determinants of balance of payment fluctuation in nigeria

DETERMINANTS OF BALANCE OF PAYMENT FLUCTUATION
IN NIGERIA
Vincent M. Nwani
Economic Associates,
Victoria Island, Lagos, Nigeria.
Phone: 234 (0) 803 382 7944
E-mail: Vincent_nwani@yahoo.com
ABSTRACT
This study investigates the long-run determinants of balance of payment dynamics in Nigeria
between 1981 and 2002, using econometric method of cointegration and error correction mechanism. It
found that all the variables except balance of payment, exhibited non-stationarity. The results also indicate
that balance of payment cointegrated with all the identified explanatory variables, suggesting that balance
of payment fluctuations in Nigeria could be caused by the level of trade openness, external debt burden,
exchange rate movement and domestic inflation. We concluded that a reduction in fiscal deficits, an
increased domestic production through private investment, inflation targeting and regulated capital market
integration are the panacea to the negative fluctuation in the Nigerian balance of payment.

1. Introduction
Balance of payment (BOP) is a statistical statement that systematically
summarizes for a specific time period, the economic transaction of an economy with the
rest of the world. It reflects changes in the claims and liabilities of an economy vis-à-vis
the rest of the world that are ascribed to transactions. Over the last two decades, there has
been growing trend in the fluctuations of the Nigerian BOP. We recognize the fact that
quite a lot of research has been carried out on the Nigerian BOP, but most of these studies
have been at the theoretical level, while others addressed the consequences of BOP
deficit. None of these studies empirically modeled the long run determinants of BOP
fluctuation. Moreover, the impact of these alleged consequences during policy
discussions has led to proposals aimed at attacking these consequences, not the causes of
BOP fluctuations. This study attempts to cover this gap.
BOP crisis distorts the workings of the entire system (economy) because it creates
disequilibrium between the supply and demand for money. BOP disequilibrium is a
refection of disequilibrium in the money market (IMF, 2000). Monetary disequilibrium
produces adverse effect on the aggregate expenditure for goods and services (absorption)
in the sense that, if the public has an excess supply of money it gets rid of it by passing its
excess cash balance to foreign countries in exchange for goods and services. If the public
desires to keep more money than it has in stock, it achieves it by reducing absorption and
ultimately passes goods and services on in foreign countries in exchange for money.
The above explanation raises several research questions; what are the causes and
solutions of BOP fluctuation in Nigeria? Does currency depreciation induce relative price
change, raising the domestic currency price of exports and imports and encouraging
resources to move into the traded goods sector? The broad objective of this study is
therefore, to find the long-run causes of BOP fluctuations in Nigeria and suggest various
ways of ameliorating this recurring phenomenon. The rest of the article is organized into
four sections II contains the causes and trends of BOP fluctuations in Nigeria. The
theoretical framework is present in section III. The methodology employed and
estimation results are contained in section IV. Section V concludes with lessons for
policy.
2

2. Determinants of Trend Balance of Payment in Nigeria
The staff note prepared by the IMF committee on balance of payment statistics
(2000) reports that BOP problems are due to the disequilibrium in the physical flows,
namely exports and imports of goods and services. Thus it could be analysed on the
basis of partial elasticities of the exports and imports and the role of exchange rate in the
adjustments of BOP to devaluation. In Nigeria, BOP fluctuation is motivated by factors
such as money illusion, terms of trade, external debt servicing and exchange rate
(devaluation) movement (Olaloku, 1979). The “time series data” presented in table 1
indicates that (BOP) was N-3020.8m (1981), N349.1M (1985), N-5761.5m (1990), N-
195216.3m (1995), N-326635m (1999) and N-565353.3m for 2002.
Table 1 Balance of payment and some other macroeconomics indicators in Nigeria 1981 – 2003
Year BOP
(N’M)
Official
Exchange
Inflatio
n rate
Balance of
trade (N’M)
Trade
openness
Real
GDP
External
debt
Term of
trade
Rate
(N/US$)
(%)
growth growth
1981 -3020. 0.63 21.4 -1816.3 0.47 0.01 0.24 329.7
1983 -301.3 0.74 23.2 -1401.2 0.02 -0.05 0.19 345.0
1985 349.1 0.99 5.5 4658.2 0.26 0.00 0.18 300.0
1987 159.2 4.19 10.2 12498.9 0.45 -0.01 1.43 252.3
1989 -22994.2 7.65 40.9 27111.0 0.39 0.07 0.79 187.3
1991 -101407.3 9.70 13.0 34515.2 0.65 0.04 0.09 135.1
1993 -24060.4 22.63 57.2 52669.7 0.55 0.02 0.16 108.2
1995 -195216.3 21.88 72.8 195533.7 0.86 0.02 0.01 100.0
1997 1077.7 21.88 8.5 395946.1 0.73 0.05 -0.035 188.7
1999 -326671.4 92.54 6.6 326499.7 0.38 0.01 3.07 107.6
2000 314139.1 109.55 6.9 982759.4 0.60 0.03 0.20 100.4
2001 24729.9 112.48 18.9 643535.8 0.61 0.03 0.03 104.3
2002 -565353.3 126.4 12.9 302141.0 0.54 0.03 0.19 108.7
2003 -162839.7 135.40 14.0 - 0.48 0.05 0.08 119.3
Sources: (1) Central Bank of Nigeria (CBN) Statistical Bulleting 2003, CBN Annual Report and Statement
of Accounts (various issues). (2) Columns 6, 7 and 8 are computed by the authors.
3

Figure1.
BOP Dynamics in Nigeria (1980-2003)
400000
200000
0
-200000
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
BOP (N’M)
-400000
-600000
-800000
In an attempt to identify the long-term causes of BOP fluctuation in Nigeria, the
vulnerability of the economy to external shocks, external debt burden and debt servicing
issues, inflationary effects, trade openness and exchange rate movements have remained
the focal issues. BOP adjustment through exchange rate changes relies upon the effect of
the relative prices of domestic and foreign goods on the trade flows with the rest of the
world (Thrillwall, 2004). This relative price, or terms of trade is defined by the ratio of
export and import prices in domestic currency from the point of view of the country as a
whole, the terms of trade represents the amount of imports that can be obtained in
exchange for a unit of exports
or the amount of exports required to obtain one unit of
imports. The terms of trade may vary both because of change in the prices expressed in
the respective national currencies and because of exchange rate changes. Thrillwall
(2004) noted that depreciation in the exchange rate at unchanged domestic and foreign
prices in the respective currencies makes domestic goods cheaper in the foreign markets
and foreign goods more expensive in the domestic market.
Soderstan (1989) contents that devaluation tends to make imports more expensive
in domestic currency terms, which are not matched by a corresponding rise in export
prices. This implies that the terms of trade will deteriorate. Deterioration in the terms of
4

trade represents a loss of real national income and can lead to BOP crisis because more
units of exports have to be given to obtain one unit of imports. Hence, the terms of trade
effects caused by devaluation lowers income. A devaluation of currency causes an
increase in the import prices and general price level. This initiates reduction in the real
value of wealth held in monetary form such that the real value of cash balance is reduced
leading to unfavourable BOP.
Chachodiades (1978) maintained that money illusion and expectation effects can
induce BOP fluctuation because real income does not change due to proportionate
increase of price and money income. The direction of the change depends on the type of
money illusion. Money illusion inhibits real activities though these effects are significant
only at the short run. Therefore, if people are unconscious of the workings of money
illusion, they will likely change their absorption. It is possible that economic agents in
Nigeria regard the increase in prices induced by currency devaluation as likely to spark
further price rises. This has consequently resulted to an increase in direct absorption,
which has worsened the country’s balance of payments.
Nigerian external debt burden and external debt servicing over the last two and
half decades has been blamed by several authors for the negative profile of the country’s
BOP. Dell and Lawrence (1980) stressed that a potential and increasingly significant
source of demand deflation, which might be induced by currency depreciation in
developing countries, arises from its effects on debt servicing. Clearly, the Nigerian
external debt obligation is large and the interest payable keeps rising over years. The
implication of this circumstance is that debt service expenditure reduces wealth and the
resource available to improve the country’s real activities, which is detrimental to BOP.
Inflationary effects caused by currency depreciation might be expected to have an
expenditure reducing impact (Dornbusch, 1992). Reduction in real expenditure will occur
only if the appropriate monetary policy is simultaneously pursued (Fakiyesi, 1996). But
over years, inflation policies and targets in Nigeria has failed to achieve its desired
objectives of correcting BOP disequilibrium due to misspecification of macroeconomic
policies and insufficient time lag. The monetary approach to the balance of payment sees
the monetary implications of exchange rate depreciation as being absolutely crucial. But
depreciation becomes unnecessary provided sufficient time (that is financing) is available
5

for automatic correction to occur. According to this approach the mechanism by which
depreciation affects the BOP is by raising the domestic price level and thereby increasing
the demand for nominal money balances (Ikhide, 1993).
The messages from the review above are many. First BOP can be caused by many
factors, notably, monetary, fiscal and structure of factors. Second, in an economy like
Nigeria that is already beset with trend unfavourable balance of payment (BOP). It would
appear more reasonable to look into the long run determinants of BOP rather than the
short run causes. Finally, it would be better to incorporate all the possible factors into one
model and disaggregate the variables to identify the exact points that deserve attention.
6

3. Theoretical Framework
A theoretical rational for the BOP variability is the disequilibrium between
domestic income and expenditure. The absorption approach focuses on the fact that
current account imbalances can be viewed as the difference between domestic output and
domestic spending (absorption).
A = X– M – Y – A ……………….. (1)
Understanding how devaluation affects both income and absorption is therefore
central to the absorption approach to the balance of payments. If devaluation raises
domestic income relative to domestic spending the current account will improve. Machup
(1960) formalized this possibility in his article “Balance of payments and the so called
dollar shortage”. He stated that if devaluation raises domestic absorption relative to
domestic income the current account will deteriorate. Machup assumes that if the
economy is below the full employment level, then there will be an increase in net exports
following devaluation. It is however not clear whether the employment effect will raise or
lower national income.
The elasticity approach holds that BOP problems are due to disequilibrium in the
physical trade flows namely export and imports of goods and services. This approach can
be analyzed on the basis of partial elasticities of the exports and imports and the rate of
exchange rate in the adjustment of BOP to currency devaluation. Thrillwall (2004)
showed that there are two direct effects of exchange rate changes on the balance on goods
and services.
• The increases in the volume (Volume effect) of exports due to the increase in
the price competitiveness of the exports and the decrease in the volume of
imports due to the decrease in the price competitiveness of imports subject to
the devaluation. The volume effect clearly contributes to improving the goods
and services account.
• Due to the devaluation (Price effect) exports become cheaper measured in
foreign currency and imports become more expensive measured in domestic
currency. The price effect clearly contributes to the worsening of the goods
and services account.
7

If the suitable conditions on the elasticities are fulfilled the balance of payments ought to
improve. However, it may happen that quantities do not adjust as quickly as prices,
owing to frictions and reaction lags of both consumers and investors; it takes time for
consumers in both devaluing country and the rest of the world to respond to the changed
competitive situation. Due to these facts the balance of payments may again deteriorate
before improving towards the new equilibrium points.
(Quantity adjustment periodic is defined as the period in which both quantities and
prices can changes)
Adopting the general framework of BOP as described by Kallon (1994), the
simple open-economy LM model is employed to derive the long-run BOP equation. The
equation is of the form:
Y t =
∝ 1 r t + ∝ 2 g t + ∝ 3 P f t + ∝ 4 Y t-1 (∝ 1 0)……………………….(2)
Where equation (2) is the commodity market equilibrium, which is assumed to
depend on the domestic interest rate (r), level of government spending (g), the
relative price of imported goods (P f ) and the real income (Y) of the previous year.
On the other hand the money market equilibrium (LM) equation is of the form:
M t = B 1 Y t + B 2 Y t + B 3 Op t (B 1 >0, B 2 , B 3

The search for a reliable BOP function continues to be an intensive activity. In
Nigeria, very little is known about the contemporary relationship between BOP and other
macroeconomics variables. The BOP function adopted in this study, therefore, combines
the structuralist, monetarist, and fiscalist approaches. Given the structure of the Nigerian
economy since financial deregulation and trade liberalization, (1985 – 1986) as well as
voluminous empirical evidence on Nigerian and other nations of the world, we specify
the following long –run BOP function.
BOP = ∝ o + ∝ 1 EXR t + ∝ 2 INF t + ∝ 3 BOT t + ∝ 4 TOP t + ∝ 5 Y t + ∝ t
EXD t + ∝ 7 TOT t + U t ………………. (5)
Where BOP = Balance of payment, EXR = Exchange rate (N/$), TOT = Term of trade,
BOT = Balance of trade (Export – Import), TOP = Trade Openness (Total trade/Norminal
EDP), Y = Real GDP Growth, EXD = External Debt Growth, U = White noise error
term.
Yearly, data were used to estimate equation (5) above and the estimation sample
is 1980 through 2003. All the time series data employed were gathered from CBN
publications such as monthly and annual reports and statistical bulletin (for various years)
and augmented by relevant publications of the Federal Office of Statistics (FOS) and the
international financial statistics of the IMF.
The main question here is whether macro economic variables such as BOP,
exchange rate, inflation rate, external debt, prices and so on, maintain a long-run trend
following a shock, or do they follow a random work? The essence of this question is that
if these variables follow random works or have different orders of integration a
regression of one variable against other variables can lead to spurious results since the
OLS estimates would definitely not yield consistent parameter estimates. Therefore, the
requirement, in modeling BOP, is an approach that captures the long run relationship
between the variables while avoiding spurious inferences. Both the Augmented Dickey –
Fuller (ADF) and Phillips Peron (PP) unit root tests were employed to determine the
order of each series. The critical t value for ADF tests was chosen by the Akaike (AIC)
9

final prediction Error (FPE) criterion while PP tests are estimated for alternative Bartlett
Karmeel truncation lags.
In the event of non-stationarity in the series, we then conduct cointegration tests.
The implication behind this series of tests is that if each time series is non-stationary,
linear combinations of such time series that are stationary might exist. In that case,
multiple times series are cointegrated and share some common stochastic trends. We can
interpret the presence of cointegration to imply that the long run movements in these
multiple time series are related to each other. Indeed, if there is a long-run relationship
between two or more non-stationary variables then the general concept of cointegration is
that deviations from this long run path are stationary. If this is the case, the variables in
question are said to be cointegrated. A time series can however, only be cointegrated if it
is integrated of the same order (Eagle and Granger, 1987).
The fact that variables are cointegrated implies that there is an adjustment
process, which prevents errors in the long run, relationship from becoming larger and
larger. Engle and Granger (1991) have shown that any cointegration series has an error
correlation presentation, suggesting that cointegration is a necessary condition for an
error correction model (ECM) to hold. Hylleberg and Mizon (1989) and Philips and
Lovetan (1991) have given a detailed analysis of cointegration and error correction
mechanism. Indeed, such models incorporate both the economic theory relating to the
long run relationship between variables and short run disequilibrium behaviours.
The starting point of ECM is a test against the null hypothesis that the residuals of
the long run (static) model are non-stationary. Here, we apply ADF tests to the residuals
of the static cointegration regressions. The evidence of stationarity of the residuals
indicates that the variables in the model are indeed cointegrated. The next move is to
switch to a short run model with an error correction mechanism. Adopting the Engel-
Granger representation, we employ an error correction dynamic specification of the form.
∆BOP t = ∝ o + ∝ 1 L (∆Z) - ∝ 2 ECM t-1 + e t …………. (7)
10

Where Z is the vector of variables that cointegrate with BOP, L is a general lag operator
and ECM is the time series of residuals from the cointegrating vector.
Equation (7) incorporates a corrective mechanism by which previous
disequilibrium in the relationship between the level of BOP and the level of one or more
determinants are permitted to affect current change in BOP. In this way, an allowance is
made for any short run divergence in BOP from its long run trend. Equation (7) can then
be reduced to a parsimonious equation through the elimination of insignificant terms and
the imposition of constraints that holds a reasonable approximation (see Adam, 1992 and
Bougton, 1991).
4. Model Results and Discussion
4.1 Unit Root Tests
Previous evidence considering the stationarity of some macro economic variables in
Nigeria provided mixed results. For instance, Folorunsho and Abiola (2000) showed that
inflation, exchange rate and real GDP in Nigeria between 1970 and 1998 are stationary at
levels. Ariyo (2004) discovered that over the period (1970 – 2001, Balance of payment is
stationary, but over a shorter sub period (1986 – 2001) the Augmented Dickey Fuller test
cannot reject the unit root –null hypothesis. Table 2 presents the ADF results with the
evidence that balance of payments, balance of trade, external debt growth, exchange rate
and real GDP series are stationary at their levels 1(0), while other variables in our model
assumed stationarity at their first difference 1(1). Also, the PP unit root results presented
in Table 3 provides that balance of payment, balance of trade, external debt growth and
inflation series are stationary at their levels, While other variables in our model are
stationary at their differences. Thus, the method suggested by Evans (1991) to decompose
macro economic variables into long run and short run components based upon the
11

assumption of unit root in the level of macroeconomic variables are applicable to this
sample.
Table 2. Augmented Dickey Fuller (ADF) Unit Root Tests
Variables T - statistics P. values Remark
Levels Differences
BOP -11.0481 * 0.00 1 (0)
BOT -6.04591 * 0.00 1 (0)
ExD -1.36839 -10.815 * 0.00 1(1)
ExDG -5.01357 * 0.00 1(0)
ExR -5.20405 * 0.00 1(0)
INF -2.33505 -4.63872 * 0.07 1(1)
TOP -0.83412 -5.04206 * 0.04 1(1)
TOT -1.58857 -493488 * 0.02 1(1)
Y -4.17209 * 0.01 1(0)
Note: one star implies significance at 1% (critical value = 4.49831).
Table 3. Philips Perron (PP) Unit Root Tests
Variables T - statistics P. values Remark
Levels Differences
BOP -3.22995* 0.03 1 (0)
BOT -4.19446*** 0.01 1 (0)
ExD -0.47767 -4.22065** 0.01 1(1)
ExDG -5.02412* 0.00 1(0)
ExR -1.35243 -8.31185* 0.00 1(2)
INF -2.34651 -5.85384* 0.00 1(1)
TOP -3.2899*** 0.09 1(0)
TOT -1.57315 -8.42084* 0.00 1(1)
Y -3.07864 -11.0368* 0.00 1(2)
Note: One star implies significance at 1% (critical value = -4.4679), two stars implies
significance at 5% (critical value = -3.64496) and three stars implies significance at 10% (critical
value = -3.26145).
Following our findings in section 4.1 that all the variables of interest are
stationary at difference {i.e.1(1) and I(2)} except for BOP, we tested for possible
12

cointegration among these variables. Adopting Engle and Granger’s two-step method, we
first estimated the long-run relation of these variables with BOP by OLS and then tested
for the stationary of the residuals. We found that there is a long-run relationship between
BOP and its determinants through the stationarity {1(0)} of the residual series at its
levels. The OLS estimations result and summary statistics is presented in table 4.
Table 4: OLS Estimation Results and Summary statistics of BOP model Eq. (5).
Variables Coefficient STD error T-statistics Probability
Constant 2028607 534412.4 3.795 0.00
BOT 0.743 0.049 6.597 0.00
EXD -0.061 25714.7 -1.2445 0.23
EXDG -47901.7 9987.6 -1.862 0.08
EXR -14222.2 1202.6 -1.932 0.027
INF -4262.6 180323.8 -3.544 0.00
TOP -207463 594.8 -1.15 0.27
TOT 1874.2 6.143 3.15 0.00
Y 18.731 6.559 2.855 0.01
Note: R 2 = 0.83, F = 8.06, DW = 1.61, Observation =22.
The cointegration results reported in table 5 indicate that all the variables in our
model are cointegrated with BOP. It was noted that the lagged series of the explanatory
variables did not survive the resulting Schwarz information criterion (not statistically
significant). We observed that both the residual and its lagged series (ECM and ECM1)
proved statistically significant. This further confirms the long-run relationship between
BOP and all the tested determinants. The cointegration result is presented in table 5.
13

Table 5: Cointegration results (Modeling BOP by Engle and Granger’s Two –step method)
Variables Coefficients STD Error T-value P-value
Constant -1184897 106415 -3.31 0.00
BOT 0.743 1.29 5.75 0.00
BOT 1 0.196 0.156 1.254 0.256
BOT 2 -0.826 0.182 -4.53 0.004
EXD -0.061 5.63 -9.01 0.00
EXD1 1286.9 376.52 0.034 0.973
EXD2 -2382.8 36838 -0.646 0.541
EXDG -47901 2.95 -6.21 0.00
EXDG 1 0.284 0.493 0.577 0.421
EXDG 2 0.811 0.448 0.634 0.389
EXR -1422 1.15 2.41 0.00
EXR 1 1725 1629 1.059 0.33
EXR 2 -11576 14175 -0.816 0.445
INF -4262 1.38 -3.09 0.00
INF 1 -1135 1586 -0.714 0.501
INF 2 353.6 1555 0.227 0.827
TOP -207.4 2.07 -3 0.00
TOP 1 0.193 3.271 1.13 0.541
TOP 2 -0.384 5.86 -0.691 0.32
TOT 1674 6.83 2.75 0.00
TOT 1 1867 1127 1.656 0.14
TOT 2 382.8 1056 0.362 0.72
Y 18.73 7.53 2.49 0.00
Y 1 10.64 12.34 0.862 0.729
Y 2 -2.68 11.02 -0.243 0.421
Note: R2 = 0.931, F = 6.263, DW = 2.596, SE = 0.090 Information criteria: SC = -2.14, FPE = -0.147.
ECM Result
Variables Coefficients STD Error T-value P-value
ECM -0.41 0.44 -2.439 0.073
ECM 1 -0.3 0.114 -2.635 0.099
The results presented in table 5 clearly indicates a well-defined error correction
term ECM t-1 and ECM t which indicates a feedback 0.30 and 0.41 of the previous and
current year’s disequilibrium from the long-run elasticities of the identified variables that
courses the fluctuation of BOP. The empirical evidence shows that these variables
14

provide explanation for the fluctuation of BOP in Nigeria through time. The effect of this
error correction is not only large but also exhibited a negative influence as expected at 1
per cent level of significance. The constant coefficients of the variables also reveal that
exchange rate, external debt, inflation and trade openness relates negatively with BOP
over time in Nigeria. Conversely, balance of trade, term of trade and real GDP
maintained a positive relationship with BOP during the sample period.
5. Conclusion and Policy Lessons
Results from our empirical analysis provide support for the unit root alternative
hypothesis hence these variables are non-stationary except for BOP, and indeed, they are
of random walk. Given the non-stationarity of these series, the cointegration equation was
estimated. The evidence confirms that balance of payment cointegrates with exchange
rate, balance of trade, term of trade, inflation rate, real GDP, external debt and trade
openness.
In order to put the Nigerian economy on the path of sustainable growth and
development as well as reduce balance of payment fluctuations, the government has to
reduce fiscal deficits, encourage
private investment, target inflation and address the
perennial external debt problem. Bearing in mind both, the practical problems associated
with exchange controls, it seems reasonable to conclude that while exchange rate
devaluation may in some cases, provide a useful short run tourniquet in the event of a
balance of payments crisis, they do not generally offer a cost effective measure of
correcting deficits. Indeed their principal effect is adverse and suppresses imports rather
than the positive one of encouraging exports.
15

REFERENCES
Adam, C. S. (1992) Recent Developments in Econometric Methods: An Application
to the Demand for Money in Kenya, African Economic Research
consortium, AERC special papers, 15, Pp. 1- 52.
Boughton, J. M. (1991), Long run Money demand in Large Industrial Countries, IMF
staff papers 38; Pp. 1 – 32.
CBN Annual reports (various issues)
CBN statistical Bulletin (2002)
Chachodiades, M. (1978), International Monetary Theory and Policy, New York,
Mcgraw Hill.
Dell A. and R. Lawrence (1980), Balance of Payments Adjustment process in developing
countries (New York: Pergoman)
Dornbusch, R. I (1992) Lessons from Experiences with High Inflation, The World Bank
Economic Review 6(1)
Engel, R. F. and C. W. J. Granger (1987), Cointegration and Error-correction:
Representation, Estimation and Testing, Econometrics 55, Pp. 251 –76.
Engel, R. F. and C.W.J. Granger (1991), Long run Economics Relations: Readings in
cointegration, Oxford University Press London.
Fakiyesi, T. O. (1996), Further empirical analysis of inflation in Nigeria, CBN Economic
and Financial Review 34 (1), Pp. 489- 499.
Hylleberg, S. and G. F. Mizan (1989) cointegration and Error Correction Mechanism,
The Economic Journal, 99, Pp.113 – 125.
Ikhide, S. I (1993) Financial Liberalization and the Inflationary process; is there a
link in Adjusting Developing Countries? International Review of Economic
and Business, 40, Pp. 157 – 174
IMF (2000), Training in the use of Balance of Payments Statistics – staff notes,
Thirteenth meeting of the IMF committee on Balance of Payments Statitistics,
Washington, D.C. October 23 – 27, 2000.
Kallon, K. M. (1994), An econometrics analysis of inflation in Sierra Leon, Journal of
African Econometrics, 3 (2), Pp. 199 – 230.
16

Machup, F. (1960), Three Concepts of Balance of Payments and the so-called Dollar
shortage, Economic Journal Vol. LX March, Pp.46 – 68.
Olaloku, F. A. (1979), structure of the Nigerian Economy, Lagos Macmillans
Soderstan, R. O. (1989), International Economics, New York, Harper and Row, PP. 226
– 241,
Thrillwall, A.P.(2004), Trade ,Balance of Payments and Exchange Rate Policy In
Developing Countries, Edward Edger Publishing, UK.
17