Descriptive Statistics
Table 2 below gives a summary of descriptive statistics of variables for the model. The reported statistics include the mean with their corresponding maximum, minimum and standard deviation. The distributional properties are also examined through their skewness and kurtosis, while the Jarque-Bera test statistic is used to test for normality in the distribution.
Table 2
a Descriptive Analysis using Per Capita Gross Domestic Product (GDPPC)
| GDPPC | LNGFCF | LNLAB |
Mean | 6.336911 | 21.82994 | 16.66539 |
Median | 6.097285 | 21.90543 | 16.72709 |
Maximum | 8.071936 | 22.82515 | 17.89236 |
Minimum | 5.030932 | 21.64558 | 15.78490 |
Std. Dev. | 0.886476 | 0.191725 | 0.566318 |
Skewness | 0.628980 | 2.810509 | 0.135678 |
Kurtosis | 2.214475 | 15.79861 | 1.782931 |
Jarque-Bera | 4.490673 | 398.9424 | 3.174569 |
Probability | 0.105892 | 0.000000 | 0.204480 |
Sum | 310.5086 | 1069.667 | 816.6041 |
Sum Sq. Dev. | 37.72033 | 1.764415 | 15.39436 |
Observations | 49 | 49 | 49 |
Source: Author’s Computation from E-view 9 |
The table above shows the descriptive statistics of the variables. The mean value for log of gross domestic product, log of gross fixed capital formation and log of labour is 6.336911, 21.82994, and 16.66539, respectively. The maximum values for log of gross domestic product, log of gross fixed capital formation and log of labour is 8.071936, 22.82515 and 17.89236, respectively with respective minimum values of 5.030932, 21.64558 and 15.78490. The table also shows the skewness values for the variables. The skewness value for log of gross domestic product, log of gross fixed capital formation and log of labour is 0.628980, 2.810509, and 0.135678 respectively. This shows a positive skewness value for log of gross domestic product, log of gross fixed capital formation and log of labour.
Also, the kurtosis values for these variables are also determined. The kurtosis value for log of gross domestic product, log of gross fixed capital formation and log of labour, is 2.214475, 15.79861 and 1.782931respectively. A distribution with a coefficient larger than 3 is said to be leptokurtic and one with a coefficient smaller than 3 is platykurtic. This then shows that all the variables are platykurtic expect the log of gross fixed capital formation which has a distribution with a coefficient less greater than 3 which is leptokurtic
Table 2
b Descriptive Analysis using LOG of Gross Domestic Product (LNGDP)
| LNGDP | LNGCF | LNLAB |
Mean | 24.80304 | 21.82994 | 16.66539 |
Median | 24.31533 | 21.90543 | 16.72709 |
Maximum | 27.06627 | 22.82515 | 17.89236 |
Minimum | 22.94049 | 21.64558 | 15.78490 |
Std. Dev. | 1.164851 | 0.191725 | 0.566318 |
Skewness | 0.670170 | 2.810509 | 0.135678 |
Kurtosis | 2.237964 | 15.79861 | 1.782931 |
Jarque-Bera | 4.853466 | 398.9424 | 3.174569 |
Probability | 0.088325 | 0.000000 | 0.204480 |
Sum | 1215.349 | 1069.667 | 816.6041 |
Sum Sq. Dev. | 65.13011 | 1.764415 | 15.39436 |
Observations | 49 | 49 | 49 |
Source: Author’s Computation from E-view 9 |
The table above shows the descriptive statistics of the variables. The mean value for log of gross domestic product, log of gross fixed capital formation and log of labour is 24.80304, 21.82994, and 16.66539, respectively. The maximum values for log of gross domestic product, log of gross fixed capital formation and log of labour are 27.06627, 22.90543 and 17.89236, respectively. Their respective minimum values are 22.94049, 21.64558 and 15.78490 respectively. The table also shows the skewness values for the variables. The skewness value for log of gross domestic product, log of gross fixed capital formation and log of labour are 0.670170, 2.810509, and 0.135678 respectively. This shows a positive skewness value for log of gross domestic product, log of gross fixed capital formation and log of labour.
Also, the kurtosis values for these variables are also determined. The kurtosis value for log of gross domestic product, log of gross fixed capital formation and log of labour, is 2.237964, 15.79861 and 1.782931 respectively. A distribution with a coefficient larger than 3 is said to be leptokurtic and one with a coefficient smaller than 3 is platykurtic. This then shows that all the variables are platykurtic expect the log of gross fixed capital formation which has a distribution with a coefficient less greater than 3 which is leptokurtic
Table 2
c Descriptive Analysis using the Growth of Gross Domestic Product (GDP_Growth)
| GDP_GROWTH | LOG(GFCF) | LOG(LAB) |
Mean | 5.384677 | 21.82994 | 16.66539 |
Median | 5.125552 | 21.90543 | 16.72709 |
Maximum | 7.071936 | 22.82515 | 17.89236 |
Minimum | 4.030932 | 21.64558 | 15.78490 |
Std. Dev. | 0.898438 | 0.191725 | 0.566318 |
Skewness | 0.547940 | 2.810509 | 0.135678 |
Kurtosis | 2.069817 | 15.79861 | 1.782931 |
Jarque-Bera | 4.218480 | 398.9424 | 3.174569 |
Probability | 0.121330 | 0.000000 | 0.204480 |
Sum | 263.8492 | 1069.667 | 816.6041 |
Sum Sq. Dev. | 38.74515 | 1.764415 | 15.39436 |
Observations | 49 | 49 | 49 |
Source: Author’s Computation from E-view 9 |
The table above shows the descriptive statistics of the variables. The mean value for growth of gross domestic product, log of gross fixed capital formation and log of labour is 5.384677, 21.82994 and 16.66539 respectively. The maximum values for the growth of gross domestic product, log of gross fixed capital formation and log of labour is 7.071936, 22.82515 and 17.89236 respectively with respective minimum values of 4.030932, 21.64558 and 15.78490. The table also shows the skewness values for the variables. The skewness value for growth of gross domestic product, log of gross fixed capital formation and log of labour is 0.547940, -2.810509 and 0.135678 respectively. This shows a positive skewness value for all the variable that is the growth of gross domestic product, log of gross fixed capital formation and log of labour are positively skewed.
Also, the kurtosis values for these variables were also determined. The kurtosis value for growth of gross domestic product, log of gross fixed capital formation and log of labour, is 2.069817, 15.79861 and 1.782931 respectively. A distribution with a coefficient larger than 3 is said to be leptokurtic and one with a coefficient smaller than 3 is platykurtic. This then shows that all the variables are platykurtic expect the log of gross fixed capital formation which has a distribution with a coefficient less greater than 3 which is leptokurtic.
Correlation Analysis Result
The study makes use of correlation analysis in other to show the relationship between gross domestic product per capita and other macroeconomic variables.
Table 3
| GDPPC | LNGFCF | LNLAB |
GDPPC | 1.000000 | | |
LNGFCF | 0.214435 | 1.000000 | |
LNLAB | 0.701656 | 0.295716 | 1.000000 |
Source: Author’s Computation from E-view 9 |
The correlation output between log of gross domestic product and each of log of gross fixed capital formation and log of labour. The correlation analysis of the log of gross domestic product and the log of gross fixed capital formation is positively related. This implies that there is a linear positive relationship between the log of gross domestic product and log of gross fixed capital formation. Specifically, the correlation coefficient between the two is 0.214435. Since the coefficient of the relationship between the two is less than + 0.5, there exist a “weak positive correlation” between log of gross domestic product and log of gross fixed capital formation. This shows that as gross fixed capital formation increases, gross domestic product also increases.
The correlation analysis of the log of gross domestic product and log of labour is positively related. This implies that there is a linear positive relationship between the log of gross domestic product and log of labour. Specifically, the correlation coefficient between the two is 0.701656. Since the coefficient of the relationship between the two is greater than + 0.5, there exist a “strong positive correlation” between the log of gross domestic product and log of labour. This shows that as labour increases, gross domestic product also increases.
Table 3
b Correlation Analysis between log of gross domestic product and other macroeconomic variable
| LNGDP | LNGFCF | LNLAB |
LNGDP | 1.000000 | | |
LNGCF | 0.240698 | 1.000000 | |
LNLAB | 0.839177 | 0.295716 | 1.000000 |
Source: Author’s Computation from E-view 9 |
Table above shows the correlation between the variables. The correlation output between log of gross domestic product and each of log of gross fixed capital formation and log of labour. The correlation analysis of the log of gross domestic product and the log of gross fixed capital formation is positively related. This implies that there is a linear positive relationship between the log of gross domestic product and log of gross fixed capital formation. Specifically, the correlation coefficient between the two is 0.240698. Since the coefficient of the relationship between the two is less than + 0.5, there exist a “weak positive correlation” between the log of gross domestic product and log of gross fixed capital formation. This shows that as the gross fixed capital formation increases, gross domestic product also increases.
The correlation analysis of the log of gross domestic product and log of labour is positively related. This implies that there is linear positive relationship between the log of gross domestic product and log of labour. Specifically, the correlation coefficient between the two is 0.839177. Since the coefficient of the relationship between the two is greater than + 0.5, there exist a “strong positive correlation” between the log gross domestic product and log of labour. This shows that as labour increases, gross domestic product also increases.
Table 3
c Correlation Analysis between growth of gross domestic product and other macroeconomic variable
| GDP_GROWTH | LOG(GFCF) | LOG(LAB) |
GDP_GROWTH | 1.000000 | | |
LNGFCF | 0.192298 | 1.000000 | |
LNLAB | 0.712252 | 0.295716 | 1.000000 |
Source: Author’s Computation from E-view 9 |
Table above shows the correlation between the variables. The correlation output between growth of gross domestic product and each of log of gross fixed capital formation and log of labour. The correlation analysis of the growth of gross domestic product and the log of gross fixed capital formation is positively related. This implies that there is a linear positive relationship between the growth of gross domestic product and log of gross fixed capital formation. Specifically, the correlation coefficient between the two is 0.192298. Since the coefficient of the relationship between the two is less than + 0.5, there exist a “weak positive correlation” between the growth of gross domestic product and log of gross fixed capital formation. This shows that as the gross fixed capital formation increases, growth of gross domestic product also increases.
The correlation analysis of the growth of gross domestic product and log of labour is positively related. This implies that there is a linear positive relationship between the growth of gross domestic product and log of labour. Specifically, the correlation coefficient between the two is 0.712252. Since the coefficient of the relationship between the two is greater than + 0.5, there exist a “strong positive correlation” between the growth of gross domestic product and log of labour. This shows that as labour increases, growth gross domestic product also increases.
Determinant Of The Sources Of Growth In The Nigeria Economy From 1970 To 2018
The study adopted the Ordinary Least Square (OLS) to estimate models of sources of growth in the Nigeria economy from 1970 to 2018.
Table 4
a Sources of Growth in the Nigeria Economy using Gross Domestic Product per Capita as the dependent variable
Dependent Variable: GDPPC |
Method: Least Squares |
Included observations: 49 |
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | -12.67654 | 10.61157 | -1.194596 | 0.2384 |
LNGFCF | 3.180008 | 1.210017 | 2.612205 | 0.0001 |
LNLAB | 1.094802 | 0.172136 | 6.360107 | 0.0000 |
R-squared | 0.898214 | Mean dependent var | 6.336911 |
Adjusted R-squared | 0.825810 | S.D. dependent var | 0.886476 |
S.E. of regression | 0.645180 | Akaike info criterion | 2.020696 |
Sum squared resid | 19.14784 | Schwarz criterion | 2.136521 |
Log likelihood | -46.50704 | Hannan-Quinn criter. | 2.064640 |
F-statistic | 22.30890 | Durbin-Watson stat | 1.724647 |
Prob(F-statistic) | 0.000000 | |
Source: Authors computation from E-View9 |
Interpretation of the Estimated Models of Sources of Growth in the Nigeria Economy using Gross Domestic Product Per Capita as the Dependent Variable
Table 4a above shows the result of the estimated model of sources of growth in the Nigeria economy. From the table, it is revealed that the two variables were statistically significant where both log of gross fixed capital formation and log of labour are statistically significant at 1% level of significance respectively. The explanatory variables, which is gross fixed capital formation and labour are positively related to the growth of gross domestic product in Nigeria. Therefore, a 1% increase in any of the explanatory variable will increase the growth of the economy by 3.18% and 1.094% respectively. The economic implication of the above result is that increase in gross fixed capital formation and labour will lead to the growth of the economy.
The R-squared of 0.898214 showed that the explanatory variables which are the log of gross fixed capital formation, and log of labour explains about 89.8% of the variation in the growth of gross domestic product in Nigeria. The Adjusted R-squared of 0.825810 means that all the explanatory variables can only explained 82.5% total variation in in the growth of gross domestic product in Nigeria. Durbin-Watson statistic of 1.724647 shows that there is no autocorrelation in the model. The F-statistic is 22.30890 with probability value (p-value) of 0.000 which is less than the significance level of 1%. This indicates the models goodness of fit to the data and thus, the overall model is statistically significant.
Table 4
b Sources of Growth in the Nigeria Economy using real GDP (LNGDP) as the dependent variable
Dependent Variable: LNGDP | | |
Method: Least Squares | | |
Included observations: 49 | | |
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | -2.961564 | 10.64269 | -0.278272 | 0.7821 |
LNGCF | 0.409668 | 0.509945 | 0.097399 | 0.9228 |
LNLAB | 1.731064 | 0.172641 | 10.02697 | 0.0000 |
R-squared | 0.704279 | Mean dependent var | 24.80304 |
Adjusted R-squared | 0.691422 | S.D. dependent var | 1.164851 |
S.E. of regression | 0.647072 | Akaike info criterion | 2.026553 |
Sum squared resid | 19.26032 | Schwarz criterion | 2.142379 |
Log likelihood | -46.65054 | Hannan-Quinn criter. | 2.070497 |
F-statistic | 54.77610 | Durbin-Watson stat | 1.976862 |
Prob(F-statistic) | 0.000000 | |
Source: Authors computation from E-View9 |
Interpretation of Result of Estimated Model of sources of Growth in the Nigeria Economy using real GDP as dependent variable
Table 4.3b above shows the result of the estimated OLS on sources of growth in the Nigeria economy. From the table, it is revealed that one variable is statistically significant where labour is statistically significant at 1% level of significance while gross fixed capital formation is not statistically significant. The explanatory variables, that is gross fixed capital formation and labour are positively related to the growth of gross domestic product in Nigeria therefore, a 1% increase in any of the explanatory variable will increase the growth of the economy by 0.409% and 1.7311% respectively. The economy implication of the above result is that increase in gross fixed capital formation and labour will lead to the growth of the economy.
The R-squared of 0.704279 showed that the explanatory variables which are the log of gross fixed capital formation, and log of labour explains about 70.4% of the variation on sources of growth in gross domestic product in Nigeria. The Adjusted R-squared of 0.691422 means that all the explanatory variables can only explained 69.1% total variation on sources of growth in gross domestic product in Nigeria. Durbin-Watson statistic of 1.976862 shows that there is no autocorrelation in the model. The F-statistic is 54.77610 with probability value (p-value) of 0.000 which is less than the significant level of 1%. This indicates the models goodness of fit to the data and thus, the overall model is statistically significant.
Table 4
c Sources of Growth in the Nigeria Economy using real growth of gross domestic product as the dependent variable
Dependent Variable: GDP_GROWTH |
Method: Least Squares |
Included observations: 49 |
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | -11.54911 | 10.59148 | -1.090416 | 0.2812 |
LOG(GFCF) | 1.194108 | 0.507491 | 2.352964 | 0.0007 |
LOG(LAB) | 1.139377 | 0.171810 | 6.631616 | 0.0000 |
R-squared | 0.507670 | Mean dependent var | 5.384677 |
Adjusted R-squared | 0.486265 | S.D. dependent var | 0.898438 |
S.E. of regression | 0.643958 | Akaike info criterion | 2.016904 |
Sum squared resid | 19.07538 | Schwarz criterion | 2.132730 |
Log likelihood | -46.41416 | Hannan-Quinn criter. | 2.060849 |
F-statistic | 23.71667 | Durbin-Watson stat | 1.571605 |
Prob(F-statistic) | 0.000000 | |
Source: Authors computation from E-View9 |
Interpretation of Estimated Model of sources of Growth in the Nigeria Economy using growth of GDP as dependent variable
Table 4.3c above shows the result of the estimated model of the sources of growth in the Nigeria economy. From the table, it is revealed that the two variables are statistically significant where labour is statistically significant at 1%. The explanatory variables that is gross fixed capital formation and labour are positively related to the growth of gross domestic product in Nigeria. Therefore, a 1% increase in any of the explanatory variable will increase the growth of the economy by about 1.194% and 1.139% respectively. The economic implication of the above result is that an increase in gross fixed capital formation and labour will lead to the growth of economy.
The R-squared of 0.507670 shows that the explanatory variables which are the log of gross fixed capital formation, and log of labour explains about 50.7% of the variation in the growth of gross domestic product in Nigeria. The Adjusted R-squared of 0.486265 means that all the explanatory variables can only explained 48.6% variation in in the growth of gross domestic product in Nigeria. Durbin-Watson statistic of 1.571605 shows that there is no autocorrelation in the model. The F-statistic is 16.53799 with probability value (p-value) of 0.000. This indicates the model’s goodness of fit to the data and thus, the overall model is statistically significant.
Relative Importance On Sources Of Growth In The Nigeria Economy
The study will adopt the result of gross domestic product per capita in determining the relative importance of the sources of growth in Nigeria economy base on high models goodness of fit to the model. The result has high explanatory power of R2 = 89.8% and ADj R2 = 82.5% compare to other results which have less explanatory power of R2 and Adj R2. Therefore, table 4.4, 4.5, 4.6 and 4.7 below show the relative importance of the sources of Growth in the Nigerian economy. The study will test for standardized coefficients or beta coefficients. Standardized coefficients or beta coefficients are the estimates resulting from a regression analysis that have been standardized so that the variances of dependent and independent variables are 1. Therefore, standardized coefficients refer to how much standard deviations of dependent variable will change, per standard deviation increase in the predictor variable. For univariate regression, the absolute value of the standardized coefficient equals the correlation coefficient. Standardization of the coefficient is usually done to answer the question of which of the independent variables have a greater effect on the dependent variable in a multiple regression analysis, when the variables are measured in different units.
Formula for standardized beta test
Where β* is the standardized beta, β is the unstandardized beta, σX is the standard deviation of independent variable and σY is the standard deviation of dependent variable respectively
Table 5
Variables and their corresponding Standard Deviation Value
Variables | Standard Deviation Values |
GDPPC | 0.886476 |
LNGFCF | 0.191725 |
LNLAB | 0.566318 |
Source: Author’s Computation from E-View-9 |
Table 6
Variables and their Corresponding Unstandardized Coefficient
Variables | Unstandardized Coefficient |
LNGFCF | 3.180008 |
LNLAB | 1.094802 |
Source: Author’s Computation from E-View-9 |
Table 7
Variables | Standardized Beta Value |
LNGFCF | 0.687 |
LNLAB | 0.699 |
Source: Author’s Computation from E-View-9 |
Table 8
Standardized Beta Ranking in Descending Order
Variables | Standardized Beta Value | Ranking |
LNGFCF | 0.687 | 2 |
LNLAB | 0.699 | 1 |
Where 1 > 2; |
Source: Author’s Computation from E-View-9 |
Going by the results in Table 4.9 above on standardized beta ranking, it can be seen that one(1) standard deviation increase in gross fixed capital formation while holding every other explanatory variables constant will increase gross domestic product by 0.687. A standard deviation increase in changes in log of labour holding other variable constant will on the average increase the standard deviation on log of gross domestic product by 0.699.
From the above result it can be concluded that labour is the most important among the explanatory variables in explaining the source of growth of the economy.
Comparison Of Findings With Similar Studies
The study found that gross fixed capital formation has a positive impact on the growth of Nigeria economy which means that increase in gross fixed capital formation will lead to increase in the growth of Nigeria economy which conforms to findings of Ajide (2014). The implications of the results is that there might have been considerable improvements in some of the components of capital investment in the country like the size of government capital formation, legal structure and security of property rights, access to sound money, freedom to trade internationally and regulation of credit, labour and business. Such improvements in the components can be explained in part by the enthronement of democratic structures in the country since 1999 till date.
More so, the current study finds that labour has a positive significant impact on the growth of Nigeria economy which is in line with the result of Oluitan (2015), Babatunde (2015) and Olatunji (2016).
Policy Recommendations And Limitation Of The Study
Based on findings of this study the following policy recommendations are put forward:
-
Investment in human capital and engaging the huge unemployed labour force in the country should be of high importance to the government
-
Investors should be highly encouraged by reducing taxes, making raw materials needed available; working tools etc. This will encourage private investors into local investments.
-
There is need for government to consciously develop the business environment by provision of necessary infrastructure, which will lower the cost of doing business in Nigeria. The recent privatization of electric power holding company may be a step in the right direction if there is an improvement in the services provided
-
Finally, government should encourage stability in macroeconomic variables and employ such growth oriented and stabilization policies especially at macro level which will induce economic growth and development
Further research can still be conducted along this line by using another model aside from the model used in this study, the study only utilized three variable which are gross domestic product, gross fixed capital formation and labour which is not enough to determine the sources of growth in Nigeria economy because they are not the only factor contributing to the growth of Nigeria economy. Also, the present study only utilize regression analysis in investigating sources of growth in Nigeria other studies can make use of some other methods in the process of the research and also indicate some other variable or factors that dictate the sources of growth in Nigeria.