How Has the Pandemic Changed the Labor Market?

Has the pandemic exacerbated trends that were already in the labor market or introduced a new range of issues?

Davis: It’s exacerbated some trends and reversed others. First, it’s been a sharp, negative blow to less-educated workers. That’s both a continuation of a decades-long trend, but something of a reversal of the past 2–3 years, in which they had been seeing a better labor market. Second, we’ve had agglomeration in cities for more than a century, and that has been reversed, at least temporarily. People are eager to get out of really densely populated areas for fear of infection risk. Thirdly, we had some slow trends toward interacting remotely and working from home before the pandemic, but they were greatly accelerated by the pandemic.

Bertrand: At the beginning of the recession, people were talking about COVID being an economic equalizer, but it’s been exactly the opposite. It has increased inequality across income, across socioeconomic groups, and between genders. There’s no doubt that the lower wage sector of the labor market is bearing, so far, the brunt. It’s somewhat different from what we’ve seen historically in terms of industrial composition: most recent recessions were more focused on manufacturing and construction. This time, it’s service-sector workers, gig workers, self-employed workers who have been heavily impacted and were not covered by a safety net prior to the CARES [Coronavirus Aid, Relief, and Economic Security] Act. What the new administration does with regard to these workers is a big, outstanding issue. The other factor is that this has been more of a “shecession” than a “mancession.” And that raises issues about women’s continued engagement with the labor force.

Notowidigdo: In terms of the gendered aspect of the COVID recession, some of this could reflect preexisting trends. The United States has been somewhat of an outlier among developed countries. At the start of the 2000s, women worked about as much in the US as in Germany, France, and the United Kingdom, and in the past couple of decades, those labor-force participation rates have diverged. Some of what’s going on during the COVID recession could just be a continuation of these trends, although, of course, there’s childcare and other issues that might be disproportionately affecting women during the COVID recession.

Davis: I agree on how the gender impact of COVID has played out so far, but there are potentially important differences over the longer run and possibly big benefits for women, especially more-educated women. My survey work with [Stanford’s] Nick Bloom and [Mexico Autonomous Institute of Technology’s] Jose Maria Barrero shows a tremendous decline in the stigma associated with working from home. There’s also been a lot of investment in making work from home better, and technological advances to improve remote interactivity. When we ask people whether they would like to work from home, it’s quite uniform across demographic groups—a lot of people would like to work from home two days a week. Working from home is more attractive for those who can. That’s mainly better-educated, professional people. That could be favorable for more-educated women.

Bertrand: I very much agree. The other silver lining is that among the higher educated who have been working from home, there’s more fathers engaging in childcare who might not do so otherwise. I’m worried, though, about what’s happening to the current generation of women who may be really missing out on a valuable time in their career and how they are going to recover from that. Also, flexibility can take very different forms. Working on the weekend and late at night is more of a norm, and that, again, is going to be harder for women, no matter what.

How effective was the US federal government’s initial support for small businesses?

Bertrand: It’s too early to tell whether the Paycheck Protection Program has been effective. It has helped, but whether it was money well spent is still up in the air. Especially for very small businesses—retail, food, restaurants—the situation is dire. There was a quick recovery up until the middle of June, but the data suggest that nothing much has improved since then. A looming issue is that a lot of small businesses face the possible end of a moratorium on evictions, and many small-business owners might be sitting on overdue rental payments.

Davis: If your metric was “spend a lot of money quickly,” the PPP succeeded. How much better did it make things than they would have been without policy of that particular form? You could have had more generous funds for low-income families or a broader coverage of unemployed workers. Would small businesses have been better off if the federal government had spent more money on trying to get a handle on the dimensions of the coronavirus itself by establishing mass testing and contact regimes? Just knowing how big the risks are, and where they are, and what types of nonpharmaceutical interventions are most effective could be helpful to all businesses. Small businesses suffered so much because we made most of them shut down. If we had figured out some way to deal with the virus while not hammering small businesses so hard, that would have been a better outcome than just giving them funds to tide them over.

Bertrand: My sense of the research on nonpharmaceutical interventions is that they do not explain much of what happened with small businesses.

Davis: It’s still an open question. Either way, it doesn’t address the counterfactual. We spent trillions in combating the economic effects of the pandemic. For a few billion or even tens of billions, we could have established large-scale randomized testing, both for antibodies and infections, in the population. You can imagine randomly sampling census tracts, and paying people $500 to take the test. What would we have learned if we had done that? We’d have learned about the prevalence of infection at a point in time, and related it to the tests that were done for health-care diagnostic reasons. We could have informed people about their actual risk. We could have learned about the infection mortality rate more quickly. We’d have learned about how far we were from herd immunity. If we’d tested for several months, we’d have learned about the duration of immunity. We’d have learned more directly about the effectiveness of these nonpharmaceutical interventions. I’m disappointed that as part of the CARES Act, we didn’t set aside resources to do that. There’s also the public infrastructure to implement contact tracing. That’s a bigger investment, but it would benefit us tremendously as we go forward with this pandemic and with future pandemics.

What have we learned about the US unemployment safety net?

Davis: One thing that the whole COVID pandemic has revealed quite starkly is how inadequate our unemployment-insurance system is as a claims-processing entity. First, in many states, the system fell way behind in processing claims or in verifying that claims were legitimate. People who lost their jobs had to wait weeks to get anything. That’s extreme hardship, and it’s an insult. Secondly, there were claims—about which I’m dubious—that there’s no way to provide additional unemployment benefits, except by a fixed amount. That’s such an easy problem to fix. What does it take for a state to send an unemployment-insurance benefit payment to somebody? They’ve either got to know the bank account, the name, and the social-security number—they already know that—[for an electronic transfer], or they have to have a physical address, a name, and a social-security number to send a check. Even if the state can’t do a finely tailored expansion of unemployment-insurance benefits, it can send that information to the federal government. The federal government already sends out checks to tens of millions of people. This seems like a billion-dollar problem, to revamp the IT infrastructure of all the UI [unemployment-insurance] systems. Thirdly, there’s the administrative shortcomings of the UI system. Like Marianne, I’m sympathetic to providing some UI benefits to gig workers, but that part of the CARES Act is especially prone to fraud, so we need some investment in systems to deter fraud and abuse, both because it’s costly and a misuse of the system, but also because fraud undermines political support for the system. There’s a lot of missed opportunities in improving the back end of our unemployment benefit system.

Bertrand: Many have been calling for building in more automatic stabilizers within our UI or safety-net system. Seeing the gridlock that we have in Washington, with people having to wait, and fearing the uncertainty of what will happen to them, it’s high time we had more of these automatic stabilizers built into the systems. The triggers could be based on the unemployment rates in a particular community, which could automatically change the level of unemployment benefits—something that would not rely on an act of Congress. Think about how much the well-being of all the people who have been affected would have increased if they had not had to wait for Congress to get its act together.

Notowidigdo: I agree. I was involved in a project in which we surveyed countries during the Great Recession. We found that in many countries, unemployment benefits became more generous during recessions, exactly when the unemployment rate crossed a specific threshold, and that benefit extension happened automatically. I certainly think this is something the US could do. We should also think about future stimulus as income maintenance or income replacement. The unemployment rate is continuing to drop, the labor market is recovering, and it may not be clear that we need more stimulus. But many Americans are still struggling, and this relates to the disparate impact of the COVID recession. That’s a good argument for targeted income relief.

What are the labor-market challenges facing the Biden administration?

Davis: There are parts of the labor market that are doing quite well. At the same time, there are parts of the labor market defined by skill groups, demographics, and geographic areas where things are really quite bad. We talked a little bit about service jobs in business districts—people working at restaurants, baristas in coffee shops, people providing personal services to office workers. That part of the economy is pretty devastated. So there’s both a skill aspect and a spatial aspect. Part of the job for Congress, the president, and local officials is, for those people who aren’t going back to the jobs they used to have, how can we create new job opportunities for them quickly? That strikes me as the central challenge from a labor-market perspective. A lot of this is going to happen on the ground level. If you look at cities like Detroit, or Gary, Indiana, they basically failed to adjust to a massive demand collapse. And of course, the massive demand collapse in steel, in the case of Gary, or the auto industry, in the case of Detroit, was the triggering event. But then the local officials in a city have to recognize, “Look, we’ve got to repurpose our commercial space, our local economies, and we’ve got to do it expeditiously.” That involves business licensing, zoning, new ways of organizing residences and businesses. Some cities do that well; some don’t. I worry a lot about the left-behind people, the left-behind places, and the left-behind people in the left-behind places. That is partly a consequence of policy, not just at the national level but also at the local level.

Notowidigdo: I’m optimistic we’ll return to an era of incremental reforms. What would that look like in terms of tackling inequality? I’m guessing that it would be a modest increase in progressive taxes, and I doubt we’ll go anywhere close to what Biden had promised during the campaign, or any of the major changes that were suggested during the Democratic primary, in terms of things like large wealth taxes. But I think we’ll make small steps to returning to what tax rates looked like under Obama and under Clinton. Is that going to make a huge dent in inequality? Maybe a little bit.

Davis: There’s opportunity for bipartisan agreement. For example, in our working-from-home research, when we ask people about how productive they are when working from home, how efficient they are, there’s a lot of heterogeneity. There’s two main reasons for that. For those who are living in a confined space with kids and feel they’re less productive working from home, that’s pretty hard for policy to address. But some people feel they’re less productive because they’ve got poor-quality internet connections. And not surprisingly, those are concentrated among lower- or middle-income, less-educated folks. We hear all this desire for spending on public infrastructure, and this is a big gap in the quality of public infrastructure. Here’s an opportunity both to improve the distribution of outcomes and to do it in a way that is actually productivity enhancing. There aren’t many of those, so hopefully we'll take advantage of that. I can imagine Republicans and Democrats coalescing around an infrastructure investment program of that sort.

What is the most enduring effect the pandemic will have on the US labor market?

Davis: The working-from-home phenomenon is huge. At the peak of pandemic-induced lockdowns in April and May, more than half of all paid hours worked were done at home—more than 60 percent by our estimates, if you weight by earnings, because higher income people were more likely to work from home. Even in November, it was still at 35–40 percent. When we ask individuals, “What does your employer plan for you to do after the pandemic’s over?” they give a number like 23 percent [of their hours worked from home]. In another survey, where we’ve asked employers directly, they give a number closer to 20 percent. Before the pandemic, it was more like 5 percent of all paid workdays. So that’s a huge, persistent shift in the structure of working arrangements that I expect will endure.

Bertrand: I agree about working from home. The crisis has also dramatically accelerated the continued reshaping of our retail sector, a trend we’ve seen for a long time. That’s more a product-market development, but it’s going to have implications for the labor market as well. It’s striking to see the rate of share in food service—the share of chains versus independent restaurants, and how that’s changing. People are going to be returning to jobs, but they’re going to be different kinds of jobs and different kinds of employers. That’s going to be fascinating to study,

Notowidigdo: We’re not just going to be working from home, we’re going to be consuming at home more—watching movies at home instead of going to movie theaters, for example. That’s going to be another enduring impact. If you think about that in the context of health care, a lot of states almost immediately started experimenting with telemedicine at the start of the COVID recession—they were reimbursing providers to offer telemedicine visits. This is a sort of forced experimentation that we’re going to keep with us for at least several years after the pandemic ends. We’re going to figure out what kinds of health care you can consume at home without having to go to the doctor’s office or a hospital. There could be some real positives to come out of that. It could increase access for people who, prior to the COVID recession, had difficulty making it to a doctor’s office or going to the hospital. Telemedicine could provide a way for them to get the health care that they need.