1.0 IntroductionHuman capital plays a crucial role in economic growth of every nation. Human capital is one of the factors claimed by economist to be source of economic growth. According to Abramowitz (2003) human capital is the most important and promising source of growth for any production industry and economic growth. In their study, Adeyemi and Ogunsola (2016) defined human development as the process of acquiring and increasing the number of persons who have the skill, education, experience which are critical for the economic and political development of a country.

Human capital development is strategic to the socio-economic development of a nation and includes education, health, labour and employment and women affairs. Investing in human capital development is therefore critical as it is targeted at ensuring that the nation’s human resource endowment is knowledgeable, skilled, productive and healthy to enable the optimal exploitation and utilization of other resources to engender growth and development. Nonetheless, no country has achieved sustained economic development without substantial investment in human capital.

According to Omotayo (2015) the concept of human capital has shifted the focus of economic development theorists to generally agree that the quality of human capital has a significant impact on economic development and growth. The quality of human capital in developing countries is very low compare to the developed countries. This can be attributed to poor investment in human capital. The United Nations recommends that 26 per cent of the total expenditure must be devoted to education but in Nigeria between 1986 to 1990 only 5.64% of total GDP were devoted to education and 5.84% between 1999 and 2003. Lack of investment in human capital will lead to inefficiency of labour force which results in low productivity. In the literature there are several theoretical models of economic growth that have emphasized the role of human capital in the form of educational attainment (Barro, 2001). These theoretical models includes Nelson and Phelps (1966); Lucas (1988); Becker, Murphy, and Tamura (1990); Rebelo (1992); and Mulligan and Sala-i-Martin (1992). There are also some empirical studies of growth for a broad crosssection of countries, such as those by Romer (1990a), Barro (1991), Kyriacou (1991), and Benhabib and Spiegel (1992) which used educational attainment as proxy for human capital. These studies have, however, been hampered by the limited educational data that were available on a consistent basis for a large number of countries. Due to this limited educational data we therefore decided to use total labour force as proxies for human capital in contrary to several studies that have used primary and secondary enrolment as proxy for human capital.2.0 Literature ReviewJaiyeoba (2015) examine the relationship between human capital investment and economic growth in Nigeria between of 1982-2011. This study employs trend analysis, the Johansen cointegration and ordinary least square techniques. The study found a long-run relationship between government expenditure on education, health and economic growth. Government expenditure on health and education, secondary school enrolment, tertiary school enrolment and gross fixed capital formation have positive and significant impact on economic growth. But the study found that government expenditure on education and primary enrolments have negative impact on economic growth. Kanayo (2013) examined the relationship between human capital development and economic growth in Nigeria using Error Correction Model as an analytical tool. The study found that investment in human capital in the form of education and capacity building at the primary and secondary levels impact significantly on economic growth, while capital expenditure on education was insignificant to the growth process. The author recommended that educational institutions in Nigeria should be re-structured for quality schooling at the primary, secondary and tertiary levels. In a competitive and globalized economy, this will require strategic planning, increase in capacity utilization by the education sector and rebasing of growth fundamentals.Adeyemi and Ogunsola (2016) examined the impact of human capital development on economic growth in Nigeria using time series data spanning from 1980 to 2013. The study employed ARDL Co-integration analysis to estimate the relationship among the variables used in the study. The study established long-run co-integration among the variables. The findings from the study revealed that there is positive long-run relationship among secondary school enrolment, public expenditure on education, life expectancy rate, gross capital formation and economic growth but it is statistically insignificant. The results also showed that there is negative long-run relationship among primary, tertiary school enrolment, public expenditure on health and economic growth. In line with the findings, the study recommended that government should put in place the required education and training policy that would guarantee quality schooling for primary and tertiary education. Government should also commit more funds to health sector to enhance human capital development. Mba, Mba, Ogbuabor and Ikpegbu (2013) examined human capital development and economic growth in Nigeria using ordinary least square (OLS) technique of estimation. The study found a positive relationship between human capital and economic growth in Nigeria. The recommendations drawn from the study centered on revisiting the man-power needs of the various sectors of the economy. Also, while workable policies should be put in place to bring about an overall economic growth, expenditures on health and public education should be utilized effectively and efficiently so that the country would experience quality health care services and quality educational system. 3.0 MethodologyIn line with the theoretical framework discussed above, the study adopted the endogenous growth model with a modified Cobb-Douglass production function assuming constant return to scale. This approach has been adopted by several authors including Eller (2005), Fink et al (2004, 2005) and Webb et al (2002). Aggregate output is specified as: Y = AK^± H^І L^(1-±-І) (1)Where Y represents the output (GDP), A denotes technology progress, K represents physical capital, H stands for human capital and finally L is the used labour force. After transforming equation (1) into the intensive form, it becomes: y = Ak^± h^І (2) By taking logarithm of both sides and differentiating equation 2 ”ln(y) = ln(A) + ±”ln(k) + І”ln(h) (3) Apart from physical capita and human capital, evidence from previous studies has shown that many other factors are significant determinants of real growth. Therefore, Trade openness, manufacturing and interest rate are included as they will also serve as control variables. €ln(y) = ln(A)+ ±_1”ln (k_f)+ +±_2 €ln (h)+ ±_3 €ln(INF)+ ±_4 €ln(INT)+e_t (4)4.0 Measurement of variables Secondary data were used in this study and the annual data covers the period of 1986 ” 2015. The data were obtained from World Development indicator (WDI). The following variables were included in the study; GDP is the real GDP (constant 2010 US$), physical capital is proxy by gross capital formation. Human capital is the total labour force. Trade openness is the sum of export and import as percentage of GDP. INT is real interest rates. Manufacturing is the manufacturing value added. 5.0 Empirical resultsIn order to analysis that free from spurious results there is the need to perform a unit root test. Therefore, both ADF and PP tests will be conducted in log level and first differences in order to determine univariate properties of the series being examined. That is, to test for the presence of the unit roots or nonstationarity. Both tests involve testing the null hypothesis of a unit root or nonstationarity of the series against the alternative of stationarity. The results from both ADF and PP statistics shows that all the variables are stationary at first difference therefore are integrated of order 1(1). Therefore, we can reject the null hypothesis of unit root. Table 1: Unit Tests resultsNote. Mackinon critical values for rejection of hypothesis of a unit root. ** Denote significant at 5% level. ***Denote significant at 1% level. The critical values for ADF are: -3.753, -2.998, and -2.639 (constant only @ level); -3.769, -3.00, and -2.642 (constant only @ 1st difference); -4.416, -3.622, and -3.249; (constant & trend @ level); -4.441, -3.633 and -3.255 (constant and trend @ 1st difference) at 1%, 5% and 10% level of significance respectively. However, the critical values for KPSS test are: 0.739, 0.463 and 0.347 (with constant only); 0.216, 0.146 and 0.119 (constant and trend) at 1%, 5% and 10% level of significance, respectively.Having established that the variables are all stationary at first different we therefore, performed cointegration test in order to determine if there is long run relationship among the variables. In order to achieve this we perform cointegration test by using both maximum “-max and trace tests and their results are presented in panel A of table 2 below. Table 2: Cointegration results (with a linear) where r in the number of co-integrating vectors. Note: t ratios are in parentheses***denote significant at 1% and * significant at 10%.In summary, the results in panel A of Table 2 shows that both maximum “-max and trace test statistics reveal that there are, at most, four cointegrating relationships among the real GDP and other variables. In essence, both the test statistics (trace and ?-max) reject the null hypothesis of ?0: r = 0, ?0: r ‰¤ 0 and ?0: r ‰¤ 0 at the 5% significance level. The conclusion from this result is that there is a long run relationship between economic growth and other variables. The co-integrating equation (normalized on growth variable) shown in panel B of Table 2 result shows that the coefficients of all the variables are highly statistically significant at 1% significance level. The results also that physical capital, trade openness, human capital and manufacturing value added have positive impact on economic growth. Given that a cointegrating relationship is present among the related variables, an error correction model is estimated. This is a model that combines both the short run properties of the economic relationship in the first difference form as well as the long run information provided by the data in level form. Therefore, we use the information provided by the likelihood ratio test to generate a set of the models that capture the short and long run behaviour of the output relationship. The changes in the relevant variables represent elasticities, while the coefficient of the ECM term represents the speed of adjustment back to the long run relationship among variables. The short run results are provided under the error correction model as shown in Table 3 below. Table 3: Error Correction Model (dependent variable ”In Y) Note. t statistics in parentheses. (*) (**) (***) denotes significance at 10%, 5% and 1% level respectively. All variables are as defined earlier. The ECM results are interesting as it shows a varying statistical significance level for some of the coefficients while some of them were also statistically insignificant. The adjusted R-squared which ranges between 0.29 and 0.44 in regression 1, 2, 3, 4 and 5 are obtained from the short-run model. This suggests that all the explanatory variables (human capital, physical capital, trade openness, manufacturing and interest rate) account for average per cent between 29 to 44 variations in the dependent variable (GDP). The F-statistics which test for the overall significance of the model is relatively high and provides a good fit for the estimated model. The Durbin Watson (DW) statistics are generally satisfactory indicating non-existence of autocorrelation problem in the model. The coefficients of ECM carry the correct sign (negative) and are statistically significant in the entire models at 10% except in model 5 where it is significant at 5% with t-statistics ranging between -1.79 and -2.06. The speed of adjustment of economic parameters to equilibrium is ranging between 20 and 36% to GDP growth rate in the short run. This implies that convergence to equilibrium after shock to the variables in Nigeria will take about two and half years. Hence, the ECM is able to correct any deviations in the relationship between GDP growth rate and explanatory variables. The results show that both contemporary physical capital and lagged one are negative but not significant in any of the models. Contemporary and lagged one interest rate has positive impact on economic growth. This is consistence with Yinusa and Akinlo (2013). The contemporary interest rate is not significant while lagged one interest rate is significant in first three models. Lagged one labour is positive in all the models but not significant in any of the models. Therefore, conclusion cannot be drawn. Labour lagged two is negative and significant at 5% level of significance in all the models. This result is consistence with the results of Adawo (2011) but contrary to the findings of Jaiyeoba (2015) and Omotayo (2015). The reason why our finding is contrary might be due to the differences in variable used to measure human. It is evidence from the literatures that some of the studies used primary and secondary enrolment while some studies total labour force. Trade openness has negative and significant impact on economic growth in Nigeria. The coefficient of trade openness is negative in all the models. The coefficients of manufacturing are positive and significant. This implies that manufacturing contributes to economic growth in Nigeria.This negative impact of physical capital might be caused by lack of technological innovation in Nigeria. It could also be due to the lack of enough investment in the as a result of capital flight due to high level of corruption. In the result, human capital has significant negative effect on economic growth. This might due to the high illiteracy rate in Nigeria and as many workers are unskilled, leading to their low productivity. Also, it could be that the personnel management system in firms and enterprises does not allow well-educated employee to contribute meaningfully to the enterprises. Government need to invest more on the labour force so as to improve the quality of the labour force. The positive impact of interest rate on economic growth is contrary to expectation. The positive impact of the interest rate might be due to interest rate liberalization or lower and stable interest rates which create a lot of economic certainties that encourages investors to borrow and invest in productivity-improving projects. This result is consistent with existing empirical findings (Oosterbaan et al., 2000). The negative effect of trade openness on economic growth is not surprising as the economy is a mono product economy. 6.0 Conclusion This study examined the impact of human capital on economic growth in Nigeria. To achieve this objective we employed Error Correction Model for the analysis. The study performed unit root tests to determine the stationarity of the data and cointegration test to determine if there is log run relationship among the variables used in the study. The results show that all the variables are stationary at first difference and that there is long run relationship among the variables. The results further show that human capital has negative and significant impact on economic growth in Nigeria. There is therefore the need for the government to improve the quality of the work force through the restructuring of the educational system, organising series of seminars, training and retraining of the work force so as to reverse the negative effect of human capital on economic growth.