Just a week ago, I did a post that questioned a recent study of the effects of ending the $300 supplemental unemployment benefits (in 25 states). The problem was not that the study was wrong, rather that it was merely one of lots of similar studies done over recent decades, and its results need to be viewed with caution. It sure didn’t take long for my warning to be validated. Here’s what I said:
And lots of other things are going on that might have effected the results. As just one example, the study occurred during the recent Covid delta variant surge that disproportionately hit the “red states”. And which are the states that eliminated the $300 bonus? Mostly red states. These states also had a significantly lower unemployment rate than average . . . making job growth more difficult.
And here’s today’s headline in the FT:
US job growth slows sharply as Delta variant hits recovery
Total of new positions drops to 235,000 in August from 1.1m in July but unemployment rate falls
Read those figures again. We went from a blowout figure of 1.1 million in July (probably aided by the expiring benefits in 25 states), to an abysmal 235,000 in August. You don’t need a PhD in economics to understand that this is mostly due to the delta variant. Indeed, a couple of days ago the American Economic Association convention that takes place in early January each year was moved online. (A silly decision, in my view.)
Last week, I saw some people on twitter criticizing UI skeptics for refusing to admit they were wrong about the effects of ending the supplemental UI benefits. In light of today’s jobs report, I wonder if those same people will admit they were wrong about the implications of these new studies.
PS. The FT article describes Goldman Sachs providing what seems like a more clear-eyed view of the effects of this policy:
Goldman Sachs economists estimate that July’s jobs growth would have been 400,000 higher had the enhanced benefits expired nationwide, and forecast next week’s termination to add 1.5m in payroll gains by the end of the year.
READER COMMENTS
Jose Pablo
Sep 3 2021 at 7:02pm
“Goldman Sachs economists estimate that July’s jobs growth would have been 400,000 higher had the enhanced benefits expired nationwide, and forecast next week’s termination to add 1.5m in payroll gains by the end of the year.”
You can be pretty sure that the payroll gains by the end of the year will NOT be 1.5m … which does not mean, in any possible way, that Goldman Sachs’s model is “wrong” on the effects of the enhanced benefits expiration.
The fact that you can be so easily fooled by your observations, is just too difficult to apprehend for partisans (on both sides). I am afraid you are fighting windmills here, Scott.
Following your co-blogger Bryan Caplan favorite strategy, what would be the best way of transforming this discussion into a properly worded bet you would be willing to put money in? One that even Goldman Such would be willing to underwrite (although I can imagine bets you will be willing to put money in, but Goldman Such would not underwrite … more than the other way around)
Scott Sumner
Sep 3 2021 at 7:50pm
I think you misunderstood what Goldman Sachs is claiming. It’s a policy counterfactual. The actual numbers depend on many different factors.
Jose Pablo
Sep 4 2021 at 6:28pm
Yes, I did. I thought the 1.5 m addition in payroll gains was Goldman Sachs’s forecast for “total” gains between now and December taking into account next week enhanced benefits expiration and all other inputs in their model.
The main point is still valid, though: Since the observable values of “economic variables” depend on many different factors, you can always fight tooth and nail the apparent “falsification” of your “favorite theory”. The impact of minimum wage regulation on employment came to mind.
Trying to word the statements on “betting form” can help. After all, “bets” should be, by definition, falsifiable at some point in time. But this can only get you so far, I guess.
Comments are closed.