Learning to Love Massive Unemployment Benefits

L ast year, as the pandemic was just striking the U.S., there was a minor kerfuffle over Congress’s plan to give the unemployed an extra $600 a week. Some Republicans initially assumed it was a drafting error, because it would give many laid-off workers more money than they’d made while working.

It was actually intentional. States’ unemployment systems run on creaky old technology, and they can’t quickly switch to complicated new formulas. When COVID-19 kicked a bunch of people out of work, the only way to boost their benefits — which normally replace only a fraction of their earnings — was to add a flat amount to what they’d normally get, and just about any decent-sized amount would pay at least some people more than they’d made before. The $600 boost, however, nearly tripled the usual benefit and paid the median worker about 45 percent more than he’d made while working, with lower-income workers doing better than that and higher-income workers doing worse.

It wasn’t crazy to think that this would discourage workers from getting new jobs if they could. I was concerned about it myself. But a new study shows that effect to be rather small. The unemployment top-up didn’t keep too many people from work, but it did boost spending at a time when the economy really needed it.

Making unemployment benefits more attractive than work would be an insane policy in normal times, and the study does nothing to change that. But it turned out okay this time. And the debate is far from over: After the $600 boost expired in August, it was brought back twice at $300, first for six weeks under an executive order and later through the round of stimulus that Donald Trump signed late last year. It’s set to expire in March, but Democrats want to keep it going.

The study, from researchers at the University of Chicago and the JPMorgan Chase Institute, relies on anonymized data from the bank accounts of Chase customers running through October of 2020. These numbers are especially good for tracking the spending of people who got unemployment benefits.

People who lose their jobs usually start spending a lot less, and Americans in general cut their spending during the pandemic. But when their benefits were boosted, unemployed workers actually started spending more. Estimating these workers’ “marginal propensity to consume,” the authors conclude that the beneficiaries spent about 30 to 40 cents of each extra dollar within a month, and about two-thirds of the extra cash within six months.

This was important to the economy, as it came at a time when businesses really needed the extra dollars. Unemployment benefits accounted for about 7 percent of all personal income in June of last year. The authors estimate that the $600 boost pumped up overall spending by 2 percent or more between April and July.

The result is also relevant to stimulus policy going forward. Giving cash to the unemployed appears to have higher bang for the buck than other ways of pushing dollars into the economy, including those free-money-for-all stimulus checks everyone is obsessed with. Giving money to people who actually lost jobs is also more defensible as a form of “relief.”

But what about the effects on job-finding? Fortunately, the Chase data also allow the researchers to see when people went off unemployment and back to work. And as with spending, the patterns here are very different from what we see in normal times, when studies show that higher unemployment benefits can drastically reduce job-seeking.

People were slower to leave the unemployment rolls, but not just while the boost was in effect. Before the pandemic, about 10 percent of unemployed workers left the program every week, but that number cratered when COVID-19 hit, rebounded somewhat, and then kept pretty steady all the way from May through October — despite the $600 boost expiring in August, with the $300 boost from the executive order being sent out haphazardly over the next couple of months.

Presumably the job-finding rate stayed so low and consistent, even as benefits changed, because it’s hard to find a new gig when the economy is in the toilet and a raging virus makes interviews dangerous. Indeed, one rationale for the expanded benefits was that we didn’t necessarily want people working; even the usual requirement that the unemployed actively search for a job was waived in most states.

However, there is one concerning aspect to these numbers. If the $600 boost was discouraging people from finding work, we should see folks abruptly getting jobs right as it expires. We do see this, especially for those whose benefits were especially high relative to their previous incomes, as shown in the August peak in this chart.

As the authors explain, however, this effect is rather small. The folks who abruptly left unemployment in the three weeks after the boost expired account for only about 3 percent of the workers who had joined the program between late March and late July. By contrast, more than half of the workers who got the $600 supplement returned before it expired. Using some complicated modeling, the authors estimate that the disincentives created by the $600 boost decreased employment by only 0.2 to 0.4 percent over the whole period from April to July.

Another interesting finding is that 50 to 75 percent of people who left unemployment during the pandemic last year returned to their old employer (depending on the month, with lower numbers later in the year), as opposed to 20 percent during normal times. By law, laid-off workers receiving unemployment have to accept an offer to return, though there were pandemic-related exceptions in effect, as well as concerns about how well this could be enforced with unemployment systems so overwhelmed. Some also worried that workers would lose the connections to their employers if they were laid off and went on unemployment, instead of continuing to receive their normal paychecks. But many, many workers were asked to come back and did.

These results help to answer two big questions. First, was the $600 supplement good policy when we enacted it? It probably wasn’t the ideal way to structure a benefit, but boosting spending by 2 percent while mildly disincentivizing employment for a few months is a good trade in a pandemic. In fact, the added spending almost certainly supported more employment than the disincentives destroyed. In the new study’s calculations, the supplement created at least $450,000 in spending for every job lost because it discouraged work, while estimates of how much spending is needed to create a job tend to come in at $125,000 or less.

Second, should we keep paying a supplement past March, when the current $300 boost expires? Maybe not: That boost still pays about 40 percent of workers more than they made before, and as we get closer and closer to normal, the highly unusual results of this study will become less and less applicable. The Congressional Budget Office already expects a historically typical unemployment rate this year, and the population is getting vaccinated rapidly.

However, the authors note that lots of the unemployed are still dealing with long-term job loss or repeated spells out of work thanks to the pandemic, which certainly warrants sympathy. I’m still leery, and if it’s going to continue, I’d like to see the supplement reduced further or at least tied to economic conditions.

Truly massive unemployment benefits are not a policy that I, or really any conservative, would normally even entertain. But COVID-19 does crazy things to all of us.