Wednesday, February 26, 1997

Stanford economics professor and Nobel laureate, Kenneth Arrow and Hoover fellow Kenneth Judd discuss the sources of the persistent and increasing inequality of American incomes.

Recorded on Wednesday, February 26, 1997

ROBINSON: Welcome to Uncommon Knowledge. I'm Peter Robinson, a fellow at the Hoover Institution. Our show today: the gap, the growing gap, between rich and poor. Back in 1968, the average household income for the richest one-fifth of American households was about seventy-four thousand dollars. Here's a seventy-four thousand dollar slice of pie. In the same year, the average household income for the poorest one-fifth of American households was about one-tenth the amount for the rich, or just about seven thousand dollars. Here's a seven thousand dollar slice of pie. By 1994, some twenty-six years later, a couple of things had changed. For one, the economy itself had grown. There was, so to speak, a bigger pie to be carved up. For another, the richest one-fifth of households had seen their average income grow in constant dollars to nearly a hundred and six thousand. I will now attempt to cut a one hundred and six thousand dollar slice of pie. (Julia Child said not to worry about a little mess in the kitchen; it's all part of fine cuisine.) One hundred and six thousand dollars. But the poorest one-fifth of American households? Well, they had seen their income grow by only a few hundred dollars, so they were still effectively at about seven thousand and change. Our guests today: two economists. Kenneth Arrow is a professor of economics at Stanford University and a winner of the Nobel Prize. Kenneth Judd is a senior fellow at the Hoover Institution. Ken Arrow and Ken Judd have given a great deal of thought to the rich getting richer while the poor just about hold their own.


ROBINSON: So, we've got two distinctive trends. One is that the rich are moving up faster than anybody else, and the second is that the poorer have been moving up and have now effectively plateaued. Maybe a little up, a little down, but effectively treading water. How come? Ken Arrow?

ARROW: Well, that's something--I must confess--economics isn't all that good at explaining. There are three major hypotheses that have been advanced. One is the foreign trade, that we have increased imports of domestic, sorry, manufactured goods from low-cost countries--Indonesia, Malaysia, and, of course, now China--low-wage countries and that there are certain industries--you can certainly see it, there's no question, if you look at the textile industry, you can see this very plainly.

ROBINSON: Or, I just bought a big, new Sony TV and the box was stamped, "Made in Mexico." This is an example of what you've said. So the wages went to low-paid Mexican workers as opposed to American workers.

ARROW: It's to the emergence of these countries--these countries are getting more productive, partly because they're managed by Japanese, as you suggest, or American or European companies, but these countries, their productive skills are going up. From a world point of view, of course, this is good, you understand. From the point of view of the American workers, who are in competing industries, this is not very good. Second, the only reason to suppose this can't be the major explanation is that the volume of these imports simply isn't that big, fewer than 3 or 4 percent of national income. It's hard to believe that that--so I think that it has an effect, but it's hard to believe it's the major thing.

ROBINSON: So Pat Buchanan and Ross Perot are overreacting? There's no giant sucking sound?

ARROW: There is a sucking sound, but it's not giant.

ROBINSON: There's a little sound, a little hiss.

ARROW: That's right.

ROBINSON: Okay, so there's a hiss--that's one item.

ARROW: A second factor is immigration. The immigrants, as least a large fraction of them, are competing for low-wage, low-skill jobs. This is a controversial question. Some experts, someone like George Borjas, who is, I think, the leading expert, does believe this is a significant effect. However, everybody agrees that these two put together can't possibly explain most of the change.

ROBINSON: Before we go on, I just want to make sure that I understand the immigrant point. So, we have, for example, in Southern California, a few hundred miles to the south of here, where in the past twenty years the immigrant population has increased by about five million or so, largely from Mexico. So the notion would be that those people are right here in American territory, competing against formerly high-paid American workers.

ARROW: And they typically go into low-paying jobs and low-wage industries, because they typically don't have as much education or skills or experience as the average American worker does.

ROBINSON: Okay, so in the world today, there are a lot of people who are willing to work for less than American workers. Some of them live in Mexico, some of them live in China, and some of them have moved here.

ARROW: Most scholars in this field feel, if you make all the allowances, you still have the majority of the growing inequality unaccounted for, and the general--now, let me give you two theories, but the one that most people stress is that the changing technology of American industry is creating a greater demand for high-skilled occupations and high abilities. We can come later to that, because there are some problems with that theory.

ROBINSON: Ah, you're a skeptic on that one, but you're--

ARROW: I'm sure it's something to it, but I'm having difficulties making a full explanation for reasons that I can enlarge on later. There's another more impalpable one, which is that there's a shift in values--that, in effect, there were fairness criteria, particularly in the giving of wage distribution.

ROBINSON: You've mentioned several phenomena which are objective things you can point to and then another phenomenon, values, that's inside people's heads. I'd like to save that one for a moment, because it strikes me as trickier, and perhaps even more interesting, ground. But, Ken Judd, Dr. Arrow has mentioned, in effect, newly available pools of cheap labor. You'd agree with that?

JUDD: Yes, I agree that that's important.

ROBINSON: And you'd agree that it only accounts for a dinky little amount of the income inequalities.

JUDD: It appears that it's related to a small, relatively small, fraction of the economy, and so there's some question, here, legitimate question as to how important that is for this issue. It's certainly a teasing factor.

ROBINSON: Are you tending to disagree with Dr. Arrow here?

JUDD: The numbers really aren't there to say who's right, who's wrong. Basically, I think there's a laundry list of reasons. I agree with the reasons that Ken has put on this laundry list; there's a couple I would add.


JUDD: For the last, you look over the last thirty years, we've had deregulation of some industries, for example, in the airline industry and trucking, for example, and regulation tended to help certain workers get higher wages than they might have gotten otherwise without regulation, and so that has blunted wages in some respects.

ROBINSON: Would you draw an analogy between the effect of regulation and the effect of unions? Both were, in effect, controlling the markets.

JUDD: Those are not unrelated, because deregulation has certainly weakened the power of some unions.


JUDD: And the other factor that is important for some of these income inequality statistics is that there have been important social changes in terms of the nature of family. For example, we like to talk about how the typical family are today and how much they're making relative to the typical family thirty years, but you're really comparing apples and oranges. The typical family thirty years ago was close to the Ozzie and Harriet TV example.

ROBINSON: There were a mom and a dad and two or three or four kids.

JUDD: Mom and Dad, and Father who had a job and Mother who stayed home and took care of the house, whereas today you have a much greater variety in terms of what we call a family. Even if you have mother and father in the house, it's very likely that both have full-time jobs.

ROBINSON: What about that one right there alone? The statistics are for households, so what we're talking about when we show this income inequality, we're talking among households. So, is it to a certain extent accounted for simply because in the upper income households women go out to work? There are two incomes?

JUDD: That's part of it. On the other side, you have a lot of single mothers trying to raise children by themselves and there is no man in the house, so you have a greater variety of family situations. And part of this is simply due to women's liberation and changes in attitudes. Thirty years ago, if a woman was unhappy in her marriage, she basically was stuck there, whereas now she has more opportunities to go out and support herself and her children, and so, well, you have more divorce--

ROBINSON: So, at the lower end, would you tend to find a lot of single working mothers without the additional income of a father?

JUDD: That's certainly is, yes.

ARROW: There's no question about it. And that's a change from the long-term history of the country. It used to be, if you looked at the people in the bottom, say, 10 to 20 percent in income distribution, they were single males, drifters, and now, what you'll find is that it is dominantly, as you say, single-headed households. One figure, which also has some great social consequences--consider the percent of children who are brought up in the lowest twenty percent of the income distribution. It used to be 15 percent. Now, that's going back quite a ways, forty or fifty years ago. Today's 25 percent.

ROBINSON: Hobos don't have that many children, but single working moms have children.

ARROW: In fact, you have more children in this lower group than there are adults.

ROBINSON: And that has implications for the way the country feels, lives--

ARROW: That gives me concerns about the future.

ROBINSON: Single heads of families may be doing badly, but the heads of corporations are doing amazingly well.


ROBINSON: Let me go back to what Ken Arrow mentioned a moment ago, the argument of values. I found a concise statement of this--striking to me--in an article by Paul Krugman, who's at MIT, formerly of Stanford, now at MIT, and he points out that in 1970, the CEO, the chief executive officer, of a typical Fortune 500 corporation earned about thirty-five times as much as the average manufacturing employee. Today, twenty-six, twenty-seven years later, it's common for CEOs to be paid a hundred and fifty times as much. So, what happened, says Krugman, and I'll read you a sentence: "Though America was a society with large disparities between economic classes, it had an egalitarian ethic that limited those disparities. That ethic is gone." His point is that what used to be considered obscene behavior--boards lavishing huge payments on CEOs--is now considered normal. Do you buy that?

ARROW: I do.

JUDD: I buy the statistics, and maybe there's been a change in attitude.

ARROW: I must say, I've been on boards of directors, and you can start to see the change.

ROBINSON: You can see it? You've seen it in your own experience.

ARROW: Yes. It's funny about CEO salaries. In some sense, what is the basic of the CEO salary? It's what other CEOs are getting. But you're competing in that market, you know, trying to track them. If you're asked what a fair salary is, the typical procedure is to get an outside consulting firm which gives you the statistics. They try to make it comparable--you control this, you control that, whatever it is--you get the range, the distribution of salaries of comparable executives.

ROBINSON: So they're just giving you the market price in effect?

ARROW: They, in effect, give you the market price. So you say, "Well, we want to pay a little more than the average," or "We want to keep him," or "He's better," or something like that. Well, this kind of a situation, since there's no very well-defined equilibrium, of course, they all go up. The relatives are unchanged. And it's very hard to assess what the value of the CEO is to the corporation as opposed to the next best. I mean, there's not exactly a great deal of evidence to that question. So, conventions, values, can play a big role in this, and I think there was a period where--there was a philosophy around in the fifties about, say, these new companies here, of being a team. I mean, some of the companies were calling themselves "associates" or using some word like that to convey the idea that they were--

ROBINSON: Teamwork?

ARROW: Teamwork. At least the engineers, the skilled workers, the engineers and so forth constituted a team. I think that's not even part of the mythology any more.

JUDD: The focus on CEO salaries is interesting, but how many CEOs are there in the Fortune 500? You're talking about 500 people. And then maybe you add on the managers, even if you add in a hundred, you're up to fifty thousand people, but one percent--the work force is, what, a hundred some million--one percent of that is a million people. So, what's going on with CEO salaries is interesting to talk about, but certainly that is not having a big impact on the statistics. Now, maybe there is a change in values, but if it's going to have an impact on the statistics, it has to be having an impact much lower down in ranks in terms of the organizations. [Fade out.]

ROBINSON: Changes in demographics, changes in international trade, changes in technology--but what a layman like me wants to know is what are the underlying values here?


ROBINSON: You look at this pattern of growing income inequality, and how do you respond to that? Is it--aside from free trade--is it, in and of itself, for the shift that it represents within American society, anything that alarms you?

ARROW: Ah, yes. Now let me see. There are two things. One is a value statement, which people may differ on. I find inequality to be, in and of itself, objectionable. Not everybody agrees to that statement. I mean, just the mere fact that, though we're a part of the same community, we differ so much is a bothersome thing. That doesn't mean that every inequality can be rectified, because sometimes the rectification creates problems that are worse than the disease, but at least, it's a problem you want to address if you can--

ROBINSON: But it is, in and of itself, somehow offensive to you that your fellow men and women should live so--

ARROW: Yeah, I mean, what I consider, one thing that I'm concerned about is the future--and we may already be seeing some of that now. Every theory of human capital suggests that one place where human capital is formed is in the home, the family.

ROBINSON: And human capital means?

ARROW: The ability of people to produce, based on their own abilities; the formation of--

ROBINSON: Of intelligent, healthy, well-educated--

ARROW: Intelligence, health, habits, aspirations. Do people care about education, for example, or are they so short-sighted they don't realize the long-run gains from education? The value systems in the family and in the neighborhood, which can militate against--and there's a lot of evidence that's suggesting the importance of these effects. Now, poverty means a deprivation of opportunity for the child. The mother isn't there, is working all the time. There's children just put in front of television sets, and this means that the workers twenty years from now are going to be very ill-formed.


JUDD: I don't regard income inequality as a per se "bad," and I don't even think all the so-called poorer people do. Some of the income inequality that we see is just a reflection of the fact that many more people go to college and go to graduate school today than they did thirty, forty, fifty years ago, and are in higher paying occupations. I look at my family, for example; there's far more income inequality in my family in my current generation than in my father's or grandparent's generation, but that was because all my ancestors basically went to school and then they were farmers and everybody pretty much had roughly the same standard of living.

ROBINSON: So, your dad is a farmer, and you hold a doctorate and teach.

JUDD: Yes, and so there's some growth in income inequality there, but I don't think anybody would regard that as bad.

ROBINSON: And your dad doesn't resent the income inequality?

JUDD: No, neither do my sisters. I mean, they're happy that I have what I have, but I have it because I went to college and got a Ph.D. So, to some extent, income inequality is just because some people choose different occupations and have different choices. And that kind of inequality is something which I don't think many people would object to or think of as being a per se "bad." So, that's why, in order to make these judgments, one, I think, has to understand where the inequality comes from. And if it comes from personal choices and even just differences in abilities, most people understand that. On the other hand, if it comes from manipulating the political process and getting favors from your friends who are in power and denying equal treatment to other people who are out of power, then that's illegitimate, and that kind of inequality is nuts.

ROBINSON: The big question at last: With income inequality growing, what can be done?


ROBINSON: There are reasons to be queasy about this pattern. You both agree on that. So, my next question is, What do we do about it?

ARROW: Well, you pose a very difficult problem, because it's hard to see measures. My own view is that there should be some more redistributive measures, of which the earned income tax credits is an example.

ROBINSON: Could you explain how that works?

ARROW: For people who have earned income up to a certain point, there's essentially a deduction on the income tax.

ROBINSON: For working people at low levels it tends to supplement their income, is what's going on.

ARROW: That's right.

ROBINSON: So, that's redistribution. You cream off from the well-to-do or at least from the rest of society--

ARROW: Minimal incentive effects compared to other forms.

ROBINSON: And an incentive effect is what?

ARROW: People might find it more profitable not to work.

ROBINSON: In other words, paying them to be idle.

ARROW: But the tax credit avoids that because they keep part of what they earn and they keep an increasing part as it gets poorer.

ROBINSON: I'm a little startled, because we've been discussing a gigantic change and a big discontinuity with American history, and it can't be your whole solution simply to tweak a program that's already in existence. Is that the case?

ARROW: Well, I'm afraid I'm convinced by the fact that the market is extremely powerful--quite apart from one's attitude towards it--it's extremely powerful, and there's not--

ROBINSON: So you feel helpless before the market?

ARROW: I feel that we can do some things, but it's probably in the long run, we just have to wait until it rides out.

JUDD: The focus of the earned income tax credit and redistributive taxation and welfare programs and unemployment programs--I think that we need some kind of safety net like that, but that is basically trying to deal with the problem after it's already been created. What we should do, what policy should be focused on, is preventing the problem in the first place, and I think the one thing that the data tells us is that your fate is--a lot of it's luck and accident--but to the extent that it's predictable, it's highly typed to your education, and it's the high school–educated worker and high school dropouts who have suffered in the last thirty years.

ROBINSON: You'd agree with this? This is not contestable.

ARROW: That's right; it's factual.

JUDD: And when you look at education, college-level education, post–high school in general--I'm not saying that everybody should go to Stanford or anything like Stanford--but post–high school education in general, from an investment point of view, has a very good rate of return. It has a rate of return at least as good as the stock market.

ROBINSON: You're telling me that if I were an eighteen-year-old with fifty thousand dollars, I could either invest it in the stock market or invest it in a good college education, and the return to me over the longer term of putting the education would be higher than sticking it in the stock market?

JUDD: That's been the historical experience.

ROBINSON: And you'll agree with that?

ARROW: Not the last two years. But over any long period of time, yes.

JUDD: But what we need to do is, if we care about the income distribution twenty, thirty, forty years from now, we need to deal with it now by seeing to it that people have an adequate education to compete in the future world, and that is something that can affect not just the few people at the bottom but affect the large fraction of the people that we're talking about that have fallen behind.

ROBINSON: Surely the field of economics can tell us whether income inequality is going to shrink or get wider?


ROBINSON: Your word a moment ago, Ken Arrow, was that the market forces are huge and we can do this and we can do that, but we need to ride it out. So, let me close by asking you a prediction. Ten years from now, the year is 2007; is the gap continuing to widen or is it beginning to close? Ten years from now--Ken Arrow?

ARROW: Let me say that, really, anything I say is a pure guess.

ROBINSON: We'll take a guess from a Nobel Prize–winner any time.

ARROW: I feel that you're going to get--as measured by overall income inequality--we're going to get a reduction.

ROBINSON: It'll start to close?

ARROW: I believe that the higher productivity that's going to be generated through all these technologies that, so far, have not paid off very well but are going to; the telecommunications, communication, information technologies are going to start reflecting themselves. This is going to show up and increase the middle incomes. I do fear that there's going to be a really lower class, not very big, that's permanent, because of the permanent damage to the family situation, the poverty today, the inherited poverty that may produce a bigger diversion at the very bottom.

ROBINSON: Okay. Ken Judd, ten years from now, income inequalities continuing to widen or starting to close?

JUDD: I guess I'm more pessimistic on that. Again, it's all a guess. And the horizon, ten years, a lot of things can happen in ten years. You think ten, twenty, thirty years, like the last thirty years. On the technology side, I'm not so sure it's going to end up turning around and helping everybody. You look at the advances in computer technology over the last twenty, thirty years. That pace is going to continue over the next twenty years; nobody really questions that. And the image of an automated factory eliminating blue collar factory labor comes to mind, and I see computers moving in there and replacing people. And then on the international side, if the United States continues a free trade policy and if countries like China want to continue an export-driven development, that's also going to spell bad news for blue collar workers. If Detroit thought they had problems dealing with Japan and Korea, just think of the image of a billion Chinese building cars. I'm not so optimistic on this, when you look ten, twenty, thirty years down the road.

ROBINSON: Well, I'm so depressed I barely have the strength to manage it, but--Dr. Ken Arrow, Dr. Ken Judd, thank you very much.


ROBINSON: Ken Arrow wants to expand the earned income tax credit. Ken Judd wants to invest more in education. But both economists agreed: For the most part, there isn't really all that much we can do about income inequality. In my own view, if the poor are going to be getting the same share of the pie, no matter what, then economic growth--baking the biggest pie that we can possibly bake--matters a lot. I'm Peter Robinson. Thanks for joining us.

More from Uncommon Knowledge