In the labor market, workers can increase their chances of meeting potential employers through costly search. My first chapter aims to empirically quantify the search intensity of workers, both employed and unemployed. My second chapter develops a theoretical model to study the optimal unemployment insurance with search intensity endogenously chosen by unemployed workers. I devote my third chapter to empirical identification of whether major illness leads to persistent household poverty in rural China.

My first chapter studies the search behaviors both on and off the job, and the effect of search intensity on wage determination. Four determinants of wages are considered: productivity, workers' bargaining power, competition between employers due to on-the-job search, and search intensity. I estimate the structural model using the 2001 panel of the Survey of Income and Program Participation (SIPP), together with supplementary information from the American Time Use Survey (ATUS). The empirical results demonstrate that search intensity declines as the worker gets a wage rise from her current job.

My second chapter addresses the efficiency issues arising from the externalities and hidden-action features of search effort. The solution to the social planner's problem may not be decentralized in a competitive market. Calibration shows that the current US unemployment insurance (UI) system generates an 8.07% welfare loss relative to the socially optimal allocation.

In the third chapter, I use a unique dataset on Chinese rural households to test whether severe illness can cripple a rural household's economic resources leading to temporary and/or persistent poverty. When health shocks are assumed to be exogenous, in the sense that households cannot control the arrival rate of adverse health shocks by choosing the amount of medical expenditures, a Markov regime-switching regression model reveals no significant evidence that a severe illness causes persistent household poverty. To endogenize health shocks and choices on medical expenses, a dynamic structural model is employed. The structural estimates support the view that major illness leads to persistent household poverty.