This thesis comprises three chapters that delve into various labor market dynamics and the policy implications in the context of the COVID-19 pandemic and its aftermath. In the first chapter (joint with Andre Kurmann and Etienne Lale), we investigate the dynamics of small businesses and employment using real-time data from the private sector throughout the COVID-19 crisis. The pandemic has led to an explosion of research using private-sector data to measure small business activity. Yet important questions remain about sample representativeness and how to identify business openings and closings. We propose new methods to address these issues by exploiting information on business activity from Google, Facebook, and Safegraph. We apply our methods to Homebase data and show that the resulting estimates closely fit official statistics. We then use the data to study whether small businesses have been hit harder by the pandemic and the extent to which the Paycheck Protection Program helped mitigate these effects. The second chapter (joint with Andreas Hornstein, Marios Karabarbounis, Andre Kurmann, Etienne Lale) focus on the effects of pandemic unemployment insurance (UI) benefits. UI acts as both a disincentive for labor supply and as a stimulus for labor demand. In equilibrium, the two effects combine, which may explain why several studies have found only small negative effects of the generous UI expansions during the pandemic on job finding rates and employment. In this paper we propose a new research design to estimate independently the disincentive effects of pandemic unemployment benefits. Using high-frequency worker-firm matched data from Homebase, we document that employment of low-wage businesses recovered more slowly from the initial pandemic shock than neighboring high-wage businesses, and that this recovery gap is significantly related to the relative generosity of UI benefits. By comparing neighboring businesses that are largely sharing the benefits of the local UI stimulus, our research design identifies more closely the disincentive effects of pandemic UI benefits. We use an equilibrium model of labor search with heterogeneity in firms and workers to translate the reduced-form estimate of the recovery gap into an unemployment duration elasticity and an aggregate employment loss. Our model, which captures well the recovery gap between low- and high-wage businesses, implies relatively low duration elasticities. Yet, the sheer size and multitude of the pandemic programs implies that the disincentive effects arising from the pandemic UI benefits are substantial and amount to 5 percent of normal employment. The third chapter studies work-from-home (WFH) work mode's implications on labor market. The COVID-19 pandemic has triggered a widespread adoption of WFH practices and accelerated the advancement of remote work technologies. Surprisingly, even after the pandemic has subsided, a substantial shift towards WFH remains evident among workers. However, the accessibility to WFH is not uniform across all types of workers. Notably, high-tech industries, characterized by a predominantly high-skilled workforce, exhibit a higher prevalence of WFH. This raises concerns about the effects of WFH on workers employed in industries where remote work is unfeasible. In this paper, I develop a spatial equilibrium model that incorporates WFH to examine the implications on workers' mobility, local market outcomes, and overall welfare. I find 3 key insights: (1) there is a productivity threshold for WFH adoption, (2) there is a one-way dependence of low-skilled workers on high-skilled workers' mobility, and (3) if workers are fully mobile, both types of workers benefit from the introduction of WFH.