Unemployment insurance reduces employment
The claim in this title is one of the most well established claims in all of economics. If you are one of those “believe the science” people, then you should believe it to be true. It is based on the following pieces of evidence:
1. Many good empirical studies that show that UI reduces employment.
2. Basic economic theory predicts that unemployment insurance reduces
3. Casual empiricism: Anyone who travels around the country frequently meets people who say, “I know this guy who refuses to go back to work because he can make more on unemployment.
4. Introspection: If I had a crappy job and someone offered to pay me more to sit at home, would I take the offer? Of course—who wouldn’t?
This claim is often treated as a “conservative” opinion, but distinguished left-of-center economists like Brad DeLong and Larry Summers also believe it to be true. (It’s also in Paul Krugman’s textbook.) The difference is that progressives are more likely to favor UI because they believe the benefits outweigh the costs.
And yet, the media doesn’t like unpalatable truths, and thus is reluctant to acknowledge this fact. Here’s a recent headline:
Ending unemployment benefits had little impact on jobs and fueled $2 billion spending cut, study finds
I guess this is supposed to be read as evidence that the consensus view in the economics profession is wrong. How should we react to this news?
Suppose there had been 100 previous studies of UI, and also assume (hypothetically) that 90 found a negative effect on employment and 10 did not. Then this study might change the ratio to 90 and 11. In other words, a good Bayesian would not meaningfully revise their view of the effect of UI on employment on the basis of this one study, unless the study employed econometric techniques that were much more powerful than previous studies.
In fact, however, the study did not find no effect; it found that UI reduced employment, in which case it would be 91 to 10 in favor of the standard view. This is from the paper cited by the article:
With those caveats and if the only impact of the policy change were through the labor supply of the previously unemployed, extrapolating from our job finding estimates (along with the 2.9 million individuals receiving UI in end of April) suggests an additional 35 thousand additional jobs were created in June and 135 thousand in July, but 25 thousand fewer (so far) in August. This implies that the unemployment rate in Withdrawal states would have been 4.8 percent in the absence of the change, as opposed to 4.5 percent in reality. Note, however, that since the federal pandemic unemployment insurance programs are slated to expire on September 6, 2021, these approximately 145 thousand additional jobs (adding June, July and August together) would have likely been created a few months later without the early withdrawal.
Because of the “caveats” mentioned in the quote, the authors believe the actual effect was even smaller. Those caveats involved things like the impact on aggregate demand and the crowding out of other job seekers such as teenagers (who don’t typical get UI). I tend to discount these caveats, as AD is certainly not holding back employment right now and teens have no trouble finding jobs in a country were low skilled employers are desperately short of workers.
It is true that the study found a smaller effect than some might have anticipated, but keep in mind that this policy change did not eliminate UI, it merely eliminated the $300/month bonus. And it only applied to some states. And it’s only been a short period of time. 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 (the national rate is 5.4%), making job growth more difficult. It also occurred at a time when many people are flush with cash due to the recent stimulus payments. In other words, it’s just one study, to be added to all the previous studies.
The recent decline in the supply of workers is probably due to a number of factors:
1. Covid scaring off workers.
2. School closures keeping parents at home.
3. Less immigration.
4. Stock gains and Covid causing increased retirement.
5. Stimulus money having an “income effect”.
6. Enhanced UI benefits.
Many countries have labor shortages right now, so that suggests that Covid is directly or indirectly a factor. (By “indirectly” I mean leading to changes in government policy as a result of the Covid recession.)
This is from the news article cited above:
States that ended federal benefits early saw larger job gains among the unemployed: Their employment jumped 4.4 percentage points relative to jobless individuals in states that kept benefits flowing, according to the paper, which analyzes data through the first week of August.However, that translates to just 1 in 8 unemployed individuals in the “cutoff states” who found a job in that time period. The majority, 7 out of 8, didn’t find a new job.
“Yes, there was an uptick [in employment],” University of Massachusetts Amherst economics professor Arindrajit Dube said. “Most people lost benefits and weren’t able to find jobs.”
I hate to pick on Dube, both because he’s a brilliant economist and because I know what it’s like to talk to reporters, but I can’t help saying that I hate the “weren’t able to find jobs” framing. Given that so many employers of low-skilled workers are short of staff, I highly doubt that 7 out of 8 unemployed workers were not “able” to find jobs. Rather, I’d guess that many remained out of work for other reasons.
PS. You can argue that employers should pay more, but that opinion has no bearing on anything in this post. Also note that this post does not discuss whether UI is a good idea, an entirely different question.