Transcampus.com

advert
Home Instructors Journals ContactUs
Home

 

Instructors

 

Journals

 

Contact Us

 

JOURNAL OF RESEARCH IN NATIONAL DEVELOPMENT VOLUME 7 NO 1, JUNE, 2009

LONG RUN EXCHANGE RATE EFFECTS ON TRADE BALANCE.

EVIDENCE FROM NIGERIA

 

BigBen Chukwuma Ogbonna

Department of Economics, Ebonyi State University, Abakaliki

 

Abstract

This paper investigates Nigeria’s exchange rate-trade balance relationship on the long-run for the period 1960-2005. We employ unit root, cointegration and error correction model (ECM) procedures to estimate the peculiar trade balance equation which includes income and money stock so that monetary and absorption approaches to balance of payments are also examined. The benchmarks results indicate that exchange rates do play a role in determining the long-run equilibrium behaviour of Nigeria’s trade balance, even though such relationship is faint. Furthermore, the results show that real exchange rate devaluation improves trade balance which suggests that ML condition cannot be rejected for Nigeria. The results further suggest that the long-run effect of exchange rate devaluation on the trade balance is enhanced if accompanied by contractionary fiscal and monetary policies.

 

 

Keywords: Trade balance; Money Stock; Real Exchange Rate, cointegration.

 


 Introduction

Prior to the early 1970s, Nigeria was a predominantly agrarian economy depending on the agricultural sector for the sustenance of its citizenry and the external sector.  The oil boom of the mid-seventies led to large earnings in foreign exchange, and changed the economic profile of the country. When Nigeria became an oil exporting country, agriculture was relegated to the background.  In the words of Atoloye, this relegation was as a result of inappropriate exchange rate policy which made the prices of agricultural output too low to give farmers the incentive to produce. During this period, the Nigerian exchange rate policy tended to encourage over-valuation of the Naira which in turn, encouraged imports, discouraged export and helped in sustaining the manufacturing sector’s over-dependence on imported inputs (Obadan, 1993). Exchange rate policy during this period was not geared towards the attainment of long-run equilibrium rate that would equilibrate the balance of payments in the medium and long term and facilitate the achievements of export diversification and discourage over-independence on imported manufacturing inputs; but rather the reverse was the case.

 

However, the collapse of oil prices in the early 1980s and the build up of unimaginable external trade and trade arrears made the change, which culminated in the adoption of Structural Adjustment Programme (SAP) in 1986 quite imperative.  For instance, Nigeria’s external sector was over heated and fragile and was characterized by over valuation of the naira exchange rate, accumulation of trade arrears, continuous decline in foreign exchange earnings and increasing debt service obligations that resulted from excessive debt burden then known as ‘debt overhang’.

The objective of exchange rate policy was derived from the overall objective of macroeconomic management to achieve internal and external balance in the medium term.  Internal balance means the level of economic activity consistent with the satisfactory control of inflation.  (Williamson, 1982), while external balance means balance of payments equilibrium or sustainable current account deficit financed on a lasting basis by expected capital inflow (Komolafe, 1996).

This objective was in the context of the second-tier foreign exchange market (SFEM) introduced in September, 1986.  The SFEM was predicated on the attainment of a realistic exchange rate of the naira through depreciation by the market exchange.  It was then expected that a realistic exchange rate would reduce excessive demand for foreign exchange, especially for importation of finished goods and services and stimulate export.

 

Although many factors may have combined to explain the general adverse macroeconomic condition, the exchange rate policy of a country has been frequently identified as a major contributor (World Bank, 1984).  Since the exchange rate policies of Nigeria has left the domestic currency in constant depreciation without stability, with the nation’s imports and  exports remaining non-responsive to this scenario, the problem which this study intents to investigate concerns the long-run exchange rate effect on trade balance of Nigeria.

This remaining part of this work will be structured as follows:  Section two, presents the review of related literature; section three, the methodology; section four, empirical results and discussion, section five: summary and conclusion.

 

Theoretical Review

There exist vast theoretical and empirical studies on developed economies on how the trade balance responds to adjustment in exchange rate (see Rosenweiig and Knock; 1998, Rose and Yellen, 1989; Kim, 2001; Bahmani-Oskooee and Ratha, 2004). Also studies on developing countries to this effect have proliterated considerably (see Bhmani-Oskooee, 1985; Upadhyaya and Dhakal, 1997; Senhadji, 1998; Rawlins and Praveen, 2000; Sinah, 2002; Musila and Newark, 2003; Naraya and Naraya, 2004 and Agbola, 2004). There are three approaches to the analysis of the effect of exchange rate depreciation on trade balance.

 

The elasticity approach

128

This approach credited to Robinson (1947) and Meltzler (1948) and popularized by Kreuger (1983) posits that transactions under contract completed at the time of devaluation or depreciation may dominate a short-run change in trade balance causing it to deteriorate in the short-term. But in the long-run (as elasticity of exports and imports rise, the adjustment in export (increase) and imports (fall) quantities cause the demand for import to fall after substantial lags and subsequently results in an improvement in the trade balance of the country devaluing its currency-the J-curve effect. Empirical evidence on J-curve hypothesis is mixed (ie having both bad and good qualities). In this regard, this work will not venture into investigating and testing for the existence of J-Curve relationship in Nigeria, but however, will estimate the long-run effect of exchange rate depreciation on the country’s non-oil balance of trade. Another aspect of the elasticity approach to exchange rate-trade balance relationship is the Marshall-Lerner (ML) condition (Marshall, 1923; Lerner, 1944). The theory that a devaluation of a domestic currency will improve current account balance is founded on a number of elasticity approach models, the most popular of which is the Marshall-Lerner condition. The condition states that for balance of trade to benefit from a currency devaluation or depreciation, the summation of the price elasticities of imports and exports of the country devaluing its currency must be greater than unity. The model is:

ηx + ηm > 1. ……………………………………………………………….……..2.1.

 

If it is equal to unity, the balance of payments remains stable, if it is less than unity the balance of payments worsens, but if it is in excess of unity, the trade balance improves. The theory states that when a country devalues its currency, the domestic prices of its imports are raised and the foreign prices of its exports are reduced. These will work to reduce the domestic demand for imports and increase the external demand for the domestic exports to improve the trade balance. Empirical evidence shows that Marshall-Lerner condition is satisfied in majority of advanced economies, but it is a general consensus among economists that both demand and supply elasticity will be greater in the long-run than in the short-run.

 

The Absorption Approach

The absorption approach emphasizes on changes in real domestic income as a determinant of a nations balance of payments –exchange rate relationship. It treats prices as constants and therefore all variables are in real term. This approach disaggregates expenditures into consumption (c), investment (i), government expenditure (g) and imports (m). The sum of these four expenditure variables is defined as the domestic absorption (a), which in equation form is expressed thus:

            a    =    c  +  i   +  g + m …………………………….2.2

A nation’s real income (y) is a reflection of the total expenditure on it’s outputs as expressed below:

 

            y   =    c   +    i   +   g   + x ……………………………2.3

Where: x, represents the value of the domestic exports. The absorption approach, though a simple theory is of great assistance in understanding a nation’s external sector performance in periods of economic contraction and expansions. For instance, if a nation experiences an economic contraction, does its current account balance necessarily improve and its currency definitely appreciate?. In the scenario of economic expansion, if the real income rises thereby increasing absorption, the direction of current account adjustment depends on the relative changes in the two variables. If real income rises faster than absorption, then the current account balance will be exposed to positive adjustment and vice-verse. On the whole, absorption approach stresses real income in balance of payments and exchange rate determination and further suggests that relative changes in real income (output) and absorption, determine a nation’s BOP and exchange rate performance. In a seminal contribution of Johnson, he pointed out that the formulation of the balance of payment as the difference between aggregate payments and receipts identifies the “monetary nature” of the balance of payment. The monetarists would argue that this ought to be expected, as under the quantity theory, there is a fundamental relationship between the money supply and real income.

 

The Monetary Approach

Although David Hume (1952) identified some of the key ideas, the monetary approach essentially is a product of works conducted in the late 1950s, the 60s and early 70s. The approach is concerned with the determination of balance of payments under a fixed exchange rate regime. In the world of monetary approach, devaluation is equivalent to a reduction in money supply; or put differently, devaluation reduces the real value of money stock. Reduction in the money stock directly affects the availability of credit needed to finance budgetary deficits; reduction in government expenditure can in the context of developing economies lead to unemployment, reduction in investment and social services, and general deflation. On the other hand, since there is hardly any developed capital market in developing dependent economies, and there is little or no access to foreign capital markets, excepting for the oil producing countries, if the authorities desire to maintain the level of expenditure in order to sustain the prevailing levels of employment, it means financing the deficit domestically from the banking system. This would mean monetary expansion, which therefore exerts further pressures on the balance of payments through the tortuous process. From a theoretical point of view, since money is a stock, not a flow, the principal contribution of the monetary approach is its emphasis on the effects of exchanges in balance of payments on the stock of financial assets in the economy.

           

129

In a developmental context, it ought to be noted that many developing countries are characterized by peasant and subsistence agric farmers. These

peasants lack the ability and facility to access the complex and sophisticated commercial banking system, and there is hardly any developed indigenous financial institutions designed to meet the demand of the poor peasants. On this note, it could be argued that a substantial portion of the outputs in these economies are perhaps uninfluenced by considerations of the monetary authorities and the money supply. This therefore imposes a limitation on the applicability of the monetary approach to the balance of payments in developing countries. The monetary approach as opposed to the elasticity and absorption approaches, regards capital account as central in the depreciation analysis, relegating the current account. The basics of this approach are that money and asset markets govern the trade balance position through the forces of demand and supply of real money. In a closed economy money stock in excess of the demand, results in excess liquidity in the private sector. This leads to increased expenditure, which if the economy is already operating at full capacity, results in excess demand in the goods market. This generates prices increase to reduce the real money balance until the excess of expenditures over output is eliminated. In the case of an open economy, the increase in the domestic expenditure spills over to the balance of payments (BOF), with the increased demand for goods satisfied by imports. This depletes the volume of foreign reserves and causes the money stock to gradually fall until money holding are reverted to its original level.

           

MS    =   D   +   R    …………………………………………………..……2.4

 

Where: MS is the aggregate money stock, D, is the domestic credit, while R represents foreign reserves. But where the demand for money is in excess of the supply, residents are forced to reduce their consumption expenditures thus improving the trade balance (Bulusivar, 1996). In the short-run, devaluation of domestic currency, raises the prices of domestic goods as the resultant increase in import prices reflects in the prices of domestic import substitutes as a result of the increase in their demands. This higher price level drags the real money stock bringing the demand for money in excess of the supply. Then, for the economic agents to restore their desired level of money balances and financial asset holdings, they reduce their consumption expenditures to spend less and save more. This translates into improved money stock and balance of payments. But as soon as people have restored their desired financial

 

holdings, expenditure will rise, eliminating the trade surplus. In essence, the monetary approach argues that changes in nominal exchange rate can have only temporary effect and that there will be no long-run equilibrium relationship between the trade balance and the real exchange rate (Salasevicius and Vicious, 2003). In effect, for cointegration analysis for trade balance-exchange rate relationship, no long-run relationship between the two variables should be found. The modeling approach in this study should be such that the vector would be able to capture the effects of the exchange rate on the trade balance in a model that puts together the elasticity, absorption and monetary approaches to the balance of payments.

 

 

Empirical Review:

Recently, an upsurge of research has occurred relating to less developed countries, predominantly emerging Asian markets. Baharumshah (2001) employs an unrestricted VAR model for the bilateral trade balances of Thailand and Malaysia with the United States and Japan for the period 1980 to 1996. He finds support for a stable and positive long-run relationship between trade balance and the exchange rate. The evidence on the short-run response of the trade balance supporting the J-curve effect is mixed. A delayed J-curve seems to apply to Thai data, whilst no support for the J-curve was found in Malaysian data. In addition, Bahmani-Oskooee and Kantipong (2001) tested on disaggregated data the J-curve between Thailand and her main trading partners Germany, Japan, Singapore, United Kingdom, and the United States for the period 1973 to 1997. They find evidence of the J-curve curve in bilateral trade with the U.S. and Japan only.

130

Upadhyaya and Dhakal (1997) test the effectiveness of devaluation on the trade balance for eight developing countries (Colombia, Cyprus, Greece, Guatemala, Mexico, Morocco, Singapore, and Thailand) applying the methodology proposed by Wickens and Breusch (1988). Their empirical finding is contradictory to Baharumshah’s findings regarding Thailand and Rincon (1998) regarding Colombia. They provide evidence that seems to suggest that only in the Mexican case devaluation improve the trade balance in the long-run. Bahmani-Oskooee (2001) investigated the long-run response of Middle Eastern countries’ trade balances to devaluation by applying the Engle-Granger and Johansen-Juselius cointegration methodology. The evidence suggests that there exist a favorable long-run effect of a real depreciation on the trade balance for all seven countries (Bahrain, Egypt, Jordan, Morocco, Syria, Tunisia, and Turkey). In the case of Morocco, this opposes the results obtained by Upadhyaya and Dhakal. Wilson (2001) examined the relationship between the real trade balance and the real exchange rate for bilateral merchandise trade between Singapore, Korea and Malaysia with respect to the United States and Japan. No evidence of a J-curve effect was found, with the exception of Korean trade with the United States. In the case of Turkey, Akbostanci (2002) finds support for a favourable long-run relationship between the exchange rate and the trade balance. The generalized impulse response function indicates in the short-run an S-shaped trade balance response to devaluation. This view with respect to the positive long-run relationship is supported in Bahmani-Oskooee (2001). However, Kale (2001) obtains conflicting results, providing evidence of a negative long-run impact of devaluation on the trade balance. In the short-run Kale finds evidence of a delayed J-curve effect.

 

Research related to the relationship between the trade balance and the exchange rate in Nigeria is still scarce. Specifically, the author is yet unaware of any literature that has diagonised Nigeria’s non-oil subsector using trade balance model that encluded income and money stock. Closing this lacuna is the motivation for this paper.

 

 

Methodology                     

 

The Model Specification

Trade balance in most instances is stated as the value of net exports (X – M). In this study we measure non-oil trade balance as the ratio of the values of aggregate non-oil export (nox) to the aggregate non-oil imports (nom). The x/m ratio has been employed in many empirical analyses to determine trade balance exchange rate relationship (Rincon, 1998; Bahmani-Oskooee and Brooks, 1999 and Gupta-Kapoor and Ramakrishnan, 1999). One reason adduced for its use according to Bahmani-Oskooee, 1991, is that this ratio is not sensitive to the unit of measurement and can be interpreted as nominal or real trade balance. In the same vein, this ratio in a logarithmic model yields the exact Marshall-Lerner condition rather than approximation (Boyd et al, 2001). Following the procedure adopted by Rincon, 1998, we specify the bilateral trade balance as a function of real exchange rate, real GDP, real domestic money supply and a (0,1) dummy variable to capture the impact of trade liberalization policies in Nigeria occasioned by the SAP of 1985. The reduced form of the equation expressed in natural logarithm is given as follows:

 

LTB = a0 + a1LRGDP + a2LRER + a3LRMS  + a4DUM + εt……………3.1

 

Where: L is natural logarithm, RGDP is real Gross domestic product, RER is the real exchange rate, RMS represents real domestic money supply, DUM represents policy shift dummy variable that takes the value of zero for the period before 1986 and one for post 1986, a0…a4 represent the explanatory powers of the variables, TB is the trade balance and εt is the stochastic error term. According to Rincon, this vector is thought to capture the effect of exchange rate on the trade balance in a model that puts together the elasticity, absorption and monetary approaches to the balance of payments.

 

Estimation Procedure

The modeling procedures adopted in this study are in three steps:

(i)         Determining the order of integration of the variables employed using Augmented Dickey-Fuller (ADF) and Phillips-Perron (1988) unit root tests,

(ii)        If the variables are integrated of the same order 1(1) we apply the Johansen-Juselius (1990, 1992, 1994) maximum likelihood method of cointegration to inquire the number of cointegrating vector(s) and lastly

(iii)       If the variables are cointegrated, we specify the vector error correction model (VECM) and estimate same using standards and diagnostic tests.

 

Data Source

The analysis is conducted over the period 1960-2005 using annual data. The primary source of data is the Central Bank of Nigeria (CBN) statistical bulletin (2004; 2005) where data are obtained for export and import values and the GDP. The remaining data are source from IMF international financial statistics CD Rom (2006) and Economic Intelligent Unit data base (2006).     

 

 Benchmark Results

In this section, we provide benchmark tests of the significance of the exchange rate in trade balance equations. The tests are derived from empirical estimates of equation 3.1.

 

Summary Statistics

Data on all the employed variables for 1960 – 2005 period are presented in table 4.1 as are their means, standard deviation (SD), and coefficient of variation (CV).


 

Table 1 Summary Statistics of variables used.

Variable

Description

Mean

SD

CV

LTB

LRGDP

LRER

LRMS

DV

Log of Aggregate Trade Balance

Log of Real Gross Domestic Product

Log of Real Exchange Rate

Log or Real Money Supply

Dummy Variable (Trade liberal)

1.302

29.806

59.267

6498.652

0.434

0.542

11.701

28.970

3512.892

0.501

0.416

0.392

0.480

0.540

1.154

Source: Author’s computation based on IMF-IFS, 2006

 


Unit Root Tests

 

In this section, we analyze the time-series properties of the employed variables for the period 1960 – 2005. It is pertinent to test explicitly for the manifestations of non-stationarity, in way as a first step in exploring the status of the data and in the other, because the presence of such non-stationarity at times has important econometric implication. The Augmented Dickey-Fuller (ADF) and Phillips-Perron (PP) Unit Root tests are employed to test the integration level and the possible cointegration amonge the variables (Dickey and Fuller, 1981; Phillips and Perron, 1988).

 

   p

 

i =2

The Phillips-Perron test procedures, which compute a residual variance that is robust to auto-correlation, are employed to test for unit roots as an-alternative to the ADF unit root test. It might seem reasonable to test the existence of a unit root in the series using the most general of the model of the form:

           


∆Yt = ao + gyt-I + a2t + S Bj∆Yt- i - 1+ et ……………….(4.1)

 

Where y is the series; t = time (trend factor); a = constant term (drift);  et = Gaussian white noise and P = the lag order (Soukhakian, 2007).

The ADF and PP tests (table 2 in Appendix) indicate integration of the order 1(1) for LRGDP, LRER and LRMS only at first difference and at 1 percent significant level, while LTB indicates non-stationarity at 1 percent significance level in both level and first difference. 

 

Cointegration Test

The results of the above unit-root tests suggest that all the variables are integrated of the order 1(1). Sequel to these results, we proceed a step further to employ the Johansen (1988, 1991) and Johansen and Juselius (1990, 1992, 1994) to test for cointegration among the variables. The results of trace and maximal eigenvalue tests are as reported in table 4.3 in the appendix. Beginning with the rejection of the null hypothesis of no cointegration ( r = o) among the five variables of FDI, NER, LENDRATE, DOMCR and GDP as both the maximal eigenvalue and trace statistic suggest r = 2. We therefore conclude that there are two co integrating relations or equations among the variables.

Estimates of the long-run cointegrating vectors are given in table 4.

 

 

Estimation of an Error-Correction Model (ECM)    

The results of the cointegration confirm the existence of an underlying long-run stationary steady-state relationship between the explained and the explanatory variables in logarithm. In this instance, ECM is considered the best option for estimation of equation 3.1 to determine the dynamic behaviour of TB, because of ECM’s ability to restrict the long-run behaviour of the endogenous variables to converg to their cointegrating relationships while allowing for short-run adjustment dynamic. For this purpose, Engle-Graiger (1987) methodology is followed and error correction model employed.

 

Table 5 in the appendix reports the ECM results, testing the exchange rate-trade balance relationship on the long-run. The model is estimated using Eview version 6.0 econometric package. The final and most statisfactory results are discussed. All the coefficients of the variables are appropriately signed. The adjusted R2 (goodness of fit) of the value 0.52 indicates that 52 percent of the variance of the dependent variable is explained by the explanatory variables. This suggests that the employed explanatory variables jointly significantly influence the explained variable. The F-statistic value of 10.31 is statistically significant at 1 percent level. The Breusch-Pagan LM test for autocorrelation is 10.48 (0.84) and suggests the presence of no first order autocorrelation at a 1 percent significant level. The estimated long-run exchange-rate elasticity value of 0.01 though miserable is statically significant at 1 percent level. This suggests that a 100 percent (real) devaluation/depreciation of the Naira keeping other variables constant, leads to 1 percent improvement in the aggregate real trade balance, and indicates that ML condition seems to hold for Nigeria. This result is in tandem with the earlier finding of Ogbonna (2008) when non-oil real trade balance of Nigeria was put in perspective. The estimated coefficient for income of -0.07 is statistically significant at 1 percent level. The result indicates that a 100 percent rise in income worsens aggregate real trade balance by 7 percent. This is consistent with what the absorption approach would say, if real income rises, thereby increasing absorption, the direction of current account adjustment depends on the relative changes in the two variables; if real income rises faster than absorption, then the current account balance would be exposed to positive adjustment and vice versa. Thus, the negative signing of the real income coefficient suggests that in case of Nigeria, absorption rises faster than the rise in income in a scenario of economic expansion. Furthermore, the negative sign of the estimated coefficient for the real income variable is inconsistent with what the monetary view would say, that income has a positive relationship with the trade balance. The parameter estimate of real money stock of -0.0002, though miserable is statistically significant at 2 percent level. This suggests that increase in real money stock worsens trade balance on the long-run, but such effect is inconsequential judging from the miserable explanatory power of the variable. The value of the error correction term lagged one period EC(-1) of -0.56 is significantly different from zero at 1 percent level, indicating the validity of a long-run equilibrium relationship among the variables of equation 3.1 and suggests that the system corrects in previous period disequilibrium by about 56 percent a year.

Conclusion

In this paper, we have investigated empirically, the role of exchange rates in determining the long-run behaviour of Nigeria’s trade balance under the alternative approaches to balance of payments. We tested the empirical validity of the hypothesis derived from elasticity, absorption and monetary approaches to balance of payments. Particularly we investigated the validity the Marshall-Lerner (ML) condition employing model specification that incorporated the trade balance, exchange rate, money and income. Implicitly, we tested the empirical relevance of the absorption and monetary approaches for the data set used. The econometric method consist of relatively new technique for analyzing multivariate cointegrated systems originally developed Johansen (1988) –unit root, cointegration and error correction model (ECM). Annual time series data for the period of 1960 – 2005 were analyzed. The results obtained indicated that exchange rates do play a role in determining the long –run equilibrium behaviour of Nigeria’s trade balance. This in effect suggests that trade balance cannot be treated as exogenous with respect to exchange rates. This finding is inconsistent with the literature claiming non existence of direct relationship between trade balance and exchange rates and the monetary view which claims that exchange rate-trade balance effect is temporary. The estimation reported one cointegrating vector between the trade balance, exchange rate, money and income, indicating that there exists a long-run equilibrium relationship between the variables. The results further showed that ML conditions were supported by the data, meaning that real devaluation of the domestic currency improves trade balance. Further more, the positive effect of exchange rate devaluation/depreciation on the trade balance seemed enhanced if accompanied with contractionary fiscal and monetary policies. However, our general finding suggests that exchange rate, money stock and income play minimal role in long-run trade balance dynamics. The current data and technique can be employed to estimate exchange rate-trade balance relationship on the short-run for a comparative analysis.    

 

 

References

 

Agbola, F.W. (2004). “Does Devaluation Improve Ghana’s Trade Balance? Contributed paper presented at the International Conference on Ghana’s Economy, M-Plaza Hotel, Accra, Ghana, 18-20 July.

 

Amisano, G. and Carlo G.  (1997). “Topics in Structural VAR Econometrics,” 2nd ed, Berlin: Springer – Verlag.

 

Bahmani – Oskooee, M. (1985). “Devaluation and the J-Curve: Some Evidence from LDC’s. The Review of Economics and Statistics, 500 – 04.

 

Bahmani – Oskooee. M and Brook T. (1999). “Bilateral J-Curve between U.S. and Her Trading Partners”, Weltwirt schaftliches Archiv Vol. 135 (1), P 156-165.

 

Bahmani-Oskooee, M. (2001). “Nominal and Real Effective Exchange Rates of Middle Eastern Countries and their Trade Performance”, Applied Economics, vol. 33(2), pp. 107-117

 

Bahmani-Oskooee, M. and Alse .J. (1994). “Short-Run versus Long-Run Effect of Devaluation: Error Correction Modelling and Cointegration”, Eastern Economic Journal, vol. 20(4), Fall 1994, pp 453 – 464.

 

Bahmani-Oskooee, M. and Rather, A. (2004). “The J-Curve: A Literature Review”, Applied Economics, 36, 1377 – 1398.

 

Boyd, D., Caporale, G.M. and Smith, R. (2001). “Real Exchange Rate Effects the Balance of Trade: Cointegration and Marshall-Lerner Condition”, International Journal of Finance and Economics, 6, 187 – 200.

 

Dickey, D. (1976). “Estimation and Hypothesis Testing in Non Stationary Time Series”, Unpublished Ph.D. Dissertation, Iowa State University.

 

Dickey, D. A. and Fuller, W.A. (1979). “Distribution of Estimators for Autoregressive Time Series with a Unit Root” Journal of American

Statistical Association, 74; 427-431.

 

Gupta-Kapoor, A. and Ramakrishnan U. (1999). “Is there a J-Curve? A New Estimation for Japan”, International Economic Journal vol. 13(4), pp. 71-79.

 

Hacker, R. S. and Hatemi .A. (2002) “Effect of Exchange Rate changes on the Trade Balance in the short and long-run: Evidence from German Trade with Transitional European Economies”, www.ihh.hj.se/ersasise2002/papers/Hackerand %20Hatermi J Emerging Markets. Pdf

 

Johansen, S. (1988). “Statistical Analysis of Cointegrating Vector”, Journal of Economic  Dynamics and Control.

 

Johansen, S. (1991). “Estimation and Hypothesis Testing of Cointegration Gaussian Vector Autoregression Models”, Econometric vol. 59.

 

Johansen, S. and Juselius, K. (1990) “Maximum Likelihood Estimation and Inference on Cointegration with Application to Demand for Money”, Oxford Bulletin of Economics and Statistics, 52, 169-210.

 

Johansen, S. and Juselius, K. (1990). “Maximum Likelihood Estimation and inference on Cointegration: With Application for the Demand for Money”, Oxford Bulletin for Economics and Statistics. Vol. 59.

 

Juselius, K. (1992). “Testing Structural Hypothesis in a multivariate cointegration Analysis of PPP and UIP for U.K.”, Journal of Econometrics, 53, 211-44.

 

Juselius, K. (1994). “Identification of the Long-run and the Short run Structure: An Application to the ISLM Model” Journal of Econometrics, 63, 7-36.

 


Kanolafe, O. R. (1996). “Exchange Rate Policy and Nigeria’s External Sector Performance”. Nigeria Journal of Economics and Social Studies, Vol. 38.

 

Kim, S. C. (2001). “Effect of Monetary Policy Shocks on the Trade Balance in open European Countries, Economic Letters, 71, 197 – 203.

 

Kruger A. O. (1983). “Exchange Rate Determination”,Cambridge:  Cambridge University Press.

 

Metzler, L. (1948). “The Theory of International Trade”, In Howards, S.E (1984), ed, 1; A survey of contemporary Economics, Philadelphia, Blakiston.

Musila, J. W. and Newark, J. (2003). “Does Currency Devaluation Improve Trade Balance in the Long-run?: Evidence from Malawi,” African Development Review, 15(2), 339-352.

 

Narayan, P. and Narayan, S. (2004). “The J-Curve: Evidence from Fiji”, International Review of Applied Economics, 18(3), 369-380.

 

Obadan, M. I. (1993), Overview of Nigeria’s Exchange Rate Policy Management Service, the Structural Adjustment Programme. CBN Economic and Financial Review. 34 (4).

 

Olugbenga, O. (2003). “Exchange Rate and Trade Balance in East Asia: Is there a J-Curve?,” Economic Bulletin vol. 5(8), pp. 1-13.

 

Paesaran H and Shin Y. (1997). Autoregressive Distributed Lag Modeling Approach to Cointegration Analysis”, DAE working papers Amalgamated Series, University of Cambridge.

 

 

Pesaran, M. H. and Pesaran, B. (1997). “Working with Microfit 4.0: Interactive Econometric Analysis”, Oxford; Oxford University Press.

 

Rawlins, G. and Praveen, J. (2000). “Devaluation and Trade Balance: The recent Experiences of Selected African Countries”, Centre for Economic Research on Africa, School of Business, Montchair State University, New Jersey, U.S.A.

 

Rincon, H. C. (1998). “Testing the Short-and-Long-Run Exchange Rate Effect on Trade Balance: The case of Colombia”, Dissertation Paper (Ph.D), University of Illinois, Urbana-Campaign.

 

Robinson, J. (1947). “The foreign Exchanges”, In essays in the Theory of Employment, 24, 53-68.

 

Rose, A. K and Tellen, J. L. (1989). “Is there a J-curve”. Journal of Monetary Economics. 24, 53-68.

 

Rose, A.K. (1990). “Exchange Rates and Trade Balance: Some Evidence from Developing Countries”, Economic Letters, 34, 271-275

 

Rosewiig, J. A. and Kock, P.D. (1988). “The U.S. Dollar and the Delayed J-Curve”, Economic Review: Federal Reserve Bank of Atlanta, 73(4), 2-15.

 

Salasevicius, R. and Vaicius, P. (2003). “Exchange Rate Relationship: Test the Marshall-Lerner Condition in the Baltic States” SSE Riga Working Papers, 2003: 13(48) Stockholm School of Economics in Riga.   

 

Senhadji (1998). “Dynamics of Trade Balance and Terms of Trade in LDCs: The J-Curve”,  Journal of International Economics, 46, 105-31.

 

Singh, T. (2002). “India’s Trade Balance: The Role of Income and Exchange Rate”, Journal of Policy Modelling, 24, 437 -452.       

 

Smal, M.M. and de Jager, S. (2001) “The Monetary Transmission Mechanism”, South African Reserve Bank, occasional paper 16, September, Pretoria.

 

Song, Y. (1997). “The Real Exchange Rate and the Current Account Balance in Japan”, Journal of Japanese and International Economies. 11, 143-184.

 

Soukhakian, B. (2007): “Financial Development, Trade Openness and Economic Growth in Japan: Evidence from Granger Causality Tests, International Journal of Economic Perspectives, 1(3), 117 – 126

 

Stucka, J. (2004). “Effect of Exchange Rate Changes on Trade Balance in Croatia”, IMF Working Paper, WP/04/65

 

Upadhyaya, K. P. and Dhakal. D. (1997). “Devaluation and Trade Balance: Estimating the Long-run Effect”, Applied Economic Letters vol. 4, pp.343 – 345.

 

Williamson, J. (1982). The Exchange Rate System: Washington D. C. Institute for International Economics.

 

Wolgom, G. (2003). “How has Inflation Targeting Affected Monetary Policy in South Africa?” The Journal of South African Economics, 71(2), 380 – 407.


 

APPENDIX

Table 2: Unit Root Tests .                       

Variable

ADF

PP

Decision

 

Intercept

Trend/Intercept

Intercept

Trend/Intercept

 

LTB

ΔLTB

-3.8091(1)*

-6.1231(1)*

-4.4811(1)*

-6.1121(1)*

-3.7481(1)*

-20.9421(1)*

-4.2941(1)*

-21.3641(1)*

1(1)

1(1)

LRGDP

ΔLRGDP

-0.7691(0)

-7.0971(1)*

-2.7801(0)

-7.0291(1)*

-0.3311(0)

-7.1431(1)*

-2.2301(0)

-7.0661(1)*

1(0)

1(1)

LRER

ΔLRER

-2.2731(0)

-5.6171(1)*

-3.2241(1)***

-5.5471(1)*

-2.2731(0)

-5.5601(1)*

-2.8261(0)

-5.4841(1)*

1(0)

1(1)

LRMS

ΔLRMS

-1.2691(0)

-5.0301(1)*

-2.4331(0)

-4.9711(1)*

-1.0261(0)

-5.5881(1)*

-2.1211(0)

-4.9411(1)*

1(0)

1(1)

Notes:   (i)   *, **, ***, represent 1%, 5% and 10% significance levels respectively.

            (ii)  The test is performed using Eview enconometric package version 6.0.

 

Table 3. Johansen – Juselius Maximum Likelihood Cointegration Test.

Hypotheses

 

 

 

Null

Alternative

Statistic

95% Critical Value

Trace Test

r = 0*

r > 1

79.17

69.81

r < 1

r > 2

39.87

47.85

r < 2

r > 3

22.44

29.79

r < 3

r > 4

5. 98

15.49

r < 4

r > 5

0. 02

3.84

Maximal Eignvalue Test

r = 0*

r = 1

39.30

33.87

r < 1

r = 2

17.43

27.58

r < 2

r = 3

16.45

21.13

r < 3

r = 4

5.96

14.26

r < 4

r = 5

0.02

3.84

Notes:   i.    r = no of cointegrating vectors

ii.  The test is conducted using Eview Econometric Package version 6.0.

iii. * indicates rejection of the null hypothesis (r = 0).

 

Table 4: Estimated Error-Correction Model

Dependent Variable: DLTB

Regressors

Parameter Estimates

Std-error

T-ratio

Intercept

DLRGDP

DLRER

DLRMS

EC(-1)

0.115

-0.07

0.01

-0.0002

-0.564

0.059

0.016

0.003

6.4E – 05

0.162

1.940

4.163

2.899

2.467

-3.481

Adj R2             =   0.52

F – Statistics    =   10.31