Peter Van Doren joins us this week for a discussion on how wages are determined in a market economy.
Is there a correlation between a worker’s productivity and the value they provide for society? Why has CEO pay increased so much lately? Should the government have a role in fixing unequal or unfair wages?
Show Notes and Further Reading
Van Doren mentions this blog post by Robert Lawrence on the gap between real wages and labor productivity. See also this link for the same discussion (only with Canadian data) on the terms of trade between what workers make and what they consume.
Here are papers by Kevin Murphy and Steven Kaplan on CEO pay.
Van Doren also mentions The Homevoter Hypothesis: How Home Values Influence Local Government Taxation, School Finance, and Land-Use Policies by William Fischel.
Aaron Ross Powell: Welcome to Free Thoughts from Libertarianism.org and the Cato Institute. I’m Aaron Powell.
Trevor Burrus: And I’m Trevor Burrus.
Aaron Ross Powell: Back on Free Thoughts today is our colleague Peter Van Doren. He’s a senior fellow at the Cato Institute and editor of Regulation magazine. Today, we’re talking about wages and workers. So start with a deceptively simple question, how are wages determined?
Peter Van Doren: In standard neoclassical economic thinking, the factors in a market economy are paid what’s called their marginal product, that is, assume a factory, assume a think tank, assume whatever kind of organization you want. And then imagine that you add as we say in class one more unit of labor or one more unit of capital. And how long would we continue to do that? And the answer in the class is you would continue adding factors as long as the output from adding one more unit of savings or one more worker to your workforce as long as the revenue produced by that augmentation exceeded the cost of those factors that you purchased in the market.
Trevor Burrus: So is this just like a price? I mean these are the same—
Peter Van Doren: In effect, as long as the output of adding factors to your organization, as long as the cost of those factors is less than the output they produce, you’re making money. And so how long would you continue to do this? And the answer is until the marginal output equals the marginal cost of the factors that you have purchased. So—and again in an introductory micro class, you would regurgitate back on the preliminary exam that factors and markets are paid their marginal product.
Trevor Burrus: So, that’s thinking about our business. We will employ workers and we will add a unit of labor up until it’s no longer worth it to us for them to work another day at a certain wage. But how does competition between employers and also potential employees then factor in what the wage is?
Peter Van Doren: So, I’m—make believe I’m farming, and that the history of the US economy in the 20th century was all of our grandparents especially mine, all of our great grandparents maybe for you, but my parents actually were farmers. So everyone in 1920—well, not every, but—
Trevor Burrus: Probably 80% of the workers.
Peter Van Doren: —75%, 80% of people in the US economy were farmers. And now it’s one point something percent. How did all that happen? And the answer is people were lured away—so the marginal product of a person on a farm given that farm prices were declining, and why were they declining? Well, the productivity, the marginal product of farms was increasing dramatically because of capital and technical innovation. So my grandfather’s farm in the ‘50s, one of his cows produced 10,000 pounds of milk per cow per year, and now dairy cows produce upwards of 30,000 pounds of milk per cow per year.
Aaron Ross Powell: Is that because of better equipment or better cows? Or both?
Peter Van Doren: Both. I mean genetic and literally the breeding of cows, they’ve been bred to become milk-producers and not much else. And then there are pictures of my father and grandfather in the 1950s using hand forks to shovel hay.
Trevor Burrus: Pitchforks, yeah.
Peter Van Doren: Pitchforks and there’s a picture of my father on a tractor without rubber tires, you know, the—well, you know, the steel tires.
Trevor Burrus: Steel tires? That seems like a really bad idea.
Peter Van Doren: Exactly. So, the productivity of labor was—labor was drawn away from farming into doing other things mainly working in cities, mainly in manufacturing or something like that before we switched to services. Because the marginal product of labor in that setting was much higher and thus their wages were higher and, in effect, farming had to compete and it couldn’t and so what it—it just lost people over time because the wage rate for people in farming basically kept going down and the wage rate for doing other things kept rising. So markets reallocated labor and capital away from farming and towards other things because, in effect, the value-added of labor and capital doing other things other than farming was much more valuable to the economy than farming.
Aaron Ross Powell: So you said workers are paid—in this story, they’re paid their marginal product. So if I am working at, say, a restaurant and if they hire me, I will create $10 in additional income for this restaurant minus whatever my costs to the restaurant I’m going to be. So, total productivity is $10, then my wages will equal $10?
Peter Van Doren: Again, the—right. So there’s—assume everything is held constant and then you add one more cook or one more server and then the question is how much does the revenue of the restaurant rise? And again, we have to talk about marginal versus infra-marginal. So—
Aaron Ross Powell: Define these for—since we’re not economists. Aaron and I are not economists.
Peter Van Doren: Infra-marginal means everything before the margin. So, firms keep hiring people as long as their output exceeds their wage. So actually that’s where surplus comes from, right? So workers actually make more than they are paid and the surplus goes—
Trevor Burrus: And they produce more, you’re saying, you know, than they are paid?
Peter Van Doren: In effect, all infra-marginal workers. So you ask, but wages are set at the margin. If I leave Cato and go to another think tank home, I mean, again, we’re not nonprofit so it’s hard to make the metaphor work. But if we were in firms and we had sales and things like that, everyone’s paid the marginal product of the last worker even though if you’re infra-marginal, you’re actually more productive than that, which is why the firm hired you.
Trevor Burrus: Now this seems all very difficult to actually carry out in a very fluid fashion. What I mean is that, you know, workers—these aren’t just particles that are being attracted to the biggest attractor without any friction in between workers.
Peter Van Doren: They’re people.
Trevor Burrus: Yeah, they’re people.
Peter Van Doren: They’re not just commodities.
Trevor Burrus: Yeah. And so they don’t want to move and they don’t want to retrain and they like their job and they have this personal satisfaction.
Peter Van Doren: All true. All true.
Trevor Burrus: So it’s not just their like—so, is the non-fluidity of labor become an issue in this or something that we should be concerned about?
Peter Van Doren: Well, certainly the data show that American workers are moving around physically a lot less than they used to, and economists are quite—well, have been exploring this and we— certainly the number—well, we know that. That’s a stylized fact that it’s true, which is the fluidity and mobility of American workers seems to have dramatically declined in the last 30 years or so.
Aaron Ross Powell: What does stylized factor mean?
Peter Van Doren: It’s—that’s a good question.
Trevor Burrus: I’ve heard you say this a lot too. It’s a—
Peter Van Doren: It’s something that intellectuals use as an adjective for “Can we agree on the following facts—”
Trevor Burrus: OK, so like—
Peter Van Doren: —for purposes of discussion?”
Trevor Burrus: Fluidity of labor has decreased.
Peter Van Doren: Yeah.
Trevor Burrus: That’s our stylized fact.
Peter Van Doren: So I’m saying there’s a bunch of papers out there whose sites I could provide that sort of agree that something’s happened. So, we call that a stylized fact, which is I don’t—that’s a good question. We could have a Free Thoughts where does—
Trevor Burrus: We’ll just assume this, but—
Peter Van Doren: —stylized facts come from?
Trevor Burrus: And so that—I mean the fluidity of labors I’m sure will come up more in this discussion, but that brings up this question where if it’s not moving around much, then there seems to be a problem with, say, these wages or—
Peter Van Doren: That breaks down the process that I described.
Trevor Burrus: Yeah.
Peter Van Doren: Absolutely crazy. I mean just as in consumer markets, there’s something called search. So if we have five or six supermarket chains now in the D.C. area, if everyone habitually shopped at the market near them and no one looked at the odds and no one at the margin shifted their behavior, then the supermarkets would have no incentives to change any of their prices. Same thing in the labor market, which is if this firm treats workers badly and doesn’t pay them very much and people know that, but—or the workers know that, but none of them changed. None of them moved. None of them go anywhere and the firm can keep doing—mistreating them, if you will.
Again, all economics markets depend on not only the existence of choice but the exercise of choice. In financial markets, right? We have index funds or I’m in the Cato 401(k). You are too. We all learned from our financial columns in The Journal or The Times or whatever we read that ordinary people can’t beat the market; therefore, it wasn’t of a index fund. Well, index fund is just buying a random sample—well, it’s buying the complete market-weighted—
Trevor Burrus: Diversified sample, yeah.
Peter Van Doren: —of all firms that are publicly traded. Well, if everyone index-funded, if every—I mean everyone, then stock prices have no meaning. Everyone is just buying the market every month. So somebody somewhere has to be an active investor. So you’ve asked the same question that labor markets or any market were all—when I used to teach a class, I said, “Why is the price of orange juice, you know, $2.50? Why isn’t it $2.51? Why isn’t it $2.49?” And the answer is—and I ask the class, “Do any of you look at the price of orange juice every week?” No one had raised their hand. Then I go, “Why don’t firms change the price?”
Trevor Burrus: They’re not searching.
Peter Van Doren: Somebody does. See if no one checks things out and no one changes their behavior, then all of the stuff we’re talking about doesn’t work.
Trevor Burrus: But I want a segue on that point because actually—so is the answer to the question—so, I mean so people might do a lot of searching on orange juice, but in general, you go to your grocery store more often and things like this and the price may vary. So there—
Peter Van Doren: But how about jobs?
Trevor Burrus: So they’re not—now jobs, they’re different. But like—but then you have people like—this is a segue, but like commodities future traders who are really looking at the price of orange juice, who are searching, who are making choices more than your average consumer, and they’re probably determined.
Peter Van Doren: We are—they are the margin—in other words, somewhere—the reason prices vary and prices matter and wages vary and wages matter is that somebody somewhere is choosing and deciding and changing and we are all mooching off that behavior. If no one thought about changing or deciding, then this thing we call the market and the prices in them would actually be (a) stagnant and (b) have no meaning if you follow what I’m saying.
Aaron Ross Powell: So, when we tell this economic story of where wages come from and this is why workers are compensated in the ways into the levels that they are, one of the pushbacks is to question this notion of the marginal productivity, to say like “Look,” you know, teachers are incredibly important and produce supervaluable things; whereas, you know, the kids who played in the national championship game last night and are going to go on to get drafted in, you know, the early rounds are going to make an extraordinary amount of money much more than the teachers, and they’re not really producing anything of value, you know. I mean—or that amount of fun is certainly not as important or valuable as what the teachers are doing. So there seems to be this disconnect between what we think of as productivity and what we think of as, you know—
Trevor Burrus: Value?
Aaron Ross Powell: Yeah, as value.
Trevor Burrus: This is the kind of question makes Peter’s head—
Peter Van Doren: No, no. I get it. I get it. It’s—economists care about and focus on how people behave rather than what they say. And ever since—I mean I can remember discussions of voting—in New York State, school budgets have voted up or down every year by school districts. It’s the only democratic part of New York State. I mean New York is a big government mostly and the legislatures corrupt and people don’t have much of a say in lots of things.
But, school district budgets are voted on every year by every school district outside of city, so Buffalo. They don’t have votes, but what I’m calling school districts like where I grew up in upstate New York, every year, there’d be a vote. A budget would go before the voters. And every year, the argument you just made would be put forth by the teachers, “We’re valuable and—” all these other pursuits like, you know, don’t spend an extra dollar on a snowmobile, which is a big thing in Northern New York. Send those tax dollars to the school district and make your child have a better future.
And every year, because of New York State politics, voters don’t get much of a chance to protest their anger at things. They would vote down the school budget often and then the board would have to offer one at a time. “All right, New York State loses the taxes of this and the budget is this” without your consent, but if you want sports, if you want afterschool music, if you want afterschool buses for your kids from varsity, if you want all this, you get to vote on a line at a time and say yes or no to here’s what the tax rate would be if we put this more into the budget.
And every year, the budget would get voted down and then they vote on it one at a time and they’d put it all back in because they realize they like stuff. But, they could never vote down or up on teacher salaries, the one thing—But again, people can move around, right? So some school districts in New York State are famous for quality and others are not, and—when my brother moved to Maryland, the same thing. He said, “I’m—” he looked at all the SAT scores by zip code. And he said, “I’m going to move here because here’s the best value for taxes and value-added and money.”
So, enough parents take into account what’s going on and then those districts spend a lot more on students than others and those—Bill Fischel is an economist at Dartmouth. He’s written about this in a book called The Homevoter Hypothesis which talks about how parents sort themselves out across space by how much their school district do spend and how much value-added they think they get. So, ironically, even though you said, “No, they don’t get paid as much as the coaches of football teams,” but teachers don’t have TV revenue. That’s the—
Trevor Burrus: Yeah, they could—I guess I could film their classes. A lot of discussion has occurred in the last 10 years, I’m going to say-ish, especially in the last I would say four, about real wages stagnating especially with the election, I’ve been taking—just finished talking about the struggling worker with Donald Trump and also Bernie Sanders discusses traveling worker and a lot of people talk about wages are flat-lining or stagnating. Is that true?
Peter Van Doren: A lot of what I bring to these podcasts is that when data—be careful of thinking that a data set with a label is actually measuring the same thing over time. So there’s something called wages and it’s produced by the Bureau of Labor Statistics and you can look up this data set and it starts in 1940—all these start after World War II. And then they’re—and you can look at the time series and the label is the same. So it looks to the reader particularly journalists that we measure something called wages and they are going up and up and up and up and then somewhere as you said in the last 10 or 20—depending on how you count—years, something has gone wrong, right? And then the style—again, I’m—
Trevor Burrus: Stylized fact.
Peter Van Doren: The stylized fact becomes part of the chattering class discussion. So The Times and The Post and so-called elite educated media keeps saying over and over again that we all know something called wages—so workers as a percent of national income are getting less and capitals getting more and “we all know that” and then we say, “Oh, we’ve got that new big book by Piketty and by—So, and that sort of somehow—that train, that train of discourse becomes very hard to stop.
But there are some interesting papers by economists that I can describe, which I’ll do now, which basically say, “Hmm, we need to correct four—” and this is four things. So, first, the nonproduction and nonsupervisory workers is technically the set of people that are being—whose wages are being examined by the BLS. Well, as a percent of the US Labor Force, that set of people is actually shrinking. So the percent of people, you and I, all of our Cato salaries are not probably in the BLS data set. And so, “Hmm, I’m not a capitalist, so how come—”
So, the notion that something called the wage database is counting everyone we wouldn’t think of as a rockafeller. That’s actually increased—less true over time, so that’s correction number one, is—
Aaron Ross Powell: What kind of jobs that are more common now are excluded from the production and nonsupervisory?
Peter Van Doren: The funny thing about these data sets is they’re all produced by the government and economists all like use them but academic economists, and I put myself in that category as well, are remarkably uninformed about the mechanics of the gathering of the data for these data sets even though we’re all very dependent on them. The set of people who actually are very good at knowing all this are the economists in the BLS. And if you find them and engage in a—I’ve met some of them at dinner parties and like, “Oh, yeah—” They spend their entire career as, “Oh, yeah, I do the gathering of the data for the Midwest and this data set.” And it’s like, “Oh, OK. Talk to me about what’s—”
Sadly, that knowledge doesn’t tend to enter the journals very much. I mean it’s literally an entirely separate important occupation which is—whose details are almost completely unknown to all of us who talk about these things, which is a sad commentary. So even—
Trevor Burrus: I know that is absolutely fascinating. And I mean, so basically—
Peter Van Doren: We are dependent on, you know, 1300 people who gather the GDP and the BLS and the wage and the income data, and yet they’re not—to be honest, they’re not considered the prestigious jobs in economics. And so what they do and how they do it is actually incredibly under-known and under-appreciated by the rest of the profession. And so, I actually can’t answer Aaron’s question other than to know in general the set of people you and I might be willing to call workers is less and less sampled by this BLS data set.
Trevor Burrus: But production and nonsupervisory seems to be trying to get to hourly wage kind of workers.
Peter Van Doren: These things were set up after World War II. It’s certainly the case that the percentage of people in the US economy fitting that definition were higher. And, the normative purpose of these data sets, which is involved in the struggles between labor and capital and The New Deal and all of that. Probably infused the way these things were defined and how they have been not altered over time. I mean there is a whole—someday we’ll talk—the politics of data and—I’ll give you—
Trevor Burrus: That’s a future episode. We just called the whole thing politics of data.
Peter Van Doren: A brief segue.
Trevor Burrus: Please.
Peter Van Doren: When Reagan took office, David Stockman was head of—he was what? OMB?
Trevor Burrus: OMB, yeah. Stockman, you said, yeah?
Peter Van Doren: So, there used to be something called the low, moderate and high standard of living budgets and I used to use them in class. Well, the low standard of living budget was very, very close every year to median family income. Well that—so, who was behind those data sets? In the politics of data, the answer was labor unions. And Reagan and Stockman, of course, did not—were not in that camp and, thus, the data—those data sets were terminated by Stockman. I mean you can go back and find these—the low, moderate and high standard of living budgets and then early in the Reagan years, these time series just ended. So, why data are the way they are and why they’re gathered is certainly a scientific enterprise in part, but there’s also a—particularly with wages and returns on capital, I mean all that contentious stuff, there is a political component. And Stockman believed that ending the existence of those data sets would in turn undermine some of labor’s attempts to increase wages beyond what Reagan thought was appropriate.
Aaron Ross Powell: You said there were four things and that was the first. What’s the second?
Peter Van Doren: The second is the difference between wages and total compensation. So, employers over time, the composition of the compensation for workers has gone from almost all wages to much more benefits in particularly healthcare. So the increase in cost of healthcare and the fact that employers for many employees pay a component of their health insurance cost and/or healthcare cost directly. That—so healthcare going from whatever, 5% of GDP to 17% of GDP, that’s not in wages .That’s not all of it. Some of it is in benefits and pension benefits and 401(k) contributions by employers, and all of those things aren’t in the BLS wage data set.
So the second correction is to take a BLS data set, include all workers and not just production workers, and then include all compensation and those data sets do exist, although they’re much less prominent and not reported much in the media.
Trevor Burrus: That seems crazy that—I mean one thing that we know in basic, I guess, labor economics but just economics to some degree is that an employer looks at an employee as all the costs associated with that employee if they have to provide them the healthcare, if they have to provide them with different sort of mandates or, you know, if it’s impossible to fire them. All these kind of labor costs, they just see them as the “how much it costs the firm to employ this person?” It doesn’t really matter to the firm that this is how much they take home and this is how much they get in benefits.
Peter Van Doren: The firm could care less.
Trevor Burrus: They could care less.
Peter Van Doren: Completely indifferent.
Trevor Burrus: So, it seems to me—and I thought this before and I’m glad you brought this up that given the skyrocketing cost of healthcare, and I blame a lot of that on the government, and if we do have take-home wages somewhat stagnating. But if last year I cost $6000 to Cato for my healthcare and then this year I cost $8000 to Cato for my healthcare, it’s going to be really hard for me to go and ask for a $2000 raise because basically I got one even though I don’t even know it if they have to provide me with healthcare.
Peter Van Doren: Correct.
Trevor Burrus: That could be a huge factor in stagnating—
Peter Van Doren: It has been.
Trevor Burrus: Yeah.
Peter Van Doren: It has been. And particularly for lower wage, lower-paid employees, healthcare cost as a percent of their income is much higher than it is for higher-paid employees. So the so-called aggrieved workers that have been subjective of debate and discussion in this year’s election, the lower your compensation, the greater this healthcare effect has been on your take-home pay. So the lowest wage employees have been hit the hardest if their employers provided healthcare coverage and, of course, for some at the very bottom, their employers never did or the employees never contributed anything to that coverage, in which case, this argument doesn’t apply. But when you do it in the aggregate and when we do this—when we put this up, right? You’ll have links to—I can put in some of the pages—
Trevor Burrus: Absolutely yes.
Peter Van Doren: —that people can read to follow the charts and data that I’m describing. They’ll see in the figures how much of a change this makes in the data.
Trevor Burrus: So, third.
Peter Van Doren: Third, this is a little more complicated. This is—there are different deflators used to control for inflation. There are consumption deflators and there are, in effect, business deflators. So there’s what’s going on in the US economy is the rate at which prices are increasing for the things people make are differing from some of the things they consume. So particularly in manufacturing because of productivity gains in manufacturing and world trade, the prices and, thus, the revenues to firms for those kinds of goods have been much more attained. The price increases have been much less than for what are called non-traded goods. So, education and healthcare in particular in the United States which are non-traded goods—
Trevor Burrus: So like HDTV—so, we brought up HGTVs before. Those prices are—
Aaron Ross Powell: HD.
Trevor Burrus: HDTV.
Aaron Ross Powell: HG is a place about buying houses.
Trevor Burrus: Oh, yes.
Peter Van Doren: It’s so close to be Home and Garden Television.
Trevor Burrus: Yeah, yeah. HDTV. That’s—yeah. Sorry, I’m a little bit cold, but—So those have gone down compared to, let’s say, your brother is purchasing is a $3000 one.
Peter Van Doren: Right.
Trevor Burrus: But, then the other kind of goods—so what are the other names for those types of goods? Or is it just manufacturing goods?
Peter Van Doren: Well, it’s often traded and non-traded goods, but basically this paper which the link will be in the podcast by Bob Lawrence, when you deflate worker’s pay by a business deflator rather than the CPI, much of the stagnation disappears. They go from—so the healthcare cost component add something. But the real kicker comes from this way to treat inflation. And this is where some of our listeners may go, “I don’t buy it” or whatever. But you’ll have to look at the paper and—
Trevor Burrus: Well, I want to translate what you just said into English for philosophy majors and lawyers. You deflate, so you have the CPI, the consumer price index, which is how you’d measure real—that’s when—
Peter Van Doren: It’s one way. So inflation has to somehow—so the CPI is a market basket of goods and it goes out and prices those goods every month. The physical quantities of those goods are specified every 10 years, literally it’s so many—
Trevor Burrus: Two pounds of flour or something like that.
Peter Van Doren: Yeah, literally.
Trevor Burrus: Yes.
Peter Van Doren: What’s the cost of a hockey stick? You know, hockey sticks are in the Boston CPI. They really are.
Trevor Burrus: Are you serious?
Peter Van Doren: Yeah, no, I am. And they’re not in—
Trevor Burrus: Are show shoes in the Minneapolis one?
Peter Van Doren: And they’re not in—they’re not in Seattle, right? These are literally physically defined market baskets of goods. In the market basket component, what people are buying is updated every 10 years. But in between, it’s not. So literally every month, the CPI is people go out to the stores in all the cities or all these regions that are defined, there’s CPIs for every metropolitan region and then the US is just an aggregation of all of those. These are physical quantities of goods that are priced.
The business deflator is a bureau of economic analysis product from the Treasury department, from the GDP data people, completely different. It is weighted by what people are producing, not what people are buying. OK? So, all the heavy machinery that’s made by John Deere—think of all those combines, all those big earth movers that are then sent around the world to dig up stuff in Australia and China. Those things are not in the CPI, OK? That’s—
Trevor Burrus: I don’t think I’ll buy a combine every 10 years.
Peter Van Doren: People aren’t buying—
Trevor Burrus: Combines, yes. Trucks.
Peter Van Doren: So those things.
Trevor Burrus: Ditch Witches, yeah.
Peter Van Doren: But they’re in what people produce. So the business deflator component of GDP is about all of that. And the prices of those things have been not rising as fast as things like housing and healthcare. And so, once you take that into account because the revenues, the firms that are making those things are also not rising that rapidly because they face world competition. The payments to labor in those firms have not been rising as rapidly as the things that they—as the things that those workers actually consume.
Aaron Ross Powell: Does the difference in these—and so what they end up—what we just look like if you control for one versus the other, does it change based on income? So we said that healthcare, because presumably healthcare is like whether you’re poor or rich, the amount of costs to insure you is roughly the same. So, a low-income earner, healthcare is going to be a larger chunk of their paycheck than a high-income person. Is there a similar thing where the kinds of goods that go into the CPI are more likely the kind of things that make up a larger portion of—so that CPI-related inflation is going to bite the lower-income person harder than the higher-income?
Peter Van Doren: First of all, these things aren’t disaggregated by where you are in the income distribution. None of these data sets are. So, the one—again remember the politics of data sets? So the one political element of inflation that is certainly discussed a lot is, “Should the elderly be treated—should elderly inflation be calculated differently than inflation for the rest of us?”
Trevor Burrus: Because they don’t buy hockey sticks.
Peter Van Doren: Well, and they buy—
Trevor Burrus: I mean, I’m being facetious, but basically, yes. Yeah.
Peter Van Doren: And they buy healthcare.
Trevor Burrus: Of course.
Peter Van Doren: And—since social security costs—social security cost of living increases are tied to the CPI, the advocates for the elderly particularly AARP argue that you should have pharmaceutical prescriptions and healthcare—
Trevor Burrus: And Matlock episodes. All included in the elderly—
Peter Van Doren: And TV land, the price of TV land.
Trevor Burrus: Not Beyonce CDs though, so Matlock—yes, Matlock. No Beyonce in the elderly CPI. That makes sense.
Peter Van Doren: But Aaron’s point is correct, which is this is—everything I’m talking about is for everybody on average, which means there are—different people will be affected differently by this depending on how close they are or not to something called the average consumer or the average business that—whose experience I’m describing. That’s correct.
Trevor Burrus: So, in that—in this third factor, the two deflators, the sort of kicker is that this means that it’s less stagnant if you include this in the wages or less stagnant than the—
Peter Van Doren: Of the corrections so far that I’ve described, this is where a lot of the action occurs.
Aaron Ross Powell: Using the business one as opposed to the CPI.
Peter Van Doren: Correct. And since this is classy mysterious and difficult to explain, this may be the most difficult to win over a non-econ nerdy or non-nerdy econ kind of listener.
Trevor Burrus: What we have—I’m looking at the graph now and we have—we’ll have the paper linked in. And so now we can move on to the fourth factor.
Peter Van Doren: The fourth people probably should be able to relate to particularly if they’re users of computers and things like that, which is—so, the output of people has—you have to, in effect, subtract depreciation is the argument of this paper. So if the capital stock, if all the stuff out there is lasting as long as it always did, neither shorter nor faster, you don’t have to worry about this. The claim of the paper is that the stuff we’re making now because it’s increasingly computer-related has an increasingly short capital lifetime and, thus, the actual rate at which the capital stock of the United States is running out of gas is depreciating is actually increasing. It’s a shorter time period. So buildings and lighting and all that still last 20 years, but software, computers, lots of—I mean this booth we’re recording in, for all I know, it’s already out of date.
Trevor Burrus: I think that’s true.
Aaron Ross Powell: For the most part, yes.
Trevor Burrus: For the most part, yeah.
Peter Van Doren: OK.
Aaron Ross Powell: We were talking about upgrades just yesterday.
Peter Van Doren: You see? All right. Well, the—so, see, the listener can relate to. Even Cato’s audio room is depreciating rapidly. So, once one corrects for this in the argument of this paper, there’s been no divergence between the compensation to workers and the increase in GDP from 1970 through 2008. There is no puzzle to explain according to this paper. Since 2008, there is a divergence and that’s even more—well, we can drift into that but it’s—
Trevor Burrus: It’s more complex.
Peter Van Doren: It’s more complicated and it’s a kind of scary Luddite kind of argument which has been in the media, which is are we getting to the point where the robotification of working life is to the point where there is insufficient demand for workers? We really aren’t coming up with something for everyone to do. Economists have always thought of this as a crazy argument because human needs are endless and we can come up with—I mean it won’t be manufacturing, but it’ll be service-based. But maybe the ability of people to demand things that require labor is, in fact, limited.
Bob Lawrence, the author of the paper we’ll link to, calls this labor-augmenting technical change which took me a while to wrap my head around that, but I finally figured out what he’s saying which is the computers and software and things are making some labor so productive that it’s actually decreasing the price of labor. It’s making them so productive that we need fewer workers and that’s the scary scenario, which is we can’t come up with something for everyone to do.
The right has tended not to worry about that. The left has forever. And the economist who wrote this paper is a center-left, I would say, but not rampant left, not like Piketty and all. And, thus, this is an argument—I mean at a minimum, we have—given his four corrections to the data, for him everything is hunky-dory up until 2008. But even when you add his four corrections, the divergence between worker’s total compensation corrected these four ways and total economic output diverges after 2008.
Aaron Ross Powell: So, to reframe that or express that in the story that you told at the beginning of how wages are determined, does this mean that—so if I’m a firm and I’ve got a certain number of workers, those workers because of the technology that I’ve accessed to in my firm, are so productive that—
Peter Van Doren: They need—
Aaron Ross Powell: —it’s not worth it to me to hire more people.
Peter Van Doren: Right.
Aaron Ross Powell: Because basically it’s going to—hiring an additional person is going to increase productivity some, but either not enough to compensate for their other costs or I’m not going to be able to find the market for that increase.
Peter Van Doren: There we go. That’s what it is. That’s the—it’s literally we can make more stuff by adding more people because of this augmented technical change. However, we can’t figure out how to sell it.
Aaron Ross Powell: Because there’s only so many iPhones you can sell because once everyone has one, you don’t need to sell more of them or—
Peter Van Doren: Now, again, this is—this is my interpretation. Lawrence never says this. He just uses this term and I’ve been reading this paper over and over again for a year. I’ve been talking with Jeff Meyer and my colleague about this. And this is my—so, we’ll get listeners writing in and saying, “Peter, you’re wrong” or “you missed this paper.” This is my interpretation of trying to translate this language into English. He does not.
Trevor Burrus: So I just want to clarify for Aaron’s point that—so if we’re making all the microwaves with one guy sitting in a factory pushing the bottom to, whatever, start the machines and then turn the machines off and that’s all the microwaves we need, that the reason we can increase that production function is because we don’t need more microwaves. Is that basically—or other things too.
Peter Van Doren: This is a—it’s an oddly—
Trevor Burrus: But then—Is this like people adjust their own behavior on this though because—? Isn’t there some sort of that you get two microwaves or you put one in your bedroom or if they’re super-cheat, you might put one in your bedroom and you might put one in your kitchen?
Peter Van Doren: That’s what I’m saying. This—
Trevor Burrus: You know, I might have one in my office so I don’t have to walk down. I’ll always use the one—
Aaron Ross Powell: Similar as the television.
Trevor Burrus: With the televisions, exactly. I mean I was thinking Back to the Future and nobody has two televisions. Well, not, you know, everyone. You have one every room of your house if it were to become that cheap.
Peter Van Doren: All of what you are saying has been the traditional economic critique of claims that ever since the Luddite concerns, I mean this has been going on a long time in western societies with industrialization, which is we need to preserve jobs and economists oppose it, “Oh, come on. We’re—no, no, no.” It’s just—the transition is difficult but everyone eventually will leave agriculture and come to do things like podcasts in a recording studio.
Trevor Burrus: Exactly.
Peter Van Doren: Right? We’ll figure out other things to do and this notion of satiating the demand and, thus, these workers will be let go and then no one will hire them to do anything. I’m not saying I buy that argument. I’m just saying this is my interpretation of what Lawrence is saying. I haven’t seen papers critiquing what he said either in economic language or my interpretation of it. I’ve talked with my colleagues about it. They, too, think I’m probably correct in interpreting—translating what he’s saying into what I think has been a fear of some folks about industrial society for a long time. And, I’m just going to have to say I leave it at that, which is all your objections make perfect sense, but it’s exactly what I would always say in class, “Economists have never—” but he’s a mainstream economist. He is not a Marxist. He’s not a radical economist. He was at Brookings. He was at Harvard. So, for this kind of argument to come from someone like him certainly got my attention.
Trevor Burrus: So I want to get on to some of the things that we discussed when we were planning this episode. We had the wage stagnation issue. Wages and workers, quality of life and the level of wages has been controversial for a while. And there are some things that you hear people say about wages that are fairly common. One is that CEOs and bosses take money from their workers or that they could easily give back and that there is something askew with the compensation regime because the guy who makes the iPhone only makes—let’s say, not the one in China, but the one in Cupertino—makes $35 an hour, but Tim Cook takes home however many billion dollars a year. There’s something wrong with that in terms of how their compensation is determined. What would economists say about that?
Peter Van Doren: There is a literature, a very vigorous literature on CEO pay and what’s happened to it. Again, the term—stylized facts—we—
Trevor Burrus: That’ll be the name of this episode. Stylized facts with Peter Van Doren.
Peter Van Doren: It is certainly the case that CEO compensation has risen dramatically. My reading of the literature is that this arose because of a crazy confluence of factors that no one anticipated and where some very smart people sort of got it wrong. And so briefly, go back to the early ‘90s, CEO pay became a political issue in the 1992 election. President Clinton ran on a platform of doing something about it and in 1993, the tax code was changed so that the tax deductibility of executive compensation was limited under the IRS rules to a million dollars or less if it was cash. Well, then the stock options became the mode of pay and the ’93 through 2001 period was one of the greatest stock booms in US history.
And so, looking backwards, the—and these stock options were not indexed to relative performance. So, basically the rising tide of stocks inflated CEO pay tremendously because boards switched from paying people cash to paying people in stock options. The literature was supportive of that initially but then Jensen and Murphy, two of the prominent people that argued in a famous paper that CEOs were made like managers in the old days in, let’s say, ‘50s and ‘60s and they should be paid incentivized somehow. And they have completely—they have argued in the literature that stock options were a big mistake. People should be paid in stock, not stock options. Boards paid in stock options because they misperceive the cost of stock options to be free or low instead of redistributing from existing shareholders out there.
So, in that sense, the boom in the US economy, more of it went to CEOs than would have gone to shareholders if the pay had been in stock and particularly stock that was indexed to relative performance. I mean the whole economy boomed and if you’re paid in stock, no one should be paid the SMP 500. You’re running one particular—so the question is how’s your company doing relative to other companies? All companies in the ‘90s were doing great. Remember? We had a budget surplus in 19—I mean there was just—it was one of the most unbelievable booms looking backwards. Productivity increased a whole lot. No one anticipated any of this. And because it is law and then the change to stock options rather than stock, and particularly stock that had to be long held, you just had people in this giant cash machine tied to the stocks in general.
Trevor Burrus: So it’s not as simple as saying that they can just give back to workers in that—although they’re taking from workers. They’re taking from shareholders.
Peter Van Doren: Probab—I mean yes. In that phase—
Trevor Burrus: “Taking,” something like that.
Peter Van Doren: There’s a paper by a guy named Kaplan from Chicago Business School that says since 2001, CEO pay has actually not been—it’s not increased dramatically. And, in fact, in 2010, CEO pay was at the 1998 level. It actually had gone below what it was in the boom. And there’s also a skewness to the CEO—it’s right hand-skewed, so the average CEO pay is very different than the median. To be sure, the median rose but the mean really rose, so there’s some particularly egregious cases that skewed the average. When you look at the median, things are less dramatic, though quite much higher pay than “average workers.” That’s certainly the case.
But, it now appears to be the case that boards and academics are getting it. Stock options are now out. Stock is in and stock that has to be held a long time—See, Rex Tillerson with his appointment as Secretary of State, he’s not supposed to be able to sell his stock. He’s supposed to have a very long term interest in Exxon Mobile that matches the interest of equity holders of that company. And ironically, his government services, he’s going to be—I mean they had to figure out this way to—he can’t—
Trevor Burrus: Require himself the stock.
Peter Van Doren: Under contract, he can’t—he’s not supposed to sell. So the board—if I’ve read correctly, actually it’s had to make him more short-term oriented in order to fit the ethics constraints of public service.
Trevor Burrus: So with all these questions about wages and the justness of them and in who makes what and what they should make, is there any sort of conclusion that seems obvious here where you say, “Oh, well, the government has a role or should be doing something about fixing this if someone perceives a problem in wages.”
Peter Van Doren: One, I think—one answer I have is when everyone asserts something, you should take a deep breath and look more carefully at data sets, how they’re constructed, why we infer what we do from them, and realize how complicated it is to construct the data set across time that’s called consistent in measuring the same thing. It’s not. It’s not that easy. And the sad thing is that less-informed discussion of data sets informs politics rather than more informed and probably can undo that, although this podcast is in attempt to do that.
Second, given our discussion today, notice nothing jumped out in an obvious way for governments to somehow intervene to do something other than the cost of housing. I mean remember we said what consumers are buying, the price of those things is going up a lot, and as you said, healthcare and housing. You see healthcare regulation as well as owning regulation we’ve talked about in other discussions are probably the reasons for the prices of those things rising so rapidly.
Aaron Ross Powell: Free Thoughts is produced by Evan Banks and Tess Terrible. To learn more about libertarianism, visit us on the web at www.libertarianism.org.