Overview

Abstract (English)

This thesis is comprised of three chapters, each in the field of labor economics. The first chapter proposes a new way of thinking about the impact of temporary help agency placement. These placements are shown to affect job match quality and resulting wages of temporary workers differently, depending on which workers self select into this form of job search. An assignment model shows that temporary help agencies will only sort and improve the job match quality of some workers. The framework is applied to the case of different post-contract wage impacts across gender. More “sortable” women choose temporary help agencies relative to men because of unobserved characteristics. As a result, placement coincides with sorting for women more-so than men. The second chapter finds that job match quality changes because of the response of firms to economic conditions. In a downturn, firms post relatively more manual skill jobs than they otherwise would, leading to an increase in overqualification. A model of job search shows that firms will exploit the relatively large share of highly educated workers that are unemployed in a downturn, hoping to attract these workers at a lower cost. In the model, this behavior leads to endogenous overqualification. Empirical estimates support the model results, finding that the changing skill requirements of newly formed jobs, rather than the economic conditions themselves, contribute to overqualification. The third chapter of this thesis uses instrumental variables estimation and unique synthetic microdata from Canada to compare the causal impact of two factors affecting the supply of crime: labor market conditions and the justice system. The findings suggest that individuals at the margin of committing property crimes and assaults consider the opportu- nity cost of offending. Similar evidence does not exist, however, for drug crimes, murder or prostitution. Crime is most elastic with respect to the probability of capture, measured by the cleared by charge rate. Offense rates respond somewhat less to labor market conditions, measured with wages and unemployment rates.

Abstract (French)

Please note that abstracts only appear in the language of the publication and might not have a translation.