Olga and Hugo explain why the Great Resignation is the wrong story to describe what’s happening in the U.S. labor market. Job turnover is a welcome sign for the economy, and wage increases for low-income workers can boost economic growth.
Olga Bitel: It seems like investors are now fretting over something that we would have loved to have had during the last decade. Today, we have a very dynamic labor market. People are not suddenly realizing that they hate their jobs. They’re moving between companies because there are better opportunities, whether that means working remotely or closer to home or having a bigger paycheck or a more interesting job. More and more people are being absorbed back into the labor market. Wages for the bottom 40% of the labor cohort are actually rising, which supports domestic demand. These are all the conditions that we wanted to have, but didn’t, in the aftermath of the Great Financial Crisis.
Hugo Scott-Gall: If this dynamic labor market that you’re describing is so good, why are so many people talking about it as if it were bad? With your growth-economist hat on, what would you say is an optimum amount of labor-market dynamism? Were we just too far below that amount before?
Olga: I don’t know how to think about the optimal level. But we can say definitively that in the last decade, the number of jobs on a monthly basis that people parted with and the number of jobs that people acquired dipped relative to the ’90s and the 2000s.
If on average, per month, in the United States, people leave about 5.5 million to 6 million jobs and they find 5.5 million to 6 million jobs, the net is what is reported. When we say the U.S. economy increased its number of jobs by 200,000, that’s not the total number of jobs that got filled. It’s the net.
The underlying numbers in the last decade were considerably lower than in prior years. We also saw very muted wage growth for virtually all cohorts of employees. There were some notable exceptions, but those were mostly in the top income bracket and the very, very top of the top income bracket. Apart from those exceptions, the traditional labor market was quite a bit slower.
That’s bad, because virtually all of your domestic demand growth comes from lower cohorts on the income distribution. These are the folks who are more likely to spend all of their income. So, the more dynamic the labor market, the higher the wage growth in those cohorts—assuming it’s not terribly in excess of productivity gains—the stronger your domestic demand and the more sustained your growth will be. In the 2010s, obviously, we had both a slow labor market and lackluster growth.
That’s why the current labor-market dynamism is a good thing. We should be celebrating it. The Great Resignation—the narrative that people don’t want to work or that COVID has somehow altered their perception of the benefit of work—doesn’t make sense. At the height of COVID, we talked about a shortage of 21 million employees. By the end of 2020, we were 10 million short. Today, we are less than one-half that. People are getting reabsorbed, coming back to their previous jobs or going to new jobs. It’s not as though we suddenly find ourselves with a population in which nobody wants to work.
Hugo: Isn’t it interesting that the narrative that emerged around the worst period of COVID—that a lot of people were going to permanently leave the workforce—seems to have been overestimated?
What’s the best way of measuring how well a country is working? Is it taking the population ages 18 to 65 and dividing that by total hours worked? Obviously, labor-force participation depends on what you think the labor force is. If the labor force, as a percentage of 18- to 65-year-olds, shrinks, you can have what looks like high unemployment only because you’ve reduced what you’re dividing by.
Olga: If we think about hours worked, then the economies of East Asia—I’m thinking specifically about South Korea, but to a lesser extent Japan as well—they would look off the charts. But we also know that past a certain threshold, your productivity drops quite considerably. It’s not as though working 12- to 15-hour days is just as productive as working the first five hours in your day.
Furthermore, if you keep doing this for an extended period, it really erodes your productivity—so much so that when we’ve seen experiments with a four-day workweek, without exception, productivity gains have been tremendous, 30%–40%. We’ve seen this experiment in Finland, Japan, and New Zealand. I think something similar is now unfolding in California.
To your point on labor-force participation, that is the number of working-age people who are actually working, apart from those who are in school or otherwise productively engaged. If we look at the Nordic countries here, so Denmark, Sweden, their participation rates are close to 80%. That is what makes your pension system sustainable. That’s what makes your healthcare benefits sustainable. That’s what makes your domestic demand growth sustainable. That seems to be the magic rate at which the cost-benefit analysis in a modern, liberal democracy adds up. In the United States, labor-force participation is in the low 60s. So, we’re quite a way off from the magic number of 80.
Hugo: Yeah, that all makes sense. And so, if wages were to rise very strongly in real terms, which they’re not doing now, would we discover that demand creates more supply, and would we see more people joining the workforce? Is full employment price-sensitive?
Olga: Ah, that’s exactly where I was going with this. The labor-market inflation theory is quite elegant and has been around for a long time. But we’ve never seen it supported empirically. There hasn’t been a real-world example in the past, say, 100 years where we can say that there is something called the wage-price spiral and that excessive wages feed into inflation.
Hugo: So, what worries you about the U.S. labor market? Is it the shift we’ve seen in the Beveridge curve, such that reducing unemployment now requires a much higher rate of job openings? We talked about the volume of people, but we didn’t talk about skills. Do you see a big gap in skills demanded versus skills supplied?
Olga: That’s one reason why you see significant price discrepancies, and that will show up in wages. There’s a bit of a barbell going on. What have been really decimated in the last several years, even before COVID, are U.S. immigration numbers. We used to invite around 700,000 highly skilled workers per year. That pool has virtually dried up. If you aggregate that shortfall over five or six years, you’re talking about several million people. These are in tech-dependent, highly skilled, information technology-services, engineering-type jobs, for which the United States is just not producing enough graduates. That shortfall is only likely to get exacerbated with time.
The shortfall in lower-skilled immigrants is much harder to measure because they are often in the country illegally. Many of them work in high-contact services, such as gardening, construction, and housekeeping, where you see a lot of shortages as well.
Hugo: In a country that is reducing immigration, doesn’t that just build up wage pressure?
Olga: Potentially. Wage growth in the lower cohort has not kept up with productivity gains, and it certainly has not kept up with inflation, in many cases. So, we could be a few years away from that. But yes, that’s the argument here. Unless we either retrain the people that we currently have or revert to encouraging immigration, eventually, there will be wage pressure in those areas of the labor market where the needs are not being met.
Hugo: How many mature economies or economies with high per capita income have ever retrained their populations? I get that countries with very low gross domestic product per capita can do it.
Olga: I can only think of Sweden after its 1994 crisis. This is a country that was effectively bankrupt. At the height of its bankruptcy, when its GDP basically collapsed by 30%, the country maintained its job-security councils, which help displaced workers with financial support and retraining opportunities. So, a lot of people went back to school or turned to vocational training and became IT professionals, developers, tech-maintenance professionals, things of that sort.
Something similar exists in the United States today. I think Cisco is running a program that takes two years and costs $30,000. You become a systems operator, and you can earn a salary of $100,000–$150,000 per year after that. The pool of potential labor in the United States is much vaster than it was in Sweden at that time. We established that our labor-force participation rate is in the 60s, not in the 80s. There’s quite a bit of underemployment.
Hugo: What’s the sensitivity of the U.S. workforce to asset prices? I’m sure plenty of people think they can retire based on the average asset prices of the last two years. If those fell a lot, would that jeopardize quite a few retirement plans and bring older people back into the workforce? Would that be a “boomer”-rang?
Olga: Well, that would be a bad reason to bring people back into the workforce. A better reason would be rehiring them as part-time consultants in areas where there’s a lot of demand; the knowledge gap is significant and they can really bring their expertise to bear, to do something they enjoy and feel rewarded and relevant.
Hugo: The United States is always assumed to be quite a friction-free place in that it’s one homogenous country. But could certain changes increase the dynamism of the U.S. labor market? For one thing, isn’t there a lot of regulation in certain professions so that credentials don’t transfer from state A to state B? And the relicensing requirements are pretty onerous. I saw a chart claiming that hairdressers and operators of funeral parlors must retrain to be licensed in other states. You wouldn’t have thought that the tonsorial requirements of someone from Kentucky would be radically different from someone from Missouri.
Olga: You wouldn’t have thought. But you’re right, these state licensing requirements have been growing quite significantly, in part because they’re a form of taxation for the state. Every time you go through a licensing process, you pay a fee. The U.S. labor market isn’t nearly as dynamic as maybe it once was—or maybe it has never been as dynamic as it has been portrayed.
I suppose the punchline for us is that the current labor-market dynamism and much-higher-than-expected wage growth, especially in the lower cohorts of the income distribution, for now is quite a welcome change from the last decade. All else being equal, this could support stronger growth in domestic demand.