Understanding the responses of businesses and local economies to pro-employment policies is of critical economic and public policy interest. Better understanding of these policies not only can improve our ability to achieve major social goals, but also can shed light on more fundamental aspects of how local economies work and of how firms respond to incentives. This dissertation focuses on two specific cases of pro-employment policies: payroll subsidies for nursing homes and fiscal stimulus during the Great Recession.

For the first policy, I study the effect of payroll subsidies offered by state Medicaid programs to nursing homes on nursing home employment and wages. I identify the effect of subsidies using within-state, across-nursing home variation in subsidy rates and find that the subsidies were very effective at inducing nursing homes to increase nurse and nursing assistant employment and wages. Subsidy effect estimates are consistent with 100% pass-through to labor, with an implied elasticity of employment to subsidies on top of average Medicaid payments of 4.5. Beyond these baseline findings, I also investigate how the effectiveness of subsidies varies by nursing home market competitiveness and across for-profit and not-for-profit nursing homes. Cumulatively, my findings indicate that nursing home payroll subsidies are substantially more effective than previously thought, suggesting that revisiting the efficacy of other payroll subsidies using firm-level subsidy variation would be valuable as well.

For the second policy, in work joint with coauthors, I examine the employment effects of the American Recovery and Reinvestment Act of 2009's stimulus expenditure during the Great Recession. We use across-county, within-state variation in stimulus to identify local fiscal multipliers, with a focus on identifying how multipliers vary by how severely a place was affected by the Great Recession. We find that the employment multiplier is more than twice as large in the half of counties most negatively affected by the recession than in the least affected half of counties. These findings demonstrate that the fiscal multiplier varies spatially across local labor markets as a function of exposure to employment reductions. They also imply that an employment-maximizing stimulus package targeted to high excess capacity counties would have created nearly twice as many jobs as were actually created by the American Recovery and Reinvestment Act.