Russ Roberts

George Borjas on Immigration and We Wanted Workers

EconTalk Episode with George Borjas
Hosted by Russ Roberts
Sam Quinones on Heroin, the Op... When is a Worker More than a W...

WeWantedWorkers_cover.jpeg George Borjas of Harvard University and author of We Wanted Workers talks with EconTalk host Russ Roberts about immigration and the challenges of measuring the impact of increased immigration on American workers and consumers. The discussion also looks at the cultural impact of immigration and what immigration in the past can tell us about immigration today.

Size:30.0 MB
Right-click or Option-click, and select "Save Link/Target As MP3.

Readings and Links related to this podcast episode

Related Readings
This week's guest: This week's focus: Additional ideas and people mentioned in this podcast episode: A few more readings and background resources: A few more EconTalk podcast episodes:


Podcast Episode Highlights

Intro. [Recording date: December 20, 2016.]

Russ Roberts: I want to remind listeners to go to and in the upper left-hand corner you can vote for your favorite episodes of 2016 and share other feedback, which I really appreciate. Today's guest is the economist George Borjas,... Today we'll be discussing his latest book, We Wanted Workers: Unraveling the Immigration Narrative. George, welcome to EconTalk.

George Borjas: Thank you so much, Russ. I don't know if you realize this but it's almost 40 years since the first time we ever met.

Russ Roberts: Yeah. I know. I do realize it.

George Borjas: [?] cool. If somebody had told me that we'd be here today talking about this, 40 years ago, I would have said, 'Not in[?] your dreams.' But somehow we're here, okay?

Russ Roberts: Yeah. I don't like to think about it because it's so long ago. The other part, of course, that's hard to imagine and couldn't have been imagined is that we'd be talking not via phone but via Skype; it would be recorded and it would be available to anyone around the world at zero direct cost--many good things in the last 40 years, at least for us.

George Borjas: I know. That's for sure.

Russ Roberts: So, let's start with the title of your book, which is clever: We Wanted Workers. Where does that title come from?

George Borjas: Okay. It comes from a quote, from a statement by Max Frisch, who was a Swiss novelist, essayist. And he was reflecting on the immigrants, on the guest workers really, that Europe tried to import during the 1950s and 1960s, particularly Germany but many other European countries as well. And those guest workers came in and they clearly contributed to the economic miracle. For example, the post-War Economic Miracle in Germany, right? But Max Frisch was looking at it from afar, and he basically said, looking at it in a more general sort of way, 'We wanted workers, but we got people instead.' And the reason that I think that's sort of one of the themes that I stress in the book is that even though I'm an economist, I tend to be a little dissatisfied with the very mechanical way in which economists view immigration. The typical--let me call it the 'economistic' way of looking at immigration looks at them as a bunch of robotic workers that you can basically move from place to place as the need arises. And that, I think, is sort of what Max Frisch is referring to when he said, 'We wanted workers.' And, it's true: Immigrants played a role. Okay? But the fact is that immigrants are human beings as well. And people make decisions. And people make decisions that are based on what is best for them. They are rational human beings, like you and I, right? And those decisions might or might not be precisely what they are receiving [?] in mind. And they may or may not increase the benefits; or they might actually create some harm on the way. The point is that, the fact that immigrants are something beyond workers, and that they play a role that is not just this robotic kind of role of moving from factory to factory, means that we have to look at the impact of immigration in a much broader framework. We have to take immigrant decision-making into account, in particular. And that, I think, was one of the things that motivated me. It's really one of the themes that motivated me as I was trying to write this book in an easy-to-explain way for a general reader: that we have to look as immigration not just--let me put it a different way. A lot of people make the analogy between immigration and trade. And in fact, immigration is like trade, to some extent. When we import a widget, a point that I make in the book, it's sort of like importing the raw labor creates a widget, only that we importing the raw labor would allow us to create the widget domestically, right? But the fact of the matter is that when widgets break down, we can throw them out. When immigrants break down or get sick or things happen, we have a responsibility. And that is one crucial way in which immigration is not like trade. And that's sort of the thematic content, what that statement is trying to capture in the book.

Russ Roberts: Yeah, well I think--we'll come back probably to that analogy between trade and immigration--which, of course, they are both related to open borders. If you have open borders, you can choose to have open borders for goods; you can choose to have open borders for capital; you can choose to have open borders for people. And the impact of those different cases have some similarities. And as you point out, they also have some differences. So, I think it's interesting how hard it is for people to think about those clearly, even s economist. Even myself.


Russ Roberts: Let's start with a standard argument you hear, that you attack in the book--correctly, I believe. People often say, 'Well, we have to have immigrants because if we don't, there will be no one to do the--certain types of jobs.' Americans, quote, 'Americans won't do these jobs'--whether it's mowing lawns, painting, basically construction, agricultural work in particular which--these are all areas that have a lot of immigration workers. What's wrong with that argument? Is that correct?

George Borjas: Yeah, the usual [?], the immigrants do jobs [?] don't want to do. And the problem with that basic argument is that [?] ignores the role of markets. And the role of incentives. I think the correct statement is really, immigrants do jobs [?] don't want to do, at the going wage. The presence or absence of immigrants basically changes the market. And markets react to either, to people coming in or people going out. The way I start this in the book, as you know, is by describing--it's a little anecdote of the firm in Georgia back in 2006 where at the time the Bush Administration was trying to look serious about enforcing immigration policy regarding illegal immigration. And they actually conducted quite a few raids on firms that hired illegal immigrants at the time. And one of the raids happened to be on a firm called Crider in Geogia, a chicken processing plant. And the nice thing about this example is that, Crider did precisely what we would teach a profit-maximizing, rational firm to do. Crider basically woke up after Labor Day weekend finding that 3/4 of its workforce had been basically--had disappeared--because of the raid. And, you know, what does Crider to at that point? Well, they put an ad in the paper. And the ad they put in the paper is: Increased wages. We want workers at increased wages. You know, less supply; higher wages. So, to me, it's not surprising that markets respond in the way that common sense tells you they respond. What I always really found very puzzling about the immigration context and the supply-and-demand issue is that most of us have no trouble, you know, given our training we have no trouble saying supply and demand--it's a very nice, unifying tool through which to explain why prices go up and down. And somehow, that idea tends to disappear from a lot of people's minds--a lot of economists' minds--when it comes to immigration. And I always found it a little puzzling. I mean, I don't quite know why there's a resistance to accepting that prices will change when supply shifts. But in the case of immigration, you tend to see the resistance quite often.

Russ Roberts: Well, I'd be in that group, to some extent. So, let's get into that, because I think that's a central issue that we should talk about. First, on the Crider example, in Georgia, it's not surprising that overnight, finding that you are out of workers you are going to pay more. The more interesting question would be 6 months to a year and 2 years later. Was that just an emergency move on their part or was that a response to the fact that there was suddenly a smaller supply of workers of low-skilled workers to that particular area of the country, and we normally wouldn't expect the loss of 100 workers in a town or a city or a state or a country to have a big impact on wages? It's a small change. But that had a big change. It was observable, partly because it was--they had a very urgent short-run demand for workers. But the more general question I guess is the following. And I want to phrase it--it's hard to phrase, because I'm a big fan of supply and demand. Of course, the question is: Supply and demand of what? Labor, with a capital "L"? Low-skilled labor? High-skilled labor? Labor that has a particular kind of skill? Certainly all those kind of things are going to have impacts on how you use supply and demand and whether you use it carefully. I guess the question I would have is that, when we talk about labor, generally--if we said, for example, if we try to imagine the kind of experiment, which is the kind of experiment you do in your work: What if the United States had not liberalized immigration in the last 25 years? Which it has--the last 30 years, I guess--post 1990 is the key when it started to become more liberal, or is it a little before that. But I think post-1990 has been a relatively liberal era for open borders. Do you want to think about that as an L--a supply and demand for Labor? Or do you worry about the fact that it depends on what kind of workers they are; it depends on whether they are complementary or substitutes? Does it depend on what parts of the country they go to. It just seems like a little more complicated than just supply and demand for labor.

George Borjas: Oh, I completely agree, Russ. Okay? One of the things in my work over the last 10, 15, 20 years has been that you really have to match the skills of immigrants to the skills of natives, if you want to sort of detect an impact. And a lot of the literature has gone the wrong way, basically because, you know, the matching is not quite proper. I mean, I give you an example of this in some work that I've done on mathematicians, which I sort of talk a little bit about in the book. You know, the Soviet Union collapsed in the early 1990s, 1992 or so, right? And a bunch of mathematicians came to the United States. Now, from an outside perspective, you'd see the supply of mathematicians increasing, right? But it turned out that the mathematicians who came to the United States happened to specialize in particular topics. So, if you were to look at the average mathematician in the United States, what actually worked on very different kinds of topics, you would see actually very little impact. But if you look at the mathematicians in the United States who specialized in the topics that mathematicians in the Soviet Union had specialized in, then you detect a negative impact of supply and demand--the sort of the basic [?] would give you. So, you know, I am very appreciative of the fact that you really have to very carefully match who it is that the immigrants are going to hit first. Because, as you said, there's complementarities involved as well. For better or worse--and I think it's for worse, actually--much of the literature has focused way too much on trying to measure the own[?] impact--in other words, try to detect that when the group of a particular skill, the number of that group goes up, the weight might go down--in the short run. And again, I'm not doing [?] distinction--short run versus long run. In the short run, the theory is very clear about that. And in the theory it's also clear that in the long run, what's going to happen is capital will adjust and with constant return to[?] return to world, the average wage will return to what it was pre-immigration. But that doesn't mean that even in the long run there are not distributional impacts. Those groups that had a relatively larger supply shock in terms of more workers will tend to be a little worse off than those groups that had a smaller supply shock. So, all these things come into the picture when one tries to estimate the impact. And that is partly the reason why the literature is actually very confusing. Different people are trying to estimate different things. Short run, long run confusion enter into the picture. And again, from the point of theory, the short-run, long-run distinction is very clear. From the point of data, it's far less clear. Because we don't really know what the long run will be in a labor market hit by a supply shock. Will it be a day? Will it be a year? Ten years? We don't really know that. No one has really ever analyzed the dynamics of what happens to labor markets as a result of a supply shock. And then, we also have to deal into the skill issue, which means that: How do you find skill groups? Could be very problematic. Because, you know, at some point the [?] nature to defining a skill group. Like, for example, in my work I've defined it in terms of location and age, basically, in [?] of what I've done. In all, there is education and there is education, right?

Russ Roberts: Yeah. It's a good place to start--

George Borjas: Yeah, go ahead.

Russ Roberts: It's a good place to start. But obviously two people, each with, say, 4 years of education, a college degree, may have very different skills. And one may compete with one kind of worker but not at all with another type. So--

George Borjas: Exactly. That's what I'm trying to point out.


Russ Roberts: I want to take this--let's take this point more generally. Because I think it comes up a lot in lots of different areas of labor economics. It drives me crazy when people say, 'Seattle passed a living wage ordinance and employment in Seattle is doing fine.' Well, most people in Seattle are not going to be affected by the minimum wage. It's only going to be the people down toward the lower end of the skill distribution. And you have to look at those workers to see if there is any impact. Because otherwise, the overall impact is going to get lost in the noise of--particularly if there is growth in the Seattle area, which there is; it's doing very well. So it's hard to know whether there is any effect. But if you are trying to measure it, you certainly don't want to look at, say, employment of all workers. Because the theory doesn't expect that to change very much, if only a small proportion of workers are affected by the minimum wage. And I think, similarly, if more low-skilled workers come in to the United States, you and I are not going to be very affected on the wage side. We're going to be affected on the price side. It's going to be probably good for us, actually, because many of the things we buy might get cheaper as a result. But, if hundreds of thousands of economists came in from the Soviet Union, that could affect our wage. And certainly if it had happened 20 years ago, would it be affecting our wage today? As long as they could teach what we teach, roughly, research what we research, approximately. And--the basic point is undeniable, that certainly you have to be careful. One of the things that comes through in your book very clearly is how hard it is to make those kind of measurements, given that our data are imperfect--as you point out, we don't have, people don't walk around with a sign on them saying 'I'm a substitute for such-and-such kind of worker.' You have to, inevitably, make assumptions. You made one just a minute ago in passing. You mentioned the phrase 'constant returns' as a certain assumption about how capital and labor combine. And it may be a decent approximation. It may not be for a particular city or state or a time period when you are trying to measure the impact.

George Borjas: That's exactly right. Actually, there's two things I want to say to what you just said. One is the role of assumptions is huge. And we're not appreciate[?] that. We don't emphasize that enough when we discuss the labor market and immigration, sort of what the general findings are. And, like, this whole literature--a lot of people, and I sort of cite in the book some examples, claim that the data have shown that on average the immigration has no impact on the long-run wages. Right? And that is true--that is what the data show. What the discussion that is unusually emphasized is that that has been assumed by the constant returns assumption. The fact that on the average, the impact of immigration, on the average, is zero, has nothing to do with data. It's all been built in by the fact that the model underlying this analysis that assumes constant returns. And the other point you raised, which is, again, extremely relevant in this context, is the minimum wage example. You raised it in the context of, you know, you have to really match what is being affected by the minimum wage or not, right? I want to raise it in a different context, which is a lot of people make what I, what is intent of pure theory, completely contradictory arguments when it comes to immigration and when it comes to the minimum wage. Somehow, you see people have no qualms whatsoever saying immigration in the short run doesn't have an effect on wages. And at the same time they say that increasing the minimum wage has no effect on employment--in the short run. As you know, those two things are completely contradictory. In the context of the minimum wage they are basically claiming that the labor market, the labor demand curve is perfectly inelastic. In the context of immigration they are claiming that labor demand curve in the short run is perfectly elastic. You know, one of those things, and probably both are wrong. Labor demand curve should probably not in that, in either extreme. And it's really sort of--I've always been fascinated by the intellectual contradiction, in sort of claiming these two things simultaneously.


Russ Roberts: Well, here's the other question I have on this; and then we'll move to some other examples and try to summarize some of the empirical work and get to this--I think we'll want to re-emphasize this point about assumptions because I think it's very clear. So, workers aren't widgets. They are not the same. Importing workers is not the same as importing the goods that workers overseas might produce, partly because workers bring their cultural habits. And that's obviously an issue that you talk about in the book. But they also bring the fact that they want to buy things--

George Borjas: But--

Russ Roberts: So, which makes them not like a widget. So, in general, if you ask me, 'Is the increase in the U.S. population between, say, 1900 and 2000, has that been good or bad for workers?' We wouldn't want to use a supply-and-demand model, because supply-and-demand model--it's a technical term; it's a partial equilibrium. And partial equilibrium is probably not the right way to simply think about--it simplifies what's going on when we think about a growing population. So, that's the problem I have when I think about an increase, say, in immigration. To me, it's a lot like an increase in population. It's true you bring in adults rather than infants--usually. But obviously it's a little more complicated than just saying, 'Well, there's a supply shock.' I'm willing to accept the fact that in the short run, that can certainly be the case where workers with skills who are close to the workers who are coming in, that it's going to be hard on them. Of course, they already have jobs; their wages, a lot of them, their wages are not going to go down instantaneously with the opportunities to hire additional workers that come with foreign immigration. So, I'm not quite: Explain to me how the supply and demand framework should be used in, say, a case like population or immigration. And why it's different. It seems to me it's very different from the minimum wage.

George Borjas: No, I agree with you, okay? In the long run, it's more of a group[?]. In the long run the real question is economic growth. What does having more people do to economic growth? Right?

Russ Roberts: Yep.

George Borjas: And the question then becomes, what do you want to assume is the underlying technology?

Russ Roberts: Say that again? What do you want--?

George Borjas: The question becomes: What are you going to assume about the underlying technology.

Russ Roberts: Correct.

George Borjas: Okay? And if you assume constant returns, then not much will happen in terms of per capita income, in the long run.

Russ Roberts: And therefore, how do you explain--that's probably not a good starting place.

George Borjas: No, no, but therefore the question is: Do immigrants completely replicate what we have now? Or is the skill distribution of immigrants different from what we had originally? And then it will depend on factor proportions, I think.


Russ Roberts: So, I'm going to disagree with--let me agree with half of that and then let me disagree with the other half.

George Borjas: Okay.

Russ Roberts: Even though it's maybe only one half there. It's certainly true--let's suppose we take a country like the United States at a point in time, and now we're going to increase the population via immigration, by 20%. A huge increase. Big, big increase.

George Borjas: Actually double it--

Russ Roberts: Great. Let's double it. And I'm going to think of two or three different cases. In case Number 1, the skill set--and I'm ignoring the cultural side of this for now, obviously; we're only going to look at the economic impact, the financial/monetary impact on wages and markets. So, obviously, if everybody is the same--excuse me--if the distribution of the new people mimics, is almost exactly the same as the current distribution, we'll have doubled everything. We'll have doubled all the low-skilled workers, all the high-skilled workers. And it's the case--

George Borjas: And the long-run capital should double as well.

Russ Roberts: Because?

George Borjas: Because--the, you know, in a global--yes, you have to say, now you have to say, well, the rate of return to capital will increase in the United States as a result of initial people coming in. Right? That will bring in initial capital somehow.

Russ Roberts: Maybe from overseas. Maybe people will save more.

George Borjas: Exactly--

Russ Roberts: Okay. So that's the case, and maybe we have another Grand U.S. next to the old U.S. So, it's twice as big; it's got twice as much capital. But we don't know if it has twice as much. But it has more capital and more labor.

George Borjas: Yeah.

Russ Roberts: Next question: The second case would be--

George Borjas: But the point is, in that scenario, per capita income doesn't really change.

Russ Roberts: Well, we don't know that exactly.

George Borjas: I think we do.

Russ Roberts: I don't think we do. The reason I'd say that is that I actually think there's, via Adam Smith, the division of labor is limited by the extent of the market.

George Borjas: Oh, but I am thinking of it, very specific production function here.

Russ Roberts: I know; and I don't think that way. I don't find that--

George Borjas: But that goes back to the beginning of what I said: It all depends on what you assume about the [?]

Russ Roberts: That's fine.


Russ Roberts: So let's leave that alone for the moment. Let's accept the fact that different assumptions about how people combine could affect the final conclusion. But it certainly wouldn't be the case--and you wouldn't argue the case--that in the long run all those extra workers would lower wages for the people who already live here. It could in the short run. It could because--for a lot of reasons. And it certainly could measure--well, anyway, that's Case One. Case Two: We double the population but every single person who comes is below the median or the average, one or the other, in their skill level. And Case Three is they are all above the median or the average of the skill level. And the question would be how different are those Case Two and Three from Case One? That is, obviously, they are different. They are not the same.

George Borjas: Oh, definitely.

Russ Roberts: So, you want to speculate about that for a minute?

George Borjas: Well, I tell you: I think about it this way. In the first case it's very clear what will happen. You have a brand new United States next to the old United States. Everything doubles. I should say, per-capita GDP (Gross Domestic Product) doubles--I'm sorry, total GDP doubles; per capita GDP doesn't change. And when we just replicate ourselves, not much happens--with constant returns. When we don't replicate ourselves, what's going to happen is, capital will still adjust. Right? In the long term, capital should still adjust; but there will now be a different [?] proportion than what we had originally. And what I would think would happen in terms of a very simple-minded model of long-run growth, right--what I think will happen is that the group that encountered the highest supply increase will tend to have relatively lower wages than the group that encountered the least supply increase. So, for example, in the [?] you said, suppose a lot of the immigrants or most of the immigrants or all of the immigrants are low-skill. If you are a low-skilled worker in the United States, you are going to have relatively low wages. But the high-skilled workers--there are complementarities involved in this production technology, right? The high-skilled workers are going to be much better off.

Russ Roberts: Yep. And--

George Borjas: But look--that's actually raising an important point. Once you go away from this very simple-minded replicating ourselves model, you recognize immediately the distributional impacts. And that introduced the notion of tradeoffs into immigration discussions. You know, not everybody will be better off. It's true that the economic pie accruing to the original population will increase, but it doesn't increase equally for everybody. Right?

Russ Roberts: Correct. But--

George Borjas: [?] Hmmm?

Russ Roberts: That's for sure. I agree with that 100%. But it seems to me that the simple short run story misses a key thing. So, I'm going to shift gears on you.

George Borjas: No, look. Before you leave this, I want to continue. The correct way of looking at this--I mean, I think both things matter, okay?

Russ Roberts: And I agree with that. [?] not irrelevant. That will be clear from this next example.


Russ Roberts: So, if I said to you--until recently, let's say, we didn't trade with China, either because they were so poor or because transportation costs were too high. Whatever the reason is, it doesn't matter for this example. And now suddenly we are trading a lot with China. Which is what happened between, roughly over the last 20 years, we increased our trade with China dramatically. I have no doubt that that was harder on people who made things that--who had the skills that were similar to the people in China than it was on me: there are Chinese professors who come here and Chinese academics and Chinese economists, but in general it's relatively small. And so I've benefited tremendously from increased trade with China because my clothes are less expensive. And so on. My toys, gadgets, etc. And clearly a lot of people haven't benefited this much. In fact, they may even have been hurt on net. It's true they also get to buy cheaper clothes, and cheaper gadgets, but they have, unfortunately, might not have a good job at all.

George Borjas: Or [?]

Russ Roberts: Right--they might not have any job. Again--in the short run. And that short run might be fairly long. So, we don't want to say--the terms 'short run,' 'long run' are kind of shorthand ways of saying 'it's more complicated.' But we all understand, if we're thoughtful, what those complications are. Now, if you said to me, 'Therefore, there's distributional consequences of trade with China, I certainly agree 100%. But I would never suggest that trade with China and the United States is something akin to a wash for the wages or wellbeing or standard of living in the United States. I'd say, in the short run it's harder on people who have those skills, but because of the gains to all the rest of the folks, those people now have more purchasing power; and they are going to demand and create or produce more things--things are going to be created and produced using the opportunity now that things are cheaper coming in from China. And therefore, lower-skilled workers who were hurt initially, they may get benefits that are not see; and in addition, their children and grandchildren will live in a much better world potentially because we are using our resources more effectively. And I would think the same thing would be true of immigration.

George Borjas: Look. Let me actually make two points about what you have said. I mean, I think it's a really nice thing to compare the analogy you made on trade with a similar example with immigration. First of all, suppose that instead of trade with China, we had gotten immigrants--and they [?] to be very low-skilled immigrants. Okay?

Russ Roberts: Correct.

George Borjas: And that will mean the same thing you said about the people being hurt by the Chinese imports. Some people will be hurt by the low-skilled immigrants. And you and I would gain tremendously. We can buy all kinds of stuff cheaper; we can hire people to clean our house, fix our rooves, and stuff like that. And you can show from the simple economic model that both in the case of trade you had in mind, and the case of immigration that I'm putting forth as an analogy, the economic pie accruing to [?] has increased. The pie increased and the plate[?] shifted. It changed.

Russ Roberts: The way the pie is divided.

George Borjas: [?] And then the question is: How much weight should we put to the fact[?]--and this is a value question now--how much weight should we put to the fact[?] that many people perhaps are much worse off right now than they would have been otherwise? You made an implicit judgment: You said that in the long run, my children, grandchildren, so on, were much better off and they have more [?].

Russ Roberts: Oh, but not just mine. My point was much broader than that. My point is that the children and grandchildren of the low-skilled workers who are hurt in the current world by Chinese trade, they are going to inherit a better world because we've grown the pie.

George Borjas: Well, what's a better world if you've tried[?] this and don't have a job? And can barely afford--you know. See, that's where you and I depart. I'm not willing to go into the next step. Because I don't know what a better world would be.

Russ Roberts: But that means you are telling me that when we decide to open our borders to Chinese goods, we should not. We should be very wary of the fact that American workers--and by the way, this is also true of innovation.

George Borjas: Right--[?] make analogy.

Russ Roberts: It's not just trade. If we have technological improvement that makes some people's skills obsolete, and they are not going to have a job, their children and grandchildren are going to live in a much better world--which is the history of the last hundred years of the United States.

George Borjas: Okay. You show me evidence that in fact the children and grandchildren of people who [?] trade are way better off than they would have been otherwise, I'm willing to buy into that. I mean, to me that sounds like an ideological argument more than a factual argument. And let me tell you another thing that you raise in your question. As you say, yes, there will be an increase in the economic pie. But this is where the distinction between workers and people comes in very, very nicely, okay? In the case of trade, the increase in the economic pie accruing to natives happens, and it's there: no doubt about it. In the case of immigration, it's not so clear. Because suppose [?] who came in have been low-skilled. Well, they are going to have impacts--forget all the other stuff that might go on. Just look at the Welfare State impact. The Welfare State impact is not trivial. And it may well be that the expenditures that they trigger on the Welfare State could easily offset the gains that accrue from them being like widgets in a sense, right? And that's why it is entirely possible for globalization[?] from trade to actually increase the economic pie, and globalization through immigration to not increase the economic pie at all.

Russ Roberts: So, that is relevant. That's a relevant point. But let me just answer your charge that I made an ideological claim. I'm going to make the argument that I made a logical claim, through the logic of economics. And I'll give you some evidence. My evidence would be the following. In 1900 the average standard of living in the United States was a fraction of what it is now. It's probably something between 10 and 30 times higher. Of course, it's very difficult, it's impossible to measure precisely, because so many products, we try to figure out the purchasing power of--income today versus income in 1900, those products were simply not available in 1900. There's no elegant or precise way to deal with that reality. Put that to the side. No one disputes that our material wellbeing today dwarfs what it was 100 years ago. Now, what happened over those 100 years is that we had an immense amount of creative destruction. And it came from 3 different sources. We had tremendous technological change. We had a huge increase in globalization of goods and capital. We had a huge increase in immigration and mobility. And we also had population change. So, hundreds of millions--maybe--well, I'll leave it at hundreds of millions--hundreds of millions of jobs were destroyed over that century. Over the last 100 years. And new jobs were created. Now, you can argue that it's certainly the case that many of the individuals who were hurt by those three changes of immigration, technology, and trade were harmed, at the time. But I think it would be hard to argue that their children and grandchildren were harmed. They live much better material lives through the process of creative destruction--through the process of growth. You can argue it wasn't worth it. But my claim would be that a farmer living in 1900, when 40% of the workers were on the farm--today it's 2% in the United States--that farmer's children and grandchildren are much better off even though the changes that caused that to happen were very tough on farmers in the meanwhile. They couldn't cover their mortgage; their prices went down because of competition and increased innovation in agriculture and economies of scale. And small farmers in particular went bankrupt. And they had a tough time. And their children had suffered through the fact that their parents didn't have a job; their farm was out of business; etc. Their farm was bankrupt. But I think overall, it's been a pretty good run? You going to disagree with that?

George Borjas: No, not at all. I actually make the argument myself sometimes. Okay?

Russ Roberts: I'm glad to hear it.

George Borjas: So, I actually agree with you. But the point that--look. There's a lot of things going on over the last century.

Russ Roberts: Fair enough.

George Borjas: So, to--so who knows? It's very hard to disentangle these facts. What I think is something that we economists have been guilty of is the following: When we teach trade and immigration in class, we always point out the models create these benefits and costs. Right? Even though the pie might increase.

Russ Roberts: Yep.

George Borjas: When you talk about trade and immigration in the public debate, public policy--especially with trade--you don't hear much about the cost.

Russ Roberts: Yeah, and I agree with you there. That's disgusting. It's dishonest.

George Borjas: And that's been a very destructive part of what we've done as a profession, I think. Because some people do get hurt. And people getting hurt, getting left behind and [?] more has consequences. Social consequences; political consequences. And, you know, we are in a world now who might be living through those consequences.

Russ Roberts: I agree with that. Actually, I agree with it very strongly. I think it's incredibly depressing how advocates for and against both sides of these issues and don't admit to various costs and benefits, depending on which side you are on. [?] a free lunch.

George Borjas: That was the--thank you for saying that. That was one of the things I wanted to get into my book: There are tradeoffs in everything, right?

Russ Roberts: Yup. I totally agree.

George Borjas: And you know that immigration is one of those things.

Russ Roberts: I'm willing--even though I'm more of an open-borders guy than you are, George, I certainly agree with that in your book. It made me think about more than I have. Which I really appreciate. And it also reminded me of something I'm very much in agreement with, which is the tendency for advocates to cherry-pick data on both sides of this debate and avoid those costs. I [?] 100%.


Russ Roberts: So, let's talk a little bit about the measurement of those costs. What would you argue is the best estimate, at least in the short run--and the short run could be long--

George Borjas: right--

Russ Roberts: of the harm to native American, low-skilled--native workers of low skills? That is, high school dropouts--

George Borjas: right--

Russ Roberts: high school graduates from the recent increase in immigration? And by the way, the other thing that drives me crazy is everyone just assumes that everything is linear--so if we made it twice as big, the effect would be the same. Or if we do it now versus 50 years ago, it doesn't matter. And of course, you point out many times in the book very well that you have to be careful about historical context, the types of workers, the countries they come from. Those are all relevant. So, talk about what we know about the impact on low-skilled workers in America when immigrants come in.

George Borjas: Look, the number I carry in my head is that what we've seen in the last 20, 30 years has basically been something on the order of like a 20, 25% increase of supply on the bottom end of the skill distribution. I think the best available number is that that has basically increased--has lowered the wage at the bottom of the skill distribution by something between 3 and 5%. Which is not a huge amount. I mean, it's not a huge impact, by any means. But it's not zero.

Russ Roberts: Agreed.

George Borjas: And, you know: The question is: How much attention should one put when one thinks about immigration on that particular loss? And that's really much more of a value question than it is an economic question. I mean, what I'm trying to resolve in the book is to sort of point out to people that, yes, you can cherry-pick data, and you can do this and you can do that. But overall, there is some evidence that the people who are most affected, once they[?] find things properly, people who are most affected by immigrants will, just like you, are going to be hurt a little bit. But I think one thing that I don't do enough in the book and that people don't do enough generally, is sort of--it comes back to something that we talked about earlier: those who [?] come along with complementarities. And we've not really put much attention on measuring the gain that you and I get as a result of [?] immigration. And those gains could be substantial. And that truly is what the debate should be about.

Russ Roberts: Well, your point about the--let's say it's 5%--it does remind me that low-skilled workers in America, the last 25 years, have been very challenging, for three reasons. And we've mentioned them all. One is the increase in immigration of low-skilled workers. The second is technology, the increase in artificial intelligence, etc. And the third, of course, is trade.

George Borjas: Right.

Russ Roberts: All three of those--trade with countries who have a lot of low-skilled workers.

George Borjas: Right. They [?]

Russ Roberts: Yeah. So, this explains why there are a lot of folks in certain parts of the country who aren't doing very well. And it's a serious issue. So, let me--let's not debate whether 5% is small or large. I wouldn't debate it. It's small to some; large to many. But if you are one of those workers, it's large, I assume. If you are poor, 5% is not a trivial amount. Potentially. It seems to me the right policy response to that is not to keep out low-skilled workers. It's not to reduce innovation. It's not to keep out products from countries with low-skilled workers. It's to try to improve the skill-set of the workers who are here. To try to encourage them to finish high school. It seems to me a very strange policy idea to say, 'You have to keep out workers from, say, Mexico or Latin America, because they compete with American dropouts; and they get hurt.' Shouldn't we just try to improve our school system and our culture to get people to graduate high school?

George Borjas: Uh, okay. That's actually a great point. And my usual answer to that is two-fold. One is: Who pays for that? Because, you know, providing education for many, many more people is not--you know, there's not a free lunch, right? Somebody pays for that. And I don't know what the estimates of that would be, but I would like to know before you make a decision as to what the right thing to do would be. What would be the cost of doing that? Right?

Russ Roberts: Yep.

George Borjas: And then the second thing that's always puzzled me about this, and I've never really thought it through completely because I haven't had the time to sort of sit down and work out a model, is sort of related to what you said. A lot of people say, 'Look, it's true that, or it may be true that the bottom end has gotten hammered because of immigration and so on. But one good thing about all this is I think can encourage them to go back to school and get more education. Right?' And there's some kind of an upgrade, a skill-upgrade they call it, or something along those lines. And one way to look at it is that way. Another way to look at it is the following: These people decided, before this supply shock, that the optimal thing they should do was to get x schooling and no more. And now, there's a shock out their own control that they now have to revisit that mathematization[?] problem. And they have to incur the cost of moving away from whatever they have picked before to some other equilibrium. Right? Is that an optimal way to run a--is that an optimal thing that's come out of immigration policy, to force people who had already pre-decided they didn't want to do certain things, to make them do them? And I don't see anybody discuss this very clearly. And, you know, in the abstract, we can all say, 'Of course, we can just get everybody--let's make everybody go to college. Let's do that.' But a lot of people don't want to do that. And how do you compensate those people who don't want to do that, through the fact that you are not changing the environment in a way that they have to do something that they don't want to do?

Russ Roberts: Yeah. I think, instead of saying 'Force,' I would say, 'Incentivize.' But it's the same point. Right? Do we want to incentivize people to go to college, say, who don't want to go now. Etc. I think the key point in answering that--and we don't know the answer precisely at all, but the key point would be it's not just that they are going to be incentivized to go to college. It will be, or to say, in high school, say, in graduating, the key point is that with more workers and more people, presumably there's more--and if there is economic growth as a result and more capital coming in, the opportunities and the return to finishing high school and going on to college will be higher than they were before. Just as you mentioned--before, that those new workers will be complementary to, say, our skills. Of course, the question is how much and how big.


Russ Roberts: George, talk for a minute about your personal story. Because it's very interesting. And I think listeners will probably guess that you were an immigrant. And that your experience affected your interest in this topic. Again, I think--that's a small part of the book, but it's a very interesting part. And, talk about that for a little bit.

George Borjas: Okay; thank you so much. I tell you: I was born in Cuba. And my family used to own a small clothing factory. They manufactured men's pants. And, like I say in the book, I'm actually one of those rare economists that spent a lot of time at a factory with capital and labor. The manufacturing product that was sold in the marketplace. So, I've actually seen that firsthand. Okay? So, the factory was confiscated very soon after Castro took over. And basically, you know, my father was very ill from my childhood; and there were rumors that Castro was going to ship out all the children to the countryside and re-educate them or something. My family wanted to get me out of the country, no matter what it involved. But my father basically got sicker over that period, and eventually passed away. And that always disrupted these plans. My mother and I were eventually able to migrate--overseas--just a week before the Missile Crisis closed the border with Cuba. So I was very, very lucky in being one of those last rights[?]. Many years go by, and some--I'm an immigration, obviously; I'm kind of predisposed to immigration topics. And I'm at Columbia U., a grad student at Columbia--this a few years before we met at, you know, in Chicago, right? And Barry Chiswick comes by; and here's a paper; and that paper is about assimilation. That paper is actually, now, you look back on it, and it's sort of the foundation of immigration economics in the modern era. The paper that sparked the whole field. And his claim was--his finding was--that if you look at immigrants who just arrived in the United States and compared them to immigrants who had been here a long time, the immigrants who just arrived earn a lot less than those who have been here a long time. And then he proceeded to interpret that finding as a proof of economic assimilation. The longer you are here, the more you learn whatever it is you have to learn--the language, the American way of life, whatever. Right? The American way in the labor market. And you improve your human capital in some sense. In that seminar I asked the question that clearly came from my own background; and it happened to be because I knew that the human flow was composed of two distinct[?] waves--those who came, like myself, before 1962, and those who came many years later. There were many, many rumors or observations in the Cuban community that those who came before 1962 were the entrepreneurs, the highly skilled; those who came years later were not. So, I said, 'You know, couldn't it be that the reason you are finding that the more recent immigrants do worse than the earlier ones, not be the result of assimilation but just be the fact that the groups are just different kinds of people? Like, in the Cuban [?] I had in mind?

Russ Roberts: [?]

George Borjas: Exactly. And that was really this--this was the mid-1970s. That was the birth of my first paper on immigration, which was published in 1985. So, I moved to California. And this idea that I asked [?] about, kept floating in my head, as California was literally changing dramatically over in that period. This is the early 1980s. This is when you and I had met; and I moved to California: the early 1980s. And you can basically see California changing. Before 1980, California was not a particularly heavy immigrant state. And I get there; and overnight, you can sort of see the town changing. Okay? And I said to myself, 'This looks really interesting.' And it was again [?] change. And I want to study this, out of curiosity. And a question that kept cropping up in my mind was that I had asked at a seminar: How, exactly, would you measure assimilation when you have different "ways" of people being different? And that's what I said in the study; that was really my answer to immigration. And that's the way I got involved.


Russ Roberts: And how much do you think we know now? And there are many, many examples in the book trying to measure this; but try to summarize them. What do we know now about people's ability to assimilate? Whatever that means--there's cultural and economics and assimilation. But let's stick to the economic side, again--the idea that the wages of the children and grandchildren become closer to the wages of the natives.

George Borjas: Right. Well, there's two kinds of estimation we are talking about. One is what happens if you are an immigrant, lifetime, and the other is what happens to the children and grandchildren. Right?

Russ Roberts: Yep.

George Borjas: The whole Chiswick paper was really what happened to some of the immigrants, lifetime. And we know a lot more about that now than we did when Chiswick began--obviously. What we know is that the groups of immigrants who came in the 1940s, 1950s, 1960s, 1970s--you know, assimilated quite well in terms of, they experienced a lot of wage growth. The groups that came in the beginning of the 20th century and the groups that came of the end of the 20th century, did not experience as much wage growth. It's sort of a curious finding. Because if you look at the 20th century in the United States, it's basically book-ended by two mass migrations. And what we tend to find is that they [?] came into mass migrations don't tend to progress as fast as those who came in a period of less immigration. Okay? So that's actually very interesting, sort of topic for future discussion. And we don't quite know why it happened. But it's interesting.

Russ Roberts: Could be a supply effect. Could be something different. Could be there was a big recession in 2008--

George Borjas: Exactly--

Russ Roberts: that set everybody back. We don't--

George Borjas: We don't know. I'm being completely honest. We don't know, but it's just a very interesting pattern. The second thing, which is what happens to the children and grandchildren. And again, the only experience we have here is what happened in the 20th century. So, what we get, we have data for, is we can track the people who came in the early 1900s, look at their children, look at their grandchildren over the century. Right? And what we see is that they are [?]. The children improve over time. And ethnic inequality--in other words, the difference between Group X and Group Y narrows down a lot. And that's what you would expect from the melting pot working. Now, you didn't raise this point, but I make this point in the book: Which is the following: A lot of people look at that 20th century experience and say, 'Look, you know, even though we are having, perhaps we are having some problems today with the new immigrants--they are not doing so well, whatever--it all worked out in the 20th century and therefore you can just extrapolate into the future, and it will work out fine. And this is one of the points I want to make in the book, which sort of came up in our conversation before. You know, conditions on the ground matter a lot. You cannot use things that happen at one place at one time, regarding immigration, and imagine that it will happen in another place at another time. Because, again, immigrants are not just robotic workers. And they example I'd like to give of the 20th century is that--sort of a few examples--is the following. One is: Look, when immigrants in 1900 came in, they went into the manufacturing sector. It's not too much of an exaggeration that the Ellis Island immigrants built up the manufacturing sector of the United States to a large extent. Now, the important thing about that is that those jobs eventually became unionized. And were very high-paying jobs. And those union jobs were, you know, were, you know, [?] transferred within the family. So if your father--one number that I always find incredible is that in 1915 or so, almost 3/4 of the Ford Motor Company workers were foreign born. Well, just imagine if you were a Ford employee and became unionized, got a great-paying job. Your children got that job. Your grandchildren. And that was really the middle class in the 20th century for many immigration families.

Russ Roberts: Yeah. That's a fantastic example. Of course, it was harder for other people's children or grandchildren to get those jobs, because the union wanted to keep them out and were eager to keep the gains very narrow. So they--

George Borjas: Exactly--

Russ Roberts: I just want to emphasize that.

George Borjas: But nevertheless the fact that the immigrants were so over-represented in the manufacturing sector clearly helped them.

Russ Roberts: Yeah.

George Borjas: And the question is: What is it now--what conditions today will lead to that kind of assimilation path in the next hundred years? I don't know. Another issue is, there were two wars. And the Germans were, in particular, highly discriminated against in WWI. And those[?] states passed legislation making it illegal to speak German in public. Making it illegal to actually speak German in public schools. And that was in effect an assimilation, right? Again, we don't want to have another World War, but how do you produce those underlying conditions? And that's one of the reasons--I think there's something you've actually talked about before in some of your work, is that--there is, the incentive to assimilate culturally doesn't really exist any more. The [?] kind of assimilation doesn't really exist any more.

Russ Roberts: It's true.

George Borjas: You know, I give the example in the book of the [?] California, [?] out a memo prohibiting people from using the world melting pot because it's a micro-aggression. Just imagine what doing that means in terms of, you know, future assimilation. It's not really clear, right?

Russ Roberts: Well, the example you also give in the book--which I think is fantastic and relevant, is, through most of a large part of American history, immigrants were eager to learn the language. But, of course it's hard to learn a new language when you are an adult. But the children, at least, would learn the new language. And the parents would often do their best to speak the native language at home to encourage their kids. In fact, my wife's grandparents came here from Germany in 1939, escaping the Nazis; and they never spoke German again. So, think about how crazy--think about that house: A husband and wife, not speaking their native language because they are so horrified and traumatized by the Holocaust; so of course their children spoke English much better than they would have otherwise. Of course, they didn't know German. That was--there was a cost to that, of course. But in today's world, we make it, for reasons that I'm not going to try to explain, we make it easier for people to not learn English. We have wonderful private and public accommodation of people where English is not their native language. Obviously Spanish being the second language. And there's something good about that; and there's something not so good.

George Borjas: Right. And look: the other point that you can make is that there are now gains to be had by making sure you retain the ethnic identity. In terms of all kinds of programs that try to divide people by ethnic groups and reward people and penalize people accordingly. And that also affects assimilation. So, I think that a usual--we all have a tendency to do this. And we all look at data; and we always look at data from the past. And we assume the world is linear and we can just extrapolate from where we are now using the data from the past. Right? It's not so clear you can do this in this context. And that's one of the warnings, the warnings that I give in the book.


Russ Roberts: Yup. Yeah, and I agree with that. So, let me make a philosophical observation and see where you stand on this. So, I'm going to give you my summary of what we know about this and why I come down as I do. And I'm going to be as honest as I can be. So, I concede that immigration is hard on low-skilled workers. I think the effects are relatively small, but I concede that, for them, small is not really the right word. It might be large to them, because 5% of a poor person might be more significant than it might be to me. And I get that. I also believe that over the longer period of time it leads to growth, and better use of resources; and that's going to enhance the opportunities of their children and grandchildren. Especially if we get rid of things like mandated second-language signs, etc. I think we should be encouraging people to learn English and use English; and I hope culturally we'll move toward a more assimilatory--I think it's okay to say, melting pot. I would like to see that. But I understand I don't have control over that. So, I think the economic effects are relatively small. For me, and others, but especially for me. So I concede that. They are bigger for people who are not like me. But for their children and grandchildren, I think that they'll be okay with it; and I think they care about their children and grandchildren. And I see a huge benefit to the people who come here. So, as somebody who is not an nth generation American--just like you, I'm not as new as you are--

George Borjas: Right.

Russ Roberts: but I have--my 4 grandparents, 1 was born in Poland and 3 were born in the United States. I think--they all came in the 1880s part. And I concede the part that I have an emotional reaction to this issue, partly because if those people hadn't come here they would have been killed in the Holocaust. And they would have been really poor, either way. And I'm really glad that we got--that my ancestors came here. That I got to be born here instead of in Poland, or somewhere else. And I'd like to see that opportunity available to other people. So, I see it as a wonderful thing for the people who come here--mostly. I see it as mostly a good thing for the people who are here. And for me the only issue is the cultural issue--is that for a net negative or a net positive, to have diverse people coming here who may not share all the cultural values of America. And there I do worry about the melting pot and assimilation. But that's where I stand. Where do you stand? And does your personal story play a role in your willingness or eagerness to let--or antagonism to people coming here?

George Borjas: You know, we are not that far away, believe it or not. I mean, I think I will--in the little summary you just gave at the very end of what you just said, you sort of didn't really talk about the people being hurt by it so much. And the question is how much wage we put on that. You know, one of the things at the end of the book that I have is I sort of say to myself, 'What would I do if I had control over immigration policy?' Right? And even though there is an impact on the low-skilled labor market, I don't come out at the end and say, 'Let's just stop all low-skilled immigration.' In fact I say we shouldn't do that. We should actually--there is something quite--in the way you put it, there is something quite historic about the United States giving the opportunity to many low-skilled people from all over the world to come to this country and live, you know, much better lives. Right? And I don't want it--I would never want to throw that away. That's part of what the United States is about. But what I think we've made the mistake on, is ignoring the fact that the long run is far away. And there is a need, right now, to address the dislocations suffered by people at that low end. And unfortunately, both in the case of trade and immigration, we've tended to be models[?] people. We tend to look forward and say, 'Look, in the long term, the economic pie will go up. Your children might be fine.' Whatever. Right. But that's 20, 30 years down the line. What do we do with the people who are being hurt today? So, I'm not that far away from you. What I would actually argue, that we cannot just dismiss the short-run impact. So willingly. I think we have to take--we have to think of immigration--if we are going to do these kinds of things, we have to think of immigration policy in a broader way. And that broader way is not just how many immigrants to admit and which immigrants to admit, but what to do about the people who are being hurt by it now. And once we address that, then how much more in your ballpark?


Russ Roberts: I think we have to deal with cultural issue, too.

George Borjas: Yeah; all of that [?] book. I mean, the cultural issue--look at Europe right now. Right? The cultural issue is at the core of all this. And is something--we wanted workers; we got people, instead. And that's what encapsulates what's really at stake here.

Russ Roberts: You have any thoughts on what we might do to--I mean, I would argue that Europe's problems are much worse than the United States's, because, for whatever reason--historical or policy-wise--workers here in the United States are much more likely to assimilate. Their children and grandchildren are, I think, as well. We can debate how fast that is relative to the past. I agree with you that it probably isn't the same. But we seem to be doing a better job--better than Europe is. Is that just good luck? Is it other things that we might do as a nation, though, to make that easier?

George Borjas: Well, look, we've done a lot better job than Europe has in the past. And we did a lot better job than Europe did in the past by basically keeping our hands off the whole assimilation process. We let the process work. The government really--you know, we've had immigration policy, but unlike many European countries we've never had an assimilation policy. We've never penalized it. We've never--you know, apart from a special case like Germany and things like that, right, we've never really done--the government never really has been that much involved in the process of assimilating immigrants and their children. Right? And, the question becomes, in today's world: Is that hands-off approach something that's wise? Or should we begin to rethink that? And I actually don't have an answer for that. I don't really know what the correct approach is.

Russ Roberts: But I think there are things the government does that reduces the incentive to assimilate? Right? They mandate certain things.

George Borjas: Oh, I agree. I think, I think, from the point of view of the country, assimilation is a great thing. I think taking actions and putting in policies that retard assimilation or that encourage ethnic identity to remain generation-after-generation is not a good thing in the long term. Okay? But, you know, I wish I had a magic wand to sort of pass by and return to--you know, one great example of this is that if you think--if you think of the [?] grandparents, and you think about the public school system in New York City back in 1900, 1910, 1920--you know, their goal was to make these immigrant kids Americans. Okay?

Russ Roberts: Yeah.

George Borjas: They learn the language and they were going to come out as American kids. Right?

Russ Roberts: Yup.

George Borjas: You say that today in a school, and you get fired. And that's a problem. That's a problem we want to address.

Russ Roberts: And those kids desperately wanted to be American.

George Borjas: Exactly.

Russ Roberts: They were eager to leave behind their customs and traditions. Often.

George Borjas: Yup.

Russ Roberts: It's been a fascinating ride, when you think about the ethnic pride that we have now, and the negatives about that in terms of seeing us each separate versus part of the same thing--

George Borjas: I know[?]--

Russ Roberts: There's a tradeoff there that's unavoidable.

George Borjas: Right.

Russ Roberts: And certainly maybe we've reached a tipping point where people don't want to be "American". They want to be something American--whatever it is.

George Borjas: And I think if you are from[?] California, say if you are saying 'melting pot,' you've done, you've conducted a micro-aggression.

Russ Roberts: Yeah, that's a fascinating--how that came to be, of course, is unknowable. It's an emergent phenomenon. But that's how I think you would--you would argue, I assume, that that's a reality we have to accept.

George Borjas: That's right.

Russ Roberts: We can't change that easily. And we need to take that into account when we think about--

George Borjas: And the important thing for policy is you really have to take that into account before you let in too many people, for example. Like Germany did. You know. It's not something you can ignore.

Comments and Sharing

TWITTER: Follow Russ Roberts @EconTalker

COMMENTS (54 to date)
pyroseed13 writes:

Excellent discussion. I have to agree with Borjas on the point that the argument that the children and grandchildren of those who were hurt by trade and immigration will be better off is more of an ideological argument than a factual. There's nothing in the theory of trade, automation and immigration that implies that future generations of the affected workers will necessarily be better off. In fact, given the dynamics of today's economy, there are reasons to think that they could be worse, unless additional interventions were made to alleviate the pain.

People like to make comparisons to other periods of industrialization to argue that these groups will be better off. But back then, most workers were what we would call low-skilled. If you lost your job in agriculture, it would be relatively easy to transition to working in manufacturing. You didn't need a college degree or certification. It's much harder for workers to transition to industries where wages are rising, and therefore is it is conceivable that those children and grandchildren of affected workers could be made worse off by trade and immigration.

rhhardin writes:

What's unique about American assimilation is suggested by a line in Bridge of Spies, to the effect that what makes us American isn't ethnicity or origin or even citizenship but agreement to American rules, in particular the Constitution.

You watch out for the other guy's rights, as a result. It's an agreed-to self-limitation on culture.

Nobody cares if you raise chickens or any other cultural thing, if you agree to that. Then you're an American.

Floccina writes:

I have an anecdote on this:

I have a friend who is a low wage earner. He is a good guy. He lives in about the worst housing in our town. His US born neighbors are generally a mess. One sold his car, he has many problems with them, but some Mexican immigrants have moved in and he has made friends with them they seem to have far problems and provided him better neighbors. The immigrants are living in that housing because they send money to relatives. The bottom live is that they have probably improved his life even if they have lowered his wages.

mike64 writes:

Mr Borjas points out that the increase in labor at the bottom end of the skill set has only effected wages by 3-5%, however there are 3 points that are as or more important that wages.

1. The minimum wage laws put a floor on the wages. If there were no floor, it is likely the wage effect would be larger.

2. The make-up of the workers in that labor pool is changing. 40 years ago, it was predominately high school and college kids trying to earn some spending money. Today it is more likely to be older workers often times with kids.

3. Government benefits are making up some of the 'difference' in wages for those workers because of child tax credits, the EITC, subsidized housing, and other benefits.

Wages aren't the only measure of success/failure of the immigration policy but the composition of the labor pool and government policy also have to be taken into account.

In summary, economic improvement for natives at the bottom of the labor pool are being crushed by the massive immigration in the last 25 years. Mr Borjas is correct that without a job, it doesn't matter that prices across the board are lower.

pyroseed13 writes:


Your second point makes me ponder a few things. First, it's important to remember that high school dropouts are only about 12% of the labor force. Not a small group by any means, but smaller than it was 40 years ago. As a result, it would probably be wrong to say that most people at the bottom are negatively impacted by immigration, at current levels that is. Even high school graduates according to Borjas see a small boost in their incomes from immigration.

Second, I am not really sure how much we know about this group of high school dropouts. If these are older workers as you suggest, then the claim that these workers will try to upgrade their human capital in response to falling wages for high school dropouts is probably false, as Borjas correctly notes. Some will probably just be pushed out of the labor force entirely. But if these are younger workers, it seems bizarre to suggest that wages are main reason that someone would choose to graduate rather than dropping out. People drop out for all sorts of reasons, none of which I would argue is particularly rational. Given this, I am torn by how much sympathy I should give to this group, even as someone who prefers less low skilled immigration.

SoCalBaker writes:

This interview was one of the few that had me agreeing with the guest more than Russ. I think that culture and intent is ofter overlooked in the success or failure of immigrants impact of on the host society

I have a Korean friend that his parents came over to America to make a better life. The Parents opened a doughnut shop and raised their son to do well in school, he went on to be a doctor.

I see economic Immigrants who come here to make money and to send it home, with the intent of going home one day to build a house in Mexico. They don't assimilate, they don't force the kids to do well in school, parents never learn the language after 30 years of being here.

Kids now working, competing for the same jobs has the parents have, not much improvement at all. The schools become a basket case, because the good kids leave and now they spiral.

You can always tell when a town in Southern California is dominated by Agriculture or low skilled manufacturing/ service because the school scores are in the toilet.

People always forget, cheap labor is not cheap, it is who pays; the employer or the society. Low wages help the employer and the consumer, but not the society that was to educate or medically treat or police the low wage worker.

Lastly, as for the transition from the farm to the factory was over a relatively long period and peoples life spans were shorter, so a society could digest a painful transition easier, now we have a rapid change from factory to high tech with a longer life span on top of that in a very short amount of time.

FredC writes:

Another great podcast.

Yes, I think things are truly different now. American workers at the low (and increasingly middle) end of the scale have been hammered by the triple whammy of trade, massive immigration, and automation.

Look at the war zone that is Chicago, opiod use in middle America, people in their 20-30's living in the parents' basements, low family formation numbers, poor labor participation and significant underemployment. Clearly, the down side has not been sufficiently accounted for.

The expansive welfare state and cost of modern medical care also skews it all. And finally, lack of assimilation and increased balkanization in the and identity politics in the USA add fuel to the fire.

I think significant turmoil is coming, regrettably.

Sam writes:

Russ repeats the argument that grandkids and grandchildren of everyone is better off now because of immigration in early 1900s or late 1800s. But is that true of African-Americans? Imagine the wave of immigration in the early 1900s hadn't happened, wouldn't the demand for labor facilitate more organic racial integration? Allowing the racism is bad for business argument to work in practice. Maybe African-Americans would get some of those manufacturing jobs that later unionized.

Also Russ' point is about aggregate lives of the grandchildren. Borjas was arguing that maybe true overall but not for the subset of people whose grandfathers were negatively affected.

Finally the successful wave of immigration Russ is happy to mention over and over again happened before the modern welfare state. That is a huge change that only gets barely mentioned (and it is Borjas who brings it up).

Bogwood writes:

Given the increasing resource scarcity, massive population overshoot and the zero sum nature of the current USA economy immigration has to be controlled. A simple fair rule would be some variation of reciprocal immigration, you come from Ireland, someone goes to Ireland. It would reduce the current moral hazard problem.

Remember that no human is literally productive but either extractive or facilitating extraction. Being top predator costs 2-3 orders of magnitude more energy than the than the primary producers. The only option is fewer top predators. Cheap energy gives prosperity borrowed from the future, not so much in dollars as in diminished future resources. That one hundred year window is closing. This type of discussion needs an ecologist or panel of ecologists to keep the discussion on track. Converting a Somali living closer to balance with nature, to an American consumer does no one any favors, not even,in the long run, the immigrant.

SaveyourSelf writes:
George Borjas said, "When immigrants break down or get sick or things happen, we have a responsibility."

Now hold on, George. Where does this assumed obligation come from? Are immigrants our children? Are they our property? Are they under compulsion to immigrate? Have they given us or you some sort of gift that you feel obligated to reciprocate?

George laid down this statement concerning "responsibility" right at the start right after declaring that immigrants are people. Fine, their people, but how does that make him their dad? His very foundation is paternalistic hubris. His solutions boiled down to "build up barriers to competition for those most vulnerable to it." He is making cowardice a virtue.

If men are free to choose and they choose and lose, let them fail. Otherwise they were never free to begin with.

Bring on the competiton, I say. If we must push values on the poor, let it be courage in the face of the unknowable and persistence in spite of repeated failure, because that is learning. And learning leads to growth. Borjas' solutions amount to protecting a pet group of people from the injustice of having to learn.

On the plus side, I appreciated Borjas's humility regarding the present state of evidence in economics. He didn't embellish.

jw writes:

An excellent guest and an excellent podcast (I found myself wondering if "extremely balanced" was an oxymoron as once something is balanced, it cannot be more balanced...). Notes:

- It was refreshing to hear Borjas' reasoned critiques of studies and offer counterfactuals and admit where there are gaps in our knowledge. How this guy is at Harvard's Kennedy School is beyond me. Maybe there is hope yet.

- As usual, there was not enough distinction between illegal immigration and legal immigration. I think that it goes without saying that the illegal immigration is primarily low skilled and that mainstream America would have little problem with immigration if it was controlled and roughly reflected an economic/cultural cross section of current America.

- I am glad that they covered the quantity and assimilation/cultural issues. It is much more difficult to assimilate a flood than a stream. It is much more difficult to assimilate completely foreign cultures (Borjas' comment about Europe wanting workers and getting people was the capstone of the podcast).

Let's not forget that the flood of illegal Mexican immigration over the past eight years was not some enlightened humanitarian vision but a crass political calculation designed to bolster future voting blocks.

- As above, historical immigration patterns pre-welfare state are almost useless now. The incentives are much greater now to immigrate and not contribute. As Borjas pointed out, we now have government incentives to NOT assimilate.

- You can keep dreaming that better schools and more education funding will somehow overcome exploding single parenthood and family breakdown, but it won't. It also won't fix the fact that by definition, 50% of people have IQ's below 100. With agriculture, manufacturing, blue collar service work, etc, they can have solid and fulfilling lives, but as Charles Murray points out, they aren't going to become rocket scientists. So without those opportunities, they are naturally stressed. As discussed, they are stressed by technology, trade and immigration. The easiest one to control is immigration.

- Make no mistake that even the highly skilled H1B visas provide Silicon Valley billionaires with lower cost labor. There is no shortage of technical labor, there is just a shortage at the price that Silicon Valley wants to pay (as high as that might seem to the rest of the country). I am not arguing for less H1B, but just pointing out that this economic argument (that I learned from Sowell) is not just about the low skilled.

- As I have said before, diversity for diversity's sake is not a valid construct. The individual components of a system must all have positive expectations. You can never know if they will have positive outcomes in the future, but you want to design your system to provide the highest probability of a positive systemic outcome. You cannot do that if you do not have some control of the inputs.

For the US, once they are inside the system (citizens protected by the Constitution), you (rightly and properly) have much less control.

FredC writes:

The massive immigration we have had over the last 30 years has coincided with a national debt that has exploded to $20trillion....largely fueled by entitlement spending. Yeah yeah, big difference between correlation and causation. But still...

As an aside, I'd love to hear how an undocumented immigrant making $12 an hour hanging drywall is offsetting the thousands $ spent on his american born children for schooling and medical care and food stamps. He could work 30 years and not cover the medicaid funded hospital cost of the birth of one child, even assuming everything went smoothly. Now, multiply this scenario by 10,000,000, and read some columns by Victor Davis Hanson documenting real life in the Calif central valley these days. (He'd be a good guest for the podcast, btw)

Mr Borjas nails it, the downside has been swept under the rug. But that $20trillion is causing an awfully big lump in the carpet now.

One more thing, we are above 320,000,000 in the US now. In the seventies it was non-stop blather about over population. Can somebody clear that up for me, do we need more people in the US, or fewer? I'd love to know for personal breeding purposes.

A.G.McDowell writes:

You compare the success of integration in the United States with that in Europe. It might be interesting to correlate this with the availability of government largess for politicians to distribute to particular communities, and to the existence of inequality, especially meritocratic inequality.

Northern Ireland is a noted example of the persistence of two distinct communities - no melting. In "Minority Verdict" Maurice Hayes notes that each community's politicians saw their duty as ensuring that government resources were distributed to the community they represented, for example in the provision of housing by the Government. I would compare this to the well-known "resource curse": the path to success for an able and ambitious individual is in acquiring and directing the resources of the government so that it flows towards themselves and their community.

If such people, in both communities, had the option of acquiring great wealth for themselves through business and technology, there might have arisen a powerful group of people, drawn from both communities, who had more in common with each other than with the less successful members of their own communities, and with the incentive, and perhaps the power, or at least the prestige, to damp down any conflict which threatened their success. (For an example of the attempted use of prestige, see "Leave the fighting to McGuigan" - or, from a very different era "Dear Kate, you and I cannot be confined within the weak list of a country's fashion: we are the makers of manners, Kate")

SaveyourSelf writes:


JW wrote, "Let's not forget that the flood of illegal Mexican immigration over the past eight years was not some enlightened humanitarian vision but a crass political calculation designed to bolster future voting blocks."

Probably, but it may also have been an enlightened economic decision designed to reduce transaction costs in US markets.

Your point about assimilating a flood vs a stream is well taken.

After thinking about it a while, I've decided I disagree with your position on the importance of illegal vs legal immigration if by legal you mean "approved by some central government committee". I'd agree if by "legal" you meant, "agreed to the terms of citizenship". Outside of the restrictions of citizenship, I don't think there is any meaningful issues with immigration in a world absent redistribution. In a world with redistribution...well, you know--it speeds up catalysm, but that's not an issue with immigration. It's an issue with unstable property rights.

One more nitpick. You wrote, "the individual components of a system must all have positive expectations." That's not necessarily true. It is very possible that a complex structure with "the highest probability of a positive systemic outcome" will include elements with negative expectations. Acceptance of your statement is useful for designing systems with deductive reasoning, but real world systems are not constrained by logical proofs. The only way to know whether a component with negative expectations would improve overall stability is through competition. Like I said, it's a nitpick.

As an aside, your writing style is consistently crisp, intelligent, thoughtful, and a delight to read. Thanks for posting.

Rob writes:

This was an interesting discussion, but I was left confused by the last few minutes. You both believe strongly in the importance of assimilation, but I don't think you explained why.

From my perspective (as someone who grew up in Britain, now living in Germany), it's not obvious what is necessarily desirable about assimilation. I think that my generation (I'm 40) has grown up surrounded by the idea that cultural diversity is valuable and makes life interesting. I am genuinely open to hearing arguments about why that view may be mistaken - so I think you missed an opportunity here to make your case.

Eric Johnson writes:

I know he was past the 1 hour mark, but it seemed to me that Russ shutdown the podcast rather abruptly at the mention of Germany allowing 2 million immigrants without a government assimilation plan (rather than allowing the guest to make the traditional closing summary). Too much political baggage for the end perhaps?

Also unmentioned (that I can recall) is the fact that the US in 2017 is a county of well over 300 million, vs the US in 1900 at 76 million. Just the raw numbers makes a huge difference for what is both possible or desirable in terms of immigration policy. Other related topics such as welfare entitlements at least were mentioned.

My question: At what point do we as a nation start biasing national policy to benefit our existing population (i.e. Americans), vs the *possible* benefit of importing populations from other places, for whatever supposed goal (intellectual capital, economic growth, humanitarian assistance, ...) ? There was some tangential discussion about these topics, but the bottom line is: if the US doesn't have a target, we'll never hit it. The US needs to have a comprehensible, and well-supported policy for these things, otherwise we're going to continue splintering into a thousand hyphenated-American tribes.

Greg G writes:

>----"Let's not forget that the flood of illegal Mexican immigration over the past eight years was not some enlightened humanitarian vision but a crass political calculation designed to bolster future voting blocks."

This is a classic example of an "alternative fact."

In reality, net immigration of Mexicans to the U.S. was much higher during the Bush administration for the very good reason that that is when the housing boom happened and those immigrants were drawn by the ease of getting jobs during that construction boom.

The net immigration of Mexicans to the U.S. reversed during the Obama administration and many illegal Mexican immigrants "self-deported" due to the much greater dIfficulty of finding a job during the Great Recession.

Turns out economic incentives matter and it is a lot more appealing to immigrate to a neighboring country during their housing bubble than their housing bust. But if your view of immigration is based on political conspiracy theories you may not believe that.

Kevin Driscoll writes:

First off, this was an excellent show. I felt you communicated the heart of your agreements and disagreements well, without making a disagreement among people with PhDs in Economics sound like a disagreement among lay people.

I was very surprised that the assimilation issue came up, and now that it did I wish you had more to say about it. I am 75% on the other side of this. I've basically bought into the liberal (or Social Justice Warrior, if you like) idea that we should not be expecting people to assimilate. I view the problem as being primarily with the intolerance of the native people; they seem to not be willing to accept a robust pluralism. I think it's hard to argue that the 'default' in the US is a kind of European Whiteness, and it seems pretty repugnant to me to be telling people that it's best if they erase their heritage and put on a coat of Americanism. And that's true especially in the case of Black Americans, many of whose ancestors were drug here against their will.

My heuristic is to trust the experts. But in this case the economics experts are saying something *diametrically opposed* to the experts in sociology, anthropology, etc.

On the economics of it, I can see why all of us speaking the same language would be more efficient. And if we all wanted similar stuff then there would be increased opportunity for economies of scale. But that all seems like small potatoes to me, and I like to hear about whatever data exists.

pyroseed13 writes:


For what's it worth, polling does show that Democrats in particular have become more supportive of immigration over the past two administrations than they were during the Clinton years. I could never imagine a Democratic politician giving the speech Bill Clinton gave on illegal immigration in today's political climate. I don't really see how anyone can deny that their support is driven by mainly concerns to create a class of voters that will vote for Democrats, especially after all these years of saying how the Republicans are finished because of the "emerging Democratic majority."

Your other points however are correct and interesting. I wonder if any research has been done to examine how immigration, both illegal and legal, may have contributed to the housing boom. Presumably, if the cost of labor goes up, fewer houses will be built, or fewer renovations will happen. Your last point about "self-deportation" also illustrates why something like E-verify could work pretty well. People will be more likely to return to their home countries when faced with diminishing opportunities.

Greg G writes:


>----"I don't really see how anyone can deny that their support is driven by mainly concerns to create a class of voters that will vote for Democrats, especially after all these years of saying how the Republicans are finished because of the "emerging Democratic majority."

Is it really beyond your imagination to think that a lot of people sincerely believe that immigration is a net positive for America? It's one thing not to believe something. It's something else entirely to think that other people don't believe it.

I have never thought that immigrants were necessarily a sure thing as Democratic voters. And even if they were, new immigrants are unlikely to become voters soon enough to do many current incumbents much good.

Regardless of what country they come from, immigrants tend to be more self-reliant and entrepreneurial than the average person from their country of origin. They often have a libertarian streak and often have conservative social values. Mexican immigrants are far less likely to divorce than the average American.

All other things being equal, the Republicans should do well with immigrant voters. But all other things are NOT equal. As George Will pointed out, if you are a politician and voters think they don't like you then you have a problem. If you are a politician and voters think you don't like them then you have no chance with them. Republicans could compete just fine for immigrant voters if they dropped their hostility towards them.

Neither political party is ever dead in a two party system. A change of power is only ever one recession away. The highest levels of support for any Presidents in my lifetime have been LBJ in 64 and Bush (41) in 1990. Neither was even re-elected.

Daniel writes:

I think the key to immigration and global trade to Mr. Borjas point is what do we do with the people impacted NOW.
The government has ignored this for a long time, where some parties hope the free market will take care of it(pull yourself up from Bootstraps mentality) and other parties building inefficient programs to retrain, and lastly the failure of government to not regulate entities profiting from misery, For Profits schools as an example.

My point is that there has to be a solution for the NOW group, because we cannot simply stick our head in the sand and protect all jobs/wages from a global economy but we have to offer a model for transition or dare i say it a "PAYOUT" for those maybe of a certain age that will have a really hard time retraining.

What we do know is that doing nothing leaves people in a terrible spot and while their children will probably benefit in later generations from the net change some of generations between NOW and Then will be lost. And that loss causes Instability which is what we seem to have now in our system.

Sasha Sasha writes:

I was a bit confused about how the "children and grandchildren will be better off" argument could be implemented into an economic model. It's wrong to just make such an assumption without any data. Not all people have children and grandchildren and we also don't know the level of concern that people have for their children. So I do see this argument work from a point of view of an economic planner who is concerned with the total welfare of the current and future citizens and is trying to maximize that function.
I don't see this argument work given the current democratic political structure. You can't assume for people that they should care more about the long run outcomes for their grandchildren than about the short run impact of a policy on their daily lives.

Also the against immigration arguments, in terms of implementation, forget that US population consists of many immigrants. The impact on wages and employment from immigration was discussed but you did not account for the fact that immigrants benefit from an open immigration policy. As an immigrant I would not vote for a policy that closes the borders even if it would lead to a lowering in my wage because the benefits outweigh this cost. I am much better off the more people speaking the same language as me immigrate here because that means more friends, more potential marriage partners, larger community events, more connections at work - so in total the benefits are much greater than the possible cost of lower wages.

pyroseed13 writes:

@Greg G

In response to your first point, so why did the opinions of Democrats toward immigration change from the 90s to the present? Was there suddenly an avalanche of research that led people to think that immigration would be a net positive for the economy?

Yes, it's true that many groups of immigrants exhibit signs of social conservatism. But do they vote on those issues? No. Many Hispanics share the same big government views as liberals, and that is what primarily drives their support for Democrats. But there are exceptions of course. Cubans, Vietnamese, and Filipinos tend to be Republican leaning groups. But there's ample evidence that migrants tend to bring the views of their home countries with them. See Garrett Jones. Yes, over time these groups may become more like us, but as Borjas points out assimilation has been slowing down with recent waves of immigration.

Chris writes:

I was a little confused by the distinction made between globalisation in trade and globalisation in people.

If a country imports goods say from China, and these are essentially inventory to be sold to domestic consumers, there is no guarentee they will be sold. If they are not, there are a range of private costs (including opportunity costs) associated with holding the inventory and eventually disposing of it. Are these costs not analogous to welfare state impacts that are discussed in the context of immigration?

jw writes:


Thank you for the kind words.

There will always be a political component to immigration control, that cannot be helped (elections have consequences). But no standard is not a better standard.

With respect to the system design analogy, no analogy is perfect. In my systems, negative expectations that are negatively correlated are quite useful, but I can't come up with an analogy for this as it pertains to immigration. What I do know is that without some defined control, we may get what was reported by many during the Mariel boatlift, where Castro emptied his jails and mixed criminals in with the refugees. I also see what has happened in Europe with their apparently failed attempt at assimilating a large group with completely different values and no vetting (as if vetting is even possible from some barely functioning governments).

There are Muslim "barrios" in England and France where it is simply too dangerous for police to enter. I imagine that most Americans do not want that here.

I am all for LEGAL immigration (already about 1M/yr), but there must be some controls.

Kevin Driscoll,

My heuristic is to trust the experts.

If you learn anything from Econtalk, and Borjas provides an excellent example in this podcast, it is to take the "experts" with a large grain of salt. Always go to the original sources and come to your own conclusions.

Greg G,

This is a classic example of an "alternative fact."

It is actually quite the opposite. Instead of relying on MSM factoids, as above, one has to look behind the headlines. The most generally cited figures come from Pew, who in turn rely almost exclusively on the US Census. They expect an illegal alien to provide their name, address and whether they are illegal or not. Naturally, this is a quite unreliable methodology. In all fairness, Pew recognizes this and "adjusts" the data by increasing it by 5-10%, based on ANOTHER poll taken more than a decade ago.

Now, I agree both that the downturn probably had a temporary negative effect on illegal immigration and that estimating people not caught is by definition difficult. ICE/BP reports 400K apprehensions per year over the past few years and the prior White House reported that they deported roughly 400K each year (BTW, that is 3.2M deportations by the same folks that say that we can't deport 11M - just sayin').

The key thing is that they are not the same people! Some are, but most aren't. When apprehended crossing, many are just given a summons to appear and the appearance rate is naturally extremely low. The prior White House also reported a doubling of criminal deportations while the total remained constant. Were twice as many criminals crossing and if so, why? Answer - they weren't, it wasn't the same people.

So one has to think that there is an apprehension failure rate. If it is 50%, then there are 3.2M new illegals that got through (there is substantial anecdotal evidence that the failure rate is much higher). Say 6.4M total minus 3.2M deported (again, not the same people) leaves a net of 3.2M over eight years (and I think higher).

But for anyone to claim that it is flat would mean that ICE/BP caught every single attempted crossing, which clearly is not the case.

Finally, the official numbers are not correlated at all with housing starts since the downturn.

Greg G writes:


The Democratic Party has always been the major party most receptive to immigration in America. The eight Clinton years in the 90's were not a time when immigration was a hot political issue. Immigrants are always resented most in times of economic distress and that was a time of robust economic growth.

I think that, as more people get to meet and actually know immigrants, they tend to fear them less. The places in America where long time Americans are least afraid of, and most welcoming to immigrants are the very places where there are the most immigrants. The rural places that have never seen an immigrant tend to be the places where people worry the most that they will soon be subject to Sharia Law.

It is true that immigrants used to be much more eager to assimilate and lose their former ethnic identities. I suspect there is some optimum point, which is impossible to identify with certainty, beyond which more immigration does tend to work against liberal values.

Robert Swan writes:

An interesting discussion that looked at immigration from different viewpoints. Ultimately though, you're (or at least I'm) no clearer on how to evaluate a given immigration policy.

  • Russ's belief that future generations will be vastly better off than we are assumes (I think) that technological advances continue apace. I'm not so sure that they will.
  • Even if they are better off, it doesn't necessarily justify the suffering imposed today. Individuals who lose out today may, because they lose out, not have any descendents. There might be a legitimate social justice question if "the rich get richer and the poor get obliterated".
  • Ditch the word "assimilate" (make similar); instead say "incorporate" (unite in one body). Who wants a homogenised society?

    For one of my off-the-wall analogies, a body is made of many different types of cell: liver, brain, fat, blood, etc. These are distinct, but incorporated. There are other distinct cells, too, which don't integrate with the body. They're called "cancer". It may well be that, alike for cancer and fifth columnists, prevention is better than cure.

  • Russ's touching belief in formal education was again on show. I'm happy to allow that a certain amount of education is needed before a person is employable. Does more education make them any more employable? That depends.
  • Abrupt ending. Not keen to look into the ups and downs of Germany's experiment with open borders?
  • I was disappointed (well, very slightly) that the transcript spelt Prof. Borjas's invented word as "economistic" -- it would have been much more fun to spell it "economystic" (or is that reserved for monetarists).

I typed up the bulk of these points then read the comments. I see similar thoughts from several other commenters, but all have been well worth reading.
rtd writes:

I am always disappointed with economists discussing these issues such as the following:
"Somehow, you see people have no qualms whatsoever saying immigration in the short run doesn't have an effect on wages. And at the same time they say that increasing the minimum wage has no effect on employment--in the short run. As you know, those two things are completely contradictory. In the context of the minimum wage they are basically claiming that the labor market, the labor demand curve is perfectly inelastic. In the context of immigration they are claiming that labor demand curve in the short run is perfectly elastic. You know, one of those things, and probably both are wrong. Labor demand curve should probably not in that, in either extreme. And it's really sort of--I've always been fascinated by the intellectual contradiction, in sort of claiming these two things simultaneously."
Why, when talking about the real world, and not an example from a principles text, is price pass-through ignored??? Ceteris is almost never paribus in reality (I could be fairly easily coerced into believing the long-run impact, after pass-through, etc that a negligible amount of jobs are lost in the aggregate economy to min wage). I do always hate arguing from anecdotes, but I lived in Seattle when the recent minimum wage law was passed and enacted. And I distinctly recall (I have pictures to prove it) coffee shops posting signs reading "due to the recent minimum wage increase, each cup of coffee has increased $0.25" or something to that effect. I'm against minimum wages from a libertarian perspective - consenting adults entering to a contract yadayadayada - but the argument about minimum wage noticibly impacting employment in that sector over short/medium term almost as a rule (and not exception) just doesn't hold... coffee.

jw writes:

Further ICE statistics here.

It shows that ICE removals are down 40% from 2012 (from 410K to 240K). Removals are a subset of total deportations - those not needing a court order - including people apprehended within 100 miles of the border and others apprehended in the interior but previously deported. It also shows the percentage of criminal deportations growing, but with the denominator (total) declining, that would be expected. Essentially, what they are trying to show is that they are now focusing on criminal deportations.

But what they also show is:

- There was NO decrease in apprehensions due to the 2008 recession
- That ICE protocols have been changed and that they have been directed to NOT remove nearly as many non-criminal crossers (as if illegally crossing is not criminal)
- These statistics don't match the White House deportation statistics unless there was a huge increase in deportation court cases.
- ICE is only catching and immediately removing 175K people per year.

SOMETHING dramatically changed during President Obama's second term.

Greg G writes:


I looked at your link for "2016 ICE Immigration Removals." I didn't see where you are finding the data for your claim of "NO decrease in apprehensions due to the 2008 recession." Where are you finding that?

I did see this there:

"Changing migrant demographics also continued to impact ICE removal operations in FY 2016, as illegal entries by Mexicans CONTINUED TO DECREASE (emphasis added) while those by Central Americans continued to increase."

Presidents have a lot of discretion in enforcement levels and priorities on immigration and deportation decisions as anyone who followed the recent election should be aware of.

Obama chose to dramatically ramp up deportations of illegal immigrants convicted of crimes, caught near the border, and caught recently. He did this, at least in part, to make it easier to defend not focusing on deporting non-criminal and illegal immigrants and those who have been long time residents with families here.

>----" Instead of relying on MSM factoids, as above, one has to look behind the headlines."

I had to Google "MSM" in order to find out it was an acronym for mainstream media. I didn't need to Google "look behind the headlines" to know that was a euphemism for giving your imagination freer reign with alternative non-mainstream "facts" and assumptions.

jw writes:

Greg G,

See Figure 1 in the link. This is removals, which I made clear. They are clearly not linked to GDP.

My quantitative statements have been made based on government provided statistics and some fairly basic math, no imagination was necessary. If that disagrees with mainstream anything, that is not my problem.

If you have similar data to support your "those who have been long time residents with families here" narrative, please provide it. Also feel free to point out where in the law it is no longer a crime if you are here illegally for a long time or have a family (or Google "anchor baby" or "birth tourism", which are real phenomena).

Again, I support legal immigration but not illegal immigration. I fail to understand why anyone would support illegal immigration.

Greg G writes:


Figure 1 is indeed removals, not apprehensions. In fact you said:

>---"But what they ALSO SHOW (emphasis added) is:

>-- There was NO decrease in apprehensions due to the 2008 recession."

They DON'T "also show" that.

Of course legal immigration is best. You do not have to "support" illegal immigration to think that attempting to deport every illegal immigrant might be bad policy.

jw writes:

Greg G,

Mea culpa, you are correct on apprehensions vs removals for that single sentence.

Further research revealed that President Obama redefined deportations in 2011. Border removals were previously never counted as deportations but were changed that year to be included. This significantly inflated the total deportation statistics.

Greg G writes:


Thanks for acknowledging that mistake. That kind of courtesy is surprisingly rare in internet comment sections an is one of several reasons why this is my favorite internet comments section.

I haven't checked your claim about the redefinition of deportation but I'm assuming it's correct. I don't think it changes anything that we disagree about but I do see how it helped give you the non-MSM impression that you had.

Ginny Alexander writes:


With respect, I invite you to recognize the bias in your point of view due to the fact that you spend much of your time in Palo Alto or suburban Maryland among people who have good federal or federal contract jobs.

I invite you to take a train, just once, across the rust belt to see how the majority of the children and grandchildren of the farmers are faring. You interact almost exclusively with those who have made it "out."

You've had guests discussing the heroin epidemic and disability. When assessing the well-being of those children and grandchildren, you've got to factor in family dysfunction, the extremely high rate of incarceration, and rampant mental illness. What of the Brookings Institution's claim that socio-economic mobility in the US has fallen sharply since the 1980s? These are inescapable parts of life for "the average American."

I assume that unemployed people are particularly vulnerable to addiction. Life is NOT better for children of addicts. Are you able to factor that into your assessment?

I wish to express frustration with the field of economics for choosing odd metrics when talking about quality of life or "life being better" for children and grandchildren. Often the discussion is framed in terms of income, employment, or per capita GDP. Who's measuring ownership of capital, or aspects that come closer to approximating actual well-being and satisfaction?

My husband and I are great-grandchildren of farmers who had enough to eat during the great depression because they had capital (land and cows.)

Like most of our generation, we are wage slaves in an apartment. Yes, many own homes; homes that are ridiculously inefficient from a production standpoint and would not necessarily be assets during a really tough economic time. I concede that we're not at the mercy of the weather like our farmer ancestors.

You seem enthralled with technological advances that make life better. There certainly is a lot to be thankful for. But I'm really irked by the suggestion that Netflix and iPhones make us strictly better off than the farmers of yore. There are a galaxy of dimensions to consider.

Are there economists out there who are talking in terms of ownership of capital, owning the means of production? What about quality of life in terms of "robustness," or ability to comfortably weather the loss of jobs, health, or infrastructure?

I'm not saying I want government intervention as a safety net - I want attention to the ways widespread ownership and now-defunct private safety nets used to work. I also want acknowledgement for the fact that millions of Americans are NOT better off because they find themselves addicted, incarcerated, mentally ill, or caring for family members who are in such predicaments.

Luke J writes:

Good stuff

Mike Riddiford writes:

Great discussion on a controversial subject - thanks both to Russ and Prof. Borjas. Both of them challenged my pre-existing thinking in many ways.

A few comments:

Russ pointed to the empirical evidence of 20th century growth in the US as evidence of long run benefits from immigration, which is reasonable enough. However, IIRC, the US had very different immigration policies over that period, so it would be instructive to compare different periods within that longer period (a difficult task given all the other confounding variables on economic growth).

I think the 'we wanted workers' tagline is brilliant - it gets to the heart of the issue. Immigration is not just an economic issue, as important as it is in that respect, it is also a social issue, and needs to be understood in those terms. In this respect it is quite different to trade,say.

For anyone interested, another good book on immigration is Paul Colier's Exodus. Collier is a Professor of Economics at Oxford Uni

Ajit Kirpekar writes:

Its interesting. There are three effects that have been linked to the cause in low income and lower middle income(and possibly middle income as well) wage stagnation - automation, trade, and immigration.

In the former two cases - we are talking about serious reform on how to stop their insidious side effects. But no one even thinks about smashing the silicon valley machines(or more appropriately, smashing the engineers' mac books).

Frankly - just because there are losers in this tradeoff doesn't mean the net benefits are negative and that ultimately is Russ' point and one we should all agree on.

The real question is - how do we help those who are hurt? And here, there doesn't seem to be an answer. There doesn't even seem to be a proper way to adduce what is the cause - since opiod addiction is a new phenomenon and immigration, trade, and automation has been a long secular trend.

Unfortunately, the anti trade anti immigrant rhetoric is being thrown around as a real "solution" when in fact it is no more of a solution than job retraining or lecturing about the importance of college.

It would be nice to give benefits to the losers of trade, but then you breed ugly dependence and it becomes a narcotic. Have we not seen what has happened to the black community since the war on poverty?

Bottom line - you're not going to bring back wage growth by stopping illegals or cutting off trade. I doubt you would do it even if you had an army of people smashing macbooks. Its inevitable, we'd better get use to it.

Amir writes:

Great podcast as usual ,well done russ.

Some where in the podcast borjas talked about the work he has done recently on the impact of immigrant Russian Mathematicians on the productivity of American Mathematics market
“Borjas, George J., and Kirk B. Doran. "The collapse of the Soviet Union and the productivity of American mathematicians." The Quarterly Journal of Economics (2012)”
I went through the article and found it interesting , in the first page these sentence showed up
“We have benefited from comments made by many economists and mathematicians, including ……..”
My question is how we can calculate these benefits ,spillover effects, its not in any databases : ISI , AMS, etc , at least up to my knowledge .other fact that worth mentioning here is the teaching impact ,for example in my studentship I take course in Differential Geometry with prof' " …… ov " who is a Russian émigré and learned something from him ,tacit knowledge - and I ended up working in a totally different subject ‘Commutative Algebra’ , I will never mention his name even in my paper’s acknowledgement but he has impact on my career.
Hope these examples show how difficult is to quantify the impact of immigration on the host market

john penfold writes:

Another great interview. Save me so much time and feelings of guilt for not reading all the important books Dr. Roberts brings to us. Thank you. Can't get in a car without loading up on podcasts, sometimes I just run errands to catch up.

First time I've agreed with the guest more than with Dr. Roberts. His real life experience clearly helped him approach the subject with broader insights.

Dmitry writes:

Dear Russ and George,

Thank you for an interesting discussion.

As an argument in favor of immigration you cite tremendous growth for the past 100 years and I agree with that.

But the devil is in the details: "... it came from 3 different sources. We had tremendous technological change. We had a huge increase in globalization of goods and capital. We had a huge increase in immigration and mobility. And we also had population change."

It is really hard to separate the influence of each of those factors. But one thing that we have seen historically is that some countries achieved tremendous growth without much immigration:

1. Take the economic miracle of post WWII Germany, which until recently hadn't had much immigration inflow going on.

2. Similarly, take Asian miracles - Japan and South Korea. As far as I know, pretty homogeneous societies.

Other European countries like France and Italy have done rather well, again being pretty mono-cultural until recently.

You have to somehow explain those outliers. To me this says in favor of the argument, that it is technological change and trade which matter. I am ignorant of current research of course, so I would be happy to have a look at papers describing empirical evidence in favor of immigration proving me wrong.

But then the cultural reverberations of immigration create a huge problem, costs of which might far exceed the benefits (even if they are as large as you think).

Any way, I don't think the picture is as clear as you, Russ, are portraying it.

[Just as an FYI, I am an immigrant myself.]

Jared Szymanski writes:

As jw astutely pointed out:
"You can keep dreaming that better schools and more education funding will somehow overcome exploding single parenthood and family breakdown, but it won't. It also won't fix the fact that by definition, 50% of people have IQ's below 100. With agriculture, manufacturing, blue collar service work, etc, they can have solid and fulfilling lives, but as Charles Murray points out, they aren't going to become rocket scientists."

It's unrealistic to hope that if we just had better education, or re-education then the low skilled people could become high-skilled people and get better jobs. I remember another podcast in which the discussion was about whether the education system was a sorting machine or a transformation machine. Even with the ideal education system I think there would remain a significant component of sorting. Some people just aren't endowed with the kind of drive or intelligence it takes to become higher earners, no matter how much education you try to impose on them.

That's the argument for some social safety net that essentially compensates these low-skilled workers for harm they suffer from immigrants lowering their wages. We collectively benefit so much from trade, immigration and technology that it is foolish and impossible to try to stop these forces, but perhaps we should find some way to cushion the effects on the low-skilled who are harmed by it.

Martin Dertz writes:

This has been a great discussion to read through. Kudos especially to Greg G & JW.

My only thought/question not addressed by comments so far is that -- to me at least -- cultural and linguistic assimilation means homogeneity. And a homogeneous, complex system makes a fragile system. Are culture and language somehow different than other system features? I would think a highly diverse -- ie high entropy -- population is desirable for long-run system success, right?


@Ginny Alexander:
You very well describe the limitations of economist arguments, where all things either have value measurable in dollars or no value at all. As Bogwood said the conversation could use an ecologist. I suggest Aldo Leopold, especially his Sand County Almanac.

@Jared Symanski:
The argument for a safety net (well, mine at least) is it allows vulnerable people to transition after shock. The only people who argue against a safety net are those optimistic (or dumb) enough to think there's 0 chance they'll ever need it. For whatever reason -- labor supply shock, tech revolution, natural disaster -- you lose your job/business. If no safety net your future's bleak. With one you've got a shot.

John Alcorn writes:

I would like to steer the EconTalk audience to two rebuttals of the case, made by George Borjas and others, against more open immigration.

1) A paper by Michael Clemens & Lant Pritchett:

The New Economic Case for Migration Restrictions: An Assessment (Center for Global Development, Working Paper No. 423, February 23, 2016).

Here is the abstract:

"Migration barriers have complex effects, among which is a cost to global economic efficiency. A recent research literature has asserted that, far from having an economic cost, migration barriers may in fact enrich the world economy. It is claimed that barriers do this by stopping the spread of impoverishing ‘culture’ or ‘institutions’ from poor to rich countries. This is the new economic case for migration restrictions. We assess the economic theory and evidence behind this claim. While it is possible in principle for such effects to arise, they would occur at orders of magnitude higher migration rates than presently observed. That is, the new efficiency case for some migration restrictions is empirically a case against the stringency of current restrictions."
2) A blogpost by Bryan Caplan, next door at EconLog:

Trillion Dollar Bills on the Sidewalk: The Borjas Critique (EconLog, July 16, 2014).

David Zetland writes:

I was surprised that the discussion of "falling wages" did not move to "falling cost of living" as a great fall in COL would more than make up for a fall in wages, for many workers.

Sure, nobody wants lower wages, but the harm is surely lower if cost deflation outpaces wage deflation.

John Alcorn writes:


"George Borjas: And the important thing for policy is you really have to take that into account before you let in too many people, for example. Like Germany did."

Compare the analysis and findings by John Kennan, "Open Borders in the European Union and Beyond: Migration Flows and Labor-Market Implications" (NBER Working Paper No. 23048, January 2017). The abstract is pasted below:
"In 2004, the European Union admitted 10 new countries, and wages in these countries were generally well below the levels in the existing member countries. Citizens of these newly admitted countries were subsequently free to take jobs anywhere in the EU, and many did so. In 2015, a large number of refugees from Syria and other broken countries sought to migrate to EU countries (along very dangerous routes), and these refugees were met with fierce resistance, at least in some places. This paper seeks to understand the labor market implications of allowing free migration across borders, with particular reference to the EU. The aim is to quantify the migration flows associated with EU enlargement, and to analyze the extent to which these flows affected equilibrium wages. The main conclusion is that the real wage effects are small, and the gains from open borders are large."
A preliminary, ungated draft of Prof. Kennan's paper (September 2016) is available here.

Max writes:

I really enjoyed the discussion, but I have to disagree with Russ' assertion that it is possible (even likely) that the new generation of immigrants have less of an incentive to learn the local language, due to new environments that promote speaking Spanish.

My anecdotal evidence that this is in fact NOT happening involves the time I have spent working at a dual language public school in Oregon. What surprised me from working at the school was that immigrant children from a young age are able to discern the social benefits of speaking English. In fact, the dual language program (where students receive classes half the time in Spanish and the other half in English) functions as much as a tool to preserve the Spanish-speaking culture of Hispanic students as it does to assimilate these immigrant students into an English-speaking society.

I was shocked to find that many children of Mexican and Central American immigrants barely speak Spanish, even though their parents don't speak a word of English. That's right, the parents speak exclusively one language (Spanish) while their children speak exclusively another language (English). People must not believe me, thinking "how do they then communicate?" Well, in many cases the parents speak to their children in Spanish, and the kids are able to understand, but respond in English. Both parents and children understand the rudimentary elements of the language to be able to get by without actually being able to speak the other's language or even understand it outside the specific context of the family home.

This, I believe, is the reality that many people fail to see when they are confronted with Spanish advertisements and customer service in Hispanic immigrant sections of towns. None of that is new, such things happened with German and Italian immigrants before them. And I believe similar social and economic incentives led the children of those immigrants to quickly identify English as a language with much greater returns, and quickly adapt.

Now, there is always the argument that "This time is different", but I challenge anyone to give me an example of an immigrant community in the United States that was so stubborn as to maintain their original language while actively avoiding speaking the local language (English). We have had over 200 years of immigration, from countries with languages as diverse as Chinese, Polish, Italian, German, Vietnamese, etc. etc. Not one of those communities was able to completely keep its language over time. In most cases their original language dies off quite quickly. How many grandchildren of these original immigrants are able to speak the language that their grandparents learned from birth? Very few.

I think the trend will continue, because incentives matter, and there is no greater incentive in the United States than speaking English. And for children it is much easier to identify and adapt to that incentive than it is for their parents, who are stuck in the path dependency of their mother tongue.

Kevin writes:

Thanks all for a great discussion.

I think culture overwhelms any impact economics has. The idea that you are American if you believe a set of ideas and those who assimilate to those ideas are Americans is known as a proposition nation. However, I became dubious of that idea the longer I have watched native Americans reject whatever the supposed proposition is with radical legal and cultural changes. Russ cannot persuade even a tiny fraction of people who grew up with his shared culture that the "idea" of freedom is the winner, but we believe we will convert people from heavily socialist countries with wildly different political cultures? From a purely political view of wanting my country to be more free and looking at the voting habits of recent SAmerican immigrants I would strongly argue that we should stop ALL immigration from those countries for decades. They vote in a way that will reduce my freedoms and my childrens freedoms, why would I want them here?

The process of assimilation is likely a function of the volume of immigration. We have had a long wave of immigration that has overwhelmed institutions and the ability of many locations to assimilate. Turning off illegal and even legal immigration seems reasonable until America stabilizes. To illustrate this here is the percent of US citizens who are US immigrants.

You can see that once the percent got high enough the US citizens realized it was disruptive and ceased the flow of immigration until the percent was reduced significantly. I argue that we are simply at that point again, its time to quit immigration for awhile and get our own house in order.

American is in the midst of a very serious struggle for freedom with a portion of the political spectrum already turning to violent demonstrations. I think we need a breather from further social stresses from immigration.

jw writes:

John Alcorn,

I read the Caplan link, and while I don't agree at all with his reasoning, I was amazed at the final sentence! He was going to have a meeting for open borders proponents, but wanted to strictly vet any attendees with opposing views!

If this didn't completely validate Borjas' "wanting (attendees) but getting people" thesis, I don't know what would.

John Alcorn writes:


Bryan Caplan explains the distinction between trespass and migration in another blogpost (at an embedded link in the blogpost that you read):

"Immigration, Trespassing, and Socialism" (EconLog, November 12, 2012).

Kevin writes:

Relistening to the discussion, Dr. Roberts asserts that because of advancements in the world the children and grandchildren of people who lose their jobs second to immigration will still be better off.

I think this is true, but has nothing to do with immigration. I saw no evidence presented that immigration contributes to an increase in GDP per capita, especially the majority of the immigration the US has experienced in the last 20 years - mostly low skilled. Trade probably does, technology obviously does. The grandchildren will be relatively richer whether or not there is immigration. Many countries with stable governments and trade have grown richer in the past 100 years whether they allowed immigration or did not. I would be interested in some data showing that countries with differing levels of immigration have differed in relative wealth creation after accounting for other freedoms of trade, innovation, and good government policies. What keeps countries rich or poor seems independent of their immigration policies and much more due to their ability to adapt innovations in technology with basic good government.

Marilyne Tolle writes:

Interesting NBER WP on the general welfare effects (positive) and distributional consequences (which depend on the substitutability or complementary of immigrant and native workers) of immigration through the lens of the IT sector.

jw writes:

John Alcorn,

I read that link as well and don't agree with it at all. He is taking an analogy and splitting hairs on the exact definitions out of context.

He skips over the first crucial distinction - an illegal immigrant occupies his space in a country WITHOUT the country's consent.

John Alcorn writes:


Reasonable people may differ, about whether peaceful migration requires a country's consent. For example, Nathanael Smith, at this link (and embedded links) rebuts the theory of collective property rights of citizens in a country's public roads, and makes the case for the right to migrate.


Return to top