Claudia Goldin on the Economics of Inequality (Ep. 133)

How to model social progress.

Harvard professor Claudia Goldin has made a name for herself tackling difficult questions. What was the full economic cost of the American Civil War? Does education increase or lessen income inequality? What causes the gender pay gap — and how do you even measure it? Her approach, which often involves the unearthing of new historical data, has yielded lasting insights in several distinct areas of economics.

Claudia joined Tyler to discuss the rise of female billionaires in China, why the US gender earnings gap expanded in recent years, what’s behind falling marriage rates for those without a college degree, why the wage gap flips for Black women versus Black men, theoretical approaches for modeling intersectionality, gender ratios in economics, why she’s skeptical about happiness research, how the New York Times wedding announcement page has evolved, the problems with for-profit education, the value of an Ivy League degree, whether a Coasian solution existed to prevent the Civil War, which Americans were most likely to be anti-immigrant in the 1920s, her forthcoming work on Lanham schools, and more.

Watch the full conversation

Recorded September 1st, 2021

Read the full transcript

TYLER COWEN: Hello, everyone, and welcome back to Conversations with Tyler. Today I’m chatting with Claudia Goldin, who is, quite simply, one of the leading economists. She is professor of economics at Harvard University, has made major contributions to economic history, the theory of income inequality, economics of education, gender economics, and much more. Most exciting of all, she has a new and excellent book out, called Career and Family: Women’s CenturyLong Journey toward Equity. Claudia, welcome.

CLAUDIA GOLDIN: Well, I’m very, very pleased to be here, Tyler.

COWEN: Let me ask a question I’ve been wondering about. Why does China, right now, have so many of the world’s self-made female billionaires?

GOLDIN: I don’t know, but I wouldn’t mind being one. [laughs]

COWEN: But it’s a big change in the payment of women, and it’s happened in one place. It hasn’t happened in America.

GOLDIN: As you well know, it’s a very large country. Let’s face it, Chinese women became more liberated during the Communist Revolution. Anyone who went to China during the period after the revolution saw that women really did work alongside men.

There’s no question then that equality has escaped China, as it has escaped much of the world, but I do get the sense . . . I was there in 1975, which is a very long story. I met many different people who were in some middle ranks, upper-middle ranks of the party, and there were a lot of women. Let’s face it, there weren’t a lot of women in 1975 in Congress, and certainly not in the White House.

COWEN: But if we look at a CCP meeting today — a photo — it just seems to be all men, right? There’s no Nancy Pelosi in there. At the middle level, women don’t seem to earn that well in China, but somehow, at the very top, there’s been this big breakthrough.

GOLDIN: Well, kudos to them.

On the pay gap

COWEN: A question about America. As you’ve shown, from 1995 to 2008 in the US, the gender earnings gap actually expanded. Why was that?

GOLDIN: Oh, it didn’t expand that much. We’re talking about blips on blips. It expanded more for the higher educated than for the lower educated — or should I say, it actually closed for the lower educated, whereas it stagnated for the higher educated, and the latter — a lot of that is due to the enormous expansion of inequality at the top. Women paid a relative price for the fact that inequality just zoomed upward.

COWEN: How much of that do you think is due to the growth of super firms? Especially in tech and finance, those seem to be fairly male-intensive areas. If they’re earning much more at the upper end, that would push up the income gap.

GOLDIN: I think it’s less that than the distribution itself, that men, for lots of different reasons, tend to be in the way-right tail. That’s going to be in finance. It’s not just in tech; it’s going to be in all areas. It’s going to be in surgery. It’s going to be in dentistry. It’s going to be in lots of different self-employment.

COWEN: It seems scalability should really matter. I could be the world’s best dentist. I can’t serve that many customers or patients well —

GOLDIN: Thank goodness.

COWEN: — but I won’t be the next Mark Zuckerberg, right?

GOLDIN: Right, but finance is certainly scalable. Let’s face it, surgeons are — when we look at the data that’s been put together for the upper, upper end — surgeons tend to be up there too.

COWEN: Now, you’ve argued in your book and in numerous articles that flexibility of hours is a critical variable for thinking about the earnings gap between men and women. If we look at the literature on firms, we seem to see that fixed costs are rising, intangible capital is rising, lower marginal cost, higher markups.

Does that mean that flexible working hours are hard to achieve when fixed costs are high? You would think when fixed costs are high, you want to basically take the people you have and drive them as much as possible, which, in fact, you see in finance and in tech.

GOLDIN: Right.

COWEN: So, should we be pessimistic about gender pay equity moving forward?

GOLDIN: The fixed cost here can also be coming from certain labor, from training, but you’re saying that the fixed costs are all that capital per labor has increased?

COWEN: Or its branding. Facebook, Google are very famous companies. They’ve achieved a certain position. At the margin, they want to work the people they hire fairly hard.

GOLDIN: Yes, but it doesn’t seem that difficult to create a small amount of redundancy to have good substitutes. All of these firms have professionals — think about consulting firms, accounting firms, banks, financial corporations, lawyers — all of them have individuals who are quite special, quite unique. And yet, I don’t think it takes that much to create a small amount of redundancy so you can have people hand things off.

That doesn’t mean that you’re creating cadres of individuals who are look-alikes and who are driving down their own price. Flexibility isn’t that difficult to achieve for the individuals who are at that upper end, whose earnings you are going to have to increase even more if they put their foot down and say, “I am not working during my vacation.” You’re going to have to increase their earnings even more. I don’t think it’s that heavy a lift.

COWEN: As skill levels rise, doesn’t it get harder to hand things off rather than easier? In a pharmacy, there’s skill required, but it’s not extreme.

GOLDIN: Well, that’s what you think. [laughs]

COWEN: Substitute in another economist for an hour of your time — that’s very hard to do.

GOLDIN: I don’t think so. I know one. I live with one who I use as my substitute. [laughs]

COWEN: Well, one, right? A coworker.

GOLDIN: First of all, I don’t think it’s in the training as much as it is in — and I think most people in the business would agree with me that, yes, there’s a lot of training for many different fields — accountants, lawyers, consultants — but most of the problem that they would claim there is, is in the handing off of information, in the passing off of information with high levels of fidelity and with trust. I think that one of the things that the Zoom world has done is it’s led me to trust you even though I can’t touch you.

COWEN: Do you think there are scenarios where marriage makes a meaningful comeback for lower socioeconomic groups? And what might those scenarios look like?

GOLDIN: Well, as someone who got married five years ago after living with someone for 26 years, I would have to get out of my own life into someone else’s. But I think that for many people, marriage creates more stability, either because they’ve made a commitment to themselves or those around them see them as more committed. I think the main benefit is to children, so that children have more stability.

COWEN: Sure, but for people without a college degree, marriage rates have fallen dramatically in the United States —

GOLDIN: Dramatically.

COWEN: and a rise in single-parent families. Can you see economic forces that might reverse that? Or is that just going to keep on going?

GOLDIN: I don’t know if it’s going to keep on going. There was a really great piece in The Times the other day about mainly women with less than a college degree who said, “I don’t want to get married or have kids right now. I want a career. I want to be financially independent.” I don’t know whether that means that they are eschewing marriage altogether and that, in fact, we have a new group of women who, even though they don’t have higher degrees, are wonderfully financially independent and can raise children in good environments.

Obviously, the big issue — and this is the one I think is in the back of this question — is what’s happening to the men. If we can lift the economic opportunities of men who don’t have college degrees, then some of the problem, I think, will be solved. But that’s a big problem.

COWEN: How do men and women respond differently to narrow financial incentives in the workplace? Or is it just the same?

GOLDIN: To narrow, meaning they —

COWEN: If you give people bonuses, do they work harder, would be one of many questions. Do men and women respond the same to that or differently?

GOLDIN: I really don’t know. It’s a very, very good question. Did they respond in terms of working harder? That’s a good question to ask some consultant. I’m trying to think about whether . . . I think my answer would be that they both would respond the same, but that women have greater constraints who are parents.

I can be in a law firm and have phenomenal incentives. I can be in an accounting firm with phenomenal incentives. We know what those types of bonuses look like. But if a parent says, “I cannot work Thursday mornings,” and therefore gives up that benefit, that isn’t because she’s a bad worker and she’s not responding to the benefit. It’s that her needs are inconsistent with the firm’s demands.

COWEN: My intuition would be that whether women work at all depends more on incentives, but how much they work at the margin — if they work — depends less on incentives.

GOLDIN: That’s a good point, but of course, that initial margin is the bigger margin than the one that gets you into the workplace —

COWEN: Oh, sure.

GOLDIN: — and so then you have the issue of selectivity. The women who are in the workplace and in certain types of workplaces are going to be selected either on the basis of personal characteristics or the characteristics of their spouses.

COWEN: Why does the wage gap flip for Black women versus Black men?

GOLDIN: [laughs] These are great questions. These remind me of questions that you might get in some generals exam, and then you think, “What does the wage gap flip?”

The Black male wage gap is a very difficult statistic because it’s a highly selected statistic. As we know from lots of very good work, if we take the population, the entire population will get one number. If we take the population that’s not incarcerated, we’ll get another number. If we take the population that’s never been incarcerated, we’ll get another number. I think that deep in that calculation are some of the nation’s largest ills.

COWEN: When I read noneconomists on wage gaps, I see the word intersectionality very often — the notion that there’s some nonlinear effect created by combining different types of discrimination. What is your take on intersectionality? Does it play a role in your argument?

GOLDIN: I would like to take a small course on intersectionality because I’m not certain — if I model this as a theorist — where the sections are going to be and how I figure out how many sections I want.

Someone has a gender at birth, they have a decided gender perhaps, they have a race, they have where they grew up. Many of these have a religion. They have the education of their parents, [laughs] and many of these covary — they’re correlated.

In fact, this would be a great PhD thesis, and for all I know, someone would tell me, sure, when Rosen wrote it 40 years ago, or maybe Eddie Lazear. I don’t know, quite frankly, what to do with intersectionality.

COWEN: What do you think of this as a simple model? We all know the standard economic model — there can be too many tolls on the medieval river, right? It’s double marginalization. That can be a nonlinear impact — too many tolls, the river is just unusable. There are negative externalities, one predator on another. Think of discrimination as a kind of tax, which it is, and then you have multiple taxes from different directions, and they interact like the tolls on the river, and maybe intersectionality makes a lot of sense, no?

GOLDIN: Yes, I think that that’s a very, very good way of thinking about it. Thank you for giving me a little bit of theory to hang my hat on. But the problem is whether these are multiplicative, whether they’re less than that, if there really is some issue having to do with functional form, and I don’t know whether they are separable. The way in which you said it, they would be separable. Thank you for the idea. I will pass it on to one of my really good graduate students to model and then test. That’s a very, very good point.

COWEN: Another question.

GOLDIN: Another PhD dissertation coming from Tyler.


COWEN: Well, that’s what we’re here for.

GOLDIN: That’s what we should call this: PhD dissertations coming from Tyler. [laughs]

COWEN: When I speak with high-achieving women I know, they tell me something frequently, more or less the same — that there’s a fear of sexual harassment, and this leads women to choose more careers where they don’t put themselves out there so much, and as a result will earn less, or quite likely they’ll be less famous, suffer other penalties, and that this is a major factor in career choice and earnings. What’s your take on that?

GOLDIN: Well, I think it can be. I in no way, shape, or form want to minimize that. My sense is that gender differences in earnings are not, in the majority, due to factors like that.

However, those factors we really want to root out. There’s obviously a difference between dating and romance, and sexual predators and sexual harassment. The fact that we have many so front-page cases about sexual harassment — both heterosexual and homosexual harassment — shows that it exists, and it should be rooted out. How much it is affecting the choices that women make and men make, I think, is not that gigantic.

COWEN: If we’re trying to explain gender pay gaps or inequities, and we think there are at least two factors — one is, for a given job, women might be paid less, right?


COWEN: Another factor is that women may end up in jobs or in majors or with interests that pay less in the longer run. How do you assess the relative proportion of those two factors?

GOLDIN: Well, we do all sorts of calculations to try to figure out what’s happening within versus between. Of course, one of the problems that we always face is the issue of aggregation. The more you disaggregate occupations or firm occupation, the more you’re going to minimize the problems that are happening within. It’s my sense, from the work that I’ve done, that the issue of occupational segregation, for example, has been vastly reduced over time.

What’s happening within occupations? The real challenge there is to understand whether individuals are being paid less for absolutely equivalent work within the same firm — and that is the standard federal guideline that one is looking for, that’s the smoking gun — or whether individuals are getting paid less, for example, because they’re working at the boutique law firm where they have amenities that they want, or whether they’re working at the Madison Avenue, Park Avenue law firm where they have fewer flexible amenities. I think that those are the challenges and the data challenges that we all have.

COWEN: As you may know, there’s a debate in experimental economics and also field experimental economics as to whether women are more averse to competition than men. One doesn’t have to think this is intrinsic; it could itself be the result of discrimination. But what’s your view there?

GOLDIN: I know all the experiments, and they’re done by extraordinary scholars. I myself have no idea how to do experimental economics because you could do just about anything, and you just have to be incredibly clever at designing these experiments.

I know that there’s lots of evidence that women take fewer risks. They are less pushy. They are more insecure, or at least we think that they are more insecure. [laughs] There are very, very interesting experiments on stockbrokers, but of course, there’s the flip side that individuals who are highly risk-taking often lose a lot.

I think that, more and more, as women are educated at lower levels to realize that they can be anything that they want to be . . . If we look at younger women versus older women, I think that we do see that women are more competitive now than they certainly were when I was a kid.

It’s interesting that there is the great book Women Don’t Ask, by Linda Babcock and [Sara] Laschever. I remember when that first came out, and I said, “How can you say women don’t ask? Go to any department store and see who’s standing there screaming about the fact that the toaster doesn’t work.” [laughs]

COWEN: As you know, if we look at undergraduate majors, the ratio — it’s like 3:1, men to women, right? It’s not obvious that it’s budging. If we look at faculty, I read one estimate that the ratio for women hasn’t really improved since the 1990s.

GOLDIN: The ratio — you’re saying in economics or in what?

COWEN: The gender ratio. I’m not sure what level they were counting or adjuncts included, but it wasn’t obvious —

GOLDIN: You’re saying 3:1 for a particular major?

COWEN: For economics majors.

GOLDIN: Oh, okay.

COWEN: I believe the ratio’s 3:1 and not really budging.

GOLDIN: Well, I put those numbers together.

COWEN: Okay, and how is it evolving?

GOLDIN: It was more like 4:1 than 3:1, now about 2:1, so it has increased by quite a lot. One of the problems that we have, as a field, is that when students — before they even come into their freshman year, and they’re asked what do they want to major in, women will — if they want to major in the social sciences — will put down psychology, and men will put down economics, so we lose them before they even unpack their suitcases.

I ran an RCT I called Undergraduate Women in Economics Challenge. We had 20 universities and liberal arts colleges begin a program for freshmen to explain to them that economics is not just about the things that their parents have told them about, which they find boring, which don’t involve people, but economics is a very people-oriented subject. It involves inequality, it involves children, it involves obesity, it involves health, it involves everything.

We moved the needle a bit. We’re still trying to figure out exactly how much we moved the needle. But the problem is, we do very, very poorly in our PR ourselves. So if you read the textbooks — and they are changing, and there are a number of recent textbooks that understood these problems — that economics is X and Y; it’s agents.

Most young people don’t want to deal with agents. They want to deal with humans, and particularly, it seems as if women, more than men, would like to deal with humans. And so women will tend to go into this field called psychology or sociology, which indicates in their writings that it’s about people. We indicate in our writings that it is about agents or Greek letters, and so we just have to do a lot better.

COWEN: How is the gender ratio for tenured economics professors evolving?

GOLDIN: CSWEP, that is, the American Economic Association Committee on the Status of Women in the Economics Profession, puts together those data, and they are creeping upward, as one would think they would. Now, in some very interesting work, we can see that they are — I said the words “creeping up.” They’re not bounding up.

The reason that they’re not bounding up at first we thought was a pipeline issue, but now it’s pretty clear that it’s less the pipeline issue. While it was there and still is there, there’s also an issue with the fact that women — more than men — will leave academia and use their economic skills extremely well but outside the tenure track.

COWEN: Should we either abolish tenure or fundamentally change it so as not to penalize women, but also for other reasons? It seems grossly unfair.

GOLDIN: Yes, I certainly wouldn’t do it because it’s a penalty for one group or another. It is an issue that arose. It solved potentially an issue from some time ago. I don’t know if the fundamentals are the same that would support it. It is clearly a difficult problem in the sense that individuals have higher productivity when they’re young and then lower productivity. I think that having long-term contracts is useful for individuals in fields where they’re so good.

If we think about sports, clearly those people just clean up at the beginning, and then they become used-car salesmen or whatever they become. So the question is, do we have a completely changed academic environment in which the superstars clean up at some point and then they go into something else? I don’t think that that would work either.

I think that a very important thing to think about is that we produce a large number of products. We produce knowledge, and that’s really the hardest thing to figure out how to do, but we produce the dissemination of knowledge and teaching. And people who are older probably do that better, and that is not generally what tenure is given for.

COWEN: I’m sure you know the [Betsey] Stevenson and [Justin] Wolfers paper, 2009. Why haven’t American women gained more in happiness?

GOLDIN: [laughs] See, I’m not a great fan of that. I remember Betsey was interviewed on Mother’s Day one year. She was a new mother. The person interviewing her said, “Your work shows that mothers are less happy than they had been before they became mothers.” The person said, “That was a great question for Mother’s Day.” Betsey was stuck, and she said, “Well, yes, but they would never give up their baby.” I think that for me, as an economist, that says it all. There’s some problem.

COWEN: So, you think it’s not true that American women are less happy than they had been? I’m not sure what your stance is.

GOLDIN: My stance is that I’m not happy with the happiness measures.

COWEN: Do you think there’s something else they’re not picking up?


COWEN: That’s an ordinal-choice thing rather than stress in the moment?


COWEN: And that if people are choosing something out of a greater set of opportunities, they probably are happier?

GOLDIN: That’s right. It’s very hard to do that in general. Sometimes we rile up a group and get them to remember the things that they’re less happy about, and sometimes we show them where their lives are safe and secure and wonderful and plentiful, and then they get happier.

I remember when Dick Easterlin first began the happiness literature. I remember very, very well — it was a very long time ago — having dinner with him in Madison, Wisconsin, and talking about it. I was enthralled with trying to understand this happiness literature.

That was across countries, and it seemed interesting, from my Chicago theory point of view, that people recenter their happiness. You can be in a place in which, if you were plunked down there from somewhere else, you’d be miserable, and yet, you’re there, and you recalibrate yourself, just like people recalibrate themselves when they have a bad health event. It takes a while. I’m not certain what to do with the happiness literature.

COWEN: What best correlates with women keeping their maiden name after marriage? You’ve written on this.

GOLDIN: Right. For the period in which it increased enormously, which was in the early ’70s, there were several things that were important. The first one was that they were given an appellation. Rather than being called Miss or Mrs., they could be called Ms. It just opened the spigot. It just meant that more and more women could do it.

But more importantly — and this gets to the title of the article that I wrote with one of my undergraduates — it was that they already made a name. If you publish, or you’re a lawyer or a doctor or a yoga instructor or a hairdresser, and you are known as Tallulah Bankhead — to pick a great name — it’s going to be hard to change it. It became your identity. It became your calling card.

Then the question is, okay, fine, keep your name, but then there were some barriers to that. The barriers then began to be taken down. There were institutional barriers. There were legal barriers in terms of having to change your name with the Social Security system, on your passport, on your driver’s license. There were retrograde states that weren’t letting women who got married not change their names.

I remember I was at Princeton at the time, and I was married to a Princeton professor. I was no longer a professor for Princeton, and the library insisted that I use his name on my card despite the fact that I had been an assistant professor there with my name. That was one of the greatest affronts. All of that disappeared. That became, once again, a great enabler.

Then finally, it’s what to do when your kids are in school, and you have a different name. It’s very easy when the plumber comes to the door, to tell the plumber, “I am not Mrs. Katz,” but it would be much harder. It’s often hard still, unfortunately, to pick your kid up at school when you have a different last name than they.

COWEN: Why is the title Ms. more or less disappeared from the scene?

GOLDIN: I don’t think it has. Where has it?

COWEN: I heard it a great deal in the 1980s. It feels today that I hardly ever hear it. I also hear Mr. less often, to be clear, but Ms. seems to have vanished, viewed as almost condescending. I don’t know, I’m asking. I genuinely don’t know.

GOLDIN: Tyler, what do you hear?

COWEN: People avoiding the designation altogether.

GOLDIN: But on forms, the list of potential appellations will always include Ms. I think that one reason that it may disappear — and I think that it’s nice if we actually get to know our first names — is that, particularly in certain communities, there may be a sense that you don’t want to use any indication of gender. That gender has become a no-no.

I’m on a national committee in which we’re very used to talking about two genders. We have two genders, X races, Y ethnicities. You multiply them together, and you get the number of cells. And now we’re into more genders, so I think that may be one of the reasons.

COWEN: How much do you think the trans movement will, in fact, disrupt feminism by breaking down the two traditional categories, men and women? If someone wants to say, “The ratio of men to women in this area is that,” the trans movement could say, “We can’t quite presume that,” and blur the entire comparison. Is that going to matter moving forward? Or is that temporary?

GOLDIN: I don’t know. I don’t think any of this is temporary. If anything, I would applaud the fact that we are smoothing over gender and moving . . . I would think that in the greater scheme of things, that is a move towards greater equality in the sense that individuals — there is fluidity in who they are. I’m not quite certain what to do with it, but I applaud it.

COWEN: When you read The New York Times wedding announcement page, what honestly do you think? I tend to think two things: “My goodness, these elites are remarkably creative and self-replicating,” and then it strikes me how much people who are not religious at critical points in their lives actually want to invoke religion for the ceremony, but then it’s often just discarded again. That’s what I actually think when I read that page. What do you actually think?

GOLDIN: The first thing is that I’ve been reading that page for a very, very long time. For those viewers who don’t know, when I did the piece on name changes, one of the datasets that we used was The New York Times Style section, now called Vows. It would list whether the bride was or was not changing her name. We used that as a dataset.

To be perfectly frank, The New York Times, in their own mysterious ways, has changed that page to be some very — I shouldn’t say very, but relative to what it was — non-elite.

COWEN: Is single-sex education for women underrated or overrated? What do you think of it?

GOLDIN: I have a personal stance that it’s highly overrated, but I wouldn’t want to get into an argument with my friend, colleague, a person I have enormous admiration for, Amy Finkelstein, who believes thoroughly in it. When I was a kid growing up in the Bronx, I couldn’t imagine a world that I wanted to be in that didn’t have boys.

At the time, girls were, I think, far less assertive, far less serious. I had girlfriends, but I enjoyed being around the boys a lot more. I went to Bronx Science, which was about two-thirds boys. When I applied to college, I just could not imagine going to an all-girls school. Now I know the arguments in reverse, and I respect them, but it was certainly not for me.

On for-profit and non-for-profit education

COWEN: Why does for-profit education seem to fail so badly? Some parts of the learning sector are for-profit. Your books are put out by for-profit publishers, textbooks are. Berlitz seems to do okay relative to the language classes at your local state college, but for-profit education, as a whole, seems to implode. What’s the model there?

GOLDIN: I think it’s some for-profit education. It’s certainly not all for-profit education. I think that the problem is that the part of for-profit education that expanded the greatest was for BAs, particularly in fields like business administration management, and they became more BA mills.

When I was studying for-profit education, I really wanted to be able to look at the materials that were given to people. What were they studying? I was never able to break into that.

The problem is twofold. One is that there’s a huge profit incentive for the company. That’s, of course, not always a problem, as we know.

The other one is, the individuals are not as aware as they should be of what they’re getting themselves into. They are often very needy. They’re low-income. They’re first-generation college. Someone is dangling in front of them something that they really want. They desperately want something that’s going to get them out of the hole that life has put them in. That person who’s dangling it is not giving them sufficient information and sufficient amount of time to get out of this contract.

I think we saw that very, very clearly in the GAO audit studies of the for-profits. I think that that’s the problem. The for-profits that give short courses that give a one-year certification in, let’s say, medical technicians — what we’ve shown in our work, those are not the big problem. The big problem is what I just said.

Let me just say that I have taken short for-profit courses on dog training, and they work really well.

COWEN: For nonprofits, I’ve seen estimates that maybe no more than 40 percent of the people who try, finish. Does that just mean we have too many people trying to attend college? Isn’t the true rate of return on education actually pretty low, at the median at least?

GOLDIN: Let me back up and say that when we were just talking about for-profit education, that the non-for-profit, the non-selective portion could use a bit of change as well. It’s not as if they’re charging as high an amount or recruiting using techniques that are as extreme and exploitative as the for-profits, but they also may not really wake up individuals and tell them exactly what they have to do and how much time they need to spend.

We do know, Tyler, there’s incredibly good, really superb research on that marginal student. This is Seth Zimmerman’s phenomenal work. The marginal student does gain a tremendous amount by getting into a good college.

COWEN: But is that true for the marginal student on the side of the margin where they don’t finish? Your own work with Katz suggests there’s this discontinuity, some people who simply can’t leap the hurdle into having a college degree. It’s somehow too hard for them temperamentally.

GOLDIN: Possibly. I’m trying to go back. We have to go back to Seth’s work to know the point of identification. I thought the point of identification was the individual who was graduating from high school. He’s, in fact, including in that the probabilities that they don’t finish, and the rate of return for that person who was able to go to this Florida International was incredibly high.

COWEN: If you look at a school, say, like Duke or Emory, is it a long-run problem that if they admit people on their merits, there’ll be too many women in the school relative to men, and some kind of affirmative action will be needed for the males?

GOLDIN: These are private institutions, and they can generally accept whom they would like to accept for various reasons of diversity.

COWEN: Should they do that? Or should they just get in 76 percent women, say?

GOLDIN: I’m brought back to the original issues that were raised by a small number of liberal arts colleges and universities in the ’50s and the ’60s about why they should become coeducational institutions.

Those reasons were that their marginal student was not going to Princeton but going to Harvard, not going to Princeton but going to Penn, not going to Princeton but going to Cornell, because that student wanted an education that was more balanced in terms of what the world would look like when they got out. And that more balanced, then, was not necessarily Blacks, Hispanics, and Jews, but the one major thing that was missing from Princeton and Yale and Dartmouth and Amherst and Wesleyan and a whole bunch of places was women.

Those institutions, in a process that I’ve described in the origins of coeducation, led these institutions to move in the direction of accepting more women. Now what’s going through your mind, I think, is, “Yes, but they weren’t lowering quality. In fact, they were increasing quality.” Diversity, in any dimension, can be thought of as a plus for everyone.

It was about 10 years ago that some dean in a small liberal arts college in the Midwest admitted to the fact that they were accepting men with lower SAT, ACT, and grade point averages to increase diversity.

COWEN: Men, probably, are not less intelligent than women, on average. What’s the pipeline problem? Is it too much homework and too many extracurriculars in high school or something else? Where are we failing our young boys?

GOLDIN: We can go back to as early as we have data on high schools and know that girls attended high schools, graduated from high schools at far, far greater numbers than boys. If there is an issue here, it’s certainly not extracurriculars. It may have to do with what’s going on in your cells and this difference between this Y and this double X.

COWEN: The value of an Ivy League degree — what percentage of that value do you think comes from signaling as opposed to learning?

GOLDIN: Very little. I think that it’s not signaling. It’s probably networks.

COWEN: Networks. Why is there a sheepskin effect, then, which is quite strong? Because you could meet people there for three years and then just not finish.

GOLDIN: I don’t know what the numbers are for the sheepskin effects. I’m always very dubious of anything having to do with real signaling. In terms of sheepskin effects, I’m willing to read up on it, but I don’t know what it is. The networks are enormously important.

Sometimes I look at my undergraduates, and I think, “I would just be such a better person if I could live with these people. These are the most accomplished people.” You would be able to listen to someone playing the cello. You would be listening to someone teaching you about Verdi operas. They’re just an incredible group of individuals. In some sense, it’s not the same networking that Teddy Roosevelt would have been able to do, or John Kennedy. It’s, instead, an ability to grasp just an enormous amount in a mere four years.

I went to Cornell, which, of course, is an Ivy League school, too. I had a phenomenal experience, but nothing like I would have if I went to Harvard right now. That’s why I want to go back to college.


COWEN: You’re there now, right? You get their classes.

GOLDIN: [laughs] I know, but I’m on the giving end and not the receiving end.

COWEN: The literature on returns to skill — there’s a piece by Beaudry, empirical work by Rob Valletta, which argues that post-2000 or so, the returns to skill are, in general, collapsing. Do you view that as stepping outside of your model with Katz in your book Race between Education and Technology? Or do you dispute the result? Or how do you see that?

GOLDIN: I would have to go back over the returns to which skills.

COWEN: Most skills. They suspect it may be due to automation or artificial intelligence.

GOLDIN: I read the work of David Deming, in which the set of skills that are in enormously great demand are a set of social skills. The question is how do you up social skills? It seems as if we know a lot about the demand for social skills; we know a lot less about the supply of social skills. It may be that when we think about the skills of the past, we’re thinking about, first, manual skills, then motor skills and mental skills. Now we’re moving into the range of social skills.

On past and future projects

COWEN: Our last segment on economic history. Why wasn’t there a Coasian solution to the Civil War, namely just buying out the slaveholders? This is going back to your work from, what, 1977?

GOLDIN: Well, Lincoln wrestled with that. When one thinks about what went on in the White House during the Civil War, one thinks about Lincoln just physically wrestling with all of these difficult issues. In fact, the slaves were purchased in the District of Columbia.

He was doing calculations about how we could pay slave owners for their property. For the longest time, one of the central issues in Lincoln’s mind was that property was extremely important to protect, and slaves were property, so we would have to buy out the slaves. He made lots of different calculations.

Of course, there were many different emancipation programs that preceded that. In fact, you may know that the North had slaves. New York did. New Jersey did, and by acts of the legislature, slaves were freed as of a certain age. That was what was called gradual emancipation. It wasn’t very good emancipation, and it could have had some horrific effects. In fact, one of the effects was that slaves were shipped from these northern areas to the South.

There were lots of different potential programs. One of the things that Frank Lewis and I did when we computed the cost of the Civil War was, we were essentially telling Lincoln, “Oops, you should have done it.”

COWEN: What best predicted anti-immigration sentiment in the 1920s?

GOLDIN: Well, what probably best predicted it was a combination — it appeared from some ancient work I’ve done — of areas that got immigrants, and had reduced the opportunities for others in the area, and for whom the immigrant groups that were there already were not the ones who were flooding in. One could think of the various parts of the country.

Ironically, the South was really pro-immigrant for the longest time because it viewed these immigrants as a wonderful substitute for a labor force that was migrating north and had, for decades, not been slave. So, the South was pro-immigrant. They became anti-immigrant only when they realized that immigrants weren’t going there, and when they were going there, they were Italians, and they were dark. These were pretty racist people.

In the North, the places that began to shift being anti-immigrant — and I was looking more at the shifts of voting for legislation that would put in place rules for literacy — those places were places that did not have large numbers of people from eastern, central, and southern Europe. The people there went on slamming the door in the face of their relatives.

COWEN: Very last question. Your book is coming out. Again, that book is Career and Family: Women’s Century‑Long Journey toward Equity by Claudia Goldin. Other than promoting your book, what will you be doing next?

GOLDIN: Oh, wonderful. I’m working on a couple of projects. This is a lot of fun. There was something called the Lanham Act that was passed in 1940 that was for infrastructure for America — lots of money. By 1942–43, it was pretty clear that America needed to — in fighting this war at home, in producing goods for the Allies, we needed to have women employed. So, the Lanham Act funds were repurposed for nursery schools and after-school programs.

What we’re doing with Joe Ferrie and Claudia Olivetti is finding out exactly where these Lanham schools were. Through the amazing ability that exists now, we can figure out whether individuals who were three to six years old during some period had some high probability of going to a Lanham school, what happened to them 50 years later. That’s one of the things I’m doing.

Another thing is trying to figure out why men who are fathers do so well relative to women who are mothers. That’s another piece to the puzzle.

COWEN: Claudia Goldin, thank you very much.

GOLDIN: It was a real pleasure to meet you almost in person, Tyler.

COWEN: Same here. Till next time.

Thumbnail and header photo credit: BBVA Foundation