Testing for Labor Rationing: Revealed Preference Estimates from Demand Shocks

National Bureau Of Economic Research Inc, Cambridge MA

Investigators

Abstract

Abstract This research project will test for and quantify the existence of surplus labor---individuals who are involuntarily rationed out of the wage labor market---in low income communities. In both the U.S. and in developing countries, many social programs are predicated on the idea of surplus labor; yet there has not been an empirical test of its existence. The research will use field experiments which involve creating many industry jobs in a nearby city to test the labor surplus theory in a rural setting. If rural workers can be absorbed into these factory jobs without any change in the wage or employment levels in the rural labor market, then this will demonstrate the existence of surplus labor. A pilot study of 36 local labor markets provides initial support for the existence of labor surplus in low income communities. The results of this research will provide a better explanation of the functioning of labor markets in low income and high unemployment communities, and therefore provide guidance on crafting policies to reduce unemployment or to combat the effects of such high unemployment rates. In addition, by revealing which types of self-employed small entrepreneurs would prefer wage work, the findings have implications for how to better target programs to enable small business growth in high unemployment settings. The results will be central to understanding many issues in development economics and will have applications in several fields of economics, such as labor economics, public economics, and industrial policy. By improving the functioning of the U.S. labor market, the results thus improve U.S. competitiveness. This research project uses field experiments to obtain revealed preference estimates of labor rationing in rural labor markets, using rural labor markets in India as the test bed. A large labor demand shock will be created by recruiting up to about 30 percent of workers in surrounding villages to work in nearby factory jobs. If the amount of labor rationing is at least as big as the demand shock, then there will be: (i) no effect on local village level wages, (ii) no change in local village employment, and (iii) positive employment spillovers on workers who do not receive factory jobs. In addition, a supplementary experiment will be conducted to estimate the short run elasticity of supply of labor, in order to check whether labor supply is perfectly elastic. Examination of who benefits from unemployment spillovers can be used to explain the two possible components of labor rationing---voluntary unemployment and disguised unemployment. Finally, comparing traditional survey-based measures of unemployment to these revealed preference estimates can help provide some validation of the reliability of traditional survey measures. The results of this research provide strong microeconomic foundations of labor market rationing and provide guidance for labor market policies in both low income and developed countries.