Module 7, Chapter 5 - Investment Climate

Final Report June 30, 2004Case Study: Inspectorate Reform in Latvia 1999-2003Foreign Investment Advisory ServiceAuthors: Jacqueline Coolidge, Lars Grava, Sanda Putnina1

Final Report June 30, 2004PrefaceThis background paper reviews the reform of the state inspections system in Latviafrom 1999 through 2003. The aim of the case study is to identify the political andinstitutional reform elements and the day-to-day management of such public sectorreform endeavors, and highlight the mechanisms and management tools that aretransferable and could be applied to similar reform efforts elsewhere.The analysis was based on a number of inputs, including: interviews with many of thestakeholders involved in the process; surveys of the business environment in Latvia in2001 and 2003 (including independent econometric findings prepared by a researcherin 2004 on which the authors have based a number of their conclusions); legalresearch; and materials generated by the inspectorates themselves, such as theirAnnual Public Reports and information posted on their web pages.The authors would like to thank everyone who contributed with their insights andcomments, and in particular Gregory Kisunko of the World Bank, Andris Liepins ofthe Ministry of Economy, Svetlana Proskurovska of the World Bank (formerly of theLatvian Bureau for Public Administration Reform), Janis Berzins of the Latvian StateLabor Inspectorate, Egils Harasimjuks of the Latvian State Sanitary Inspectorate,Igors Ponomarjovs of the Latvian State Fire and Rescue Service, Juris Kanels, MarisElerts, and Martins Zemitis of the Latvian Investment and Development Agency,Iveta Reinholde of Zygon Consulting (formerly of the Latvian Bureau for PublicAdministration Reform), and Aminur Rahman and Domagoj Ilic of FIAS. The studywas jointly financed by FIAS and the LIDA..The case study is structured as follows: section A provides an overview of the overallreform context, section B presents the results of the reforms and section C thendiscusses how these results and changes where achieved. This is followed byconcluding section D, where the implementation experience is assessed and lessonslearned are distilled.2

Final Report June 30, 2004TABLE OF CONTENTSAbbreviations............................................................................................................. 5Executive Summary ....................................................................................................... 6Introduction.................................................................. Error! Bookmark not defined.A. Reform Context.......................................................................................................10B. Key Reform Impact.................................................................................................261. Impact on the Ground ......................................................................................271. Changes in frequency and duration of on-site inspections. .........................282. Changes in administrative sanctions and penalties. .....................................293. Changes in the quality of on-site inspections and interaction be tween thebusiness and inspectors. .......................................................................................304. Changes in the instances of corruption reported. .........................................312. Impact on the Operating Environment for Inspectorates .................................33C. Managing the Reform Process ................................................................................14I. Reform Formulation.........................................................................................141. Overview......................................................................................................142. Initial Formulation of Reforms (January 1999 – July 1999) .......................14Steering Committee..........................................................................................15Action Plan.......................................................................................................163. Initial Implementation Experience and Refining of Reform Proposals(August 1999 – December 1999) .........................................................................164. Development of a Comprehensive Inspectorate Improvement Program(January 2000 – September 2000) .......................................................................20II. Reform Implementation. Implementation of the Inspectorate ImprovementProgram (April 2000 – June 2002) ..........................................................................211. Implementation of the CM Instruction On the Preparation Of InternalOperating Regulations In Inspections (adopted on January 18, 2000) ................212. Training for the Heads of Inspectorates and Inspectors...............................223. Provision of Information to Clients .............................................................234. Creation of the System of Performance Indicators and Internal Control inthe Inspections ......................................................................................................245. Coordination of the Work of Inspectorates..................................................24D. Assessment of the Implementation Experience: Lessons Learned in Latvia .........35References:...................................................................................................................41Annex A: Inspectorate Reform: Summary of Legislative Changes and Procedural orAdministrative Changes Undertaken by the Government and Individual Inspectorates.......................................................................................................................................43Annex B: Availability of resources..............................................................................47Annex C: “Costs and benefits” ....................................................................................483

Final Report June 30, 2004LIST OF BOXESBox 1 : A Comprehensive Government Program for Improvement of the BusinessEnvironment in Latvia .........................................................................................13Box 2 : The Investment Climate in Latvia ...................................................................26Box 3 : Better Late Than Never: The Municip al Police ..............................................25LIST OF FIGURESFigure 1. Average Number of Inspections During Last 12 Months. ...........................28Figure 2. Average duration of inspections in hours, 2001 and 2003. ..........................28Figure 3. In comparison to two years ago the quality of work done by … has… .......30Figure 4. Percent of Time that Unofficial Payments are Required, 1998....................32Figure 5. Incidence of bribery in relation to inspections, 2003. ..................................33LIST OF TABLESTable 1. Select performance indicators of State Labor Inspectorate (2001-2003). .....28Table 2. Performance indicators of State Labor Inspectorate and Fire and RescueService (1999-2003) .............................................................................................294

Final Report June 30, 2004AbbreviationsARCS Administrative and Regulatory Costs SurveyBPAR Bureau of Public Administration ReformCM Cabinet of MinistersFIAS Foreign Investment Advisory ServiceFICIL Foreign Investors’ Council in LatviaGoL Government of LatviaLDA or LIDA Latvian Development Agency (as of 2004: Latvian Investment andDevelopment Agency)PSAL Programmatic Structural Adjustment Loan5

Final Report June 30, 2004Case Study: Inspectorate Reform in Latvia, 1999 – 2003Executive SummaryLatvia began a comprehensive reform of government inspectorates in 1999, in thecontext of a broader program for removal of administrative barriers to investment. In1998, both the World Bank’s corruption survey, and the study of administrativebarriers to investment carried out by the Foreign Investment Advisory Service (FIAS)found that government inspections caused substantial problems for businesses,including disruption of their business activities, risk of possibly severe sanctions, andcorruption.Most government inspectorates tended to operate with a “control mentality,” focusingon enforcing government regulations (however arbitrary), discovering infractions, andimposing fines or other sanctions (e.g., freezing bank accounts, seizing equipment, oreven forcing a business to suspend operations).In 1999, the Government of Latvia entered into a formalized process of dialog withrepresentatives of the business community, and agreed on, inter alia , reform of themajority of national- level inspectorates as a high-priority item which was included inan Action Plan adopted by the Council of Ministers in May 1999.Subsequently, in negotiations between the Government of Latvia and the World Bankfor a new Structural Adjustment Loan, the former proposed that some items from theAction Plan, including reform of the inspectorates, be included among theconditionalities. The World Bank agreed, which served to reinforce the importance ofthis set of reforms, and also led to increased resources (both domestic andinternational) being devoted to the effort.The reform program included:• A new government “instruction” for the inspectorates to specify the rights andresponsibilities of government inspectors vis-à-vis inspectees (in particularprivate sector businesses);• Increased information available to businesses about the inspection process ingeneral and about each specific inspection in particular;• Annual (regional) meetings between inspectorates and client groups forselected inspectorates;• A requirement for written inspection reports after all inspections;• Formation of an Inspection Coordination Council;• Compulsory annual performance reports that are publicly available;• Training for inspectors in how to improve strategic focus and develop a “clientorientation” in their work;• Development of performance indicators in several key inspectorates.Most of the above reforms were implemented in 2000 and 2001. At the end of 2001, abusiness survey of administrative and regulatory costs found that most businesses no6

654321Final Report June 30, 2004longer considered government inspections to be a serious obstacle. A series of“templates” filled out by each of the main inspectorates showed that several of themwere able to supply detailed data for purposes of monitoring performance in terms ofefficiency (e.g., ratio of inspectors to inspectees). They also compile data about thenumber sanctions imposed, number of appeals, outcomes of appeals, etc.The success of the national- level inspections reform program can also be contrasted tothe lack of such a program for municipal level inspections (primarily municipal policeinspections). Survey data from the 2001 business survey showed that municipal policeinspections were still a significant source of complaints from businesses, includingharassment and extortion of bribes. The municipal police in Riga have now pledged toimplement a reform program modeled on the national program, and recent findingsindicate some progress.A second business survey carried out in 2003 showed a continuation of progress inboth qualitative and quantitative measures, as outlined below.FIGURE 1:In comparison to two years ago the quality of work done by … has …Environmental Inspectorate(N=45)-1.211Labor Inspectorate (N=138)-1.912.6Sanitary Inspectorate (N=149)-411.5Fire and Rescue Service(N=191)-2.26.1Construction Inspectorate(N=36)-4.44.4Municipal Police (N=96)-10.49.5-15 -10 -5 0 5 10 15DecreasedIncreasedSome of the more striking findings of a comparison of the 2001 and 2003 businesssurveys include the following:• The burden of inspection in terms of incidence of inspections from labor, sanitary,construction, and municipal police has significantly decreased in 2003 compared7

Final Report June 30, 2004to 2001. The correspo nding reduction in labor, sanitary, construction, andmunicipal police inspection probabilities are 10%, 7%, 4%, and 7% respectively;• There has also been a substantial reduction in hours spent in an average inspectionbetween 2001 and 2003.FIGURE 2: Average duration of inspections in hours14.012.010.08.06.04.02.0SanitaryInspectorateLaborInspectorateConstructionInspectorateEnvironmentalInspectorateFire and RescueService0.02001 2003• Interestingly, the burden of inspection is less if the firm is located in Riga vis-àviselsewhere in the country.• While there have been some improvements, municipal police still seems to exertits strong discretionary power and this is more prominent in Riga than elsewherein the country.Focus group discussions in early 2004 confirmed the survey findings, and addedthat:• participants found the professional standards of most inspectorates tohave increased over the past five years;• cash bribes associated with inspections had decreased, although insome cases it may have been replaced with more sophisticated forms,such as pressure to take on a business partner or hire someone“recommended” by an inspector.Thus we can conclude that the inspectorate reform program has been largelysuccessful so far, in terms of maintaining and improving the Government’s goals ofsafeguarding public health, safety and the environment while reducing the associatedcosts and risks imposed on businesses.These reforms are noteworthy because most transition and developing countries havefound it very difficult to implement a coordinated reform program of this natureacross a range of different agencies. In the case of Latvia, they appear to havebenefited from the following circumstances and measures:8

Final Report June 30, 2004• A consensus across most political parties to accelerate reforms in order to jointhe European Union;• Elections bringing in a new government immediately after the World Bankand FIAS reports, which was not embarrassed by the reports, was interested insolving the problems identified, and was led by the Prime Minister as“Champion” of the reform effort;• A strong, well-organized, and constructive private sector ready to contribute tothe effort of designing the reforms and cooperating with their implementationand monitoring; and• A strong and active layer of “technocrats” in government who maintainedcontinuity as governments changed, provided critical intellectual input fromwithin government, and willingly handled all the details of the reform process.Over a five year period, it is clear that the key inspectorates affecting businesses haveevolved towards a “compliance orientation”, focusing on helping their “clients” tounderstand government requirements and to live up to them. The reforms arecontinuing, with regular monitoring to assess performance and to revise the ActionPlan regularly to keep it on track and effective in delivering needed improvements.Lessons Learned: The following factors all contributed to a significant extent to thesuccess of the inspectorate reform program.A. EU Accession and other external stimuli: By 1998, the consensus of almostall participants in the political process in Latvia was accession to the EU.Secondarily, the reforms were supported by the World Bank.B. The new government elected at the end of 1998 had a fresh mandate, and noreason to feel “embarrassed” about the criticisms of the previous government.This was reinforced by an internal motivation for change in various ministries,departments and agencies, including some of the inspectorates themselves.C. There was a solid group of pro-reform “technocrats” (relatively a-politicalcivil servants) and senior- and middle- level. This cadre provided criticalcontinuity as various governments came and went. This included the LatvianDevelopment Agency (which had credibility with the business community)and the Bureau of Public Administration Reform (which had credibility withthe national government).D. The business community played an active and constructive role as “dialogpartners” and participants in the development of various Action Plans forreform.Recommendations : FIAS’ one key recommendation at this stage is the need for allinspecto rates to develop performance indicators based on the mandates of therespective institutions (e.g., reductions in workplace accidents, illnesses and fire). Allinspectorates should, as soon as possible, be able to document that they are achievingthe same or improved outcomes (e.g., number of workplace accidents) with the sameresources.9

Final Report June 30, 2004Case Study: Inspectorate Reform in Latvia, 1999 – 2003A. Reform ContextThis section summarizes some relevant aspects of the investment context and climatein Latvia, the pre-reform situation and the origin of the inspectorate reform efforts.The desire to integrate into the economic and security structures of Western Europeafter regaining independence in 1991 had created a broad consensus for reform inLatvia 1 . Early on, economic policy was based on the tenets of a liberal marketeconomy with a focus on creating the necessary institutional structures andmaintaining a consistent macroeconomic policy. Some progress had already beenachieved by 1998 when Latvia joined the World Trade Organization.The main foreign and economic policy goals of the country were driven by a desirefor accession to the European Union. The accession program led to an acceleration ofreforms in a number of areas. In addition to the already existing consensus of aliberal economic policy, a new consensus was developed for acceleratedharmonization of legislation with that of the EU and the need for a broad set of publicadministration reforms. The key implementers of these strategic goals were to begovernment technocrats, who knew they would be judged no only by the legal contentof reforms, but also by the quality of their implementation.Just as in many post-communist transition countries, in spite of all the legislativereforms, the investment climate in Latvia in 1998 was characterized by recurringproblems in the day-to-day realities of conducting business. These problems wererelated to burdensome and unmotivated visits by government inspectors, delays intransferring property rights, unclear immigration procedures for foreign citizens,unnecessarily large numbers of permits for construction approval, etc.All these area-specific problems had in common, to a greater or lesser extent, suchcross-cutting issues as inconsistent interpretation and implementation of laws andregulations, a general lack of reliable and timely recourse by enterprises, little or noexchange of information among government institutions or between the public andprivate sectors. The reforms that had been undertaken up until then (at a very fastpace) were largely structural and macro-level, geared towards setting up the overallframework and legislative bases. There had not been much time to think through andplan for the implementation of the new laws and regulations, and this was left by thewayside.Thus, by the end of 1998, the main complaint of businesses in Latvia was thatalthough the legislation was in place and of generally good quality, theimplementation and application of the legislation caused problems for conductingday-to-day business. Also, the culture of the public administration, including theinspecting institutions, in many cases still revolved around the ideas of policing anduse of administrative authority to enforce the rules, with little or no concept of the1 Latvia is a unitary state and a parliamentary democracy with a proportional electoral system. ThePrime Minister presides over a coalition government, typically formed by 3-5 parties that compose amajority in the parliament.10

Final Report June 30, 2004enterprise as “the client”, who should be encouraged and helped to comply withgovernment regulations intended to safeguard public welfare.Since the complaints by businesses were often similar and began to fit a pattern,partic ularly regarding administrative barriers, the Director of the InvestmentDepartment of the Latvian Development Agency (responsible for investmentpromotion) thought that a more comprehensive and methodical approach to reformwas necessary. As a result, in 1998 the Ministry of Finance and the LDAcommissioned a Report on Administrative Barriers to Investment in Latvia 2 , preparedby the Foreign Investment Advisory Service (FIAS) of the World Bank. The FIASReport analyzed and critically evaluated many problems related to employment,immigration, taxation, customs, purchase of real estate, construction, inspections, andother issues of importance to entrepreneurs and it became a critical catalyst indocumenting the procedures and problems and suggesting solutions.An understanding of the need to focus on the inspectorates grew out of this FIASStudy. FIAS found that “inspections were one of the most often- mentionedcomplaints of Latvia when describing the administrative environment forinvestment.” 3FIAS wrote that the activities of the inspectorates could be characterized by thefollowing 4 :• Frequent and uncoordinated checks.• Intimidating and often aggressive behavior by inspectors during on-sitevisits.• Unclear rights and obligations of inspectors and inspectees during on-siteinspection procedures.• Ineffective and non -credible appeal procedures.• Burdensome rules• excessive discretion by inspectors.• For some inspectors, the problems described above created fertile groundfor solicitation of brib es 5 .To further illustrate the status and attitudes of the pre-reform inspectorates, somepoliticians, technocrats and even inspectors used (often colorful) phrases like these todescribe the situation in the late 1990s:• “No one disturbed the inspectorates in their grazing” (politician)• “They were each a kingdom unto themselves” (technocrat)• “Each was a servant to a different master” (inspector)• “Each inspectorate had been left marinating in its own juices” (inspector)2 Research undertaken in 1998 and 1999, published and distributed in 1999.3 Also the World Bank study “Corruption in Latvia: Survey Evidence” (1998, World Bank) drewattention to the relationships between businesses and inspectors in general.4 These problems, and the relevant improvements, are further described below in the section “KeyReform Impact”.5 See, e.g., World Bank Report “Corruption in Latvia: Survey Evidence” (1998, World Bank) .11

Final Report June 30, 2004The consequences of these proble ms were a disruption of business and frequentincidents of harassment and corruption.The inspectorate reform efforts in Latvia grew out of and benefited from the systemfor development and monitoring of the implementation of the Government ActionPlan for Improving the Business Environment (see Box 1: “A ComprehensiveGovernment Program for Improvement of the Business Environment in Latvia”). Theinspectorate reform efforts that evolved through this process of dialogue with thebusiness community were later institutionalized as a separate “InspectorateImprovement Program” by the Government. 6The constellation of stakeholders in the inspectorate improvement program includedpolitical leaders, technocrats, various state institutions, non-governmentalorganizations like business associations, and international institutions, in particularthe World Bank. The core constant underlying, coordinating and sustaining theseefforts was the technocratic level staff at the Latvian Development Agency 7 (LDA)and the Bureau of Public Administration Reform 8 (BPAR). Both institutions enlisteddomestic resources like government experts and business professionals and regularlytapped into other sources, such as the political leadership, EU on-going technicalassistance or the World Bank as necessary 9 . The sections below will discuss in moredetail the role of each of these stakeholders.6 The reform efforts described in this background paper were not the first efforts at improvement.There had been some cooperation among the inspectorates, including combined visits, for example, bythe Labor Inspectorate, the Environmental Inspectorate and the Fire and Rescue Service. There hadalso been some reform programs, but mostly in the context of the line ministries (not with otherinspectorates).7 The Latvian Development Agency had been an independent legal entity wholly owned by theGovernment of Latvia from 1993 until January 2004. It was reorganized into the “Latvian Investmentand Development Agency” beginning February 2004. The Agency was and remains functionallysubordinate to the Ministry of Economy.8 The Bureau for Public Administration Reform was established in 1997. In January 2000 it wasformally merged into the Secretariat for the Special Tasks Minister for Public Administration andMunicipal Reform Issues. From May 2000 the name of the institution was the Secretariat for theSpecial Tasks Minister for State Reform. The functions of the Bureau for Public AdministrationReform were fully integrated into the Secretariats for two successive Special Tasks Ministers. For thesake of clarity, this case study will refer to the Bureau for Public Administration Reform throughout,although as indicated here, at various points in time the precise name of the institution was different.9 The approximate costs of a project like this are presented in Annex B “Availability of Resources”.12

Final Report June 30, 2004Box 1 : A Comprehensive Government Program for Improvement of the Business Environment inLatviaThe reform of the state inspections system took place in the context of broader government-wide efforts toimprove the investment climate by identifying and reducing administrative barriers to investment.Latvia has implemented an effective and sustainable mechanis m to identify and reduce administrative barriersto investment through the process of structured dialogue among stakeholders that follows a “cycle of reform”consisting of:1) systematic identification of problems;2) structured dialogue between government an d the business community to agree on the priorities for reformand draft an Action Plan;3) adoption of the Action Plan and implementation of necessary reforms by policy -makers;4) ongoing monitoring of the business environment and assessment of the impact of reforms.The targeted reforms of the business environment thatstarted in 1999 have produced tangible changes andhave also been noted by the business community and anumber of international organizations for theirinnovative nature and effectiveness 10 . The mostsignificant results since 1999 have been achieved inimproving government inspections, as well as enterpriseregistration, tax administration, customs and bordercrossing,real estate and expatriate residency.1.Identification ofproblems4. Monitoring ofimpact of reformsimplemented2. Agreement onreform prioritiesand drafting anAction Plan3. Enactment ofAction Plan,ImplementationThe key element and operational expression of the cycle of reform in general and these reforms specifically isa legal instrument: the Government-adopted Action Plan for Improving the Business Environment. TheAction Plan is an inter-ministerial planning document established first in 1999 and updated regularly, whichsummarizes:(a) the problems expressed by businesses,(b) the specific tasks to be undertaken in order to solve the problems, agreed to in structured dialoguebetween public and private sector technical experts,(c) the institutions responsible and(d) implementation criteria and deadlines.The Action Plan includes amendments to legal acts, revision and simplification of procedures, improvement ofcoordination between different institutions, preparation and publication of information as well as training ofstate and municipal officials. The Plan is a "living document,” as it is regularly updated to include new itemsand to remove those that have been implemented. By December 31, 2003, 91 of 106 tasks included in ActionPlan have been successfully implemented, which represents a very high (approximately 88 percent) successrate of reforms.10 The European Commission’s Annual Reports on Latvia’s Progress Towards Accession since 1999and onwards have consistently praised the "serious efforts of Latvia to improve the business climate".The European Commission’s Directorate General of Enterprise awarded these reform activities "CCBEST" status, meaning that the policy of Latvia in this area is an example for other EU candidatecountries for improving their competitiveness. And the World Bank – and in particular its FIASprogram – to a significant extent have based the development of its tools for reduction ofadministrative barriers on the experience gained and pilot projects undertaken in Latvia.13

Final Report June 30, 2004B. Managing the Reform ProcessThis section presents a detailed overview and analysis of the design andimplementation process of Inspectorate reform and highlights critical aspects of thestrategic and day-to-day management of such reforms.I. Reform Formulation1. OverviewThere were several waves of reform formulation, followed by implementation,assessment, and another round of reform formulation. Oftentimes the formulation andimplementation proceeded hand in hand. Like most real-life examples of successfullyimplemented reforms, this one does not follow a linear pattern, and trying to describeand analyze it in such a fashion would not do justice to the lessons that can be learned.Nevertheless, we present the reform formulation process with some chronology toreflect the multifaceted nature of such a process, inevitably reducing the full impact ofsharp turns and dead ends encountered.The formulation of the Inspectorate Improvement Program started as one componentof the effort to Reduce Administrative Barriers to Investment (which later wasrenamed the Business Environment Improvement Effort, although the focus continuedto be on administrative barriers). The foreign chambers of commerce and businessassociations like the Foreign Investors’ Council in Latvia (established in the summerof 1999) also played a role in the initial stages in formulating the problems andsuggesting solutions.Certain aspects of inspections reform were included in the Government’s Action Planto Reduce Administrative Barriers in 1999, based on the FIAS Report. However,when an opportunity arose to strengthen these efforts and find them an appropriateinstitutional champion, the inspectorate improvement tasks were incorporated in theWorld Bank’s PSAL.The following text is a more detailed review of the critical points and events in thedevelopment of the Inspectorate Improvement Program.2. Initial Formulation of Reforms (January 1999 – July 1999)To initiate an open discussion about the problems identified in the World Bank’sCorruption Report and the FIAS Report on Administrative Barriers to Investment, theLDA invited business representatives and government officials to a presentation ofthe findings in January 1999. Soon after, the Prime Minister established a SteeringCommittee to come up with solutions to the issues identified. The SteeringCommittee, using both the recommendations from FIAS and the World Bankdeveloped the first Action Plan to Improve the Business Environment in Latvia,14

Final Report June 30, 2004containing more than 30 specific activities and designating the institutions responsiblefor implementing each activity.Steering CommitteeThe Steering Committee was composed of senior civil servants from all key ministriesand state institutions affecting the investment climate and served as an umbrellainstitution whose status of direct reporting to the Prime Minister provided it with amandate to invite other state instit utions to participate as deemed necessary. It alsoincluded representatives of the private sector such as the Foreign Investors Council inLatvia (FICIL,which included the foreign chambers of commerce), the LatvianChamber of Commerce and Industry and other business associations.The tasks of the Steering Committee were not sector-specific, but covered a broadspectrum of issues in the business environment. Therefore, to carry out the evaluationof the problems and recommendations expressed in the FIAS Report and elaborate anAction Plan, four technical subcommittees were created :(1) for inspections problems,(2) for expatriate residency and work permit issues,(3) for registration, tax and customs matters, and(4) for land acquisition and construc tion issues.Upon creation of the Steering Committee, the Prime Minister also designated theBusiness Environment Improvement Unit at the LDA to serve as the Secretariat of theSteering Committee, be in charge of the daily work and later monitor the refo rmprocess.The work on the Action Plan proceeded in the four technical subcommittees, each ofwhich was headed by a member of the Business Environment Improvement Unit atthe LDA. The timeframe was extremely tight for the ambitious goals and the draftAction Plan for Removal of Administrative Barriers to Investment needed to besubmitted to the Government by May 1999.The subcommittee on inspections included a number of heads of State Inspectorates,key civil servants from some of the relevant line ministries, representatives from theBPAR and designated members of the business associations and chambers ofcommerce, including both foreign and local businesses. These were the groups thatwere involved in the formulation of the reform proposals.At the political level it was the Prime Minister who initially championed the effort.The political leadership shifted to the Minister of Economy after a change inGovernment in July 1999 and remained there with successive Ministers of Economy.The crucial support of the Prime Minister at the initial phase was explicitlydemonstrated in various ways, including opening the conference in January 1999,when the FIAS report was presented and findings discussed, issuing a decree onestablishing the Steering Committee and launching the work of the Committee byparticipating in its initial meeting. However, the inception of the idea, the strategicdirection, formulation of alternatives and day-to-day management of the overallreform effort came from the technical leve l at the LDA and later from BPAR for theinspections reform activities. This ownership of reform by the technical level, along15

Final Report June 30, 2004with continuous pressure by business associations, explains why the efforts related tothe Improvement of the Business Environme nt, including Inspectorate Reform, couldsustain 4 or 5 changes in the Government between 1999 and through 2002.The public and private sector members of the inspections subcommittee debated anddiscussed in great length the problems raised in the FIAS Report. Heated discussionsemerged between the heads of inspectorates and businesses (especially thoserepresenting the FICIL), between the heads of inspectorates and other ministries andother inspectorates as well. The meetings were prepared, chaired, facilitated andfollowed up by the representatives of the LDA. These discussions resulted inproposals for improving the work of the inspectors and became one of the four mainsections of the Action Plan, adopted by the Government in May 1999.Action PlanThe measures included in the Action Plan for Removal of Administrative Barriersrelating to inspectorate reform were the following:• To address the problem of inconsistent treatment of inspected enterprises bythe inspectors and to ensure that an administrative statement is always issuedon-site in writing, BPAR must by 1 September 1999 develop and submit to theCabinet of Ministers a draft instruction obliging the controlling institution toissue administrative statements in writing on-site or, in case of absence of anadministrative statement, to prepare, issue and place in the archives adescription of each inspection, including the name, address of the inspectedentity; the object of the inspection; the results of the inspection; detectedbreaches (with reference to relevant legislation) and indicating actions takenby the inspector.• To address the problem of the lack of information to the public and businesseson inspectorates and their functions, all the inspections are required to developtheir own internal regulations for operations, develop informational materials(on rights, appeals, legal requirements, etc.) and submit those to the BPAR byJuly 1, 1999. The BPAR was expected to assemble the information materialsand the LDA to place them in the p ublic domain.3. Initial Implementation Experience and Refining of Reform Proposals (August1999 – December 1999)During the process of gathering and analyzing the reports on the Implementation ofthe Action Plan in August of 1999, the LDA became aware that there was a need formore guidance for the inspectorates in preparing the internal operating regulations andinformation materials and the need for a more systemic approach to this task. Theinspectorates turned out not willing or able to fulfill these tasks on their own. Theirreports, which attempted to explain why the internal operating regulations had not yetbeen drafted, at best noted the need for more guidance and more typically indicatedthat all the pertinent information was already available in the laws and regulations.16

Final Report June 30, 2004At the same time, due to a change in government in July and the effect of the summerdoldrums, much of the inertia for the reform was lost and the efforts threatened to fallapart. With the departure of the Prime Minister who supported these efforts, the LDAnow had to search for new political champions to bolster the whole program forimprovement of the business environment. The logical first attempt was appealing tothe newly appointed Minister of Economy. It turned out that the interests and agendasof the two parties at this point coincided and complemented each other very well. Thereform effort was thus sustained at this stage by the energetic drive of the newMinister of Economy.The Minister of Economy had a specific agenda, initially unrelated to the efforts toreduce administrative barriers: his motivation to champion the overall businessenvironment improvement effort arose from his disapproval of the hitherto almostsingle-minded focus of the Ministry of Economy on the activities of the PrivatizationAgency. The Minister thought that a new effort was necessary to redefine the role ofthe Ministry of Economy if it truly was to be the policy institution guiding economicdevelopment with a long-term strategy (since privatization, almost by definition,would have a mandate of limited duration). This new effort would turn out to be theprogram for removal of administrative barriers to investment.The interventions and activities of the Minister of Economy in 1999 therefore servedas the political catalyst for revitalizing the reform efforts. This was the second timethat the efforts to remove administrative barriers were actively championed by apolitical leader. His tenure also marked the first successful attempts atinstitutionalizing the work on the Action Plan for Improvement of the BusinessEnvironment. Further on the political leadership played a lesser role in guiding theprocess and the reform efforts soon became mostly technocrat-driven (with ongoingmonitoring and pressure by the business community and overall backing by thesuccessive Ministers of Economy).Due to this new drive, the LDA secretariat was mandated by the Minister of Economyto organize meetings with the relevant institutions included in the Action Plan andtheir supervising ministers. The team from the Ministry of Economy included theMinister, representatives from the LDA, and representatives from businessassociations. The purpose was to hammer out the specifics of reform, agree on a newtimetable for the unimplemented measures in the revised Action Plan that was thensubmitted to the Cabinet of Ministers (CM) for adoption. As the Minister ofEconomy later explained, the reason that his fellow ministers allowed him tointervene in the activities of their subordinate institutions was that administrativebarriers, including problems with inspectorates, were not their political priorities:“Politicians don’t care about these problems.” They were aware of the problems, butthey were not painful for them politically, so there was no compelling reason to stopthe Minister’s (and LDA’s) efforts.In the context of inspectorate reform, such a meeting was held between the Ministerof Economy and State Reform Minister on September 22, 1999. The aim was torefine the measures included in the Action Plan given the feedback from theinspectorates on the difficulties they were facing with implementing the assignedtasks. An agreement was reached on the following items and is reflected in the report17

Final Report June 30, 2004to the Cabinet of Ministers on the implementation of the Action Plan, approved by theCM in October, 1999:• BPAR must prepare a draft Cabinet of Ministers Instruction OnPreparation Of Internal Operating Regulations In Inspectorates andsubmit it to the CM by November 30 (The instruction was adopted by theCM on January 18, 2000.).• BPAR must prepare a report on the appeals mechanisms in inspectoratesby November 23.• BPAR must prepare a report on the functions of inspectorates includinganalysis of overlapping functions by November 23.The first of the agreed tasks meant that the BPAR itself had to learn the problemsrelated to development of the internal operating regulations and assume a much moreactive role in guiding the inspectorates. The latte r two tasks compelled the BPAR todelve into the details of regulation of inspection activities in Latvia and analyze thework of these institutions in a comprehensive and systemic manner. Altogether, thesemeasures resulted in a process of institutional learning that helped to create the muchneeded analytical capacity at the center of government and turned out to be critical infurther design of the reform.Simultaneously, while the World Bank and the Government of Latvia werenegotiating the PSAL conditionalities in the spring of 1999, the LDA (acting throughits parent ministry, the Ministry of Economy) determined that one means to ensureimplementation of the recommendations from the FIAS Report and the Action Planwould be inclusion of the measures in the PSAL conditionalities, particularly theissues relating to reform of the inspections system and construction procedures. Thiswould both provide pressure to meet the conditionality in full and by the agreeddeadlines, as well as additional resources (PHRD grant) to assist with implementation.This initiative was supported by FIAS and by the World Bank staff working on theproject at the time, but it is critical to note that the initiative for including this topic inthe conditionality came from the Government of Latvia, not the Bank; thus theGovernment had a strong feeling of “ownership” of these reforms.The Policy Reform Program, signed by the Government as part of the PSAL 2 thatwas concluded with the World Bank in 2000, mandated the BPAR to work on theinspection issues. The PSAL and the related reform program as agreed with theWorld Bank was adopted by the Government in February 2000 11 and contained adescription of problems and suggested measures originally included in the ActionPlan for Removal of Administrative Barriers.In December 1999 FIAS, LDA and BPAR organized a seminar for the inspectorateson the process of inspectorate reform, with a specific focus on the draft Instruction OnPreparation Of Internal Operating Regulations In Inspectorates. Representatives ofFICIL, the American Chamber of Commerce in Latvia and other business associationswere asked to speak about their members’ experiences with inspectors during the firsthalf of the seminar (they did not participate in the second half). Although somemembers of the inspectorates had been present in the technical working groups under11 Cabinet of Ministers protocol of 17.02.2000, no. 8 23.§18

Final Report June 30, 2004the Steering Committee, this was the first time that the heads of all the identifiedinspectorates (total of 28) had been called together, and at times the discussionsbecame extremely tense since there was a critical mass to voice resistance to theproposed reforms. The difficulties lay in the fact that there may have beeninsufficient preparation to allow the inspectorates to understand that in terms ofprocedures carried out on-site at business premises, the inspectorates did in fact havea great deal in common, especially in terms of the burdens and risks they collectivelyimposed on businesses.Despite the difficulties, a positive result was achieved. For the first time, theinspectorates had been brought together as a group. They were challenged en massewith the problems raised by the business community and understood that many of thecomplaints from businesses are the same, regardless of which specific inspectorate isat issue. During the discussions there was a gradual understanding and acceptance bythe participating inspectors of the concept that regardless of the technical specifics oftheir field of regulation, the administrative procedures they employ on-site should besimilar, and that they could all improve their effectiveness by helping their clientscomply with requirements rather than focusing on discovering and punishing“infractions”.The seminar also gave the opportunity for inspectors to express their specific concernsabout the draft Instruction, which was a novelty in Latvia after independence. In fact,many of those involved within the GoL believed that such reforms were mostefficiently imposed from the top-down, and that discussion would just be a waste oftime. In fact, there was quite a bit of “venting” during most of the seminar. But asparticipants realized that their concerns were being taken seriously, they became moreconstructive, suggesting modifications to the reform program rather than dismissing itoutright. In response to this, some of those who had been skeptical of the participatoryapproach to the design of reforms began to change their stance, and the conceptgradually gained wider acceptance.The Instruction was adopted by the Cabinet on January 18, 2000. The Instructionrequired each inspectorate to adopt internal operating regulations based onstandardized requirements, including:• The mandate for the inspectorate’s activities.• Standardized operating procedures within the inspectorate.• Defined scope of rights and obligations of the inspectors vis-à-vis theenterprises they visit.• Standardized inspections procedures to be followed on-site.The Instruction On Preparation Of Internal Operating Regulations In Inspectorateswas never meant to be an end in itself. Instead, the goal had been to standardize onsiteinspections procedures. One means to do this was to establish standard guidelinesfor the operating regulations of each of the inspectorates. But in order to create suchstandard regulations, the circumstances in Latvia at that time may have necessitatedimposition from above. Although this was undeniably a formal and legalisticapproach, it is difficult to imagine the Latvian inspectorates – at that particular stageof their development and in light of their lack of awareness of good governanceprinciples – voluntarily complying with such an initiative, particularly based on their19

Final Report June 30, 2004pre-reform attitudes (discussed above). A welcome by-product, whether planned orinadvertent, was that they were compelled to cooperate to learn from each other theirexperiences in implementing the Instruction.4. Development of a Comprehensive Inspectorate Improvement Program(January 2000 – September 2000)In January of 2000 a series of informal meetings began among the heads of theinspectorates. These meetings were moderated by the BPAR and the LDA. Duringthis process it became clear to all stakeholders that the goal of improving the work ofthe inspectorate s required a multi-pronged approach, including legislative change,training in procedures and client-orientation, experimentation through pilot projects,cooperation among the inspectorates and the government.At this point the targeted assistance of FIAS and the WB proved to be of tremendousvalue. An external management consultant was hired to participate in the December1999 seminar with the task of preparing proposals for a work program onimprovement of the inspectorates. With inputs from LDA, BPAR, FIAS, WB, and afew heads of inspectorates, the proposals were finalized for the InspectorateImprovement Program (February 2000), including:• A training program for inspectors and heads of inspectorates, includingenlisting the support of the School of Public Administration• designing and setting up an institutional structure for coordination ofactivities and implementation of the program: the InspectorateCoordination Council• development of common inspecting procedures• clarification of mission, mandate s, and development of strategies toimprove compliance in the regulated area through better riskassessment• development of systems for assessing performance of inspectorates• development of client orientation through training• design and implementation of client information policy• implementation of internal operating regulations of inspectorates (inthe beginning in two inspectorates)Through ongoing regular informal meetings among the heads of the inspectorates todiscuss these proposals, support was generated for the establishment of theInspectorate Coordination Council and implementation of the InspectorateImprovement Program. This resulted in the formal establishment of the InspectorateCoordination Council by a decree of the State Reform Minister on April 14, 2000 12 .The members of the Council included heads of 8 key inspectorates, representatives ofthe BPAR and LDA. The Chairman of the Council was elected by the members andwould be rotated after every 6 months. With time, the meetings of the CoordinationCouncil became open to the rest of the heads of the 24 central inspectorates, andeventually the inspectorates amended the Statutes of the Coordination Council toinclude as members heads or their first deputies of all central inspectorates. This12 A new decree on the Council was issued on October 17, 2001 by another Minister for State Reform.20

Final Report June 30, 2004indicated the voluntary wish of inspectorates to be included into an association thatpresented them with the possibilities to learn and discuss substantial issues related toregulatory policy.During the first formal meeting of the Inspectorate Coordination Council on April 18,the State Inspectorate Improvement Program was adopted. The first Chairman of theCouncil was the Head of the State Labor Inspectorate, who had been a champion ofthis stage of the reform effort, and who had already initiated several successfulreforms to improve efficiency within his own agency. He was behind mobilizing thesupport of a critical number of inspectorates for participation in such a program andthrough his example demonstrated to his peers that there was real value inparticipating in the pilot programs, engaging in innovative training, etc.The implementation of the agreed Inspectorate Improvement Program was alreadyunderway when it was formally adopted by the CM on September 15, 2000. Thistypifies how the who le effort developed – bottom-up in terms of the interactionbetween the technical and political levels of government.II.Reform Implementation. Implementation of the InspectorateImprovement Program (April 2000 – June 2002)The Inspectorate Coordination Council monitored the implementation of the Programin 22 inspectorates 13 . The Council was responsible for drafting an action plan for theimplementation of the Program involving all the interested parties and responsibleinstitutions. The key components of t he program were the following:• Implementation of the CM Instruction On Preparation Of Internal OperatingRegulations In Inspectorates• Training for the Heads of Inspectorates and Inspectors• Provision of Information to Clients• Creation of a System of Perfor mance Indicators and Internal Control in theInspections• Coordination of the Work of InspectoratesThe following text is largely based on the Program, with additional comments andobservations to relate it directly to the focus of this Case Study.1. Implementation of the CM Instruction On the Preparation Of InternalOperating Regulations In Inspections (adopted on January 18, 2000)As envisaged in the original Action Plan for Improvement of the BusinessEnvironment, on the basis of this Instruction the inspectorates developed andimplemented Internal Operating Regulations, which govern inspectors’ activitiesrelated to the physical persons or legal entities to be inspected. The followinginspectorates prepared their Internal Operating Regulations:13 Later 24 – the number of inspectorates changed as new inspections related to implementation of EUpolicies have been established.21

Final Report June 30, 20041) State Quality Control Service forCrop Production;2) Plant Protection Service, SeedControl Department;3) State Pedigree BreedingInspectorate;4) Lottery and Gambling MonitoringInspectorate;5) Quality Control Inspection forMedical Care and Working AbilityExamination;6) Consumer Rights ProtectionCenter;7) Sanitary Border Inspectorate;8) State Construction Inspectorate;9) State Labor Inspectorate;10) State Railway TechnicalInspectorate;11) State Electrical Transmission andCommunication Inspectorate;12) State Pharmaceutical Inspectorate;13) State Education Inspectorate;14) State Culture Monument ProtectionInspectorate;15) State Metrology Inspectorate;16) State Inspectorate for Monitoringof Precious Metals’ Standards;17) State Sanitary Inspectorate;18) State Technical MonitoringInspectorate;19) State Fire and Rescue Service;20) State Veterinary Service;21) State Environmental Inspectorate.These regulations were crucial in establishing a unified inspection procedure and as asource of information for those inspected on the scope of inspectors’ rights, duties andresponsibilities, as well as on-site inspections procedures. This in turn facilitated acommon understanding between inspectors and businesses about the officialinspection procedure. The regulations state that one of the duties of the officer duringthe inspection is to inform the inspected person about the procedure for appeal of thedecision, or the decision imposing an administrative penalty should containinformation about the appeals procedure available. Implementation of theseregulations fostered accountability on the part of the inspectors and improved thetransparency of their actions, as well as standardized the inspection procedure itself.2. Training for the Heads of Inspectorates and InspectorsSince inspectorates are legally part of the civil service, the School of PublicAdministration could be involved in their training. However, what was unique in thecase of the Inspectorate Improvement Program was that the School agreed to tailorsome of the training courses for inspectors specifically and arrange inspectorate-onlycourses. The School also accommodated the interests of the inspectorates to conducttraining in their own premises.Altogether 492 participants attended the seminars organized within the framework ofthe Training Program for the Heads of Inspectorates and Inspectors.Training of inspectors was carried out in the following areas – on how to establish aninternal audit system, assess risks, prepare informational materials for the clients andcreate an information sys tem. Eight inspectorates organized seminars specifically onClient-oriented communication delivered by the School of Public Administration aspart of the civil servants’ training program.Two inspectorates – the State Labor Inspectorate and the State Fire and RescueService hosted a pilot training course on Strategic Planning conducted by a Swedish22

Final Report June 30, 2004management consultant financed by the WB through a PHRD grant. This trainingwas very well received by these two inspectorates, who carried on with the ideas andactually developed their strategic plans for 4-5 years, which now form the basis fortheir performance management systems. Strategic Planning helped to analyze therole and place of the inspectorate in the public administration, to identify the needsand problems of the clients and to plan activities so as to ensure the most efficientimplementation of the legislation. Following the success of the pilot training andusing the local trainer trained during the work with the above two inspectorates,seminars on Strategic Planning took place in seven other inspectorates - StateConstruction Inspectorate, Quality Control Inspection for Medical Care and WorkingAbility Examination, Consumer Rights Protection Center, State Inspectorate forMonitoring of Precio us Metals’ Standards, State Sanitary Inspectorate, StateTechnical Monitoring Inspectorate, Sanitary Border Inspectorate.3. Provision of Information to ClientsThe initial skepticism evident in 1999 was overcome and later most of theinspectorates paid ser ious attention to client information and education. At thebeginning of 2001, 16 inspectorates prepared Client Information Strategies for 2001 .The Client Information Strategies envisage provision of information to the immediateclients (physical persons and legal entities working in the sector monitored by theinspectorate), as well as the general public on the following issues:• activities of the inspectorate, main objectives, services provided, rights andduties;• requirements established in the legislation and recent changes;• rights and duties of the clients.For public information campaigns, the inspectorates mainly used mass media andpress conferences. To inform their immediate clients, inspectorates used a variety ofmedia – leaflets, informational stands in their offices, Internet home pages, meetingswith clients, provision of information on the phone or at the office. Someinspectorates (the State Construction Inspectorate, State Environmental Inspectorate,State Pharmaceutical Inspectorate) organized seminars in which current issues,requirements of the normative acts and changes in legislation referring to the sectormonitored or controlled by the inspectorate are discussed.Implementation of the client information strategies ensured that clients receiveinformation and training on legal requirements, control and appeals procedures,reducing the number of infringements due to lack of information about the normativeacts.To evaluate the impact of informative materials on the business environment and todetect what kind of information businesses need, how useful the information includedin the informative materials is, as well as how effective the dissemination policy is forthe clients, the need for client surveys was recognized. Some inspectorates alreadyhave conducted their client surveys to evaluate the performance of the inspectorate(Sanitary Border Inspectorate, Lottery and Gambling Monitoring Inspectorate). Forexample, its client survey helped the Sanitary Border Inspectorate to determine which23

Final Report June 30, 2004service provision areas require more attention, what clients would like to know andthrough which media. The Lottery and Gambling Monitoring Inspectorate used thesurvey to identify the main areas where the inspectorate could improve itsperformance. The State Labor Inspectorate used the Study on Employers’ andEmployees’ Awareness of Labor Regulations to identify the information needs of theclients, the media preferences and the level of awareness and knowledge on selectedlegal aspects of labor relations.4. Creation of the System of Performance Indicators and Internal Control in theInspectionsThe Program envisaged creation of inspectorate performance indicators and internalcontrol systems. On 25 April 2001 during the meeting of the InspectorateCoordination Council it was decided that performance indicators and annual reportsare priority areas. Taking into account that Latvian public administration institutionsdo not have the necessary experience to define performance indicators and developrelated reporting mechanisms, the LDA and the BPAR approached the World Bank toprovide guidance and assistance from international experts on this topic. The muchneeded assistance was provided and in November 2001 the School of PublicAdministration organized a seminar “Result-oriented Approach for Institutions / USAExperience” for representatives of public administration institutions. Representativesof 3 inspectorates took part (State Sanitary Inspectorate, State Labor Inspectorate,State Inspectorate for Monitoring of Precious Metals’ Standards) in the seminar. Theefforts were continued at the beginning of 2002 when a separate training program forinspectorates was prepared on how to define performance indicators.Seven inspections established internal audit divisions, which carry out evaluations ofinternal control systems and provide recommendations on improvements in order toensure effective public administration performance. In nine other inspectorates,internal audit is carried out by the internal audit unit of the respective ministry.Analysis of the public annual reports of inspectorates for 2002 indicates thatrecommendations of internal auditors for elimination of identified shortcomings anddeficiencies have been useful and most have been implemented. Internal auditors havealso developed several recommendations for improvements of the inspectionprocedures, which were accepted by the heads of inspectorates and have already orwill help to improve the performance of inspectorates.5. Coordination of the Work of InspectoratesBefore the introduction of the Inspectorate Improvement Program one of the keycomplaints was the lack of cooperation and coordination of inspectorates. Thereforesignificant attention was devoted to this issue in the course of implementation of theProgram. The Inspectorate Coordination Council facilitated coordination with otherinspectorates of similar competence and other institutions in order to improveinformation circulation and organize joint and coordinated inspections. Jointinspections are carried out by many of the inspectorates and they are regarded asuseful because they save time for the enterprises and create an integrated picture ofthe operation of the businesses for the inspectorates.24

Final Report June 30, 2004Box 2 : Better Late Than Never: The Municipal PoliceThe success of the Inspectorate Reform Program can also be demonstrated by counterfactual evidence: contrasting itwith the example of an “inspection” institution that was not initially involved in these reform efforts – the municipalpolice 14 . It is noteworthy that steps similar to those taken by other inspectorates as part of their reform program havenow also been taken by the municipal police (especially in Riga).The 2003 Self-Assessment Update of Adm inistrative Procedures on Doing Business in Latvia found that the overallevaluation of the work of the inspectorates in Latvia is favorable and they do not create significant problems forenterprises. One exception, however, was the municipal police; the report found evidence of abusive and corruptbehavior on the part of some municipal police. The 2001 ARCS yielded evidence of two clear categories of problemsassociated with the municipal police: First, they were ranked among the highest of groups of pub lic servants in theincidence of bribery, and second, they are ranked among the lowest of agencies within the public sector in terms of“quality of service”.The municipal police were found to have an unclear and often overreaching mandate and poor coordination withother institutions. The municipal police were reported to often engage in inspections-type activities, but without themandate to do so. The 2003 Self -Assessment Reports states:“In focus groups, business representatives repeatedly expressed a notably negative opinion regarding the workof the municipal police. It was mentioned that the municipal police have a very broad mandate, that thatmandate is unclear and that there is a lack of coordination in the institutional framework of the municipalpolice, as well as in terms of cooperating with state and other municipal institutions. Business representativesexpressed concern that as a result they get visited and controlled often, with the municipal police searching forinfractions, and if they cannot impose a penalty, they threaten to warn the relevant institutions that can.”These complaints were almost identical to the ones that FIAS heard in 1998 regarding state-level inspections.Therefore, the 2003 Self-Assessment Report recommended: “Other institutions may also benefit from these reformactivities [of the inspectorates]. For example, the municipal police sometimes overstep their mandate and performregulatory checks, thereby overlapping in functions with the respective inspections.” These problems identified withthe municipal police in Riga received substantial press attention and interest from the Riga City Council followingthe release of the report. Perhaps following the example of the reform activities undertaken by the inspectorates, theRiga Municipal Police reports (in a letter dated November 2003) that it has now undertaken the following reforms:• The Riga Police have signed an agreement on cooperation with the State Police.• The Riga Police have issued an instruction “On Work Organization”. The purpose of this instruction is toarticulate and standardize the rights and responsibilities of the municipal police when undertakinginspections functions.• The Riga Police have set up a database of penalties imposed.The 2003 ARCS results indicate that although the burden of inspection in terms of incidence of inspections frommunicipal police has been reduced in 2003 compared to 2001, the probability of municipal police inspection is stillabout 11% higher for a firm located in Riga than for a firm located elsewhere in the country. This frequency patternin Riga has also not changed significantly between 2001 and 2003 (table 2.2 in Annex D). The probability ofimposition of fines from the municipal police is about 32% higher for a firm in Riga than elsewhere (table 1.5 inAnnex D). Overall, the municipal police received the most polarized evaluations from businesses that were asked toassess whether “in comparison to two years ago the quality of work done by municipal police has increased,remained the same or decreased” (see Figure 3), suggesting that some positive changes are starting to occur, whilenegative aspects have not disappeared. These results and the focus group discussion still imply the presence of overlydiscretionary power of municipal police in the capital city.14 There may be a number of reasons why the Municipal Police did not participate in the inspectionsreform process – there is no central state-level institution responsible for their activities as these aremunicipal institutions; and also the Municipal Police may not have been regarded as an inspectorateearly on. Only later, with the commencement of the reform initiatives, the municipal police graduallybecame acknowledged as an institution with inspections functions.25

Final Report June 30, 2004C. Key Reform ImpactThis section will highlight what key changes have taken place as a result of theinspectorate reform efforts both on the ground and within the operating environmentfor the inspectorates.Before the discussion of the improvements that have taken place since the start of thetargeted reform initiative we will first present the summary of key problems asexperienced by businesses at the end of 1990s. At the time specific activities of theinspectorates were overwhelmingly characterized by the following 15 :• Frequent, uncoordinated and inconsistent checks. There were complaints that there waslittle coordination or exchange of information among inspectorates either at the samelevel (the central government inspectorates, for example) or different levels (centralgovernment inspectorates and regional inspectorates implementing the same mandate).There was reported to be very little consistency from inspector to inspector – eveninspections from the same inspector were reported to have different results at differenttimes.• Intimidating behavior by inspectors. Businesses expressed complaints about theaggressive and unpleasant attitudes of inspectors during on-site visits.• Nontransparent inspection procedures. There was little understanding by enterprises ofwhat were the rights and obligations of both the inspector and the enterprise itself. Often,the mandate of the inspectorate and the objectives of the visit were not clear. Few of theinspectorates required their inspectors to complete reports that would be available to thebusiness – particularly when the findings were positive.• Ineffective appeal procedures. There werecomplaints that there were few realisticmechanisms for recourse in the case ofdisagreements. The typical mechanism –appealing to the head of the inspectorate –was often unattractive to investors becauseof potential problems in the future with theinspector whose decision had been calledinto question. The alternative mechanism,the courts, was perceived to be lengthy andexpensive.• Burdensome rules. Investors complainedthat fines could be given for unimportantreasons and inspectors were reported toBox 3 : The Investment Climate in Latvia inthe 1990sIn the mid-/late 1990s, Latvia had a relativelyunfavorable reputation for its investmentclimate compared to some Central andEastern European countries. For example,Latvia was rated at “3” by the HeritageFoundation in the mid/late 1990s (where 5 isworst and 1 is best); and 2.7 on TransparencyInternational’s corruption perception index(where 0 is “very corrupt” and 10 is “highlyclean”) and a “D” rating by the EIU’s CountryRisk Ratings. As shown in the text below andin Annex C (“Costs and Benefits”), most ofthese figures have demonstrated a notableimprovement over time.have a wide latitude of what could be assessed as an infraction and what the fines shouldbe. Although the fines tended to be small, more problematic for busines ses was thenuisance factor, and the opportunity cost of a company manager spending time with aGovernment inspector.15 See FIAS. (1998). Report on Administrative Barriers to Investment in Latvia; Cabinet of Ministers.(1999). Action Plan for Improving the Business Environment; Cabinet of Ministers. (2000).Inspectorate Improvement Program.26

Final Report June 30, 2004• Administrative corruption. Because the inspectors had wide latitude and low levels ofaccountability, investors claimed that on-site inspections were often opportunities tosolicit bribes. The 1998 World Bank Corruption Report 16 presented a statistical basis forthese qualitative and anecdotal assertions and summarized the problem as follows:enterprises “suggest that there is excessive discretion, a factor known to be a cause ofcorruption.” 17Although no single instrument was designed at the outset of the reform initiatives toestablish a baseline to measure the impact of the inspectorate reforms on the businessenvironment in Latvia, various sources of information have been used to assess andquantify the practical impact of these reforms. These sources include:• 1998 World Bank Report “Corruption in Latvia: Survey Evidence”.• 1998 FIAS Report “Administrative Barriers to Investment in Latvia”.• Results of the 2001 Administrative and Regulatory Cost Survey of a representativesample of 541 businesses.• 2002-2003 Self-Assessment Report on Administrative Procedures for Doing Businessin Latvia, including on-site inspection procedure templat es completed by keyinspections as part of the Self-Assessment Report.• Interviews with the key stakeholders conducted in the fall of 2003 for this case study.• Results of the 2003 Administrative and Regulatory Cost Survey of a representativesample of 502 businesses and data analysis conducted for the purposes of this casestudy in March 2003.• Results of focus group discussions on the results of the 2003 Administrative andRegulatory Cost survey held in March and April 2004.• Statistical information and performance indicators from key inspectorates that werethe primary focus of the earliest reform process (Labor Inspectorate and Fire andRescue Service).1. Impact on the GroundThe reforms have resulted in significant changes to the way the inspectorates operateinternally and in relation to the organizations and enterprises they control. In sum, thesurvey data available to assess the pre-reform and post-reform situation indicates thatbetween 2001 (and before) and 2003 the burden of inspections on businesses has beenreduced to a significant extent 18 , including:• The burden of inspections in terms of incidence of inspections from labor,sanitary, construction, and municipal police has been significantly reduced in2003 compared to 2001 and before.• There has also been a substantial reduction in hours spent in on-siteinspections between 2001 and 2003.• The probability of fine imposition from the labor, sanitary, fire and safety andmunicipal police have reduced.• The burden of inspections has reduced significantly for small firms and tosome extent for medium firms as well.• The burden of inspections is lesser now for the firms located in the capital city(Riga) then elsewhere in the country, which may indicate that reforms have16 “Corruption in Latvia: Survey Evidence” 1998, World Bank.17 Ibid, p. 9.18 See Annex D for summary of data analysis27

Final Report June 30, 2004progressed more rapidly in the center and are spreading to the territorialoffices of the inspectorates more gradually.6. Changes in frequency and duration of on -site inspections.Figure 1. Average Number of Inspections During Last 12Months 19 .Figure 2. Average duration of inspections in hours,2001 and 2003 20 .4.54.03.53.02.52.01.51.00.50.02001 2003SanitaryInspectorateConstructionInspectorateEnvironmentalInspectorate14.012.010.08.06.0Fire and Rescue4.0ServiceInspections2.0LaborInspectorate 0.02001 2003SanitaryInspectorateLaborInspectorateConstructionInspectorateEnvironmentalInspectorateFire and RescueServiceThe survey data indicates that while the average frequency of inspections has notchanged substantially for key inspectorates (Labor and Fire) between 2001 and 2003(see Figure 1), the average duration of on-site inspections has decreased substantiallyfor all major inspections (see Figure 2).If we compare this information with the statisticsprovided by the inspectorates themselves, we see areduction in the relative number of inspected entitiesand more targeted inspections as well. For example,data from the Labor Inspectorate (see Table 1)indicates that while the number of enterprises underits supervision in 2003 has increased by 17% ascompared to 2001, the number of actually inspectedentities in the same period has increased by only 5%.This is consistent with the survey findings discussedabove. It is also noteworthy that at the same time thenumber of inspectors has increased by only 2% (from96 to 98 inspectors). The number of repeatinspections is also fairly small – at 6.7 % for 2002and has decreased by 3% compared to 2001.Table 1. Select performance indicators ofState Labor Inspectorate (2001-2003).Select Indicators % changes2001 vs. 2003# of inspectors + 2%# of identified organizationsand enterprises undersupervision# of employees in thesupervised organizations andenterprises# of inspected organizationsand enterprises# of organizations andenterprises inspected more thanonce a year+ 17%+ 25%+ 5%-3% 21In interviews, the heads of inspectorates indicated their belief that the implementationof better “risk targeting” methods accounts for these changes. There is also a general19 Based on the results of the Administrative and Regulatory Cost Surveys in 2001 and 2003.20 Based on the results of the Administrative and Regulatory Cost Surveys in 2001 and 2003.21 Data for 2001 vs. 2002.28

Final Report June 30, 2004understanding that the overall public health and workplace safety situation has notdeteriorated because of such changes in the frequency and duration of on-siteinspections.Another factor that accounts for these measurable changes is a qualitativeimprovement in the professionalism and competence of inspectors. In addition to theinformation provided by inspectorates indicating that training of inspectors has beengiven much more attention than previously, this assertion was also confirmed in focusgroup discussions with businesses held in March 2004. Businesses noted thefollowing: “The length of the inspection depends on the qualifications of inspectors –if they know where to look and what to look for, the inspection is shorter. Theinspectors have become more professional and they do not look just for anything buthave developed their routines and standard procedures.”7. Changes in administrative sanctions and penalties.The practice and procedures for application of fines was one of the focuses of thereform efforts. One aspect of this was the development and implementation of acompliance-oriented approach in the inspectorates (at least to some extent), so that theimposition of sanctions is no longer seen as an end in itself, but as one of the meansfor achieving compliance with and enforcement of safety regulations (and, crucially,one not to be applied without consideration of alternatives). Another aspect wasimprovement of operational procedures for handling decisions on application of fines– so that they are properly registered and recorded.As shown by the results of business surveys 26 theprobability of imposition of fines from thefollowing inspectorates have been reduced by thefollowing amounts between 2001 and 2003:• Labor Inspectorate - 84%,• Sanitary Inspectorate - 82%,• Fire and Rescue Service - 87%, and• the Municipal Police - 73%The focus group discussions provided more detailon these figures. Businesses indicated that “theinspectorates are no longer penalizing for littlemistakes, they are just writing or telling that theseproblems have to be averted.” These conclusionsare also supported by the performance data of theState Labor Inspectorate (see Table 2), which reveals that the number ofadministrative penalties awarded has decreased by 23% from 2001 to 2003.Similarly, the number of warnings issued for suspension of activities has decreased22 Data for 2001 vs. 2002.23 Data for 2001 vs. 2002.24 Data for 2001 vs. 2003.25 Data for 2 001 vs. 2003.26 See Annex D for summary of data analysisTable 2. Performance indicators of State LaborInspectorate and Fire and Rescue Service (1999-2003)Select Indicators# of administrativepenalties awardedAmounts ofadministrativepenalties, Lats# of issued warningsfor suspension ofactivities / equipment# of suspendedequipmentFire andRescueService,(% change1999 vs. 2003(forec.))LaborInspectorate(% changes2001 vs. 2002and whereavailable 2003)-12% -23% 223% -7% 23-52% -22% 24N/A -44% 2529

654321Final Report June 30, 2004substantially (by 22%), as has the number of cases of suspended equipment (44%).The situation with the Fire and Rescue Service follows the same pattern.8. Changes in the quality of on -site inspections and interaction between thebusiness and inspectors.Complaints about intimidating behavior, such as aggressive and unpleasant attitudes,by inspectors during on-site inspections and lack of transparency in inspectingprocedures were major concerns described in the 1998 FIAS Report. As discussedabove there are clear indications that the quality and professionalism of inspectors hasimproved (see also Figure 3), as well as the internal operating procedures.Figure 3. In comparison to two years ago the quality of work done by … has… 27 (in %)Environmental Inspectorate(N=45)-1.211Labor Inspectorate (N=138)-1.912.6Sanitary Inspectorate (N=149)-411.5Fire and Rescue Service(N=191)-2.26.1Construction Inspectorate(N=36)-4.44.4Municipal Police (N=96)-10.49.5-15 -10 -5 0 5 10 15DecreasedIncreasedNote: answers for “Don’t know/Hard to say” and “Remained the same” not shown.Some inspectorates received very high evaluations during the business focus groupstesting these findings, such as “the best inspectorate, in my view, is the Fire andRescue Service; they are the most organized and they know what they want.” At thesame time both the survey analysis and the focus group discussions confirmed that“there are differences between different regions, the inspections have improved inRiga, but not so much in the countryside.”Survey results regarding the quality of service provided by various inspectorates areprovided below:27 Based on the results of the Administrative and Regulatory Cost Survey in 2003.30

Final Report June 30, 2004Now, please evaluate the aspect:responsiveness to requests for advice, explanation, clarifications ofinspectors. Was it . . .?(Base = enterprises, that have been visited by respective inspection)* -the number of respondents is not large (N

Final Report June 30, 2004• Sanitary inspections: about 12%• Fire inspections: about 12%• Inspection by tax authorities or audits: about 8%Figure 4. Percent of Time that Unofficial Payments are Required, 1998.Source: WB “Corruption in Latvia: Survey Evidence”, 1998.Although the data are not directly comparable, some inferences about the trends since1998 can be made looking at the most recent survey data from the ARCS of 2003 29 .The figure below indicates that the incidence of corruption has decreased as regardsthe inspections. In particular, comparable data from the 2001 ARCS sho ws a markedreduction in reported bribes associated with municipal police inspections, from about9% in 2001 to just over 2% in 2003.29 It should be noted that the data from the two sources is not directly comparable due to differentmethodologies, different samples, different phrasing of questions. Nevertheless, with these caveats inmind, a comparison between the two results indicates that there has been improvement.32

Final Report June 30, 2004Figure 5. Incidence of bribery in relation to inspections, 2003In the last year, did you give a gift or pay a bribe for …?(Base = enterprises, that have been visited by respective inspection)* -the number of respondents is not large (N

Final Report June 30, 2004The following key improvements in the inspections system have taken place as aresult of the reform efforts (for more detail on some of these see Annex A:“Inspectorate Reform: summary of legislative changes and procedural oradministrative changes undertaken by the government and individual inspectorates”):Establishment andimprovement of keymanagement systemsand approachesImprovements inadministrativeculture of dealingwith the clients(inspectees)Gradual shift offocus frompunishment forviolations tocompliance basedactivities• Clarification of mandateand documentation ofinternal processes andprocedures• Standardization ofoperating procedures andforms across inspectorates• Application ofperformance measures• Creation of a forum forthe regular exchange ofinformation amonginspectorates• Application ofstandardized inspectionsprocedures to be followedon-site• Clearly defined scope ofrights and obligations ofthe inspectors vis-à-visthe enterprises they visit• Implementation of clientsurveys and feedbackforms• Increase in openness andtransparency ofinspectorates as well astheir receptiveness forchange• Focus on compliance basedactivities aimed at ensuringmeaningful observance ofhealth and safety rules• Ensuring there is an increasedprovision of constructiveadvice to inspectees• Implementation of strategicand work time planning 30 ;improved targeting ofinspections related to criticalcompliance risks• Application of performanceindicators and performanceassessments to align the goalsand the implementation.Increased information sharingand coordinated inspectionplanning to reduce overalltime that businesses spendwith government officials• Extensive training incross cuttingadministrative procedureissues and clientorientation• Overall burden of inspectionsis decreasing and inspectionsare more targeted, number ofpenalties has decreasedsubstantially. At the same time• Initiation of strategicplanning and structuralchanges to implement itthe number of accidents andthe general health and safetysituation has not deteriorated(for details see Section C).⇒ Increased satisfaction of inspectors with the work environment (except wages!)⇒ Increased compliance of business⇒ Improved performance of inspectorates (supervising a steadily increasing pool ofbusiness with the same human resources)30 For example State Labor Inspectorate assigns ~30% of inspectors’ time for advisory activities.34

Final Report June 30, 2004D. Assessment of the Implementation Experience: Lessons Learned in LatviaThe implementation experience of the inspectorate reform effort clearly indicates theneed for flexible approach, involvement of wide range of stakeholders and the abilityto assess the interim progress at various stages and reformulate and reshape thereform measures as needed.The following factors all contributed to a significant extent to the success of theinspectorate reform program. Some factors may be specific to Latvia, others may bebroadly applicable to other cases.1. EU accessionAs noted above, by 1998 the consensus of almost all participants in the politicalprocess in Latvia was accession to the European Union. At tha t point, the EU wasrelegating Latvia to the presumed “second wave” of EU accession countries (alongwith Lithuania, Slovakia, Bulgaria and Romania), while Estonia was included in thepresumed “first wave.” The body politic of Latvia was determined to catch up. Alongwith this objective came a number of underlying assumptions about improving theinvestment climate, fixing legislation and eradicating corruption that would serve asvirtually indisputable arguments for further administrative reform.2. New governmentAfter Parliamentary elections in October 1998, a new government took office inJanuary 1999. The new government did not feel embarrassed by the criticisms raisedin the FIAS Report. It was willing to both listen to investors as well as address theshortcomings and deficiencies of the previous government. Therefore, the timing ofthe election cycle proved to be a crucial factor for initiating the reform efforts.However, frequent subsequent changes in government meant that for the reformefforts to continue, they needed to be supported (and preserved) at the technocraticlevel, particularly since the coalition governments had fairly short life spans (since1993 there have been 11 governments). The time period reviewed in this study –1998 to 2003 – was no exception. It should also be noted that up until 2002 there wasno real change in the overall political leadership, as more or less the same parties(sharing similar, if not identical, political platforms) formed successive coalitiongovernments and ministers rotated from one sectoral ministry to another. This provedto be an important factor in ensuring political support for the inspectorate reformprocess.3. Critical role for technical staff within governmentWhile there were frequent changes in the Latvian government, the civil serviceremained relatively stable (there was a relatively high non-political level – ministrystate secretaries were for the most part the highest levels of career civil servants, andonly ministers and their parliamentary secretaries were political). The role of the civilservants and the technocrats inevitably increases in such a system, since they are theones that ensure continuity and smooth functioning of the system at a time of change35

Final Report June 30, 2004in government. This meant that the reform process developed a technocrat-driven,bottom- up approach, almost regardless of the political changes taking place above.4. A central institution responsible for management of administrative reformFor the inspectorate reform to be implemented, there was a need for an institutionwithin the government that could own this reform as a part of wider administrativereform portfolio. The status of the Bureau of Public Administration Reform reportingto the Minister of Public Administration and Local Government Reform was wellrecognized and respected by the inspectorates. This created the necessary conditionsfor the BPAR to exercise leadership in inspectorate reforms. This role was criticalespecially in the early stages of designing the reform and setting it in motion. Furtherdown the road the BPAR acted as a coordinator and resource center supporting theactivities of the Action Plan for Improvement of the Business Environment.5. Participation of businessesWhile all of the above factors were government-related, there was another significantsource of pressure in the reform process – complaints and proposals from businessesthat were channeled through the business associations and the Latvian DevelopmentAgency. The input of businesses developed into a crucial control mechanism to keepconstant pressure on the government, and the business community (in particular,certain business associations such as the Foreign Investors’ Council of Latvia) alsobecame a very active constituency to which the government was obliged to report.This relationship was in fact designed into the reform process through the Cycle ofReform and the Action Plan (described in Box 1 “A Comprehensive GovernmentProgram for Improvement of the Business Environment in Latvia”).6. Structured Dialogue: public -private sector and intra- governmentalThe inputs of the business community proved to be most effective when channeledthrough a structured dialogue with the relevant government counterparts. In fact,since 1999 the essence of the reform process to reduce administrative barriers toinvestment in Latvia has been a structured dialogue among the stakeholders. In manycases, this meant dialogue between public officials and the private sector experts onissue-specific topics. The LDA, with a strong mandate to improve the investmentclimate, played the role of “honest broker” between the government and the businesscommunity, and earned the trust of both sides.Another crucial aspect of “structured dialogue” is discussion among governmentrepresentatives from different institutions. A lack of dialogue among theinspectorates was a key deficiency and its introduction was the necessary step tobegin to resolve the problems of the inspectorates, particularly in overcoming theperceived lack of commonalities among the inspectorates. At the outset, manyinspectorates understandably argued that they implement very different substantivelaws on sanitary standards, work safety, fire safety etc. The intra-governmentaldialogue, fostered through the efforts of the BPAR and through the forum of theInspectorate Coordination Council, instead created an awareness among the inspectorsthat certain aspects of their activities – in particular procedural behavior on-site and36

Final Report June 30, 2004the rights and obligations of inspector vis-à-vis inspectee (the business) – werecommon to all inspectorates.7. External motivation: World Bank PSAL conditionalitiesThe World Bank’s PSAL conditionalities regarding inspectorate reform – which wereincluded in the policy reform matrix at the suggestion of the Government of Latviaitself – were an important external pressure that helped get the BPAR activelyengaged in guiding the process of reform. Once on board, the existence of PSALconditionalities helped the BPAR convince the inspectorates and other institutions ofthe importance of this effort.8. Flexibility to adopt a variety approaches within the overall reform process,based on prevailing circumstancesThere were times when a formal and “legalistic” approach was necessary to open thedoor for subsequent reform steps, since such an approach was better understood andaccepted by civil servants. By means of formal legal documents, such as theGovernment Instruction issued by the Cabinet and the Operating Regulationssubsequently prepared by the inspectorates themselves, the first critical steps towardinstitutionalizing the evaluation and reform process were achieved.At the same time, it was the combination of home- grown ideas and external advicethat produced the mix of reform proposals that ultimately satisfied stakeholders, weresuccessfully implemented and produced positive results. The flexibility to assessvarious alternatives, the ability to leverage resources and adapt the inputs of var iousstakeholders were essential in the process of formulating reform measures andensuring their implementation. It was necessary to maintain communication channelsand hold discussions with various stakeholders at various points in time – otherministries, the inspectorates themselves, businesses.9. Internal motivation for changeA crucial element leading to the success of the reform at the implementation stagewas the motivation of the heads of the inspectorates themselves for change. Thismotivation was hard fought for and gained only after lengthy discussions and debateson the underlying problems and ways forward. It proved to be necessary to enlist theinitial support of one or two heads of inspectorates who helped to “break the ice” andconduct the first meetings on a neutral topic like implementation of internal auditsystems in government institutions. Supporting the champions from within theinspectorates and allowing them to undertake “formal” leadership in coordination ofreforms (although supported by the BPAR) acted as an entry point to engaging othermore skeptical inspectorates.In fact, due to the support of its head, the Labor Inspectorate played an especiallyimportant role as a pilot and a sounding board for BPAR and LDA in designing andimplementing the reform.37

Final Report June 30, 2004The motivation also partly came from an appreciation on the part of inspectorates ofthe attention paid to improvement of their work from the center of government. Andpartly, the training opportunities were significantly increased due to the inclusion ofthe School of Public Administration as a responsible institution for realization of theInspectorate Improvement Program.As pointed out by the heads of some inspectorates in subsequent interviews, theirmotivation center ed on the following:• awareness of the changing external environment, including EU accession;need to oversee implementation of a whole new set of standards and legal norms,including a shift away from an “enforcement orientation” (focusing on numbers ofinspections, numbers of infractions found, revenues from fines received, etc.) to a“compliance orientation” focused on stimulating the enterprises to take moreresponsibility themselves to implement the better defined health, labor safety, foodand hygiene standards for their own benefit and stimulating enterprise to implementfunctioning self-assessment and internal control systems for various safety relatedrisks.• the reform program empowered them, provided training and supplied the skillsneeded to respond to a changing business environment.10. Start simpleTo introduce change, it was necessary to start out first with simple managementtechniques – internal operating guidelines, a requirement to always document the onsiteinspection procedure regardless of the outcome, design of informational materialsand their distribution. Over time, demand grew for more input and assistance inapplying increasingly complex management systems as tools for improving theperformance of the inspectorates. The inspectorates now have developed strategicplans, have started introducing performance management systems and a few are in theprocess of implementing Quality Management Systems based on ISO 9000 seriesstandards.11. “Winners and losers”An additional broad topic to be covered in assessing the implementation experience inLatvia is the general issue of “winners” and “losers” in reform programs – those whoultimately benefited versus those who did not. As the 1997 World DevelopmentReport has noted, “resistance to reform is often triggered by the potentialredistribution of resources among different groups, whose precise composition willdepend on the reform in question.” 31The Latvian case, however, does not seem to have been predicated upon a clear-cut“winners and losers” paradigm; many of the reforms relating to removal ofadministrative barriers in Latvia (including the improvement of the inspectorates’activities) overcame the distributional problems inherent in this paradigm. This31 World Bank, 1997, World Development Report, p. 146.38

Final Report June 30, 2004possible exception to a ge neral rule can be attributed to two factors: first, the nature ofthe consensus to drive the reform efforts forward and second, the fact that many of thereform efforts were not sufficiently radical or costly (and therefore not especiallynoticeable) to any one interest group that such groups would self- identify as the“victims” of the potential reforms and consequently act as an insurmountable sourceof resistance or, upon successful implementation of reform, become the clear (andsore) “losers”. Both exceptions touch upon the issue of those inspectors who couldhave had an interest in maintaining the status quo and could have formed a basis ofresistance to reform.One of the primary factors that makes the “winners and losers” model difficult toapply in Latvia was the underlying political consensus: EU accession. Related – andalso generally accepted – goals included the improvement of the investment climateand the reduction of administrative corruption. Through various outreach campaignsto stakeho lders, the press and the general public, the term “removal of administrativebarriers” came to be understood as an integral component in the nexus that leads toimprovement of the investment climate and ultimately economic growth anddevelopment. These were the arguments that were developed at the beginning of thereform initiatives and became generally adopted by all stakeholders.Another factor contributing to the lack of resistance to the reforms (and thereforeminimizing the possibility that there would be definable “losers”) was the nature ofthe reforms – they were sufficiently low- intensity, low-profile and cross-cutting thatfew politicians – or inspectors – became anxious about any one of these issues,whether reform of construction, inspectorates, border crossing, etc. A leadingtechnocrat involved in the reform efforts noted that for there to be resistance toreform, the reform would have to be important or costly for someone. This couldpotentially also apply to individual inspectors who could have formed a grass-rootssource of resistance – it is possible that they did not anticipate the implications of theincremental reforms in terms of their effect on individual corruptive and rent-seekingactivities. In effect, though, these individual inspectors would no longer be ableengage in corrupt practices due to a more open environment with effectiveinstruments of control, accountability and peer pressure 32 . Therefore, there did notappear to be significant resistance by the inspectors because the inspectors did nothave a natural base on which to mobilize themselves against the apparentlyincremental changes which led to more noticeable improvements in transparency andindividual accountability.Of course, another manner of anticipating and countering potential resistance couldhave been an increase in civil service pay, which can serve as an essential tool toovercome opposition of “losers”. This, however, was not used in Latvia as an integralpart of the inspectorate reform program, although the problem of low salaries waswell-known and acute. But reform was accomplished even without this mechanism.It remains for discussion whether the reform could have been more effective if therewas also an increase in inspectors’ salaries, but in any event this case study appears to32 It should be noted that there were some lateral factors that could also have weakened the possible“rebellion” of potential losers. The Inspectorate reform was unfolding on the background of the Anti-Corruption Program of the Government and an EU project aimed at raising awareness of corruptionand providing instruments of detecting and preventing corruption in the government. Without thatbackground, the strength of the potential losers could be difficult to neutralize.39

Final Report June 30, 2004demonstrate that (limited) reform is possible even without addressing such anapparently fundamental issue as pay reform.It is likely that some individual inspectors could have been benefiting from bribesprior to the reforms (and this assertion is supported by survey data and anecdotalevidence that indicate an overall reduction in bribery later). Such inspectors mayhave lost some revenue from the reforms, although this cannot be provenconclusively. Therefore, the heads of inspectorates strongly raised the issue ofappropriate remuneration for civil servants to ensure attractiveness and retention ofthe trained staff. Low salaries were also seen as the key reason for temptation toabuse the administrative role and engage in corrupt practices.12. Limiting corrupt practices via administrative checks and mechanisms foraccountabilityThe dynamics of bribery associated with inspections can be seen as falling largelyinto two categories:1) Bribes extorted by inspectors from businesses who may have been innocent ofserious wrong -doing: In many such cases, according to interviews withentrepreneurs, inspectors would seize upon minor, technical infractions and threatenmajor sanctions, and drop the threats only if they received a high enough bribe.Examples of “infractions” claimed by inspectors included over-paying taxes,submitting information on the wrong color form, etc.; and possible sanctions couldinclude large fines, freezing bank accounts or suspending business activity.This kind of extortion was made much more difficult once all inspectors wererequired to make a written report after all inspection visits, with a record of any“infractions” and the outcome of the investigation. Targeting of “vulnerable”businesses was also made more difficult with the introduction of more systematicinspection schedules and risk-assessment targeting of inspections.2) Bribes offered by businesses to inspectors to ignore or bend important regulationssuch as pollution controls or building standards. In this case, there is a greater degreeof “mutual advantage”, but monitoring of outcomes, and the ability to relate outcomesto individual inspectors has at least a ‘deterrent effect’ on inspectors who may betempted to accept such bribes. For example, the ability to compare amounts of finesimposed by different inspectors may reveal significant shortfalls associated withparticular inspectors. Unfortunately, in areas such as building standards, there is oftena lengthy delay between actual construction activities, and incidents such as buildingcollapse, such that monitoring is less likely to deter bribery to bend regulations in thisarea.Either way, according to focus group discussions and survey data, it appears that themajority of bribes involved were relatively small. On the other hand, the heads ofthree of the main inspectorates indicated overall improved performance and morale.One of the more interesting sources of satisfaction was that the operational regulationsand information materials prepared as part of the whole effort actually ease the workof the inspectors since they can now refer people to these materials and the websites.40

Final Report June 30, 2004References:1. Baltic Data House for the FIAS and the Latvian Development Agency. (2002). Results of the 2001Administrative and Regulatory Cost Survey.2. Cabinet of Ministers. (May 1999). Government Action Plan for Improving the BusinessEnvironment.3. Cabinet of Ministers. (September 2000). Inspectorate Improvement Program, September 15, 2000.4. Cabinet of Ministers. (January 2000). Instruction On Preparation Of Internal OperatingRegulations In Inspectorates, adopted on January 18, 2000.5. Cabinet of Ministers. (February 2000). The Policy Reform Program of PSAL 2 (Cabinet ofMinisters protocol of 17.02.2000, no. 8 23.§).6. Cabinet of Ministers. (January and June 2002). Implementation of Inspectorates ImprovementProgram. September 2000 - June 2002: Summary.7. European Commission. (1999). Annual Report on Latvia’s Progress Towards Accession.8. FIAS. (1998). Report on Administrative Barriers to Investment in Latvia.9. FIAS and Latvian Development Agency. (2003). Self-Assessment Report on AdministrativeProcedures for Doing Business in Latvia.10. Fire and Rescue Service. (2000). Annual Report.11. Fire and Rescue Service. (2001). Annual Report.12. Fire and Rescue Service. (2002). Annual Report.13. Grava, Lars and Putnina, Sanda. (2001). The Cycle of Reform. Latvian daily newspaper “Diena”,Commentary and Opinion (2 nd p.), 12 November 2001.14. Latvian Facts for the Latvian Development Agency and FIAS. (2004). Results of the 2003Administrative and Regulatory Cost Survey.15. Latvian Investment and Development Agency. (April 2004) Summary results of focus groupdiscussions on the results of the 2003 Administrative and Regulatory Cost survey. (unpublished)16. State Labor Inspectorate. (2000). Annual Report.17. State Labor Inspectorate. (2001). Annual Report.18. State Labor Inspectorate. (2002). Annual Report.19. State Reform Minister. (April 2000). Decree establishing the Inspectorate Coordination Council,April 14, 2000.20. State Reform Minister. (October 2001). Decree on the Inspectorate Coordination Council, October17, 2001.21. State Sanitary Inspectorate. (1999). Annual Report.22. State Sanitary Inspectorate. (2000). Annual Report.23. State Sanitary Inspectorate. (2001). Annual Report.24. State Sanitary Inspectorate. (2002). Annual Report.25. World Bank. (1997). 1997 World Development Report.26. World Bank. (1998). Corruption in Latvia: Survey Evidence.41

Final Report June 30, 2004Annexes toCase Study: Inspectorate Reform in Latvia, 1999 – 200342

Final Report June 30, 2004Annex A: Inspectorate Reform: Summary of Legislative Changes andProcedural or Administrative Changes Undertaken by the Government andIndividual Inspectorates.I. Legislative change and direct consequences of the legislative change.1) Adoption of CM Instruction On the Preparation Of Internal OperatingRegulations In Inspections.The Instruction was adopted on January 18, 2000. It was ultimately the result ofconsultations and discussions with the inspectorates themselves, and theimplementation of the guidelines in the Instruction was one of the first concretebases for cooperation among the inspectorates.2) Adoption of internal operating regulations.21 inspectorates adopted internal operating regulations, pursuant to the CMInstruction. The inspectorates are listed above (see Section B, pg. **).3) Adoption of the Inspectorate Improvement Program.The Inspectorate Improvement Program was adopted on September 15, 2000 bythe Cabinet of Ministers. Although much activity was already underway, thispolitical support was a crucial success.II. Procedural or administrative change based on legislative change.1) Setting up an Inspectorate Coordination Council.The Inspectorate Coordination Council was a forum for exchanging experiencesand ideas and it had begun to function informally even before its official creationon April 14, 2000 by decree of the Special Tasks Minister.2) Training of employees.Under the aegis of the Inspectorate Improvement Program unprecedented trainingefforts were implemented. Altogether 492 inspectors attended the seminarsorganized within the framework of the Training Program for the Heads ofInspectorates and Inspectors. Training of inspectors was carried out in thefollowing areas:• on how to establish an internal audit system,• assess risks,• prepare informational materials for the clients and• create an information system, on client-oriented communication andstrategic planning.43

Final Report June 30, 2004On an ongoing basis, the select inspectorates apply the following methods ofcommunicating significant changes in legislation and requirements to theiremployees 33 :Communicatingrequirements tostaffCopies of theamendedlegislation aredistributedSpecial trainingis providedManuals andinstructions areupdatedInspectorateLabor x x x xSanitary x x xFirefighting andxxRescueConstructionx x xInspectorateEnvironmental x x xx x xState RevenueServiceEmployeesfamiliarizethemselves withthe changesSource: 2003 Self Assessment Report on Administrative Procedures for DoingBusiness in Latvia3) Training of business representatives and preparation of informational materialsfor businesses.The self-assessment templates of 2002 asked the responding inspectorates toidentify which methods they use to communicate information on the requirementsfor this procedure to enterprises. The responses were as follows:Communicatingrequirements tobusinessesNo additionalinformation isdisseminated,because therelevant lawsandPrintedleaflets andbrochuresSeminarsandtrainingareorganizedInformation isposted on theinternet homepageAnsweringindividualinquiries (byphone, withletters, inperson, etc.)Inspectorateregulations aresufficientLabor x x x xSanitary* x xFirefighting andx x xRescueConstructionx x xInspectorateEnvironmental x x x xState RevenueServicexx* The Sanitary Inspectorate also uses the mass media to inform businesses about requirements.Source: 2003 Self Assessment Report on Administrative Procedures for DoingBusiness in Latvia33 In response to the question “what is the method of communicating significant changes in legislationand requirements to the employees of this institution?” in the Self -Assessment templates of 2002.44

Final Report June 30, 2004As seen in the table, many of the inspectorates now have internet home pages withinformation on the inspectorate and its functions, such as:State Labor Inspectorate (www.vdi.lv);State Fire and Rescue Service ( www.vugd.gov.lv);State Construction Inspectorate (www.varam.gov.lv/VBI );State Environmental Inspectorate (www.vvi.gov.lv);State Revenue Service (www.vid.gov.lv).4) Public relations, outreach and feedback.The self-assessment templates indicated that the inspectorates have used thefollowing methods to receive feedback from enterprises (the results are from2001/2):Feedback fromenterprisesInspectorateSuggestionsandcomplaintssubmitted byenterprisesFeedbackformsSurveysMeetings andworkshops withbusinessassociations andnongovernmentalorganizationsAnalyses ofinformationfrom massmediaLaborSanitary 2 1 13 2Firefighting and “not counted” “not counted” “not counted”RescueConstructionInspectorate6 10Environmental 4 23State RevenueServiceSource: 2003 Self Assessment Report on Administrative Procedures for DoingBusiness in Latvia5) Better understanding of their mission and the purpose of their procedures.In 2002 various inspectorates and audit institutions were asked to completetemplates assessing their impact on the business environment in Latvia. The“purposes” of the on-site inspections activities were as follows:Inspectorate or auditinstitutionState Revenue ServiceState Labor InspectoratePurpose of inspection/audit procedureTo check the correctness and compliance with normative acts of thecalculation, payment and deposit in the budget of a particulartaxpayer’s taxes.To ensure the observance of legislation relating to labor rights,workplace safety and monitoring of dangerous equipment, and as aresult to reduce the number of those harmed in accidents as well asoccupation illnesses.45

Final Report June 30, 2004Inspectorate or auditinstitutionState Sanitary InspectorateState Fire and Rescue ServicePurpose of inspection/audit procedureTo accomplish the implementation of normative acts defining publichealth and consumer safety on a nationwide basis in the areas of theinspectorate’s authority, and to reduce possible risks to the health ofthe public and consumers.To monitor that institutions, enterprises, organizations andinhabitants fulfill the requirements of normative acts regarding thereduction of fire hazards. To control that planning and constructionorganizations, enterprises and inhabitants comply with therequirements of design and construction standards, norms andregulations for any type of construction, reconstruction, and capitalrenovation, as well as technical modernization activities.State Environmental InspectorateState Construction InspectorateState Language InspectorateControl and monitoring of the implementation of normative acts inthe area of environmental protection and use of natural resources.To carry out state control over construction and state constructionexpertise, to control the fulfilment of the requirements in laws andother normative acts relating to the quality of construction, the safetyof hydrotechnical constructions in hydroelectric powerstations.To ensure the implementation of the State Language Law and otherlegal acts governing the use of the state language.Source: 2003 Self Assessment Report on Administrative Procedures for DoingBusiness in LatviaFrom the above table, it is clear that some inspectorates are still describing their“purpose” primarily in terms of enforcement of legislation, but most also note theoverall goals (e.g., for the State Labor Inspectorate: “… to reduce the number ofthose harmed in accidents as well as occupational illnesses.”).46

Final Report June 30, 2004Annex B: Availability of resourcesIn addition to main GoL staff resources, including the considerable LDA and BPARstaff-time, and other major in-kind contributions from the Inspectorates themselves,FICIL and other business associations, the following list of projects were the mainexternal sources of support to the effort in Latvia:FIAS projects:FY99 Study of Administrative Barriers to Investment: $135,600 34FY00-01 Implementation Assistance: $211,800 35FY02 Self-Assessment update of Administrative Barriers ; $139,200 36FY04 Assistance in Monitoring Impact: $60,000 37World Bank resources:FY00 PHRD grants: $39,500FY00 Finnish CTF: $36,000It should be noted that most of the FIAS projects were not exclusively designed forinspections reform, but rather a portion of the financing was able to support theseactivities.34 Including co-financing from the LDA, drawing on EU/Phare, Swedish CTF and other resources35 Including co-financing from the LDA, EU/Phare and SIDA36 Including co-financing from the LDA and the Government of Luxe mbourg37 Including co-financing from the LDA and World Bank (PREM)47

Final Report June 30, 2004Annex C: “Costs and benefits”While the reforms of inspectorates represented only one aspect of a larger and broaderset of reforms to improve the investment climate, the combined impact of all suchreforms in Latvia has been very impressive. Table C.1 and Table C.2 below showshow Latvia’s ratings have improved across a large ra nge of measures of theinvestment climate.Business environment ismeasured in a variety ofways. There are numerousperiodic indicators andratings on various aspects ofthe environment for doingbusiness, compiled, amongothers, by the EconomistIntelligence Unit, theHeritage Foundation, theWorld Bank and the EBRD,A.T.Kearney, the WorldEconomic Forum, etc. Manyindicators, if not all of them,are based on the individualCountryTable C.1. Business Competitiveness IndexBCI rankingBusinessenvironmentCompanyoperations andstrategy2003 2001 2003 2001 2003 2001Estonia 28 28 27 26 36 32Latvia 29 41 31 42 29 35Slovenia 30 32 34 35 27 28Czech Republic 35 34 38 31 33 41Hungary 38 27 37 25 45 33Lithuania 40 50 41 47 41 47Slovak Republic 42 40 42 36 44 57Poland 46 42 44 40 43 55Turkey 51 35 54 33 50 44Romania 67 61 64 60 72 63Bulgaria 68 68 67 65 73 70perceptions of respondents and are at times difficult to compare. Nevertheless, theyrepresent the existing sentiment of the markets towards the business environment at apoint in time. A rather representative indicator follows. Table 1 on the WorldEconomic Forum’s Business Competitiveness Index shows that Latvia has made someof the largest improvements over the last two years of all the EU Accession countries.48

Final Report June 30, 2004Table C.2. LATVIA: Summaryof RatingsHeritage Foundation's Indexof Economic Freedom1/1995 1996 1997 1998 1999 2000 2001 2002 20033.05 2.95 2.85 2.75 2.65 2.65 2.5 2.45Transparency International's2/Corruption Perception Index2.7 3.4 3.4 3.4 3.7 3.8EIU's Country Risk Service 3/ 75 51 53 53 50 43 41Euromoney's Country CreditRatings 4/ 31.09 47.12 57.71 37.81 50.41 53.11 52.56 58.84 62.11Institutional Investor'sCountry Credit Ratings 5/ 23.4 25.7 32.6 36.3 40.8 47.9 48 52 51.5EBRD's Transition Report 6/a) EBRD rating of legalextensiveness (company law) 3.3 3.3 3.7 4 3.7 n.a.b) EBRD rating of legaleffectiveness (company law) 3 2 3 3.7 4 n.a.1/ The Index is based on 10 broad factors of economic freedom (Trade policy, Fiscal burden of government, Governmentintervention in the economy, Monetary policy, Capital flows and foreign investment, Banking and finance, Wages and prices,Property rights, Regulation, and Black market).Each country is given a score ranging from 1 through 5 for all 10 factors; then these scores are averaged to get its final Indexof Economic Freedom score. Countries with a score between 1 and 2 have the free st economies; those with a score around3 are less free; those with a score near 4 are excessively regulated and will need significant economic reform to achievesustained increases in economic growth; and those with a score of 5 are the most economically repressed.2/ All countries in the CPI obtain scores between 0 (very corrupt) and 10 (highly clean).3/Letter scores range from “A” (the lowest risk) to “E” (the highest risk). Overall scores are awarded in one-point i ncrements,and can range from 0 (“A” category) to a maximum of 100 points (“E” category) for the highest-risk countries. The “A”49

Final Report June 30, 2004ratings category encompasses scores from 0-20; the “B” ratings category, 21-40; the “C” ratings category, 41-60;the “D” ratings category, 61-80; and the “E” ratings category, 81 -100.4/ To obtain the overall country risk Score, Euromoney assigns a weighting to 9 categories (political risk - 25%;economic performace - 25%; debt indicators - 10; debt in default or rescheduled - 10%; credit ratings - 10%;access to bank finance - 5%; acess to ST finance - 5%; access to capital markets - 5%, and discount on forfaiting - 5%)The higher the obtained total score, the better the credit rating.5/ Each country has been graded on a scale of zero to 100, with 100 representing those countries that have the least chanceof default.6/ The EBRD's Office of the General Counsel has developed measures to assess the extent to which key commercialand financial laws have reached internationally accepted standards (extensiveness) and the degree to which these lawsare implemented and enforced (effectiveness). Scores range from 1 (bad) to 4+(excellent).50

Final Report June 30, 2004Annex D: Econometric Analysis of Latvia Survey Data on InspectorateReform51

Final Report June 30, 2004In this appendix, we empirically focus on the outcomes of the inspectorate reformprocess. Key objectives of this empirical investigation are as follows:1. To see whether inspection outcomes have improved significantly between2001 and 2003.2. To examine whether benefits of inspection reform are only concentrated inthe capital Riga or have diffused across all regions in Latvia over time.3. To examine whether the inspection outcomes are biased against particulartypes of firms or all types of firms have more or less gained from theinspectorate reform process.4. To investigate whether there are any differences among variousinspectorates as far as inspections outcomes are concerned, and if yes, thenwhat has been the direction of change in different inspectorates between2001 and 2003.We focus on five different outcome measures to proxy for various dimensionsof the burden of inspections. The outcomes are:• incidence of inspection,• frequency of inspection,• number of inspectors from the same inspectorates who inspected the firm• hours spent in inspection• incidence of fine imposition upon inspectionWe analyze these inspection outcomes for seven different inspectorates:• labor• sanitary• fire and rescue• construction• environment• state revenue system• municipal policeAs far as firm characteristics are concerned, firms are categorized into four typesbased on their activities –• manufacturing• commerce• services• construction and others (henceforth construction)52

Final Report June 30, 2004Depending on its labor force, the firm’s size is categorized into:• small (1-9 employees)• medium (10-49 employees)• and large (50 or more employees)Other two key attributes of the firm are:• whether it is an exporter or not• whether it has some form of foreign ownership or notWe estimate following three models for each of the inspection outcomes:yi=0year + α1α location+∈(1)iyi= α year + α location + ( year * location)+ ∈(2)0 1α2iyni∑j=1=ωn∑j=11jβ1 j( typetypejj+m∑k=1* year ) +β2km∑k=1+ ω ( location * year ) + ε5isizeω2kk+ λ exp ort + ∂ × foreign + γ × location + δ × yrinbus( sizek* year ) + ω3(exp ort * year)+ ω4( foreign * year )i+ ϖ year +1where, yiis the inspection outcome for firm i, year is the dummy variable thatassumes the value 1 if the survey year is 2003 and 0 otherwise, location is thedummy variable that takes the value 1 if the firm is in Riga and 0 otherwise, typeis the dummy variable for type j, sizekis the dummy variable for size k, export andforeign are the dummy variables representing whether the firm exports or not andwhether it has foreign ownership or not, yrinbus i is the number of years the firmhas been in business, and ∈iis the stochastic error term.Model (1) helps us to determine whether the inspection outcomes are significantlydifferent (better or worse) in Riga than elsewhere and whether outcomes havechanged significantly between 2001 and 2003. Model (2) further allows us toanalyze whether inspection outcomes have significantly changed in Riga between2001 and 2003 as well as whether the inspection outcomes in Riga aresignificantly different than elsewhere in 2003. The underlying hypothesis that weare interested to test is that even if immediate post reform benefits were initiallyconcentrated in the capital Riga in 2001, whether these benefits have beensubsequently spread over the country in general. If the inspection situation hasimproved in all parts of the country between 2001 and 2003, one would thenexpect that the coefficient α2in model (2) to be insignificant (at least in the(3)j53

Final Report June 30, 2004economic sense, that is, the coefficient would be of negligible magnitude, eventhough it is statistically significant), implying that inspection outcomes inelsewhere in the country is as good as in Riga in 2003.Finally, model (3) helps us to analyze whether the inspection outcomes are biasedagainst particular types of firms or not, and if some types of firms are yet to bebenefited from the inspectorate reforms in 2001, whether their situations haveimproved in 2003. As we estimate the same model for the inspection outcomes ofdifferent inspectorates, this in turn helps us to comparatively assess the outcomesof different inspectorates. The estimates of these models are reported in tables 1.1-3.4. When the outcome variable is a 0-1 categorical variable as in the case ofincidence of inspection by different inspectorates, we obtain a probit estimate ofthese models, and when the outcome variable is a continuous variable likeinspection frequency, number of inspectors who visited the factory and hours spentin inspection, we estimate the models using OLS (adding a constant term).Some of the key findings of our analysis are as follows:• Significant reduction of incidence of inspection between 2001 and 2003.• Substantial reduction in time spent in inspection between 2001 and 2003.• Significant reduction in the probability of fine imposition between 2001and 2003.• Burden of inspection is lesser for firms in Riga than in elsewhere in thecountry (with the exception of municipal police inspection).• Burden of municipal police inspection is higher in Riga than in elsewherein the country.• Inspectorate reform in Latvia to a great extent benefits the small andmedium firms, compared to large firms.• Inspectorate reform to some extent also tends to benefit export orientedfirms (compared to non-exporters) and firms under some form of foreignownership (compared to domestic firms).• Compared to construction firms, the burden of inspections seems to behigher for the manufacturing and commerce firms.In the following we provide a detail analysis of these key findings 38 .Change in the Burden of Inspections between 2001 and 2003The burden of inspection in terms of incidence of inspections from labor, sanitary,construction, and municipal police has been significantly reduced in 2003compared to 2001. The corresponding reduction in labor, sanitary, construction,38 We focus on the estimates that are statistically significant at 10% or below unless otherwisementioned.54

Final Report June 30, 2004and municipal police inspection probabilities are 10%, 7%, 4%, and 7%respectively (see table 1.1). The reduction in inspection frequency between 2001and 2003, however, is only significant for the sanitary inspectorate. While in 2001the average number of sanitary inspections for a firm outside Riga was about 3, in2003 it has reduced to only about 1 (see table 1.2).There has also been a substantial reduction in hours spent in inspection between2001 and 2003. For a firm outside Riga, the average time spent in labor inspectionhas reduced from about 34 hours in 2001 to about 2 hours in 2003, that fromsanitary inspection from about 55 hours to about 7 hours, that from fire and safetyinspection from about 15 hours to about only 2 hours, that from environmentinspection from about 10 hours to less than an hour, that from state revenue systemfrom about 153 hours to only about 5 hours, and that from municipal policeinspection from about 7 hours to less than one hour (see table 1.3). Not just thetime spent in inspections from different inspectorates has been reduced between2001 and 2003, there has also been a marginal decline in the number of inspectorsfrom labor, sanitary, and municipal police inspectorates who visited the firm (seetable 1.4).Finally, the probability of fine imposition from the labor, sanitary, fire and safety,state revenue system and municipal police have reduced respectively by 84%,82%, 87%, 64%, and 73% between 2001 and 2003.Burden of Inspection for Firms in Riga vis-à-vis elsewhere in the CountryLocation seems to have a strong effect on the burden of inspection. Apart from thecase of municipal police inspection, for all other inspectorates, burden ofinspection is less if the firm is located in Riga vis-à-vis elsewhere in the country.This in turn implies that there is enough room for improving the inspectionsituation outside the capital and thus to diffuse benefits of inspectorate reformthroughout the country. On the other hand, municipal police still seems to exert itsstrong discretionary power and that power is more prominent in Riga than inelsewhere in the country. This in turn deserves serious attention and reforminitiative for municipal police inspectorate.If a firm is located in Riga, the probability of inspection from labor, sanitary, fire,environment, and state revenue system are respectively about 13.5%, 12%, 8%,4.5%, and 24% lower than its counterpart which is located elsewhere in thecountry (table 1.1). On the contrary, the probability of municipal police inspectionis about 11% higher for a firm which is located in Riga than its counterpart inelsewhere in the country. These contrasting evidences in turn reflect thediscretionary power of municipal police in Riga. This inspection situation in Rigadoes not significantly differ between 2001 and 2003 (table 2.1).55

Final Report June 30, 2004While the frequency of labor inspection is significantly lower in Riga thanelsewhere, that of municipal police inspection is however significantly higher inRiga than elsewhere (table 2.1). This frequency pattern in Riga has also notchanged significantly between 2001 and 2003 (table 2.2). However, hours spent ininspections from different inspectorates seem not to be significantly different inRiga than elsewhere and neither the situation in Riga differs significantly between2001 and 2003 in terms of hours spent in inspection (table 1.3 and 2.3).While the number of inspectors visited from labor, fire and state revenue system issignificantly lower in Riga than elsewhere, that from municipal police is againsignificantly higher in Riga than elsewhere (table 1.4). However, the latter tends togo down slightly in 2003 compared to 2001 in Riga (table 2.4). Finally, theprobability of imposition of fine from the municipal police is about 32% higher fora firm in Riga than elsewhere (table 1.5). The corresponding probability for otherinspectorates, however does not differ significantly between Riga and the rest ofLatvia.Firm Size and the Burden of InspectionsSmall FirmOne of the notable aspects of inspectorate reform in Latvia is that in comparisonwith the large firm, the burden of inspections seems to reduce significantly forsmall firm and to some extent for the medium firm as well. In 2003, the probabilityof labor inspection is about 51% less for the small firm than that for the large firm,which is a further improvement of the situation from 2001 39 . Similarly, theprobability of fire, construction, and environment inspections are also about 45%.9%, and 20% lower, respectively for the small firm than that for the large firm in2003. While the probability of municipal police inspection did not differsignificantly between small and large firm in 2001, subsequently in 2003, themunicipal police inspection probability has reduced by 15% for the small firmcompared to the large firm (table 3.1).The burden of inspection frequency also seem to improve for the small firm inrelation to the large firm. Inspection frequencies from sanitary, fire, construction,environment, and state revenue system are about 4, 1, 1.5, 2, and 4 points lower,respectively for the small firm than that for the large firm in 2003. While in 2001,labor inspection frequency was about 0.45 points lower for the small firm than thatfor the large firm, the situation for the small firm in relation to the large firm hasfurther improved in 2003. Now compared to the large firm, labor inspectionfrequency for the small firm is about 1 point lower (table 3.2).39 In 2001, the corresponding probability of labor inspection was about 31% less for the small firmthan that for the large firm.56

Final Report June 30, 2004On the other hand, time spent in inspection by all the inspectorates (with theexception from labor inspectorate, for which the time spent in inspection does notdiffer significantly between the small and large firm) has substantially increasedfor the small firm compared to the large firm in 2003 from 2001. While initially in2001 time spent in inspection from different inspectorates (other than the laborinspectorate) was significantly lower for the small firm compared to the large firm,the situation becomes completely opposite in 2003. Now (i.e., in 2003) the timespent in sanitary, fire, construction, environment, and state revenue system areabout 124, 36, 59, 80 and 430 hours more, respectively for the small firm than thatfor the large firm (table 3.3). Similarly the number of inspectors from sanitary,fire, environment, and state revenue system who inspected the firm hassignificantly increased for the small firm (compared to the large firm) in 2003from 2001 (table 3.4).Medium FirmsCompared to the large firm, the burden of inspection in some dimensions alsotends to be significantly lower for the medium firm. For instance, the probabilitiesof fire and safety inspection and environment inspection are respectively about18% and 7% lower for the medium firm than that for the large firm in 2003.Construction, environment, and state revenue inspection frequencies are also about1.1, 1.6 and 3.1 points lower, respectively for the medium firm compared to thelarge firm in 2003. The burden of labor and fire inspection frequencies furtherlessens for the medium firm compared to the large firm in 2003 from 2001. Whilein 2001, the labor inspection frequency was 0.22 points lower for the medium firmthan that for the large firm, the frequency is about 0.79 points lower for themedium firm than that for the large firm in 2003. Similarly in 2001, while the fireinspection frequency was about 0.57 points lower for the medium firm than thatfor the large firm, it has further gone down by 0.6 points in 2003 for the mediumfirm in relation to the large firm.While initially in 2001, hours spent in various inspections were much less for themedium firm than that for the large firm, the situation has almost reversed in 2003.Compared to the situation where time spent in sanitary, fire, construction,environment, and state revenue inspections were respectively about 109, 31, 56,72, and 382 hours less for the medium firm than that for the large firm, in 2003 thecorresponding time spent in these respective inspections are about 119, 29, 54, 71,and 368 hours more for the medium firm than that for the large firm.Similar reversal is also observed in terms of number of inspectors from the sameinspectorate who inspected the firm. While the sanitary, environment and staterevenue inspectors, who visited the medium firm were 6, 1, and 6 less than thecorresponding inspectors who visited the large firm in 2001, the number ofsanitary, environment and state revenue inspectors who visited the medium firm in57

Final Report June 30, 20042003 are respectively 8, 1, and 8 more than the corresponding number ofinspectors who visited the large firm.Firm type and the Burden of InspectionsManufacturingIn terms of firm type according to its activity, the probability of sanitary and fireinspections seemed to be about 23% and 33% higher for the manufacturing firmthan that for the construction firm in 2003 (table 3.1). The burden is also worse forthe manufacturing firm (compared to the construction firm) in terms of frequencyof fire inspections and time spent in labor and fire inspection. The frequency offire inspection is about 0.67 points higher and time spent in labor and fireinspections are respectively about 65 and 19 hours more for the manufacturingfirm compared to construction firm in 2003 (table 3.3) 40 . Similarly, the number offire inspectors inspected the factory in 2003 is also marginally higher for themanufacturing firm compared to the construction firm (table 3.4).Trade and CommerceThe inspection burden also seems to be higher for the trade and commerce sectorfirm (compared to the construction firm). The probability of sanitary inspection isabout 47.5% higher for the trade and commerce sector firm in 2003. Number ofinspections from sanitary, fire and safety, and state revenue system are also about4, 0.5, and about 3 points higher for the trade and commerce firm than that for theconstruction firm in 2003. Similarly, time spent in sanitary and state revenueinspections are respectively about 91 and 6 hours more for the commerce sectorfirm compared to the construction firm in 2003. The number of sanitary inspectorsis also about 7 points higher for the commerce sector firm in 2003 than that for theconstruction firm.ServiceService sector firms on the other hand tend to do better than construction firms insome aspects of inspection burden. Although initially in 2001, the probability ofsanitary inspection was about 43% higher for the service firm than that for theconstruction firm, it has gone down by 26% for the service firm compared to theconstruction firm in 2003. The probability of construction inspection is about 10%lower for the service sector firm than that for the construction firm in 2003. While40 Although these corresponding frequency and time spent on inspections have been reducedsubstantially for the manufacturing firm compared to construction firm between 2001 and 2003,they are not statistically significant58

Final Report June 30, 2004the time spent in state revenue inspection was also about 319 hours less for theservice sector firm than that for the construction firm in 2001, there has been a netincrease in the state revenue inspection time by about 2 hours for the service sectorfirm in 2003 (compared to construction firm).ExporterThe probability of sanitary inspection is about 15% lower for the exporter than thatfor the non-exporter in 2003. Although initially in 2001, the probability of staterevenue inspection was 24.5% higher for the exporter than that for the nonexporter,it has subsequently declined for the exporter (in relation to the nonexporter)by 22% in 2003. However, in comparison to the non-exporter, theprobability of fire and environment inspections are respectively about 23% and24% higher for the exporter in 2003.Although initially in 2001 sanitary and state revenue inspection frequencies wereabout 6 points higher for the exporter than non-exporter, they have also declinedby about 6 and 5 points, respectively for the exporter (compared to the nonexporter)in 2003. Thus in 2003, the magnitude of these inspection frequencies arealmost same for both the exporter and non-exporter.Time spent in construction, environment, and state revenue inspections arerespectively about 48, 46, and 173 hours less for the exporter than that for nonexporterin 2003 41 . The burden of inspection in terms of the number of sanitaryand state revenue inspectors has appeared to reduce for the exporter compared tothe non-exporter between 2001 and 2003. While in 2001, the number of sanitaryand state revenue inspectors were about 11 and 10 more for the exporter than thatfor the non-exporter, these corresponding numbers have declined by the samemagnitude in 2003, implying even though in 2001 export firms were in acomparatively disadvantageous position in relation to non-exporter, this problemhas tended to disappear by 2003.Foreign OwnershipThe probability of municipal police inspection in 2003 is about 11% lower for thefirm under some form of foreign ownership than that for the domestically ownedfirm. On the other hand, compared to the domestic firm, while in 2001 sanitary andstate revenue inspection frequencies were 5 and 3 points lower, respectively for theforeign ownership firm, these corresponding inspection frequencies have beenincreased by 6 and 4 points, respectively for the foreign ownership firms (in41 In 2001, construction, environment and state revenue inspections were about 48, 46, and 173hours less for exporter than non-exporter, while in 2003 these inspection time have been increasedby the similar magnitude. But none of these increases are statistically significantly different fromzero.59

Final Report June 30, 2004relation to the domestic firm) in 2003. Fire inspection frequency has also increasedby 1 point between 2001 and 2003 for the foreign ownership firm compared to thedomestic firm. However, while construction and environment inspectionfrequencies were about 2 points higher for the foreign ownership firm than that forthe domestic firm in 2001, each of these frequencies have gone down by the samemagnitude in 2003 for the foreign ownership firm in relation to the domestic firm.Although in 2001 construction, environment, and state revenue inspections were75, 70, and 218 hours more for the foreign ownership firm compared to thedomestic firm, time spent in construction and environment inspections havedeclined by the same magnitude for the foreign ownership firm in 2003 than thatfor the domestic firm. For the state revenue system, the inspection time is about221 hour less for the foreign ownership firm than that for the domestic firm in2003, implying a net decrease of 3 hours of inspection time for the foreignownership firm (in relation to the domestic firm) between 2001 and 2003.Furthermore, compared to the domestic firm, sanitary and municipal policeinspections are respectively about 76 and 15 hours less for the foreign ownershipfirm in 2003.While compared to the domestic firm, 8 less sanitary inspectors visited the foreignownership firm in 2001, the number of sanitary inspectors who visited the foreignownership firm in 2003 has increased by the same magnitude in relation to thedomestic firm. On the contrary, while initially in 2001 the number of constructionand environment inspectors were about 2 more for the foreign ownership firm thanthat for the domestic firm, these numbers have declined by the same magnitude in2003 for the foreign ownership firm (compared to the domestic firm). Finally, thenumber of state revenue and municipal police inspectors, who visited the foreignownership firm in 2003 are about 7 and 1 less than the corresponding inspectors,who visited the domestic firm.Thus, there is much room for further improvement, for a large number ofindicators, particularly as some inspection burdens seem to have increasedbetween 2001 and 2003. Nonetheless, for a good number of indicators, as seenabove, the burden of inspections seem to be lesser for the small and medium firms(compared to the large firms) and for the exporter and foreign ownership firms(compared to the non-exporter and domestic firms).SummaryTo summarize the impact of inspectorate reform process in Latvia, it appears thatbetween 2001 and 2003, the burden of inspection has been reduced to a greatextent. However, there is yet much room for improvements. It is also necessary tofurther investigate why the burden of inspection from some inspectorates, whichwas initially lower for some types of firms, such as exporter and foreign ownershipfirms compared to their respective counterparts (i.e., non-exporter and domestic60

Final Report June 30, 2004firms) has tended to increase between 2001 and 2003. Unless these issues areaddressed seriously and appropriate measures are in place, it could have a negativeimpact on the country’s future foreign direct investment inflow and export. This inturn could adversely affect Latvia’s overall growth and investment path. In termsof activity, to a great extent, the burden of inspection seems to be comparativelyheavier for manufacturing and trade and commerce sector firms. Such burdenneeds to be eliminated in order to ensure level playing field for all types ofbusiness activities and to achieve a balanced industrial growth.Benefits of inspectorate reform also tend to be concentrated largely in the capital,while for a broad-based investment led growth and to ensure that benefits ofgrowth and increased investment diffuse over all parts of the country, theinspectorate reform process needs to be continued in all parts of the country.While the burden of inspection from most of the inspectorates seem to be lesser inRiga, the burden of municipal police inspection rather contrastingly is much higherin Riga, implying the strong discretionary power of municipal police in the capitalcity. Municipal police inspectorate reform should be thus on the top of the prioritylist. On the other hand, whereas in many transition countries, tax administrationand in particular tax inspection seems to be one of the most critical barriers tobusiness, the improved situation of state revenue inspection in Latvia isnoteworthy and highly reassuring.Finally, one of the most noticeable aspects of inspectorate reform is that smallfirms are the principal beneficiary of such reform process. Also the burden hasbeen reduced to some extent for medium firms. To some extent, export orientedfirms and firms under some form of foreign ownership have also been benefited byinspectorate reform. This is very encouraging as it should not only have a positiveimpact on growth and investment but should also facilitate the improvement ofabsolute and relative poverty situation in Latvia in the future. However, there hasbeen a tendency of increasing hours and numbers of inspectors visit for small andmedium firm in 2003, compared to 2001. To make the inspectorate reform processfruitful, such a tendency should be curbed.61

Final Report June 30, 2004Tables:Table 1.1: Probability of Inspection-Model 1(1) (2) (3) (4) (5) (6) (7)Labor Sanitary Fire Construction Environment SRS PoliceRiga -0.135 -0.123 -0.084 -0.006 -0.045 -0.238 0.107(4.10)*** (3.79)*** (2.46)** (0.32) (1.99)** (6.82)*** (3.69)***2003 -0.098 -0.068 -0.033 -0.037 -0.020 0.007 -0.070(3.06)*** (2.15)** (0.98) (1.91)* (0.91) (0.20) (2.50)**Observations 886 886 886 886 886 886 886Note: Marginal effects of probit estimates are presented in the table. Heteroskedasticity corrected z statistics arein parentheses. * significant at 10%; ** significant at 5%; *** significant at 1%.Table 1.2: Frequency of Inspections-Model 1(1) (2) (3) (4) (5) (6) (7)Labor Sanitary Fire Construction Environment SRS PoliceRiga -0.125 -0.752 -0.008 0.296 0.208 -0.796 0.412(2.03)** (1.03) (0.07) (0.98) (0.69) (1.49) (2.94)***2003 -0.062 -1.547 0.004 -0.250 -0.235 -0.755 -0.169(1.07) (2.13)** (0.04) (1.19) (1.11) (1.25) (1.33)Constant 0.549 2.661 0.620 0.309 0.396 2.230 0.512(12.67)*** (3.15)*** (8.94)*** (2.83)*** (3.52)*** (2.97)*** (4.43)***Observations 882 878 879 880 881 878 880Adjusted R- 0.00 0.00 0.00 0.00 -0.00 0.00 0.01squaredNote: Robust t statistics are in parentheses * significant at 10%; ** significant at 5%; *** significant at 1%.Table 1.3: Hours spent in Inspections -Model 162

Final Report June 30, 2004(1) (2) (3) (4) (5) (6) (7)Labor Sanitary Fire Construction Environment SRS PoliceRiga -0.951 -15.500 -2.471 10.922 9.670 -0.976 3.160(0.12) (1.03) (1.01) (0.98) (0.87) (0.03) (1.19)2003 -32.256 -47.725 -12.874 -10.926 -13.241 -147.481 -7.214(3.91)*** (3.07)*** (5.19)*** (1.44) (1.72)* (4.66)*** (3.38)***Constant 33.726 54.580 14.686 7.437 10.224 152.917 6.549(3.38)*** (2.92)*** (5.01)*** (1.87)* (2.43)** (4.60)*** (3.36)***Observations 871 870 863 878 875 865 870Adjusted R- 0.01 0.01 0.03 0.00 0.00 0.02 0.01squaredNote: Robust t statistics are in parentheses. * significant at 10%; ** significant at 5%; *** significant at 1%.Table 1.4: Numbers of Inspectors Visited -Model 1(1) (2) (3) (4) (5) (6) (7)Labor Sanitary Fire Construction Environment SRS PoliceRiga -0.165 -1.337 -0.189 0.195 0.196 -1.777 0.385(2.24)** (1.20) (2.57)** (0.64) (0.66) (1.78)* (3.58)***2003 -0.218 -2.333 -0.082 -0.297 -0.279 -1.841 -0.330(3.13)*** (1.89)* (1.13) (1.42) (1.35) (1.57) (3.83)***Constant 0.716 3.434 0.751 0.367 0.377 3.676 0.539(11.17)*** (2.24)** (11.48)*** (3.18)*** (3.40)*** (2.45)** (8.06)***Observations 875 875 878 877 879 875 876Adjusted R- 0.01 0.00 0.01 0.00 0.00 0.00 0.03squaredNote: Robust t statistics are in parentheses. * significant at 10%; ** significant at 5%; *** significant at 1%.Table 1.5: Probability of Imposition of Fines(1) (2) (3) (4) (5) (6)Labor Sanitary Fire Environment SRS PoliceRiga 0.004 -0.120 0.107 0.071 -0.074 0.316(0.06) (1.23) (1.96)* (0.77) (1.08) (2.94)***2003 -0.837 -0.821 -0.871 -0.644 -0.734(6.85)*** (6.78)*** (5.90)*** (6.76)*** (5.42)***Observations 176 191 239 60 374 139Note: Marginal effect of probit estimation are reported in the table. Robust z statistics are in parentheses. *significant at 10%; ** significant at 5%; *** significant at 1%.Table 2.1: Probability of Inspection-Model 2(1) (2) (3) (4) (5) (6) (7)63

Final Report June 30, 2004Labor Sanitary Fire Construction Environment SRS PoliceRiga -0.108 -0.125 -0.070 0.003 -0.053 -0.191 0.105(2.48)** (2.89)*** (1.50) (0.10) (1.76)* (4.03)*** (2.74)***2003 -0.076 -0.070 -0.021 -0.029 -0.026 0.046 -0.071(1.94)* (1.79)* (0.50) (1.20) (0.99) (1.08) (1.97)**2003 X Riga -0.064 0.004 -0.033 -0.022 0.020 -0.108 0.005(0.96) (0.06) (0.47) (0.56) (0.41) (1.52) (0.08)Observations 886 886 886 886 886 886 886Notes: Robust z statistics are in parenthes es. * significant at 10%; ** significant at 5%; *** significant at 1%.Table 2.2: Frequency of Inspections-Model 2(1) (2) (3) (4) (5) (6) (7)Labor Sanitary Fire Construction Environment SRS PoliceRiga -0.110 -1.298 -0.044 0.594 0.495 -1.177 0.393(1.50) (0.99) (0.40) (1.06) (0.90) (1.25) (2.03)**2003 -0.050 -1.984 -0.025 -0.013 -0.005 -1.059 -0.184(0.75) (1.94)* (0.29) (0.18) (0.05) (1.14) (1.25)2003*Riga -0.033 1.188 0.078 -0.647 -0.626 0.827 0.040(0.26) (0.87) (0.31) (1.15) (1.12) (0.82) (0.14)Constant 0.543 2.878 0.634 0.192 0.282 2.380 0.520(11.79)*** (2.84)*** (9.60)*** (3.91)*** (4.67)*** (2.60)*** (3.99)***Observations 882 878 879 880 881 878 880Adjusted R- 0.00 0.00 -0.00 0.00 0.00 0.00 0.01squaredNote: Robust t statistics are in parentheses. * significant at 10%; ** significant at 5%; *** significant at 1%.Table 2.3: Hours spent in Inspection-Model 2(1) (2) (3) (4) (5) (6) (7)Labor Sanitary Fire Construction Environment SRS PoliceRiga -1.764 -27.699 -4.605 20.409 18.005 -2.456 6.014(0.12) (0.99) (1.01) (0.99) (0.88) (0.04) (1.22)2003 -32.908 -57.401 -14.559 -3.395 -6.544 -148.657 -4.958(2.72)*** (2.54)** (4.11)*** (1.95)* (2.87)*** (3.82)*** (2.28)**2003Xriga 1.758 26.282 4.564 -20.587 -18.099 3.175 -6.142(0.12) (0.93) (1.00) (1.00) (0.88) (0.05) (1.24)Constant 34.054 59.427 15.537 3.713 6.876 153.511 5.422(2.82)*** (2.63)*** (4.39)*** (2.13)** (3.01)*** (3.94)*** (2.50)**Observations 871 870 863 878 875 865 870Adjusted R-squared0.01 0.01 0.03 0.00 0.00 0.02 0.0164

Final Report June 30, 2004Note: Robust t statistics are in parentheses. * significant at 10%; ** significant at 5%; *** significant at 1%.Table 2.4: Numbers of Inspectors Visited -Model 2(1) (2) (3) (4) (5) (6) (7)Labor Sanitary Fire Construction Environment SRS PoliceRiga -0.204 -2.119 -0.129 0.412 0.424 -2.606 0.566(1.85)* (1.03) (1.11) (0.74) (0.77) (1.41) (3.05)***2003 -0.250 -2.955 -0.034 -0.125 -0.097 -2.500 -0.187(2.85)*** (1.57) (0.35) (1.60) (1.45) (1.35) (2.26)**2003* Riga 0.086 1.695 -0.131 -0.471 -0.495 1.794 -0.392(0.60) (0.82) (0.92) (0.84) (0.89) (0.96) (1.92)*Constant 0.732 3.744 0.728 0.282 0.287 4.004 0.468(9.98)*** (2.00)** (9.66)*** (4.10)*** (5.00)*** (2.18)** (6.71)***Observations 875 875 878 877 879 875 876Adjusted R- 0.01 0.00 0.01 0.00 0.00 0.00 0.04squaredNote: Robust t statistics are in parentheses. * significant at 10%; ** significant at 5%; *** significant at 1%.65

Final Report June 30, 2004Table 3.1: Probability of Inspection-Model 3(1) (2) (3) (4) (5) (6) (7)Labor Sanitary Fire Construction Environment SRS PoliceManufacturing 0.058 0.529 0.327 -0.062 0.096 -0.139 -0.031(0.52) (3.58)*** (2.79)*** (1.50) (1.22) (1.15) (0.35)Trade 0.070 0.475 0.241 -0.044 0.025 0.157 0.126(0.76) (3.69)*** (2.39)** (1.08) (0.44) (1.61) (1.62)Service -0.058 0.427 0.083 -0.103 -0.030 -0.048 0.016(0.64) (3.29)*** (0.82) (2.49)** (0.54) (0.48) (0.21)2003 X 0.115 -0.304 -0.016 0.045 -0.022 0.209 0.075Manufacturing(0.69) (2.70)*** (0.09) (0.49) (0.26) (1.31) (0.51)2003 X Trade 0.051 -0.218 0.023 -0.017 -0.016 0.113 -0.024(0.37) (1.58) (0.15) (0.28) (0.20) (0.81) (0.22)2003 X Service 0.076 -0.259 0.146 0.018 0.034 0.090 -0.013(0.56) (1.89)* (0.99) (0.26) (0.40) (0.65) (0.12)Export -0.036 -0.152 -0.079 -0.024 0.019 0.245 -0.087(0.47) (2.10)** (0.93) (0.61) (0.40) (2.91)*** (1.44)2003 X Export 0.046 0.134 0.233 0.036 0.247 -0.221 0.031(0.37) (1.03) (1.80)* (0.46) (2.45)** (1.74)* (0.28)Foreign 0.020 0.004 -0.034 0.001 0.011 -0.022 -0.114(0.25) (0.05) (0.39) (0.02) (0.23) (0.25) (1.87)*2003 X Foreign -0.138 -0.040 0.022 -0.064 -0.093 -0.073 0.148(1.17) (0.32) (0.16) (1.12) (2.08)** (0.53) (1.14)Year in 0.003 0.003 -0.000 0.002 0.001 0.006 0.002business(0.87) (1.06) (0.06) (1.41) (1.02) (1.82)* (0.87)Small -0.306 -0.059 -0.453 -0.091 -0.197 -0.130 -0.006(4.01)*** (0.78) (5.61)*** (2.15)** (3.95)*** (1.58) (0.09)Medium -0.112 0.071 -0.182 -0.034 -0.071 -0.034 0.06566

Final Report June 30, 2004(1.55) (0.90) (2.24)** (0.94) (2.02)** (0.40) (0.95)2003 X Small -0.196 -0.164 0.035 0.026 0.034 -0.090 -0.155(1.74)* (1.49) (0.26) (0.40) (0.49) (0.67) (1.76)*2003 X -0.126 -0.118 -0.013 0.039 0.031 -0.081 -0.093Medium(1.12) (1.07) (0.10) (0.55) (0.44) (0.60) (1.07)Riga -0.188 -0.118 -0.182 -0.021 -0.096 -0.252 0.139(4.01)*** (2.55)** (3.51)*** (0.82) (3.40)*** (4.81)*** (3.38)***2003 X Riga -0.027 -0.026 0.030 0.004 0.072 -0.058 -0.032(0.36) (0.37) (0.39) (0.09) (1.34) (0.76) (0.56)2003 0.033 0.363 -0.129 -0.056 -0.066 0.067 0.079(0.22) (2.13)** (0.80) (0.80) (0.79) (0.42) (0.62)Observations 886 886 886 886 886 886 886Note: Absolute value of z statistics are in parentheses. * significant at 10%; ** significant at 5%; *** significant at1%.Table 3.2: Frequency of Inspections-Model 3(1) (2) (3) (4) (5) (6) (7)Labor Sanitary Fire Construction Environment SRS PoliceManufacturing 0.013 -1.881 0.664 -1.166 0.179 -3.131 0.009(0.07) (0.77) (2.14)** (1.43) (0.22) (1.61) (0.02)Commerce 0.026 4.317 0.484 -0.627 0.592 2.772 0.440(0.17) (2.10)** (1.86)* (0.91) (0.88) (1.71)* (1.20)Services -0.194 2.214 0.122 -0.563 0.626 0.423 0.347(1.24) (1.07) (0.46) (0.81) (0.92) (0.26) (0.93)2003*0.040 1.351 -0.613 1.006 -0.395 2.979 0.012Manufacturing(0.15) (0.39) (1.38) (0.87) (0.34) (1.08) (0.02)2003 *0.317 -3.366 0.240 0.580 -0.868 -1.425 0.002Commerce(1.43) (1.14) (0.64) (0.60) (0.90) (0.61) (0.00)2003* Service 0.330 -1.850 0.279 0.521 -0.917 0.200 -0.283(1.50) (0.64) (0.76) (0.54) (0.96) (0.09) (0.55)Export -0.071 6.084 -0.171 -1.348 -1.069 6.184 -0.137(0.53) (3.43)*** (0.76) (2.32)** (1.83)* (4.38)*** (0.44)2003 * Export 0.308 -5.805 1.024 1.345 1.474 -5.264 -0.066(1.53) (2.19)** (3.04)*** (1.55) (1.69)* (2.50)** (0.14)Foreign 0.002 -4.702 -0.064 2.073 2.168 -2.945 -0.337(0.02) (2.66)*** (0.28) (3.55)*** (3.70)*** (2.09)** (1.08)2003* Foreign 0.078 6.201 0.717 -2.302 -2.525 4.153 0.324(0.36) (2.20)** (2.00)** (2.48)** (2.71)*** (1.85)* (0.65)Year in business 0.003 0.398 -0.012 0.004 0.019 0.423 -0.004(0.60) (6.06)*** (1.46) (0.20) (0.90) (8.21)*** (0.39)Small -0.456 -4.243 -1.000 -1.459 -1.868 -3.672 0.075(3.55)*** (2.49)** (4.64)*** (2.63)*** (3.35)*** (2.75)*** (0.25)Medium -0.219 -1.433 -0.570 -1.072 -1.623 -3.120 0.02567

Final Report June 30, 2004(1.65)* (0.81) (2.54)** (1.85)* (2.80)*** (2.24)** (0.08)2003* Small -0.633 4.249 -0.329 0.905 1.308 2.132 -0.559(3.17)*** (1.61) (0.98) (1.05) (1.51) (1.02) (1.20)2003*Medium -0.577 1.742 -0.595 0.692 1.123 1.739 -0.199(2.81)*** (0.64) (1.73)* (0.78) (1.26) (0.81) (0.41)Riga -0.226 -2.157 -0.242 0.143 -0.092 -2.206 0.530(2.77)*** (1.99)** (1.76)* (0.41) (0.26) (2.58)** (2.78)***2003* Riga 0.039 2.053 0.169 -0.196 -0.023 1.880 -0.110(0.33) (1.32) (0.86) (0.39) (0.04) (1.53) (0.40)2003 0.215 -2.872 0.077 -1.412 -0.427 -2.134 0.381(0.85) (0.86) (0.18) (1.28) (0.39) (0.81) (0.64)Constant 0.970 1.252 1.215 2.095 1.335 1.913 0.177(5.29)*** (0.52) (3.95)*** (2.57)** (1.68)* (1.00) (0.41)Observations 882 878 879 880 881 878 880Adjusted R- 0.10 0.08 0.10 0.02 0.03 0.13 0.01squaredNote: Absolute value of t statistics are in parentheses. * significant at 10%; ** significant at 5%; *** significant at1%.Table 3.3: Hours spent in Inspection-Model 3(1) (2) (3) (4) (5) (6) (7)Labor Sanitary Fire Construction Environment SRS PoliceManufacturing 64.748 74.690 18.719 -13.965 1.047 -158.531 4.245(2.26)** (1.39) (2.21)** (0.47) (0.04) (1.43) (0.54)Commerce 15.359 90.648 11.349 7.558 26.933 -265.398 16.326(0.64) (2.02)** (1.62) (0.30) (1.08) (2.87)*** (2.47)**Services 2.524 19.868 2.937 11.095 29.679 -319.552 12.260(0.10) (0.44) (0.42) (0.44) (1.18) (3.43)*** (1.84)*2003*-65.373 -77.202 -17.895 13.605 -1.438 161.844 -3.714Manufacturing(1.60) (1.01) (1.49) (0.33) (0.03) (1.03) (0.33)2003* Commerce -16.238 -93.592 -10.803 -7.392 -27.273 271.719 -15.673(0.47) (1.46) (1.08) (0.21) (0.77) (2.07)** (1.67)*2003* Services -3.073 -22.785 -2.338 -11.295 -30.064 321.630 -12.145(0.09) (0.36) (0.24) (0.32) (0.86) (2.48)** (1.31)Export -26.648 2.012 -7.655 -48.394 -46.506 -173.354 2.284(1.28) (0.05) (1.24) (2.30)** (2.17)** (2.10)** (0.40)2003* Export 26.484 -3.101 7.498 48.210 47.318 183.506 -2.694(0.86) (0.05) (0.82) (1.53) (1.48) (1.52) (0.32)Foreign -0.699 -75.768 -6.639 75.517 70.159 218.428 -14.567(0.03) (1.92)* (1.07) (3.55)*** (3.24)*** (2.60)*** (2.55)**2003* Foreign 2.098 79.526 5.797 -76.114 -69.945 -221.433 14.302(0.06) (1.28) (0.60) (2.27)** (2.05)** (1.72)* (1.58)Year in business 0.993 3.522 -0.057 -0.063 0.070 -1.932 -0.205(1.30) (2.46)** (0.25) (0.08) (0.09) (0.64) (0.99)68

Final Report June 30, 2004Small -28.470 -112.037 -38.582 -60.405 -80.541 -447.330 -16.006(1.41) (2.96)*** (6.39)*** (2.97)*** (3.87)*** (5.87)*** (2.91)***Medium -20.337 -108.859 -31.125 -55.842 -72.022 -382.443 -6.202(0.97) (2.75)*** (4.95)*** (2.64)*** (3.33)*** (4.79)*** (1.08)2003 * Small 28.514 124.344 35.957 58.689 79.526 430.262 14.911(0.92) (2.15)** (3.94)*** (1.87)* (2.49)** (3.65)*** (1.77)*2003 * Medium 20.983 118.745 29.414 54.396 71.106 368.270 5.825(0.66) (1.99)** (3.13)*** (1.69)* (2.17)** (3.03)*** (0.67)Riga -9.825 -42.552 -11.857 4.483 -2.529 -143.131 5.787(0.77) (1.79)* (3.16)*** (0.35) (0.19) (2.91)*** (1.67)*2003*Riga 10.028 42.562 11.733 -4.634 2.412 142.607 -6.020(0.55) (1.25) (2.20)** (0.25) (0.13) (2.04)** (1.21)2003 -45.039 -113.585 -40.303 -51.455 -56.071 -811.174 -5.287(1.16) (1.56) (3.54)*** (1.28) (1.39) (5.45)*** (0.50)Constant 41.805 90.666 43.216 53.677 57.214 835.836 7.270(1.47) (1.70)* (5.16)*** (1.81)* (1.93)* (7.66)*** (0.93)Observations 871 870 863 878 875 865 870Adjusted R- 0.02 0.03 0.08 0.02 0.02 0.08 0.02squaredNote: Absolute value of t statistics are in parentheses. * significant at 10%; ** significant at 5%; *** significant at1%.69

Final Report June 30, 2004Table 3.4: Numbers of Inspectors Visited-Model 3(1) (2) (3) (4) (5) (6) (7)Labor Sanitary Fire Construction Environment SRS PoliceManufacturing 0.257 -5.138 0.807 -0.322 0.045 -6.457 0.149(1.10) (1.27) (3.29)*** (0.40) (0.06) (1.69)* (0.48)Commerce 0.258 6.703 0.521 0.179 0.565 4.895 0.882(1.32) (1.97)** (2.53)** (0.26) (0.85) (1.52) (3.41)***Services -0.066 2.392 0.144 0.169 0.668 0.458 0.267(0.33) (0.70) (0.69) (0.25) (1.00) (0.14) (1.02)2003 *-0.081 4.741 -0.593 0.123 0.160 6.211 -0.129Manufacturing(0.24) (0.82) (1.69)* (0.11) (0.14) (1.14) (0.29)2003 * Commerce -0.020 -6.917 -0.160 -0.369 -0.537 -4.999 -0.738(0.07) (1.42) (0.54) (0.38) (0.57) (1.10) (2.00)**2003 * Services 0.189 -2.809 0.119 -0.400 -0.620 -0.813 -0.306(0.68) (0.58) (0.41) (0.42) (0.66) (0.18) (0.84)Export -0.123 10.700 -0.195 -1.275 -0.927 10.226 -0.265(0.72) (3.63)*** (1.09) (2.20)** (1.61) (3.67)*** (1.20)2003 * Export 0.177 -11.101 0.446 1.277 1.281 -10.310 0.159(0.70) (2.52)** (1.67)* (1.48) (1.49) (2.50)** (0.48)Foreign -0.055 -7.577 -0.143 2.204 1.876 -6.672 -0.596(0.32) (2.53)** (0.80) (3.76)*** (3.25)*** (2.36)** (2.66)***2003 * Foreign -0.152 8.395 0.159 -2.313 -2.007 7.261 0.561(0.56) (1.78)* (0.56) (2.50)** (2.19)** (1.64) (1.58)Year in business 0.009 0.788 0.005 0.003 0.011 0.838 -0.002(1.44) (7.25)*** (0.75) (0.13) (0.52) (8.32)*** (0.23)Small -0.534 -7.643 -0.858 -1.491 -1.869 -6.349 -0.210(3.29)*** (2.69)*** (5.00)*** (2.70)*** (3.40)*** (2.45)** (1.00)Medium -0.312 -6.056 -0.457 -1.192 -1.613 -6.001 -0.284(1.85)* (2.05)** (2.56)** (2.06)** (2.82)*** (2.21)** (1.29)2003 * Small -0.359 10.313 0.474 1.400 1.762 8.993 -0.143(1.42) (2.35)** (1.79)* (1.63) (2.07)** (2.22)** (0.44)2003 * Medium -0.221 8.326 0.380 1.148 1.621 8.154 0.222(0.85) (1.85)* (1.39) (1.30) (1.85)* (1.95)* (0.66)Riga -0.322 -3.655 -0.298 -0.040 -0.118 -4.315 0.714(3.10)*** (2.04)** (2.75)*** (0.11) (0.34) (2.57)** (5.28)***2003 * Riga 0.177 3.541 -0.018 -0.006 0.022 3.841 -0.560(1.19) (1.38) (0.12) (0.01) (0.04) (1.60) (2.88)***2003 -0.018 -7.377 -0.404 -1.071 -1.264 -7.726 0.313(0.06) (1.34) (1.21) (0.98) (1.18) (1.50) (0.75)Constant 1.028 2.365 1.063 1.483 1.393 3.133 0.204(4.44)*** (0.59) (4.37)*** (1.83)* (1.78)* (0.83) (0.67)Observations 875 875 878 877 879 875 876Adjusted R- 0.06 0.10 0.07 0.01 0.02 0.12 0.08squaredNote: Absolute value of t statistics are in parentheses. * significant at 10%; ** significant at 5%; *** significant at 1%.70