CYCLICALITY OF THE US SAFETY NET: EVIDENCE FROM THE 2000s AND IMPLICATIONS FOR THE COVID-19 CRISIS

In this paper, we explore the cyclicality of the U.S. safety net over the 2000s through the economic peak in February 2020 before the onset of the COVID-19 crisis. We compare the effects of means-tested programs with those of social insurance programs, separately and combined. We find, on a per capita basis, Unemployment Insurance (UI) is by far the most cyclical, particularly when fully funded federal extensions are included. A 1-percentage-point increase in the unemployment rate leads to a 17 percent increase in monthly real UI spending. Overall, the social insurance programs provide an additional $31 (2019$) in per capita real spending for each percentage point increase in the annual unemployment rate, while the means-tested programs provide a statistically insignificant $8.50 per capita for each percentage point increase in unemployment. The means-tested programs without SSI provide a significant $12 for each percentage point increase in the unemployment rate. Thus, the parts of the means-tested safety net that can respond quickly are also providing modest countercyclical stabilization. We conclude by speculating what this means for the current response to COVID-19.