Capitalisn’t: Can Labor Markets Save Capitalism?

David Autor: The idea of using tools to make less expert people able to do more expert things, if we can pull it off, would be a very powerful use of this technology.

Bethany: I’m Bethany McLean.

Phil Donahue: Did you ever have a moment of doubt about capitalism and whether greed’s a good idea?

Luigi: And I’m Luigi Zingales.

Bernie Sanders: We have socialism for the very rich, rugged individualism for the poor.

Bethany: This is Capitalisn’t, a podcast about what is working in capitalism.

Milton Friedman: First of all, tell me, is there some society you know that doesn’t run on greed?

Luigi: And, most importantly, what isn’t.

Warren Buffett: We ought to do better by the people that get left behind. I don’t think we should kill the capitalist system in the process.

Bethany: Over the last four decades, one of the biggest issues and challenges to capitalism has been growing inequality. You can’t have a healthy economy or healthy society with terrible and rising inequality. We already knew this was a problem, and when the pandemic hit, the experts said it was going to get worse.

How could it not get worse? Low-wage jobs disappeared because industries like hospitality and restaurants were disproportionately affected by the shutdowns. Clearly, this was a disaster for inequality.

Luigi: But this is not what happened. All the experts were totally wrong. By early 2022, the employment-to-population ratio among adults with only a high-school education had already attained a hundred percent of its early-2020 level, which by the way, was already a very high level historically. People with only a high-school education recovered all their jobs by the beginning of 2022. This is one of the factoids that emerged in a recent paper by David Autor.

In his paper, he writes, “Rapid relative wage growth at the bottom of the distribution reduced the college-wage premium and counteracted approximately one quarter of the four-decade increase in the aggregate 90-10 log wage inequality.”

Translated into English, that means that the change in the differential between what the people in the 90th percentile of the income distribution make and what people in the 10th percentile of the income distribution make has increased disproportionately in the last 40 years. Now, one-quarter of that increase has been made up just in the last two years.

Bethany: It’s a pretty stunning statistic. Yes, inflation has risen so fast that the fall in inequality has not meant, actually, earnings growth for almost anyone, but Autor and his colleagues found that young high-school workers constitute the only group that has not seen its post-pandemic earnings gains entirely eroded by inflation.

This is a big deal, but what happened? Why and what does it mean for the future?

Luigi: This seems to me the definition of socialism, that we become equally miserable and that you should rejoice in the fact that we’re equally miserable.

Jokes aside, David Autor is regarded as one of the foremost experts on employment and technological change. He’s an economist, public-policy scholar, and professor of economics at the Massachusetts Institute of Technology, where he also acts as a codirector of the School Effectiveness and Inequality Initiative. And he is joining us on the podcast to discuss his paper and much, much more.

Bethany: Before we bring David in, Luigi, let me ask you a few questions.

What makes this research, this new paper he did, so groundbreaking? Is it the conclusion that inequality, strangely enough, in the aftermath of the pandemic has actually narrowed, or is there something also about the way in which he did the research that is different?

Luigi: First of all, I think that I would like to spend a little bit of time describing his overall research because it’s very important.

First of all, he really nailed what we now call the China shock. He nailed how important the fast liberalization of trade with China was in impacting employment in the United States, but particularly employment in Midwest regions that were affected very negatively. He is basically the father of the China-shock literature, and also, he is the one who has closely studied not only the displacement effect that technology has on work, but also the creation of new jobs that come with technology.

Bethany: Just a couple of technical things before we move on to actually talking to David. In his paper he talks about this notion of imperfect labor-market competition.

What does that mean, and how is that different from how most economists think about the labor market?

Luigi: It’s actually stunning. I’m not a labor economist, but I read a lot of this literature recently, and even the top labor economists say, until recently, we thought that there weren’t a lot of frictions in the labor market. More or less, people were paid their competitive wage.

Now, people are realizing that there are a lot of frictions. Some of them have to do with mobility; some of that is with segmentation of information. Some of that may be even with dual-career issues, et cetera. Many employers succeed in holding onto people paying less than their market wage. That’s what having some market power means, and this is more diffuse than we make it out to be.

If you remember, when we interviewed Eric Posner about his book, his book was all about the fact that even the antitrust authorities were not looking for the exercise of market power in the labor market because it was thought to be so weak. Now, I think people have changed their minds, and they’re looking at it much more carefully. Surprise, surprise, when they look, they actually do find something.

Bethany: Without further ado, let’s bring in David and talk about his work, and then you and I can circle back and discuss what we think.

Luigi: David, you are one of the few economists I know who has been fired from a real job, and for sure the only one I know who was fired because he was wearing nail polish.

To what extent has your experience in the real world changed your work as a labor economist?

David Autor: I started as a student at Columbia. I dropped out after three semesters. I was a very unhappy kid. I was not suitable for employment, and so I was just up to something nutty every weekend, and one day I came in and I had decided, “Oh, wouldn’t it be fun to wear nail polish this week?”

I was doing microelectronic assembly, meaning I was putting together integrated circuits with my hands, and my boss came in and saw me handling the integrated circuits with nail polish, and that was just not OK. That was the end of that job.

I had a number of jobs, and one of those jobs was, I spent several years as an educational director at a nonprofit in San Francisco, working at a Black Methodist church, where we did computer-skills training for poor kids and adults. This was back in the late 1980s, and we were funded by Apple and a lot of the other Silicon Valley companies.

There was a sense of this digital divide. We didn’t know what that meant, and I don’t think if we knew if we were effective, but we believed strongly that the tools required for work were changing, and many people who were going to need those tools didn’t have access to them.

The combination of that work and my own work . . . I also worked at McDonald’s, I also worked as a temp, I also worked at a pizza place for a long time, I also did software development. That’s how I got so old.

It made me very cognizant of the way the kind of winds of change in the labor market affect the livelihoods of lots of people. This is why I’ve spent a lot of my career thinking about technology, also trade, also temporary-help agencies. Things that impinge on the quality of jobs and often move things around really quickly.

Bethany: You’ve talked about reimagining capitalism in order to save it.

How important is getting the nitty-gritty of labor markets right to saving capitalism?

David Autor: Oh, I think you can’t save capitalism without it. Nitty-gritty is, I would call . . . This is the biggest in the big picture. I think Dani Rodrick has put this very well. He says, you know, you can think about three ways you can sort of intervene in income distribution. One way is through what you would call premarket activities like educating people, training people, investing in people. The other is postmarket, where you tax and transfer. Then there’s the direct within-the-market where you actually affect the quality of jobs.

This is the one we tend to do the least, and understandably, it’s a difficult place to work, but in some sense, I think it’s the most important. I think nobody likes taxation and transferring, neither the people who pay the taxes nor the people who receive the benefits. We have an interest in making sure that work works for most people. The reason is 60 percent of GDP goes to labor, and that’s our primary means of income distribution. If we didn’t have that system, if money came out of a hole in the ground, and then we all had to redistribute it, it would be a war of all against all.

Let’s say it came out of a hole in one person’s lawn, and then we had to tax it and redistribute it. It would be a terrible, terrible problem. Work as a method of income distribution, as a means of structuring society, of promoting social goods, things that are useful and productive and give people esteem and meaning and structure and all kinds of things—it has so many benefits, I don’t know of a good alternative to it.

When labor markets work well, a lot of other social problems decline a bit. Crime tends to decline. Families tend to work a bit better. I think we should take a moment to actually be happy about the situation we’re in, in the sense that, I think many of us thought coming out of the pandemic that we would be in a state of depressed labor markets, where all the gains that we were making were going to be undone.

Instead, we’re in a period where inequality, by many measures, is coming down, and more importantly, the quality of work for people who are low-paid is improving. Some of this we should credit to our effective policy—not perfect policy, but effective policy for helping us guide ourselves out of the pandemic. We often talk about how government can’t do anything, but in fact, it turns out government could just spend 10 percent of GDP to assist households, to assist businesses, to assist the unemployed. That was pretty remarkable.

We’re in a much better position now, three years later, than we thought we would be in. The notion that inequality just must forever rise, as many people have begun to think, has been disproved. Then we all of a sudden have a kind of vista of technological possibilities that we hadn’t seen. We should be less fatalistic in saying, well, the future is whatever we can predict and more appreciative of, actually, how much agency we have.

I think the notion of the future as a prediction exercise neglects the fact that the future is a creation exercise. It’s something that we’re building, and so recognizing our agency in this—and we’ve just demonstrated our agency in this, in so many ways—we ought to take an active role in making these highly consequential choices about how labor markets will work, how we’ll use these tools to invest in skills, and how we’ll take this bounty of possibilities and turn it into a set of outcomes that we want to kind of inhabit and we want our children to inherit.

Bethany: David, throughout history, pandemics have reshaped labor markets.

Is what we see happening now a result of the pandemic, or would it have happened anyway?

David Autor: It was happening, but it wouldn’t have happened like this.

We were in a period of tightening labor markets. That reflected a number of things. An important thing it reflects is demographics. We have a rapid increase in retirements, and we have very few young people entering the labor market. Fertility is low, small cohorts, and then vastly restricted immigration. Then, there really is a discontinuity with the pandemic, where all of a sudden, things got much tighter, much faster. Inequality really compressed. We saw some pretty sharp changes, especially in the way low-paid labor markets among young workers, high-school grads, are working, and I think many of us were extremely surprised by this. I know I was surprised. I’m on record as being surprised because I predicted exactly the opposite in a paper I wrote early in the pandemic called, “The Post-Pandemic Labor Market: Too Few Low-Wage Jobs.” I felt entitled to call my paper “The Unexpected Compression” because clearly it was unexpected for me.

The tightness of the current labor market is good. It is making bad jobs or low-paid jobs less low-paid, but it’s not transformative. People are earning a little bit more; they’re getting somewhat better benefits. Employers are doing a little more to accommodate them, but this is not going to rocket people into the middle class. We need to do more if we want that to happen.

Luigi: As the only immigrant in the group, let me raise a difficult question because from your descriptions, you are completely ignoring the variable of immigration. I’m not a labor economist, but I don’t think you need to be a labor economist to know that if you increase supply, prices tend to go down. We know that the United States has been very open to immigration, especially low-skilled immigration, and prices on low-skilled immigration have been very low.

Surprise, surprise, around the time of the pandemic, where number one, a lot of illegal immigrants were forced to go back because they couldn’t survive in New York because they had no support, and after that they were not allowed to come back, we see a massive increase in wages at the low end of these distribution. Everything fits one picture, which is immigration depresses wages at the low end, and if you want to ensure a more equitable distribution, you have to restrict immigration.

Why am I wrong?

David Autor: I did actually mention immigration as one of the sources of the scarcity, the restrictions in immigration.

Now, this is a contested topic in labor economics because I think any trained economist shares your prior, that of course, you restrict supply, price goes up. The evidence on this has been surprisingly hard to produce. Decades of research by David Card and others has primarily found that immigrants are more complementary than competing with many native workers.

In my own opinion, many people, even people who believe the things I just said, will say exactly what you said—in fact, what I said a minute ago, which is restrictions on immigration are likely contributing to labor-market tightness. I think our immigration policies are inhumane and a mistake for the country, but that doesn’t mean I think that if we have unrestricted immigration, we should allow wages to fall to any level they might fall to.

I still support things that affect the quality of work for those who do it, including minimum wages, because they don’t appear to be very adverse for employment. It’s a very imperfect instrument, and it’s a very incomplete instrument. It’s easy politically because it appears to be free, like rent control.

Why don’t we pass a law that makes prices better or wages better? It’s always popular nominally. You don’t have to raise your taxes to do it. It turns out the minimum wage, despite not having a strong economic justification in the kind of blackboard version of the world, at least in many high-income countries, appears to work better than you’d expect. It appears to reduce inequality, to raise wages at the bottom, to reduce poverty a bit, without having substantial adverse effects on unemployment.

I think it’s inconsistent to say immigrants on the one hand have no effect on the market, and on the other hand, restricted immigration supply is contributing to labor-market tightness. I think that would be inconsistent. It is a contributor. I think we should have more immigrants, but I think labor-market tightness is not our only tool for affecting the quality of jobs. We shouldn’t have to actually go into an inflationary spiral to improve wages for low-paid workers.

Luigi: Wait, why do you say we should have more immigrants? We should have more immigrants at the upper end of the income distribution, but not at the lower end, because that clearly affects the lower-middle class.

David Autor: I think we should allow many more immigrants for two reasons.

One, it’s incredibly welfare-improving for them. There’s no question that the ability to migrate to a high-income country massively improves people’s standards of living. But again, I think that needs to be done in conjunction with things that affect the quality of work.

Many European countries have higher immigration rates than we do in the United States, and yet they don’t have the same deterioration of labor standards for in-person service work, and that’s because they make a choice about that.

Luigi: I don’t want to come across as this big anti-immigration guy.

But first of all, look at Japan. Japan had a very strong policy of anti-immigration, and they are increasing the quality of the jobs at the low end by a lot. Why? Because they use a lot more machines.

If in the United States, you wash cars by hand, it’s because it’s cheaper to do it by hand. A lot of jobs are done by hand because labor is so cheap. The moment you actually put on pressure and create a sustained wage at the bottom, people find new ways to actually save jobs and make jobs more productive and most of the time, also better and more pleasant. Washing a car by hand ain’t particularly pleasant to me.

I think restricting entry is not so crazy, especially if you associate it with the fact of whose welfare you’re maximizing. I think we are in a democracy, and in a democracy, you should maximize the welfare of the people inside the country. That’s what democracy is about.

You are saying that, oh, this is great because you maximize the people who are going to come here. But that’s kind of very, if you want, enlightened, but—

David Autor: Sure, it’s easy for me to say. It just makes my life easier, more immigrants, cheaper services.

I don’t think Japan is a comparison that works in the favor of this argument. Japan has a desperate labor shortage. Their population is shrinking. They don’t have enough workers to do what they need to do, and even elders are being undercared-for. They are trying very hard and better than other countries to use technology to compensate.

People have a right to disagree. I can understand we’re not here to vote on the welfare of the world. Citizens have every right to make decisions about who comes into their country, but I do think there are benefits to it economically, and it would address some of our problems about having a larger population to help care for the elderly, to help pay taxes.

A shrinking population is a very difficult problem for an industrialized country. It’d be one thing if we just culled 10 percent of the population of all ages, but that’s not what happens. When a population shrinks, it gets older, and that creates a very difficult imbalance, and that’s what Japan is going through.

Very few industrialized countries at this point have replacement-rate fertility. Certainly, Italy does not. All of Asia does not. The only countries in Europe that do are the Scandinavian countries, arguably because they have such favorable parental and childcare and educational situations that people don’t find having children as costly or intimidating.

Reasonable people can disagree about this, but I think at this point, our restrictions are too onerous, and I don’t think we should believe that our only tool for improving the quality of low-paid jobs is labor-supply restrictions. We could also improve jobs somewhat through minimum wages, through the earned-income tax credit, through labor standards. We have done this before, and we don’t have to . . . It’s not either-or.

Bethany: Given the huge cost to our society of having an older workforce and not having a replacement rate, isn’t it incredibly shortsighted of us not to enact more policies to encourage people to have children, through better help from others and all the things that we resist doing? Given that the long-term cost of that is so immensely high, per your point?

David Autor: It is a puzzle that we don’t fully understand about why, when countries get wealthier, fertility falls so much. I have my own theories, which is that children are an inferior good. And I have three kids of my own, and they’re inferior. No, I have three kids of my own, and they’re wonderful.

It is the case that adulthood has never been as good as it is. There are so many fun and interesting things to do, so much travel, so many experiences, and you have good health for decades and decades. The opportunity cost of having children and encumbering yourself in that way, I think, rises. But pronatalist policies have a pretty mixed track record, ironically. Many, many countries that have tried to do this, they’ve turned out to be expensive, they’ve had limited success and they often—and this is a very politically incorrect thing to say—but they often encourage people to have children who were not the people they were intending.

We could do more, and I think it would be somewhat better. I think if we made higher education less expensive, if we made childcare less expensive, if we made work policies more family-friendly, we would probably increase fertility a bit. And that might very well be worth it, but I don’t think it’s going to actually be sufficient.

I actually think immigration . . . If we want to even maintain our current population over a longer period of time, immigration is going to be required. I think if we just open the doors and don’t worry about the labor-market consequences, some of them will be adverse.

Luigi: A lot of people, including your colleague Daron Acemoglu, talk about the future of technological unemployment. And in fact, they advocate some form of universal basic income to make up for that. You seem to be going in the opposite direction.

What is the source of this disagreement?

David Autor: He and I would probably agree on the following. The issue really isn’t so much the quantity of jobs but the quality of jobs. I can’t speak for Daron here, so I’ll just speak for myself. I don’t envision a future in which we’re running out of work to do. All the time we’ve been talking about the robot apocalypse and the AI apocalypse and so on, labor markets have been getting tighter and tighter, and many European countries are seeing the lowest unemployment rate they’ve seen in decades.

There really is an issue of what type of work people are doing, and the quality of jobs in the United States has deteriorated over the last four decades for people without a college degree. You can think of middle-skill workers, people with high-school degrees or some college, some college graduates, doing kind of expert work in factories, in production lines, in offices, in administrative support, in clerical, in sales, and a lot of that work has been automated.

Those workers, it’s not that they’re necessarily unemployed, but many of them have moved into personal services, food service, cleaning, security, entertainment, recreation. Nothing problematic about that work, but it pays poorly. The reason it pays poorly is it doesn’t require expertise.

Think of a crossing guard and an air-traffic controller, two people whose job it is to basically stop collisions between vehicles and people or vehicles and vehicles occurring. Air traffic controllers make about five times what crossing guards do. Crossing guards are paid very little. They’re doing the same work, but one requires specialized knowledge that’s scarce.

Technology, by the way, has made that knowledge valuable. An air-traffic controller without a GPS and a radar and a two-way radio is basically a person in a field staring at the sky. Often, expertise and technology go hand in hand, but not always. Technology can also eliminate the need for expertise, and that’s what happened for many of these middle-skill jobs.

What I’m concerned about looking forward is not a jobless future, but a future where a lot of expertise is potentially devalued, and people are doing low-value jobs, and we can have lots and lots of productivity growth coming from AI and other technologies, but if it doesn’t accumulate to labor through the scarcity value of labor, then we have a real distribution problem on our hands, and I think that’s something to be very concerned about.

We’re not good at handling redistribution through our political systems. If labor suddenly were not scarce because all of these activities could be done by machines, we would have fabulous wealth, but most people wouldn’t have a claim on it. Then we would need another system for distributing that wealth. It frightens me to think what that system would look like.

Bethany: Even the way you laid it out just now, it almost seems inevitable to me. Certainly, jobs like home healthcare and all sorts of service jobs, the need for those jobs isn’t going to go away. And yet, to your point, without some kind of intervention, nor are they going to pay more.

Or maybe a different way of asking this, why are these sorts of utopian trends that we’re seeing right now in the wake of the pandemic, where wages in this segment of the population are rising, how is there a way that that could continue?

David Autor: One thing is low-wage jobs in the United States are worse than low-wage jobs in other high-income countries. That’s a policy choice. That’s not intrinsic.

Every rich country has people working at McDonald’s. The people in Norway earn twice what McDonald’s workers do in America. They’re flipping the same burgers. Now you might say, OK, Scandinavia, it’s paradise, not relevant. But even if you look to our neighbor to the north, the people at Tim Horton’s are making 30 percent more an hour flipping burgers in Canada. One thing is we’ve made choices about the quality of work and what distributions can look like. We could make better choices.

A second thing is most other countries also do a lot more to fill out that middle of work by creating skilled apprenticeship systems, skilled vocational training systems. We’re going to need lots of people to do plumbing, electrical work, solar installations. There’s going to be a lot of physical work and a lot of hands-on work in medicine as well. I do not think it’s all going to be automated away, but those things, we actually lack good pathways into them.

The United States education model is built around college or bust. We don’t invest sufficiently in skilled vocational work, and there’s no future where people without education beyond high school, without post-secondary education, have great career prospects. Sometimes, if you’re going to college, we have an eight-lane highway that goes from your high school to a college. If you’re not going to college, then it’s just over the mountains and far away, you may find the way into the labor market.

I do think there’s a technological angle to this as well about how we use the technology, and there, I think, there’s a lot of room for making good choices. I’m one of many, many people who thinks that AI really is a big deal, but how we use it . . . It’s incredibly malleable. You could apply it to make people who are doing care work improve the set of care tasks that they could do, not just empathy but also medicinal tasks. Reduce the physical demands of that work; it’s actually dangerous, a lot of people get injured lifting people, bathing people. You could use it to enable people with a set of some expertise to do a broader set of activities.

I’ll give a concrete example. This is not as much technology as other institutions. The job of nurse practitioner . . . A nurse practitioner is a registered nurse with an additional master’s degree, and a nurse practitioner works alongside a doctor and can examine patients, can diagnose, and can prescribe under the doctor’s pen.

NPs were basically not present 20 years ago. Now there are about 400,000 of them. They make about $150,000 a year at the median, which is a good income. The employment of NPs is expected to grow 40 percent in the next 10 years, which is much faster than the rest of the labor force. The main thing is this is a person who has less . . . They didn’t spend three years in medical school after college plus seven years in residency. It’s a good example of taking expert work and allowing someone with less expertise but sufficient training and support to do more of it.

We’ve gotten really good at making expert people really expensive. Over the last 40 years, we’ve kind of hollowed out the middle of the income distribution and the work distribution and then made people who were doctors and lawyers and engineers and computer scientists, we’ve made them more and more scarce, effectively. Their earnings have gone up enormously.

It would be nice to have a world where, actually, some of that expertise was less expensive, but then people who were skilled in some foundational way could do a broader set of activities. You could imagine a future in which you would go for a four-year degree in medicine and then with this four-year degree in medicine and a lot of technology tools, you could do a lot of medical-care tasks, and the technology would be a support to that. It’s possible that it doesn’t always eliminate expertise. It actually broadens its applicability.

For example, I could go on YouTube this weekend and figure out how do I rewire and replumb my house, but if I don’t really know how to do those things, I’m going to drown myself or electrocute myself. If I actually know something, I know some basic skills, then having access to that information makes me much more effective. It’s not a substitute; it’s a complement.

The idea of using tools to make less expert people able to do more expert things would be really, if we can pull it off, a very powerful use of this technology. We’re going to get pretty much what we aim for. Technologies are not going to appear if we’re not trying to make them, and so it really matters where we put our money.

Bethany: Is your belief in the value of work automatically antithetical to UBI, or is there a way to make the two work hand in hand? Is there a way to make the two congruent?

David Autor: Yeah, it’s not antithetical to that. We should recognize there’s a bit of a continuum here. There’s lots of things that we provide in kind, not in terms of cash, that in a sense you could think of as being a universal basic income. Access to education, access to certain health services, access to even vaccination during the COVID pandemic. There are many ways in which we are making transfers to all citizens that they have a right to, regardless of whether they’re working or not.

UBI is another step in giving them cash on top of that. Done at a very large scale, it would be incredibly expensive. We would have to tax ourselves a lot to do it. That itself is challenging to pull off. Then the question I think one would want to ask is, well, if you’re willing to tax yourself that much, is that how you’d want to use the money, or would you want to use it for something else? I would like to see a lot more money spent in improving the quality of work. If you ask me, do I want to give a thousand dollars a month to everybody, or do I want to give $1,500 a month to boost wages in low-paid employment? I might go for the latter.

When I started studying economics and I had the kind of temerity to say this, where we studied utility functions, and there’s increasing utility in consumption and declining utility in work, I always said, oh, they have it right, but the sign is wrong on the work part because, actually, people benefit from work. It’s not just income. It gives them structure, it gives them identity, it gives them social status, it gives them a circle of friends, a community. Work is valuable in many, many ways.

Losing work, even if you’re “compensated” for it, is just not full compensation. I joke to my colleagues, I say, imagine you, Daron Acemoglu. You’re an eminent economist, and we’re going to end your job today, but don’t worry, MIT will send a check to your home address, so you’re good. Well, that’s insane. Of course, that’s not the entire value of Daron’s job to him. He takes incredible joy in what he does and pride in what he does, and that’s true for most people.

Most people do not want to receive a check as compensation for their crappy job. They want to have a job where they’re proud of what they do, and they feel good about being paid for it. My halfway house is, even if you think you know where you want to go and you want to integrate, don’t do it as shock therapy. We just have seen over and over again how badly that works out.

Luigi: You are very famous for your work on the so-called China shock: the devastating impact that the opening up to the WTO had on American manufacturing and on the families that live in the Midwest.

Last year, we had as a guest Glenn Hubbard, who was, as you probably remember, at the head of the Council of Economic Advisors for the president when the China shock took place. We asked him, if you knew then what you know now, what would you have done differently? Let me ask the same question to you.

David Autor: Are you going to tell me what Glenn said?

Luigi: I will later. I don’t want to bias the answer.

David Autor: Sorry. It’s unquestionably the case that China’s rapid growth and especially after it joined the WTO, the World Trade Organization, in 2001, produced a flood of low-priced goods with which American manufacturers could not compete. That reduced employment in US manufacturing by, depending on who you ask, between 1 and 2 million workers in a very short period of time.

You might say, well, that’s a small number, it’s a labor market of 150 million people. That would be true, except that those are very concentrated losses. It’s not a thousand people in every county, it’s the furniture industry in Tennessee or something. These were economic bombs that went off over certain communities, and people didn’t recover from that very quickly. Even today, a lot of those areas are just . . . They’re kind of blighted. It was not anticipated that it would be this traumatic. I don’t think the people, the architects of this policy, thought that this would be nearly as devastating to those places as it turned out to be.

If we were doing it differently, if we were starting over again and we know what we knew now, I would say there are two things I would do differently. The less radical is that I would put it in place a lot more adjustment assistance to allow people to retrain, to move around and so on. I think that’s pretty inadequate. Those things don’t tend to work out that well. I would have made it go much more slowly, and I think the rate of change matters enormously. It’s not just where you’re going but how fast you go there.

We all know that 25 years from now, there will not be the several million commercial drivers and long-haul truckers that we have in the United States. Autonomous vehicles will make those positions much rarer. That’s OK, because that’s a 25-year problem, and in 25 years, many people will retire from that activity, fewer people will enter, it’s not going to be overnight, and so, that’s a manageable problem.

If it were going to happen today, that would really be a problem if, come Monday morning, all those millions of people were out of work.

The bottom line, in addition to the standard package of social supports, which I think are necessary but not sufficient, I would have slowed it down. I don’t think it would make sense or be feasible or in any way good to prevent China’s development, but I would have slowed the rate of transition such that we could have made a much healthier movement of labor into other activities over a longer period of time.

Luigi: Thank you for your time, David.

David Autor: Well, that was a lot of fun. Thank you guys very much.

Bethany: What did you think was the most interesting takeaway, Luigi?

Luigi: I think I have a lot of reactions, so maybe I will have to do one at a time.

The first is we have interviewed a lot of economists. I think that he is probably the one that comes across as most real. He’s also trying to appear good, but I think he really cares about what he does This is where his credibility is much bigger because he’s not just an ivory-tower economist. He is somebody who spent time with manufacturing workers and understands them directly. Most people want more than just a salary; they want a job in the sense of something to be proud of, something that gives them context, that makes them happy, then makes them get up in the morning.

I think that is not just the money. Money is important—I would be the last one to underplay that—but it’s not the only thing going on. I think that this is where he has the credibility to assert that in a way that other people don’t.

Bethany: It’s a part of the conversation that is often left out by people who think they have solutions to what’s taking place, and that’s, I think, a very human desire, a very human need, to be of use. I’ve always been astounded by how many thinkers are willing to disassociate from that and just assume that if you can give people money and enable them not to work, that somehow, we’ll be in this utopia.

I think I mentioned on a previous podcast that I have a whole conspiracy theory surrounding UBI. I think it’s being pushed by technology billionaires for a reason because in an attention economy, they’d love nothing more than to have people who have nothing to do but sit at home and be on their devices and be funneling their attention into the technology economy. I think it’s very Matrix-like. You could write a whole sci-fi book about this.

Luigi: I would actually like to buy into your conspiracy theory. It probably might be right.

As an economist, I think that it’s funny because now, conspiracy theory has this negative connotation. But as an economist, we are trained to think about incentives, even when it comes to adoption of ideas or pushing of ideas. To me, what you are describing is not a conspiracy theory. It is sort of pure rationality, but that’s a discussion for another day.

What I want to stress is the following fact. You remember, then, when we had Glenn Hubbard, I asked him twice what he would have done had he known at the time what the China shock would have created. The first time, he punted the question. The second time, he said a little bit of retraining but nothing much. I do appreciate the fact that David, the first time around, said very openly, we should have slowed down the process.

As economists, we don’t pay enough attention to the adjustment costs, and the adjustment costs are basically our lives. The Industrial Revolution was great for humankind ex post, but it left a lot of people poorer and created a lot of disruption along the way. If we can try to be more cognizant of that, I think it’s very important. He was very open and honest about that, even willing to go against the natural tendency of economists that free trade is good, and so, the sooner you get something good, the better it is, and so on and so forth. He was willing to sort of buck the trend there.

Bethany: The former math major in me very much likes the idea that the rate of change is as important as the change itself. I did wonder about that a little bit because being able to adapt, if you slow down the pace of change, seems very appealing and logical on the surface, that if we move more slowly, then you can have more time to figure it out. But that presupposes a government and a society that has an interest in figuring it out.

I worried that while it was an incredibly appealing answer, it was a little bit Pollyanna-ish in the end, that if you slow it down, we’ll figure it out, but will we? Unfortunately, there’s not much in the last 20 years of American life . . . Maybe I’m being too cynical, but that gives me the comfort that we would be able to figure it out, even given time.

Luigi: No, no. The way I interpret it is not so much that we would have figured out and done something differently. It is that it is easier for people to adapt if you do it slowly. Then, on top of that, you could do a lot of initiatives. First of all, we don’t know exactly what initiatives besides retraining, but the reality is, in my view, if you are a 40- or 50-year-old worker in a manufacturing plant, and the plant closes, you’re not going to reinvent yourself as a nurse or as a web designer the next day, probably not even the next decade. Because you don’t have a lot of flexibility, and you are toward the end of your working life, really, the opportunities are very limited.

What you want is to give time for those guys to retire and make sure that not a lot of young people go into the same profession, so that you have a transition, or at least the people that go into the same profession know what they’re getting into. They invest more in education to be more flexible down the line.

But all this requires time because we don’t fully appreciate how long it takes for ideas and skills to get embedded in the population. You need to train the teachers, and that takes a while, and then the teacher needs to train the next generation, and the next generation takes a little while to reach the labor market, so this stuff is incredibly slow.

Bethany: What’s interesting is that, to me, my brain went in a different direction, and I immediately thought what the government, what policymakers, should do, given time, in order to ease the transition. Whereas I think he meant, and you are speaking to, a more organic form of transition, that if it moves more slowly, people will be able to figure it out for themselves. That’s interesting.

Luigi: No, no, but you are right that this gives more time for the government to do anything.

I’m not denying that it’s an important additional benefit. What I’m saying is that even if you think that additional benefit is going to be wasted, because these days we’re not that optimistic about the government’s ability to do stuff . . .

In spite of what David said, because David was actually very optimistic, he said in COVID, we did a lot of things right. I think that there is room for public policy, and sometimes we get it right, but even if public policy were not to work, I think that slowing it down a bit would be better. Now, the issue—and he was very honest about this—is once you start slowing, there are a lot of forces that say, why don’t you block it? Slowing but not blocking is really the difficult political-economy question.

Bethany: Perhaps a way to summarize his point of view on this is, move slowly and don’t break things, which is very anti- our current, modern, Silicon Valley ethos of move fast and smash everything.

Luigi: Absolutely, but also move. He’s very clear that we don’t want to stop because some people say, we want to stop change, it’s too fast, we want to stop it.

No, no, you want to slow down. I wouldn’t say move slow, but it’s slow down. Honestly, I think that the change—maybe this reflects that I’m getting old, and when you’re older, you see the world in sort of . . . Oh, in the old days, it was different. Historically, change has taken place less rapidly than it is today.

Actually, one favorite picture of mine is how long it took for different innovations to spread in the world. You start with electricity, that in order to get the first 50 million people to adopt, it took like 50 years. The phone, similar. Then you go to the internet, and now you’re down to 10 years, and then you go to Pokémon Go, and in 10 days, 50 million people were using Pokémon Go. I think that that’s an indication that, really, change does move. Now, Pokémon Go, of course, is not that important, but new apps, they go so fast, they spread so fast, in a way that was unprecedented before.

Bethany: Do you feel optimistic after our conversation with him that this dynamic of narrowing inequality that we saw is sustainable? Or do you question that? Do you think it might be a blip? I lean toward the latter, but I’d like to hear something optimistic on this gorgeous, sunny Thursday.

Luigi: Actually, let me take a middle-of-the-road view here, because basically, he was really, really trying not to sort of endorse a view that immigration is bad. “As an economist, we don’t have good evidence of this. Yes, in first principle, it should be there, but we don’t know.”

Versus on trade, where he said, “Oh, the reason is because people thought that trade was good and so they didn’t want to say trade was bad and they didn’t . . .” Wait a minute, aren’t you doing exactly the same thing about immigration?

I think that, honestly, this is really at the heart of the tension for the, if you want, upper-class Democrats or the Democratic elite. I’m sympathetic to his heart when he says, when somebody we bring to another country, we are giving him or her so much in value that it is literally winning the lottery. It says when they put up a lottery, the green card, you literally win the lottery. Wouldn’t you like to expand this gift to more people? The answer is yes, if you care about the world, that’s what you should do.

However, this immigration, in my view, does have an effect on income distribution, especially at the lower end. I don’t think it’s a coincidence that the periods where you see a lot of catching up at the low end of this income distribution appears where immigration is more restricted. So, between the ’30s and the ’80s, everybody celebrated the magic period after World War II, et cetera, et cetera, but there was much less immigration than there is today.

Surprise, surprise, all wages went up tremendously. If you have more people coming in, it is certainly more convenient for you, David and me, because we buy those services, and if they’re cheaper, it’s good for us. But from the point of view of sharing the wealth produced, it is not working very well, and also from the point of view of feeling we are part of the same group. How can you say we belong to the same country? Then, every time there is an opportunity, you take somebody from abroad, and you have them compete at the lower end of the income distribution, but not so much at the upper end. For the doctors, there is not such a thing.

Bethany: Yeah, I think that’s a really important point, and I think I can maybe explain the cognitive dissonance, which is that trade is people, but the effect on people’s lives is second- or third-order, whereas immigration, the effect on people’s lives is first-order. If you are a caring person, it’s hard to see them both through the same lens.

This does seem to be a bit of an untouchable thing in the economics profession because, remember, our previous guest who wouldn’t really answer that question argued that, over time, immigration was good, that it took longer for immigrants to achieve, but they didn’t take jobs from Americans. It will be interesting to see how the Democratic-elite part of the economics profession wrestles with the data from the pandemic.

Even immigration aside, I don’t know, maybe I’m too cynical. I’m just worried that it was this blip. That now, as we head into a much tougher economy as a result of the interest-rate increases needed to blunt inflation, as well as the crisis in the banking sector, my attention is just gravitating toward the news of the layoffs. That seems to be a pretty constant flood, and for a while the belief was it was just in the technology industry, and then you’ve had Disney and so many other companies since then. Just when everybody is starting to embrace this . . . One of the themes that seems to me very clearly to emerge from the pandemic is how often the experts are wrong. I worry that just when we’re starting to embrace this idea that inequality is narrowing, that it’s going to flip around on us again.

Luigi: Yeah, but I think that if you have some significant reduction in immigration longer term, I think that that will change a structural balance and will make it more equal in terms of the income distribution. That’s my prediction.

Then, I think, it’s a political decision. If we were to take my preferred immigration policy, that is, for 10 years, we only import college graduates, no people with no college degree, I think that the income distribution in this country would change dramatically, and we would benefit tremendously from technology, et cetera. We could have the cake and eat it too.

Now I understand that we’re not helping people in Guatemala who are trying to run away from an evil government—which, by the way, we helped create over time—but that’s a different story. I think that if you care about the income distribution, I think that that’s the easy solution. Now, whether this will be implemented, I think depends very much on the politics, but I don’t see it as out of the question that a Trump 2 or a DeSantis presidency would do something like that.

Bethany: It’s a really interesting point, and a lot of people who want to both retain their belief that inequality is bad, and they would do anything to narrow inequality and help low-wage earners in America, and that immigration throughout the spectrum is good, are going to have to wrestle with those dueling beliefs.

The question is whether people will or whether the data will become one of those things that you just can’t say, even if it does bear out.