• No results found

include Mankiw, Romar and Weil, Arrow, Villanueve Rebelos A. K. model. The increasing returns theorist opposed the one classical growth theory that are subject to decreasing returns and said that the investment in some new area product, power source or production technology proceeds through time that each new increment of investment is more productive than the previous increment. The source of these increasing returns can be seen through cost and idea. Investment in the early stages of development may create new skills and attitudes in the work force whose cost may be lower than the previous investment at the initial stage. Also each investor may find investment because of the infrastructure that has been created by those who came before.

Finally, the new growth do not predict convergence and hence, countries with abundant physical and human capital will grow permanently faster that countries with small capital in contrast to the Solow model, the new growth model predicts divergence as implied in (Romer, 1996) and (A.O. Jenur, 2008).

the trade channel. The nature of the effect however, runs in either position or negative direction. According to IMF (1994), and European commission (1990), empirical evidence in favor of a systematic positive (or negative) effect of exchange rate stability on trade (and thereby growth) in small open economics has remained mixed. Gravity models have been used as framework to quantify the impact of exchange rate stability on trade and growth. Schnabel (2003) found evidence that exchange rates ability is associated with more growth in the EMU periphery. The evidence, according to him, is strong for emerging Europe which has moved from an environment of high economic instability to macro-economic stability during the observation period. Other empirical studies examine the role of capital market in ensuring exchange stability and economic growth.

The study undertook an investigation, aimed at finding any relationship before regional trade agreement (RTA) and growth. He focused on whether openness size of the population and the gross domestic product (GDP) affected growth of countries that have entered into RTA. The results shows that economy’s with open economics grow faster. He also provided evidence that he level of development in neighboring open economics have some spillover effect. By contrast, the lead level of development in open economics has no little on

domestic growth. Similar studies were done by Langhamer and Hienmenz (1990).

Their empirical work found out that regional agreement made up of developing nations has had no significant contribution to trade expansion.

Arron and Sala-Martins (1995), estimated the impact of trade protection on growth. Using tariff on capital goods and intermediate goods as a measure to protect their result indicated negative impact between trade, liberalization and growth countries with low results according to them grow faster than those with high tariffs. This confirms the earlier theoretical literature in favour of trade liberalization the forgoing literatures examined has known all positive relationship between trade and growth. In the words of Onah (2002), trade liberalization policy in Nigeria, was accompanied in 1987 budget and the result has been encouraging.

In his own view, the rate of inflation has been reasonably controlled though not reduced thoroughly. In spite of their effort to reduce prices, the local industries are collapsing because of inadequate demand for their products.

However, Boardiary and Trenderick (1987), using static applied general equilibrium (first generation) found that removal of tariffs in Canada would cause welfare to decline by about trade deterioration, resulting from an import tariff reduction as implied by national product differentiation. Assumption has led

Broom (1987) to conclude rather criterically that unilateral trade liberalization is E (>o) and (<o) minus (-) the income elasticity of demand for export and imports respectively.

Moreley (1992), analyzed the effect of real exchange rates on output for twenty-eight (28) devaluation experiences in developing countries using a regression framework. It was explicitly concluded that the exchange rate devaluation is a major factor for the upsurge of inflation. Kamin (1996) showed that the level of the real exchange rate was a primary determinant of the rate of inflation in Mexico during the 1980’s and 1990’s. (anetic and Green (1991), Falokun (1994), reached similar conclusions for some selected African countries including Nigeria.

Dell Africa (1999), examined the effect of exchange rate fluctuation on the bilateral trade of European Union members plus Switzerland over the period of 1975-1994, using several definitions of volatility. In the basic OLS regression, exchange rate fluctuation had a small but significant negative impact on trade;

reducing volatility to zero in 1994 would have increased trade by an amount ranging from 10 to 13%, depending on the measures of fluctuation used. Usually both fixed and random effects, the impact of fluctuation was still negative and

significant but smaller in magnitude. The author found that elimination of exchange rate fluctuation would have increased trade by about 36 in 1994.

Mauna and Reza (2001), studies the effect of trade liberalization, real exchange rate and trade diversification on selected north American countries like Morocco, Algeria and Tunisia. By decomposing changes in real exchange rate into fundamental and monetary determinations, and by using both standard statistical measures of exchange rate fluctuation and the measures of exchange rate risk developed by puree and Steiner (1989). They reached the conclusion that exchange rate depreciation has a positive effect on the quantity of manufactured exports while exchange rate misalignment, volatility has a negative effect. According to them, the motivating result is that all manufacturing sub-sectors are responsive to exchange rate changes but the degree of responsiveness differs across sectors.

In their study, Broda and Romatis (2003) they found that real exchange rate volatility depresses trade in differentiated goods. The study used bilateral trade made where the OLS (ordinary least square) and GMM (Generalized method of moment) methods were sued after taken into account the direction of causality, they ascertained that a 10% increase in volatility depresses differentiated product trade by 0.7%, while a 10% increase in trade reduces exchange rate volatility by

0.3%. The OLS estimated results showed that the effect or volatility on trade is reduced by 70%. They justified the result by arguing that much of the correlation between trade and change to the effect that trade has a depressing fluctuation.

Their study further revealed that a 10% increase in the intensity of bilateral trading relationship reduces the volatility of the associated exchange rate by 0.3%. Moving to the studies of exchange rate volatility on trade in less developed countries (LCD’s) Carter (1981), who used a log-level model specification to examine Brazilian exports, used annual data for 1965-1979 to arrive at the conclusion that a significant reduction in exchange rate uncertainty in Bazillion’s economy during the crawling era was adopted in 1963.

Philips (1986), Granger and Newboold (1974) found that export and exchange rate risks are related, however, they criticized the use of a log-level model when the data is non stationary.

Osuntogen et al (1993), in their analysis of strategic issues in promoting Nigeria’s non-oil exports, determined the effects of exchange rate uncertainty on Nigeria’s non export performance as a side analysis. This is the pioneering effort in Nigeria to determine the effect of exchange rate risk on export. However, their model did not take into consideration the cross price effect. Exchange rate acts as

shock absorber if rigidly fixed, the shock of inflation and deflation from abroad are transmitted to internal economy system. But variations in the exchange rate can wind off the invasion of the inflationary and deflation any forces. If demand and supply could work excellently in economic sense, it would be better to allow exchange rate to be freely determined by both demand and supply.

In conclusion, most of the econometric analysis indicated that devolutions (either increases in the level of real exchange rate or in the rate of depreciation) were associated with a reduction in output and increase in inflation.

Nigeria is regarded as the largest oil producing nation in Africa and the tenth largest in the world in terms of oil reserves with a production level of close to 2 million barrels per day, though this level has been seriously affected due to crises in the oil production region. Nigeria benefited handsomely from likes in the oil, since the beginning of second world war. The balance of payment portion of the country remains highly favorable with over 20 month of imports, which translates tower 55 billion of reserves. Exchange rate was moderately stable between 2000 and 2008, while GDP growth averaged 5.01% within the same period.

However, oil consumption in the country heavily relies on the import of refined petroleum and products since the collapse of local refineries in the late

1980’s. Thus over 90% of the country domestic requirements of oil are sourced from imports. The near collapse of the power generation and distribution industry in the country further accentuates the acute shortage of energy. The burden on the government to unwisely and between 1999 and 2000, the federal government of Nigeria has reduced its subsiding approximately 9 times. This seriously affects production, consumption and instruments in the country between 1986 and 2007, while figures 23 and 4, all in the appendix, represent the trends in the various in natural log.

2.3 LIMITATIONS OF THE PREVIOUS STUDIES

The impact of unstable exchange rate and devaluation on the economy have been a matter of concern to many scholars, researchers and business entrepreneurs.

Another major problem is the issue of appropriate definition of the concept of equilibrium. This portion of this project reviews the studies of different people on aspects of exchange rare devaluation and lack of appropriate definition of the concept of equilibrium in the measurement and analysis of the real exchange rate.

Egon (1963), examined the effects of exchange rate on price level balance of

payment and economic interaction. He rightly pointed out how these economic variables are affected by variations in exchange rate of the currency.

Aluko (1988) in his own view on the appreciation and depreciation of he naira since 1970 with regards to its effect on balance of payments and external reserve of the Nigeria, concluded depreciation of the naira which he said was overvalued was necessary for the implementation of SAP. He did not, however consider the developing nature of the Nigerian economy. And as a developing country or economy, Nigeria mainly producers primary product and imports machinery and some (major) raw materials for its industries. He did not consider the attendant high cost of imports with depreciation, devaluation would impose on such imports which would in turn, lead to high inflation rate. Kanyo (1988) in his work on inflation blames competitive price linking on free floating exchange market. This, he said is necessary due to the developing nature of the Nigerian economy.

Eze (1988), in his appraised of foreign exchange rate fluctuation on the Nigerian economy recommended that the central bank of Nigeria should stabilize the value of naira exchange at efficiently approved rate to the public. The action of the black market in which foreign exchange is sourced faster than at the banks. He

however suggested what the government should influence in the foreign exchange rate positive economy reforms that will reduce the adverse effects on unstable foreign exchange rate on the Nigeria economy.

The big push strategy the proponent of the big push strategy are of the view that the economics of developing countries like Nigeria cannot only be described as being stagnant but also lack the enthusiasm and courage to take the great leaper to the exponents of this theory, the less developed countries needs to get out of its underdevelopment and the only way is to out of its is to use a huge amount of resources in order to start the process of development. The less developed economics need to use more than half of the national income of the economy for all out investment. According to the proponent of this strategy, the idea of bit progress or step by step is not possible to help development countries to achieve their goal of self sustaining growth. The advocate of the strategy stress that as a car needs a big push therefore, will come from is it public sector on private sector.

The contribution of these authors is still in order to study the economic implication of exchange rate instability and how a less developed countries can achieve economic growth.

CHAPTER THREE 3.0 RESEARCH METHODOLOGY

The methodology is the background against which the reader evaluates the findings and conclusions (Osuala: 1992). The choice of the appropriate technique to be used in a research work depends on the research problem as well as the objectives of the study (Koutsoyiannis: 1997). Econometric method of regression analysis was employed in this study.

3.1 MODEL SPECIFICATION

We shall employ the single equation technique of econometric simulation for this study. The model specification involves the determinant of the dependent and independent variables were included in the model the priori expectation of the signs and sizes of the parameters of the functions, the functional form of the model, the mathematical form of the equation.

The model that will be adopted is the classical least regression model that will be used (OLS). The choice of this method is predicted on the basic features of OLS (BLUE).

MODEL 1

The model will be used to capture these objectives.

Objective 1: The econometric model is stated as;

GDP = b0 + b1 ER + b2 INT + b3 DOP + ei Where: ER = exchange rate

INT = Interest rate

DOP = Degree of trade openness = GDP = Gross domestic product Ei = The stochastic error term.

EXP+ IMP GDP

ER = is the exchange rate, INT = interest rate, DOP = degree of trade openness which are the independent variables causing variations on the dependent

variables.

GDP = Gross domestic product is the dependent variable,

BO is the intercept parameter and B1, B2, B3, are coefficient of the variables. Ei = stochastic error term.

3.2 METHOD OF DATA ANALYSIS

The result of this work shall be evaluated in three ways namely economic, statistical and econometry criteria.

3.3.1 ECONOMIC CRITERIA

The economic criteria test shall be conducted to enable us examine the magnitude and size of the parameter estimate. This evaluation is guided by economic theory to ascertain if the parameter estimate conforms to expectation.

The variable for real interest rate represents the user cost of capital. There exists a negative relationship between interest rate and investment on economic growth.

The variables for political risk are expected to exhibit a positive impact on free flow of export. This is informed by the fact that trade will move freely into areas of the economy with stable political system. The variable for trade openness which represents the measure of trade in the economy, is measured as trade to output ratio. Countries with high trade potentials will attract inflow of capital into the country. So there exist a positive relationship between trade openness and economic growth. exchange rate is expected to be positive because depreciation of the currency which is increase in exchange rate boost export and this brings about economic growth.

Variables Expected Signs

Exchange rate (ER) Positive (+)

Interest rate (INT) Negative (-)

Degree of trade openness (DOP) Positive (+)

3.3.2 STATISTICAL TEST (first – order)

Under the statistical test (first –order), we will test for the goodness of fit, the individual significance of each regression using the f- test and finally, significance of the regression model using the f-test.

(a) Goodness of fit test: We shall make of the coefficient of multiple determination R2 to find how well the sample regression line fits the data. R2 measure how the variations in the explanatory variable effect the dependent variable.

(b) Student t-test: It is used for testing the significance. We shall make use of 5% level of significance with n-k degree of freedom and where necessary, the probability value will be used as a rule of thumb. Where a = 0.05 (n-k), n = number of observation (sample size), k = total number of estimated parameter.

(c) The f-test: This will be used for testing the overall significance of the regression model. In order words, it will be used for testing joint impact of the independent variables on the dependent variable. The regression might not have influence on the dependent variable except in conjunction with other regression. We shall make use of 5% level of significance with (k-1) (n-k) degree of freedom where vi = k-1, v2 = n-k

3.3.3 ECONOMETRIC (Second-order test)

Econometric test will be used for empirical verification of the model. This will range from test including autocorrelation normality, heteroscedasticity and multicollinearity test.

(1) Autocorrelation: The classical linear regression model assumes that autocorrelation does not exist among the disturbance terms. In order to find out where the error terms are correlated in the regression, we will use the Brush – Godfrey serial correlation test. Brush-Godfrey test is the test for detecting autocorrelation. It allows for autoregressive (AR) and moving average (MA) error structure. It was jointly developed by Breusch Godfrey (Gujarati, 2004).

(2) Normality test: This test will be conducted to find out if the error term was normally distributed with zero mean and constant variance ie it ei N (0,52).

This is one of the assumptions of the classical linear regression model. The Jargue Bera test will be used to test for normality in the time series variables used. This test will be conducted by augmenting the equation by adding legged values of the dependent variables.

(3) Heteroscedasticity test: Heteroscedasticity occurs when the variance of the error term additional of the chosen values of the explanatory variables is not constant. In order to capture heterscedasticity and specification bias, the cross-product term will be introduced among auxiliary regressions.

(4) Multicollineaerity test: This test is used to detect linear relationship among the variables. This is a situation where the explanatory variables are highly interconnected when they are highly correlated, it becomes difficult to separate the effect of each of them on the dependent variable.

3.4 NATURE AND SOURCE OF DATA

The data used for this study are annual. Time series from 1980-2015, they are sourced from the Central Bank of Nigeria (CBN) Statistical Bulletin (2015)

CHAPTER FOUR

4.0 PRESENTATION AND ANALYSIS OF RESULTS 4.1 PRESENTATION OF RESULTS

Two models were estimated in this research work based on the topic the researcher is discussing. The models were estimated using the ordinary least square (OLD) method. The result of the models are presented below as thus:

Model I

Table 4.1.1 Result presentation Dependent variable: GDP

METHOD: Least square Sample: 1980-2015 Included observation 31

Variable Coefficient Std. Error T. Statistics Prob

Constant 12.082283 0.320608 37.68720 0.0000

ER 1.193588 0.051041 23.38486 0.0000

INT -0.077045 0.015944 -4.832137 0.0000

DOP 0.159790 0.174463 0.915901 0.3678

R-squared = 0.959414

F- statistics = 23.51945 (3,27) Durbin – Watson stat. = 0.313026 (0.0000) Number of observations = 31 Number of variables = 4

4.2 RESULT INTERPRETATION

4.2.1 ANALYSIS OF RESULTS BASED ON ECONOMIC CRITERIA Model I

The above result in terms of coefficients of the regression can be interpreted as follows:

The intercept is 12.08283. This shows that if all the explanatory variables are held constant, GDP will be 12.08283 ceterus paribus.

Exchange Rate (ER)

The coefficient is 1.193583. This indicates a positive relationship between real exchange rate and GDP, showing that a unit increase in exchange rate (ER) will increase GDP by 1.19588.

Interest Rate (INT)

Interest rate has a negative coefficient of -0.077045. This indicates that interest rate has a negative relationship with GDP, showing that a unit increase in interest rate (INT) will reduce GDP by 0.077045.

Degree of Trade Openness (DOP)

The coefficient is 0.159790. This shows that the degree of trade openness has a positive relationship with GDP, showing that a unit increase in the degree of trade openness (DOP) will increase GDP by 0.159790.

4.2.1.2 ANALYSIS BASED ON THE A PRIORI CRITERIA

This test is carried out to ascertain if the parameter estimates conform with what economic theory postulates in terms of sign and magnitude. The test is

summarized below:

Table 4.2.1.2 Model I

Variable Expected sign Obtained sign Conclusion

ER Positive (+) Positive (+) Conforms

INT Negative (-) Negative (-) Conforms

DOP Positive (+) Positive (-) Conforms

4.2.2 ANALYSIS BASED ON STATISTICAL CRITERIA

4.2.2.1 THE COEFFICIENT OF MULTIPLE DETERMINATION (R2)

In our model, mode I has R2 of 0.959414, which implied that about 95%

of the variation in real GDP is explained by the independent variable (real exchange rate, interest rate and degree of trade openness).

4.2.2.2 The T-test statistics

The T-test is used to determine the significance of the individual parameter estimates and to achieve this, we have to compare the calculated t-value in the regression result with the t-tabulated at n-k degree of freedom, at 5% significance level.

Test Hypothesis

H0: B1 = 0 (The parameters are statistically insignificant) H1: B1 ≠ 0 (The parameters are statistically significant).

Decision Rule

Reject Ho if t-cal > t-tab Accept Ho if otherwise

From our data n = 31 and k = 4 Therefore d.f = n-k =31-4 = 27

Critical t-tabulated at 0.05 significance level is equal to 2.052 MODEL I

Variable T-calculated T-tabulated Decision rule Conclusion

ER 23.38486 2.052 Reject Ho Significant

INT -4.832139 2.052 Reject Ho Significant

DOP 0.915901 2.052 Accept Ho Significant

4.2.2.3 The F-statistics Test

The test is carried out to determine if the independent variables in the model are simultaneously significant or not. It has k-l degree of freedom in the numerator and n-k degree of freedom in the denominator. Hence, the analysis shall be carried out under the hypothesis below:

Ho: X1 = X2 = X3 = 0 (The model is

insignificant) H1: X1 ≠ X2 ≠ X3 ≠ 0 (The model is significant)

Using the correlation Matrix

GDP ER INF EXPT DOP INT

GDP 1.0000 0.837062 -0308417 0.991147 0.185848 -0.003738 ER 0.837062 1.00000 -0342273 0.837436 0.232210 0.167966 INF -0.308417 -0342273 1.00000 -0.303448 0.170291 0.430320 EXPT 0.991147 0.837436 -0303448 1.0000 0.216589 -0.012541 DOP 0.185848 0.2322210 0.170291 0.216589 1.00000 0.307901 INT -0.003738 0.167966 0.430320 -0.012541 0.307901 1.0000

Decision Rule

From the rule of thomb, if correlation coefficient is greater than 0.8, we

conclude that there is multicollinearity but if the correlation coefficient is less than 0.8, there is no multicollinearity

Conclusion: Multicollinearity only exist between

ER and GDP EXPT and GDP EXPT and ER Decision Rule

Reject Ho if f-cal > f-tab otherwise accept Ho. V1 = K-1 = 4-1 = 3 (numerator)

V2 = n-k = 31-4 = 27 (denominator) MODEL I below analysis the result

F-calculated T-tabulated Decision rule

212.7502 2.9604 Reject Ho

From the result, since f-cal > f-tab (i.e. 212.7502 > 2.9604), we therefore reject the

null hypothesis Ho and accept the alternative hypothesis H1 and conclude that at

5% level of significance the overall regression is statistically significant.

(2nd order Test)

4.2.1.1 TEST FOR AUTO CORRELATION

This test is aimed at ascertaining if autocorrelation occurred in the model. To achieve this, we assume that the values of the random variables are temporarily independent by employing the technique of Durbin-Watson (d) statistics

Decision Rule

Null Hypothesis (Ho) Decision If

No positive autocorrelation Reject 0 < d < du

No positive autocorrelation No decision DL ≤ d ≤ du

No negative autocorrelation Reject 4 – dL < d ≤ 4

No negative autocorrelation No decision 4 – du ≤ d ≤ 4-dL No autocorrelation (positive or negative) Do not reject Du < d < 4 ≤ dL

Where dL = lower unit du = upper unit

d = Durbin Watson calculated

From the Durbin Watson table.

Model I model II

dL = 1.160 dL = 1.160 du = 1.735 du = 1.735 d*= 1.108538 d*= 1.49057 Decision rule

Model I: 0< d < dL

0 < 1.108538 < 1.160 Conclusion

The Durbin Watson shows that there is no positive autocorrelation in the two models. Therefore, we reject the null hypothesis for both model.

4.2.3.2 NORMALITY TEST

This test is carried out to test if the error term follows normal distribution. It is done using the Jarque-Bera statistic which follows chi-square distribution with 2 degrees of freedom at 5% level of significance.

Test Hypothesis

Ho: ei = 0 (The error term is normally distributed) H1: ei ≠ 0 (The error term is not normally distributed).

a = 5% (0.05 significant level) Decision Rule

Reject Ho if X2 cal > X2 tab otherwise accept Ho

From the result, obtained from Jarque-Bera test of normality, (JB) = 0.289133.

That is X2 cal = 0.239133 X2 tab = 5.99147 Conclusion:

We accept and conclude that the error term is normally distributed since X2 cal < X2 tab i.e. (0.289133 < 5.99147).

4.2.3.3 HETEROSCEDASTICITY TEST

This test is carried out to test if the error term has a constant variance. The test follows chi-square distribution with degrees of freedom equal to the number of regression in the auxiliary heteroscedasticity regression, excluding the error term.

Test Hypothesis

Ho: Homoscedasticity (The variance is constant) H1: Heteroscedasticity (the variance is not constant) Decision rule

Reject Ho if X2 cal > X2 tab otherwise accept Ho.

From the heteroscedasticity test result X2 cal = 450.7.76 and X2 tab = 16.919

From the result, X2 tab > X2 – tab (i.e. 16.919 > 4.500776) therefore reject the null hypothesis of homoscedasticity and accept the alternative hypothesis of heteroscedasticity showing that error term have a constant variance.

Related documents