new predictions of effects of alternative UI benefit supplements

The UI supplements have expired. Congress is considering a range of options.

What will happen to
1) *consumption*
2) *UI replacement rates*

Thread w/ @JoeVavra @pascaljnoel

1) with @Farrell_Diana @FionaGreigDC @maxliebeskind using #JPMCInstitute data, we estimate a two-week MPC on UI benefits of 73 cents. Multiplying $600 by the number of UI recipients and dividing by total consumption gives an estimate of % change in cons. https://institute.jpmorganchase.com/institute/research/labor-markets/report-consumption-effects-of-unemployment-insurance-during-the-covid-19-pandemic

The decline in consumption from "no supplement" is greater than the entire peak-to-trough decline of the Great Recession.

This reflects just how much the US economy is on life support right now through UI benefits.

2) What will happen to replacement rates?

Although in the past we have focused entirely on replacement rates in % terms, it is also useful to see the *difference* between benefits and prior earnings. We do this because some of the people who have replacement rates > 100% are actually quite close to replacement in $ terms

The reason we released this update is that we now take into account wages of people unemployed during this recession using the April, May and June 2020 CPS. (Our earlier draft relied on people unemployed in 2019). https://bfi.uchicago.edu/working-paper/2020-62/

Unemployment has skewed in this recession towards workers with low earnings. This plot shows you how the update from 2019 to 2020 data affects our estimates (we find replacement rates that are a bit higher)

These estimates come with many assumptions, as is the case for all real-time research. For example, we don't know how many people are getting UI benefits because of some data challenges. We use a rough estimate of 25M from @bencasselman @NelsonSchwartz https://www.nytimes.com/2020/07/02/business/economy/jobs-unemployment-coronavirus.html

Likewise, our UI MPC estimate uses all bank account outflows. If people are transferring $ to their savings accts then we overstate the MPC. I think this is unlikely and will have new results soon, one more open question.
\\end{thread}