Distributive policy

Is it better to support jobs or workers?

America has seen five waves since the start of the pandemic.

The March-April 2020 pandemic recession was a different creature from previous post-World War II recessions: different in cause, duration, depth, and the types of social and economic changes that occurred. The appropriate economic policy response was also different. Instead of the standard anti-recession policy of stimulating the whole economy, it is more useful to think of pandemic recession policy as a form of social insurance. A key question is whether this social insurance should work primarily by supporting the unemployed or by supporting jobs.

Lest this distinction sound like a pun, consider this real-world difference. During the pandemic, most European countries responded with “part-time work” schemes. The idea is that the employer does not need to fire or lay off workers. Instead, it dramatically reduces their hours, and the government makes up the difference. It’s a kind of partial unemployment, except that when the worst of the pandemic’s short, sharp blows to the economy passed, workers were still employed at their previous jobs and employers could increase their hours again. In contrast, the American approach emphasized larger and longer unemployment benefits for those who were out of work. U.S. employers (with the exception of some small state-level programs) did not have the option to transition to short-time working.

Giulia Giupponi, Camille Landais, and Alice Lapeyre discuss the trade-offs between these two approaches in “Should We Insure Workers or Jobs During Recessions?” (Spring 2022, Economic Outlook Journal, 36:2, 29-54). Here is one of their figurines. The solid lines show the share of the population receiving unemployment insurance, the blue line showing the United States, and the red line showing a weighted average for Germany, France, Italy and the United Kingdom. Note that the share of workers receiving unemployment insurance during the pandemic is increasing in the United States (solid blue line) but barely moving in European countries (solid red line). Conversely, the share of short-time workers is increasing in European countries (red dashed line) but barely budgets in the United States (blue dashed line).


The bottom panel of the figure shows the difference in the share of the “unemployed” adult population, which peaks in the United States, but does not move much in European economies, because part-time workers still have jobs.

Giupponi, Landais, and Lapeyre examine the trade-offs in detail. I will try to offer a quick summary here. The Unemployment Insurance-based approach has provided a higher level of assistance to people (indeed, the United States’ response to the pandemic has in some cases involved Unemployment Insurance payments in excess of previous wages ). However, the incentive for firms to lay off workers and for workers to become unemployed was much weaker under the short-time working system.

The great advantage of the short-time work approach is that it preserves the previous job match between employer and employee, which preserves the professional knowledge and skills contained in the special relationship between a certain employer. and a certain worker, and also avoids dismissal and rehire costs. On the other hand, since the pandemic recession (like all recessions) involves a reallocation of workers from one sector to another, partial unemployment tends to inhibit this reallocation more than an approach through unemployment insurance.

Finally, workers generally benefit from a “tight” labor market, which means that there are a relatively small number of job vacancies for available workers. In an American context, the consequences of the pandemic recession have thus been historically unusual, with unemployment rates having quickly returned to very low levels and the American economy in a situation where job creations exceed jobs. It seems plausible that high unemployment insurance payments in the United States, by encouraging some workers to leave their jobs, have contributed to these tight labor markets in the United States, thereby benefiting other workers.

The authors are also quick to point out that a society need not make a binary choice between these approaches. For example, it may be that short-time working is better for a short and sharp negative shock that has a high probability of reversing, while unemployment insurance is better for a shock that seems likely to last longer and less likely to reverse. One can imagine that one approach would work better for workers in some industries than others.

But there is another dimension to this choice: when an unexpected shock like a pandemic recession hits and a rapid policy response is needed, what is the government already prepared to do? The administrative political apparatus of the American government was not designed to resort to partial unemployment, even if it had been desirable to do so. A group of co-authors – David Autor, David, David Cho, Leland D. Crane, Mita Goldar, Byron Lutz, Joshua Montes, William B. Peterman, David Ratner, Daniel Villar and Ahu Yildirmaz – export this question in “The $800 Billion Program Paycheck Protection: Where Did the Money Go and Why? » (Spring 2022, Economic Outlook Journal36:2, p. 55-80).

These authors point out that when the pandemic hit in March 2020, providing financial relief to the economy seemed more important than taking the time to implement a carefully designed and targeted program. Moreover, the US government did not have the pre-existing administrative capacity to accelerate an immediate program of partial unemployment. So, alongside higher unemployment benefits and stimulus checks, the U.S. government enacted the $800 billion Paycheck Protection Plan, with the goal of providing funds to small businesses. The funds were provided by guaranteeing loans from banks: indeed, any company with up to 500 employees could obtain a loan of up to $10 million, which would then be canceled if the company maintained employment and wages. at about the same level for 2 – 6 months.

Remember that while some economic sectors were very hard hit by the pandemic recession, others were less hard hit or recovered more quickly. For these companies, PPP was essentially free money. Unsurprisingly, 94% of eligible companies signed up at the cost to the government of $800 billion. The authors summarize:

The Paycheck Protection Program has had measurable impacts. This significantly mitigated pandemic-related job losses, preserving between 1.98 and 3.0 million job-years during and after the pandemic at a substantial cost of $169,000 to $258,000 per job-year saved. The PPP also reduced the rate of temporary closures among small businesses, although it is less clear whether it reduced permanent closures. However, the majority of PPP loan dollars issued in 2020 – 66% to 77% – were not disbursed to paychecks, but instead were disbursed to business owners and shareholders. And because business ownership and shareholding is concentrated among high-income households, the program’s impact on the overall household income distribution has been highly regressive. We estimate that about three-quarters of the benefits of PPP accrue to the top quintile of household income. By comparison, the impact of federal pandemic unemployment insurance and household stimulus payments was much more evenly distributed.

However, the authors hesitate to settle for PPP with the luxury of hindsight. As they point out:

The regressive distributional impact of PPP and its limited effectiveness as an economic stimulus stem from the program’s lack of targeting. This absence, in turn, reflected necessity. Given time constraints and, more profoundly, the lack of existing administrative infrastructure to oversee targeted federal support for the entire US small business population at the start of the pandemic, we strongly suspect that Congress would not have could not better target the Paycheck Protection Program without significantly slowing its delivery. We therefore agree with Bartik et al. (2021) that policymakers made a defensible trade-off between speed and focus in PPP design.

Of course, the underlying lesson here is that if the United States is to have the ability to use furlough in response to a future recession, the administrative infrastructure must be put in place in advance. But one of the main frustrations of current policy-making is that there seems to be so little focus on learning the policy lessons of the pandemic – not just on employment issues, but also on regulatory and public health issues – on institutions, arrangements, and priorities must be established now, before they are urgently needed.