Are manufacturing workers really worth their pay?

Much attention has been focused in recent years on the importance of education in raising labor productivity. Policy makers repeatedly stress the value of investing in education because it is believed that formal schooling and on-the-job training raise the quality of workers. Countries with highly skilled work forces are more productive than those with less skilled workers and have a competitive advantage in several strategic areas. Empirical evidence suggests that workers with above average levels of education are quicker to implement new technology, are more complementary to physical capital, and improve both technical and allocative efficiency. Moreover, numerous studies have revealed a significant relationship between the accumulation of human capital and per capita growth rates, particularly in developing countries. For example, Lau et al (1990) estimate that a one year increase in the average years of schooling of the workforce in East Asia and Latin America could raise real GDP by 3–5%. While there is little doubt that the expansion of high quality education at the national level has very high returns, no empirical study to date has estimated the direct effect of education on physical output at the firm level. This void in the literature is surprising given the fact that all human capital theory rests upon the assumption that education raises the marginal physical product of workers. This paper aims partially to fill this void by presenting evidence on the direct impact of education on productivity using applied production analysis. Specifically, the paper has two aims: (1) to provide empirical evidence on the effect of different levels of education on marginal productivity at the firm level, and (2) to test how closely wages proxy marginal product. To date, there is no other study to the author's knowledge which estimates the elasticity of firm output with respect to human capital using production analysis. Instead, the conventional approach has been that earnings are used as a proxy for productivity and then earnings functions are used to estimate the effect of education on productivity. The organization of this paper is as follows. Section 1 discusses the data that are used for analysis.Section 2 presents the theoretical background on the effects of education on production and discusses the advantages of a production function approach. Section 3 specifies a modified form of a Cobb-Douglas production function which permits the inclusion of educational categories and other variables affecting productivity. Section 4 presents the estimates obtained from the production function analysis which are associated with various levels of formal and non-formal schooling. Section 5 contains the concluding remarks.