The effect of government educational expenditure on labor productivity in Turkish manufacturing sector

Labor productivity is very important phenomena to maintain the sustainable economic growth. Actually, there are many studies in the economic literature about determining the factors affecting labor productivity. At this point, this study aims to investigate the relationship between labor productivity and government educational expenditure by using the data covering the period of 1971-2015 for Turkey. To this aim, it is applied linear and non-linear autoregressive distributed lag model (ARDL). By using the bound testing approach, the long run relationship between variables has been investigated. Both models have been applied to determine whether the relationship between variables is linear or nonlinear. An asymmetric relationship between labor productivity and government educational expenditure was determined. For this reason, nonlinear model is preferred. The empirical findings indicate that positive changes in educational expenditure has negative effect on labor productivity and negative changes in educational expenditure has positive effect on labor productivity. The long run coefficient of openness is significantly positive; the short run coefficient of it is significantly negative for all lags. On the other hand, the effect of GDP growth on the labor productivity is significantly positive in both short and long run. While the effect of average annual labor worked hours is significantly negative for the first two lags, it is significantly positive the last two lags in short run, it is positive but not significant in long run.