Muchas gracias. Voy a empezar en español pero cuando llegue a mi presentación formal voy a cambiar al inglés. Quiero empezar en español porque soy economista, y a los economistas nos gustan los modelos, y la teoría, y todo tiene que ser organizado perfectamente.
Quiero comenzar con una historia de inmigración que pasó en Miami, a siete cuadras de aquí, en la “Freedom Tower,” cuando yo y mi familia llegamos de Cuba hace como unos cincuenta años. Me acuerdo que el día después de que llegamos tuvimos que ir allí a llenar unos formularios y luego iban a decidir qué hacer con nosotros. La persona que estaba a cargo de nuestro caso quería que mi mamá se fuera para California.
La conversación entera fue para convencer a mí madre que no debería quedarse en Miami y debería mudarse para California. Y yo … en mi mente — era un niño — pensaba en la piscina que íbamos a tener en la casa en California, y las estrellas de Hollywood iban a estar ahí conmigo, y cosas como esas. Pero cuando mi mamá estaba ya casi lista para decir “Sí, vamos pa’ California,” en ese momento una mujer en línea detrás de nosotros dijo: “No vayan pa’ allí, que hay terremotos.” Y ese fue el final de California.
La historia es interesante, porque en la teoría económica todos siempre tenemos un epsilón (ε), un error que surge de variables no identificadas, y en mi caso, mi epsilón siempre ha sido terremotos (earthquakes).
La vida entera cambió por un evento aleatorio (random event).
De la misma manera quería compartir con ustedes otra historia de inmigración que también es verdad. La historia … goes back to when Jimmy Carter was president, in January 1980. This actually happened, because I know somebody who was in the Oval Office and he told me the story. Jimmy Carter invited Deng Xiaoping, who was the vice-premier of China, to visit the White House. This was the very first time that a Chinese at that level had visited the United States. The reason why Jimmy Carter wanted so badly the China visit was his concern about human rights, and especially human rights in China. Jimmy Carter had with him literally a binder of human rights complaints that he wanted to bring up with Deng Xiaoping, addressing all kinds of human rights issues in China. Deng Xiaoping and Carter are speaking in the Oval Office and Carter opens the binder and starts going page by page complaining about human rights in China.
Eventually they get to the page that has to do with emigration. As all of us know, communist countries did not allow their people to leave easily for many, many years, and Carter told Deng Xiaoping: “How can you possibly think you can join modern civilization, with modern human rights, if you don’t let your people go?” Deng Xiaoping sat back, thought about it, and smiled. Then he said: “You know, Mr. President, you’re right. Absolutely right. How many Chinese nationals do you want? 10 million? 20 million? 30 million?” That was the end of the discussion.
This actually happened. The reason that it is important to tell the story, like the other story, is because not only do random events that economists could never really model sort of drive migration flows and waves. Deng Xiaoping put his finger on a question that any industrialized nation today has to answer: “How many immigrants are we willing to admit?” There are arguments over the fence, over amnesty, but nobody really argues, or even talks, about that fundamental question.
Many, many more people want to come to the U.S. than the U.S. would ever be willing to admit. So the question is, how many are we willing to admit? This is a reasonable question to ask. Maybe economics has something to say about it. And just suppose that Jimmy Carter had actually thought about this before his meeting with Deng Xiaoping, and decided that admitting maybe 10 million Chinese nationals over the next 10-20 years would not be so bad. He could have told Deng Xiaoping: “10 million, that’s my answer to the question.” But Deng Xiaoping could have said: “Which 10 million do you want? We have a billion people to choose from.” And that is the other central question on immigration that we never talk about either.
Immigration policy really consists of two questions: How many people does a country want to admit, and out of the many who want to come to a country, which formula are you going to use to select the winners and the losers? What method of discrimination are you going to use to allocate visas?
U.S. IMMIGRATION POLICY
I want to start with a little bit of the history of U.S. immigration policy. A lot of people, even people that I have had in my class, who presumably make immigration policy in Congress — as senators and congressmen — don’t even know how U.S. immigration policy works.
I want to show you what we think we know about economics and immigration and then return at the end to those two questions I raised earlier and ask: what does economics have to say about the kind of policy we should have? Let me start by giving you a little bit of history and put it in context.
In 2010, about 214 million people, or about 3.1% of the world population were immigrants. I know in Miami it doesn’t sound like a lot — the number here could be 40% or whatever — but 3.1% of the world’s population is a lot of people. It only used to be 1% in 1980. So there has been a really remarkable increase. And equally important, the U.S. is by no means exceptional anymore when it comes to this. Table 1 shows that a country like Greece is 10% immigrant, Germany 13%, France 11%, Sweden 14%. The U.S. at 13.5% is in the middle of the pack. The fact is that the most industrialized economies are now what we used to call the U.S.: they are nations of immigrants. It is a worldwide phenomenon; large-scale immigration is prevalent all over the industrialized world.
Table 1. Share of the Population of “More Developed Regions” That Are Immigrants (percent)
Figure 1 shows historical legal immigration to the United States from the 1820s to the 1990s. The crucial qualifier here is legal. We have kept records of legal immigration since the 1820s. The way to think about this graph is by imagining the immigration officials in Washington mailing out green cards. The graph shows how many green cards were mailed out in each decade since the 1820s. The big peak around 1900 is the one we all know from the Hollywood movies —Ellis Island, the boats, people looking up at the Statue of Liberty, crying and so on and so forth. And you can see the reaction to that very quickly; by the 1930s almost no immigrants were coming into the U.S. legally. Since the 1930s, the number of legal immigrants has been going up very rapidly; right now we are admitting over 10 million people legally per decade. That’s basically a million people a year admitted legally.
Figure 1. Number of Legal Immigrants to US, by Decade
The history of U.S. immigration policy has been told in many books and articles. Ignoring a lot of unnecessary details, the main immigration policy regimes are:
- Before 1875, no restrictions.
- Between 1875 and 1924, Congress got involved and began to add to a list of people who should not be allowed into the U.S. (“excludables”), including Asians, convicts, idiots, political radicals, public charges. None of those words are my words; those are the words used in the law. I don’t call them “idiots”; that’s the word the law used. So anybody who was an Asian, or a convict, or an idiot, or a radical, or a public charge, was not admissible.
- By 1924, that list was actually very, very, very long. And combined with the first great migration, it had a reaction. Congress said for the first time: “Let’s make a new law” and answered those two questions I posed earlier. They created what was called the National Origins Quota System. That system put in place a numerical limit on admissions and a system to allocate visas based on “national origins.” At the time, two-third of the U.S. population was of either British or German ancestry. That means that two-thirds of the visas went to Great Britain and to Germany. It was a pure racial system. Depending on what the national origin of the U.S. population was, that country would get so many visas. That system was in existence for 40 years.
- In 1965, as part of the civil rights movement, it was decided that the National Origins Quota System was discriminatory — which it was, by definition, because it limited people in getting visas — and to replace it with the system that we have now, the Family Preference System. Again, the number of immigrants increased a little, but basically what changed was the way of allocating visas. Instead of using national origin to separate out the winners and losers, they used family preferences. For most people in the world today, about the only way to come to the U.S. legally is by having a family member already here sponsor their entry. Absent that, it is extremely difficult to obtain a green card.
Classes of Admission
In the last decade the U.S. admitted 10.5 million people, a million per year basically. The four major parts of the current law and the number of persons admitted are:
- Family Preferences, which I just talked about and will come back to later, which accounted for 6.8 million, about 70% percent of immigrants to the U.S.
- Employment-Based Visas, which accounted for about 1.6 million visas; actually this is a misleading category because if you get an employment based visa you can also bring your spouse and your children, and therefore your espouse and children count as part of those 1.6 million. So the number of visas actually given to skilled workers or to employees is actually much less than that.
- Refugees — many of us in this room came through that category, including myself. Refugees and asylees accounted for 1.3 million.
- Diversity Visas: In the U.S., believe it or not, we have a lottery. Anybody in the world can apply to this lottery. And if you ever want to see it at work, check the website. The State Department opens up a website with open enrollment for the Diversity Visa and anybody in the world can apply. We raffle out 45,000 visas a year. Anybody can win. Just to give an idea of the demand for entering the U.S., for the 45,000 visas that have been raffled the last couple of years, we’ve received between 10 and 15 million applications per year. That gives you an idea of why a limit has to exist. Absent a limit, the number of immigrants entering would be really quite dramatic.
Returning to family preferences, the way this section of the current law works is that if you are a U.S. resident or citizen, you are entitled to sponsor the entry of certain people in your family. For example, if I were a young man, and suppose I went to Cuba as a foreign student and met a nice Cuban girl that I wanted to marry and bring her back. That Cuban girl would be allowed to come in with me because she would be my wife. That’s part of the Family Preferences. But what’s really interesting is that within five years my wife will become a citizen, which means her siblings are now qualified to enter the U.S., as she can sponsor their entry. And once her siblings come, that means the siblings’ spouses can come in as well. But if the siblings’ spouses can come in, so can the siblings’ spouses’ parents. And so can the siblings’ spouses’ parents’ siblings. So the siblings preference actually opens up the way toward a triangle, a pyramid, an open-ended pyramid.
A mathematical friend of mine actually worked out a formula to find out to how many people in the world Family Preferences would actually apply if you would let this go on for 100 years: the entire world qualifies for entrance in a hundred years. This cannot happen, however, because there are numerical limits in the law. To give you an idea again of how crazy this whole system is, if you applied for entry for your brother around 1991, from the Philippines, your number would have just come up. We have a 20-25 year waiting period for this kind of visa.
Remarkably, if the Comprehensive Immigration Reform Bill right now before Congress passes, the siblings preference would disappear. Even though you do not hear about this in the press, that is one of the crucial changes. Siblings preference would be replaced by something else, that I will talk about at the end.
In addition to legal immigration, we also have illegal immigration. The Department of Homeland Security (DHS) annually provides an estimate of how many illegal immigrants live in the U.S. (Table 2). These are the official illegal immigration estimates.
Table 2. Illegal Immigration (DHS estimates) (millions)
There is a big debate over how many illegal immigrants there actually are in the U.S. Let me tell you how DHS makes the estimates because that is part of the debate as well. We obviously don’t go out and count how many illegals there are. We keep very good track of how many green cards we mail out. We apply a mortality table to the number of green cards mailed out and that gives a number of how many foreign-born people should be alive in the U.S. today. Put that number on one side of the page. The Census Bureau goes out and asks people: “Are you foreign born?” This gives you a count of how many foreign-born people are living in the U.S. You take the difference and that’s the number of illegal immigrants. Basically, the numbers in Table 2 imply that the Census Bureau, when they went out and counted people, found about 12 million more people in January 2011, than they expected to find given the number of green cards they mailed out.
Right there you can see a problem. The problem is that when the Census Bureau goes out and counts, they can’t count everybody. And it would not be unreasonable to presume that the number of people they can’t count probably includes people who do not want to be counted, particularly illegals. So therefore, they have to make an assumption about what the undercount rate is. If you open up the report and you read the text you will not find the undercount assumption, but it you go to the footnotes, in very tiny print you will find the assumption they make and it’s an assumption they have made forever, based on no data whatsoever: the assumption is that the undercount rate for illegals is 10%. That is, we only miss counting 10% of illegals. If you believe this undercount rate is too high, or too low, you can blow up these numbers accordingly and that’s what part of what the argument is about.
Figure 2 does not use fancy econometrics. It reports just facts. It is basically gives the wage gap between immigrants and natives in each census. Believe it or not, back in 1960, just before some of us in this room came here, the typical immigrant in the U.S. earned about 5% more than the typical native in the U.S. You extend forward 40 or 50 years, the typical immigrant in the U.S. is earning about 20% less than the typical native in the U.S.
Figure 2. Percent Wage Gap between Immigrant and Native Men (age-adjusted)
And you might say, why do we care? Well, we care because this is reflecting something about the skill distribution of immigrants. And if there’s anything we’ve learned about economics is that it is the interaction between the skill distribution of immigrants and the skill distribution of natives that determines the impact of immigration. Believe me, we would be having a very different kind of debate today if that line were upward-sloping instead of downward-sloping. It is the downward-sloping nature of the line that is really driving a lot of the debate.
It is clear that a lot of people are moving. An important question is “who moves?” What kind of selection is going on? One of the remarkable things about immigration is that not everybody moves, even if they could. There are many, many examples in the world where there are no legal restrictions preventing people from moving and people still don’t move.
I’ll give you two examples. One is Puerto Rico. Puerto Rico’s wage is about 30%-40% lower than the U.S. wage for equal skills, even after controlling for price differences. If the world was composed of income maximizing rational human beings, Puerto Rico should be empty right now, but it’s not. A lot of people left, some 30% since World War II, but even more chose to stay.
Another example is Germany before the Berlin Wall was built in 1961. Before the Berlin Wall was built, people already knew what East Germany and West Germany were like. All that an East German had to do to emigrate to West Germany — before the Berlin Wall was built — was literally get to Berlin and cross the street. Once in West Berlin, he or she could take a train or a plane or some other means and go to West Germany. That’s all it took, no documents of any kind, no permission of any kind, and yet people didn’t do it.
There are many other documented examples of people who are free to leave choosing not to do so. If you believe in the economic model, migration costs must be huge for this to happen. Given that only some people leave, and most people don’t, what is that telling you about selection? The standard economic model basically tells you that the kinds of people who leave will depend on the payoff to leaving: if the country has a high rate of return to skills, skilled people would not want to leave their country, and it would be the unskilled who would want to leave. In contrast, if a country has a very low rate of return to skills, then the people who are really skilled are being taxed very heavily and they will want to get out.
This is actually not a bad description of the world. This is what happens if you compare the earnings of immigrants with the Gini coefficient of the source country (Figure 3). The Gini coefficient more or less measures the return to skills. In countries that have very equal income distributions — for example Japan and Sweden at the top left of Figure 3 — people who leave those countries tend to be highly skilled, very highly skilled compared to the people who stay. This is not surprising: if you are skilled in Sweden, their system, for whatever reason, does not really reward your skills very much while the U.S. will reward you more. However, if you look at a country at the very lower right, like Haiti, the return to skills in Haiti is probably pretty high. If you are well off in Haiti, you are really well off. Unfortunately, most people are not. And the people who tend to leave Haiti are not the most well off people. That’s actually just the data, again not using any fancy econometrics. It’s just the data showing you what happens if you compare the entry earnings of immigrants to the U.S. as a function of the measure of the return to skills.
Figure 3. Entry Wage of Immigrants and Gini Coefficient in Source Country, 1995-99
A second question in every immigration debate all over the world is: What happens to the labor market in the receiving country when the immigrants get there? I’d like to start with a quote from Paul Samuelson, the author of the textbook that most people my age, or a little older or younger, used when taking economics in college. I have taken the quote from the 1964 edition of Samuelson’s Economics textbook. The reason I picked that year is because it was before 1965, when U.S. immigration policy changed. Samuelson wrote: “After World War I, laws were passed severely limiting immigration. Only a trickle of immigrants has been admitted since then … By keeping labor supply down, immigration policy tends to keep wages high.” Supply and demand at work — something we teach in Economics 101. This is actually a very controversial thing to say, however. Because if you believe that statement and you read it from today’s perspective, it would say: “Since 1965 laws were passed liberalizing immigration. Many, many immigrants came. Therefore by keeping the labor supply high, wages were kept low.” This would be very controversial to say today in the U.S., that somehow immigration would have a negative impact on the wage structure, even though it’s a simple-minded implication of supply and demand that we teach on the first day of Principles of Economics class. Whether this is true or not is a different matter; this is what the theory tells you.
In the last 30 years many labor economists have spent a lot of time trying to determine if this is true or not. The best data on this relationship is plotted in Figure 4, a scatter diagram relating wages and immigration by skill groups for 1960 to 2010. The figure plots wage differences for a skill group over a decade between 1960 and 2010, as a function of how many immigrants went into that skill group. What you see is that the groups that received the most immigrants tended to have lower wages or lower wage growth. In other words, the demand curve — if you want to think of it that way — is downward-sloping. The groups that experienced the biggest shift in supply experienced the smallest wage growth. Now the slope of that relationship is not that large: a 10% increase in supply lowers the wage by 3% or 4%. So, in the big scheme of things, we are not talking about a huge impact. But recall that the kinds of immigrants coming in today have changed, with most of them coming in at the two ends of the skill distribution. The U.S. actually gets many, many highly-educated immigrants and many, many, many more low-educated immigrants, and very few people in the middle of the skill distribution. So, the supply and demand argument would tell you this impact is really most dramatic among the groups that had the largest supply shifts, which happen to be low-educated people.
Figure 4. Wages and Immigration by Skill Group, 1960-2010
Table 3 shows the predicted impact of 1990-2010 immigrant inflows on wages. The first column tells you the impact in the short run. Thus, for high school dropouts, basically it is about a 6% wage drop in the short-run, in other words overnight. Over time it goes down to 3% because during a longer time period, the labor market adjusts, capital adjusts, and so on and so forth.
Table 3. Predicted Impact of 1990-2010 Immigrant Influx (Theory-Based Analysis) (percent)
There is a crucial distinction between the data behind Figure 4 and Table 3. Figure 4 contains data from the Census Bureau. Nothing fancy going on here. Meanwhile, behind Table 3 there is a lot of fancy econometrics and that’s why the table states that it originates from “theory-based analysis.” Personally, I think that Figure 4 is much more trustworthy than Table 3 because the latter involves all kinds of other assumptions, for example, Cobb-Douglas aggregate production function. But nevertheless, it is telling you that the impact is going to be concentrated at the lower end of the skill and education end, which is also what Figure 4 is telling you.
GAINS AND LOSSES FROM IMMIGRATION
On net, the U.S. makes a small gain from immigration. Workers lose 2.8% of GDP, or $400 billion, while employers gain 3.0% of GDP, or $430 billion. The net gain is relatively small, around $30 billion dollars per year, or about $110 per native-born person. While $30 billion is a substantial amount of money, it is really a rounding error in comparison with the U.S. economy, which has a multi-trillion dollar GDP.
The core of the immigration debate is about distribution. Once you realize this, it puts the whole debate over immigration in a new focus. Some people win from immigration and some people lose. The people who lose are those who compete with immigrants; those who win are people who “use” immigrants. I mean “using” in a very broad sense — the user could be the employer, or it could be me when I go to the supermarket. Whoever uses immigrants to some extent, gains; whoever competes with immigrants, loses. That’s what the debate is about.
Let me turn to the “public charge” issue. Again it is very nice to put this issue in historical context. The public charge concern over immigration goes back to 1645, 130 years before the Declaration of Independence in 1776.
- The first public charge restrictions were put in place in the very liberal and very enlightened Massachusetts colony in 1645 and 1655. The orders prohibited the entry of poor immigrants.
- In 1691, the New York colony hired an economist to look into the issue of immigration. This economist said to the New York colonists: “the way to avoid getting poor immigrants is by putting up a bonding system,” and they actually enacted such a system into law. If you wanted to bring in your relative into New York after 1691, you had to put up a bond with the government, and that bond would be funds that the government would use if the relative brought in got sick and needed some kind of help.
- In 1882, Congress banned the entry of “any persons unable to take care of himself or herself without becoming a public charge.”
- In 1903, Congress approved deportation of immigrants who become public charges within two years after arrival “for causes existing prior to their landing.”
As it now stands, the law is that public charges are not admissible and when immigrants become a public charge after entry, they are deportable. Needless to say, none of that is ever enforced, because the data tells you a very different picture.
Figure 5 is based on data from the Current Population Survey (CPS). Again, no fancy econometrics. The CPS began in 1994 to ask about immigration background. The figure shows the fraction of native-born and foreign-born households receiving assistance: cash assistance, food stamps, Medicaid. Currently, something like 20% of native-born households receive one of the three types of aid and so do 30% of immigrant-households. Those are the facts. I am not saying that the U.S. is a welfare magnet. I am not saying that these immigrants come to take advantage of the U.S. I’m not saying any of that; just presenting facts.
The U.S. has developed a very highly developed welfare system, designed to help particular kinds of people when they need help. If you let in a lot of low-skilled people all kinds of things happen: people get sick, they move around, they have families, they have children. These were not the original reason for migrating; it’s just that we have a system designed to help people in certain circumstances.
What does all this imply for immigration policy? I’ve thought about this for a long time and I have a very simple answer: nothing at all. We economists tend to be very misleading, if not actually dishonest, about what our work means for public policy. Let me tell you what I mean by that. Suppose you believe everything I just told you—immigrants come in, they tend to be low-skilled, they have an impact on low-skilled workers, they use welfare, and so on and so forth.
Figure 5. Trends in Percent of Households Receiving Assistance (cash assistance, food stamps, or Medicaid)
Let me put on a hat of defender of native-born interests. By native-born I actually mean something much broader — since many of us in this room are immigrants, we are trying to set up a system as of right now that would close the door to more immigrants. So by definition all of us are natives. What should we do tomorrow afternoon? Suppose I happen to care about the people in this room; the people in this room are my concern. If this were the case, then I don’t really want to add a group of people who are going to hurt the least-skilled amongst us or a group of people that is going to increase our taxes, right? All the facts that I have given you point to an immigration policy of enforcing the law — public charges should be deported (they are deportable) and we should not let in any more low-skilled people because they increase inequality. That’s one implication of the data.
Let me take off that hat and put on another hat, one that says “I really care about poor people all over the world.” I’m very humanitarian minded — poor people wherever they are really make me feel for them, I care about them a lot. This implies very different immigration policies. Depending on the hat I put on, my objective function changes.
This is where economists are never really very honest. Policy doesn’t come out of data. Policy comes out of values. And it’s your values that really determine what we should do tomorrow. If your values are that you care about who your neighbors are — your neighbors right now — that implies a particular set of policies. If by neighbor you mean a much broader, global perspective, it implies a different kind of policy. And who am I to say which is the right perspective? I am nobody; I have my opinions on this topic but they are just my opinions. They are not informed by any data. Each of us has his own set of values and we can never agree on those. We can agree on the facts, but we can never agree on what to do with the facts because what to do has nothing to do with the facts — it has to do with values.
With that in mind, it is worth returning to the questions we posed earlier: how many immigrants to admit and which kinds of immigrants to admit. There is actually nothing in economics that tells us about the “how many.” To be sure, the easy answer for an economist is “until marginal cost equals marginal revenue.” But how do you measure this? There is nothing in the literature that would tell us whether a million new immigrants right now is too many or too few. There is just nothing available empirically to determine that.
The “which kinds” we know more about, but again that depends on what you have in mind. To show you what I mean by that, take another country as an example. A great thing for all of you to do when you get back to your hotel rooms or to you home tonight is to search the internet for “Canada immigration.” That will bring you to the Canadian immigration website. Act as if you want to move to Canada and a web page will open up that will give you a test which will be instantaneously graded. They will ask you: “How old are you?” If you respond over 55, like myself, too bad: “We really just want younger people.” They will ask you, “How much education do you have?” I’ll say, “Ph.D.” — great answer! If you respond “high school drop-out,” they say, “go to the U.S.” They will ask you, “What do you do for a living?” If you say “chef,” that’s a great answer; they seem to looking for cooks all the time. Every time I open up the website they want cooks. If you respond say, “economist,” not such a good answer.
The point is somebody in Canada has thought through what it is they want to accomplish through their immigration policy. And what they want to accomplish is to attract immigrants with a particular set of characteristics that would improve the well-being of the native population. They are not embarrassed to say it: we want to maximize the economic well-being of the native population, again not the population born in Canada, but people in Canada today looking forward to the next 20 years. With respect to the U.S., we don’t admit it, but we also ask questions and we also have a point system.
Before we leave Canada, there’s one more thing I should tell you. At the end of the website test, the system adds up all points associated with the answers and gives you a final score. For example, it will say: “This month the passing grade is 68. If you got more than 68, go see the consular officer for your city. If you got less than 68, sorry.” They actually change these numbers all the time. So not only have they thought this through, but depending on economic conditions, they will change matters accordingly.
In the U.S. we cannot do that. Everything is written into law. We have to have the same immigration policy whether the unemployment rate is 10% or 2%. It doesn’t make any difference. Everything is written into law, and as you know by looking at the debate right now, it is very hard to change immigration policy. It does not change that often. The House has to approve a change, the Senate has to approve it, and the President has to approve it, all at the same time. This is a very hard set of coincidences for it to happen. So, we are stuck with whatever we put into law for 50 years. And if we change the law this year, we will be stuck with the new law for 50 more years.
Let me tell you a couple of things about the current debates that are interesting. In the current immigration reform debate, as I said earlier, the siblings preference is likely to disappear and it is going to be replaced, believe it or not, by the Canada-style points system. If this issue has not surfaced in the press it is because the press is fixated on the amnesty aspect of the law. But the plan is to put in place a point system that depends on education and experience. This system could actually distribute up to a quarter-million visas per year once it gets fully going. If this actually happens, for the first time ever, the U.S. would actually begin to move towards adopting a Canadian-style policy. And that, I think, is far more important than the whole issue of amnesty. Because that would set the U.S. in a path that would be very, very hard to change in the future. Given everything that we know today, the future demand for highly skilled workers is going to be very high. We are getting rid of the siblings preferences that allows for an exploding pyramid and we are going replace it with a larger point system if immigration reform actually goes through.
As I said earlier, we should do what’s best for us. But each of us has a different opinion on it. Let me tell you how difficult this question really is. Think of the world as being composed of three populations: (1) U.S. natives; (2) immigrants; and (3) rest of the world. Most people in the U.S., when they think about immigration policy, really have U.S. in mind. We should do what’s best for us. Canada, for example, has clearly thought about it and set out what is best for them. The fact of the matter is that any immigration policy you set is going to make the lucky few who get the visas very, very lucky. They’re going to be better off. The immigrants are going to be better off. But then you have the rest of the world to worry about. Should we be concerned about that at all? Let me tell you a story of why I think we should be concerned.
Every country in the world that receives immigrants seems to want to receive many highly skilled immigrants. That’s what all countries are in the market for: the H-1 visa debate in the U.S., the point system in Canada, the new point system they have instituted in Germany, Australia, New Zealand, and Singapore. All of these countries are competing for high skilled immigrants. The problem is that there really aren’t that many highly skilled people in the world. There’s a limit to them; they are a scarce commodity.
I was in London a few years ago and read a newspaper story. I don’t know if this is true or not, but the story in the paper reported the fact that the National Health Service (NHS), the socialized medicine system in that country, had a hard time attracting native-born British people to be doctors. Because of the payment system, I guess, that the NHS has put in place, natives do not want to do the job of a doctor. The way they solved the problem is by importing a lot of foreign-born doctors. They imported so many foreign-born doctors that there were a couple of African countries where something like 95% of all the doctors they’ve ever produced are now practicing in London. To be sure, this is great for the NHS and for the British people: they are healthier, they got somebody else to pay for the education of doctors, and they can enjoy the output. But think about the people left behind in these countries. There are no medical professionals left. Should we be worried about that?
It is impossible to devise an immigration policy that makes everybody better off. Thus, we have to make a really important decision: whose well-being do we want to maximize? And that is not an economic question. That’s a values question. If you tell me the answer to the question, I can tell you what the immigration policy should be. Many of my friends, who know that I have worked on immigration policy for many years ask me: “What should we do?” And I always say to them: “Who are you rooting for?” You tell me who you are rooting for and the answer about what immigration policy we should adopt is very clear.
The next time you hear the immigration debate just keep in mind that arguing over a fence, over a green card for a few hundred people, is not what immigration policy is really about. Many people are hurt by immigration policy decisions, and a lot are helped. But that’s not really what’s at the core of this. At the core is who you care about? What kind of country do you want the U.S. to be? You answer that and everything about immigration policy will follow in a very linear fashion.
Thank you very much.
1. This paper is a transcription of the Ernesto Betancourt Keynote Address delivered by Professor Borjas. The transcript was prepared by ASCE’s President with assistance from Sjamme van de Voort and has been approved by Professor Borjas.