The effects of private equity buyouts on employment, productivity, and job reallocation vary tremendously with macroeconomic and credit conditions, across private equity groups, and by type of buyout. We reach this conclusion by examining the most extensive database of U.S. buyouts ever compiled, encompassing thousands of buyout targets from 1980 to 2013 and millions of control firms. Employment shrinks 12% over two years after buyouts of publicly listed firms—on average, and relative to control firms—but expands 15% after buyouts of privately held firms. Postbuyout productivity gains at target firms are large on average and much larger yet for deals executed amid tight credit conditions. A postbuyout tightening of credit conditions or slowing of gross domestic product growth curtails employment growth and intrafirm job reallocation at target firms. We also show that buyout effects differ across the private equity groups that sponsor buyouts, and these differences persist over time at the group level. Rapid upscaling in deal flow at the group level brings lower employment growth at target firms. We relate these findings to theories of private equity that highlight agency problems at portfolio firms and within the private equity industry itself.

This paper develops a quantitative theory of how people weigh the risks of infections against the benefits of engaging in social interactions that contribute to the spread of infectious diseases. Our framework takes into account the effects of public policies and private behavior on the spread of the disease. We evaluate the model using a novel micro panel dataset on consumption expenditures of young and older people across the first three waves of COVID-19 in Portugal. Our model highlights the critical role of expectations in shaping how human behavior influences the dynamics of epidemics.

In statistics, samples are drawn from a population in a data-generating process (DGP). Standard errors measure the uncertainty in estimates of population parameters. In science, evidence is generated to test hypotheses in an evidence-generating process (EGP). We claim that EGP variation across researchers adds uncertainty—nonstandard errors (NSEs). We study NSEs by letting 164 teams test the same hypotheses on the same data. NSEs turn out to be sizable, but smaller for more reproducible or higher rated research. Adding peer-review stages reduces NSEs. We further find that this type of uncertainty is underestimated by participants.