We develop several pieces of evidence about the reallocative effects of the COVID-19 shock on impact and over time. First, the shock caused three to four new hires for every ten layoffs from March 1 to mid-May 2020. Second, we project that one-third or more of the layoffs during this period are permanent in the sense that job losers won't return to their old jobs at their previous employers. Third, firm-level forecasts at a one-year horizon imply rates of expected job and sales reallocation that are two to five times larger from April to June 2020 than before the pandemic. Fourth, full days working from home will triple from 5 percent of all workdays in 2019 to more than 15 percent after the pandemic ends. We also document pandemic-induced job gains at many firms and a sharp rise in cross-firm equity return dispersion in reaction to the pandemic. After developing the evidence, we consider implications for the economic outlook and for policy. Unemployment benefit levels that exceed worker earnings, policies that subsidize employee retention irrespective of the employer's commercial outlook, and barriers to worker mobility and business formation impede reallocation responses to the COVID-19 shock.