Granite Exports of Pakistan, 2005‐2009 (US$ Million)
US $ Million
45
40
35
30
25
20
15
10
5
0
2005 2006 2007 2008 2009
Source: UN Commodity Trade Statistics
82 UN Commtrade Statistics (includes category HS 6802, 6815, 2515,2516,2517)
83 Ibid
265

3.3.11.1 Value Chain Analysis
Value chain analysis recently conducted by the industry is presented below (see Figure
3-28). The analysis reveals that that around 73 percent of the potential volume of the
stone is wasted at the extraction and transportation stage due to underdeveloped
quarrying methods and antiquated technology. Stone that is extracted through
indiscriminate blasting is inconsistent and in irregular shapes. Of this, 3 percent is cut
into square blocks for the purpose of exporting. However, the process of cutting square
blocks still leads to 50 percent wastage, so only the remaining half is exported. Out of
the 97 percent that is locally processed, there is an additional wastage of 45 percent; 5
percent is converted into slabs and the remaining 50 percent into tiles. Three percent of
the slabs are exported and the remaining slabs are consumed by the local market,
whereas 2 percent of the tiles are exported and the remaining 98 percent is consumed
by the local market. The cumulative wastage is 85 %.
Figure 3‐28: Value Chain of Marble and Granite Industry
27%
73
Net
Extracted
97%
Irregular
Blocks
3%
Square
Blocks
Local
Processi
ng
45%
266
5%
50%
50
Slabs
Tiles
3%
2%
97%
Export
Local
Export
98
Local
Export

Source: SMEDA Sector Strategy, 2007
3.3.11.2 Broader Issues in the value chain
3.3.11.2.1 Low Productivity
The figure above reveals that at the quarrying level, productivity suffers from
indiscriminate blasting, poor quarrying techniques and lack of infrastructure for
handling and transportation. In majority of the mines basic machinery and equipment
like compressors, drill sets and lifters are not available. This not only leads to colossal
wastage and poor quality blocks but also to low production at mines. Average
production of the majority of mines is 10 tons per day and none of the mine owners and
mining experts at mines knows the reserves, topography of the mine and chemical or
geological analysis of the stone therein.
Products thus obtained are irregular blocks. The irregular blocks produced are difficult
to handle and create problems in lifting, transportation, storage, and fabrication. The
shape of the blocks complicates quality control of the rift or preferred cutting direction
during the cutting process, which is important for strength, background colour
uniformity, and uniform texture. The handling of material is done with basic equipment
like hand-pressed oil jacks, flexible tripod lifting cranes, and cable winding machines.
Such labor-intensive handling of the material is unsafe and not cost-effective.
3.3.11.2.2 Limited Value Addition
Lack of modern technology and skills are the main reasons for limited value addition in
processing. Very few units operate with complete range of machinery and equipment
267

capable of processing stone in accordance with international standards. Utilization of
these units is nearly half of their installed capacity due to inappropriate raw materials.
There are only about 25-30 units which have all appropriate machinery to do one of the
sequential activity, either of cutting slabs section or cross cutting. Thus, the final
product is uncompetitive compared to lower priced, higher quality imports. Pakistan, in
order to realise its potential will have upgrade its industrial structure by benchmarking
activities against the marble and granite sector of Italy (see Box 3-8)
Box 3‐8: Italian Marble Industry
Italy has a strong base of model mine with highest average for quarry production and state of art
processing industry. Its contribution in research and development and human resource development
for mining and processing is noteworthy. Due to prolonged association with this trade Italian has
developed cultural affinity with stone development and the methods and techniques have been
perfected on scientific lines. Italian Marble & Granite industry can be characterized as follows:
• Highly integrated with sophisticated Demand – Consumers, Architects and Construction
Companies
• Industry focused on quality and training
• Diverse Processing Capabilities
• Focus on Cut-to-Size Jobs & Higher Value Addition
• Focus on Environmental impact of quarrying (Driver for Innovation)
• Investment in innovative technology
• Tradition of Quarrying
268

In order to upgrade the sector, SMEDA in partnership with USAID launched sector
competiveness imitative in the marble and the granite sector. Funding was provided by
USAID and sector working group comprising all major shareholders was formulated.
This group was supported by J.E. Austin an American consulting firm. The working
group develop an up gradation plan and recommendations to improve competitiveness
were submitted to the government.
In line with recommendations developed, Ministry of Industries and Production
established Pakistan Stone Development Company (PASDEC) under the umbrella of
Pakistan Industrial Development Corporation (PIDC) to implement the up-gradation
plan. PASDEC has now initiated different projects for the production and development
of different Marble, Granite and Mosaic goods through the use of latest technology. The
establishment of ten Model Quarries throughout the country has been approved, and
three Quarries have already started operations at Khuzdar, Chitral and Bunir. The
feasibility studies of the remaining seven have also been completed. Moreover, the
strategy recommended twenty Quarry up-gradations, out of which five projects at
Mastung-1, Mastung-2, Thatta, Bunir and FATA are operational while the studies of
other fifteen have been completed. Two Machinery Pools have been set up at Risalpur
and Gaddani. Four Common Facility and Training Centers (CFTC) were approved at
Gaddani, Risalpur, FATA and Karachi out of which studies for two centers in FATA and
Karachi have been completed, while for the other two are in process. Two warehouses,
one in Baluchistan and the other in Khyber Pakhtunkhwa were approved out of which
the one at Gaddani is operational while the other is in process.
The industry expects a major increase in productivity and profitability throughout the
value chain by implementing the various initiatives mapped by working group. For
example, the model quarries at Kuzdar, Chitral and Buner has provided substantial
benefits in terms of productivity and quality (see Figures 3-29 and 3-30).
269

Figure 3‐29: Traditional Quarrying Method
Source: Pakistan Stone development Company
BEFORE: Quarrying marble using the age-old and inelegant blasting technique. Waste is
significant, and large blocks of marble are rarely obtained.
Figure 3‐30: Advanced Method of Quarrying
Source: Pakistan Stone development Company
AFTER: Quarrying marble using the sawing process. Using this advanced method of extraction,
waste is minimal, and larger, more valuable blocks of marble are obtained.
270

3.3.11.3 Policy Recommendations:
The marble and granite industry is poised to catalyze economic growth in remote areas of
Pakistan not yet benefiting from the country’s current economic growth cycle. Therefore, we
recommend the following:
• Keep on supporting PSDC and ensure funding is provided for all planned
strategic interventions.
• Establish seven more model quarries in identified areas of Balochistan and KP.
• Support up-gradation of existing quarries.
• Enhance the common machinery pool for extraction of square blocks.
• Operationlaize the planned marble cities at Risalpur and Karachi and provide
requisite infrastructure in partnership with the provincial governments.
• Fund establishment of common facility and training centers at Gadani, Risalpur,
FATA and Karachi.
• Regularly monitor the performance of the company against established targets
and goals to ensure the company delivers to the needs of the sector.
3.3.12 Light Engineering Sector
The engineering industry is generally sub-divided into two sub-categories in Pakistan.
(1) Heavy engineering, producing products like cement, sugar plants, industrial boilers,
construction equipment and transmission towers, and dominated by public sector
companies; and (2) Light engineering, encompassing a range of products such as
agricultural implements, home appliances, iron & steel pipes and tubes, pumps,
electrical fittings and automobile parts etc. Most of the firms in the light engineering
sector are privately owned and/or operated. Pakistan has a significant labor cost
advantage when compared with most similarly situated countries with some type of
domestic engineering and manufacturing industry, and has capable management,
technical and administrative resources.
In order to upgrade the sector EDB worked with the sector stakeholders to map the
extent of existing issues. These were presented to the government in a report prepared
by the EDB as titled National Engineering Exports Development Strategy (NEEDS). The
stakeholder meetings that were conducted in writing this report confirmed the finding
271

of the NEEDS. Below we have summarised a whole list of issues that are faced by the
light engineering sector.
3.3.12.1.1 Non‐Availability of Export Houses
As the sector majorly comprise of small to micro scale businesses there is hardly any
possibility to keep a systematic track of exports and the quality of the goods being sent
abroad. There are no trading houses available which restricts the small businesses to
export to their desired markets. A combination of small business should be supported
and recognized as export houses.
3.3.12.1.2 Inadequacy of Market Surveys and Information
The small size of businesses does not allow them enough financial flexibility to purchase
credible market information. Moreover, there is no concerted effort by the government
to conduct market surveys and collate meaningful market information. The industry is
not connected to its consumers in the export markets and hence is also dragged down by
lack on customer feedback and input.
3.3.12.1.3 Inconsistent Presence at Trade Fairs/ Exhibitions
As the sector is largely fragmented and small in nature it has not benefitted significantly
from TDAP’s facilitation to participate in trade fairs and trade exhibitions. Consistent
year after year presence at trade fairs is extremely important to build credibility with
buyers. The sector does not have enough resources to participate at these fairs and
exhibition solely on their own. Government will have to assist them financially and
technically to represent their products to foreign buyers. Additionally government
through EDB can also facilitate the linkage of these producing clusters with buyer
clusters in foreign market. This can provide a jump start to exports.
3.3.12.1.4 Trade Delegations
EDB and the government will have to work on strategic plan to use diplomatic ties to
create trade links. The high powered delegations from Pakistan should also include a
representative of the engineering sector (possible EDB) to work on developing trade ties
via diplomacy. Currently no such representations follow such visits.
272

3.3.12.1.5 Insufficient Investment
All most all product clusters in light engineering suffer from year of little or no
investment resulting in outdated technology, unskilled workers, poor production
processes and inadequate quality control and production management. This lack of
investment has been due to unavailability or expensive availability of credit from formal
sources. Additionally, the high cost of doing business, regulatory impediments and poor
governance has also caused investment to remain low. In order to produce exportable
products worth more than US$ 10 billion as set out in the NEEDS, the industry will
require significant amount of additional modern technologies and productive assets.
3.3.12.1.6 Limited Capacity of Commercial Banks
The commercial banks have little specialization to cater to export business and
practically no inclination to provide services to Engineering Industry. This result in
unnecessary delays, loss of export orders and increased cost of doing business. There is
a need to establish a financial institution specializing in export related financial
products. Therefore, the NEEDS has emphasized on establishing an EXIM Bank of
Pakistan.
3.3.12.1.7 Insufficient Export Credit Insurance
The risk coverage and insurance needs of engineering exports are diverse and wide
spread as compared to other export sectors. The current spectrum of insurance business
in Pakistan cannot possibly serve the needs of securitizing bank loans, risk coverage of
bid bonds & performance guarantees, credit risk coverage, etc. The current available
insurance is too cumbersome and cause unnecessary delays, therefore, a concerted
effort is required to resolve this issue.
3.3.12.1.8 Sub‐optimal use of Domestic Market
There is a large enough domestic market for the produce of light engineering sectors,
however, no government has ever supported or made it attractive to supply to the local
industry. The contract enforcement and quality standards must be strengthened in the
local markets so that the firms find it equally viable to supply in the local market and
build on a solid foundation.
273

3.3.12.1.9 Levy of Federal Excise Duty on Technology Acquisition
A federal excise duty at the rate of 10% of the charges has been levied on franchises.
However, it has been levied in such a manner that it has become chargeable on
technology acquisition agreements entered to by Engineering Industry as well. This has
increased the cost of doing business and is also impeding investments and exports.
3.3.12.1.10 Lack of Common Facility Centres
Common Facility Centres are essential for manufacture of quality exportable goods by
micro and small scale engineering concerns. Some of these centres have been
established by the Government of Pakistan. However, this activity needs to be further
promoted through involvement of private sector. Where ever possible such centres must
be established through partnerships with the private sector and must be managed by
professionals selected by industry leaders.
3.3.12.1.11 Inadequacy of Testing Laboratories and Standard Certification
Almost every manufactured product that is exported has to comply with some form of
standard, be it quality, health and safety or design. Most of these standards need to me
certified by certified laboratories. Hence, the sector faces multiple problems including;
(i) ability to meet standards and awareness on standards; (ii) testing of standards for
laboratories and to bear associated costs and (iii) the validity of the tests obtained
through local laboratories. The government will have to build awareness and capacity of
the sector to meet these requirements and also build credible laboratory infrastructure
that offer certified test results which are acceptable in destination markets.
3.3.12.1.12 Inadequate Management Capacity
The managerial capacity of the business involved in the light engineering sector is
extremely poor. Most of the businesses are run and managed by single individuals and
their absence is a great threat to continuity of the business. The key person risk is
extremely high and hence, the lending agencies are fearful when extending credit
facilities beyond a certain limit. Moreover, the sector firm do not use employ
professional management and record keeping is at best informal pieces of paper or
small books controlled by the owner. This lack of managerial skills results is poor
costing, inadequate ability to manage business and judge profitability.
274

3.3.12.1.13 Inefficient use of Energy
Energy is an important input for Engineering Industry and has become a large part of
the cost of manufacture. Since, most of the productive assets and manufacturing
practices are old, they tend to consume more energy pushing the cost of manufacture
upward. NEEDS has proposed that ENERCON and NPO should play a proactive role in
raising the awareness level regarding energy efficiency. It has also been proposed that
they should provide assistance in getting the energy efficiency audits conducted and
facilitate in adoption of measures recommended in audit reports.
3.3.12.1.14 Adverse Trade Agreements
The Government of Pakistan has signed a number of trade agreements on bilateral and
multilateral level. Further developments like NAMA are also under discussion at
international scene. The knowledge about such agreements is very scanty among the
Engineering Industry. Moreover, when such agreements are developed and signed the
sector is not taken in confidence before developing the list of provisions being offered –
the FTA with China is a critical example.
3.3.12.1.15 Shortage of Skilled Human Resource
There is a dearth of availability of skilled human resource starting from basic level skills
up to managerial level labour. Good flour managers are not available and technicians
are not available. Some trainings are being offered by TEVTA but the graduates of such
training programmes are not finding jobs in the industry. The TEVTA’s training courses
are mostly outdates and do not match up to the requirement of the industry demands.
In the section below we have illustrated issues and policies for key sub-sectors of the
light engineering sector.
3.3.12.2 Agriculture Implements
The agriculture implements (excluding tractors and tractor parts) industry of Pakistan
consists of a large number of micro and small scale manufacturers throughout the
country. Most of these entities operate out of their back yards and small workshops with
outsourcing for components to other small scale operators. The manufacturers are
clustered in and around Daska, Faisalabad, Okara and Mian Channu. Due to years of
operating in the informal sector and lack of investment has resulted in outdated
275

production assets and technology. The workforce is largely without any formal training,
however, they are rich in innate abilities and raw skills. As peculiar in the informal
sector, the management skills are weak. The manufacturing processes are mostly based
on reverse engineering and hit and trial methods. This limits the design and engineering
capacity and also results in high wastage. The production is not standardized which on
one hand offers the benefit of flexible production but on the other hand limits the
advantages of scale economies. Finance is also a critical impediment impeding growth of
the sector. Against all these odds, the industry has done well in meeting a large local
demand and in accessing international markets. The equipment manufactured in
Pakistan is currently being exported to Afghanistan and in small quantities to Africa.
3.3.12.3 Policy Recommendations:
The agricultural implements industry operates on a small scale and has been a vehicle
for marshalling indigenous savings/investible funds, for the development of
entrepreneurial and managerial talent, for training of skilled and semi-skilled labor, and
for application of new technology. Therefore, we recommend the following:
• Establish a design & innovation and training centre in Daska to provide technical
skills and to help the sector move towards product standardization. The centre
should also be provided with funding to develop a common facility centre.
• SMEDA and PSQCA should facilitate businesses putting up new production lines
to shift towards standardized common parts on large scale required by the
agriculture implements industry.
• Ensure that NPO conducts a productivity benchmarking for the sector and impart
training on lean production methods.
• Support equitable business partnerships through the design & innovation
enabling the developer of a new product to partner with investors for putting up
large commercial units.
• Establish and strengthen a national association for agriculture implements and
help them establish national sales centres especially in remote areas.
• Advocate with the State Bank of Pakistan to design both short term and long term
credit schemes based on the requirements of the sector.
276

3.3.12.4 Automobile Parts
The auto parts industry employs more than 180,000 people and demand for labor is
growing rapidly as capacity expands. The 18 automotive manufacturing companies
active in assembling operations in Pakistan are supported by 800 vendors in the formal
sector and 1,200 in the informal sector. Domestic production consists of manufactured
parts such as pistons, engine valves, gaskets, camshafts, shock absorbers, struts,
steering mechanisms, cylinder heads, wheel hubs, brake drums, wheel bumpers,
instruments and instrument panels, gears of all types, radiators, cylinder liners,
electrical systems, door locks, and auto air conditioners. Approximately 90% of the
automotive parts industry is made up of SMEs, of which about 95% are self-financed84 .
These units produce a wide range of parts for the replacement market.
For illustrative purposes, we have considered the value chain of a typical automobile
radiator manufacturer (see Figure 3-31). The production of radiators requires complex
assembly using semi-skilled and skilled labor combined with technical equipment. This
is an area where it is believed that Pakistan can build a comparative advantage in the
international marketplace.
The analysis reveals that whereas, most international markets have shifted to all
aluminium radiators, Pakistan continues to produce copper tube/brass fin
(copper/brass) radiators. This technology is getting obsolete in international markets.
The copper/brass radiator costs around $68.30 to produce, whereas, the aluminiumradiator
costs $78.00/unit and above, depending on type and function.
Figure 3‐31: Value Chain for Radiator Production
\
Actual material cost
= 74%
Overhead = 18%
Direct Labor = 7%
84
‘Diagnostic study of Auto parts cluster, Lahore’, SMEDA, 2007: 6
277

Source: ‘Pakistan: Growth and export competitiveness’, World Bank Report, 2006: 86
Note: The production of automotive radiators can be broken down into 7 primary stages:
fin production, tube production, header/footer plate production; core assembly; core
finish; tank production and full assembly. The value chain analysis indicates that tube
production, fin production, and tank production constitute the largest value added in the
production process. Fin and tube production constitute over 54% of the total value added,
followed by tank production, core assembly, and full assembly, which account for another
34%.
3.3.12.4.1 Critical Issues
As discussed in the section on auto sector, the policy inconsistency regarding the
deletion programme has resulted in lack of investment in the sector. Inadequate
investment has led to inconsistent quality, out of market products and greater
production inefficiency. Government will be required to provide a consistent policy
support encouraging investment and indigenization in the sector.
3.3.12.5 Policy Recommendations:
The automobile parts industry is a crucial element in the overall supply chain of the
automobile makers. There is a dire need for the manufacturers to redefine their
relationships with this industry. This can be done at two levels: (a) Automakers can shift
more of the responsibility for inventory programs to their suppliers. This will allow the
assemblers to focus their resources on their ‘core capabilities’, which include overall
system design, final assembly and the marketing of the completed vehicle. (b)
Automakers should increase the size and complexity of those items of the vehicle that
are sourced from suppliers from individual parts and components to entire
subassemblies, such as acceleration, braking, steering, handling, and seating systems, or
even larger modules such as integral automobile interiors that include carpets,
headliners and dashboards.
• Establish a design & innovation and training centre in Karachi to provide
technical skills and to help the sector move towards product standardization. The
centre should also be provided with funding to develop a common facility centre.
278

• SMEDA and PSQCA should facilitate businesses putting up new production lines
to shift towards standardized common parts on large scale required by the
agriculture implements industry.
• Ensure that NPO conducts a productivity benchmarking for the sector and impart
training on lean production methods.
• Support equitable business partnerships through the design & innovation
enabling the developer of a new product to partner with investors for putting up
large commercial units.
• Assist leading firms in this sector with development of domestic and
international brands.
• Establish and strengthen a national association for agriculture implements and
help them establish national sales centres especially in remote areas.
• Advocate with the State Bank of Pakistan to design both short term and long term
credit schemes based on the requirements of the sector.
3.3.12.6 Home Appliances
The domestic home appliance engineering industry consists of medium to large size
units in organized sector with enough managerial capacity. The main products include
refrigerators, freezers, air conditioners and washing machines. Most of the
manufacturing is done through assembly relying on imported CKD/SKD kits. The
critical issue facing the sector is the inability to add greater value during assembly.
3.3.12.7 Policy Recommendations:
The quality of the major durable home appliances products must be made as close to as
possible to the international standard. Other factors for success in this industry are
technology importation, its successful assimilation, absorption and customized
development. We recommend the following:
• Establish CAD-CAM designing facility to develop new product designs, improve
quality and bring standardization in production parts.
• Support provision of market information and linkages with international design
houses to assist the sector add value through design innovation.
279

• Explore possibility of increasing the potential of intra-industry trade within
South Asia region.
• Share cost of compliance with international standards such as ISO, CE, UL and
other such marks conditional on companies meeting their export targets.
• Ensure that NPO conducts a productivity benchmarking for the sector and impart
training on lean production methods.
3.3.12.8 Iron & Steel Pipes and Tubes
Iron & Steel pipe and tube industry in Pakistan produces a large range of products from
cast iron, galvanized iron pipes, cold rolled tubes, square Cathode Ray tubes,
rectangular CR tubes, elliptical CR tubes, API line pipes, black pipes, structural pipes
and spiral welded pipes. In seamless tube category the industry produces heat exchange
tubes, pressure tubes, boiler tubes, low temperature service tubes, precision shaft tubes,
linked tubes, etc. as per international standards. The sector is sharply fragmented with a
few large scale corporate units having sufficient managerial, technological and financial
capability and several medium and small scale manufacturers. The industry needs to
expand its production capacities to produce enough exportable surpluses for growth in
exports. This will require small and medium scale firms upgrading their product range,
increasing their scale and becoming compliant with international quality standards.
Moreover, the current strength of the industry is in producing large diameter pipes,
which attract prohibitive cost of freights and cannot be exported to far distance markets
as they cannot remain price competitive. The government will have to facilitate the
sector in moving into innovative products that are more competitive globally.
3.3.12.9 Policy Recommendations:
• Share the mark up costs of small scale producers expanding their production
facilities to enhance their product mix.
• Provide technical support to companies through PSQCA in improving quality,
standardization and compliance requirements.
• Provide support to PSQCA to establish domestic standards of quality and
standardize product range.
280

• Improve the logistical facilities and advocate for cost efficient freight services on
long routes.
• Establish a product design and development centre to facilitate innovation in the
sector. This centre will also provide common facilities for manufacturers.
• Ensure that NPO conducts a productivity benchmarking for the sector and impart
training on lean production methods.
3.3.12.10 Pumps
The pump manufacturing is dominated by an unorganized SME sector. The industry is
concentrated and clustered in and around Karachi, Lahore, Gujranwala and Faisalabad.
There are about 70 registered and relatively organized units. The number of
unregistered and unorganized units is estimated to be over 500. This number also
includes indirectly involved businesses of vendors supplying to manufacturers operating
at medium, small and micro level. The industry has been experiencing falling
competitiveness in spite of the rise in product prices. This is attributed to be rising
production costs, high prices of raw materials and a cut throat competition through
price cutting in the domestic market. The other issues impeding the growth of the sector
include unskilled labour, outdated technology and lack of coordination with support
agencies. Lack of standardization and quality control has also halted the growth
potential of the sector, especially in the export markets. Limited investments have
generally resulted in low scales of production, a weak product mix and a general
dependence on obsolete designs and quality.
3.3.12.11 Policy Recommendations:
• Develop local quality and product standards to shift production towards better
quality and standardized products. This will be done in consultation with the
sector.
• Share cost of complying with established quality standards.
• Establish a centre of excellence in Gujranwala to work on design development,
quality improvement and standardization of products.
• Ensure that NPO conducts a productivity benchmarking for the sector and impart
training on lean production methods.
281

3.3.12.12 Electrical Fittings
The electrical fittings manufacturers are concentrated in Sargodha (70%), Lahore (20%)
and Karachi (10%). There are around 20 small scale units, around 200 small workshops
and around 1,000 manufacturers in cottage sector85 . The total estimated monthly
production is nearly 10 million pieces of various fittings per month. The sector faces
acute competition from import of similar goods from China. The critical impediment to
growth in the sector is its inability to meet the quality and standards requirements. The
quality of these fittings is extremely poor and in most cases the durability is not beyond
few weeks. In addition, the design and shapes vary significantly making most fittings not
suitable for use with majority of electric equipment.
3.3.12.13 Policy Recommendations:
The important factors for competition in the electrical fittings industry are product
design, performance , price and technical services. In the sale of electrical apparatus, the
reputation of the manufacturer as a producer of high quality and dependable products is
a factor of overriding importance. Therefore, we recommend the following:
• Establish a product design centre and testing facility in Sargodha that should also
be responsible for defining national standards on quality, design, shape and
durability. The design centre should over time refine domestic standards to
become in line with international standards such as CE mark.
• Ensure strict compliance with these domestic standards and ban sale of noncompliant
goods in the domestic market.
• SMEDA and PSQCA should support companies upgrading their products to
comply with defined standards.
3.3.12.14 Steel Structures/Towers
Pakistan steel structures industry produces towers for electricity transmission and
towers to support telecom operations (fixed line as well as cellular). A number of
companies are manufacturing standardized towers and other structures for this
purpose. There are around 10 units operating in the organized sector. They manufacture
steel structures using design drawings and use the required quality of steel as per
85
‘Electrical fittings cluster, Sargodha’, SMEDA, 2005: 4
282

specifications. Exporting steel towers to African countries offer significant potential.
However, steel towers cannot be exported in assembled form and need to be assembled
on site. Therefore, the prospective exporters need to establish a business of installation
of structures in the country of import or establish partnership with an installation
company which can market their products as well as undertake installation. Unless, this
relationship is established, export may not become possible.
3.3.12.15 Policy Recommendations:
• Share cost and facilitate through the foreign office the companies establishing
installation facilities in export markets and should facilitate/coordinate these
efforts through foreign office support.
• Conduct detailed market studies to provide a marketing strategy for the sector.
3.3.12.16 Prefabricated Buildings
Prefabricated buildings industry has recently picked up momentum in Pakistan. These
buildings are increasingly being used as grain storage silos, sheds, industrial
applications, commercial activities & warehousing and hospitals/ schools/ emergency
housing in disaster hit areas. It is a critical support sector to develop low cost industrial
sheds and housing. The critical challenge faced by the sector is lack of awareness in
using these prefabricated buildings in the market.
3.3.12.17 Policy Recommendations:
Advocate the use of pre-fabricated buildings in all government schemes of low cost
housing and infrastructure projects where applicable.
3.3.13 Fisheries Sector
The Pakistan fish and seafood industry is worth $1.2 billion with exports of fish and fish
products from Pakistan amounting to $213 million per annum (TDAP, 2008). With its
1,100 km coastline and fishing area of more than 3000 sq. kms, Pakistan has a rich
marine life. While larger than Sri Lanka, Pakistan’s fisheries sector is around half that of
Bangladesh and a fifth of India’s in terms of fish caught but even smaller in terms of
number of fishermen. The domestic market consists of a large number of informal and
formal, small-scale fishermen, combined with a small number of medium and large-
283

scale commercial fishing vessels. It is estimated that over 12,000 boats and vessels are
registered, but fewer than 4,000 of various sizes are currently active. The fisheries
sector provides employment to about 361,000 people, of whom almost 40% are engaged
in marine fisheries, and over 60% 86 in inland fisheries. Most fish are caught in Sindh,
landed at fish harbours in Karachi and Korangi, and in Balochistan at harbours in
Gawadar and Pasni. In particular, the Karachi Fish Harbour was originally built in 1956
to accommodate 400 vessels, but the demand now exceeds 2,000 vessels, overcrowding
the already over-utilized facility.
3.3.13.1 Critical Issues
The seafood industry is currently facing two critical issues (i) depleting fish stock and
(ii) poor quality control and inadequate hygiene measure which result in the value of the
catch not being maximized and also results in high levels of wastage. As a result of these
hygiene issues all types of sea food export to EU is currently banned.
3.3.13.2 Policy Recommendations:
The aim of the intervention in the fisheries sector is (a) to generate future employment
opportunities through the exploitation and processing of marine products; (b) to
increase production to satisfy local fish and marine products demand; (c) to increase the
value of fish for export and (d) to regulate and control the exploitation of fishing
products. Keeping these in view, we recommend the following:
• Create awareness on Sanitary and Phytosanitary (SPS), health and safety
standards that are required to comply with international requirements.
• Invest in expansion of the existing laboratory infrastructure to provide testing
and certification facilities acceptable to destination markets
• Provide financial support to revamp fish jetties at Gadani, Dam, Pasni and
Jewani.
• Ensure provision of facilities such as landing stations/jetties/port along Makran
Coast in partnership with the provincial government of Balochistan.
• Support Korangi Seafood processing companies in developing traceability system
for smoked, canned, fresh and frozen fish products.
86 ‘Action plan for fish quality and value adding at Karachi Fisheries Harbour’, Competitiveness Support Fund, 2007:
18
284

4 Spatial Concentration of Economic Activity
4.1 Introduction
The most striking feature of economic activity in Pakistan is the geographic
concentration (or clustering) and location of factors of production in few cities,
including unequal spatial distribution of income, poverty, education, health and
physical infrastructure, among others. The concentration of economic activity in few
metropolitan areas symbolizes the coexistence of development and
underdevelopment within and between regions. Another manifestation of the
problem is that the share of urban population has dramatically increased since 1960.
While there are a number of theories that explain why geographical concentration of
economic activity takes place87 , little is known about the factors that drive internal
economic geography of developing countries, what course this inequality takes, and
whether this inequality is socially desirable. In this chapter we examine economic
geography of Pakistan in more detail. In this regard, we explore regional and spatial
inequality by documenting various mapping measures that highlight spatial
distribution of population, regional poverty and income inequality, spatial disparities
in social infrastructure, road infrastructure, and concentration of manufacturing
industries.
By regions we mean four provinces, namely Punjab, Sindh, Khyber Pakhtunkhwa
(KP) and Balochistan, which represent the first-level of administrative jurisdiction in
Pakistan. By spatial units we mean districts, which represent the third-level of
administrative jurisdiction after administrative divisions. Because the area boundary
changes for the districts are quite common, the number of administrative districts
has increased from 43 in 1951 to 106 in 1998. To avoid distortions and inconsistency
87 For early works, see Marshall (1890), Weber (1909), Hotelling (1929), Florence (1948), Hoover
(1948), Fuchs (1962), Henderson (1974), among others. For more recent contributions, see among
others, Krugman (1991a, 1991b), Krugman and Venables (1995), Kim (1995), Krugman and Livas
(1996), Ellison and Glaeser (1997), Fujita et al. (1999), Puga (1999), Fujita and Thisse (2002), Hanson
(2005).
285

overtime, we freeze the district boundaries at the time of 1981 population census
when the number of administrative districts was 65. However, we select 56 districts
for the analysis to allow consistency in all the data with household surveys which
identify all households by their district codes, but households residing in Balochistan
are identified by their administrative divisions instead of districts. We adopt a
consistent definition and select only 56 districts where four administrative divisions
of Balochistan are also treated as districts while Islamabad and Rawalpindi are
merged to form a single district.
This chapter is broken into two parts, followed by conclusions. In the first part we
review the literature on geographic concentration and spatial inequality around
regional and urban economics known as New Economic Geography and document its
implications on further research. When regional and spatial divisions align with
political and ethnic pressures they give rise to political and social instability. But
despite obvious policy concerns there has been no systematic evidence on regional
and spatial mapping of key economic activities in the country. Therefore, in the
second part of this chapter, we document a number of mapping measures of regional
and spatial inequality to present the nature and magnitude of spatial and regional
inequality in the country.
286

4.2 New Economic Geography and Development
We begin by briefly reviewing the literature on geographic concentration and spatial
inequality, premised on New Economic Geography popularized, among others, by
Krugman (1991a, 1991b). In recent years, progress made in successfully modeling
increasing returns to scale has made it possible to analyze the economics of
agglomeration [see, Dixit and Stiglitz (1977), Krugman (1991b), Fujita et al. (1999)].
These innovations have guided scholars to formalize some traditional concepts in the
literature to explain geographic concentration of industry. They include the
Marshallian concepts of the role of technological spillovers, pooling of market for
skilled workers, and availability of non-tradable intermediate inputs as well as
Hirshman’s (1958) nonpecuniary externalities on account of forward and backward
linkages and market access due to Myrdal (1957) and Arthur (1990).
In general, increasing returns to scale play an important role in explaining why
economic activities are spatially concentrated. The ‘folk theorem’ of spatial economics
[see, Fujita and Thisse (1996)] says that under non-increasing returns when high
transportation costs are present, industry would locate at diversified places to
minimize cost of reaching consumers. In this case, many firms would operate at small
scale to produce for the available market. However, when increasing returns to scale
are present, firms would benefit by locating at concentrated places where they could
enhance production to cater increasing demand. Increasing returns tend to decrease
per unit cost due to specialization of labor and improvements in technology leading to
internal economies [Lall et al. (2004)].
The agglomeration economies consist of localization economies and urbanization
economies. Localization economies refer to within-industry or intra-industry benefits
accruing through knowledge-diffusion, buyer-supplier networks, subcontracting
facilities and a pool of skilled workers. Urbanization economies arise from acrossindustry
spillovers such as supply of other complementary services, e.g., financial
institutions, marketing and advertising agencies, and other cheap infrastructure, etc.
287

The phenomenon of localized versus dispersed industries is explained by the
centripetal and centrifugal forces. The centripetal forces of increasing returns lead to
concentration of activities arising from internal and external economies from
interaction, while the centrifugal forces lead to dispersion of activities arising from
over concentration of firms in an area that increases costs of immobile factors, e.g.,
higher land prices and land rents, higher wages and higher commuting time for
workers. Rising costs of agglomeration deter further concentration of firms in the
surrounding areas and pull economic activities in the opposite direction [Fujita and
Thisse (1996)]. A balance between centripetal and centrifugal forces leads to
equilibrium [Kruger (1991, 1998)].
Spatial inequality outcome is determined by the balance between centripetal and
centrifugal forces in a region. The theoretical models suggest that this balance would
depend on model parameters [Krugman (1991b), Henderson et al. (2000), Kim
(2008)]. For example, when transport costs are extremely high, the industry would
be highly dispersed. With immobile labor, an initial decrease in transport cost would
lead to concentration of industry. However, when transport cost is extremely low and
labor is immobile, industry would tend to spread across regions because further
agglomeration of industry would increase prices of immobile factors. So
agglomeration of firms would be highest at intermediate levels of transport costs
because the firms would like to exploit cost and demand linkages [Puga (1999)].
A number of policy implications emerge from the existing literature that helps our
understanding of the spatial inequality in developed and developing countries [Kim
(2008)]. Firstly, the magnitude of localization economies (within industry spillovers)
is much more than urbanization economies (across industry spillovers). If policy
makers want to influence spatial inequality they may use “industry-specific” subsidies
to get desired results [Henderson et al. (2001), Kim (2008). Secondly, in general,
“extractive industries” are more concentrated followed by manufacturing while
services are most dispersed [Henderson (1988), Chen (1996), Henderson and
Kuncoro (1996), Henderson et al. (2001)]. In manufacturing, traditional industries
288

such as textiles are more concentrated while high-tech industries are more dispersed.
However, there is no consensus on the most important source of agglomeration
economies [see also, Rosenthal and Strange (2004), Overman and Venables (2005)].
Thirdly, political institutions matter in determining regional and urban spatial
inequality [Kim 2008)]. Presence of strong state and local government in a country
tends to promote greater spatial equality as compared with countries where federal
government is relatively strong. Centralized versus decentralized governments have
different political motivations due to which they prefer different set of public
infrastructure that have bearing on spatial inequality. Fourthly, investments in
transport and communication infrastructure seem to promote spatial equality across
regions. In this regard, Rosen and Resnick (1980) indicate the link with railroad
investments, and Henderson (2002) and Baum-Snow (2007) with road and highway
investments. Fifthly, a policy meant to reduce EU spatial inequality has remained
ineffective [Puga (2002)], while the Korean policy to reduce excessive agglomeration
of industry around Seoul has been successful [Henderson et al. (2001)]. Finally,
there is weak empirical evidence on the inverted-U pattern of regional inequality, but
spatial inequality has the tendency to rise and fall with development patterns [Kim
(1995), Krugman and Venables (1995)].
While the empirical evidence on spatial inequality in developed countries may have
immense value, the patterns of industrial development, market efficiency and
political and institutional histories in countries like Pakistan are fundamentally
different from the developed countries. When regional and spatial divisions align
with political and ethnic pressures they give rise to political and social instability in
the country. But despite obvious policy concerns there has been no systematic
evidence on regional and spatial mapping of key economic activities in the country.
We turn to these questions below.
289

4.3 Mapping Measures of Regional and Spatial Inequality
In this section, we document the nature, trends and magnitude of regional and spatial
inequality in the federating units and the districts. We employ different measures of
concentration to gain insight into spatial inequality as a function of different push and
pull factors that explain concentration versus dispersion of economic activities.
4.3.1 Regional Spatial Distribution of Population and Density
The population growth rate of Pakistan increased from 1.57% in 1951 to 3.29% in 1961,
but subsequently fell to 3.03% in 1981 and 2.17% in 1998 [Afzal (2003)]. While life
expectancy at birth significantly improved, the fertility rate remained constant from the
early 1960s to the late 1980s; however, as fertility rate began to decline in late 1980s the
population growth rate fell [Feeney and Alam (2003)]. The fertility rate fell more rapidly
in regions that experienced more rapid income growth because the opportunity cost of
procreation increased with increased cost of raising children and more
employment/income-earning opportunities for women.
The percentage share of Punjab in total population has consistently declined from
60.9% in 1951, 57.6% in 1972 to 55.6% in 1998. In the same period, the share of Sindh
province in total population has continued to grow from 17.9% in 1951, 21.7% in 1972 to
23% in 1998. However, the intercensal share of Khyber Paktunkhwa has remained
constant while the share of Balochistan has increased from 3.5% in 1951 to 5% in 1998
[Khan (2003)]. Urban population growth has been much faster than rural population
growth in the intercensal period: urban population in 1998 as a ratio of 1951 urban
population was 7.2 times while the corresponding ratio for rural population was 3.2
times [Afzal (2003)].
The spatial distribution of population provides some valuable insights on the changing
characteristics of the provinces and regions determined over time by a number of
290

historical, socioeconomic, cultural factors and strategic concerns, among others. The
regional distribution of population has key significance to provinces due to its
repercussions on their political representation and rights in the federation, distribution
of resources and employment quotas as provisioned in the Constitution. The needs of
the metropolitan cities, urban and rural areas at the time of planning at the federal and
provincial levels for basic school and hospital infrastructure, road network,
telecommunication, civic amenities, etc. are all determined by human habitation apart
from strategic considerations.
Population density indicates concentration of population and is measured by the
persons per square kilometer of land area. Table 4.1 shows population density of
Pakistan by regions. Due to diversity in population growth rates and internal
migrations, the change in density across regions has not been uniform. For example,
population density of Pakistan increased four-fold between 1951 and 1998: it increased
from 42 persons per kilometer in 1951 to 166 persons in 1998. Punjab was the most
densely populated province in 1998 with 359 persons as against 100 persons per square
km in 1951. This was followed by Khyber Paktunkhwa with 238 persons per square km
in 1998 against 61 persons in 1951. The population density in Sindh increased five-fold
from only 43 persons per square km in 1951 to 216 persons in 1998. Islamabad capital
territory experienced eight-fold increase in population density followed by Balochistan
with six-fold increase in density while FATA experienced only two-fold increase in
density that was slowest of all.
Table 4‐1: Population Density by Area
Province/Area Population density measured by persons per square
kilometer
1998 1981 1972 1961 1951
Pakistan 166 106 82 54 42
Islamabad 889 376 262 130 106
Punjab 359 230 183 124 100
Khyber Pakhtunkhwa 238 148 113 77 61
Sindh 216 135 100 59 43
FATA 117 81 92 68 49
Balochistan 19 12 7 4 3
291

Source: GoP (2001)
Khan (2003) discusses district level variation in population density on the basis of 1998
population census. He shows that Lahore, Gujranwala, Faisalabad, Sialkot and Multan
are the most densely populated districts in Punjab while Southern districts such as
Rajanpur, Bahawalpur, Rahim Yar Khan, and Layyah are the least dense districts.
Similarly, Karachi is the fastest growing area in terms of population density in Sindh,
followed by Hyderabad, Nausharo Feroze, Shikarpur and Mirpur Khas; all have
population density of more than 300 persons per square km. In Khyber Pakhtunkhwa,
Peshawar remains highly densely populated district and Chitral remains the least dense
district. Likewise, Quetta being the capital city is the most densely populated district in
Balochistan. This is followed by Jafferabad and Killa Abdullah.
In general, districts with population density of more than 600 persons per square km
are characterized by industrial development, improved education and health
infrastructure and better sanitation facilities, e.g., Karachi, Lahore, Peshawar,
Charsadda, Gujranwala, Faisalabad, Sialkot, Mardan, Islamabad, Multan, Swabi, Gujrat
and Rawalpindi [Khan (2003)]. It is also noted that districts with medium population
density, i.e., 300–600 persons per square km, are endowed with agricultural resources
and some industry presence. These factors serve as important pull factors for the
population to gradually move from other regions to these regions. By contrast, push
forces are rampant in relatively less densely populated districts that drive population to
migrate to more developed areas. For example, districts with lowest population density
(or below 30 persons per square km) are characterized by absence of job opportunities
due to low education, poor agricultural endowments, barren or mountainous land and
no or very little presence of industrial units [Khan (2003)].
Do spatial variations in provision of schooling infrastructure and hospitals influence
population growth? Does spatial distribution of manufacturing production indeed serve
as a pull factor for population growth? We evaluate these questions by using simple
292

scatter plots relating district level population with district-level education, health and
manufacturing production.
In Figure 4.1, the measure of post-primary school system is obtained from the principal
component analysis based on data of ten educational indicators. To illustrate, we collect
data of ten educational indicators from the Provincial Development Statistics, published
by the Bureaus of Statistics of the four provincial governments. The first five variables
represent the size of schools (primary, middle and secondary) and colleges
(intermediate and degree colleges). 88 The next five variables represent quality of
school/college system in respective districts measured by teacher-student ratios in the
same schools and colleges. In doing so, we assume that public spending is positively
associated with school/college size and teacher-student ratios. The principal
components were obtained by using the Factor command in Stata software version 10.
Four factors were retained using Kaiser eigenvalue criterion after varimax rotation that
account for 70.9% variation. Factor 1 accounts for 23% weight in the total variation
explained in the data and represents post-primary school system index as it is
characterized by high positive loadings on size of degree college (0.878), size of
secondary schools (0.807), size of intermediate colleges (0.789) and size of middle
schools (0.538).
Figure 4‐1: Post‐primary school system and current population, 2005‐06
88 Since these institutions vary by size, we multiply the number of schools/colleges by their respective
enrolments and also weigh them by district area and population to generate school/college size in 1985-
86.
293

Log of population 2005-06
13 14 15 16 17
badin
karak
jhelum
sibi
faisalabad
T.T singh
karachi
rawalpindi
lahore
-5 0 5 10 15
Quantity of post-primary school system 2005-06
Log(population) Fitted values
Figure 4.1 plots district level post-primary school system index and district population.
The measure on the horizontal-axis is post-primary school system index of each district
in 2005-06; the vertical-axis shows the logarithm of population in 2005-06 in each
district. In the scatter plot each data point represents a district. With a correlation of
2
0.104 ( SE = 0.025; R = 0.223),
the upward sloping line in Figure 4.1 shows that unequal
government investment on post-primary school system is currently positively correlated
with district level population.
Figure 4.2 plots hospital quantity and district population in each district. 89 The gently
upward sloping line shows that the districts with more hospitals have higher current
levels of population.
Figure 4‐2: Hospital size and current population, 2005‐06
89 Since public spending on hospitals vary by the number of hospital beds, we construct hospital size
variable by multiplying the number of hospitals with number of beds in each district, weighted by district
area and district population.
294

Log of population 2005-06
13 14 15 16 17
sibi
karachi
faisalabad
gujranwalamultan
karak
rawalpindi
kohat
abbottabad
0 2 4 6
Quantity of hospitals 2005-06
Log(population) Fitted values
peshawar
Figure 4.3 relates value of large-scale manufacturing production and population growth. On the
horizontal-axis, the value of manufacturing production is taken from the enterprise level data in
the original data file of the Census of Manufacturing Industries (CMI), 2005-06. 90 The measure
of population growth on the vertical-axis is the log difference of district population in 2005-06
with population in 1981-82. This figure indicates that the value of manufacturing production is
currently positively correlated with growth of district level population from 1981 to 2005: with a
correlation of 0.0004 the fitted values have
2
SE = 0.00018; R = 0.094 .
These results make sense since public sector infrastructural investments in schooling and health
are not uniform across spatial units that serve as push and pull factors to determine
demographic composition of districts, while concentration of large-scale manufacturing and
ancillary industries in a few districts also serves as strong pull factors for population to gradually
move from other regions to these districts.
Figure 4‐3: Value of large‐scale manufacturing production and population growth
90
Again, we retain 56 districts as in 1990-91 district classification and accordingly adjust district level data
on value of production obtained from CMI.
295

Log of population growth from 1981-82 to 2005-06
.2 .4 .6 .8 1
quetta
karak
nawabshah
T.T singh
kalat
tharparker
lahore
faisalabad
karachi
0 200 400 600 800
Value of production 2005-06
Log(population growth) Fitted values
4.3.2 Measuring Regional Poverty and Income Inequality
Knowledge about the regional location of the poor is essential if the government is to
adopt a sound industrial development policy that is more effective in attacking poverty.
The distinction must be made between poverty and inequality. Poverty relates to
absolute living standards of a section of the society in a country. Inequality is concerned
about relative incomes or standards of livings vis-à-vis the whole society. Both poverty
and inequality have significant regional dimensions in developing countries. Accurate
answers to two questions are critical in guiding future policy that aims at faster
convergence in living standards between most deprived and least deprived regions. One,
how much poverty (and inequality) is there in Pakistan, across provinces and across
urban and rural dimensions? And two, what are the trends in poverty (and inequality)
over the last two decades? Unfortunately, the answers to these questions are blurred by
inconsistent estimates on poverty (and to some extent on inequality) reported in the
existing literature.
296

By far the most commonly used measure of poverty is the headcount index, which
measures the proportion of population lying below the poverty line. But, most poverty
studies conducted in the last two-decades on Pakistan have come up with significantly
different poverty estimates, which make these numbers controversial. While there is
controversy on the number of poor across survey years, there seems to be a consensus
that rural poverty has continued to be higher than urban poverty. Significant differences
across the four provinces have also been documented in the literature. What is most
surprising in this debate is that despite same data source, the outcome in the form of
poverty estimates has always been different. What are the factors that explain why
poverty estimates significantly differ across studies in 1990s? To answer this question it
is necessary to examine the underlying assumptions, methodology and data cleaning
protocols used by the previous studies.
Table 4.2 lists most of the studies that have been conducted on poverty in Pakistan
during the last two-decades. It reveals that these studies have mostly used calorie-food
expenditure function, suggested by Ercelawn (1990), to estimate poverty lines, however,
the calorie norms used in these studies vary widely ranging from 2150 calories to 2550
calories. To smooth consumption
Table 4‐2: List of poverty studies and their methods
Study/period Calorie norms
Used
Amjad and Kemal;
1963/64-1992/93
Anwar; 1987/88 and
1990/91
FBS; 1992/93-
1998/99
FBS; 1992/93-
1998/99
Overall 2550; Urban
2550; Rural 2550
Method Applied Base-year
for
Poverty
Lines
Average consumption of lowest
three Consumption brackets
1984-85
2550 Calorie-total expenditure function 1984-85
Rural 2550 Urban
2550
Rural 2150 Urban
2150
Calories-total expenditure
function
Calorie-total expenditure function
excluding expenditure on
marriages, funerals and durable
items
1998-99
1998-99
297

Ghaus-Pasha and
Jamal; 1996-97
Jafri; 1986/87-
1993/94
Planning
Commission (Official
Poverty Line)
1998/99
Qureshi and Arif;
1993/94 and 1998/99
Qureshi and Arif;
1998/99
Rural 2550; Urban
2230
Rural 2450; Urban
2150
Overall 2350; Rural
2450; Urban 2150
Rural 2550; Urban
2295
Rural 2550; Urban
2295
Source: Anwar and Qureshi (2002) and other sources
Calorie-total expenditure function 1996/97
Calorie-food expenditure function
& Average non-food expenditures
given that food expenditures
equal to the minimum prescribed
Calorie-total expenditure function
excluding expenditure on
marriages, funerals and durable
items
Estimated
for each
year
1998-99
Calorie-food expenditure function Estimated
for each
year
Calorie-food expenditure function
Average non-food expenditures
given that food expenditures
equal to the minimum prescribed
1998-99
differentials on the basis of gender, state of health and age these studies assign different
weights to account for adult equivalence in calories and expenditures. Some studies also
use different calorie norms for urban and rural populations. Few poverty studies opt for
basic needs approach and work out basic needs poverty that takes into account both
food and non-food basic needs. Researchers have also included or excluded large
infrequent expenditures, e.g., marriages, funerals and expenditure on durable items, in
the list of expenditures that can clearly distort the outcome for a household in a
particular year. The poverty results are also likely to differ across studies on account of
various base-years used to calculate poverty lines as well as due to constructing regionspecific
(rural-urban) poverty lines. The data cleaning protocols adopted by previous
studies are like a black box since most studies do not report their criteria for cleaning of
the outliers in the data. However, a meaningful comparison of poverty ratios across
studies demands that these studies report how many observations were deleted from the
full sample and what was the basis to delete them.
298

Moreover, most of these studies do not take into account regional inflation to make
adjustments for spatial price heterogeneity, which is now a standard procedure used in
recent studies.
Even though various studies have been conducted to evaluate poverty in the 1990s, the
studies by FBS (2001) and World Bank (2002) were the most comprehensive accounts
of the poverty profile in Pakistan. Table 4.3 reveals the poverty ratios reported by
previous studies where we can see that the estimates are very sensitive to the methods
employed and yardsticks used for poverty lines. 91 In the light of these estimates, it is
very difficult to reach at a consensus on the impact of public policy on poverty.
Table 4‐3: Poverty estimates for the 1990s reported by previous studies
Year 1987- 1990- 1992- 1993- 1996- 1998- 2000-
88 91 93 94 97 99 01
Ali and Tahir (1999) 19.18 23 28.11 27.93 -- -- --
Amjad and Kemal
(1997)
17.32 22.1 22.4 -- -- -- --
Anwar (1996) 13.81 17.26
Anwar and Qureshi
(2003)
-- -- -- -- -- 27.7 30.9
Arif, Nazli and Haq
(2001)
-- -- -- 27.4 29.6 35.2 --
FBS 2150 calories -- -- 22.2 25 21.8 28.2 --
FBS 2350 calories
(official poverty line)
91 See Anwar and Qureshi (2002) for a review.
-- -- 24.9 27.7 24.5 30.6 --
299

FBS 2550 calories -- -- 26.6 29.3 26.3 32.2 --
Ghaus-Pasha and
Jamal (2001)
-- -- -- -- 31 -- --
Jafri (1999) 29.2 26.1 26.8 28.7 -- -- --
Qureshi and Arif
(2001)
-- -- -- -- -- 35.2 --
World Bank (2002) 30.7 34 26.7 28.6 24 32.6 --
Source: Anwar and Qureshi and others
-- not available
4.3.2.1 Evidence on regional poverty
To dispel some misconceptions about the nature of poverty and its trends, this study
adopts a consistent methodology to estimate a time-series of headcount poverty. This
methodology is very similar to the official methodology used to estimate poverty
headcount rates by the Centre for Poverty Reduction and Social Policy Development
(CPRSPD) based in the Pakistan’s Planning Commission.
The head count ratio of poverty gives the relative incidence of the poor, or the number of
people below the poverty line. If expenditure (income) is denoted by y and subscripts
i refers to individuals, and p refers to the poverty line, then head count denoted by H is
given by the number of individuals so that yi< p.
The head count ratio ( p 0 ) is written
as p ( H n)
0
= × 100 , where n is the total population in the reference group.
Our analysis uses the official calorie-based poverty line corresponding to calorie intake
of 2350 calories per day per capita. Total expenditures include expenditure on food,
fuel, education, health, non-durables, etc. Tax and fees along with durable goods
expenditure are excluded from the household expenditures. We calculate adult
equivalence by assigning weight of 0.8 to household members younger than 18 years old
and full weight to all others [see Cheema (2005 for more details)]. The official poverty
line is inflated or deflated by the consumer price index (CPI) to keep the real value
constant. To adjust for regional heterogeneity in prices (regional inflation) we use PSU
level spatial price index (Paasche index) to normalize the effects on prices of food items.
300

All members of a household are categorized as poor if the household expenditure per
adult equivalent is below the official poverty line.
The data comes from Pakistan’s Household Income and Expenditure Surveys (HIES)
administered by the Federal Bureau of Statistics, Government of Pakistan. 92 We use
data drawn from seven rounds of HIES surveys from 1990-91 to 2005-06. The included
survey rounds are PIHS-HIES 1990-91, HIES 1992-93, HIES 1993-94, HIES 1996-97,
PIHS-HIES 1998-99, PIHS-HIES 2001-02, and PSLM-HIES 2005-06. 93 The survey
selects households by using stratified random sampling methods by identifying primary
sampling units (PSUs) in 8 metropolitan cities, other urban stratum and rural stratum.
The inflation adjusted poverty lines used to estimate poverty for the intervening period
are given in Table 4.4.
Table 4‐4: Inflation-adjusted poverty lines per adult equivalent per day used for poverty
estimates
Year Poverty line (Rs)
1990-91 326.5
1992-93 400
1993-94 501
1996-97 618
1998-99 673.40
2001-02 723.40
2005-06 948.47
Based on our estimates, Table 4.5 and Figures 4.4 to 4.6 give a comparative picture of
headcount poverty of Pakistan and the four provinces from 1990-91 to 2005-06. The
results reveal that the incidence of poverty has significantly decreased from 1990-91 to
1996-97, it has increased in 1998-99 and has fallen again in the subsequent period. We
note that poverty in Pakistan is largely a rural phenomenon. While rural poverty has
closely matched the trend of overall poverty, urban poverty has decreased in the
92 Most of the studies on poverty and inequality in Pakistan have used Household Income and
Expenditure Survey (HIES), and Pakistan Integrated Household Survey (PIHS) data.
93 These surveys, except PIHS-HIES 1990-91, are representative at the province level as well as at the level
of rural and urban areas. However, caution is warranted since PIHS-HIES 1990-91 survey is only ruralurban
and national representative.
301

eginning; it has remained constant until 2000-01 before falling again in 2005-06 (see
Figure 4.4).
Table 4‐5: Regional poverty in Pakistan, 1990-91 to 2005-06
Province/overall 1990-91 1992- 1993- 1996- 1998- 2000- 2005-
93 94 97 99 01 06
Overall 37.41 32.28 30.94 27.11 38.89 34.36 22.36
Punjab 40.49 31.07 30.83 26.41 41.31 32.38 18.78
Sindh 27.16 29.34 25.45 19.22 31.10 34.16 21.74
KP 44.09 41.23 38.38 37.64 46.64 41.75 27.57
Balochistan 26.42 34.89 39.64 34.38 28.19 37.36 50.74
Urban Areas 26.51 24.97 19.97 20.69 21.00 22.32 13.74
Punjab 28.29 27.72 21.51 21.99 24.39 23.13 12.51
Sindh 22.94 20.29 17.23 17.27 14.24 19.20 11.88
KP 32.14 23.08 22.66 22.41 26.75 29.44 23.98
Balochistan 21.99 35.56 21.04 23.24 24.94 26.75 32.18
Rural Areas 42.51 35.18 35.73 35.47 46.33 39.27 26.73
Punjab 45.51 32.33 34.52 32.62 48.17 36.13 21.76
Sindh 31.17 36.53 32.93 36.51 44.02 43.73 31.38
Khyber
Pakhtunkhwa
46.49 43.76 41.10 42.82 50.06 43.87 28.24
Balochistan 27.48 34.79 42.00 40.42 28.67 39.59 56.48
Source: Authors’ calculations
Figure 4‐4: Regional Poverty headcount in Pakistan, 1990-91 to 2005-06
302

The provinces present a stark picture of regional disparities as they differ enormously in
their poverty levels and trends as compared with the overall poverty. Table 4.5 and
Figure 4.5 show that the extent of urban poverty greatly varies within the country.
Whereas the prosperity of urban Sindh and Punjab, consisting of the metropolitan areas
of Karachi, Lahore, Faisalabad, Multan, Gujranwala and Rawalpindi, is well known,
Figure 4.5 displays that lowest percentage of urban poor live in Sindh followed by
Punjab. Moreover, urban poverty levels consistently fell in both provinces in the study
period. However, Balochistan province represents area of concentrated poverty since the
proportion of urban population living below the poverty line has generally increased. By
contrast, the percentage of population living below the poverty line in Khyber
Pakhtunkhwa province presents a mixed picture.
Poverty in rural Paksitan has always remained a critical factor in the overall incidence
and variation in poverty. While the extent of rural poverty levels tend to be worst as
compared with urban poverty levels, the proportion of rural population below the
poverty line has widely fluctuated during the study period (see Figure 4.6). Despite the
importance of rural population
303

Figure 4‐5: Poverty headcount in urban Pakistan, 1990-91 to 2005-06
Figure 4‐6: Poverty headcount in rural Pakistan, 1990-91 to 2005-06
in trigerring additonal demand for industrial goods produced in the country, the
importance of rual poverty has not very well understood. Whereas the acute deprivation
of KP is quite evident from Figure 4.6, the fluctuating fortunes for rural population
304

generally indicate the seasonal nature of poverty in rural areas where opportunities for
non-farm employment are few and where poor often live in isolated areas in evey sense.
Most of these regions have promising natural resource endowments, but they are
deprived in access to social services including education, health, road and transport
infrasture and market centers.
4.3.2.2 Evidence on regional income inequality
Inequality of income or expenditure is a broader concept than poverty and is defined
over the entire population rather than the population below a certain benchmark, i.e.,
poverty line. Trends in income inequality during the last four decades have been
analyzed by many previous studies. 94 In general, these studies show that income
distribution in Pakistan has improved in the sixties and eighties but has worsened in the
seventies and nineties [GoP (2000), Anwar (2005)]. However, it is important to note
that these studies are based on grouped data, which does not take into account
inequality within groups. It also appears that a similar comparison of inequality during
the last one decade cannot be made due to data limitations. For instance, GoP (2006)
notes that the “aggregative nature of income data collected in PSLM 2004-05 is strictly
not comparable with the corresponding data collected in PIHS 2000-01.” Hence to
make inequality comparisons of the recent period with the previous periods, one must
rely on the consumption based inequality. However, a consistent series of consumption
based inequality for the last two decades has not been worked out by any previous study.
The goal of this section is to document changes in national and regional inequality for
the period 1990-91 to 2005-06. We focus on per capita household consumption
expenditures to estimate inequality.
While various summary measures are used to describe inequality, Gini coefficient is the
most widely used single measure of inequality. The Gini coefficient is based on the
Lorenz curve, which is a cumulative frequency curve comparing a specific distribution
with a uniform distribution representing equality. In other words, the Gini coefficient is
94 For a review of income inequality literature, see Anwar (2005).
305

a ratio equal to twice the area between the Lorenz curve and the equality line and it can
be given by:
1
G = y − y
µ N( N −1) ∑∑
i. j j
i j
where µ is for mean income of the distribution, N is for the number of observations
and yi and y j represent income of ith and jth groups. The value of Gini ranges from 0
(perfect equality) to 1 (perfect inequality), but for per capita expenditure based
inequality it typically remains in the range of 0.2 to 0.5 and fulfills many desirable
properties, e.g., mean independence, population size independence, symmetry and
Pigou-Dalton transfer principle. A problem with this measure, however, is that it cannot
be easily decomposed to indicate the sources of inequality. Moreover, Gini coefficient is
most sensitive to changes in the middle of the distribution and ignores changes in the
distribution taking place at the tails. For the same reason, Gini estimates are often
supplemented by quintile income shares computed to analyze inequality trends at the
extremes.
4.3.2.3 Regional inequality trends in Pakistan
For this study we use household data (rather than grouped data) from seven rounds of
HIES from 1990-91 to 2005-06 to calculate Gini expenditure inequality measure
described above. The time profile of Gini inequality measure across rural and urban
areas can be envisioned in Table 4.6 and Figure 4.7 where the progress made through
the periods shows two types of patterns. One type displays that in the beginning of the
period national inequality followed a fairly continuous and smooth upward trend
showing worsening income distribution. This trend was generally corroborated by other
studies [Anwar (2005), Jamal (2009)]. And the other pattern displays a reversal of this
trend after 1998-99 when Gini coefficient sharply decreased from 34.27 in 1998-99 to
30.41 and 30.55 in 2000-01 and 2005-06, respectively.
306

The quintile shares in Table 4.6 are remarkable which indicate marginal decline in the
share of poorest 20% from 1990-91 to 1996-97 with a considerable decline in 1998-99
but still with a larger share at the end as it was at the beginning of the period. The
middle 40% of the population eroded its income share from 1990-91 to 1998-99 before
its recovery at the end of the period. The
Table 4‐6: Changes in Gini inequality in Pakistan and urban and rural areas
Gini index a
1990-
91
1992-
93
1993-
94
1996-
97
1998-
99
2000-
01
2005-
06
Urban 32.39 35.97 34.00 33.74 39.18 35.16 33.30
Rural 26.71 28.73 29.34 35.12 26.23 24.79 25.41
Overall 29.79 32.11 32.49 33.89 34.27 30.41 30.55
Quintile share b
Q1 8.5% 8.4% 8.2% 8.4% 7.7% 8.5% 8.4%
Q2 12.0% 11.5% 11.3% 11.5% 11.0% 11.9% 11.8%
Q3 15.2% 14.6% 14.3% 14.4% 14.3% 15.1% 15.1%
Q4 20.1% 19.3% 19.1% 18.7% 19.4% 20.0% 20.3%
Q5 44.2% 46.3% 47.1% 47.0% 47.6% 44.5% 44.4%
Ratio of richest
20% to poorest
40%,
(Q5/(Q1+Q2)
2.15 2.32 2.41 2.36 2.55 2.18 2.20
Note: Authors’ calculations based on household level micro data
a Gini-coefficient× 100.
b Percentage of the total.
major beneficiaries, however, were the richest 20% of the population that made
significant gains in their income shares at a time when all other income groups (i.e.,
80% of the population) were losing their income shares. Between 1990-91 and 1998-99,
the ratio of richest 20% to poorest 40% of the population declined from 2.15 to 2.55.
These results are corroborated by Anwar (2005) who noted that the period of 1990s
307

egistered slightly increasing magnitude of inequality in Pakistan because the
distributional changes allowed richest income groups to increase their relative income
shares in 1990s, which were highest in more than 50 years.
Figure 4‐7: Changes in inequality in Pakistan and by regions
We note that urban and rural areas differ in the evolution of inequality. The results
suggest that inequality in urban areas of Pakistan was much higher than rural areas.
This is not surprising in a country like Pakistan where due to presence of extremely rich
and extremely poor households in the urban areas, incomes and wages are much more
diversified. By contrast, most households in rural areas possess uniform skill set and the
bulk of the landowners are either small or very small. It leaves incomes to be relatively
more evenly distributed. Whereas inequality in urban areas always remained higher
than rural areas, it increased sharply in the first two years, fell from 1992-93 to 1996-97,
increased again in 1998-99 and then fell again in 2000-01 and 2005-06. In the rural
areas, however, inequality consistently increased from 1990-91 to 1996-97 and then
308

sharply declined over the 1996-97 and 2005-06 period. The increase in inequality in
urban areas especially after 1996-97 may be attributed to the implementation of the
Structural Adjustment Program in Pakistan. The decline in national inequality after
1996-97 is explained by a sharp decline in inequality in the rural areas and a marginal
decline in inequality in urban areas.
The time profile of the measures of inequality for the four provinces can be visualized in
Table 4.7 Figure 4.8 where the results based on 1990-91 survey should be interpreted
with caution since this survey was not representative at the provincial level. In the net,
we find that inequality levels in Punjab and Sindh are much higher than KP and
Balochistan. Like the national level trend, inequality in the provinces follows two
patterns: it depicts an increasing trend from 1990-91 to 1996-97 and a decreasing trend
from 1996-97 to 2005-06.
Table 4‐7: Changes in Gini inequality by provinces
Year Punjab Sindh Khyber
Pakhtunkhwa
Balochistan
1990-91 29.70 31.85 23.76 24.86
1992-93 32.61 33.60 27.22 24.83
1993-94 33.38 33.57 24.83 27.81
1996-97 34.78 33.20 28.59 29.01
1998-99 34.78 36.61 28.45 23.33
2001-02 30.04 35.18 23.33 22.09
2005-06 30.39 33.09 25.86 23.54
Note: Authors’ calculations based on household level micro data
Gini-coefficient× 100.
Figure 4‐8: Changes in inequality in the four provinces
309

4.3.3 Evidence on Spatial Disparities in Human Capital Infrastructure
In the industrial organization literature, the emphasis of recent studies on human
capital accumulation as key to firm survival makes sense [Mata and Portugal (2002),
Karlsson (1997)]. However, social spending in Pakistan has always suffered due to
competition with other development heads and stagnant revenue generation leaving
little capacity in the hands of the policy makers to meet the increasing backlog of
investment in education and health infrastructure in the country. A vast empirical
literature suggests that investment on human capital infrastructure alleviates poverty
[e.g., Klasen (2008), Gustafsson and Shi (2004)] while few others confirm that “the
poorest quintile benefits enormously from growth and from high average levels of
education and physical capital accumulation” [Birdsall and Londono (1997)]. While
spatial deprivation and rank ordering of districts in Pakistan has been attempted before
[see, among others, Jamal (2003), Wasti and Siddiqui (2008)], but their adverse impact
on industrial development and poverty alleviation has not been studied before.
310

We apply the principal components method to select significant principal components
to be used as education and health indexes representative at the district level. The
resulting principal components would be used as mapping measures of regional
concentration of education and health infrastructure, rank each district by making
comparison over time and to use them for further estimation in subsequent chapters.
We begin by outlining the principal component technique, describe the data on
education and health (E&H) indicators, present and discuss the results of this exercise. 95
4.3.3.1 Selecting significant principal components from E&H Indicators
We use a multivariate statistical weighting approach known as the method of principal
components (PC), which transforms original variables into linear combinations that
explain as much variation as possible with no loss of relevant information [Manly
(1994), Sharma (1996), Greene (1997)]. The weights in the principal component method
are dictated by the data in an objective manner. The first principal component is the
linear combination accounting for the largest variance, the second principal component
accounts for the greatest amount of the remaining variance, and so on. Only small
number of principal components from large number of variables is chosen that
contribute highly in explaining the variance. Components associated with smallest
eigenvalues are discarded because they are least informative. We adopt Kaiser-Gutman
Rule whereby only principal components associated with eigenvalues greater than one
are retained. The retained principal components are uncorrelated with each other,
which is useful in principal component regression analysis.
We collect district level data of education and health indicators from various reports of
Provincial Development Statistics published by the Bureaus of Statistics of the
Provincial Governments of Punjab, Sindh, Khyber Pakhtunkhwa and Balochistan. The
data on 26 education and health indicators was collected for 56 districts of Pakistan
from 1991–2006. Table 4.8 shows the 20 indicators chosen for inclusion in the analysis
along with their definition. The first five variables capture the size of school and college
95 This sub-section draws heavily from Burki (2010).
311

system in each district. Since these institutions are not homogenous in size and vary by
their enrolment levels, we multiply the number of schools/colleges by their respective
enrolments and weigh them by district area and population. Here we assume that public
spending is positively associated with size of these institutions. The next five variables
represent quality of the school system captured by teacher-student ratios in respective
districts.
We use teacher-student ratios (instead of student-teacher ratios) because public
spending on teachers is also positively associated with these ratios just like the size of
school and college system while student-teacher ratio moves in the opposite direction.
Public spending on hospitals also varies by their size, measured by the number of
hospital beds. Therefore, we construct hospital size variable by multiplying number of
hospitals with beds in hospitals weighted by area and population. The rest of the health
indicators are used as numbers and weighted only by population because their size and
coverage is based on a uniform government policy for rural localities. For each of the
chosen indicators we had 896 observations from 56 districts for 16 years.
Table 4‐8: District level variables used in principal component analysis
Explanatory variable Definition
1 Primary school size Number of primary schools× enrollment in
primary schools/(district area× district
population)
2 Middle school size Number of middle schools× enrollment in
middle schools/(district area× district
population)
3 High school size Number of high schools× enrollment in high
schools/(district area× district population)
4 Intermediate college size Number of intermediate colleges× enrollment in
intermediate colleges/(district area× district
population)
5 Degree college size Number of degree colleges× enrollment in
degree colleges/(district area× district
population)
6 Teacher student ratio in primary schools Teaching staff in primary schools/enrollment in
primary schools
7 Teacher student ratio in middle schools Teaching staff in middle schools/enrollment in
middle schools
8 Teacher student ratio in high schools Teaching staff in high schools/enrollment in
high schools
9 Teacher student ratio in intermediate Teaching staff in intermediate
312

colleges colleges/enrollment in intermediate colleges
10 Teacher student ratio in degree colleges Teaching staff in degree colleges/enrollment in
degree colleges
11 Hospital size Number of hospitals× beds in hospitals/(district
area× district population)
12 Dispensaries per 10,000 population Number of dispensaries/district population
13 Dispensary beds per 10,000 population Beds in dispensaries/district population
14 Rural health centers per 10,000 population Number of rural health centers/district
15 Rural health center beds per 10,000
population
population
Beds in rural health centers/district population
16 T.B. clinics per 10,000 population Number of TB clinics/district population
17 T.B. clinic beds per 10,000 population Beds in TB clinics/district population
18 Basic health units per 10,000 population Number of basic health units/district population
19 Basic health units beds per 10,000
population
20 Mother-child care centers per 10,000
population
Beds in BHUs/district population
Number of mother-child care centers/district
population
The principal components were obtained by using the FACTOR command in Stata
software version 10. After varimax rotation, six principal components were retained by
using Kaiser eigenvalue criterion96 which accounts for 74.47% of total variance. The six
retained principal components along with rotated factor loadings derived from the
correlation matrix of the twenty education and health variables are presented in Table
4.9. Numbers in bold indicate high positive coefficient. It needs to be clarified that the
order in which the six principal components are listed in the table only reflects the order
in which they were extracted and it does not necessarily indicate their relative
importance in explaining industrial development and the state of poverty.
Table 4‐9: Principal component rotated factor loadings
Explanatory variable PC1 PC2 PC3 PC4 PC5 PC6
Primary school size -0.180 0.060 0.047 -0.024 0.896 0.032
Middle school size 0.030 0.512 -
0.230
-0.599 0.368 -
0.042
High school size 0.013 0.817 -0.184 -0.222 0.171 -
0.052
96 Kaiser criterion suggests to retain those factors with eigenvalues equal to or higher than 1.
313

Intermediate college size -0.047 0.667 0.088 0.056 0.239 0.048
Degree college size -0.159 0.850 0.046 0.074 -0.115 0.012
Teacher student ratio in primary
schools
Teacher student ratio in middle
schools
Teacher student ratio in high
schools
Teacher student ratio in
intermediate colleges
Teacher student ratio in degree
colleges
0.046 0.103 -
0.272
0.215 -0.041 -
0.052
0.677 -0.191 0.145
0.824 -
0.060
-
0.061
0.153 -0.119 0.827 0.0186 0.039 -
0.013
0.002 -
0.008
0.279 -
0.086
-
0.274
-0.199 -0.378 -
0.137
0.774 -0.076 -
0.024
-
0.033
Hospital size -0.055 0.823 0.088 -0.069 -0.059 -
0.043
Dispensaries per 10,000 population 0.910 0.023 0.229 0.135 -0.057 -
0.048
Dispensary beds per 10,000
population
Rural health centers per 10,000
population
Rural health center beds per 10,000
population
0.567 0.015 0.369 -0.401 -0.163 0.148
0.909 -0.169 0.185 0.140 -0.015 -
0.007
0.789 -0.182 -
0.268
T.B. clinics per 10,000 population 0.265 -0.105 -
0.099
T.B. clinic beds per 10,000
population
Basic health units per 10,000
population
Basic health units beds per 10,000
population
Mother-child care centers per
10,000 population
Proportion (relative weight of each
factor in the total variance)
-0.011 -
0.036
0.093 -0.091 0.049
0.784 0.265 -
0.108
0.021 -0.039 0.029 0.961
0.934 -0.156 0.120 0.063 -0.035 0.012
-0.181 -0.210 -
0.784
Note: Numbers in bold face indicate high positive loadings
0.171 -
0.049
-
0.104
0.841 0.314 0.185 0.042 -0.053 -
0.039
22.47 15.27 12.55 12.32 6.71 5.15
The first factor (PC1) accounts for 22.47% weight in the total variance explained in the
data. PC1 represents principal component basic health, dispensary & rural health index
as it is characterized by high positive loadings on basic health units (0.934),
dispensaries (0.910), rural health centers (0.909), mother child care center (0.841),
314

ural health center beds (0.789) and dispensary beds (0.567). PC1 has high correlations
with all health variables, except hospitals, TB clinics and beds in basic health units.
The second factor (PC2) was characterized by high positive factor loadings on degree
college size (0.850), hospital size (0.823), high school size (0.817), intermediate college
size (0.667) and middle school size (0.512). It explains 15.27% of the total variation and
represents principal component post-primary school system and hospital index.
The third factor (PC3) represents principal component teachers in high school and
college index as it is characterized by high factor loadings in teacher student ratio in
high schools (0.827) and degree colleges (0.774) and accounts for a further 12.55% of
the total variance. Similarly, the fourth factor (PC4) represents principal component
quality of middle and primary schools with high factor loadings in teacher student ratio
in middle schools (0.824) and teacher student ratio in primary schools (0.677) and
accounts for 12.32% of the total variance. The fifth factor represents principal
component primary school index as it accounts for 6.71% of the total variance and is
characterized by high factor loading in primary school size (0.986). Finally, the sixth
factor represents principal component beds in TB clinic index with high factor loading
(0.961) and accounts for 5.15% of the total variance.
4.3.3.2 Most and least developed districts by social infrastructure
Of all the factors, factor 2 or principal component post-primary school system and
hospital index stands out as the most important measure in terms of coverage and share
in government spending on education and health sectors. Therefore, in the subsequent
analysis we use principal component post-primary school system and hospital index for
mapping of districts by their level of development.
Table 4.10 lists the 20 most and least developed districts for selected years on the basis
of principal component post-primary school system and hospital index. Note that the
most developed districts
315

Table 4‐10: Most and least developed districts based on rankings from principal component post‐primary school system
and hospital index
Districts PC
value
, 90-
91
PC
value,
93-94
PC
value,
97-98
PC
value,
00-01
20 most developed districts (from most developed to least developed districts)
Lahore 21.19 Lahore 25.54
1
Karachi 13.76 Karachi 12.78
9
Rawalpindi 3.88 Rawalpindi 4.528 Rawalpin
di
Peshawar 2.11 Hyderabad 3.270 Hyderaba
d
Lahore 14.85 Lahore 15.00
4
PC
value,
05-06
Lahore 20.23
2
Karachi 10.59 Karachi 6.485 Karachi 8.552
3.57 Rawalpin
di
4.883 Rawalpin
di
2.47 Peshawar 1.717 Faisalaba
d
7.420
Abbottabad 1.39 Sibi 1.695 Peshawar 1.494 Faisalaba
d
0.951 T.T singh 1.511
Quetta 1.37 Peshawar 1.521 Sibi 1.452 Sibi 0.505 Sibi 1.201
Sibi 1.36 Faisalabad 1.263 Faisalaba
d
0.976 Hyderaba
d
Jhelum 1.25 Jhelum 1.134 Jhelum 0.423 Abbottaba
d
2.105
0.325 Peshawar 0.700
0.286 Abbottaba
d
D I Khan 1.19 Multan 0.818 Multan 0.377 Sargodha 0.284 Gujrat 0.513
Multan 1.18 Sialkot 0.789 Thatta 0.230 Quetta 0.143 Jhelum 0.503
Faisalabad 1.17 Quetta 0.247 Quetta 0.216 TT Singh 0.033 Multan 0.405
Hyderabad 1.02 Gujranwala 0.223 Sargodha - Jhelum - Quetta 0.354
0.006
0.133
Bannu 0.63 Gujrat 0.182 Attock - Multan - Sargodha 0.276
3
0.158
0.134
Kohat 0.30 Abbottabad 0.055 Sialkot - Sanghar - Sialkot 0.245
3
0.184
0.188
Gujrat 0.179 Sargodha - Abbottaba - Sialkot -0.191 Gujranwal 0.056
0.105 d
0.201
a
Sialkot 0.158 Attock - TT Singh - Gujranwal - Khanewal -
0.156
0.223 a
0.235
0.207
Larkana 0.143 Swat - Chakwal - Khanewal - Jhang -
0.208
0.271
0.285
0.427
Gujranwala 0.10 Shikarpur - Khanewal - Gujrat - Attock -
6
0.242
0.281
0.402
0.537
Khanewal - Chakwal - Gujrat - Bannu - Hyderaba -
0.02
7
0.246
0.541
0.558 d
0.554
Chakwal - Bahawalnag - Tharparke - Layyah - Bannu -
0.09
7
ar
0.312 r
0.557
0.664
0.633
20 least developed districts (from most developed to least developed districts)
Bahawalpu - RY Khan - Badin - Shikarpur - Mardan -1.373
r
0.63
9
0.729
0.990
1.234
Kalat -
0.66
Mekran - Kasur -1.011 Larkana - Larkana -
0.583
316

Okara -
0.66
4
Karak -
0.73
3
Muzafargar
h
3 0.749 1.278 1.438
-
0.82
8
Layyah -
0.83
5
Dir -
0.85
8
Khushab -
0.89
6
Mianwali -
0.90
8
Dadu -
0.96
9
Bhakkar -
0.98
3
Mansehra -
1.001
Vehari -
1.005
Sahiwal -
1.06
0
Mekran -
1.101
Jacobabad -
1.128
Sukkur -
1.364
Tharparker -
1.490
Khairpur -
1.545
Nawabshah -
1.580
Muzafargar
h
-
0.817
Tharparker -
0.863
Okara -
0.871
Kohat -
0.882
Sahiwal -
0.907
D I Khan -
0.967
Sanghar -
1.060
Bhakkar -
1.082
Vehari -
1.098
Layyah -1.013 Bahawaln
g
Jhang -
1.038
D.G khan -
1.052
Okara -
1.056
Tharparke
r
Jacobaba
d
-
1.288
-
1.294
Sanghar -
1.439
Dir -
1.558
-1.311 DG Khan -
1.585
Dadu -1.327 Bhakkar -1.595
Mardan -1.061 DG Khan -1.367 Shikarpur -
1.650
Karak -1.077 Sukkur -1.376 RY Khan -1.719
Sukkur -1.117 RY Khan -1.379 Swat -1.726
R.Y. Khan -1.133 Bahawalp
ur
Rajanpur -
1.239
Mardan -1.130 Khairpur -
1.378
Mansehra -1.156 Jacobaba
d
-
1.407
Karak -
1.435
Swat -
1.500
Jacobabad -1.170 Bhakkar -1.414 Badin -
1.534
Karak -
1.243
Dadu -
1.387
Swat -1.418 Dir -
1.645
Vehari -
1.420
Dir -1.481 Mansehra -
1.449
Sukkur -
1.506
Nawabshah -
1.968
Khairpur -
2.070
Muzafarg
arh
Nawabsha
h
-
1.507
-
1.640
-1.414 Tharparke
r
Muzafaga
rh
-
1.730
-1.857
Dadu -
1.863
Bhakkar -1.519 Mansehra -
1.869
Muzafarg
arh
D I Khan -
1.833
Mansehra -
1.846
Khairpur -
1.982
Dir -1.721 Nawabsha
h
Rajanpur -
1.953
Sukkur -
1.993
-1.671 Jacobaba
d
-
2.004
-
2.086
Thatta -2.131
Khairpur -
2.198
Badin -
2.320
Nawabsha
h
-
2.669
Note: Before 1997-98, data on schools and colleges run by the Armed Forces, Railways,
WAPDA and other such institutions and government corporations was also reported in the
Provincial Development Statistics along with other schools and colleges in each district.
However, this practice was discontinued in 1997-98. To construct a consistent time-series, we
wanted to collect data of government run schools and colleges also for the period before 1997-
98, but this information was separately not available from any published source. A visible
drop in the principal component values in 1997-98 is primarily explained by this inconsistent
317

eporting of data. However, district rankings remain more or less unaffected.
include eight metropolitan cities and districts in Northern and Central Punjab. Figure
4.9 displays complete mapping of districts on the basis of the index for 2005-06 along
with development rank (in parenthesis) for selected districts. The three categories
include the most-developed districts, i.e., where the index is 1 standard deviation (SD)
above the mean, least-developed districts, i.e., where the index is 1 SD below the mean,
and the medium-developed districts that consist of all other districts. Again, we can
clearly see that investment in social infrastructure is highly concentrated in
metropolitan cities, big cities and their surrounding districts while districts located away
from these urban demand centers (e.g., southern Punjab, interior of Sindh and remotely
located districts in Khyber Pakhtunkhwa and Balochistan) are lagging behind. As
predicted by the literature on New Economic Geography, only the better ranked districts
(that have a pool of skilled workers) are concentrated by large-scale manufacturing
clusters (see Figure 4.10 based on plant level data from CMI 2005-06).
However, in least developed districts industry composition is determined by their
natural advantage in resource based industry, e.g., agro-based industries, handicrafts
and resource-based randomly scattered plants such as petroleum plants, sugar mills,
ginning factories, marble & minerals, and the like. Figure 4.11 displays that most
resource-based industry clusters are located in districts where the principal component
post-primary school system and hospital index (given in parenthesis) is at the lower end
of the distribution.
It is interesting to note that our human capital infrastructure based rankings of districts
closely overlap with two other recent studies (e.g., the development rank ordering of
districts by Wasti and Siddiqui (2008) and district-wise index of multiple deprivations
by Jamal et al. (2003). If any thing, these evidences help in understanding the
relationship between infrastructure development and growth and in determining the
areas of need.
318

Figure 4‐9: Location of districts by education and hospital index
319

Figure 4‐10: Industry clusters and development ranking of districts, 2005-06
Figure 4‐11: Resource based clusters and development ranking of districts, 2005-06
320

The mapping of districts indicates that the human capital infrastructure gap between
the most developed and least developed districts was increasing with the passage of time
(see Table 4.10). These results suggest that if balanced development is the objective of
government policy then the decision-makers may want to adopt a policy of
“geographical targeting” whereby the development funds are disproportionately
allocated to least developed areas on the basis of non-income poverty mapping of
districts. This view has been corroborated by Jamal et al. (2003) who has previously
noted that due to widening spatial deprivation of Pakistan, geographical targeting of
scarce funds in least developed areas may be a viable option for poverty alleviation.
321

4.3.4 Market Access and Spatial Inequality in Road Infrastructure
4.3.4.1 Introduction
Market access is determined by the ease of connectivity with the market centers in
spatial vicinity of the firm, which in turn depends on the availability of good road
infrastructure, firm’s distance from the market, size of the market and the availability of
quality transport networks. Absence of all or some of these factors would limit the extent
of the market for a firm because the firm would be unable to connect to a wider market
area, i.e., other cities and districts, other provinces or the rest of the world. Therefore,
spatial inequality in road infrastructure would constrain market efficiency and promote
market failures by creating factor scarcities and distorting factor prices, which in turn
would prevent these spatial units to specialize in production by their comparative
advantage or dynamic comparative advantage. Moreover, lack of market access would
also results in poor supply of goods and services and higher prices. A combination of
these factors may lead to a vicious circle of chronic and persistent poverty in deprived
regions that in turn may reproduce patterns of regional and spatial inequality. The
recent literature also suggests that improvements in roads at the regional level can
significantly contribute to the pursuit of socially inclusive growth [e.g., Khandker et al.
(2009), Jacoby and Minten (2009)]
Even though there are popular concerns about the sufferings of regional and spatial
units, there is little or no systematic documentation of what has happened to spatial
development of road infrastructure in Pakistan over the last two decades. Empirical
evidence on the importance of road infrastructure for spatial and regional inequality in
the country does not exist. Our objective here is to document spatial concentration and
the changes taking place in the road infrastructure from 1990-91 to 2005-06. Much of
this material is summarized from Burki (2010). In the subsequent chapter, we also
relate spatial concentration of road network to variation in rural poverty and industry
location.
322

4.3.4.2 Evidence of spatial inequality in road infrastructure
A consistent time-series data on road density at the district level is not available from
any published source. Part of the problem is that besides armed forces and state-owned
corporations, more than a dozen government institutions at the federal, provincial,
district and municipal government level, are involved in construction and maintenance
of road infrastructure in the country that makes the data collection exercise extremely
cumbersome. While Provincial Development Statistics occasionally report district level
data on road density, concerns on its reliability and consistency make it suspect for all
practical purposes. Fortunately, the Punjab Highway Department maintains a consistent
district time-series data on national highway roads, farm to market roads and district
government roads. However, they do not cover the road network maintained by
cantonment boards and defense housing authorities mostly located in Karachi, Lahore
and Islamabad. We obtain district level road density data of the districts of Punjab from
the Punjab Highway Department for the period 1992 to 2006. Since we are concerned
about the impact of road density on poverty and firm location, we assume that the
omission of data on cantonment and DHA roads would not directly affect poverty
incidence of rural people and firm location in other districts.
The most striking pattern that emerges from the data is that while spatial inequality in
road density has been large, it has widely fluctuated over time. Figure 4.12 displays
mapping of the districts of Punjab on the basis of road density for 2005-06 as well as
their relative rank from most dense to least dense. It appears from there that districts in
Southern Punjab are most deprived in road density while districts in Northern Punjab
are the ones with highest road density.
That road density has fluctuated over time can be observed from Figure 4.13, which
plots road density of 35 districts relative to the road density of Lahore in 1992-93 and in
2005-06. Here we consider road density of each district divided by the road density of
Lahore district. We have chosen Lahore district because it had the highest road density
in both 1992-93 and
323

Figure 4‐12: Spatial inequality in road density in Punjab, 2005‐06
2005-06 due to which the units of relative road density can be readily interpreted.
Changes in road density across districts are illustrated by departures from the 45-degree
line. The share of Lahore in road density relative to Lahore is 100%. Districts above the
45-degree line experienced improvement over time while those below the line
experienced decline in their relative shares. Some Southern districts, e.g., Rajanpur,
D.G. Khan, Bhakkar, Layyah and Rahimyar Khan, with relative road density of less than
40% of Lahore district in 1992-93 experienced no change in their relative road density
by 2005-06. Sargodha, Faisalabad and Rawalpindi districts improved their relative
shares from around 70% of Lahore to more than
Figure 4‐13: Relative road density of the districts of Punjab with Lahore district, 1992‐93 vs. 2005‐06
324

Relative road density, 2005-06
.2 .4 .6 .8 1
BHP RJN
Relative road density in Punjab, 1992-93 vs. 2005-06
MNW
DGK
MZF
BKR
LYH
JHG
BHW
RYK
ATK
CHK
KHB
JHL
SHP
SRG
FSD
RWP
KHW
GJT
TTS
KSR
GJW
VHR OKR
SHW
MLT
.2 .4 .6 .8 1
Relative road density, 1992-93
80% mainly due to construction of 400 km motorway and other ancillary roads. Jhang,
Chakwal, Muzaffargarh and Mianwali districts are other noticeable exceptions where
relative road density has substantially increased during this 13-year period. Sialkot,
Gujranwala, Sahiwal, Multan and Gujrat districts are some of the districts that have
experienced a large decline in their relative road density. Sialkot had the second highest
road density after Lahore in 1992-93 with road density equal to 90% of Lahore. But, by
2005-06, Sialkot’s road density had fallen to below 50% of Lahore.
Similarly, in Khyber Pakhtunkhwa province the large and fluctuating spatial disparity
across districts further reveals this trend. Figure 4.14 displays mapping of districts and
their relative rank in Khyber Paktunkhwa province for which the data for selected years
was obtained from the Provincial Development Statistics. The spatial inequality in road
density across districts
Figure 4‐14: Spatial inequality in road density in Khyber Pakhtunkhwa, 2005‐06
SIA
LHR
325

indicates that Peshawar, Abbottabad, Bannu and Kohat districts have highest road
density. In Figure 4.15, we plot relative road density of all the districts in 1993-94 and
2005-06 with Peshawar district. We note that Mardan, Karak, Mansehra and Dir have
below 40% of the road density of Peshawar. On the other hand, Kohat and Bannu
significantly increased their relative share in road density between 1992-93 and 2005-
06, but Abbottabad significantly lost its share from more than 100% of Peshawar in
1993-94 to around 70% level in 2005-06.
How spatial disparities in road infrastructure influence firm location and poverty?
Based on the evidence presented above, it can be argued that spatial inequality in road
density may be responsible for higher poverty and lower industrial concentration in
Southern districts in
326

Figure 4‐15: Relative road density of the districts of KP with Peshawar district, 1993-94 vs.
2005-06
Relative road density in Khyber Pakhtunkhwa, 1993-94 vs. 2005-06
Relative road density, 2005-06
.2 .4 .6 .8 1 1.2
MRD
SWT DIK
DIR
MNS
KRK
BNU
KHT
.2 .4 .6 .8 1 1.2
Relative road density, 1993-94
Punjab and remote districts of KP because lack of connectivity of these districts with the
demand centers may be promoting market failures in factor and product markets
leading to higher incidence of poverty and low industrial concentration. We relate
spatial and inter-temporal variation in road density to variation in rural poverty and
industrial concentration in the next chapter.
4.3.5 Agglomeration of Manufacturing Industries in Pakistan
4.3.5.1 Introduction
Since Krugman’s (1991b) seminal paper published over two decades ago, significant
progress has been made in theoretical models on the New Economic Geography, but
empirical studies, especially from the developing countries are limited. Our aim in this
section is to provide mapping measures of industrial concentration by exploring
whether manufacturing industries in Pakistan are agglomerated, and if so, which ones.
PSH
ABT
327

Moreover, we also present evidence on how geographic concentration emerges from the
dynamic process over time and what is the nature of agglomeration economies.
However, we begin by displaying a map of spatial distribution of manufacturing activity.
4.3.5.2 Local spatial distribution of manufacturing activity
We present a map of spatial distribution of manufacturing activity in Pakistan. As
before, we concentrate on districts as spatial units. We measure spatial concentration of
manufacturing activity by taking district level employment shares of all the industries
from CMI 2005-06 plant level data. We classify districts into three categories guided by
the mean and the standard deviation (SD) of district employment shares. We regard a
district having high- and low-concentration if its employment share is one SD above and
below the mean, respectively. All others are classified as medium-concentrated districts.
The average employment share of districts is 0.018 with SD of 0.0319. Figure 4.16 shows
spatial employment shares for 2005-06 at the district level. Note that highly
concentrated districts are mostly clustered around metropolitan cities of Karachi and
Lahore. Even medium-concentrated districts are also clustered near two big cities. Only
exceptions are Muzaffargarh and Swat districts. Muzaffargarh is located at the centre of
the cotton growing belt in Punjab and due to natural advantage in cotton there is a large
concentration of cotton ginning and textile manufacturing plants. Moreover, there is
also a huge cluster of petroleum refining industry. Similarly, Swat is home to marble
and mineral industries due to its natural advantage. In addition, there is a large
concentration of pharmaceutical and plastic industries, which explain its large share.
We also examine concentration level and change in concentration in 2-digit industry in
Punjab by the Gini index for which more than one year data was available. Note the
highest raw concentration is found in other industry (PSIC 39), basic metal (PSIC 37),
paper products (PSIC 34) and wood product (PSIC 33). However, the concentration in
all these industries falls dramatically between 1995-96 and 2005-06, except in PSIC 34
where the concentration level further increases by 57%. However, Ellison and Glaser
(1997) has proposed a much better measure of geographic concentration of industry to
which we now turn to.
328

Figure 4‐16: District level employment shares in Pakistan’s manufacturing sector
329

Table 4‐11: Change in concentration by Gini index in Punajab
Gini index, Gini index,
PSIC Name of 2-digit industry
1995-96 2005-06 Change
31 Food, beverage & tobacco 0.0252308 0.0264939 5.01
32 Textile & leather 0.0157225 0.01127 -28.32
33 Wood & wood products 0.179632 0.1507073 -16.10
34 Paper products and printing 0.1968457 0.3089328 56.94
35 Petro/chemical products 0.0645884 0.0701416 8.60
36 Mineral products 0.1072151 0.1008915 -5.90
37 Basic metal industry 0.364066 0.2285998 -37.21
38 Metal products, etc. 0.1247627 0.1541017 23.52
39 Other industry & handicrafts 0.735229 0.599017 -18.53
The Gini index is given by gi M
2
= ∑ ( s ) ,
i 1 ij −x
=
i where i = 1, K, M for spatial units
(i.e., districts) and j is for industry.
4.3.5.3 Measuring geographic concentration by Ellison and Glaeser index
To measure the extent of geographic concentration we follow the method proposed by
Ellison and Glaeser (1997). Their index is based on a rigorous statistical model that
takes random distribution of plants across spatial units as a threshold to compare
observed geographic distribution of plants. Ellison and Glaeser (1997) assume that
plants make location decisions to gain from internal and external economies peculiar to
a particular location. Because the industrial structure in Pakistan consists of many
small, medium and large plants, proper weights are required to correct for the diverse
sizes of plants and this is taken care of in the Ellison and Glaeser (hereafter EG) index.
They present the following estimator to measure the agglomeration of industries
γ
j
=
s x
⎛
⎝
x
⎞
H
⎠
⎛ 2 ⎞
⎜1−∑xi ⎟(
1−H
j)
⎝ i ⎠
M
2
∑( − ) − 1−∑
ij i ⎜
2
i ⎟ j
i= 1
i
where s ij is the share of industry ' j s employment located in district i ; xi is the share of
industry’s overall manufacturing employment in district i ; ( ) 2
s x − is an index of
raw geographic concentration given by the sum of squared deviations of employment
∑
i
ij i
330

shares of the industry j known as Gini-coefficient;
H =∑ Z is a Herfindahl-style
j k
2
kj
measure of the industry j 's
plant level concentration of employment, where Zkj is the
kth plant’s share in industry j 's
employment.
In practice, the value of the EG index indicates the strength of agglomeration
externalities in an industry. Usually a γ score of more than 0.05 indicates highly
agglomerated industry; a score between 0.05 and 0.02 suggests moderate
agglomeration and a score of less than 0.02 shows randomly dispersed industry.
In Pakistan, the Census of Manufacturing Industries (CMI) is the only source of detailed
enterprise level data on firm characteristics and covers all registered manufacturing
firms employing over 10 workers in the country. This survey is conducted by the Federal
Bureau of Statistics, Government of Pakistan in collaboration with provincial statistical
bureaus. The survey is often conducted after five-year intervals. The present study uses
plant level data taken from the Census of Manufacturing Industries (CMI), 2005-06
provided by the Federal Bureau of Statistics, Government of Pakistan. The CMI provides
data for 2-digit, 3-digit and 5-digit classifications under the Pakistan Standard
Industrial Classification (PSIC) according to geographic subdivision at the district, the
province and the national levels.
Due to large changes in geographic concentration and high turnover of manufacturing
plants, it is important to locate variation in agglomeration of industries by using data of
more than one time period. Therefore, we also use data of three time periods to examine
how geographic concentrations emerge from the dynamic process. Because the plantlevel
CMI data of Sindh, Khyber Pakhtunkhwa and Balochistan provinces for 1995-96
and 2000-01 was not available from any source, we decided to use the CMI data of only
the Punjab province for 1995-96, 2000-01 and 2005-06 to present agglomeration of
industries overtime. This supplementary data was provided by the Punjab Bureau of
Statistics.
4.3.5.4 Agglomeration of manufacturing industries in Pakistan, 2005­06
331

Next, we discuss the geographic concentration of 3-digit industries performed at the
district level by the Ellison-Glaeser index. As suggested by Ellison and Glaeser (1997),
we take values higher than 0.05 as high concentration, values in the range of 0.02 to
0.05 show intermediate concentration and values lower than 0.02 represent low
concentration. Results of the EG index are also compared with raw geographic
concentrations known as Gini index and Herfindahl index.
From the results summarized below we find that agglomeration of the 3-digit
manufacturing industries is widespread in Pakistan while a small number of industries
fall in the category of low concentration industries. For example, Table 4.12 shows that
35.3% of the industries are highly agglomerated, 38.2% are moderately concentrated
and only 26.5% industries are not agglomerated. This is further corroborated in Figure
4.17 which plots the frequency distribution of the EG index across 3-digit industries,
where each bar is for the number of industries for which the index is in that interval.
The tallest bar is at the EG value of 0.03 and 0.04 and a large number of industries are in
the range that corresponds to highly concentrated industries. 97
Turning to specific 3-digit industries, we find that the most highly concentrated industry
is other manufacturing (PSIC 394) with the EG index of 1.04 and raw concentration of
0.97 indicating that the industry is located in only one district. This is not a surprising
result since this industry classification relates to ship-breaking industry, located only in
Gadani (Kalat division) of Balochistan. Sports and athletic goods (PSIC 392) is the
second most highly agglomerated industry with the EG index value of 0.90, where high
value for the Gini index (0.842) indicates that the industry is located in few districts (5
districts) and the low value for Herfindahl index (0.077) shows that the employment is
distributed across many plants. Located in Sialkot and its surrounding districts, the
driving force for concentration of sports and athletic goods industry is natural advantage
of specialized labor, inter-industry spillovers, local transfer of knowledge and market
access in export markets due to close ties with international sports brands.
97 The most highly concentrated industries indicate that they are more concentrated than would be
expected if the industries were randomly located across spatial units.
332

Table 4‐12: Agglomeration of 3‐digit manufacturing industries in Pakistan, 2005‐06
PSIC Industry Ran
k
Number
of
Plants
Number
of
District
s
Ellison-
Glaeser
index
Gini
coefficie
nt
Herfind
ahl
index
394 Other manufacturing
industries
1 30 1 1.043 0.965 0.035
392 Sports and athletics goods 2 51 5 0.901 0.842 0.078
332 Furniture and fixtures 3 34 8 0.231 0.268 0.076
385 Scientific instruments 4 95 7 0.193 0.218 0.052
350 Pharmaceutical industry 5 213 22 0.171 0.171 0.017
322 Wearing apparel 6 236 12 0.156 0.164 0.025
393 Handicrafts and office
supplies
7 43 12 0.151 0.177 0.048
342 Printing and publishing 8 43 5 0.141 0.271 0.176
361 Pottery and china products,
etc.
9 97 7 0.139 0.175 0.058
341 Paper and paper products 10 131 22 0.124 0.210 0.117
325 Ginning and bailing of fibers 11 540 27 0.099 0.096 0.004
383 Electrical machinery 12 240 19 0.068 0.125 0.072
372 Non-ferrous metals 13 41 7 0.047 0.108 0.073
380 Fabricated metal, cutlery and
aluminum products
14 75 15 0.044 0.092 0.058
382 Non-electrical machinery 15 206 25 0.042 0.075 0.041
354 Petroleum refining,
petroleum products and
coal
16 30 12 0.041 0.164 0.142
369 Other mineral products 17 311 32 0.039 0.059 0.026
352 Other chemical products 18 150 24 0.038 0.081 0.051
356 Plastic products 19 141 20 0.035 0.048 0.017
381 Copper and brass industrial
products
20 111 17 0.033 0.067 0.041
321 Made-up textiles, knitting
mills, carpets and rugs
21 261 23 0.030 0.049 0.024
323 Leather and leather products 22 227 22 0.029 0.052 0.028
311 Dairy products and processed
food
23 1190 55 0.026 0.033 0.010
384 Transport equipment 24 186 15 0.021 0.057 0.041
362 Glass and glass products 25 34 13 0.020 0.121 0.113
320 Spinning and weaving of
cotton & wool
26 1081 44 0.019 0.023 0.006
324 Footwear manufacturing 27 35 10 0.015 0.247 0.255
331 Wood and cork products 28 15 10 0.014 0.139 0.138
351 Industrial chemicals 29 111 21 0.003 0.029 0.028
371 Iron and steel industries 30 198 18 -0.005 0.426 0.463
314 Tobacco industry 31 13 6 -0.012 0.207 0.232
333

313 Beverage industry 32 36 16 -0.020 0.080 0.105
355 Rubber products 33 30 10 -0.030 0.161 0.198
312 Animal feed & ice factories 34 65 20 -0.035 0.067 0.104
Figure 4‐17: Distribution of 3-digit Ellison-Glaeser index
Other most concentrated industries represent sectors where it is critical for the industry
to spread out to reach to the final consumers or the suppliers, e.g., furniture and
fixtures, scientific instruments, pharmaceutical industry, wearing apparel, handicrafts
and office supplies, printing and publishing, pottery and china products, paper and
paper products, etc. On the other hand, the demand for least concentrated industries is
diversified across many districts due to which they have substantial raw concentration,
e.g., iron and steel (0.426), footwear (0.247) and tobacco (0.207), but their employment
is distributed across few large plants, e.g., iron and steel (0.463), footwear (0.255),
tobacco (0.232) and rubber (0.198).
334

4.3.5.5 Geographic concentration of industries over time
How industrial concentration has evolved overtime? We also study this question by
asking whether the most agglomerated industries in 2005-06 were also agglomerated in
previous years. To make this dynamic analysis possible we select three data points over
the 10-year period from 1995-96 to 2005-06 for which CMI data of Punjab was
available. The CMI 1996-97 and 2000-01 were conducted on the basis of same PSIC, but
the classification was changed before conducting the CMI 2005-06. To produce
consistent and comparable estimates overtime, we regrouped CMI 2005-06 data
according to CMI 1995-96 classification. As before, we use the 1981 district boundaries
to identify our spatial units and focus on 3-digit industries to measure industrial
concentration.
Table 4.13 demonstrates the on-going dynamic process in 3-digit industries in Punjab
where we can clearly notice declining agglomeration levels in the most concentrated
industries in 1995-96. Except for industry 381 and 383, the industrial concentration
generally fell between 1995-96 and 2005-06 when more than 50% of the most
concentrated industries experienced a sharp decline in their concentration index while
five of these industries experienced an increase in the first five years and a decline in the
next five years. Entry of more firms in an industry explains the steady decline in the
concentration index as we note that the industry specific Gini index of these industries
gradually falls during the study period. Of the fifteen least concentrated industries in
1995-96, in general, the moderate to high concentration industries again show a
consistent decline in their concentrations. Most other industries experienced a
moderate increase in their concentration index, but the magnitude of the increase in
least concentrated industries was much smaller than the decline in most concentrated
industries. Whereas dramatic movements in the concentration index are rare, there are
only two noticeable increases in the concentration index that are worth mentioning. We
notice a dramatic increase in the concentration index of tobacco and non-ferrous metal
industries between 1995-96 and 2000-01, which is the result of exit of few plants that
makes a dramatic impact on their Gini index as well as on the absolute value of the EG
index.
335

Table 4‐13: Geographic concentration of 3‐digit industries in Punjab, 1995‐96 – 2005‐06
PSIC Industry
Fifteen most concentrated industries in
1995-96
1995-96 2000-01 2005-06
392 Sports and athletics goods 1.047087 1.068879 0.913328
385 Scientific instruments 0.8610103 0.589197 0.280299
362 Glass and glass products 0.5038906 0.30704 0.191059
361 Pottery and china products, etc. 0.415917 0.327417 0.164882
371 Iron and steel industries 0.4151095 0.391224 0.276167
322 Wearing apparel 0.3793302 0.208751 0.077318
350 Pharmaceutical industry 0.3547366 0.351063 0.339121
384 Transport equipment 0.2308846 0.364937 0.145223
355 Rubber products 0.2160143 0.089129 -0.00429
380 Fabricated metal, cutlery, aluminum and products 0.2126588 0.108821 0.173101
321 Made-up textiles, knitting mills, carpets and rugs 0.1896874 0.214717 0.037884
381 Copper and brass industrial products 0.169055 0.184687 0.249403
331 Wood and cork products 0.1624373 0.164167 0.045256
383 Electrical machinery 0.1582761 0.193224 0.20588
332 Furniture and fixtures
Fifteen least concentrated industries in 1995-
96
0.1406263 0.229663 -0.15062
324 Footwear manufacturing -0.1127579 -0.17016 -0.06546
312 Animal feed & ice factories -0.106044 3.36E-05 -0.07352
372 Non-ferrous metals -
0.0379526
0.216773 -0.00275
354 Petroleum refining, petroleum products and coal -0.0311225 0.054006 0.192322
313 Beverage industry - -0.00215 -0.02943
0.0057071
314 Tobacco industry 0.0068856 0.4708 0.064257
351 Industrial chemicals 0.0151802 0.010317 -0.00265
311 Dairy products and processed food 0.0179265 0.04199 0.021544
356 Plastic products 0.0179869 0.042642 0.055971
382 Non-electrical machinery 0.0185093 0.041723 0.108921
393 Handicrafts and office supplies 0.0209904 0.044206 0.077944
369 Other mineral products 0.0213563 0.057317 0.098664
352 Other chemical products 0.0277417 0.009313 -0.02643
336

320 Spinning and weaving of cotton & wool 0.0340134 0.049388 0.031043
323 Leather and leather products 0.0953683 0.050376 0.0288
These results are further corroborated by the declining mean values of the geographic
concentration levels. Table 4.14 reports the mean values of industrial concentration
across 3-digit industries in Punjab. The mean value of EG index remained roughly
constant from 1995-96 to 2000-01, but the index drastically fell by about 33% in the
next five years. The decline in industrial concentration is mainly explained by the
decline in raw concentration measure (Gini index), and the change in average plant size
measured by the Herfindahl index is less important in explaining
Table 4‐14: Mean values of industrial concentration measures overtime
Concentration measure
Concentration in Punjab
1995-96 2000-01 2005-06
Ellison-Glaeser index 0.1833 0.1789 0.1189
Gini index 0.2794 0.2656 0.2186
Herfindahl index 0.1465 0.1406 0.1374
Survey Year
2000-01 0.8626
Correlation of EG index
2005-06 0.7820 0.7873
the decline in industrial concentration. The correlation coefficient of the EG index
further corroborates these results, which indicates a declining trend in the dynamic
industrial concentration levels in Punjab across industries. It reveals that the correlation
coefficient of EG index between 1995-96 and 2000-01 was 0.86, which fell to 0.79
between 2000-01 and 2005-06. It suggests that the agglomeration of industries
followed a declining trend overtime.
4.3.6 Localization versus Urbanization Externalities
Localization and urbanization externalities are related to local scale externalities that
arise from local information spillovers linked with input and output markets and local
technological developments. If firms in a district learn from local firms in their own
industry, this is called localization externalities; if firms learn from all firms in the
337

district, it is termed as urbanization economies [Henderson et al. (2001)]. While several
urban areas dominate in Pakistan, the relative strength of “localization economies”
versus “urbanization economies” is not known. In the present age, there is a greater role
for information spillovers leading to greater cross-industry learning economies. These
relationships and bonding may explain presence of mega industries around large cities.
Moreover, if these externalities are dynamic, then one may expect that presence of
industry at a location in the past would affect productivity today. In these cases, industry
would be unwilling to move to far off regions where there is no “built-up stock of
knowledge” that makes diversification of industries more and more difficult.
We follow Henderson et al. (2001) to measure the scale of local externalities by a diversity index
written as
∑
( ) ( ) 2
s N
i =
n=
1 in i − n
g ⎡⎣ E E E E ⎤⎦
where i and n index district and industry, respectively, g is the index of localization
and urbanization economies in ith district, E n is for employment in industry n, E is for
total national manufacturing employment, and Ei is for employment in ith district and
Ein is for employment in ith district in nth industry. A lower value of
s
i
s
g i (minimum value
is zero) would imply that the city is non-specialized (high diversity) while a higher value
(approaching two) would indicate complete specialization. In other words, as
s
gi goes up
specialization increases and the diversity tends to fall. Henderson et al. (2001) find a
negative relationship between
s
gi and productivity in Korea.
As before, we use CMI data of Punjab for 1995-96, 2000-01 and 2005-06 and calculate
localization and urbanization index of 3-digit industries for 29 districts of Punjab. In
Table 4.15 we show localization versus urbanization externalities across districts. Our
results suggest that localization economies are much greater in magnitude than
urbanization economies since the diversity index has higher values for most of the
districts with a mean of 0.253 in 1995-96. It shows that there is a greater role for within-
338

industry externalities and that technological spillovers and inter-industry learning has
little role in the manufacturing industries in Punjab.
The on-going dynamic process further reveals that the raw diversity index, g , fell
overtime indicating that the forces of agglomeration were encouraging the diversity of
local industries at an increasing rate. For example, while the index fell moderately
(3.6%) in the first five years, it drastically fell (15%) in the second five year period
indicating that the forces of diversity were increasing that may have led to increased
productivity.
s
i
339

Table 4‐15: Localisation versus urbanization externalities in Punjab, 1995‐96 to 2005‐06
g , 1995-96 g , 2000-01 g , 2005-06
s
i
Sheikhupura 0.0169 0.0539 0.0312
Khushab 0.0407 0.0385 0.0623
Khanewal 0.0493 0.0448 0.1891
Multan 0.0529 0.0452 0.0811
D.G. Khan 0.0625 0.2092 0.0843
Attock 0.0806 0.0735 0.0937
Faisalabad 0.0892 0.076 0.0489
Jhang 0.0919 0.1259 0.2931
T.T. Singh 0.1013 0.2746 0.2907
Gujranwala 0.1196 0.1620 0.0373
Chakwal 0.1381 0.1149 0.1386
Bahawalpur 0.1504 0.2804 0.4388
Kasur 0.1612 0.1286 0.1142
Muzaffargarh 0.1718 0.1975 0.0427
Lahore 0.1724 0.2279 0.1372
Sargodha 0.1967 0.1762 0.2775
Rawalpindi 0.2057 0.0249 0.0280
R.Y. Khan 0.2256 0.2566 0.4002
Sialkot 0.2574 0.5092 0.3548
Gujrat 0.2963 0.3342 0.2082
Mianwali 0.3126 0.4481 0.4431
Vehari 0.3465 0.1553 0.1695
Jhelum 0.3471 0.2824 0.1634
Bahawalnagar 0.3955 0.4615 0.2429
Sahiwal 0.4233 0.0532 0.2364
Okara 0.4896 0.4567 0.2252
Layyah 0.6518 0.7641 0.7094
Bhakkar 0.7365 0.1244 0.1238
Rajanpur 0.9589 0.9778 0.3583
Mean 0.253 0.244 0.208
s
Finally, we regress the specialization index, g () t , on district population along with
controls for survey years. By pooling data of 3-digit industries in three surveys, our
i
s
i
s
i
340

esults reported below (standard errors in parentheses) clearly indicate that district
population is negatively correlated with the index of specialization.
s
gi ( t) = 1.09 − 0.061 ln ( Dist pop)
+ 0.004Year −0.021
Year
(0.128) (0.0086) (0.0113) (0.0109)
00−01 05−06 2
N = 901; R = 0.062.
The time dummy for the second five year period is statistically significant, which
confirms the results (Table 4.15) that the index declined in 2005-06, compared with
1995-96.
341

5 Poverty Impacts of Public Investments and Causes of Industry
Agglomerations
5.1 Introduction
In Pakistan, a key objective of public investment on social and physical infrastructure is
to increase the living standards of the people. The literature shows that poverty
alleviation measures in developing countries are affected by higher income inequality
[Birdsall et al. (1997)]. Some studies find a positive association between income
inequality and poverty based on data from developing countries [e.g., Wagle (2010),
Jamal (2000)].
From the above, it is unclear whether spatial inequality in the provision of social and
physical infrastructure help or hurt poverty alleviation measures? Moreover, we also do
not know whether in circumstances when social and physical infrastructure is given,
how high and low income inequality across spatial units would affect poverty alleviation.
Similarly, we measure industry agglomeration in Pakistan by the Ellison-Glaser index
(see Chapter 4), however, the determinants of industry agglomeration and the nature of
scale economies in agglomerated industries is unknown.
The main focus of this chapter is to predict poverty impacts of public investments and
determinants of industry agglomeration in Pakistan. The next Section asks whether
infrastructure disparities across spatial units influence poverty incidence in Pakistan.
We measure spatial disparities by social and road infrastructure, and poverty means
rural poverty. In the next Section, we study the nexus between infrastructure, income
inequality and poverty where we use graphical and econometric methods for evaluation.
This is followed by exploration on the factors that cause agglomeration in Pakistan’s
manufacturing sector. Finally, we present empirical estimates on the nature of scale
economies and patterns on agglomeration, followed by a compendium of policy notes.
342

5.2 Are Regional Infrastructure Disparities Influencing Incidence of
Poverty? 98
5.2.1 Background
A number of studies in the literature address the questions of pro-poor growth where
the argument is that economic growth alone is not sufficient to alleviate poverty and
that there are a number of other factors that would critically determine whether or not
the growth is pro-poor [see, Ravallion and Datt (2002), Ravallion and Chen (2003),
Kraay (2006), Klasen (2008), Suryahadi et al. (2009)]. While Pakistan is a federation
where significant economic power rests with the federal government, its four provinces
are free to make their independent choices for public policies towards regional economic
development. The heterogeneity in inter-provincial and intra-provincial public policies
over time and space allows us to study the links between public policy leading to
regional infrastructure disparities and its outcomes on poverty. However, we begin by
outlining the underlying causes of poverty in Pakistan, then we document the
determinants of rural poverty by using the regression analysis, followed by an
investigation of the link between infrastructure disparities and poverty.
5.2.2 The causes of poverty in Pakistan
The World Bank (2002) highlights the importance of a range of household
characteristics that are found to be the cause of poverty in Pakistan, which include,
among others, land and livestock assets, education of the household head, size of the
household and nature of employment of household head. We use recent household data
to verify the influence of these factors on the incidence of poverty in Pakistan. We
explore the characteristics of poverty among households by using PSLM 2005-06 data.
In Table 5.1 we show the incidence of poverty on the basis of these household attributes
by taking data of both urban and rural areas. The numbers in Table 5.1 illustrate that no
land households are 68% of the total, but their incidence of poverty is highest at 23% in
the full sample. However, the incidence of rural poverty in landless households is much
higher at 31%. In general, there seems to be a negative association between the size of
98 Sections 5.2 and 5.3 draw from Burki (2010).
343

landholding and the incidence of poverty and this relationship holds more strongly in
rural than in urban households. Similarly,
Table 5‐1: Incidence of poverty by household characteristics, 2005‐06
Poverty by household land ownership:
Head
count
poverty
(%)
Urban Rural Overall
% of
populat
ion
Head
count
poverty
(%)
% of
populat
ion
Head
count
poverty
(%)
% of
populat
ion
Household has no land 13.6% 96.0% 31.1% 54.3% 22.8% 68.4%
Household has subsistence land
holding (1–12.5 acres)
20.6% 3.3% 22.8% 39.0% 22.7% 27.0%
Household has economical land
holding (12.6–25 acres)
2.4% 0.4% 14.2% 4.3% 13.7% 2.9%
Household has large land holding
(more than 25 acres)
Poverty by livestock ownership of
household:
0.0% 0.3% 13.1% 2.4% 12.3% 1.7%
Livestock household (Yes=1, no=0) 15.5% 3.2% 23.4% 36.3% 23.0% 25.2%
No livestock household (yes=1, no=0)
Poverty by education of household
head:
13.7% 96.8% 28.7% 63.7% 22.1% 74.8%
Illiterate head (no education) 26.3% 32.4% 33.0% 56.8% 31.5% 48.6%
Lowly educated head (1–5 years) 14.2% 16.0% 24.7% 17.0% 21.3% 16.7%
Medium educated head (6–10 years) 7.5% 31.0% 16.6% 19.9% 12.6% 23.6%
Highly educated head (more than 10
years)
3.0% 20.6% 8.1% 6.3% 4.9% 11.1%
Poverty by household size:
Small sized household (1–3 members) 1.3% 4.7% 3.9% 4.7% 3.0% 4.7%
Medium sized household (4 – 8
members)
7.3% 58.8% 20.2% 53.3% 15.6% 55.2%
Large sized household (more than 8
members)
Poverty by employment status of
household head:
25.8% 36.5% 37.6% 42.0% 34.0% 40.2%
Unemployed 16.8% 19.5% 23.9% 16.2% 21.2% 17.3%
Self employed 12.8% 32.2% 22.2% 50.9% 19.9% 44.6%
Paid employee 13.0% 48.1% 35.5% 32.0% 25.7% 37.4%
Unpaid family worker 40.0% 0.2% 21.9% 0.9% 24.1% 0.7%
Note: Authors’ calculations from PSLM-HIES 2005-06
344

incidence of poverty in rural livestock households is relatively much lower than nonlivestock
households. Head education has the most powerful but negative effect on
poverty incidence in both urban and rural households. The incidence of poverty on
households with illiterate head is 57%, lowly and medium-educated head, 17% and 19%,
respectively, and for highly educated head poverty is only 5%. A similar picture emerges
from the urban and rural households. Poverty incidence on large-sized households is
34% compared with 16% in medium-sized and 3% in small-sized households in the
national sample. The incidence in urban and rural households is a similar story. Finally,
the employment status of household head is also an important determinant of urban
and rural poverty. We can see that the incidence of poverty is high in paid employees
and unpaid family workers as compared with self-employed and unemployed household
heads. Next, we move on to explore the determinants of poverty by using the regression
framework.
5.2.3 Methodology and data
We investigate the impact of regional infrastructure disparities on rural poverty in
Pakistan by taking individual and household level data. To investigate the nexus
between regional infrastructure and poverty, we consider an empirical model written as
( )
y = α + β( RI ) + γX + η × τ + D + ε
ijt jt ijt k t j ijt
y ijt where is an index for individual i living in district
j
in province k , and t is an index
for the year of the survey. The dependent variable ijt y
equals 1 if the individual falls below
the official poverty line during the survey year,
RIkt
is a measure of regional
infrastructure (measured by principal component post-primary school system index,
and road density variables) in district j at timet . Our variable of interest is, , that
allow s us to test how variation in regional infrastructure affects poverty. Several
RI jt
345

ijt
household and individual level control variables are represented in vector X , the
( )
ηk × τ t
interaction term for period fixed-effects and province fixed-effects capture
variation in individual poverty due to changes in macroeconomic indicators over-time
j
and province-specific heterogeneity, while D is for time-invariant district fixed-effects.
We estimate binary probit regressions by assuming that the dependent variable is a
latent variable (0, 1 variable) that can be expressed as a linear function of other variables
that affect the probability that the individual is poor versus non-poor. The probit model
εijt assumes that the error structure has a cumulative normal distribution function where
ijt
the probability of an individual falling below the poverty line p can be written as
1 2
1 α + β( RI jt ) + γXijt + ( ηk× τt)
+ D j − t
2
pijt = Pr ( yijt = 1)
= ∫
e dt
−∞ 2π
where t is a standardized normal variable. The coefficient estimates are obtained from
the maximum likelihood estimation procedure.
As already discussed in Chapter 4 (Section 4.3.2), the household and individual level
data was obtained from six rounds of the Household Income and Expenditure Survey
(HIES) to estimate poverty trends. The combined data file for rural households
consisted of 373991 individual level observations. Altogether, 24002 observations were
discarded from the original sample because they did not report food expenditure needed
to measure poverty, and 694 observations were deleted from the working sample due to
faulty data on land records. After these adjustments in the total sample, we were left
with a sample of 349295 observations. The distribution of our working sample by survey
years is presented in Table 5.2.
Table 5‐2: Distribution of Rural Sample by Household Survey
Survey year Individual
observations
HIES 1992-93 57725
HIES 1993-94 47638
HIES 1996-97 45310
HIES-PIHS 1998-99 65029
346

HIES-PIHS 2000-01 66445
PSLM 2005-06 67148
Total sample 349295
We augment household survey data with external information on infrastructure
variables at the district level on education and health, and road density. We use
principal component post-primary school system and hospital index by districts that is
more apt to be exogenous to a household. For details on how this variable was
construction, see Chapter 4, Section 4.3.3. Reliable data on road infrastructure was
available only for the Punjab province. Therefore, we use road infrastructure data of
Punjab for the regression analysis (see Chapter 4, Section 4.3.4 for further details).
Table 5.3 shows the definition of the variables while Table 5.4 reveals the descriptive
statistics of all the individual, household and other control variables used in the
regressions.
5.2.4 The basic regressions
The basic regression results for the probit maximum likelihood are presented in Table
5.5. In all regressions, we include household and individual controls as the set of
regressors, which consist of four dummy variables to capture the education level of the
head of household (the omitted category is illiterate household head); three dummy
variables to capture the effects of household size (the omitted category is small-sized
households consisting of 1 to 3 members); four dummies to capture household
landholding (the omitted category is no landholding); one dummy variable each to
indicate if household owns livestock and farm equipment, the value of financial assets
owned by household; age of the individual; a dummy for gender; days worked during
the last month; and twelve sector of employment dummies (the omitted category is
unemployed).
The estimation results presented in column (1) are marginal effects from probit
regression obtained without including any covariates other than household and
347

individual characteristics. Column (2) presents results when we include in regression
fifty-six district indicators to capture time-invariant changes across districts (the
omitted category is Vehari district). The estimation results in column (3) are obtained by
deleting from the regression district effects but including
348

Table 5‐3: Definition of variables used in the Probit Regressions
Variables Description
Dependent variable:
Poverty =1 if individual falls below the basic need poverty line
Policy variables:
Principal component basic health,
dispensary & rural health index
Principal component post-primary
school system and hospital index
Principal component teachers in high
schools & colleges index
Principal component teachers in
middle and primary schools index
Principal component primary school
index
Principal component beds in TB
clinics index
Other Independent variables:
Household level controls:
Principal component index, varying by districts, with high
positive loadings on basic health units, dispensaries, rural
health centers, mother child care center, rural health center
beds and dispensary beds.
Principal component index, varying by districts,
characterized by high positive factor loadings on degree
college size, hospital size, high school size, intermediate
college size and middle school size.
Principal component index, varying by districts,
characterized by high factor loadings in teacher student
ratio in high schools and degree colleges.
Principal component index, varying by districts, with high
factor loadings in teacher student ratio in middle schools
and teacher student ratio in primary schools.
Principal component index, varying by districts,
characterized by high factor loading in primary school size.
Principal component index, varying by districts, with high
factor loading in TB clinics.
Head is illiterate =1 if individual is from a household where head is illiterate
Lowly educated head, 1 – 5 years =1 if individual is from a household where head has 1–5
years of schooling
Medium educated head, 6 – 10 years =1 if individual is from a household where head has 6–10
years of schooling
Highly educated head, 11 years or =1 if individual is from a household where head has 11 or
more
more years of schooling
Small sized household (1–3
=1 if individual belongs to a household comprising of 1–3
members)
members
Medium sized household (4–8 =1 if individual belongs to a household comprising of 4–8
members)
members
Large sized household ( 9 or more =1 if individual belongs to a household comprising of more
members)
than 8 members
No land =1 if individual belongs to a household that does not have
any inherited or rented-in land
Household has subsistence
=1if individual belongs to a household that has 1–12.5 acres
landholding, 1–12.5 acres
of inherited or rented-in land
Household has economical
=1if individual belongs to a household that has 12.6–25
landholding, 12.6–25 acres
acres of inherited or rented-in land
Household has large landholding, =1if individual belongs to a household that has more than
more than 25 acres
25 acres of inherited or rented-in land
Livestock owned (yes=1, no=0) =1 if individual belongs to a household that owns livestock
Farm equipment owned (yes=1, =1 if the individual belongs to a household that owns farm
no=0)
equipment
349

Financial assets owned by household
(Rs. million)
Individual level controls:
Value of financial assets owned by an individual’s
household in millions of rupees
Age Age of the individual in years
Male =1 if an individual is a male
Female =1 if an individual is a female
Days worked during last month Number of days worked by the individual during the last
month
Employment in agriculture or
forestry
=1 if individual is employed in agriculture or forestry sector
Employment in mining =1 if individual is employed in mining sector
Employment in manufacturing =1 if individual is employed in manufacturing sector
Employment in electricity or gas =1 if individual is employed in electricity or gas supply
sector
Employment in construction =1 if individual is employed in construction sector
Employment in wholesale or retail =1 if individual is employed in wholesale and retail trading
trader
sector
Employment in transport =1 if individual is employed in transport sector
Employment in real estate =1 if individual is employed in real estate sector
Employment in social or personal =1 if individual is employed in provision of social or
services
personal services
Unemployed =1 if individual is unemployed
Self employed =1 if individual is self-employed
Paid employee =1 if individual is a paid employee
Unpaid family worker =1 if individual is an unpaid family worker
No employment status =1 if individual has no employment status
Province-year effects:
Survey year 1992-93× Punjab =1 if individual belongs to the province of Punjab and the
survey year is 1992-93
Survey year 1992-93× Sindh =1 if individual belongs to the province of Sindh and the
survey year is 1992-93
Survey year 1992-93× Balochistan =1 if individual belongs to the province of Balochistan and
the survey year is 1992-93
Survey year 1992-93× NWFP =1 if individual belongs to the province of NWFP and the
survey year is 1992-93
Survey year 1993-94× Punjab =1 if individual belongs to the province of Punjab and the
survey year is 1993-94
Survey year 1993-94× Sindh =1 if individual belongs to the province of Sindh and the
survey year is 1993-94
Survey year 1993-94× Balochistan =1 if individual belongs to the province of Balochistan and
the survey year is 1993-94
Survey year 1993-94× NWFP =1 if individual belongs to the province of NWFP and the
survey year is 1993-94
Survey year 1996-97× Punjab =1 if individual belongs to the province of Punjab and the
survey year is 1996-97
Survey year 1996-97× Sindh =1 if individual belongs to the province of Sindh and the
survey year is 1996-97
Survey year 1996-97× Balochistan =1 if individual belongs to the province of Balochistan and
the survey year is 1996-97
Survey year 1996-97× NWFP =1 if individual belongs to the province of NWFP and the
survey year is 1996-97
Survey year 1998-99× Punjab =1 if individual belongs to the province of Punjab and the
survey year is 1998-99
350

Survey year 1998-99× Sindh =1 if individual belongs to the province of Sindh and the
survey year is 1998-99
Survey year 1998-99× Balochistan =1 if individual belongs to the province of Balochistan and
the survey year is 1998-99
Survey year 1998-99× NWFP =1 if individual belongs to the province of NWFP and the
survey year is 1998-99
Survey year 2001-02× Punjab =1 if individual belongs to the province of Punjab and the
survey year is 2001-02
Survey year 2001-02× Sindh =1 if individual belongs to the province of Sindh and the
survey year is 2001-02
Survey year 2001-02× Balochistan =1 if individual belongs to the province of Balochistan and
the survey year is 2001-02
Survey year 2001-02× NWFP =1 if individual belongs to the province of NWFP and the
survey year is 2001-02
Survey year 2005-06×Punjab =1 if individual belongs to the province of Punjab and the
survey year is 2005-06
Survey year 2005-06×Sindh =1 if individual belongs to the province of Sindh and the
survey year is 2005-06
Survey year 2005-06× Balochistan =1 if individual belongs to the province of Balochistan and
the survey year is 2005-06
Survey year 2005-06×NWFP =1 if individual belongs to the province of NWFP and the
survey year is 2005-06
District effects:
Abbottabad =1 if individual belongs to Abbottabad district
Attock =1 if individual belongs to Attock district
Badin =1 if individual belongs to Badin district
Bahawalnagar =1 if individual belongs to Bahawalnagar district
Bahawalpur =1 if individual belongs to Bahawalpur district
Bannu =1 if individual belongs to Bannu district
Bhakkar =1 if individual belongs to Bhakkar district
Chakwal =1 if individual belongs to Chakwal district
Dadu =1 if individual belongs to Dadu district
Dera Ghazi Khan =1 if individual belongs to Dera Ghazi Khan district
Dera Ismail Khan =1 if individual belongs to Dera Ismail Khan district
Dir =1 if individual belongs to Dir district
Faisalabad =1 if individual belongs to Faisalabad district
Gujranwala =1 if individual belongs to Gujranwala district
Gujrat =1 if individual belongs to Gujrat district
Hyderabad =1 if individual belongs to Hyderabad district
Jacobabad =1 if individual belongs to Jacobabad district
Jhang =1 if individual belongs to Jhang district
Jhelum =1 if individual belongs to Jhelum district
Kalat =1 if individual belongs to Kalat district
Karachi East =1 if individual belongs to Karachi East district
Karak =1 if individual belongs to Karak district
Kasur =1 if individual belongs to Kasur district
Khairpur =1 if individual belongs to Khairpur district
Khanewal =1 if individual belongs to Khanewal district
Khushab =1 if individual belongs to Khushab district
Kohat =1 if individual belongs to Kohat district
Lahore =1 if individual belongs to Lahore district
Larkana =1 if individual belongs to Larkana district
Layyah =1 if individual belongs to Layyah district
Mansehra =1 if individual belongs to Mansehra district
351

Mardan =1 if individual belongs to Mardan district
Mekran =1 if individual belongs to Mekran district
Mianwali =1 if individual belongs to Mianwali district
Multan =1 if individual belongs to Multan district
Muzzafargarh =1 if individual belongs to Muzzafargarh district
Nawabshah =1 if individual belongs to Nawabshah district
Okara =1 if individual belongs to Okara district
Peshawar =1 if individual belongs to Peshawar district
Quetta =1 if individual belongs to Quetta district
Rahimyar Khan =1 if individual belongs to Rahimyar Khan district
Rajanpur =1 if individual belongs to Rajanpur district
Rawalpindi =1 if individual belongs to Rawalpindi district
Sahiwal =1 if individual belongs to Sahiwal district
Sanghar =1 if individual belongs to Sanghar district
Sargodha =1 if individual belongs to Sargodha district
Sheikhupura =1 if individual belongs to Sheikhupura district
Shikarpur =1 if individual belongs to Shikarpur district
Sialkot =1 if individual belongs to Sialkot district
Sibi =1 if individual belongs to Sibi district
Sukkur =1 if individual belongs to Sukkur district
Swat =1 if individual belongs to Swat district
Tharparker =1 if individual belongs to Tharparker district
Thatta =1 if individual belongs to Thatta district
Toba Tek Singh =1 if individual belongs to Toba Tek Singh district
Vehari =1 if individual belongs to Vehari district
352

Table 5‐4: Descriptive Statistics – full sample
Explanatory variables Mean
Std.
Dev. Min Max
Poverty 0.351 0.47 0 1
Principal component basic health, dispensary & rural
health index
0.667 6.690 -7.213 33.737
Principal component post-primary school system and
hospital index
-0.205 2.477 -2.669 25.541
Principal component teachers in high schools & colleges
index
0.540 2.648 -4.726 12.051
Principal component teachers in middle and primary
schools index
0.108 2.077 -3.244 8.402
Principal component primary school index 0.128 1.182 -4.287 4.837
Principal component beds in TB clinics index -0.038 0.928 -1.587 8.978
Head is illiterate 0.630 0.48 0 1
Lowly educated head, 1 – 5 years (yes=1, no=0) 0.16 0.37 0 1
Medium educated head, 6 – 10 years (yes=1, no=0) 0.150 0.357 0 1
Highly educated head, 11 years or more (yes=1, no=0) 0.049 0.217 0 1
Medium sized household, 4 – 8 members (yes=1, no=0) 0.553 0.497 0 1
Large sized household, 9 or more members (yes=1, no=0) 0.379 0.485 0 1
Household has subsistence landholding, 1 – 12.5 acres
(yes=1, no=0)
0.305 0.460 0 1
Household has economical landholding, 12.6 – 25 acres
(yes=1, no=0)
0.048 0.213 0 1
Household has large landholding, more than 26 acres
(yes=1, no=0)
0.169 0.128 0 1
Livestock owned (yes=1, no=0) 0.297 0.45 0 1
Farm equipment owned (yes=1, no=0) 0.007 0.083 0 1
Financial assets owned by household (Rs. million) 0.021 0.128 0 9.734
Age 22.83 18.68 0 99
Male (yes=1, no=0) 0.510 0.499 0 1
Days worked during last month 6.517 11.528 0 30
Employment in agriculture or forestry (yes=1, no=0) 0.171 0.377 0 1
Employment in mining (yes=1, no=0) 0.0009 0.312 0 1
Employment in manufacturing (yes=1, no=0) 0.019 0.136 0 1
Employment in electricity or gas (yes=1, no=0) 0.0017 0.041 0 1
Employment in construction (yes=1, no=0) 0.022 0.146 0 1
Employment in wholesale or retail trader (yes=1, no=0) 0.0233 0.151 0 1
Employment in transport (yes=1, no=0) 0.0141 0.118 0 1
Employment in real estate (yes=1, no=0) 0.0008 0.029 0 1
Employment in social or personal services (yes=1, no=0) 0.035 0.184 0 1
Self employed (yes=1, no=0) 0.105 0.307 0 1
Paid employee (yes=1, no=0) 0.098 0.297 0 1
Unpaid family worker (yes=1, no=0) 0.089 0.285 0 1
Survey year 1992-93× Balochistan 0.018 0.135 0 1
Survey year 1992-93× NWFP 0.036 0.186 0 1
353

Survey year 1992-93× Punjab 0.073 0.260 0 1
Survey year 1992-93× Sindh 0.037 0.189 0 1
Survey year 1993-94× Balochistan 0.015 0.123 0 1
Survey year 1993-94× NWFP 0.030 0.172 0 1
Survey year 1993-94× Punjab 0.060 0.238 0 1
Survey year 1993-94× Sindh 0.029 0.169 0 1
Survey year 1996-97× Balochistan 0.015 0.124 0 1
Survey year 1996-97× NWFP 0.031 0.174 0 1
Survey year 1996-97× Punjab 0.055 0.229 0 1
Survey year 1996-97× Sindh 0.026 0.160 0 1
Survey year 1998-99× Balochistan 0.031 0.173 0 1
Survey year 1998-99× NWFP 0.042 0.201 0 1
Survey year 1998-99× Punjab 0.069 0.254 0 1
Survey year 1998-99× Sindh 0.043 0.203 0 1
Survey year 2001-02× Balochistan 0.029 0.170 0 1
Survey year 2001-02× NWFP 0.041 0.198 0 1
Survey year 2001-02× Punjab 0.069 0.254 0 1
Survey year 2001-02× Sindh 0.049 0.216 0 1
Survey year 2005-06×Balochistan 0.0287 0.167 0 1
Survey year 2005-06×NWFP 0.045 0.208 0 1
Survey year 2005-06× Punjab 0.072 0.258 0 1
Survey year 2005-06× Sindh 0.042 0.201 0 1
Abbottabad 0.023 0.150 0 1
Attock 0.011 0.104 0 1
Badin 0.012 0.112 0 1
Bahawalnagar 0.012 0.109 0 1
Bahawalpur 0.0148 0.120 0 1
Bannu 0.016 0.128 0 1
Bhakkar 0.010 0.101 0 1
Chakwal 0.006 0.081 0 1
Dadu 0.018 0.135 0 1
Dera Ghazi Khan 0.010 0.102 0 1
Dera Ismail Khan 0.017 0.130 0 1
Dir 0.028 0.166 0 1
Faisalabad 0.02 0.145 0 1
Gujranwala 0.015 0.133 0 1
Gujrat 0.0202 0.140 0 1
Hyderabad 0.017 0.132 0 1
Jacobabad 0.019 0.137 0 1
Jhang 0.017 0.132 0 1
Jhelum 0.006 0.077 0 1
Kalat 0.031 0.173 0 1
Karachi East 0.009 0.096 0 1
Karak 0.009 0.097 0 1
Kasur 0.014 0.118 0 1
Khairpur 0.017 0.130 0 1
Khanewal 0.010 0.101 0 1
Khushab 0.007 0.085 0 1
354

Kohat 0.012 0.111 0 1
Lahore 0.008 0.090 0 1
Larkana 0.0216 0.145 0 1
Layyah 0.006 0.078 0 1
Mansehra 0.019 0.139 0 1
Mardan 0.025 0.156 0 1
Mekran 0.019 0.136 0 1
Mianwali 0.007 0.086 0 1
Multan 0.019 0.138 0 1
Muzzafargarh 0.015 00.122 0 1
Nawabshah 0.027 0.162 0 1
Okara 0.015 0.122 0 1
Peshawar 0.028 0.166 0 1
Quetta 0.052 0.223 0 1
Rahimyar Khan 0.020 0.140 0 1
Rajanpur 0.005 0.073 0 1
Rawalpindi 0.020 0.142 0 1
Sahiwal 0.020 0.140 0 1
Sanghar 0.015 0.122 0 1
Sargodha 0.014 0.118 0 1
Sheikhupura 0.019 0.136 0 1
Shikarpur 0.011 0.106 0 1
Sialkot 0.022 0.149 0 1
Sibi 0.037 0.188 0 1
Sukkur 0.022 0.147 0 1
Swat 0.045 0.209 0 1
Tharparker 0.025 0.158 0 1
Thatta 0.012 0.112 0 1
Toba Tek Singh 0.012 0.109 0 1
Vehari 0.012 0.109 0 1
Number of observations 349295 -- -- --
355

Table 5‐4: Effects of household and individual characteristics on rural poverty in Pakistan, basic results
Explanatory variables
Probit regressions (average marginal effects)
(1) (2) (3) (4)
Lowly educated head, 1–5 years (yes=1, no=0) -0.101*** -0.089*** -0.093*** -0.089***
(-11.71) (-12.38) (-11.64) (-11.49)
Medium educated head, 6–10 years (yes=1, no=0) -1.166*** -0.159*** -0.161*** -0.155***
(-17.65) (-18.48) (-17.80) (-17.72)
Highly educated head, 11 years or more (yes=1, -0.256*** -0.252*** -0.251*** -0.249***
no=0)
(-19.12) (-19.00) (-20.21) (-20.48)
Medium sized household, 4–8 members (yes=1, 0.347*** 0.355*** 0.347*** 0.356***
no=0)
(33.71) (34.98) (31.64) (32.45)
Large sized household, 9 or more members 0.574*** 0.584*** 0.574*** 0.585***
(yes=1, no=0)
(36.74) (40.83) (39.74) (40.57)
Household has subsistence landholding, 1–12.5 -0.109*** -0.103*** -0.095*** -0.092***
acres (yes=1, no=0)
(-8.49) (-7.85) (-7.69) (-7.51)
Household has economical landholding, 12.6–25 -0.148*** -0.159*** -0.146*** -0.157***
acres (yes=1, no=0)
(-5.60) (-7.26) (-7.03) (-7.88)
Household has large landholding, more than 25 -0.193*** -0.205*** -0.184*** -0.198***
acres (yes=1, no=0)
(-5.93) (-6.99) (-5.47) (-6.51)
Livestock owned (yes=1, no=0) -0.034** -0.048*** -0.041*** -0.051***
(-2.45) (-4.04) (-3.63) (-4.33)
Farm equipment owned (yes=1, no=0) -0.101** -0.098** -0.098*** -0.097***
(-2.38) (-2.16) (-2.96) (-2.79)
Financial assets owned by HH (Rs. million) -1.74*** -1.784*** -1.777*** -1.738***
(-6.19) (-5.88) (-5.81) (-5.49)
Age (years) -
-
-
-
0.0009** 0.0008** 0.0008** 0.0008**
*
*
*
*
(-9.39) (-9.26) (-11.02) (-10.07)
Male (yes=1, no=0) -0.007*** -0.007*** -0.007*** -0.006***
(-3.17) (-3.21) (-3.42) (-3.11)
Days worked during last month -0.0004 -0.0002 -0.0002 -0.0002
(-0.58) (-0.40) (-0.43) (-0.70)
12 sectors and employment status indicators Yes Yes Yes Yes
District level controls included No Yes No Yes
Province× year fixed effects included No No Yes Yes
Pseudo R 2 0.109 0.126 0.129 0.153
Number of observations 349295 349295 349295 349295
Notes: All regressions are estimated by probit maximum likelihood. Numbers in parenthesis are
asymptotic t-values obtained from robust standard errors adjusted for clustering at the district level.
The regressions include intercept terms, but they are not reported. **, and *** denote statistical
significance at the 5 and 1% levels, respectively. Sector of employment indicators are agriculture &
forestry; mining; manufacturing; electricity or gas; construction; wholesale and retail trade; transport;
real estate; and social and personal services while employment status indicators are self-employed,
paid-employees and unpaid family workers. Province*year fixed effects are 24 province-year
interaction terms to introduce province level trends, which capture variation in economic conditions at
the province level during each year of the survey that may affect rural poverty. District fixed effects are
specified by introducing district dummy variables, one each for each district, which are meant to
capture the effects of time-invariant district specific characteristics on poverty in our sample.
356

twenty-four province×year indicators to control for province-specific time-varying
factors (the excluded category is Sindh×2005-06). Column (4) presents results when we
control for both province×year fixed effects and district fixed effects. Finally, the
numbers in parenthesis in Table 5.5 indicate asymptotic t-values obtained from robust
standard errors, corrected for clustering at the district level.
The estimation results presented in Table 5.5, columns (1)–(4), indicate that the
parameter estimates are highly robust to alternative model specifications. The estimated
coefficients indicate consistent negative impact of education of household head and
ownership of household assets (including land, livestock, farm equipment and financial
assets) on poverty. For instance, there is unequivocal evidence to show that poverty
coexists with illiteracy of heads of household and lack of asset ownership. Our results
show that the incidence of poverty dramatically falls with increase in education of
household head. The results in column (4) suggest that compared with illiterate heads,
poverty falls by 8.9% when head has one to five years of schooling, 15.5% when heads
has six to ten years of schooling and 25% when head has more than secondary
education. The probability of being poor progressively falls as household land wealth
increases from subsistence landholding, to economical landholding and then to large
landholding. Ownership of farm equipment is also negatively associated with the
incidence of poverty in rural Pakistan. Livestock households are significantly less likely
to be poor than non-livestock households.
In household level factors, the family size makes the most significant difference on
poverty. Increase in family size consistently increases poverty in our sample. Relative to
small sized households (1 to 3 members), the likelihood of poverty increases by 36% for
individuals coming from medium sized households (4 to 8 members) and to almost 59%
for individuals from large sized households. In other words, scale economies often
attributed to larger households do not exist in rural Pakistan. In the individual level
control variables, the coefficient on age is significantly negative suggesting that younger
population is more likely to be poor. Given that the majority of Pakistani population is
young, this result is alarming. The incidence of poverty on male and female population
357

is roughly similar, except that males are 0.06% less likely to be poor than females.
Increase in the number of days worked significantly decreases the probability of poverty
in our sample. All of the components of Table 5.5 together verify the causes of poverty
that has received a great deal of attention in the previous poverty literature and that
gives us a great deal of confidence that the
5.2.5 The effect of spatial inequality in social infrastructure on rural poverty
While current policy attention to human capital accumulation for poverty alleviation in
developing countries is clearly appropriate, the long standing inattention of the
governments in these countries to the effect of distribution of social infrastructure on
poverty has been costly. However, the magnitude of these effects on poverty remains
ambiguous. Therefore, the central question we wish to answer in this part of the report
is straightforward: does spatial inequality in social infrastructure affect rural poverty in
Pakistan?
We examine this relationship by evaluating the impact of principal component postprimary
school system and hospital index. We consider the relationship by estimating
an augmented probit model where in addition to the variables considered in the basic
model we also introduce six factor component education and health indices discussed in
Chapter 4, Section 4.3.3. We rely on district level factor component education and
health indices that are more apt to be exogenous to a household. Because factor
component education and health indices vary across districts and are correlated with
district dummy variables, we do not include district effects in these regressions.
Moreover, all standard errors account for district level clustering and also allow general
form of heteroskedasticity.
The estimation results in Table 5.6, column 1 reveal that improvements in local
education and health facilities do not produce uniform effects on poverty alleviation.
Only principal component post-primary school system & hospital index is negatively
and significantly correlated with rural poverty: an increase in this index by 2.48 (its
standard deviation) decreases the probability of poverty by 1.98%. However, the
358

probability of rural poverty is not affected by increase/decrease in local supply of any
other education or health indicator including primary
Table 5‐5: Marginal effects of the impact of education and health infrastructure on rural poverty
Explanatory variables E&H index
Initial
conditions
E&H index
(1) (2)
Principal component basic health, dispensary & rural health 0.002
0.002
index
(0.76)
(0.84)
Principal component post-primary school system and hospital -0.008** -0.008*
index
(-2.20)
(-1.81)
Principal component teachers in high school and college index -0.005
0.004
(-0.72)
(0.37)
Principal component quality of middle & primary schools 0.005
-0.002
(0.51)
(-0.21)
Principal component primary school index -0.014
-0.007
(-1.32)
(-0.72)
Principal component beds in TB clinics index -0.003 0.012**
(-0.44) (-2.23)
Lowly educated head, 1 – 5 years (yes=1, no=0) -0.093*** -0.094***
(-11.58) (-11.80)
Medium educated head, 6 – 10 years (yes=1, no=0) -0.159*** -0.160***
(-18.67) (-17.85)
Highly educated head, 11 years or more (yes=1, no=0) -0.251*** -0.251***
(-20.23) (-20.38)
Medium sized household, 4 – 8 members (yes=1, no=0) 0.349*** 0.349***
(32.36) (32.13)
Large sized household, 9 or more members (yes=1, no=0) 0.576*** 0.576***
(41.27) (40.27)
Household has subsistence landholding, 1 – 12.5 acres (yes=1, -0.098*** -0.097***
no=0)
(-8.01) (-8.10)
Household has economical landholding, 12.6 – 25 acres (yes=1, -0.150*** -0.149***
no=0)
(-7.15)
(-7.01)
Household has large landholding, more than 26 acres (yes=1, -0.189*** -0.189***
no=0)
(-5.75)
(-5.54)
Livestock owned by household (Rs. million) -0.043*** -0.043***
(-3.72) (-3.65)
Farm equipment owned by household (Rs. million) -0.099*** -0.098***
(-3.01) (-2.95)
Financial assets owned by household (Rs. million) -1.773*** -1.777***
(-5.70) (-5.78)
Individual level controls included Yes Yes
12 sector of employment dummy included Yes Yes
District level controls included No No
Province*year fixed effects included Yes Yes
Number of observations 349295 349295
359

Pseudo R 2 0.131 0.131
Notes: All regressions are estimated by probit maximum likelihood. Numbers in parenthesis are
asymptotic t-values obtained from robust standard errors adjusted for clustering at the district
level. The regressions include intercept terms, but they are not reported. **, and *** denote
statistical significance at the 5 and 1% levels, respectively. For definition of control variables, see
Table 5.5.
school index, health facilities through basic health units, dispensaries and rural health
facilities. The impact of increased supply of primary schools and TB clinics is also
negative, but no statistical significance could be attached to these results. The obvious
concern with these findings is whether we can interpret them as the causal effect of
spatial inequality in public education and health facilities on rural poverty. It may be
argued that investments in education and health infrastructure may be linked through
lagged effects on poverty. To address this concern, we also run a regression where we
include only the initial condition education and health infrastructure variable.
Our results reported in column (2) indicate that qualitatively there is no change in the
results, except that principal component beds in TB clinics index variable also becomes
significantly negative. These results indicate that improvements in the supply of TB
clinics have a lagged effect on rural poverty, which is understandable given the longterm
treatment needed for cure in TB disease. For subsequent analysis we select only
principal component post-primary school system index to examine its effects on poverty
under various local level inequality and polarization regimes.
5.2.6 Spatial inequality in road infrastructure and rural poverty
In this sub-section, we empirically examine the relationship between local road density
and rural poverty. We relate this spatial and inter-temporal variation in relative road
density to variation in rural poverty using the individual and household level data of
Punjab province from 1992-93 to 2005-06 supplemented by external information on
district level road density for the corresponding data points (see Chapter 4, Section 4.3.4
for more details).
Table 5.7 shows the descriptive statistics of the relevant variables based on the data of
the Punjab province. The sample consists of 139944 individuals. We run several
360

egressions by varying the empirical specifications to evaluate the effect of road density
on poverty under alternative assumptions. Results presented in Table 5.8 correspond to
the model where road
Table 5‐6: Summary Statistics of Punjab’s Data
Explanatory variables Mean
Std.
Dev.
Min Max
Poverty 0.326 0.468 0 1
Road Density 0.265 0.125 0.044 0.697
Head without education 0.618 0.485 0 1
Lowly educated head, 1 – 5 years (yes=1, no=0) 0.168 0.374 0 1
Medium educated head, 6 – 10 years (yes=1, no=0) 0.181 0.385 0 1
Highly educated head, 11 years or more (yes=1, no=0) 0.031 0.175 0 1
Small sized household 0.760 0.265 0 1
Medium sized household, 4 – 8 members (yes=1, no=0) 0.621 0.485 0 1
Large sized household, 9 or more members (yes=1, no=0) 0.302 0.459 0 1
Household has no landholding 0.615 0.486
Household has subsistence landholding, 1–12.5 acres
(yes=1, no=0)
0.316 0.465 0 1
Household has economical landholding, 12.6–25 acres
(yes=1, no=0)
0.481 0.214 0 1
Household has large landholding, more than 26 acres
(yes=1, no=0)
0.020 0.140 0 1
Livestock owned (yes=1, no=0) 0.366 0.481 0 1
Farm equipment owned (yes=1, no=0) 0.009 0.094 0 1
Financial assets owned by household (Rs. million) 0.018 0.059 0 2.28
Age 24.300 19.633 0 99
Male (yes=1, no=0) 0.503 0.499 0 1
Days worked during last month 6.951 11.869 0 30
Employment in agriculture or forestry (yes=1, no=0) 0.185 0.388 0 1
Employment in mining (yes=1, no=0) 0.0003 0.017 0 1
Employment in manufacturing (yes=1, no=0) 0.028 0.167 0 1
Employment in electricity or gas (yes=1, no=0) 0.001 0.034 0 1
Employment in construction (yes=1, no=0) 0.021 0.144 0 1
Employment in wholesale or retail trader (yes=1, no=0) 0.026 0.160 0 1
Employment in transport (yes=1, no=0) 0.013 0.114 0 1
Employment in real estate (yes=1, no=0) 0.0009 0.030 0 1
Employment in social or personal services (yes=1, no=0) 0.034 0.183 0 1
Self employed (yes=1, no=0) 0.121 0.326 0 1
Paid employee (yes=1, no=0) 0.098 0.298 0 1
Unpaid family worker (yes=1, no=0) 0.096 0.295 0 1
Survey year 1992-93 0.182 0.386 0 1
Survey year 1993-94 0.150 0.357 0 1
361

Survey year 1996-97 0.139 0.346 0 1
Survey year 1998-99 0.173 0.378 0 1
Survey year 2001-02 0.174 0.379 0 1
Survey year 2005-06 0.180 0.384 0 1
Attock 0.027 0.164 0 1
Bahawalnagar 0.030 0.171 0 1
Bahawalpur 0.036 0.188 0 1
Bhakkar 0.026 0.159 0 1
Chakwal 0.016 0.127 0 1
Dera Ghazi Khan 0.026 0.160 0 1
Faisalabad 0.053 0.225 0 1
Gujranwala 0.045 0.208 0 1
Gujrat 0.050 0.218 0 1
Jhang 0.044 0.206 0 1
Jhelum 0.014 0.121 0 1
Kasur 0.035 0.184 0 1
Khanewal 0.025 0.159 0 1
Khushab 0.018 0.133 0 1
Lahore 0.020 0.141 0 1
Layyah 0.015 0.123 0 1
Mianwali 0.018 0.135 0 1
Multan 0.049 0.216 0 1
Muzzafargarh 0.038 0.191 0 1
Okara 0.038 0.191 0 1
Rahimyar Khan 0.050 0.218 0 1
Rajanpur 0.013 0.116 0 1
Rawalpindi 0.052 0.222 0 1
Sahiwal 0.049 0.217 0 1
Sargodha 0.035 0.185 0 1
Sheikhupura 0.047 0.212 0 1
Sialkot 0.057 0.232 0 1
Toba Tek Singh 0.030 0.170 0 1
Number of observations 139944 -- -- --
362

Table 5‐8: Impact of Road Density on Poverty in Punjab
Explanatory variables
Probit regression marginal effects
Road density Starting road
density of
1992–93
(1) (2)
Road Density -0.336*** -0.513***
(-3.63)
(-4.55)
Lowly educated head, 1 – 5 years (yes=1, no=0) -0.081*** -0.080***
(-7.59)
(-7.68)
Medium educated head, 6 – 10 years (yes=1, no=0) -0.155*** -0.155***
(-11.46) (-11.30)
Highly educated head, 11 years or more (yes=1, no=0) -0.246*** -0.246***
(-9.52)
(-9.63)
Medium sized household, 4 – 8 members (yes=1, no=0) 0.301*** 0.302***
(22.78)
(22.56)
Large sized household, 9 or more members (yes=1, no=0) 0.560*** 0.561***
(35.61)
(34.87)
Household has subsistence landholding, 1 – 12.5 acres (yes=1, -0.146*** -0.145***
no=0)
(-11.12) (-10.86)
Household has economical landholding, 12.6 – 25 acres (yes=1, -0.189*** -0.190***
no=0)
(-6.66)
(-6.81)
Household has large landholding, more than 26 acres (yes=1, -0.213*** -0.215***
no=0)
(-4.06)
(-4.17)
Livestock owned (yes=1, no=0) -0.032*** -0.033***
(-2.76)
(-2.84)
Farm equipment owned (yes=1, no=0) -0.087* -0.084*
(-1.82)
(-1.76)
Financial assets owned by household (Rs. million) -2.340*** -2.316***
(-6.10)
(-6.07)
Age -0.001*** -0.001***
(-8.32)
(-8.27)
Male (yes=1, no=0) --0.005* -0.006*
(-1.91)
(-1.95)
Days worked during last month --0.0004 -0.001
(-1.07)
(-1.05)
12 sector and employment status dummies Yes Yes
Individual level controls included Yes Yes
District level controls included No No
Year fixed effects included Yes Yes
Pseudo R2 0.154 0.155
Number of observations 139944 139944
Notes: All regressions are estimated by probit maximum likelihood. Numbers in parenthesis are
asymptotic t-values obtained from robust standard errors adjusted for clustering at the district
level. The regressions include intercept terms, but they are not reported. **, and *** denote
statistical significance at the 5 and 1% levels, respectively. For definition of control variables, see
Table 5.5.
density and starting road density of 1992-93 are used as variables of interest where the
set of regressors include household and individual controls, sector and employment
363

status dummies and province-year fixed effects. Column (1) presents our estimation
results for road density while column (2) presents results for starting road density of
1992–93.
Table 5.8 shows a strong negative relationship between regional road density and
individual poverty: the higher the road density the lower is the probability of poverty.
This negative relationship is statistically significant at the 99% confidence level or
better. The magnitude of the association between road density and poverty is large. A
one standard deviation (0.125) increase in road density decreases the probability of
poverty by 4.2% which is substantially higher than the effect of regional investment on
post-primary school system. Column (2) estimates the impact of initial level or starting
level road density of districts in 1992-93 on long run poverty. Our results indicate that
the starting level road density is associated with a decline in poverty of a larger
magnitude: a one SD increase in starting road density leads to 6.4% decline in poverty
or almost 2.2% point more poverty reduction in the long run than the short run decline.
It thus appears that lower district-level road density substantially increases rural
poverty. It is clear from these results that investments in road infrastructure are most
pro poor.
5.3 Nexus between Infrastructure, Income Inequality and Poverty
Given investments in regional infrastructure, regions facing inequality in income may
face greater difficulties in reducing poverty. Some empirical papers [see, for example,
Ravallion (1997), Ravallion and Datt (2002), Ravallion and Chen (1997)] have already
established that, for a positive rate of economic growth, regional income disparities may
lead to varying effects on poverty. Our goal is to let the data tell us that given
infrastructure investments how increase in regional measures of inequality affect
incidence of poverty?
364

5.3.1 Does high inequality cause high poverty in Pakistan?
Before turning to a more rigorous analysis, we begin by exporting the link between high
inequality and poverty in Pakistan. Because economic inequality and poverty are
interrelated concepts, they are often studied together. Theoretically, inequality tends to
constrain availability of resources at the bottom end of the population distribution
making them more likely to be poor. However, this is an empirical question which may
or may not be true in Pakistan. Some earlier studies confirm the importance of
inequality in poverty alleviation exercise [World Bank (2002), Jamal (2006)].
Our evidence suggests that significant changes in income inequality and poverty have
been observed in the country from 1990-91 to 2005-06. However, the link between
growth, inequality and poverty is not very well understood. Even though Pakistan
enjoyed low to moderate growth rates in per capita income in the 1990s, its progress
against poverty has been affected by the cyclical nature of inequality. Based on our
calculations from seven household surveys, Figure 5.1 shows that inequality was
procyclical from 1990-91 to 1998-99 as Gini coefficient increased with moderate growth
in per capita income.
Figure 5‐1: Distribution of inequality and per capita income
Rising inequality with income growth did not exacerbate the prospects of poverty
alleviation during this period because the income share of the poorest 20% remained
365

unchanged until 1996-97. This is revealed in Figure 5.2 which shows that the number of
poor people kept on decreasing from 1990-91 to 1996-97. The largest increase in poverty
took place in 1998-99 at a time when per capita GDP growth slowed down to less than
1% due to economic sanctions imposed on Pakistan following the nuclear detonation
leading to worsening of the income share of bottom 20% population falling below 8% for
the first time in eight years. Inequality became countercyclical after 1998-99 as Gini
coefficient sharply declined while per capita income rapidly increased. Falling inequality
associated with rapid income growth significantly improved the positive effects of
income growth on poverty alleviation. As shown in Figure 5.2, low inequality augmented
the progress againt poverty as the number of poor fell from close to 50 million to a little
more than 30 million, or about 40% fall in poverty. These results suggest that inequality
does matter for poverty reduction.
Figure 5‐2: Distribution of Poverty and Per Capita Income
Further evidence confirms that inequality is a constraint on poverty alleviation in
Pakistan because the progress against poverty suffers from worsening distribution of
income. This is shown in Figure 5.3 where the solid line shows poverty with actual Gini
inequality while the broken line depicts the number of poor holding Gini inequality as
constant. We project that had Pakistan maintained the same income distribution
throughout the 1990s as it was in 1990-91 (i.e., 29.79%), the increase in the number of
366

poor would have been much smaller, especially between 1993-94 and 1998-99. We note
that even small inequality changes experienced by Pakistan made considerable impact
on poverty.
Figure 5‐3: The effect of inequality on poverty in Pakistan, 1990-91 to 2005-06
5.3.2 Method of Analysis
Toward this end, we develop a framework that allows us to study these relationships
( )
y = α + β( RI × g ) + γX + η × τ + D + ε
ijt jt jt ijt k t j ijt
jt where g measures income inequality in district j at time t , while all other variables
have been defined in Section 5.2 above. The variable of interest is the interaction of a
K K
µ i= 1 j=
1
benchmark values and written as
( )
RI jt × g jt
measure of regional infrastructure with a measure of income inequality, i.e., .
As discussed in Chapter 4, regional income inequality is measured by the most
commonly referred measure known as the Gini coefficient, which provides us some good
G = 1 ∑ ∑ f( yi) f( yj) yi − yj
where i y depicts
the value of an indicator in the tth region, µ represents average value of the indicator for
367

the whole country,
f ( yi )
is the population share of the ith region in total population of
the country and K indicates the number of regions. The value of Gini-coefficient range
from 1 to 0, where 1 indicates perfect inequality, and 0 indicates perfect equality.
Because the dependent variable is a binary variable, the model would be estimated by
the probit maximum likelihood. Our assumption here is that the dependent variable can
be expressed as a linear function of other variables that affect the probability of an
εijt individual being poor or otherwise. We further assume that the error structure has a
cumulative normal distribution function where the probability of an individual falling
ijt
below the poverty line, p , can be written as
1 2
1 α + β( RI jt × g jt ) + γXijt + ( ηk× τt)
+ Dj − t
2
pijt = Pr ( yijt = 1)
= ∫
e dt
−∞ 2π
where t is a standardized normal variable. The coefficient estimates are obtained from
the maximum likelihood estimation procedure.
5.3.3 Social infrastructure and poverty: How inequality affects poverty?
The existing literature provides no evidence on how local level improvements in
provision of public schools affect rural poverty under different inequality regimes. We
take this step and empirically examine this relationship. We introduce three interaction
terms by interacting principle component post-primary school system with three
dummy variables to represent (1) low inequality districts; (2) medium inequality
districts, and (3) high inequality districts. The districts are placed into low- and highinequality/polarization
regions if they are ½ standard deviations below or above the
mean, respectively. All other districts are classified as medium inequality/polarization
regions.
Table 5.9 provides the results where the three interaction terms between principal
component post-primary school system×inequality are the variables of interest. Column
(1) presents the results when districts are segmented by the Gini index into low-,
medium- and high-inequality regions while column (2) evaluates this effect when
districts are classified into low, medium and high-inequality regions on the basis of
368

initial inequality of 1992-93. In general, the results in column (1) and (2) provide
unequivocal support to the view that investments in post-primary school system do not
produce pro poor effects in high inequality districts. The initial level of Gini inequality
has
Table 5‐7: Post‐primary school system and poverty under different inequality regimes
Probit regressions (average
marginal effects)
Explanatory variables
Gini index Starting Gini of
1992–93
(1) (2)
Principal component post‐primary school index× low inequality ‐0.008 ‐0.017***
(‐1.27)
(‐2.65)
Principal component post‐ primary school index× medium inequality ‐0.009** ‐0.005**
(‐2.33)
(‐2.39)
Principal component post‐primary school Index× high inequality ‐0.008
‐0.012
(‐1.25)
(‐0.93)
Lowly educated head, 1–5 years (yes=1, no=0) ‐0.093*** ‐0.093***
(‐11.66) (‐11.58)
Medium educated head, 6–10 years (yes=1, no=0) ‐0.159*** ‐0.158***
(‐18.53) (‐18.54)
Highly educated head, 11 years or more (yes=1, no=0) ‐0.251*** ‐0.251***
(‐20.54) (‐20.46)
Medium sized household, 4–8 members (yes=1, no=0) 0.348*** 0.349***
(31.89) (31.94)
Large sized household, 9 or more members (yes=1, no=0) 0.575*** 0.575***
(39.61) (39.15)
Household has subsistence landholding, 1–12.5 acres (yes=1, no=0) ‐0.098*** ‐0.099***
(‐7.93)
(‐8.04)
Household has economical landholding, 12.6–25 acres (yes=1, no=0) ‐0.149*** ‐0.150***
(‐7.2)
(‐7.31)
Household has large landholding, more than 26 acres (yes=1, no=0) ‐0.187*** ‐0.188***
(‐5.65)
(‐5.71)
Livestock owned (yes=1, no=0) ‐0.042*** ‐0.042***
(‐3.67)
(‐3.69)
Farm equipment owned (yes=1, no=0) ‐0.100*** ‐0.101***
(‐3.03)
(‐2.91)
Financial assets owned by household (Rs. million) ‐1.782*** ‐1.784***
(‐5.79)
(‐5.77)
Age ‐0.001*** ‐0.001***
(‐11.29) (‐11.28)
Male (yes=1, no=0) ‐0.007*** ‐0.007***
(‐3.5)
(‐3.48)
Days worked during last month ‐0.0001
‐0.0001
(‐0.35)
(‐0.36)
12 sector and employment status dummies Yes Yes
Individual level controls included Yes Yes
District level controls included No No
369

Province× year fixed effects included Yes Yes
Pseudo R 2 0.130 0.131
Number of observations 349295 349295
Notes: All regressions are estimated by probit maximum likelihood. Numbers in parenthesis are
asymptotic t-values obtained from robust standard errors adjusted for clustering at the district level.
The regressions include intercept terms, but they are not reported. **, and *** denote statistical
significance at the 5 and 1% levels, respectively. For definition of control variables, see Table 5.5.
stronger pro poor effects in low and medium inequality regions. The magnitude of the
effect suggests that a one SD increase in post-primary school system index leads to a
0.26% reduction in rural poverty in low-inequality districts as compared with 0.08%
poverty reduction in medium-inequality districts. The impact of present levels of Gini
inequality on poverty is relatively weak in our sample where low inequality districts do
not immediately benefit from these investments while the impact on medium-inequality
districts roughly remains unchanged.
5.3.4 Road infrastructure and poverty: Does inequality matter?
How investment in regional road infrastructure affects rural poverty under different
inequality regimes? We investigate the effects of changes in regional road infrastructure
on poverty by classifying districts into three categories of inequality. Low inequality
districts are those where Gini index is 0.5 standard deviations below the mean; high
inequality refer to districts where Gini index is 0.5 standard deviations above the mean
and medium inequality districts refer to all other districts. We incorporate road density
in respective districts into the model by interaction of these three dummy variables with
the road density to capture the slope differentials of road density. Thus the effect of a
change in the district level index of road density on the probability of poverty under
alternative inequality regimes is captured by these interaction terms.
Regression results shown in column (1) of Table 5.10 reveal that holding all else as
constant increase in road density significantly decreases the probability of poverty in all
the low, medium and high inequality districts. For example, a one standard deviation
increase in road density (i.e., 0.125) demonstrates highest levels of poverty reduction in
the low-inequality districts (5.8%), followed by the medium-inequality districts (3.9%)
and then the high-inequality districts (3.5%). Column (2) presents the results of increase
in road density with starting levels of inequality and shows that a one standard deviation
370

increase in road density leads to 9.3%, 4% and 4.3% reduction in poverty in low-,
medium and high-inequality districts, respectively. By implication, these results suggest
that the long run poverty reduction potential of investment in road infrastructure almost
doubles when we move from high-inequality and medium-inequality districts to lowinequality
districts. In other words, policies that support more equal income
distribution offer high gains in poverty alleviation.
Table 5‐8: Road density and poverty under different inequality and polarization regimes
Probit regressions
(average marginal effects)
Explanatory variables
Gini index Starting
Gini of
1992–93
Road density× low Gini index
(1)
-0.467***
(2)
-0.741***
Road density× medium Gini index
(-4.00)
-0.313***
(-7.77)
-0.320***
Road density× high Gini index
(-3.01)
-0.278***
(-5.19)
-0.344***
(-3.42) (-3.14)
Lowly educated head, 1–5 years (yes=1, no=0) -0.079*** -0.075***
(-7.33) (-6.52)
Medium educated head, 6–10 years (yes=1, no=0) -0.153*** -0.148***
(-11.95) (-11.61)
Highly educated head, 11 years or more (yes=1, no=0) -0.245*** -0.242***
(-9.52) (-9.67)
Medium sized household, 4–8 members (yes=1, no=0) 0.301*** 0.299***
(22.61) (22.45)
Large sized household, 9 or more members (yes=1, no=0) 0.560*** 0.558***
(35.07) (34.02)
Household has subsistence landholding, 1–12.5 acres (yes=1, -0.145*** -0.143***
no=0)
(-11.21) (-10.85)
Household has economical landholding, 12.6–25 acres (yes=1, -0.188*** -0.188***
no=0)
(-6.52) (-6.77)
Household has large landholding, more than 26 acres (yes=1, -0.212*** -0.214***
no=0)
(-3.95) (-4.2)
Livestock owned (yes=1, no=0) -0.034** -0.036***
(-2.94) (-3.24)
Farm equipment owned (yes=1, no=0) -0.092** -0.095**
(-1.97) (-2.03)
Financial assets owned by household (Rs. million) -2.34*** -2.329***
(-6) (-6.10)
Age -0.001*** -0.001***
(-8.29) (-7.78)
Male (yes=1, no=0) -0.005** -0.005*
(-1.95) (-1.86)
Days worked during last month -0.0004 -0.001
(-0.99) (-1.13)
12 sector and employment status dummies Yes Yes
371

Individual level controls included Yes Yes
District level controls included No No
Year fixed effects included Yes Yes
Pseudo R 2 0.1551 0.1592
Number of observations 139944 139944
Notes: All regressions are estimated by probit maximum likelihood. Numbers in parenthesis
are asymptotic t-values obtained from robust standard errors adjusted for clustering at the
district level. The regressions include intercept terms, but they are not reported. **, and ***
denote statistical significance at the 5 and 1% levels, respectively. For definition of control
variables, see Table 5.5.
5.4 What Factors Cause Agglomeration of Manufacturing Industries?
5.4.1 Background
What forces drive agglomeration of manufacturing industries in Pakistan? From the
literature we come to know that the benefits of agglomeration of industries are often
associated with reduction of transport costs. In this regard, Marshall (1920)
distinguishes three types of transport costs that play an important role in firm location.
These costs are associated with “moving goods”, “moving people” and “moving ideas” or
knowledge spillovers.
Firstly, the firms would like to locate near the demand centers and input suppliers to
save shipping cost. Secondly, agglomeration of industries also offers the advantage of
scale economies on account of labor market pooling due to inherent benefits of pooling,
which also allow labor to optimally allocate time to maximizes productivity. Finally,
agglomeration allows firms to gain from the free flow of ideas or technology spillovers.
In this regard, Ellison et al. (1999) describe the finance industry in urban centers where
density speeds up the flow of new ideas.99 However, these models cannot be easily
translated into variables that could be used in the empirical models trying to find out the
determinants of agglomeration.
Natural advantage is another key factor that may motivate firms to locate at particular
regions where the considerations of Marshall’s agglomeration forces may be weak or
non-existent. To illustrate, some regions offer natural environments that are suited to
certain industries due to significant natural cost advantages. For example, milk
99 Similarly, Arzaghi and Henderson (2008) draw our attention to the benefits of networking to marketing
firms in Manhattan.
372

processing industry has a natural advantage to locate in regions where milk supply is
surplus and farm gate price of milk is relatively low. Similarly, petrochemical industry
may save shipping costs by locating near a port. One may like to offer some anecdotal
examples of industries that have otherwise agglomerated in a particular region to reap
benefits of natural advantage or any of these transport costs. However, what is desirable
from an empirical standpoint is to examine the relative significance of these factors in
explaining the causes of industry agglomeration.
In this section, the goal is to analyze the factors that help explain the causes of
agglomeration of manufacturing industries in Pakistan. We use an empirical
specification that allows us to relate the Ellison-Glaser index of industry concentration
to industry characteristics and agglomeration forces put-forward by the theory.
5.4.2 Explaining agglomeration
The goal of this sub-section is to study the causes of agglomeration. The empirical
specification used is γ = α + βX+ δ + λ+ ε where γ depicts industry agglomeration index
that has been discussed in Chapter 4, X is a vector of industry characteristics that
explain agglomeration, δ and λ are for fixed year and industry effects where the
industry effects refer to 2-digit industries in each 3-digit industry, and ε is a random
error term.
We use pooled data of 3-digit EG index of Punjab based on CMI 1995-96, 2000-01 and
2005-06 (see Chapter 4 for further details). The New Economic Geography literature
views locating near demand centers and major input suppliers as a major driver behind
industry agglomeration aimed at saving transportation cost. We take district population
as an indicator of market access and district level data on road density to proxy for
market access and transportation cost. However, we expect a strong correlation between
the two. An industry would decide to locate in areas where the supply of educated and
skilled labor force is high so that they easily find industry specific skilled labor force at
market wage rates in its area of production activity. We construct formal and technical
education variables by employing external information from the Labor Force Survey.
373

We take corresponding data from Pakistan’s Labor Force Survey for 1995-96, 2000-01
and 2005-06 on 29 district level formal education variables and skill-specific variables.
We use factor component analysis to select significant principal components (for further
details on factor component analysis, see Chapter 4). After varimax rotation, we retain
four principal components using Kaiser eigenvalue criterion that accounts for 82.2%
variation in the total variance. However, 73.5% of the variance was explained by only
first two factors. Therefore, we select factor 1 and factor 2 where factor 1 account for 51%
variance and is characterized by high factor loadings on formal education (e.g., primary,
secondary, intermediate, under-graduate and graduate degrees, and professional
degrees), while factor 2 has high factor loadings on technical skills (e.g., vocational
training, technician, garment making, leather works, polishing and soldering, interior
decoration and carpentry, etc.).
Table 5.11 shows descriptive statistics on agglomeration and sources of agglomeration
variables. The district level industry agglomeration index (the EG index) is worked out
by taking the product of each industry’s agglomeration index, γ , and the industry’s share
of manufacturing in each district, i.e., γ × s
. Road density and population data are also
taken from Chapter 4. The principal component formal education and technical
education indexes have high standard deviations, which indicate spatial inequality
across district.
Table 5‐9: Descriptive statistics for agglomeration regression
Variable Mean Std. Dev. Min Max
Ellison and Glaeser index (γ) 0.0179 0.076 -0.098 1.068
Road density 0.3420 0.146 0.044 0.697
Population (millions) 3.5329 1.632 0.852 7.419
Principal component formal
education index (F1)
0.5459 1.284 -0.725 4.575
Principal component technical
0.4313 2.129 -0.815 10.381
education index (F2)
Year 2000-01 (yes=1, no=0) 0.3125 0.463 0 1
Year 2005-06 (yes=1, no=0) 0.3637 0.481 0 1
Industry 31 (yes=1, no=0) 0.1714 0 1
Industry 32 (yes=1, no=0) 0.2513 0.434 0 1
Industry 33 (yes=1, no=0) 0.0389 0.193 0 1
Industry 34 (yes=1, no=0) 0.0422 0.201 0 1
Industry 35 (yes=1, no=0) 0.1690 0.375 0 1
Industry 36 (yes=1, no=0) 0.0745 0.262 0 1
Industry 37 (yes=1, no=0) 0.0355 0.185 0 1
Industry 38 (yes=1, no=0) 0.1868 0.390 0 1
374

Industry 39 (yes=1, no=0) 0.0300 0.170 0 1
N 899 ‐‐ ‐‐ ‐‐
Note: Industry 31=food beverage and tobacco; 32= textile and leather industry; 33= wood and wood
products; 34= paper and paper products; 35= chemicals, rubber and plastic; 36= mineral products;
37=basic metal; 38 = metal products; 39= other industry & handicrafts.
Table 5.12 shows Pearson’s correlation matrix between the agglomeration index and the
sources of agglomeration. Note that simple correlation between the EG index and its
correlates is always positive. However, caution is warranted since the correlation
coefficient between road density, population and F1 is also very high, which is expected
to create robustness issues in the full regression models.
Table 5‐10: Pearson’s correlation
EG (γ)
Ellison and Glaeser index (γ) 1
Road
density
Road density 0.0827 1
Population (millions) 0.1303 0.6169 1
Principal component formal
education index (F1)
Principal component technical
education index (F2)
Populatio
n F1 F2
0.0861 0.6757 0.849 1
0.0716 -0.0537 0.2404 -0.0001 1
We estimate the model explaining the causes of industry agglomeration by using pooled
data consisting of 899 observations (Table 5.13). We present four models for different
industry characteristics: in column (1) we present the full model, in column (2) we
include only population variable, in column (3) we include only road density variable,
and in column (4) we include the two principal component variables, F1 and F2. All
models include a complete set of year and 2-digit industry fixed effects. Overall, the
empirical estimates in Table 5.12 are quite robust. Total variation explained by these
models is about 9%.
The results tell quite a consistent story. Note a declining dynamic concentration levels in
3-digit industries in Punjab, which are in line with the results reported in Tables 4.13
and 4.14. For example, in the first five year period (i.e., 1995-96 to 2000-01), the
agglomeration of manufacturing industries in Punjab shows no change as implied by a
statistically insignificant coefficient on dummy variable for 2000-01. However, in the
375

second five year period, industry agglomeration significantly declined as revealed by a
statistically significant coefficient on year
376

Table 5‐11: OLS specification for agglomeration regression
Variable Full model Model 2 Model 3 Model 4
(1) (2) (3) (4)
Road density 0.0365*
‐‐ 0.0549**
‐‐
(1.79)
(2.14)
Population (millions) 0.0046 0.0047**
‐‐ ‐‐
(0.93) (2.06)
Principal component formal -0.0029
‐‐ ‐‐ 0.0044
education index (F1)
(-0.44)
(1.40)
Principal component technical 0.00050
‐‐ ‐‐ 0.0014*
education index (F2)
(0.37)
(1.70)
Year 2000-01 (yes=1, no=0) -0.0008 -0.0016 -0.0036 -0.0001
(-0.31) (-0.62) (-1.33) (-0.05)
Year 2005-06 (yes=1, no=0) -0.0129** -0.0117** -0.0174** -0.0107**
(-2.75) (-2.86) (-2.81) (-2.07)
Industry 32 (yes=1, no=0) 0.0038 0.00380 0.0044 0.0043
(1.04) (1.08) (1.24) (1.21)
Industry 33 (yes=1, no=0) 0.0086* 0.0090* 0.0101** 0.0100**
(1.82) (1.83) (2.18) (2.13)
Industry 34 (yes=1, no=0) 0.0193 0.0194 0.02121 0.0204
(1.32) (1.34) (1.39) (1.40)
Industry 35 (yes=1, no=0) 0.0062 0.0061 0.0078 0.0073
(1.35) (1.40) (1.61) (1.60)
Industry 36 (yes=1, no=0) 0.0284** 0.0273** 0.0294** 0.0273**
(2.35) (2.26) (2.51) (2.25)
Industry 37 (yes=1, no=0) 0.0334 0.0328 0.0360* 0.0343*
(1.60) (1.60) (1.68) (1.67)
Industry 38 (yes=1, no=0) 0.0194** 0.0196** 0.0210** 0.0210**
(2.15) (2.12) (2.17) (2.11)
Industry 39 (yes=1, no=0) 0.1129 0.1134 0.1159 0.1158
(1.26) (1.26) (1.29) (1.26)
Constant -0.0179 -0.0075 -0.0080 0.0045
(-1.19) (-0.94) (-1.06) (1.29)
R2 0.093 0.0904 0.0896 0.0877
N 899 899 899 899
Note: All the models are estimated by the OLS. Numbers in parenthesis are t-values obtained
from robust standard errors corrected for clustering at the district level. ***, **, and * denote
statistical significance at the 1%, 5% and 10% levels, respectively. Industry 31=food beverage and
tobacco; 32= textile and leather industry; 33= wood and wood products; 34= paper and paper
products; 35= chemicals, rubber and plastic; 36= mineral products; 37=basic metal; 38 = metal
products; 39= other industry & handicrafts.
2005-06. This is consistent with a 33% fall in the EG index in the second five year
period (see Table 4.14). Due to high correlations between some explanatory variables,
the coefficients of population, F1 and F2 are all statistically equal to zero in the full
model, while the coefficient on road density is marginally significant. Our estimates in
model 2 imply that the size of district population increases agglomeration; in model 3
377

oads density variable is positive and statistically significant indicating that increased
road density promotes more agglomeration. The estimates using F1 and F2 in model 4
are also positive, but statistically significant only for the technical education index at the
10% level, this suggesting that formal education is not a constraint for industry
concentration in Punjab, decrease in technical education index does lead to a significant
decrease in concentration levels.
5.5 Nature of Scale Economies and Patterns of Industry Agglomeration
5.5.1 Background
While agglomeration of industry in Pakistan is widespread, the nature of scale
economies and the pattern of agglomeration in different manufacturing industries are
unclear. Our goal in this section is to examine the nature of local scale externalities
(localization versus urbanization) in major 2-digit industries by taking CMI data of
Punjab from the 1995-96, 2000-01 and 2005-06 census years.
5.5.2 Empirical specification
We estimate a value-added production function relating industry factors of production
by controlling for variables that capture local industry spillovers. To investigate these
relationships, we follow Henderson et al. (2001) to specify the following value-added
production function
( ) ( )
y = A W f k
in in in
where
yin is the value added output per production worker in ith district in nth industry;
kin is capital per worker;
f (.)
represents production technology in nth industry;
Win represents a vector of shift factors including variables for externalities associated
with localization and urbanization economies (spillover effects) and employment in each
district in respective industries. The scale of urbanization externalities is measured by
an index written as
∑
( ) ( ) 2
s N
i =
n=
1 in i − n
g ⎡⎣ E E E E ⎤⎦
378

where i and n index district and industry, respectively,
s
gi is the index of urbanization
economies in ith district, n E and E are employment in industry n, and total national
manufacturing employment, i E is employment in ith district and in E is employment in
s
g
ith district in nth industry. A lower value of i indicates urbanization and a higher value
(approaching two) suggests complete specialization of industry.
The basic econometric model is given by
where
( V t ) ( k t ) ( E t )
s
g ( t)
+ δ () t + u () t
ln ( ) = α + α ln ( ) + σ ln ( ) + κ + ρ + µ
()
in j j in j in j i n ij
in
Vin t
is value-added output per production worker in ith district in nth 3-digit
industry in each 2-digit industry,
j = 1, K , 7 kin ( t )
Ein () t
; is capital per worker; is total
employment in own industry to measure localization economies;
()
s
git measures
urbanization economies; n ρ
is a control variable sub-industry fixed effects given
α j ij
intercept term ;
µ δ ( t)
controls for district fixed-effects; and
is a dummy variable to
capture time-effects. We assume that the technology is same across all 2-digit
industries. We use three-year pooled data of CMI 1995-96, 2000-01 and 2005-06 for the
Punjab province, consisting of 901 observations from 29 districts with an average of
about 10 industries per district in each year of the census.
5.5.3 Empirical results
We estimate the model by the OLS and present basic results for seven 2-digit industries
in Table 5.14. Two industries, paper products and printing (PSIC 34) and basic metal
industry (PSIC 37), had less than 32 observations, therefore, to increase the degrees of
freedom, we merge PSIC 34 (paper and paper products) with PSIC 39 (other industry
and handicrafts), and PSIC 37 (basic metal industry) with PSIC 38 (metal products
industry). We run OLS regressions for each 2-digit industry by including in all models a
379

complete set of sub-industry (3-digit industries), year and district fixed effects. We also
run the model and report the results of the regression for all industries, where the
coefficients of all n industries are restricted to be the same which represent the average
effects across all industries. The explained variation in each OLS model is more than
70%.
Table 5‐12; Scale Externalities and Productivity in Manufacturing Industries in Punjab, 1995‐96 – 2005‐06
Variable Industr
y (31)
lnK 0.406**
*
Localization
((lnE)
Urbanization,
gi s
(4.52)
0.338**
*
(3.76)
1.284
(1.42)
Year 2000-01 0.346*
(1.84)
Year 2000-01 0.434**
(2.16)
Industr
y (32)
0.450**
*
(7.83)
0.424**
*
(7.30)
0.164
(0.22)
0.126
(0.80)
-0.162
(-1.04)
Industr
y (33)
0.301**
*
(3.95)
0.409**
*
(3.58)
-1.121
(-0.82)
-0.165
(-0.94)
-0.473
(-1.65)
Industr
y (35)
0.417**
*
(5.77)
0.385**
*
(5.06)
-0.579
(-0.38)
0.145
(0.66)
-0.386*
(-1.73)
Industr
y (36)
0.349**
(2.54)
0.288**
(2.12)
2.570
(0.88)
-0.100
(-0.34)
0.162
(0.51)
Industr
y (37&
38)
0.437**
*
(8.05)
0.365**
*
(6.29)
1.838*
(1.98)
0.135
(0.86)
0.253
(1.58)
Industr
y
(34&39
)
0.357**
*
(3.80)
0.435**
*
(3.89)
1.167
(0.61)
-0.003
(-0.10)
0.013
(0.05)
All
industri
es
0.432**
*
(16.55)
0.407**
*
(15.30)
0.628
(1.45)
0.161**
(2.04)
-0.039
(-0.48)
R 2 0.784 0.826 0.822 0.772 0.798 0.819 0.861 0.773
N 154 226 35 152 67 200 67 901
Note: Industry 31=food beverage and tobacco; 32= textile and leather industry; 33= wood and wood
products; 35= chemicals, rubber and plastic; 36= mineral products; 37 & 38 = basic metal and metal
products; 34 & 39= paper and paper products and other industry & handicrafts. All models include a
constant term and a complete set of sub-industry and district fixed effects. ***, ** and * indicates
statistically significant at the 1%, 5% and 10% levels, respectively.
The impact of externalities due to localization economies (ln E) is always positive and
always statistically significant, including the coefficient for all industries, at least at the
5% level. Since in our model we have logs on both sides, the coefficient on (ln E)
indicates that a 1% increase in own-industry employment leads to a σ -% increase in
value added output per worker. On the basis of the coefficient of all industries, our
results predict that, holding all else as constant, a 10% increase in own industry
employment increases the value-added output per worker by an estimated about 4%. At
sample means, our estimates imply that local scale economies are present in all
380

industries, but, localization economies are highest in industry 34 & 39 (paper and paper
products and other industry & handicrafts), followed by industry 32 (textile and
leather), and then industry 33 (wood and wood products).100 In general, industries
where local scale economies are present are expected to be most agglomerated and
concentrated across districts. Likewise, these industries are also most highly
agglomerated and concentrated industries in Pakistan across districts (e.g., see industry
341, 342; 392, 393; 322, 325, 321, 323; and 332 in Table 4.12). Localization economies
are lowest in industry 36 (mineral products), which is relatively least agglomerated
industry (see industry 362 in Table 4.12). Therefore, a key feature of the industrial
geography of Pakistan is that industries that offer highest local scale economies are also
the most agglomerated industries across districts. Thus, higher magnitudes of
localization economies correspond well with inter-district agglomeration and
concentration of industry, which is in line with expectations.
Based on the prevailing evidence in other countries, we expect that urbanization or
Jacob economies are likely to be highest in industries that are regarded as high-tech and
vice-versa. But, the urbanization economies are positive and statistically significant in
only one industry, i.e., basic metal and metal products industry (37 & 38). An increase in
urbanization index by one standard deviation (0.139) increases productivity in basic
metal and metal products industry by 29%. These results seem to suggest that even
though there is a greater role for inter-industry learning from technological spillovers,
basic metal and metal products industry would greatly benefit from increase in the
urbanization index.
The average share of capital in value added is 0.432 in all industries, which is not far
from the average share of 0.37 in South Korea [Henderson et al. (2001)]. The highest
share of capital is in the textile and leather industry (0.45), followed by basic metal and
metal products (0.44) and petrochemical industry (0.42). The lowest share of capital is
in wood and wood products (0.30) followed by mineral products (0.35).
100 It is possible that some of the industries may by under-reporting contractual labor to avoid negative
fall-outs of the labor laws, which may be leading to over-stating of the magnitudes of localization
economies.
381

The two time dummies reflect productivity growth, which averages at only 1.6% per
annum from 1995-96 to 2000-01 for all industry and remains stagnant in the later
period. This pattern hides high productivity industry behind the stagnant industries.
Looking at the individual industry effects, we note that average productivity growth in
food, beverage and tobacco industry over the 10-year period was 4.4% per annum.
Unfortunately, this is the only 2-digit industry where the average productivity
consistently increased over the study period. However, productivity growth was
stagnant in all other industries (see Henderson et al. (2001)].
5.6 Compendium of Policy Notes
The objective of this Section is to prepare a compendium of policy notes consisting of
various policy instruments that would be helpful in facilitating spatial aspects of
industrial development. The policy notes below build on various mapping measures of
spatial concentration and other cross-cutting questions reviewed in Chapter 4 and
Chapter 5. While the literature does not provide specific policy recommendations, but it
does provide broad guidelines and lessons that could be used to frame specific policies.
5.6.1 Policy for pro poor infrastructure investment
Pulling all the evidence together, we conclude that infrastructural investments in
Pakistan may or may not be pro poor. We demonstrate that infrastructural investments
are a key determinant of poverty reduction in Pakistan. More concern earlier with the
factors that are leading to spatial inequality would remove a fundamental constraint to
poverty alleviation. Moreover, the causes and consequences of spatial inequality in the
distribution of income also need immediate attention of the policy makers because not
only high inequality causes high poverty, but also because high inequality interacts with
infrastructure investments in a manner that hurt the welfare of the poor, as detailed
below.
First, the government emphasis on human capital accumulation and social and physical
infrastructure development as tools for poverty reduction makes sense. The econometric
evidence shows that poverty coexists with illiteracy of household heads and lack of
household asset ownership. Policies that seek universal access to education by
382

increasing the quantity and quality of schools and colleges seem to have a strong power
to reduce poverty.
Second, spatial inequality in education and health infrastructure across districts is
rampant throughout Pakistan. While we have evidence of convergence in the leading
districts, the gap between the leading and the lagging districts is increasing with the
passage of time, which should worry policy makers who are interested in bringing about
spatial equality. The mapping of districts reveals that most leading districts in education
and health infrastructure are located near the metropolitan cities while most lagging
districts are remotely located further away from major urban centers in the country.
Third, while current attention to investment in social infrastructure is clearly
appropriate, policy inattention to spatial inequality in income is very costly, especially in
more deprived districts. Our empirical analysis of the evolution of poverty on the basis
of high quality district level data on post-primary education and hospital infrastructure
index indicates that increased public sector investments on education and health
infrastructure lead to a sharp decline in poverty in all but highly unequal districts where
these investments are not associated with a decline in poverty. Policies that encourage
investment in social infrastructure by also promoting more spatial equality in the
distribution of income are likely to yield higher returns. Even though investment in
social infrastructure matters, these investments are not always pro poor as they do not
lead to uniform welfare gains.
Fourth, persistent unequal allocation of funds on road network in Pakistan also
produces spatial concentration of road infrastructure in few leading districts at the cost
of lagging districts. Again, mapping of districts shows that leading districts in road
density are also the ones where other social amenities have unequal presence. In needs
to be stressed that public policy at the federal and provincial levels is instrumental in
shaping leading and lagging districts, although it is hard to understand how this policy
is designed.
Fifth, the econometric results based on high-quality data further show that there is a
strong negative association between road density across districts and poverty incidence.
For example, our empirical results suggest that a one standard deviation increase (i.e.,
383

0.125) in road density of a district leads to around 4.4% reduction in poverty. The
magnitude of fall in poverty due to road density is substantially higher than the effect of
investment on post-primary school system and hospitals. We also find that the long run
effects of road density on poverty are far greater than the short run effects.
Finally, we conclude that income equality matters as far as public investment on
infrastructural projects is concerned. Unequivocal empirical evidence shows that the
long run poverty alleviation potential of investment on roads almost doubles when we
move from high income inequality districts to low income inequality districts. Therefore,
we reiterate that pubic policies that seek more regional equality in incomes are far more
desirable for pro poor growth policies.
5.6.2 Location policy for industrial development
We have examined the nature and scope of geographic concentration of manufacturing
industries in Pakistan by using for the first time the Ellison and Glaeser concentration
index to understand spatial inequality across districts and the causes of industry
concentrations across time and space. While there is little doubt that increasing returns
to scale associated with agglomeration externalities do exist at a wider scale in Pakistan,
it is much more difficult to identify factors that cause industry agglomeration. We also
explored how geographic concentration of manufacturing industries emerges from the
dynamic process overtime and what is the nature of agglomeration economies. Based on
our analysis, we reach on the following conclusions.
Firstly, we find that the economic geography does matter in Pakistan. The most highly
concentrated districts in manufacturing (large scale) employment are mostly clustered
around the two metropolitan cities of Karachi and Lahore, and their surrounding
districts. The only exceptions are Muzaffargarh and Swat districts where cotton ginning
& textile industries and marble and mineral products industries have natural advantage,
apart from their clusters of petrochemical and pharmaceutical and plastic industries.
Here an important policy question arises that what is the advice to policy makers if they
want to bring more spatial equality across leading and lagging districts. As Henderson et
al. (2000) succinctly put it: “whether it is more cost-effective to move people or to move
jobs.” However, despite many empirical studies on convergence and divergence in many
384

countries, we cannot offer a concrete guide on how the policy makers in Pakistan can
move to achieve the objective of faster convergence. Nor do the existing literature offer
any benchmarks of remoteness below which the chances of viable development activity
disappear.
Secondly, we find that agglomeration of 3-digit manufacturing industries is widespread
where only a small proportion of industries fall in the category of low concentration
industries. Our results suggest that 35% of the industries are highly agglomerated (EG
index > 0.05) and 38% of the industries are moderately concentrated (EG index
between 0.02 and 0.05), while only about 27% of the industries are not agglomerated.
Ship-breaking and sports and athletic goods industries are two most highly
agglomerated industries where the driving force is their natural advantage of specialized
labor, inter-industry spillovers, local transfer of knowledge and access to international
supplier and buyer networks. Other most concentrated industries represent sectors
where it is critical for the industry to spread out to reach to the final consumers or the
suppliers, e.g., furniture and fixtures, scientific instruments, pharmaceutical industry,
wearing apparel, handicrafts and office supplies, printing and publishing, pottery and
china products, paper and paper products, etc. On the other hand, the demand for least
concentrated industries is diversified across many districts due to which they have
substantial raw concentration, e.g., iron and steel, footwear and tobacco, but their
employment is distributed across few large plants, e.g., iron and steel, footwear, tobacco
and rubber.
Thirdly, we asked whether the most agglomerated industries in 2005-06 were also
agglomerated in previous years. The evidence shows that the industry concentration
dramatically fell between 1995-96 and 2005-06. The mean value of EG index remained
roughly constant from 1995-96 to 2000-01, but the index drastically fell by about 33% in
the next five years. This is further corroborated by the correlation coefficient of EG
index: the correlation in first five years (1995-96 to 2000-01) was 0.86, which fell to
0.79 in the second five years (2000-01 to 2005-06). The econometric results further
confirm this trend. Entry of more firms in an industry explains the steady decline in
concentration.
385

Fourthly, the literature tells us that the benefits of agglomeration of industries are often
associated with reduction of three types of transport costs, e.g., “moving goods”,
“moving people” and “moving ideas” or knowledge spillovers. One, the firms would like
to locate near the demand centers and input suppliers to save shipping costs. Two,
agglomeration of industries also offers the advantage of scale economies on account of
labor market pooling. And, last, agglomeration allows firms to gain from the free flow of
ideas or technology spillovers. We examine the factors that help explain the causes of
agglomeration of manufacturing industries in Pakistan. The results tell quite a
consistent story. We find that the size of district level population, increase in road
density in a district, and increase in the pool of technically trained workers in a district
all help promote agglomeration of manufacturing industries.
Finally, the determinants of industry agglomeration guide us on the causes of dense
economic activity across spatial units, and the difficulties faced in attracting
manufacturing activities in remote districts. However, a range of policy instruments
have already been tried in Pakistan as well as other developing countries, e.g., tax
holidays, building infrastructure in industrial estates, free trade zones, export
processing zones, etc. However, it goes without saying that there is no evidence on their
systematic success or failure from any country. Therefore, it is advised that more
empirical research need to be conducted to evaluate the effectiveness of the past policies
so that we are able to conclude that under what circumstances these programs and
policies are likely to succeed.
5.6.3 Agglomeration economies, scale externalities and growth of firms
Scale externalities generate localization and urbanization economies that arise from
local information spillovers with input and output markets and local technological
developments in region. Localization is when firms in a district learn from other firms in
their own industry and urbanization refers to a phenomenon when firms lean from all
firms in the district. While some urban centers dominate in Pakistan, the relative
strengths of localization versus urbanization economies have not been explored by any
other study, to the best of our knowledge. The policy implications of our investigation
are summarized below.
386

Firstly, we find that in Pakistan, as in some developed countries, localization economies
or within-industry externalities are much more important than urbanization economies,
or inter-industry spillovers, which shows that there is much less role for technological
spillovers and inter-industry learning. By implication, policy makers should be able to
make a dent on spatial inequality by focusing on “industry-specific subsidies or
infrastructural investments”.
Secondly, a key feature of the industrial geography of Pakistan is that industries that
offer highest local scale economies are also the most agglomerated industries across
districts. Thus, higher magnitudes of localization economies correspond well with interdistrict
agglomeration and concentration of industry, which is in line with expectations.
Thirdly, the urbanization economies are rarely found in only basic metal and metal
products industry where an increase in urbanization index by one standard deviation
increases productivity by 29%. These results seem to suggest that even though there is a
greater role for inter-industry learning from technological spillovers, basic metal and
metal products industry would greatly benefit from increase in the urbanization index.
Fourthly, we find that from 1995 to 2005, productivity growth in the large scale
manufacturing sector of Pakistan remained stagnant in all industries, except food,
beverage and tobacco industry where productivity increased at 4.4% per annum. In all
other industries, productivity remained stagnant during this period.
Finally, the average share of capital in value added is 43.2% in all industries, which is
not far from the average share of 37% in the manufacturing sector in South Korea. The
highest share of capital is in the textile and leather industry, followed by basic metal and
metal products and petrochemical industry. The lowest share of capital is in wood and
wood products followed by mineral products.
387

References
Ahmad, H., M. Mahmud, N. Hamid, T. Rahim (2010). A Strategy for Reversing Pakistan’s
Dismal Export Performance. Lahore School of Economic: Center for Research in
Economics and Business, Policy Paper Series No. 1.
Afzal, M. (2003). Size and Growth Rates of the Population of Pakistan: A Study Based on 1998
and Previous Censuses. In A.R. Kemal, M. Irfan and N. Mahmood (eds.). Population of
Pakistan: An Analysis of 1998 Population and Housing Census. Islamabad: Pakistan
Institute of Development Economics.
Ali, S.S., S. Tahir (1999). Dynamics of Growth, Poverty and Inequity in Pakistan, Pakistan
Development Review. 38(4), 337–358.
Altbach, Philip G., V. Selvaratnam (1989). From Dependence to Autonomy: the Development of
Asian Universities. Kluwer Academic Publishers, Norwell.
Amjad, R., A.R. Kemal (1997). Macroeconomic Policies and their Impact on Poverty Alleviation
in Pakistan. Pakistan Development Review, 36(1), 39–68.
Anwar, T. (1996). Structural Adjustment and Poverty: The Case of Pakistan. Pakistan
Development Review, 35(4), 911–926.
Anwar, T. (2005). Long-term Changes in Income Distribution in Pakistan: Evidence Based on
Consistent Series of Estimates. Centre for Research on Poverty Reduction and Income
Distribution, Islamabad.
Anwar, T., S.K. Qureshi (2002). Trends in Absolute Poverty in Pakistan: 1990 to 2001. Pakistan
Development Review, 41(4), 859–878.
Arif, G.M., H. Nazli, R. Haq (2000). Rural Non-Agriculture Employment and Poverty in
Pakistan. Pakistan Development Review, 39(4), 1089–1110.
Arthur, B. (1990). Silicon Valley Location Clusters. Mathematical Social Sciences, 235–251.
Asian Development Bank (2005). Annual Report 2004. ADB 2005.
Asian Development Bank (2008). Private Sector Assessment Pakistan. ADB 2008.
Asian Development Bank (2010). Asian Development Outlook 2010: Macroeconomic
management beyond the crises. Philippines.
Asian Productivity Organization (2009). Retrieved from .
Arzaghi, M., J.V. Henderson (2008). Networking off Madison Avenue. Review of Economic
Studies, 35(4), 1011–1038.
388

Balassa, B. (1965). Trade Liberalisation and ‘Revealed’ Comparative Advantage. The Manchester
School, 33, 99-123.
Barro, R. J., Jong-Wha Lee (2000). International Data on Educational Attainment: Updates and
Implications, manuscript. Harvard University, February.
Baran, P. A. (1957). The Political Economy of Growth. New York: Monthly Review Press.
Baum-Snow, N. (2007). Did Highways Cause Suburbanization? Quarterly Journal of
Economics, 122, 775–805.
Birdsall, N., J. L. Londono (1997). Asset Inequality Matters: An Assessment of the World Bank’s
Approach to Poverty Reduction, American Economic Review, Papers and Proceedings of
the Hundred and Fourth Annual Meeting of the American Economic Association, 32–37.
Bleaney, M. F. (1996). Macroeconomic Stability, Investment and growth in developing counties.
Journal of Development Economics, 48, 461-477.
Bruton, H. J. (1998). A Reconsideration of Import Substitution, Journal of Economic
Literature, 36(2), 903 – 936.
Burki, Abid A. (2010). Exploring the Links between Inequality, Polarization and Poverty: Does
Public Sector Infrastructural Investments Produce Uniform Spatial Effects?. Report
prepared for the South Asia Network of Economic Research Institutes (SANEI), Dhaka.
Cadot, O., Carrere, C., & Strauss-Kahn, V. (2007). Export Diversification: What’s behind the
Hump?. The Selected Works of Olivier Cadot. Available at
http://works.bepress.com/ocadot/12
Chang, Ha Joon (2003). East Asian Development Experience. In Rethinking Development
Economics. Anthem Press, London.
Chaudhry, A.A, (2009). Total Factor Productivity Growth in Pakistan: An Analysis of the
Agricultural and Manufacturing Sector. Lahore Journal of Economics, 14, 1-16.
Cheema, A. (2005). A Profile of Poverty in Pakistan. Centre for Research on Poverty Reduction
and Income Distribution, Planning Commission, Islamabad.
Chen, Y. (1996). Impact of Regional Factors on Productivity in China. Journal of Regional
Science, 36, 417–36.
Competitive Support Fund (2007). The State of Pakistan’s Competitiveness Report, Islamabad,
Competitiveness Support Fund.
Competitive Support Fund (2009). The State of Pakistan’s Competitiveness Report, Islamabad,
Competitiveness Support Fund.
389

Competitiveness Support Fund (2010), The State of Pakistan’s Competitiveness Report 2010-11,
Islamabad, Competitiveness Support Fund.
Dao, M. Q. (2008).The Impact of Investment Climate Indicators on Gross Capital Formation in
Developing Countries. Journal of Developing Areas, 42(1), 155 - 163.
Department for International Development (2010). Private Sector Development Strategy for
the Punjab. Crown Agents.
Dixit, A. K., J.E. Stiglitz (1977). Monopolistic Competition and Optimum Product Diversity.
American Economic Review, 67, 297–308.
Dollar, D., M. Hallward-Driemeier, T. Mengistae (2005). Investment Climate and Firm
Performance in Developing Economies. The World Bank, Washington, D.C.
Dorosh, P., M.N. Khan, H. Nazli (2006). A Social Accounting Matrix for Pakistan, 2001-02:
Methodology and Results. Islamabad: Pakistan Institute of Development Economics.
(2009, July 14). Dr. Ishrat Hussain predicts economic revival in 2010. Dawn News.
Retrievedfrom.
(2010, July 3). Revival of Tevta: Ministry moves summary cabinet rejected last year. Dawn
News. Retrieved from
Drine, I.. N., M. Sami (2010). Public external debt, informality and production efficiency in
developing countries. Economic Modelling, 27(2), 487-495.
Durlauf, Steven N. and Quah, Danny T. (1999). The new empirics of economic
growth, Handbook of Macroeconomics, in: J. B. Taylor & M. Woodford (ed.), Handbook
of Macroeconomics, edition 1, volume 1, chapter 4, pages 235-308 Elsevier.
Easterly, W. E. and Schmidt-Hebbel, K. (1993). Fiscal Deficits and Macroeconomic Performance
in Developing Countries. World Bank Research Observer, 8(2), 211-237.
Ellis, P. D., E. Dowell (2009). Urban Land and Housing Markets in the Punjab, Pakistan. Urban
Studies, 46(11), 2277 – 2300.
Ellison, G., E. Glaeser (1997). Geographic Concentration in U.S. Manufacturing Industries: A
Dartboard Approach, Journal of Political Economy, 105, 889–927.
Ellison. G., G. Edward, K. William (2007). What Causes Industry Agglomeration? Evidence from
Coagglomeration Patterns, NBER Working Paper 13068.
390

Energy and Infrastructure Sector Unit South Asia Region. Pakistan Transport Sector Assistance
Note June 20, 2002 (Report No. 24354 – PAK). The World Bank Group, Islamabad,
2002.
Engineering Development Board (2009-10), National Engineering Exports Strategy (NEEDS),
Islamabad, Pakistan
Bari, F., A. Cheema, E. Haque (2005). SME Development in Pakistan: Analyzing the Constraints
to Growth. Pakistan Resident Mission Working Paper Series, Working Paper 3, Asian
Development Bank, October.
Federal Bureau of Statistics (2001). Poverty in the 1990s. Statistics Division, Government of
Pakistan, Islamabad.
Federal Bureau of Statistics (2007). Pakistan Integrated Household Survey 2005-2006.
Statistics Division, Government of Pakistan, Islamabad.
Federal Bureau of Statistics. Retrieved from
Feeney, G., I. Alam (2003). Fertility, Population Growth and Accuracy of Census Enumeration
in Pakistan: 1961–1998. In A.R. Kemal, M. Irfan and N. Mahmood (eds.). Population of
Pakistan: An Analysis of 1998 Population and Housing Census. Islamabad: Pakistan
Institute of Development Economics.
Felipe, J. (2007). A Note on Competitiveness and Structural Transformation in Pakistan. ERD
Working Paper Series No. 119, Economics and Research Department, Asian
Development Bank.
Felipe, J., T. Lim (2008). An Analysis of Pakistan’s Macroeconomic Situation and Prospects.
ERD Working Paper. Manila: Asian Development Bank.
Fischer, S. (1993). The role of macroeconomic factors in growth. Journal of Monetary
Economics, 32, 485 – 512.
Fischer, S., W. Easterly (1990). The Economics of the Government Budget Constraint. World
Bank Research Observer, 5, 126-142.
Florence, P. S. (1948). Investment, Location and Size of Plant. London: Cambridge University
Press.
Fuchs, V. (1962). Changes in the Location of Manufacturing in the US since 1929. New Heaven:
Yale University Press.
Fujita, M., P. Krugman, A.J. Venables (1999). The Spatial Economy: Cities, Regions, and
International Trade. MIT Press.
Fujita, M., J.F. Thisse (2002). Economics of Agglomeration: Cities, Industrial Location, and
Regional Growth. Cambridge: Cambridge University Press.
391

Gelman, A., J.B. Carlin, H.S. Stern, D.B. Rubin (2006). Bayesian Data Analysis (2nd ed.).
Chapman and Hall.
Ghaus-Pasha, A., H. Jamal (2001). Incidence of Income Poverty in Pakistan, Research Report
No. 39, Social Policy and Development Center, Karachi.
Ghaus-Pasha, A., H. Jamal, M.A. Iqbal (2002). Dimensions of the Non-profit Sector in Pakistan
(Preliminary Estimates). Social Policy and Development Centre.
Government of Pakistan (2000). Economic Survey of Pakistan, 1999-2000, Economic Adviser’s
Wing, Finance Division, Islamabad.
Government of Pakistan (2001). Census Report of Pakistan, Statistics Division, Population
Census Organization, Islamabad.
Government of Pakistan (2003). Labor Force Survey 2001-02, Statistics Division, Islamabad.
Government of Pakistan (2004a). Economic Survey of Pakistan, 2003-2004, Economic
Adviser’s Wing, Finance Division, Islamabad.
Government of Pakistan (2004b). Labor Force Survey 2003-04, Statistics Division, Islamabad.
Government of Pakistan (2004c). Medium Term Development Framework (MTDF) 2003-10:
Report of the Working Group on Basic and College Education. Planning and
Development Division.
Government of Pakistan (2005). Towards a Prosperous Pakistan: A Strategy for Rapid
Industrial Growth. Islamabad: Ministry of Industries and Special Initiatives.
Government of Pakistan (2006a). Economic Survey of Pakistan, 2005-06, Economic Adviser’s
Wing, Finance Wing, Islamabad.
Government of Pakistan (2006b). Engineering Sector of Pakistan, Board of Investment,
Islamabad.
Government of Pakistan (2007a). Ceramics, Industry Brief, SMEDA, Lahore.
Government of Pakistan (2007b). Economic Survey of Pakistan, 2006-07, Economic Adviser’s
Wing, Finance Division, Islamabad.
Government of Pakistan (2007c). Marble & Granite Sector Competitiveness Strategy Paper,
SMEDA, Lahore.
Government of Pakistan (2007d). Pakistan Furniture Sector Strategy Narrative, Strategic
Working Group (SWOG), SMEDA, Lahore
Government of Pakistan (2007e). Strategic Plan for Pakistan’s Gems and Jewelry Industry,
Strategic Working Group (SWOG), SMEDA, Lahore
392

Government of Pakistan (2008). Labor Force Survey 2007-08, Statistics Division, Islamabad.
Government of Pakistan (2009a). Economic Survey of Pakistan, 2008-2009, Economic
Adviser’s Wing, Finance Division, Islamabad.
Government of Pakistan (2009b). National Electronic Policy 2009. Engineering Development
Board.
Government of Pakistan (2009c). National Steel Policy 2009. Engineering Development Board.
Government of Pakistan (2009d). Strategic Trade Policy Framework 2009-12. Islamabad:
Ministry of Commerce.
Government of Pakistan (2009e). The steps taken by the Government to curtail unemployment
in the country. Islamabad: Ministry of Labor and Manpower. Retrieved from
Government of Pakistan (2010a). Economic Survey of Pakistan, 2009-2010, Economic
Adviser’s Wing, Finance Division, Islamabad.
Government of Pakistan (2010b). Private sector development task force interim report.
Greene, W.H. (1997). Econometric Analysis, New Jersey: Prentice Hall International.
Gustaffson, B., and L. Shi (2004). Expenditures on Education and Health Care and Poverty in
Rural China. China Economic Review, 15, 292–301.
Hanson, G. H. (2005). Market Potential, Increasing Returns, and Geographic Concentration.
Journal of International Economics, 67, 1–24.
Haque, N. U., A.M. Husain, P.J. Montiel, P. J. (1994). An empirical "dependent economy"
model for Pakistan. World Development, 22, 1585-1597.
Haque, U. (2010). The Value Every Business Needs to Create Now - Umair Haque – Harvard
Business Review. Business Management Ideas - Harvard Business Review Blogs.
Retrievedfrom.
Hausmann, R., & D. Rodrik (2003). Economic Development as Self-Discovery. Journal of
Development Economics, 72, 603 – 633.
Hausman, R., Pritchett, L., & Rodrik, D. (2005). Growth accelerations. Journal of Economic
Growth, 10, 303-329.
Henderson, J.V., & Z. Shalizi (2000). Geography and Development. Policy Research Working
Paper No.2456, The World Bank, Washington, D.C.
393

Henderson, J. V., Todd Lee, Yung J. Lee (2001). Scale Externalities in Korea. Journal of Urban
Economics, 49, 479–504.
Henderson, J.V. (1988). Urban Development: Theory, Fact and Illusion. Oxford University
Press.
Henderson, J.V., A. Kuncoro (1996). Industrial Centralization in Indonesia. World Bank
Economic Review, 10, 513–40.
Henderson, J.V. (2002). Urbanization in Developing Countries. World Bank Research
Observer, 17, 89–112.
Henderson, J. V. (1974). The Sizes and Types of Cities. American Economic Review, 64, 640–
656.
Hirschman, A. (1958). The Strategy of Economic Development. New Haven, CT: Yale University
Press.
Hoover, E.M. (1948). The Location of Economic Activity. New York: McGraw Hill.
Hotelling, H. (1929). Stability in Competition, Economic Journal, 39, 41–57.
Institute of Public Policy (2009). State of the Economy: Emerging from the Crisis, Second
Annual Report. Lahore: Beaconhouse National University.
Jacoby, H.G., B. Minten (2009). On Measuring the Benefits of Lower Transport Costs, Journal
of Development Economics, 89, 28–38.
Jafri, S.M. Younas (1999). Assessing Poverty in Pakistan. In Mahbubul Haq Human
Development Centre and United Nations Development Program, Islamabad.
Jamal, H. (2006). Does Inequality Matter for Poverty Reduction? Evidence from Pakistan’s
Poverty Trends, Pakistan Development Review, 43(3), 439–459.
Japan International Cooperation Agency (2006). Towards a Vision 2030: Direction of Industrial
Development in Pakistan. International Development Center of Japan.
Karlsson, C. (1997). Product Development, Innovation Networks, and Agglomeration
Economies. Annals of Regional Science, 32, 235–258.
Khan, Rana Insaf Ali (2003). Spatial Distribution of Population with Special Reference to 1998
Population Census. In A.R. Kemal, M. Irfan and N. Mahmood (eds.). Population of
Pakistan: An Analysis of 1998 Population and Housing Census. Islamabad: Pakistan
Institute of Development Economics.
Khandker, S.R., Z. Bakht, G. B. Koolwal (2009). The Poverty Impact of Rural Roads: Evidence
from Bangladesh. Economic Development and Cultural Change, 57, 685–722.
Kim, S. (2008). Spatial Inequality and Economic Development: Theories, Facts and Policies.
Working Paper No.16, Commission on Growth and Development, World Bank.
394

Kim, S. (1995). Expansion of Markets and the Geographic Distribution of Economic Activities:
The Trends in U.S. Regional Manufacturing Structure, 1860–1987, Quarterly Journal of
Economics, 110, 881–908.
Kingdon, G., M. Soderbom (2007). Education, Skills, and Labor Market Outcomes: Evidence
from Pakistan. Background paper prepared for the World Bank study, “Linking
Education Policy to Labor Market Outcomes.” Washington, DC.
Klasen S. (2008). Economic Growth and Poverty Reduction: Measurement Issues using Income
and Non-income Indicators. World Development, 36 (3), 420–445.
Klinger, B., L. Daniel (2004). Discovery and Development: An empirical exploration of ‘new’
products, World Bank.
Kraay, A. (2006). When is Growth Pro-poor? Evidence from a Panel of Countries. Journal of
Development Economics, 80 (1), 198–227.
Krugman, P. (1991a). Geography and Trade. Cambridge: MIT Press.
Krugman, P. (1991b). Increasing Returns and Economic Geography, Journal of Political
Economy, 99, 483–99.
Krugman, P. (1998). What’s New about the New Economic Geography. Oxford Review of
Economic Policy, 14, 7–17.
Krugman, P. and Livas, R. E. (1996). Trade Policy and the Third World Metropolis. Journal of
Development Economics, 49, 137–50.
Krugman, P. and A.J. Venables (1995). Globalization and the Inequality of Income, Quarterly
Journal of Economics, 110, 857–80.
Lall, S. (2000). The technological structure and performance of developing country
manufactured exports, 1995-1998. Oxford Development Studies, 28 (3), 337-369.
Lall, S. V., S. Chakravorty (2005). Industrial Location and Spatial Inequality: Theory and
Evidence from India. Review of Development Economics, 9, 47–68.
Lall, S. V., Z. Shalizi, U. Deichmann, U. (2004). Agglomeration Economies and Productivity in
Indian Industry. Journal of Development Economics, 73, 643 – 673.
Lall, S., J. Weiss (2004). Industrial Competitiveness: The Challenge for Pakistan. Seminar
Paper No 2. Islamabad: Asian Development Bank Resident Mission and Asian
Development Bank Institute.
Lee, Jhong-Wha, R.J. Barro (1997). Schooling quality in a cross section of countries. National
Bureau of Economic Research.
395

Levine, Ross and Renelt, David (1992). A Sensitivity Analysis of Cross-Country Growth
Regressions, American Economic Review, American Economic Association, vol. 82(4),
pages 942-63.
Lewis, W. A. (2003). The Theory of Economic Growth. London: Taylor and Francis.
Lin, J., Ha Joon Chang (2009). Should Industrial Policy in Developing Countries Conform to
Comaparative Advantage or defy it? A debate between Justin Lin and Ha-Joon Chang.
Development Policy Review, 27(5), 1 – 20.
Manly, B. (1994). Multivariate Statistical Methods: A Primer, London: Chapman and Hall.
Marshall, A. (1890), Principles of Economics, MacMillan, London.
Mata, J., P. Portugal (2002). The Survival of New Domestic and Foreign-owned Firms. Strategic
Management Journal, 24, 323–343.
Mckinsey (2009). Still Green for Growth.
Middleton, J., A. Ziderman, A. Van Adams (1993). Skills for Productivity: vocational education
and training in developing countries. Oxford University Press, New York.
Myrdal, G. (1957), Economic Theory and Underdeveloped Regions. London, Duckworth.
Mumtaz, K., Saleem, N., Shujat, S. & Qureshi, J. (2010). Informal Economy Budget Analysis in
Pakistan and Ravi Town, Lahore. Women in Informal Employment: Globalizing and
Organizing (WIEGO). Retrieved from
National Vocational and Technical Education Commission (2009). The National Skills Strategy
2009-2013. Islamabad, Pakistan.
National Fertilizer Strategy 2010- 2025, ADL.
Nasir, Z. M., H. Nazli (2000). Education and Earnings in Pakistan. Research Report No. 177,
Pakistan Institute of Development Economics, Islamabad.
Nathan, N (2007). Relationship-Specificity, Incomplete Contracts and the Pattern of Trade.
Quarterly Journal of Economics, 122(2), 569-600.
Nurkse, R. (1953) Problems of capital formation in underdeveloped countries. Oxford
University Press.
Overman, H., A.J. Venables (2005). Cities in the Developing World. CEP Discussion Paper No
695, London School of Economics and Political Science, London.
Overseas Development Institute (2007). Policy Space: Are WTO rules preventing development?.
Briefing Paper ODI, Vol. 14. “Pakistan Country Profile", Enterprise Surveys - The World
Bank Group. Retrieved from
396

Pakistan Petrochemical Market, Lotte Pakistan.
Prebisch, R. (1950). The Economic Development of Latin America and Its Principal Problems.
New York: United Nations.
Puga, D. (1999). The Rise and fall of Regional Inequalities. European Economic Review, 43,
303–34.
Puga, D. (2002). European Regional Policies in Light of Recent Location Theories, Journal of
Economic Geography, 2, 372–406.
Qureshi, S. K., G.M. Arif (2001). Profile of Poverty in Pakistan, 1998-99. MIMAP Technical
Paper Series, No.5. Islamabad: Pakistan Institute of Development Economics.
Raheman, A., T. Afza, A. Qayyum, M.A. Bodla (2008). Estimating Total Factor Productivity and
its Components: Evidence from Major Manufacturing Industries of Pakistan. Pakistan
Development Review, 47(4).
Ravallion, M. (1997). Can High Inequality Developing Countries Escape Absolute Poverty?
Economics Letters, 56, 51–57.
Ravallion, M., G. Dutt (2002). Why Has Economic Growth Been More Pro-poor in Some States
of India than Others? Journal of Development Economics, 68, 381–400.
Ravallion, M., S. Chen (1997). What Can New Survey Data Tell Us About Recent Changes in
Distribution and Poverty? World Bank Economic Review, 11 (2), 357–382.
Ravallion, M., S. Chen (2003). Measuring pro-poor growth, Economics Letters, 78, 93–99.
Rodrik, D. (1994), Getting Intervention Right: How Korea and Taiwan grew Rich, NBER
Working paper Number 4964.
Rodrik, D. (2004). Industrial Policy in the Twenty First Century. UNIDO
Rodrik, Dani (1995). Trade Strategy, Investment, and Exports: Another Look at East
Asia, NBER Working Papers 5339.
Rosen, K. T., M. Resnick (1980). The Size Distribution of Cities: An Examination of the Pareto
Law and Primacy. Journal of Urban Economics, 8, 165–186.
Rosenstein-Rodan, P. (1943). Problems of Industrialization of Eastern and South- Eastern
Europe. Economic Journal, 53(210/211), 202-211.
Rosenthal, S. S., W.C. Strange (2004). Evidence on the Nature and Sources of Agglomeration
Economies. In J. V. Henderson and J-F. Thisse, eds. Handbook of Regional and Urban
Economics, Volume 4, Amsterdam: North Holland.
Sakakibara, M., M.E. Porter (2001). Competing at home to win abroad: evidence from Japanese
industry. Review of Economics and Statistics, 83(2), 310 – 322.
397

Shabbir, T., A.H. Khan (1991). Mincerian Earning Functions for Pakistan: A Regional Analysis.
Pakistan Economic and Social Review. 29(2).
Shah, N. A., and I. Shariff, I. (2010). Journey of Quality Management in Pakistan. Pakistan
Institute of Quality Control, 1(3).
Shamyl (2010). Energy Crunch in Pakistan: Cause, Effect & Solution. Retrieved from
.
Sharma, S. (1996). Applied Multivariate Techniques. New York: Wiley.
Singer, H. (1950). The distribution of gains between investing and borrowing countries.
American Economic Review, Papers and Proceedings, 40, 473–485.
Suryahadi, A., D. Suryadarma, S. Sumarto (2009). The Effects of Location and Sectoral
Components of Economic Growth on Poverty: Evidence from Indonesia. Journal of
Development Economics, 89, 109–117.
Taylor, A. M. (1996). On the Costs of Inward-Looking Development: Historical Perspectives on
Price Distortions, Growth and Divergence in Latin America from the 1930a to the 1980s.
NBER Working Paper 5432.
Technology Based Industrial Vision and Strategy for Socio-economic development, Pakistan
Institute of Development Economics (PIDE).
The Conference Board and Groningen Growth and Development Centre, Total Economy
Database, January 2006. Retrieved from < http://www.ggdc.net>
The International Monetary Fund (2010a). Pakistan: Poverty reduction strategy paper.
Washington, DC: International Monetary Fund.
The International Monetary Fund (2010b). Poverty Reduction Strategy Papers.
State Bank of Pakistan (2007). SBP Annual Report 2006-07. SBP Vol I. The State Bank of
Pakistan, Karachi.
State Bank of Pakistan (2010). Retrieved from
United Nations Commodity Trade Statistics Database. UN Comtrade. Retrieved from
United Nation Environment Programme (2009). Global Trends in Sustainable Energy
Investment 2009. Renewable Energy Policy Network for the 21 st Century.
United Nations Industrial Development Organization (2009). Industrial Development Report
(2009). Breaking In and Moving Up: New Industrial Challenges for the Bottom Billion
and the Middle-Income Countries. United Nations Industrial Development
Organization.
398

World Bank, Energy and Infrastructure Unit, South Asia Region (2005). Highway and Railway
Development in India and China, 1992-2002. TN-Draft March 2005.
Weber, A. (1909). Theory of the Location of Industries. University of Chicago Press, Chicago, Ill.
Wang, Miao (2009), Manufacturing FDI and economic growth: evidence from Asian economies,
Applied Economics, 41(8), 991 — 1002
Wilson. J. S., T. Otsuki (2004). Standards and Technical Regulations and Firms in Developing
Countries: New Evidence from a World Bank Technical Barriers Trade Survey.
Waskington DC: World Bank.
Wojkowska, E. (2006). Doing Justice: How Informal Systems Can Contribute. United Nations
Development Programme.
World Bank (1993). The East Asian Miracle: Economic growth and public policy. Oxford
University Press, New York: 1993.
World Bank (2002). Poverty in Pakistan: Vulnerabilities, Social Gaps, and Rural Dynamics.
Poverty Reduction and Economic Management Sector Unit, South Asia Region, Report
No. 24296-PAK
World Bank (2006). Growth and Export Competitiveness. Washington, DC: World Bank.
World Bank (2007). Transport Competitiveness in Pakistan. Islamabad: World Bank.
World Bank (2004). Doing Business Report (2005). Removing Obstacles to Growth. The World
Bank.
World Bank (2008). Doing Business 2009: Country Profile for Pakistan.
Washington, D.C : World Bank.
World Bank (2009a). Doing Business 2010: Reforming Through Difficult Times.
Washington, D.C : World Bank.
World Bank (2009b). Pakistan’s Investment Climate Laying the Foundations for Growth.
Volume 1: The Main Report. Poverty Reduction and Economic Management
Department Finance and Private Sector Unit, South Asia Department.
World Bank (2009c). The Investment Climate Report by the World Bank. The World Bank.
World Bank (2010a). Doing Business in Pakistan. Washington, D.C: The World Bank.
World Bank (2010b). Economic Growth and Structural Change in South Asia: Miracle or
Mirage. Washington, D.C: The World Bank.
World Bank (2010c). World Development Indicators 2009. Washington, D.C: The World Bank.
399

World Development Report (2000). Entering the 21st Century. The World Bank.
World Development Report (2004). Making Services Work For Poor People. The World Bank.
World Development Report (2008). Agriculture for Development. The World Bank.
World Development Report (2009). Reshaping Economic Geography. The World Bank.
World Electronics Industries 2008-13, Decisions Consulting.
Zaidi, S. A. (2005). Issues in Pakistan's Economy. Oxford University Press, Karachi.
Zaman, A. (2001). The Economics of Stateless Nations: Sovereign Debt and Popular Well-being
in Pakistan. Pakistan Development Review, 40(4), 1121 – 1134.
400