Newly published paper by @danascoot@finamor_lucas which has I think the best evidence to date on the incentive effects of the $600 weekly supplement (🧵)
The paper uses time clock data from small biz, many of which are restaurants. they compare workers with higher and lower earnings in 2019 & ask "were workers with lower earnings in 2019 (and therefore higher benefit replacement rates) slower to return to work after expiration?"
the answer is no, workers with higher replacement rates were not slower to return to work. this holds even among firms that were hiring (see plot above).
This suggests that the \$600 supplement was not a constraint to firms that wanted to (re)hire during the summer.
the study has some limitations which it does a good job of acknowledging: don't know which workers actually get UI or what their benefit levels are, and only focusing on workers who worked substantial hours in 2019.
It will take some time for the economics profession to fully sort out the effects of benefit supplements. This is the best study I have seen so far on the topic.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
(I've seen a few places where folks have been confused about this so I wanted to provide a brief explanation of composition bias.)
The best plot I know showing this is by @ElizaForsythe & Matias Cortes
\{thread}
A big Uchicago-Fed Board-Princeton team including @JohnRGrigsby@Ahu_Yildirmaz has shown the same fact using payroll data from @ADP
Why might someone incorrectly think wages are rising? If you take average earnings *per current worker*, they have risen sharply during covid. That is what the blue line in Eliza's plot above shows.
You get the same thing from the BLS's Current Employment Statistics.
In the spring, the US Congress orchestrated the largest ever expansion of weekly unemployment benefits via a $600 per week supplement to the unemployed. This supplement expired at the end of July.
What happened to spending and savings in August in #JPMCInstitute data?