Comment

In this provocative paper, Hall and Kudlyak show that during a typical recession in the United States from 1949 to 2019, a short-lived burst of job loss explains much of the rapid increase in the unemployment rate. During the subsequent expansion, unemployment declines comparatively slowly and steadily, primarily due to a sharp and persistent decline in the job finding rate, both for the initial group of job losers and for other workers who became unemployed only later in the business cycle. Hall and Kudlyak argue that the elevated jobless rate for the latter group is evidence that unemployment is “contagious” or “infectious.” For the most part, I will not take issue with their facts, although I will make the (well-known) observation that the 2020 pandemic recession and subsequent expansion featured the fastest increase in unemployment on record, followed by the fastest decrease on record. Although this recession was different in many ways from past ones, the fact that unemployment fell so quickly during the early stages of this expansionmay be useful for diagnosing why unemployment declined so slowly during prior expansions. I will focusmy attention on the claim that unemployment is contagious or infectious. Hall and Kudlyak are very clear about what they mean by this: “We consider negative feedback from high unemployment to the job finding rate as a key mechanism behind the slow unemployment recoveries.” The function g(u) in equation (14) exemplifies this logic. SectionVIII of their paper sketches a number of endogenousmechanisms that cangenerate such feedback. Iwill not try to critique each—or indeed any—