The role of human capital in the private manufacturing sector productivity in the developing and transition economies.

Abstract

This thesis principally seeks to provide empirical examination of the contribution of human capital, particularly in the form of education, to productivity at the micor level, through the lens of human capital and production theories, for a pooled sample of countries from the Middle East and North Africa (MENA), in comparison with both the Eastern Europe and Central Asia countries (ECA). This research mainly aims to establish substantive empirical evidence on the varying effects of human capital on growth, across regions. It also aims to investigate the role of human capital investment in the productivity gains, mainly through efficiency and labour productivity, in the formal private manufacturing sector, in the aforementioned regions. The thesis takes into account the variations in per capita income levels, based on the World Bank classifications of countries by income group. In addition, this research recognises and takes into account the heterogeneity which exists throughout the selected sample of countries. The main objective of evaluating the impact of human capital is to untangle the existing differences in the firms’ performance, partly on account of employing different workers with varying levels of education, with distinctive regional socio-economic changes, and different political conditions. The stochastic frontier analysis (SFA), as a fully parameterised model is used, in order to address and examine the determinants affecting production efficiencies, especially from a human capital point of view, and in the light of Vandenbussche, Aghion, and Meghir’s 2006 assumptions, on growth, distance to frontier, and composition of human capital, which remains untested in MENA and ECA at the firm-level. The SFA was applied following the approach of Caudill, Ford, and Gropper (1995) (CFG) by estimating and testing stochastic frontier production functions, assuming the presence of heteroscedasticity in the one-sided error term (inefficiency), and by following the approaches of Hadri (1999) for cross sectional data assuming the existence of heteroscedasticity in both error terms (the one-sided inefficiency term and the two-sided symmetric random noise), in order to obtain more accurate measures of technical efficiency. However, the rationale for this choice of the two different regions, is the heterogenous organisational structures, and the dissimilarities between production functions across economies in different developmental phases, which can be used as a suitable platform for analysing the distinctive effects of human capital composition on efficiency, and growth in each region in comparison with the other. In addition, the applied methodology also involves the incorporation of two matching methods consisting of a completely randomised experimental design, propensity score matching (PSM), and a fully blocked experimental design, Mahalanobis distance matching (MDM), using a cross-sectional firm level dataset, in order to examine the causal effects of formal training on productivity in MENA, and in ECA. The main conclusion of the empirical analysis suggests that highly-educated labour proxied by workers with tertiary education and those with university degree, appear to have a positive and statistically significant impact on efficiency in the two regions. Noting that the closer is the country to the frontier, the more important this level of human capital tends to be. As a country becomes closer to the frontier, it depends more on innovation and knowledge creation, which leads to the reallocation of labour from unskilled-complementary technology production activities, to skilled-prejudiced and technology-intensive activities. This result appears to confirm the association between high levels of human capital and growth, and chimes with the relevant literature about the link between human capital and growth in the developing and developed countries. It was also found that low-skilled labour component, denoted by workers who attended secondary school, seemed to have positive and statistically significant contribution to efficiency only in the less developed countries, such as MENA. This is due to the fact that the further the country is from the technological frontier, the more reliant the country becomes on imitation activities, and this seemed to corroborate the ideas posited in the literature about the sources of growth and the proximity to the world’s technological frontier. The low-skilled labour in the private manufacturing firms, in MENA, is positively associated with high levels of efficiency, and its impact appears to be significant, especially in high- technology firms. Although in the more affluent countries, such as the high-income economies in Eastern Europe, and the middle-income economies in ECA at large, the impact of secondary school workers gives the impression of being insignificant on efficiency. With respect to the intermediate-skilled labour, which is represented by the proportion of workers who have been trained in technical schools, or received on-the-job training, the maximum likelihood estimates point that their effects on efficiency have a propensity to be statistically insignificant, in MENA and ECA, in reducing the effects of inefficiency in firms’ performance. In fact, intermediate-skilled labour is found to have a positive and significant relationship with higher levels of inefficiency, especially in MENA. Put simply, it impedes efficiency improvements in the manufacturing firms, particularly, in the low and medium-technology plants in MENA. Furthermore, the effects of highly-skilled workers on efficiency were found to be positive and of a high level of significance in the low and medium-technology firms, and this is quite clear, especially, in the high-technology manufacturing firms in this region. All in all, the results of this study are in line, and compare well with the hypotheses of endogenous growth models of Lucas (1988) and Romer (1990), and with the assumptions of Benhabib and Spiegel (1994), that the economic growth is conditional on the human capital accumulation to improve efficiency and increase productivity in order to catch up with the technological frontier and shift it upward.