Capitalisn’t: The Real Cause of Wage Stagnation

Episode Transcript

Arin Dube: This type of monopsony power or wage-setting power is really built into the nature of this market, different from markets that may be relatively easy to switch to, like what brand of cereal, although even there, as we know, there's brand loyalty and other sources of differentiated products that give some degree of pricing power. Jobs are a lot more so because there's a lot more at stake, and we don't typically change jobs in the same way we change breakfast cereal, and as a result, these kind of mobility and other frictions are just likely to be a lot harder to just wish away.

Bethany McLean: We are delighted to have Arin Dube, professor of economics at the University of Massachusetts, Amherst, and author of the forthcoming book, The Wage Standard: What's Wrong in the Labor Market and How to Fix It.

Luigi Zingales: My impression is that in the book, you use the term monopsony to indicate a number of frictions in the labor market. To what extent do you think the monopsony is due to excessive concentration of the employer? To what extent is it due to other factors?

Arin: Generally, monopsony, the word, of course, quite literally would mean a single employer. It's actually historically going back to Greek of a single seller of fish, as it turns out, but it may be the case that in some situations, the number of employers is relatively small. In other words, the market is concentrated. Really interesting work by Ioana Marinescu and co-authors suggests that typically, most workers in the US are operating in effectively a labor market with three effective employers that they're choosing from. That's a very small number if you think about.

At the same time, as I argue in the book, that it's probably not the dominant way of how monopsony power arises in the labor market. Especially not in thick urban labor markets. There are two key reasons besides concentration why we see a lot of monopsony power, which just simply means the ability of employers to influence wages and pay whatever wages they're paying without losing all their workforce. That is, first and foremost, there's some amount of search friction.

Quitting a job, searching for a job, and moving to another job are all costly, and that can really throw some sand in the wheels of the functioning of the labor market. Then there's the fact that the same job, in fact, what looks identical on paper, could be valued differently by you and me. The simplest example, for example, is commute time. There are lots of other factors like that besides commute that can make the job somewhat different.

If I change my wage, maybe as an employer, I choose to pay $1 an hour less, I'm going to lose some people, maybe who valued them just at the margin, but then I'll keep some other people who actually valued the job a little more. This also gives an important amount of wage-setting power that's just built into modern labor markets.

Bethany: Do those caveats around monopsony help explain why salaries at the bottom but not at the top of the distribution have not increased much?

Arin: I think it does in a couple different ways. First, what we see is, if you went back to 1980, if you got a job at a large firm, you pretty much felt like you got lucky because you got a better-paying job. That actually stopped being the case over the '80s and '90s. By 2000s, for those in the middle and the bottom, it just wasn't the case anymore. One of the factors behind that is firms ended up changing some of their wage policies when they didn't actually have to pay a higher wage, perhaps because of other pressure or social norms or, for example, collective bargaining in certain sectors.

They reacted by, as much as possible, lowering wages for those at the bottom. We've seen some really interesting work that show that in the '80s and '90s, for instance, when a company switched its CEO from being someone typically hired internally who was promoted to a CEO versus someone who was hired with an MBA from the business school, just that one decision led to very substantial reductions in wages, but particularly for blue-collar workers.

Luigi: Sorry, I'm a little bit confused because I think that if there is excessive concentration, or if there is variety in jobs, this should apply almost as much at the top as at the bottom of the [unintelligible 00:05:21] distribution. I think that the University of Chicago is certainly a monopsony in Hyde Park, as far as I'm concerned. Even if there is another university in town, which is Northwestern, it's not exactly identical. Maybe for commuting reasons is one, I can say other reasons.

You see that there is a lot of heterogeneity. Why, most of the effect, or maybe not most of the effect, maybe everybody's underpaid? The result that people are paid way below their marginal productivity is a result that applies only at the bottom of the distribution, or is it everywhere?

Arin: No, that's a great question. I think generally speaking, there's a good amount of evidence that there's monopsony power throughout the labor market. However, it is also the case that we see that the extent of monopsony power is surprisingly strong at the bottom at the labor market. These markets are particularly ones where there's a lot of wage markdown and wage difference between higher and lower-paying firms. That may be surprising. Why is that the case? I think it's a broader, more macro story.

What is that story? That story is that we actually, interestingly, spend a lot more time with relatively high unemployment rates generally. After 1980, we spent about two-thirds of the time below a fairly commonly used metric of full employment, which is constructed by the Congressional Budget Office. If you went back to 1950 to 1980 period, it was about a one-third of the time. We used to actually have relatively tighter labor markets. Why does this matter? This matters because a tighter labor market is particularly tight for low-end workers.

As a result, if you spend more time with relatively slack labor markets, that's a market where employers are going to actually have greater amount of wage-setting power. The really most compelling example of this really arose after the pandemic because it actually went the other direction. After the reopening of the pandemic, we had an extremely tight labor market, which was particularly so for workers in the bottom third of the wage distribution.

As my co-authors David Autor and Annie McGrew document, and I talk about this in the book, this period led to a very sharp and dramatic increase in wages at the bottom of the labor market, driven almost entirely by increasing quits amongst this workforce, as workers moved from lower-paying to somewhat better-paying job, pushing up wages by increasing competitive pressure.

Bethany: There's this long tradition in economics of criticizing any form of price control, from rent control to minimum wages. The logic being that the rationing effect is so large as to offset any potential benefit to people benefiting from lower rents or higher wages. In your book, you summarize the long intellectual journey to disprove this idea empirically. Can you share some highlights from your own journey?

Arin: I did my PhD at the University of Chicago. When I was there, the idea that the labor market was relatively best understood as a competitive market was pretty prevalent in economics generally. There were definitely a lot of work that was suggesting that maybe wage-setting was more complicated. That included, of course, the very famous or infamous, depending on one's perspective, work by Card and Krueger on minimum wages comparing New Jersey and Pennsylvania.

Just to be clear for listeners, what they found was that New Jersey, which raised its minimum wage by between 20% and 25%, actually did not see job loss among fast food workers as compared to neighboring Pennsylvania. One of the things I talk about in the paper is that we can actually do more of a meta-analysis in the sense of just summarizing the evidence from a bunch of studies. Nearly 60 studies report what happens in the US when the minimum wage rises to wages and employment.

Generally, what you find is this. On average, if you take the typical study, it suggests that if the minimum wage pushed up wages by 10%, maybe there would be a 1% reduction in employment from that. That means that roughly 90 cents of the dollar increase in earnings that would happen without any job losses actually occur, and that's the typical study. Most studies find either positive, close to zero, or very small negative effects.

That is very reassuring that at least for the type of minimum wages that have been studied, and that's an important caveat, that we have seen wages rise without costing jobs a lot, and one of the reasons for that is we have seen sharp reductions in turnover. This is now exactly what the theory of a monopsonistic labor market predicts, that when you pay a higher wage, what happens? Firms find it easier to fill vacancies, and they actually have to worry less about more workers leaving.

Separations or quits fall, and vacancies are filled fast. You end up to a point killing vacancies more than killing jobs. It's also the case that when the minimum wage rises, some of the labor cost is passed on as higher output prices. Now, importantly, that doesn't seem to reduce the product demand for burgers and retail goods and services very much, and as a result, that doesn't harm low-wage workers. It does mean that essentially middle and higher-income consumers are bearing some of the costs of paying a higher wage and that is one of the ways in which we actually absorb a minimum wage increase.

I think what we've learned is to a point, you can actually make the labor market work better for most middle and low-wage workers without causing a lot of harm.

Bethany: Given that it seems that overall raising the minimum wage doesn't lead to this dramatic loss in employment that conventional economists might have once said it did, given that we're all supposedly concerned about inequality, why has Congress not updated the minimum wage since 2009, and why did 20 states not pass a state minimum wage above the federal one? In other words, why are we not changing policy if the academic evidence seems to be pretty clear overall at this point?

Arin: That's a great question. Of course, the main reason they're doing that is in order to run a natural experiment for a generation where half the country has a minimum wage, and the other don't, because otherwise, how can we really credibly estimate the long-run effects of the policy? I am kidding.

Bethany: It's all for the benefit of economists.

Arin: [laughs] Must be because I can't think of any other positive reason to set policy this way. Look, the minimum wage is a very popular policy. It's popular among voters who are Republican, who are Democrats, who are Independents. Anytime it actually goes up as a ballot initiative, almost all the time it actually has passed, and that includes in red and purple states. It's just the case, however, that the Republican Party leadership, including in state legislatures, are quite reluctant to increase it, in part because I think it reflects interests and thoughts of relatively narrow set of businesses.

Actually, at this point, even larger businesses, especially, and even some smaller businesses, are open to a higher minimum wage, but the restaurant lobby, for instance, is very strongly opposed to it. There are different possibilities, but I think it's largely a question of political economy, not about voter preferences. Now, that being said, there is a question of exactly how high can you go. Obviously, just because the minimum wage to date, I think, has been shown to be relatively effective, doesn't mean that it would continue to be the case.

That requires, I think, taking a dispassionate look at the evidence, and at some point, especially as we explore more ambitious minimum wage standards, we will learn that maybe this went too far, and this is probably a good place to stop.

Luigi: I certainly buy that if there is a monopsony power in the marketplace, a minimum wage could be the right policy to follow, but I'm still struggling about the causes of this monopsony power. I understand that if I am Walmart and I see that I can make a little bit more money by hiring an extra employee, but in order to attract the extra employee, I need to push up the wage of that extra employee and probably the wage of everybody else in Walmart, that makes it not worthwhile to do it.

Arin: That's right.

Luigi: That's one version of the monopsony. Imagine I am a baker, and I'm alone, and I bake bread and cookies, and I really desperately need some help. I can quickly calculate that if I add another person, that person will more than pay for himself or herself. In that particular case, I should be hiring an extra person because I don't have any employee whose wage will rise, and I have no reason not to hire that person. It must be that the highly productive jobs are in limited quantity, and the jobs that are in plenty of supply are much less productive, because otherwise, everybody will start bakeries and hire people and beat up the wages.

Arin: I think this is the key thing that you highlighted is the role of this differences in productivities across firms because that's a key prediction of the model that I think actually is born out in the data in a couple of contexts, that in a market where you have a degree of monopsony power, you actually have workers in relatively low-productivity firms working. This is actually one of the really interesting things that happened when either there's a minimum wage increase, for instance, you see workers reallocated from the low-productivity firms to higher-productivity firms, which is one reason why overall employment effect is quite muted, even though if you are a low-productivity firm, you very well may see some job losses.

It's just the case that those job losses aren't just losses to the market. They're actually workers who end up taking jobs somewhat somewhere else. That's a very important mechanism of the reallocation aspect of what the minimum wage is doing. Interestingly, it's exactly also what happens when you have a tight labor market. We saw this in 2021, '22, '23. We have this huge reallocation from those bottom-paying employers when the market's tightening. That's one of the important mechanisms through which you are increasing pay.

What's also fascinating is that it turns out one reason why so many workers are at very low-wage firms, and there are many reasons for that, by the way, but one reason is that sometimes workers don't actually know how much better opportunities there may actually be out there. As it turns out, this is particularly true for workers who are in low-wage firms. As opposed to high-wage firms, workers know that they're in a high-wage firm. If you're in a low-wage firm, sometimes you actually don't.

Then, if there's some shock that, actually, suddenly you see other people switching jobs as you did in 2022, '23, that might actually accelerate and increase some degree of competition by this reallocation process. More generally, the idea that there are differences in productivities across firms is quite important. One of the implications of spending a lot of time in relatively slack labor market is that it allows for firms that frankly would only survive with a low-wage strategy to be able to actually recruit and retain workers.

If something happens, either a policy that increases the minimum wage, for instance, or that increases generally the tightness of the labor market, we will see a job ladder process where workers end up leaving those jobs and joining to higher-productivity firms. In a very important sense, if you want to make the market work well, you want policies that I believe should really enable workers to be better matched with higher productivity employers.

Luigi: You are suggesting that there is a market failure in the sense that workers are not properly allocated at their highest value use. I can see even a way to make money because you are saying that if you can inform the people working for lower wages, that they can reallocate more profitably, so you can go to Walmart and say, "I advise you how to move and work for Target, and you're going to increase your wage, and I am going to advise you and ask a small fraction of your increased wage to pay for my service." There is a business to be made in reallocating workers. Why don't we see this happening?

Arin: Just to be clear, the question is, why do we not see workers getting reallocated to the right place by some intermediary who sees that, "well, gee, John here is working at a low-productivity firm. Why doesn't John move to this higher-productivity firm where they could potentially have a higher output?" Is that the question?

Luigi: Exactly. Yes.

Arin: That is exactly the right question. Fundamentally, it may be the case that initially John doesn't really want to change jobs, but it also may be, like I said, that maybe there are certain aspects of the jobs that John feels like, "Well, maybe I'm not exactly sure how I'll get along with my supervisor at this higher-productivity firm, et cetera." All of this is to say that this just adds more-- Again, throwing more sand in the wheel. Just like that high-productivity firm just offering a higher wage couldn't attract the worker, any intermediary, it's going to be hard for them to magically make these frictions go away.

Fundamentally, I think what that gets to is this type of monopsony power or wage-setting power is really built into the nature of this market. Different from markets that may be relatively easy to switch to, like what brand of cereal, although even there, as we know, there's brand loyalty and other sources of differentiated products that give some degree of pricing power. Jobs are a lot more so because there's a lot more at stake, and we don't typically change jobs in the same way we change breakfast cereal.

As a result, these kind of mobility and other frictions are just likely to be a lot harder to just wish away. This is important because, look, if we could reduce some of the frictions, then that may seem as the first best. I think that that's a great idea, and we should pursue policies to do that whenever possible. One of the things I talk about in the book is another source of monopsony power, which is in some ways the most obvious one, one that Adam Smith talked about, which was actually outright collusion or just other forces of artifice. There was a collusion practice, for example, by major tech firms in Silicon Valley who basically all agreed not to hire from each other.

There was this infamous case in around 2007, 2008, where there was an email, for instance, from Steve Jobs writing to Eric Schmidt in Google saying, "Hey, one of your guys was trying to recruit one of our engineers. I would really appreciate if you stopped this." Then Eric Schmidt looked into it and wrote back saying, "It's done. That was an error. The person is already being fired." Steve Jobs responded with just a smiley face. Anytime that happens, we should certainly try to address it.

Another example is just non-compete agreements, which 30% of workers in the US end up signing just in order to get a job, which basically prevents them from moving to another competitor. The idea is supposed to be that this is helping protect trade secrets, but there are these cases. Lots of these jobs are hairdressers, or sandwich chains, or summer camps. Now, you don't exactly have to protect a trade secret to work at a summer camp as a camp counselor. Getting rid of these non-competes or certainly making them much less frequent would be a good idea.

The FTC under the Biden administration did propose such a policy, but then, it has been challenged in courts, and it's unclear. Generally speaking, any kind of monopsony by artifice, if we can fix it, we should. However, one of the arguments I make in the book is that even all said and done, after all of that, it's not likely to be the case that we'll get rid of the problem or even most of the problem. Historically, one of the things that we have done in this is basically having institutions that are countervailing.

Historically, the unions and the labor movement did play a role of that. At this point, private sector unionism is extremely low, barely about 5%, and it's very hard to imagine that changing in any sort of quantitatively meaningful way. We need other policy and institutional approaches to have that sort of countervailing power. In my book, I talk about the role of macro policy because if we actually had a tighter labor market, that would go part of the way. We have seen pressure campaigns have changed wage-setting behavior, including among large employers, Amazon and Walmart, but also banks in the US have instituted voluntary minimum-pay standards.

Those are helpful, but again, I think each of these can only go so far, which is why we need a broad set of tools and broad set of policies that push back and provide some countervailing force that offsets the impact of the monopsony power.

Bethany: I was thinking, as you were talking, of how crazy it was that we ever thought that labor markets were efficient because, of course, there are all these frictions, and I was thinking that your job is not like your brand of toothpaste, it's like your bank. It can commit an awful lot of sins before you're going to leave it. [chuckles] As you mentioned, macroeconomic policy, I'd love for you to expand on that a bit more. Is that actually the most important thing? Then what should the macroeconomic policy be in order to create these tight labor conditions? Is that the most important tool we have to push back on monopsony power?

Arin: I argue that you need all these prongs. You need a macro, you need a micro, and then also you need what I call meso, which is in the middle. The reason is because anytime you push too hard on one, you're more likely to run into unintended consequences. If we pursue relatively more full-employment policies, we are likely to see more broad-based wage growth. Can we just rely on that? No, because it comes with its own risks. Those risks include pushing too hard on those and then having inflation.

Similarly, if we think that we can just encourage employers to do the right thing, by increasing maybe public pressure, then we actually have some examples of that. As it turns out that these voluntary minimum wage policies are quite effective at raising pay at these major companies. Really interesting work by Laura Duranencourt and David Weil that I talk about in the book shows that it actually sharply raised wages in these companies. At the same time, they found that it didn't lead to big spillovers in wages in other employers in the market.

If you thought that, "Gee, if we took the biggest employers and really helped build a wage standard at these companies, that alone could spill over and solve the problems of a monopsonistic labor market, et cetera," then you would be somewhat disappointed. That's the limit of the micro strategy. It has a role, but it will only get you so far. As a result, I argue that it's important to think about wage floors more broadly. Here, we don't simply want to increase the minimum wage forever.

A minimum wage also has an important role, but at some point, the amount of compression you would need to raise the minimum wage closer and closer to the median wage becomes quite extreme. Instead, what about if we build wage standards and floors by different types of jobs more broadly, in the economy, if that sounds very radical? It turns out most high-income countries have some version of this, be it through collective bargaining or through policy. All of that is to say that I think we need a broad approach.

Luigi: Given the variety approach, I'm surprised that you don't even mention in the book another level, which is immigration. I find it particularly surprising because if you look at the correlation, the time series, like you had done for the Fed, the immigration is really correlated with low jobs. In a sense, if you look at before the 1980, less than half a million people were coming into this country. By 2010, more than 2 million people, legal and illegal, were coming every year into this country.

There was really an explosion of immigration since the 1980. Then we saw that during the first Trump administration, there was a cut down on immigration. Then there was the COVID that forced a lot of illegal immigrants to go back home, and then reduced, initially, at least, the entry of illegal immigrants. Then we saw an explosion of the wages at the bottom. Then we saw that toward the end of the Biden administration, we had a huge inflow of immigrants. Then we saw, all of a sudden, actually, inflation disappearing and wages not rising at the bottom, all of a sudden. At least at a very superficial level, there seemed to be quite a bit of a correlation between immigration and low wages. Why you didn't even touch this topic?

Arin: That's a good question. I think a large body of literature has looked into this, and my reading is that it finds a relatively mixed impact of immigration on wages. What I mean by mixed is relatively small. It would have a very hard time explaining, for instance, the magnitudes of the wage changes that we are actually talking about. Most directly, and this played an important role, I actually really took that idea very seriously. We tried to look at that in the more recent period when we actually saw this unexpected compression in wages occur.

It's very hard to actually correlate that even in time series, because if you look between 2016 and the present day, you have all kinds of huge increases and total shutdowns that happened. The path of the wage compression that happens in this period, really concentrated in these two, three years, that doesn't correlate well at all with time series of immigration. That is not dispositive. I think it's an important thing to understand and study better, but it gave me enough pause that I think we have lots of levers to think about and work with.

At the same time, obviously, we are seeing some very serious changes in immigration policies. What sort of impact this has remains to be seen. What I can tell you is that in the last year, for instance, when we have seen very sharp reductions in immigration, if you look at the wage growth, the wage growth at the bottom continues to be relatively similar to those at the top, but the extent of compression has certainly not increased over the last year, even as immigration grew a lot, which again goes to show that I would be guarded in drawing too much by any broad time series over the last 40 years when it comes to immigration and low-wage outcomes.

Bethany: There's a word you mentioned in the book, but you don't ever actually define what you mean by it. We haven't addressed it here, so I thought I'd ask you, and that's the concept of fairness. What is fair to you, and why does it matter?

Arin: That's a great question. Why have I not defined it is because it's very hard to define. Fundamentally, what I really mean by fairness is really nothing about what I personally think is fair. What I try to really do is think about how is fairness really approached and thought about by people generally, more broadly? One thing, for instance, is what is seen as, for example, a fair wage. Obviously, if you go ask people, people will have different opinions. However, it is the case that if you ask people, "Is a particular wage, for instance, fair or not?" people have opinions on that.

My broad perspective is that, as economists, we should respect the idea that there may be voter preferences over certain outcomes because they deem certain outcomes as fair. That, while not dispositive about something, we have to do something because it's fair or not, means that it's something we should spend time thinking about how to actually design policies to help achieve it, at least the fairness concerns playing one role in thinking about that design.

Luigi: One different interpretation, I think there are some elements that you bring to the table, is that in the old days, there was more internal fairness in firms, so that the janitors at Goldman Sachs were paid pretty well.

Arin: That's right.

Luigi: Then what happened is that the janitor's got outsourced, back to your initial interest, and so they can be paid much, much less than the past because now all of a sudden, they're not part of Goldman Sachs, but they're part of a janitorial firm. At some level, is it fair that a janitor doing the same job is paid more because he's cleaning up the floor at Goldman Sachs? Not clear to me. I think that this movement might have reallocated wages in a way that certainly is negative in terms of inequality, there's no doubt about this, but is it really unfair in that respect?

Arin: I think the issue of fairness as it affects firm wage-setting is very interesting. One of the things that there's a lot of evidence on, I've done some work on this myself, is how workers within a firm interpret wage differences and what is seen as a fair difference versus unfair difference. What's really, there's just a lot of work on this, clear is that if two people are doing similar work and they're paid differently, that is seen as unfair. Not maybe surprising, but that's the case.

Also, there's some degree to which if you are working in the same company, like you said, there's a limit on how differently you can pay people. At least that seems to be the case. As a result, companies have, a lot of times, outsourced workers, so they are getting perhaps lower wages, but they're not directly hired and working as subcontract employees. Is that fair? On the one hand, if you have within-janitorial wage differences, that could be deemed unfair, but then there's also within-firm differences that could be deemed unfair.

Here's what we know. When we moved from having more within-firm similarity in wages and blue-collar workers in relatively similar firms, overall, that meant both lower inequality and higher wages for blue-collar workers. That fissuring of the workplace where we have seen workers increasingly segregated in different types of companies, depending on if they're blue-collar or white-collar, has overall made it difficult for a bunch of workers to really have what in economics we call rent sharing, being able to actually gain from having some degree of higher productivity.

Now, it's true that rent sharing can increase dispersion of wages between janitors, but it also means the average janitorial wage is higher from that. That actually ends up mattering, empirically speaking, quite a bit more than the fact that within-janitorial wage differences may be somewhat higher.

Luigi: I thought that the richness of the analysis and the breadth was quite interesting. I found him fairly balanced in his summaries. It was not too partisan.

Bethany: What do you mean by balanced? Where would you have seen partisanship? I thought he was very data-driven as well, but I'm not sure I would know where partisanship would show up in this discussion.

Luigi: For example, the part where I thought he was a bit partisan is in excluding immigration completely from the discussion. His answer to my question in the interview was a pretty good one. He says, "It is true that the literature on immigration is more controversial. I fear that the data are terrible because one important part is illegal immigration. We don't have very good data on illegal immigration." He says, "During the pandemic, a lot of illegal immigrants in the big cities went home, because if you're illegal immigrants and you don't have work and you don't have Social Security, how can you survive?"

That's the reason why, immediately after the pandemic, I remember going to Boston. There was a Starbucks in Kendall Square, which is one of the richest places on earth, that was closing at 3:00 PM for lack of personnel. That was shocking. Then all of a sudden, the personnel reappeared. That coincided with a massive inflow of immigration. Something was going on. I don't think that the immigration is terrible for wages. I don't think there is evidence of that type. However, the vast majority of economists want to claim that there's no impact on wages whatsoever, which defies a bit gravity and says it is true that when you add labor force, you also add demand.

If you have more workers searching, it takes longer to search, and people are less likely to search. As a result, there is less that competition effect that drives up wages.

Bethany: Yes. The more you think about it, the more obvious it is that there are all these frictions that would impact the labor market. That, to me, I think, is the biggest transformation in the way I think before reading his book and talking to him than afterwards. Not just that I'm a lot more knowledgeable, but also that I think I would never think of the labor market in purely simplistic economic terms again. Does that make sense? I don't mean that economists are simplistic, by the way, [laughs] but just that it is so multifactorial.

Luigi: No, I think that you're right. Economists started with a simpler model. I think that that's what you do when you start. You start with a simpler model, and then you introduce some frictions. I think that, surprisingly, economists stay longer than was reasonable with this simplistic view. To what extent the persistence of this assumption is, to some extent, serving some vested interest? Clearly, under the assumption of perfect competition, any restriction that you impose is very negative. It's very easy to say you should put no restrictions. Under the alternative model, you can have a lot of benefits of introducing, for example, a minimum wage.

Bethany: Yes. It is in keeping with some of the other discussions we've had on this podcast recently about capitalism being created by the state. This book is, in a way, a furtherance of that argument because it shows how policy choices shape the labor market, and in so doing, shape capitalism itself. The idea that the state should just be hands-off or that policymakers should be hands-off because the whole model works in this state of perfect competition is just obviously not true, as it is not true more broadly.

I've been thinking about, out of all the jobs created last year, very few jobs were created, but 80,000 of them were in health care. If that's the direction of our labor market, then we have to start thinking about how we make health care a good job because otherwise, more and more jobs risk becoming not good jobs. These are policy choices. They're not just to the free hand of the market, be damned.

Luigi: Yes, absolutely. Also, because, especially in health care, the government plays a disproportionate role, because a lot of your compensation is driven by Medicare reimbursement, and what you are authorized to do and what you're not authorized to do. If you are a registered nurse, you can do certain things, but not certain others. You need a doctor to do certain others. This is clearly not set by the market, but it's set by regulation.

Bethany: Right, very clearly. Are there any of his-- Policy prescription is the word that's coming to mind, but they're not prescription. Maybe options. Any of the policy levers, how's that for a better word, that he discusses, do any of them resonate with you more than any others? Do you think that he is artificially minimizing the role of the macro-economic policy choices because that is so fraught?

Luigi: I don't know. I think that reading through the lines, I thought that there was an idea, but he doesn't push it, and he didn't respond consistently to my questions. My idea is that there's an externality in searching for a job because if I search for a job and you work with me, you're going to get a raise some way or another, because if I get a raise, eventually probably you get a raise as well. If I leave, probably the employer starts to panic that you'll leave as well, and so increase your wage.

I do all this effort in this situation, and you get part of the benefit. That's the ultimate definition on the externality. The fact that you need some level of inflation to get better wages is because inflation is a tax on not searching for a job, because unless your wages index-- If you just sit tight, your wage doesn't stay put. It goes down in real terms. The higher the inflation, the more I am designed to look for a job, and the more you do as well. We help each other pushing up our wages, because both of us decide to get up and look for a job.

Now, can we find some less costly way to achieve the same results? When I was going the direction, for example, of the mediator, it's not because necessarily I think that there's a lot of money to be made, but maybe the government can play some role in providing information. In today's world, we have so much more data, including what people are paid in other jobs. In 1980, it was reasonable that you had no idea what Walmart was paying versus Target, but today is much harder. Maybe we should make it even easier for people to get this information.

Bethany: Yes, maybe that's a pretty simple fix is just companies have to publish their salary data at all ranks. [chuckles] Then that would create some of that old-fashioned shame that, as you know, is my sticking point these days. That would also create transparency for workers to be able to know what they could and should be getting paid. Though, on that note, I did think he shied away from my question about fairness. I appreciate his answer that it's very, very difficult to define, but I did want to hear an answer from him as to what it actually was because I'm in search of that too. I guess there are a multiplicity of ways to think about it.

Luigi: Yes. To be fair, pun intended, I doubled down with the example of the janitor at Goldman because, honestly, what is fairness in that particular case is very dependent on the starting point.

Bethany: Yes. I thought that was a really fascinating question. Should the janitor at Goldman's salary be fair in the context of other workers at Goldman Sachs, or should the janitor's salary at Goldman be seen as fair based on salaries of janitors everywhere? If everywhere, is it everywhere in Manhattan? Is it everywhere as a national index? What's the everywhere? [chuckles] This question of fairness does get complicated really quickly.

Luigi: Yes. The other point that I wish he had discussed a little bit more is the geographical variation, because he's very much in favor of some form of negotiated wages. This works well in areas that are fairly homogeneous. The moment you have a lot of heterogeneity, then I think you do end up creating distortions that are pretty expensive in my view.

Bethany: Yes. Maybe fairness is best described as being like pornography. You know it when you see it. [chuckles] Right now, everybody knows that what they see isn't fair, and that in and of itself is a problem.