Labor Market Impacts Of Social Security Policies In Developing Countries: A Case Study Of Pakistan

This dissertation studies the impact of changes in social security policies on labor market outcomes in Pakistan. In the first chapter, I use a firm-size contingent policy change to examine the effects of Social Security extension to previously uncovered workers in the informal sector. In Pakistan, the social security law that provides for old age pensions and permanent disability benefits, the Employees' Old Age Benefits Act, was enacted on July 1, 1976 and was applicable to all private sector firms employing ten or more workers. Under this law, firms have to make mandatory contributions to the Employees' Old Age Benefits Institution at the rate of 6% of the federal minimum wage for all workers under their employment. In July 2008, the law was amended to include under its jurisdiction all firms employing five or more workers. This setting provides a natural experiment wherein the cost of employment for firms with five to nine employees increased after the policy change while remaining fixed for firms with less than five employees. Using individual level data from the Pakistan Labor Force Survey, I employ a difference-in-differences approach to identify the effects of the cost increase on wages and employment in firms exposed to the law. I find evidence of full cost shifting to wages with negligible effects on employment. My results are consistent with a positive valuation of social security benefits by workers and downward wage flexibility in the Pakistani labor market. The Employees' Old Age Benefits Institution (EOBI) follows a Pay-As-You-Go pension system, which requires the institution to balance its efforts of Social Security extension with the financial constraints posed by these initiatives. The most recent actuarial valuation of EOBI fund shows that the system may not be sustainable in the long run. The threat of long term insolvency coupled with a desire to extend Social Security ben- efits to a majority of the working population provides the motivation for the second chapter of the dissertation. As Social Security affects labor force participation decisions over the life-cycle, I develop and estimate a dynamic structural life-cycle model to evaluate policy proposals that can help to reduce the EOBI deficit without exerting a negative impact on individual labor supply or aggregate welfare. The preferred policy proposes a reduction in the payroll tax rate accompanied with a reduction in the generosity of pension benefits. The former helps to increase labor force participation at younger ages, while the latter increases labor force participation of older workers. The second chapter uses Social Security coverage to group workers into formal (covered wage employment) and informal (uncovered wage employment and self-employment) sector jobs. The third chapter takes this approach a step further and tests for the existence of labor market segmentation using a more holistic definition of formality and informality. First I offer an analysis of earnings differentials between formal and informal sector wage employees using an endogenous regime switching approach. I find positive earnings differentials between formal and informal sector jobs, and a 10.7% wage premium in the formal sector. In addition, there is evidence of queuing for formal sector jobs; individuals who prefer to work in the formal sector have longer unemployment durations. Second, I examine the patterns of worker mobility across various labor market segments and compare these patterns to the ones predicted by the segmented labor market model. My results suggest that the labor market is only partially segmented along the formal-informal dimension. However, the existence of two distinguishable segments emerges within the informal sector itself