Self-Employment and Self-Selection in Developing Country Labor Markets

Abstract

Labor markets in Less Developed Countries (LDCs) have traditionally been characterized as non-competitive. This view is expressed in well-known models in the development economics literature. For example, the labor surplus model of Lewis [16] asserts that labor in rural areas of LDCs in early stages of development is so abundant that if the wage rate were determined competitively it would be below the subsistence level. Hence institutions are assumed to arise to determine the wage rate non-competitively, and to provide the resulting excess supply of labor to agriculture with a subsistence level of living. The HarrisTodaro [11] model of rural-urban migration asserts that labor unions, minimum wage laws, public sector hiring practices and other institutional factors tend to keep formal sector wages in urban areas above labor-market clearing levels. The resulting excess supply of labor in urban areas is assumed to survive at a subsistence level by becoming selfemployed in the low productivity informal sector. Recent empirical studies provide little support for the non-competitive view of LDC labor markets. Evidence from India, Thailand, Malaysia, Kenya, and Guatemala suggests that labor markets are much less imperfect than often assumed.' An important empirical finding in several of these studies is that self-employed workers appear to have higher earnings than their wage employee counterparts in urban areas.2 This fact is inconsistent with the notion that self-employment is typically a low-productivity occupation engaged in by individuals who are searching for scarce, well-paying formal sector jobs. The purpose of this study is to analyze and estimate a model of individual choice between wage and self-employment in the LDC context. The model incorporates a production function in the self-employment sector that includes managerial ability as an input. Analysis of the influence of managerial ability on the choice of self-employment leads to a theory that can explain the observed pattern of earnings of the urban self-employed compared to wage employees. Implications of the model for estimation of production function