THE LINK BETWEEN FINANCIAL LIBERALIZATION AND ECONOMIC GROWTH IN NIGERIA

Abstract

The study examine the link between financial liberalization and economic growth in Nigeria. The study adopt Ordinary Least Square method, co-integration test and Augmented Dickey fuller (ADF) procedure. The descriptive statistics for the dependent and independent variables, RGDP, DEEP, OPEN, LENDING, EXCHR and INF all have a positive mean value which ranges from14.71844 to 2135488 with a 44 observations. The highest standard deviation of 5562937 is recorded by RGDP rate while the least standard deviation of 6.345961 is recorded by LENDING. The probabilities of Jarque-Bera test of normality for variables are all greater than 5% level of significance which indicates that the data are normally distributed. The regression estimates of financial liberalisation equation shows that three of the coefficients of the explanatory variables DEEP, OPEN and EXCHR have positive signs while the coefficient of LENDING and INF indicates a negative signs, this implies increase in the bank lending rate reduces loanable fund thereby decreasing growth rate. On the other hand, the effect of bank credit to public and private sector in Nigeria, the regression result obtained shows that the coefficients of the variables PUBLIC indicates a negative sign while the coefficient of the variable, PRIVATE exhibit positive signs. The granger causality test is use to determine the predictive content of the variable beyond that inherent in the explanatory variable itself, the result suggest that there is no direction of causality between RGDP and DEEP, which is the financial liberalization. T-test procedures shows that the t-value of variables (OPEN, LENDIING, and EXCHR) are significant while others are not. The result of the F-test shows that since F-calculated of 10.04928 are greater than F-tabulated of 4.04 for the first regression and F-calculated of 75.52089 are greater than F-tabulated for the second regression, we reject H0 and concluded that the overall estimate of the regressions are adequate statistically. The unit root test was conducted to establish that the time series data on all variables are stationary and integrated of order one at 5% level of significance in ADF. The co-integration test procedures conducted indicates at most six co-integrating vectors. Multicolinearity test was conducted to ascertain the degree of relationship between the dependent (GDP) variable and the independent (DEEP, OPEN, EXCHR, INF, LENDING, PRIVATE AND PUBLIC) variable, and the result shows that the three of the variable (OPEN, EXCHR and LENDING) have positive relationship with GDP and the relationship are actually at 1%, 4%, and 60% respective while DEEP and INF are negatively related with value 16% and 14% respectively. Normality test was also conducted to ascertained the normality distribution of the error term of the variable under consideration, the result of the test shows that chi-square tabulated is 48.12885 while chi-square calculated is 5.99147, since the chi-square calculated is less than chi-square tabulated, the variable under consideration are not normally distributed. The heteroscedasticity test was also conducted to ascertain which of the hypothesis should be accepted or rejected, the test explain that if chi-square calculated is less than chi-square tabulated, we accept H0 otherwise we reject. On the other hand, the result of the test shows that chi-square tabulated with degree of freedom (10) is 11.91592 while the calculated chi-square is 18.3070. Since the chi-square calculated is less than chi-square tabulated, the error term of the variable under consideration are homoscedastic.