An Evaluation of Endogenous Model on Wage among Farming Households in Nigeria

Abstract

Some variables may be endogenous in wage equation. However, this endogenous hypothesis is not usually tested in most applied studies. Therefore, this study empirically applied two stage least squares (TSLS) to examine the return to education among farming households in Nigeria using the Nigerian household living standard data of the National Bureau of Statistics (NBS) in the World Bank sponsored project. Tests of endogeneity, instrument relevance and over-identifying restriction were carried out. The test of endogeneity confirms ordinary least squares (OLS) should not be relied on as it does not give a consistent and reliable estimates. The instruments included are exogenous and not correlated with the error term, but correlated with the independent variable confirming the relevance of the instruments. The empirical result affirms education is an endogenous determinant of wage rate. One unit increase in year of education will increase the wage rate by #439.67 (Four Hundred and Thirty Nine Naira, Sixty Seven Kobo), holding other variables constant. It suggests that test for endogeneity is necessary for an unbiased and consistent estimate to be obtained. Thus ascertaining the endogenous nature of explanatory variable(s) is important to guiding the appropriate estimation method to be applied. This also affects the statistical inference and policy recommendation.

Keywords:     Instrumental, Variable, Two stage least squares, Ordinary least squares, Education, Wage Rate