Listen: Economist Jeffrey Frankel '74 on TPP, Trade, Inequality, and the Election

This October, Jeffrey Frankel '74, one of the leading experts on international trade, delivered the Claire Wilcox Lecture. In his lecture, TPP, Trade, Inequality, and the Election, he gives an overview of changes in trade theory and addresses the connection between trade and inequality, which Frankel says "is especially salient now." 

Frankel is the James W. Harpel Professor of Capital Formation and Growth at the Harvard Kennedy School and the director of the program in international finance and macroeconomics at the National Bureau of Economic Research. From 1993-96 and from 1999 until now, he has been a member of the NBER’s Business Cycle Dating Committee, which officially declares recessions.  He was a member of the President’s Council of Economic Advisors from 1996-1999 and chief economist in 1996-97. Frankel's research interests include currencies, commodities, crises, international finance, monetary policy, fiscal policy, regional trade blocs, and international environmental issues.  has co-authored one of the leading textbooks in international economics, written numerous academic articles and op-eds, and maintains a blog in which he shares his views on the economy and the world

Jeffrey Frankel '74 graduated from Swarthmore with a B.A. in economics and a Ph.D. in economics from MIT. Before joining Harvard's faculty, he was a professor of economics at the University of California at Berkeley. 

This lecture honors Clair Wilcox, who was on the faculty of the College from 1927 to 1968 and chaired the Department of Economics for 37 years. 


Audio Transcript

Mark Kuperberg:  So welcome to the Clair Wilcox lecture. My name is Mark Kuperberg, I'm in the economics department. So Clair Wilcox toured at Swarthmore from 1927 to 1968. If you do the math that was 41 years. He was chairman of the department for 24 of those years so I assume that was a time in which people thought being chairman was a good job. With all due respect to my current colleagues and former colleagues, I think it's fair to say that Clair Wilcox was the most eminent economist to teach at Swarthmore College for any extended period of time. His two fields of expertise were industrial organization and, as an undergraduate I read, the textbook for my course was, "Public Regulation of Business", which is one of his ... one of the main books he wrote.

The other field of expertise, which I was totally ignorant about as an undergraduate, was international trade. And it turns out that his influence on public policy or his attempt to influence public policy was bigger in international trade than it was in industrial organization and it came in two spurts. So the first spurt was, this is from the New York Times in 1930, he organized a petition of 1,028 economists to protest the Smoot-Hawely Tariff. Because there was only 1,028 economists, of course congress ignored them, but he ... So if you think about it, 1,028 economists, there's no email, and what he did was he got Swarthmore students to write individual letters to all of these economists to see if they would sign on for the petition. And, as I said, it didn't work. The Smoot-Hawely Tariff passed.

The other main event was, he led the United States negotiating team post World War II at the International Trade Conference in Havana, Cuba, that established the GATT, the General Agreement on Tariffs and Trade, which was a system by which tariffs throughout the industrial world were lowered post World War II. Jeff is gonna talk a little bit more about that. I'll just do the funny part.

So it turns out that, during the conference, this is from the New York Times, Clair Wilcox was acting chairman on the United States' delegation when the conference voted to cancel all New Years Day meetings. Professor Wilcox called it an act of irresponsibility in the wake of having two days off for Christmas. "We've had adequate time for play," he said, "I propose that we work." So I would bet the seminars ran overtime.

So today we're joined, as we have for many years, by Clair Wilcox' daughter, Andrea Wilcox-Palmer and her husband Clarkson Palmer, and we're happy to have them here. But now for our speaker, Jeff Frankle. Jeff Frankle graduated from Swarthmore College in 1974 with High Honors. He then went on to MIT for his PhD where he palled around with Paul Kreugman and people like that so ... here they are partying in Lisbon, Portugal. So without the hair, it may be hard to tell, this guy over here is Paul Kreugman and here's Jeff Frankle. If you want to know the other two people, come see me after.

So Jeff, after partying in Portugal, taught many years at the University of California Berkeley and then moved on to the Harvard Kennedy School where he is now. So as it said on our ad, which brought you here, he's one of the world's leading experts on international trade and that is true. I downloaded his CV to prepare for this talk and I have to tell you, it was not a green thing to do. So here's his CV. 

So his CV is 51 pages long. Jeff is also a member of the NBER's business cycle dating committee so they're the people who determine when a recession begins and ends. So if you have problems with the dating of the Great Recession, he's the guy to complain to. Any case, it's a pleasure to have Jeff here to talk about international trade which combines both one of the major fields that Clair Wilcox was involved in and where he clearly worried about tariffs and trade and a subject that is so relevant for this year's election. Thanks, Jeff.

Jeff Frankel:  Well thank you for that introduction, Mark. As I understand it, I'd be really flattered at the standing room only crowd but Ec 1 students been told this is going to be on the final exam ... so that's what I will attribute the attendance to.

My title, which is mostly given to me, is that fair?  [crosstalk 00:05:36] Negotiated, is on many peoples minds, the election and, particular, trade and inequality. Well, who was Clair Wilcox? Professor Kuperberg has already done an admirable job of explaining that. Let me just add, on the effort to start the International Trade Organization, which the victors, the allied countries after [inaudible 00:06:07] all thought that they were gonna do, and which culminated in that Havana Charter, was supposed to establish the International Trade Organization and which Clair Wilcox chaired, the reason why it's not more famous is the Senate did not ratify it. There was protectinist, isolationist forced then, as always. In my mind it sort of goes down in a long history of admirable, noble efforts led by the US in the cases I'm gonna name, to negotiate an international or multi-lateral agreement and to have very substantial success at getting what we want, you could have though a hundred percent, and then the president brings it back home to the Senate and they turn it down.

So that happened with the League of Nations, [inaudible 00:06:56] and a number of other agreements. Now the story had a happy ending because for a while the GATT, the General Agreement on Tariffs and Trade, which Clair Wilcox was also, had already been very active in leading to success, that kind of filled in for quite a few years until the WTO came into existence in 1994 which was the ITO that had originally been envisioned. And I can't help but see TPP as perhaps in line with that history because we pretty much got what we wanted out of it but the usual view is it does not stand a good chance of success in the Congress. I think it's still possible after the election.

If there are questions about that, if you want to talk about it but I'm going to talk about trade and inequality.

So I'm going to give you an overview of all my remarks before I get into the detail. It seems to be there's pretty wide agreement that trade is good if your criterion is GDP, is overall economic growth. Agreement among economists, and surprisingly, actually, the poll results show most Americans seem to support trade.

The doubts seem to concern things other than GDP, like labor rights, the environment, and inequality and the connection between trade and inequality is particularly salient now in the era. So that's what I'm going to focus on. I think it's fair to say everybody was caught by a surprise by Donald Trump's success and, a bit of an analogy, perhaps with Brexit in Britain, and plenty of the elites ask themselves, why didn't they see it coming? The usual story is, well, we didn't appreciate the people who feel that they've been left behind, workers who feel that they've been left behind by trends beyond their control, such as globalization. And we underestimated how much they've been hurt by it and how angry they are, and if we had known that or had paid more attention and hadn't been so out of touch, maybe we could have predicted, should have predicted Trump's nomination.

I don't see it that way. I'm gonna talk first about trade and maybe I'll break for questions when I finish the part about trade, and then about inequality and it's causes, and then conclude by talking about the election.

So what do I mean, why we agree that trade is good for economic growth? Well, let's see, how many of you are taking international economics or trade ... mostly, not a few. I'll give you a whirlwind tour of the last 200 years. There have been four big waves in trade theory. The first was classical comparative advantage, David Ricardo, countries to better by producing what they're best at, and that would give a very strong result that free trade is good for everybody. It's so good that a country should unilaterally remove it's tariffs, even if other countries are not going to, which in fact is what Britain did when it repealed the Core Laws in the 1840s. But the classical theory of comparative advantage left a lot out. Assumed that everybody was the same in a country. It assumed perfect competition, it assumed constant returns to scale, it assumed that technology didn't lessen change.

And so the second wave is the Heckscher–Ohlin theory, which I'm going to put aside for a moment. The third wave is bringing in imperfect competition and increasing returns to scale which my MIT classmate Paul Kreugman got a Nobel Prize for doing. He wrote those papers a few years after MIT, after that picture was taken in Lisbon. And then, most recently, the new trade theory which differentiates, some firms are more productive and better at exporting and it gives you adenogenous productivity, that's Mark Millet's.

Pretty much throughout all this, trade's good. I mean, under the Kreugman version, unilateral liberalization is not so clear, but if we all together agree on liberalizing trade, like through the GATT or the [inaudible 00:11:59] that's good for everybody. So it didn't really change the fundamental bottom line. And Millet's are ... really comes down in bottom line to an additional reason.

So that's theory, which is all pretty much pro. What about empirically? There's lots of econometric studies. There's one that I particularly like, just to make it very concrete, if you estimate openness to trade by the ratio of trade to GDP, every .01 increase in that ratio raises real income by about three and a half percent over the next 20 years. So that really adds up. There are countries where trade is 100% of GDP and it makes a big difference, exports plus imports we're talking about, it could be more.  And most Americans seem to agree. It's actually quite surprising, and especially most Democrats. So a pretty substantial majority support free trade.

So what about the effects other than total GDP such as inequality? Trade maximizes the size of the pie but what about the equity of how the pie is distributed? Well, first thing ... Actually, before I do this, I'm gonna make a statement people might find surprising. Inequality had declined very sharply over the last 20, 30, 40 years. What's the trick? Was that slide long enough that you could figure out the trick?

Speaker 3:  Global inequality?

Jeff:  Global inequality. Now, of course, how foolish of us to care about people in other countries. We are talking election here, and so I won't spend more than a slide or two on that, but it is Swarthmore and presumably we do care about other countries. I do anyway. It is worth mentioning. And trade, I think, is really the major reason. Globalization is the major reason. So here's, this is the number of people in poverty. As of 1990, the mode of the distribution had most people under ... this is dire poverty. This is $1 per day. By now, by 2010, the mode is at the poverty line and the average, I'll show you another chart in a minute. A billion people have been lifted out of poverty over the last 30 years and trade and investment have a huge part to do about that.

So here's the global poverty rate. In 1990, 43% of the populations of developing countries met the definition of poverty, then defined as $1 per day, 1.9 billion people, and by 2010 it was down to 21%. So it fell by more than half and there's no reason why it can't fall by half again over the next 20 years. So that's important to remember. But nobody would ever say that in a presidential debate. All we care about is Americans so I'm gonna assume from now on that all we care about is Americans and we care both about the size of the pie and the distribution of the pie among Americans.

Well, so I already said this sort of, after the surprise, the typical story goes globalization, trade, and immigration has contributed to inequality and some feel left behind, have been left behind, and they're really angry and they support radical change and if only we'd been more in touch we would've seen that coming. My take, to summarize it very briefly, there's no denying that inequality has increased a lot within the US and within some other advanced countries, our trade and immigration probably play roles or negligible roles, it's hard to pin it down exactly, along with a long list of other factors, which I'm gonna go through.

I personally don't see that being concerned about the issue of equality logically leads you to Trump. Now maybe it politically led there. I'm not a political scientist and politics are not logical. I'm not saying that there wasn't something happening there politically but if we use some economic logic, I just don't see it at all. Excuse me. There are some clear answers to the question, "How can we address the well being of workers who have been left behind?" And I'm going to talk about those towards the end.

Needless to say, trade creates winners and losers. Everything creates winners and losers. There's really no major change you can make that doesn't help some people and hurt some other people. For example, putting up tariffs would create a lot of big losers. If you saw industries, exporting industries or industries that are dependent on imports, they would lose a lot and the average person would lose. So just saying that there are winners and losers doesn't begin to answer the question.

Pareto superiority is sometimes demanded in order to make a statement that policy A is better than policy B. Some economist theorists say you have to say is Pareto superior, which has to mean that every single person is better off or at least as good off, at least as well off. You can't compare two different choices if even one person is worse off. Well, if we stick to that, you know I teach at Kennedy School of Government, called their policy school, I feel like I should sometimes take positions, have positions on issues. We could never take positions on issues if we required Pareto superiority.

But if we say what we care about, the size of the total pie and we care about equality as measured by GD coefficient or the poverty rate or some other measure, then often there's a trade off between growth and equity. But sometimes there isn't. Sometimes they're are win, win situations and so we can't approve of a policy that raises total income and also improves some measure of inequality. And these don't always go the same way, these different measures but per our purposes most of the time this evening, they will go the same way.

All right, so does trade worsen inequality? Are those losers, the people who lose from trade, relatively concentrated more among lower income people? Well eventually I'm gonna get to the Second Welfare Theorem of Economics but let me just put that aside and start first by asking, are the losers from trade concentrated in the lower segments of the income distribution? Just as a first pass, to not have to learn any fancy trade theory, but just sort of think about the effects, I would list three effects.

Trade is good for consumers and we're all consumers. It gives us the opportunity to import at lower prices than we'd have otherwise, sometimes higher quality and wider variety. Kreugman's contribution was the role of getting the consumer a wider variety.  This is gains to people at lower income levels. Going to Wal-Mart, buying a bunch of stuff made in China. It's gains at the upper levels too, imported luxury products, wine or whatever. My guess, there's research on this, but my guess is that it helps lower income families maybe slightly more in percentage terms than those at the top.

The one that everyone's concerned about, of course, is that imports tend to hurt those in import competing sectors. In the short run you lose jobs. In the longer run, you have lower real wages. And on the other side, those loses are offset, at least partly or maybe entirely, by the gains for exporting sectors. You create more jobs in exports and eventually they pay higher real wages and there is this estimate that export jobs pay an estimated 18% more than other jobs so if people shifted gradually out of import competing industries into export competing industries, that's one measure of how we'd be better off.

What does real trade theory have to say about it? Well this second wave that I mentioned is called Heckscher–Ohlin Samuelson. When I was first studying international economics at Swarthmore in the 70s, that was kind of the dominate theory. It was the dominate theory in the 50s, 60s, and 70s which is kind of ironic because that is a period when there was maximum American support for free trade. Why do I say it's ironic? The journalists love to write stories about economists. There's some maverick economists who are fundamentally rethinking classical trade theory and are part of new things which change the verdict that change is good for everybody.

Well the strongest target in real trade theory that workers get hurt by trade is in the classical version, which is Heckscher–Ohlin. It divides the world, the population into labor versus capital or more sophisticated later versions, maybe into unskilled workers versus skilled workers. And there is a theorem called the Stolper–Samuelson theorem that specifically predicts that workers in the capital abundant country, like the United States, will lose from trade. Not even that they don't gain as much as the capitalists, but that their wages will actually fall which is really quite striking and you would think would give all the ammunition that you need.

Now it's an important theory. It never did fit the facts very well or correctly, at least not if the other factor is capital. You can get a little farther with skilled versus unskilled labor. I think it's got some insight. It's got some relevance but it doesn't capture everything. It doesn't capture the advances in trade theory that have been made since 1980 having to do with imperfect competition and increasing returns to scale and Millet's thing about productivity. These new theories you can get openness to trade gives you a permanently higher growth rate whereas in the classical comparative advantage, openness to trade just gave you a one time increase in real income.  Still, there is something about this theory that seems to be, lately, kind of ringing true. That unskilled workers are, perhaps, not benefiting from trade.

But what about NAFTA? Let me spend one slide on that because everybody seems to think it's obvious that NAFTA had a negative effect on workers. And this is just the critics, you know when Obama comes out with his TPP and tries to explain why it's good he says, well of course it's not like NAFTA. No one wants to defend NAFTA. But we economists are happy to defend NAFTA and let me just point out some things that might surprise you.

If NAFTA were actually so bad for workers, you might think it would show up in the data, things like incomes of workers and wages. Here are some statistics [crosstalk 00:23:42] That's not the statistic, that's the claim. Keep it in mind. It's definitely true that inequality has increased and real wages, particularly for white males, unskilled by the way. I don't like the word unskilled, it means without completing college but, you know, there's a lot of people that haven't completed college who have a lot of skills that can fix my plumbing and stuff that I couldn't do at all. If I were a waiter I'd fail the job. There are a lot of skills but just not what you get in a great college education like Swarthmore.

Anyway, there has been this vague increase in inequalities, stagnation in the wages of unskilled workers, since 1980 and especially since the turn of the century. But there is a period of five or six years in a row that is the exception. Guess when that period is? It's the five or six years after NAFTA went into effect. That period featured the most job creation we've had in the last 40 years although we're getting close to it again now. From 1996 to 2000, GDP growth averaged 4.3% which is the highest it's been since the 70s. Productivity growth two and a half percent. Unemployment got down below 4% by the end of 2000. It is the only period since the 1970s when workers shared fully in the gains. As productivity growth at two and a half percent, workers should get a lot of that. They should get much more than half of it. They did. Real compensation per hour rose 2.2% during this period. Median family income rose strongly from $26,400 in 1993 step by step by step to $35,000 in 2000.

We just got a good number three weeks ago about median family income rising 5% last year but that's the only good number we've had in 20 years except for these good numbers. And the poverty rate declined every year from 1993, 94, 95, by the way NAFTA went into effect in 94, from 33% to 22 1/2% in 2000. Now, I'm not saying that NAFTA gave us all those benefits. The effects of NAFTA good or bad weren't that big. There were lots of other things going on. But this idea that it's so obvious that NAFTA hurt workers is based on nothing as far as workers in the average. Of course individual workers in factories and so on.

[00:26:31]

Now there is this wave of recent research that is causing people to rethink some things and this is David Archer, David Dore, and Gordon Hanson, a number of papers, and they find that a sizable portion of the decrease in manufacturing jobs, I'll show you a graph later for the decreasing manufacturing jobs. Can indeed be associated with imports and they're talking about China. Trump video, China China China ... I think the evidence is pretty strong that when China joined the world trading system and there's a lot of exports of cheap manufacturers, that was big enough that that did have an effect on a lot of industries in the US and elsewhere.

It had an effect in that, can explain part of the US loss of manufacturing jobs. Nowhere near as much as productivity can explain, but still their other finding is that employment and income in areas that were hit by these job loses can stay depressed for a much longer time. Workers don't rapidly move into some big IT job, some great export job, which apparently caught some people by surprise. I do want to emphasize that even though we've probably gained from this research greater insight, a greater appreciation for numbers to go with the images of factory towns, steel towns, or whatever, that have been depressed for a long time.

People are reading a bit too much into it. These authors don't say anything about export jobs that were created. It could be that just as many jobs as were destroyed by imports and manufacturing were created in other cities by the opportunity to export. We know that total employment went up during this period. So it's not a final verdict by any means.

So no question that inequality has gone up. If you can't read it, the red line is the top 1%. The yellow, orange line is the top 0.5%. Inequality came down down down between the 1920s and the 1960s and then it's been going up up up since 1980 and we're basically back to where we were in the roaring 20s which is pretty high inequality. So it's clearly true and various other measures which show a similar thing again, just within the US and to some extent, within individual European countries.

Why has inequality risen in the US? Trade probably does play a role so in this Stolper–Samuelson story, especially with this evidence from David Ator and his co-authors. It's really associated with when China joined the system. But I'm gonna give you a list of seven other factors and I'm not scraping the bottom of the barrel. Each of these are really important factors it seems to me, that also contribute to the increase in inequality.

So number two, which most economists have thought was probably the most important, is technological change. That you have to have, in the IT world, you have to have skills and jobs of unskilled factory workers have gotten automated away. I'm gonna list it here and then I've got approximately one slide on each of these to elaborate a little bit but here's the summary. This says technological change is steadily raising the demand for skilled workers and that bids up the wages of skilled workers relative to unskilled workers. Education is our best tool, best measure of the supply of workers and education, number of years of schooling grows. US was sort of the head of the rest of the world on this with high school in the late 19th century and for years, but it has slowed down. So demand is not keeping up ... Supply of skilled workers is measured, let's say, by college education or years of higher education, is not keeping up with demand. So these two sort of go together.  There's a book by Larry Katz and Claudia Goldin that document that. It's like a race between the supply and demand.

Here's a different one, winner take all labor markets. This sort of explains the gap between those with college and those without. That's not much help for the upper 1% or the upper tenth of 1%. If we want to understand why the upper one tenth of 1% get such a huge fraction of income, that won't really help. Winner take all labor markets will help with that.

Assortive bating is number five. Reduced competition, among corporations, more monopoly power, higher rents, is number six. Excessive executive compensation, you can put excessive in quotes, I'm never going to exactly define it, and finance I mean also. And hey, I'll put in Thomas Picketty because he's made this whole issue so front and center and done so well with his book, which I actually did read from beginning to end, whole thing. And he says it's wealth accumulation from generation to generation because the rate of return on capital is greater than other income and so it very gradually over time, over decades, generations, centuries, the wealthy get wealthier.

Let me go through, in case any of these are mysterious and I haven't defined these two, really. Let me give you a little elaboration on ... Number two, the best way of telling about the supposed gap between skilled and unskilled workers is the wage premium. So I hope you all will be pleased or your parents will be pleased to know that your expensive Swarthmore education is gonna pay off. The premium to college earnings has gone up up up since 1980 and so that has gotta be almost by definition, a major determinate explanation for the gap between the upper whatever, 40% and the lower 60%.

Half of that is technology raising the demand for skilled labor. The other half is the supply. For years, as I mentioned, throughout most of the 20th century, the number, mean years of schooling completed by age 30 was steadily rising and then this trend fell abruptly over the last 30 years. That is consistent with what we saw that the supply of skilled workers is not keeping up with the demand which would explain the higher premium, higher wage that workers who have higher education seem to command.

Number four, winner take all. What's the winner take all labor market? Forty years ago, if you were the best surgeon in New York, nobody knew it. No one even tried to measure who was the best surgeon in New York and if they did, no one ... it didn't have any effect. And so the best surgeon in New York did not earn a higher salary than the 100th best surgeon or the 1,000th best surgeon, I presume, just making up this example. But then we started to have lists of who's the best surgeon in New York. It was like each city has one. New York magazine would have these things and the local newspaper and then, of course, the internet has really done this. So now everybody has some idea of who's the best surgeon and they earn a lot more than the 100th best surgeon. And this is true in every market. It's true among CEOs and it's probably true among plumbers thanks to all the online rating services.

Within every profession, the ones who are the very best earn more than the average. Entertainers, I looked up who is the most paid celebrity last year and it was Taylor Swift according to them $170 million making her the world's highest paid celebrity. I don't think relative to the average entertainer, I think that's a much higher premium than you would have had 40 or 50 years ago. There's only one Taylor Swift and the number of fans has now ...

Assortative bating, do you know what that is? So actually I wanted to get a picture of Mad Men to illustrate this. So I'm .... just like the last one with the surgeon, I'm telling a stylized version rather than trying to use any economic theory or whatever. Stylized version is much better. So crudely put, higher educated and highly paid male professionals used to marry their secretaries so I'm gonna be the advertising men in Mad Men. But now they're more likely, and now I'm sort of equalizing income because the secretary can necessarily come from the same social, educational income class. But now, Swarthmore is completely excluded from this of course, now everybody is more likely to marry people within their own social class. A highly educated, high income men are more likely to marry highly educated, high paid women, relatively, not quite caught up yet.  When the couple passes the advantages onto their children, so that's one explanation to add to the list and I think there's something to it.

Number six, well this is a graph of the labor share of income, which we used to think was pretty steep but has had a downward trend and in particular, has kind of plunged since the turn of the century. How much of total income goes to workers and wages and salaries versus capital? A lot of these graphs, if you can't read the little credit at the bottom, are stolen from a speech that Jason Firman gave, he's chairman of the council of economic advisors up on his TEA chairman. I was on Clinton's council of economic advisors and this is one of ... I've had the experience now a couple of times, but someone who used to work with me now outranking me. But anyway, he gave a great, he's great on everything, capsule review of all this in a speech just a few days ago.

He's using this to argue that one piece of evidence among many others, that corporations have got monopoly rents and that are deviating from perfect competition, getting a larger share of income.

This is the "excessive" compensation. This is a bar chart saying, who exactly are these people earning the top 1% and they're pretty heavily ... the two big categories by far are executives and the CEOs or other top executives, and financial professionals. I think if we had the top tenth of a percent it would be even more striking, although I have a feeling some major sports figures and entertainers might figure into that one.

Now what does excessive mean? It's no question that the amounts of money we're talking about just sort of seem excessive, seem obscene. But there is a question whether they're earning a marginal product. The argument is the CEO if his or her decision, business decision, really is worth millions or hundreds of millions, billions of dollars to the company, then they might be worth their salary. Surprisingly, there are some studies that say this is the case. The alternative is that this is another example of rents, that executives have captured the compensation committee and they're all buddies. Actually, in that sense it doesn't matter. I'm not gonna choose how much it is, excessive or not, it's just really high and obviously contributing a lot towards inequality.

And particularly when you hear that executives or the board want their CEO to be paid more than the average, and everybody wants to be paid more than the average, which of course they can't be, and so it's this race upward, that to me is like they're begging to be taxed because they can still do the game of competing with each other as to who gets paid more and so it could be the market outcome. And they're still just begging for higher income tax rates. They can still do their competition and government can get some of the revenue.

Supporting Piketty, number eight, wealth concentration is also gone back up along with income concentration. On the other hand, Piketty does a lot of very careful work, a lot of numbers including some of these other theories, but in terms of the book, his major argument is the one about capital accumulation from generation to generation. And he makes a persuasive case by citing Jane Austen and Balzac that describes the first half of the 19th century in Britain and France quite well. I'm not sure it yet describes us because the inequality within labor and also the inequality within capital, seems to be much bigger than this phenomenon of, early in Europe, of any shift between the shares of capital versus labor. We're nevertheless, I'm happy to add Piketty to the list of eight.

So what weights do we place on each of these eight factors in explaining increased inequality? And, by the way, I've got some appendices slides, which I probably won't get to unless it comes up in question, graphs that might answer questions. But one of them is not a graph, it's just for each of these eight, one can imagine a policy response that's particularly targeted at that problem. For example, for the executive compensation, you'd want to have a rule like some other countries do that the CEO can't also be the chairman of the board and you want to have say on pay, which we did get after our global financial crisis even though it hasn't had much effect, and other reforms to reign in insider, buddy buddy levels of compensation. Make sure you put outsiders on the compensation board and so on, and each of the eight has their targeted factors.

I don't know what is, since this isn't, by the way, the field where I do my own research. I'm not obligated to have an estimate or to know what weight to put on these eight different factors. My guess is they're all of them, have some truth. All of them merit some weight. All eight of them, here it is again to remind you: trade, technology, education, winner take all, assortative bating, non competitive rents, "excessive" executive compensation, and Piketty's argument about wealth accumulation. I think there's probably truth to all eight of them. Now you might say, well how on earth, we must be able to decide which one is the go in or what the relative weights are in order to decide what policies to do to respond to the inequality.

Well, you would think so, but I'm actually going to say no, that you don't have to. Let me backtrack now and tell you about the second fundamental welfare theorem of economics which I'm gonna phrase a little loosely. When individuals are free to engage in trade, this size of the economic pie increases enough that the winners could, in theory, compensate the losers. In which case, everyone would be better off. So then you can make it Pareto superior if you have these transfers, these ways of compensating the losers.

Now skeptics of globalization are, among them down the hall from me Danny Rodgrick, he understands this completely, and he points out, reasonably, that the compensation is hypothetical. It doesn't actually take place in practice. So the skeptics then suggest that we should take the political failure to compensate the losers as given and so try to roll back globalization. Well, I would suggest that the better strategy is the alternative, the reverse strategy. Take globalization as given and instead work on trying to help those left behind and I want to convince you that this is not just because I've a soft spot for free trade and the things that Clair Wilcox fought for, but that is a practical, hard-headed strategy. This one makes sense and the other one doesn't.

This is a sensible strategy, why? For one thing, it would be difficult to reverse globalization, even if we wanted to. Let's leave aside the fact that if you put on 40% tariffs against trade with China and other countries, that there would be a trade war that we would lose exports of similar amount and that that would hit overall economic growth and the geo-political damage would be pretty bad and by the way, I can tell you all about TPP, about the economic and legal effects, but most of us, what we really think is most critical, is the geo-political. If Congress does not approve TPP, the countries of Asia are gonna conclude, and this won't be their first piece of evidence, that the US no longer has the will or desire to stay engaged in Asia and they're gonna look around like for who else can they get close to, and it's gonna be China.

But anyway, let's leave all that aside. Nothing a president does would be likely to bring trade back to the levels of 60 years ago. Let me be  clear, the president could do a lot, actually, there's certain things where the president needs congressional approval and certain things where he doesn't. A president who really wanted to launch a trade war could do a lot of damage and could ... We did it in the 30s, we turned the clock back on globalization in the 30s with the Smoot-Hawely Tariff, then emulated by other countries and world trade collapsed and it created the great depression.

So it can happen. But could we really go back to the level of 60 years ago? No, we certainly can't go back to the number of manufacturing jobs we had back then. Manufacturing employment has declined. This is the percentage of employment in US manufacturing. This is just from 1970 to 2012, but it goes back farther. There's no way ... So it's 25% in 1970 and 32% I think in 1950. Most of this is due to increased productivity. Productivity means producing more automobiles or steel per worker. Basically a good thing, but it also means hiring fewer workers per automobile. I don't think we're just gonna blow up the technological progress we've had since then and go back to requiring more workers to produce an automobile. Even if you could put out trade barriers to keep out Asian and European automobiles which would have to be like, a few hundred percent trade barrier, I don't know that consumers would put up with that to have American cars so expensive.

But even then, you're not gonna go back to anywhere near the number of workers you had back then because of the productivity growth in the meantime. 32% of national total down to 10% in 2010. That's all manufacturing. The really high wage jobs were autos and steel in the 50s. Those jobs, it's true that if we lived in Michigan or Ohio or Pennsylvania, you don't need a college degree if you were lucky enough, and it helped to be white by the way, if you were lucky enough to get a job in a steel mill or an auto plant. You were in pretty good shape. You earned roughly double what other American workers earned and vastly more than workers in other countries. And that's true and that's gone, no question about it.

Since 1960, the number of jobs in sectors like steel and autos and textiles and apparel, have fallen by a half. Meanwhile, the number of jobs in healthcare, which already was greater, even then was greater absolutely, but that's increased six fold. We're always creating jobs in some sectors and losing jobs in other sectors. International trade was one factor that helped put an end to those high paying jobs but another one was productivity growth.

And, by the way, another one was relocation from the north to the south. Long before there was competition with China, there was competition with the south. So, for example, textiles and apparel and shoes, which were big industries in New England where I'm, where they in the 19th century, those firms mostly moved to the south long before we had to deal with the competition from Asia.

Now we have to mention that part of this process was we now have cars that are far more reliable, break down far less often, far more fuel efficient, and far more affordable. Not just because we're importing them directly, but because Detroit or the auto industry, did finally learn to compete and today's American auto industry is a competitor auto industry and it matches foreigners on price and reliability and the rest of it. So it's extra hard to imagine going back to 1950.

So remember I said second welfare theorem says any change probably has winners and losers but it allows free exchange and competition, increases the total size of the pie, and it does it by enough that you can compensate the losers and everyone comes out ahead. What does it mean to compensate the losers? Don't we need to know very specifically about who the losers are and what the cause is before addressing it? Well, I would say, no. I mean first, to the extent of trade I listed eight factors. To the extent of trade we have a program to specifically help those who lose their jobs due to trade, it's called trade adjustment assistance. Democrats are always trying to expand it. The Republicans are always trying to cut it. It's not my favorite program. It has a bit of ... It sounds like you're going to adjust workers into new jobs but it really is what the unions call burial insurance. It's money and it doesn't really, isn't that effective particularly if they were in a very high paying job before, it doesn't transition them, typically, into jobs that pay as much.

So that's one problem with it. Another problem is why should we only help the number of workers who have identifiably lost their jobs due to trade agreements, which is actually pretty small number of workers where you can make that identification. There's far more workers who have lost their jobs due to technology or something else. Why would there be a stronger moral case in helping those who lose their job due to trade than those who lose their jobs due to technology or something else. So one of the things I would do about trade adjustment assistance would make it more general, well that's sort of like unemployment insurance I guess. But the other is, there are these points about incentives and these programs have the undesirable feature sometimes that they discourage you from going out and getting a new job because you lose the trade adjustment assistance and lose the unemployment insurance.

But we want to do it in a way that still rewards work. Sensible policies to help those, I'm getting to an important slide now, to those left behind include, here's my list. And I'm leaving some off that the answer isn't as clear. I'm putting up ones that to my mind are just really crystal clear. Maybe some of the members of the Swarthmore economics department can tell me if they agree that this represents something of a consensus among economists.

Number one, wage insurance which president Obama proposed in the State of the Union message. So wage insurance says that when you lose your job you get some money but then if you take a new job, you don't lose the benefits unless the new job is at the same wage as the old one. If you have to take a pay cut of 30% to take the new job, the wage insurance keeps paying you the difference. So it compensates them and it gets them into new jobs which is what we want. That's wage insurance.

Number two, obvious, more progressive income tax system, which is central to Hillary Clinton's plan. Let's see, I don't know, we might have some Republicans in the audience so I will confess, I'm a Democrat and if you haven't figured that out yet, you will be pretty clear by the end of this slide. More progressive income tax system. So that means, very specifically, expand the earned income tax credit, which low income workers get as an extra benefit. It basically Milton Friedman's negative income tax, and raise marginal tax rate on people in the upper 1% or the upper tenth of a percent. And I would say, also make the payroll tax more progressive so let low income workers not pay payroll tax. Currently backs out on the payroll tax at, what is it ...

Speaker 4: 1185

Jeff: 1185. So after that [crosstalk 00:54:40] he's not paying additional. It changes every year. But it seems like you should have to still pay somewhat higher payroll taxes if you go into the upper 1%. All of these are things that Democrats in general, Clinton in particular, are in favor of.

Universal health insurance, such as Obamacare but we have to go beyond it. That's like the biggest thing you can do, maybe single policy, to help the people who feel they've been, and correctly have been left behind by modern trends, whether it's globalization or technology or back luck or whatever.

Universal preschool education, high quality preschool education. Again, that's to put people on ... That deals with the families and the assortative bating and the fact that you breaking into different social classes. A kid is, by the time he or she starts elementary school, let alone college, it's already been put on a more beneficial path.

Number five, heavy infrastructure investment spending. We should've been doing this, we started with the fiscal stimulus of 2009 but then that got reversed after two years, basically when Democrats lost the Congress. We need it in terms of crumbling bridges and out of date airports and all the rest of it. It's not profligate, it's not borrowing from the future because we're gonna have to do this stuff anyway. It's all deferred maintenance. So it's a form of saving and investment and it would put to work the construction workers who lost their jobs when the housing bubble burst.

Financial regulation such as the Dodd Frank bill and more of the other things they would've liked to do that they had to take out in order to get the thing passed that got chipped away over time. I mean one that I like, Hillary Clinton's proposal to impose a risk fee on large financial institutions because they pose a systemic risk to the rest of us. So rather than just banning large institutions, say you can't go above a certain size, you put a tax on them and that will encourage them to stay small but if for some economic reason they really need to be big, they'll be big but at least they'll be paying. It's like tax inclusion externalities, you'll be taxing the externality and while you're at it, raising money for the treasury.

The other party consistently tries to block these measures. Obamacare actually made it through. But they sometimes succeed in blocking them and they sometimes don't. Some of them, I think if Hillary gets in, and if she gets enough votes in Congress, there's definitely a stroke to cooperate with Republicans on expanding the earned income tax credit and our infrastructure investment spending. The other ones, I don't know. We need really big election outcome. I'm not going to tell you who to vote for but the thing is, we're actually kind of lucky.

So I started off by mentioning Brexit. The Brits are in a really bad situation now because let's say you believe strongly after thinking it all over that you're in favor of Brexit or against Brexit. What do you do? You're not gonna have another referendum. They can have an election but both parties are thoroughly mixed up, thoroughly muddled. When the next election is, people aren't gonna get a chance to vote, I'm in favor of this vision versus I'm in favor of that vision. They're not gonna get to vote pro globalization or anti globalization, big state or small state, the parties are too muddled. We have a clear difference in the parties and it's not over globalization, it's over whether they want to take these steps to help compensate the losers. Because one party does and the other party doesn't and even though this election is completely unprecedented in many respects, it's not different in that respect.

There is the other side of the story, the argument is that everything I'm talking about is big government, government overreach and that's bad for growth. So I don't want to say that there aren't two valid sides to the argument or two debatable sides to the argument, but it is basically the same argument we've been having all along and if you're concerned about trade, is that we're not compensating the losers. All you have to do is choose the party that wants to compensate the losers and then we're done.