uh it's uh it's my great pleasure to welcome you to this joint pram jobs knowledge platform seminar with professor george borges the job knowledge platform is a joint undertaken by four vice presidencies pram fpd hd and dec in partnership with leading international think tanks over the past year the jobs knowledge platform has been active in building a community of individuals and institutions to share experiences and provide solutions to the jobs challenge immigration policy is central to the jobs agenda and we are indeed fortunate to have speaking with us today professor george bojas one of the leading experts on immigration and labor issues dr bojas is the robert w scrittner professor of economics and social policy at harvard candid school he has written extensively on immigration labor issues as we know he has written over 125 art articles in several books including friends or strangers the impact of immigrants from the u.s economy heaven's door immigration policy in the american economy and are widely used in their own textbook on labor economics and we are glad to know that it has just finished a new one on immigration the economics of immigration that we will be very glad to divorce when it becomes available and bring you back here hopefully his work always uh well also appears very frequently in major magazines and newspapers including the editorials in new york times the wall street journal and le monde and his research on the economic impact of immigration is widely perceived as playing a central role in the debate over immigration policy not only in the united states but also abroad businessweek and the wall street journal in front page feature articles have called him code america's leading immigration economist and code in today's talk professor bojas will address such a question as what does economics teach us about the determinants and consequences of immigration what are the economic impacts in both source and destination countries are these impacts generally win-win or are there real real winners and losers are these lessons robust and what are the implications of the lessons for the construction of immigration policy professor bulhes was born in havana havana cuba in 1950 and migrated to the united states in 1962 he's married it has three children it's also my pleasure to introduce vladis lopez uh who's one of the lead coordinators for the jobs knowledge platform and she will make comments after uh professor gohan speaks uh before uh and then after uh after that we'll open the floor for questions and in comments today's session is being live streamed in social media channels and gladius will be helping me also uh take questions from the social media audience uh uh uh last before i hand over to to uh professor george bullhouse a reference to uh the global knowledge partnership on migration and development nomad which is a multi-donor funded initiative of the world bank and that has been his age as uh to be a global public good sort of a global hub of knowledge and policy expertise on migration development issues and the swiss development cooperation in gtz are contributing to the trust fund and the idea is that the nomad will draw on experts on migration development from all parts of the world to synthesize existing knowledge and generate new knowledge for use by policy makers in sending and receiving comments and the nomad will enter a five-year implementation phase in april uh 2013. uh as i was talking to professor josh before this meeting he mentioned to me uh the remarkable difference in in terms of interest and widespread uh attention in the institution with regard to migration issues and that i was very glad that he notes this so without additional ado let me hand over to professor george borges let's greet him as he deserves thank you very much and thank you very much for inviting me to the world bank uh i was discussing earlier the fact that when i first came here back in the 19 mid 1980s late 1980s i think at the time the world bank didn't really wasn't much concern with immigration they were just beginning to think about it as to whether it's something they should get into or not i'm really glad that you're getting into it it's a very important topic and it'll probably become very important over the next few decades anyway thank you very much let me let me start the way i usually like to start these talks with issues by telling a little anecdote that has the the nice feature that's actually true and very much related to policy and it happened back in 1980 when jimmy carter was president and jimmy carter invited then xiaoping who was then the vice premier of china for the very first high stake high level visit by such a high-level official from china to the u.s so it was quite a big deal you know front page of the new york times washington post and so on and so forth and jimmy carter wanted to get den xiaoping into the oval office because jimmy carter wanted to lecture dan xiaoping about human rights in china and jimmy carter indeed brought into the room a big binder with issues about human rights in china that he wanted to discuss with den xiaoping and down the pages they went until jimmy carter got to the page on immigration from china china at the time like most communist countries did not allow its nationals to leave the country and jimmy carter told then xiaoping how could you think of china as wanting to join the club of so-called civilized nations that you know give human rights to its citizens if you don't let your people go and then xiaoping sat back and thought about it and smiled and he said you know mr president you're absolutely right how many chinese nationals do you want 10 million 20 million 30 million and needless to say that was the end of the discussion about human rights in china and uh the reason that's important is because then xiaoping put his finger on a question that's the core of immigration policy how many people is a host country willing to admit every i mean we discuss all kinds of things in the debate in the u.s if you read the papers every day now but nobody really asks that question you know we sort of talk about this and we talk about that and we talk about illegal immigrants when we talk about welfare use and we talk about stuff like that but we never really discuss how many immigrants is the u.s willing to admit because that's the question that's really much deeper and much more profound and in fact it's not the only question that's at the core of the debate there are two questions at the core of the debate that's one but suppose the jimmy carter to get back to my anecdote and this i'm making up now okay so until that was true but now i'm making up the hypotheticals suppose jimmy carter had actually given even 10 seconds thought to what he would have said to denshopen's question you know he could have said you know maybe we need 10 million chinese nationals over the next 10 years that may be something to consider right and the point is that densho pink could again have sat back and smiled and said and said you know which 10 million do you want we have a billion people to choose from and that is the second question at the core of the debate how many do you want and of the many people who want to enter the advanced economies in the world right now which people should these economies accept in other words there has to be some kind of numerical limit that people decide on and an allocation system by which to allocate the limited number of visas to the many many potential applicants and those are the two questions at the core of the debate and needless to say that whole debate over immigration policy in the us today and in 2006 and since 1990 has never really addressed these two questions uh so what i want to do is give you a little bit of what i call immigration economics tell you what economists have done in this and then return at the end to these two questions and ask have we learned anything from economics that will help us answer these questions in a rational way that that seems based at least on fact and not so much on hatred or emotion or things like that so let me start by giving you a little description of immigration in the world today according to the un about three percent of the world's population today lives in a country where they were not born that's quite a big number and remarkably if you look at the countries in that list the u.s is not particularly exceptional anymore in this pop in its share of foreign born people the us is like a 13.5 proportion of foreign born but sweden is even higher for example and greece is not far behind and germany is about the same so many many countries in the industrialized world would be what we used to call nations of immigrants exclusively in the u.s context in fact it's a very common development across the developed developed countries so it's a worldwide phenomenon in developed countries that is a big resurgence in immigration uh in the us and uh for for for various reasons the u.s has been the country that's been studied the most even though talk along the way about what's been happening in other countries the fact of the matter is that much of the empirical research done in the world is u.s based simply because we have a lot of data and we have a long history of this and the fact that we have a lot of data is illustrated graphically in this in this graph we've actually kept data on how many quote-unquote legal immigrants we admit since the 1820s so one way to think about this is how many green cards if you want to think of it that way have been mailed out by the us immigration officials in each decade since the 1820s and we have that number and this light is that number and as you can see there are two peaks in this graph one is in the early 1900s and that's sort of the mythic immigration that we all see in the movies you know you see the boat the stash of liberty people looking up people crying new york city and so on and so forth right and then we also have the current situation which is actually above the peak in terms of just raw numbers and we're more or less now admitting roughly speaking about a million people per year legally in the us this is the history of u.s immigration policy now let me uh put one caveat if you go to the public library or the university library and you look at the section on immigration policy there'll be tons of volumes most of which are several thousand pages long it seems very very thick what i've done here is something that will make bill gates proud i've used powerpoint to distill several thousand spaces of information into four bullet points so obviously this leaves a lot out but nevertheless i think it's indicative of what has really happened everything else is sort of minor compared to these things before 1875 in the u.s had no restrictions on immigration anybody who could somehow get their body into the u.s was a legal immigrant beginning in 1875 there became a move towards increasing restrictions on immigration so congress passed many many bills along the way restricting entry to particular groups and i list some of the groups there one important group was in particular asians congress basically prohibited the entry of asians into the us during that period beginning 1924 congress for the very first time sat down and addressed those two questions so that was the very first time in the us that somebody answered the two questions by what is called the national origins quota system congress passed legislation in 1924 that put a numerical limit given that you know you saw the graph on the previous slide where there was a big peak in the 1900s a reaction to that was let's put in a numerical limit to avoid those numbers again so for the very first time there was a numerical limit and because the numerical limit that implies that congress had to decide how to allocate the limited number of visas among many people who want to come and the way they decided that was by going through national origins so they said in 1920 the u.s population is so much german so much irish so much uh british and we're gonna let people come in according to those proportions at the time two-thirds of the us population was of german or british stock that means that germany and great britain got two-thirds of the allocated visas so it was a national origins allocation obviously it discriminates but let's be realistic given that you put a numerical limit every system you pick must discriminate by definition the only question is how do you want to discriminate here they picked national origin move on to 1965 national origins was seen to be discriminatory which it obviously is and in the spirit of the civil rights movement at the time congress said let's change the way we discriminate they still left numbers into the system they actually increased the numbers but they still had to pick a way of selecting the winners from the losers right and what they they picked was what's called the family preference system so in the us today and this is true until today in that yesterday we're guided by a system that has a numerical limit and where the winners among the many many applicants are chosen mainly on the basis of family connections with people already here and this is to show you the ways the system works in the year in the decade in the first decade of the 21st century we admitted 10.5 million legal immigrants okay 7 million of which almost 70 percent are due to family preferences uh 1.6 million is due to what's called employment-based immigration that's actually very misleading number because employment-based immigrants include not just the immigrant who got the employment-based visa but the family as well so many of those people are wives husbands and children uh 1.3 million refugees we basically accept a hundred thousand a year roughly speaking and what's called diversity visas which is just a fancy bureaucratic washington dc word for lottery we have a lottery that allocates about 45 000 visas a year randomly and if you ever want to see a system at work uh it's a nice thing to see it's like it's exactly the same way that once a year you get a an email from your employer saying the open enrollment for your health insurance and all that stuff has opened up right well people all over the world get that email in a sense and the state department opens up a website and anybody all over the world can get into the website and apply and you apply for one of the slots that's only been raffled out in that year of 45 000 slots now just to give you a rough idea of what the demand for entry is into the u.s in the last few years 10 to 15 million people annually have applied for one of those 45 000 slots it is way way harder to get one of these visas than to get into harvard college that'd be one way of imagine sort of seeing the the the chances and again that that is the reason this is a huge demand for entering to the us and that is one of the reasons why numerical limits are necessary okay so on top of all that we have another system of illegal immigration and this is what's going on these are these are not my numbers these are official estimates that the homeland security department puts out every year on the number of illegal immigrants and you can basically see it's been going up uh sort of stabilized actually since 2005 actually so between 10 and 11 million people and a quarter live in california and 60 come from mexico so on top of everything that i've just shown you legally we have about 11 to 12 million people who are here illegally and that will make basically roughly speaking the rate of entry into the u.s if you count legal or illegal one million legal and maybe half a million little per year if you divide the stock over the last you know 10 20 years so it's about 1.5 million people a year now you might want to know also because it's all a matter of debate you might want to know exactly how how does one come up with these numbers i mean after all it's not like we have a census that goes out and asks people are you legal or not right so the way that this comes around is that uh this is the statistical technique in which we go out we literally go out every year on survey people right like the census or the cps or the american community survey and a lot of these recent numbers are from the american community survey so we go out and with people and we ask them who were you born and we know how many people are in the u.s were born abroad from the very first slide i showed you we keep very good records of how many green cards we mailed out so you can imagine applying some kind of mortality table to how many green card holders should be alive today well those numbers are the difference between how many foreign-born people we counted in any given census and how many green cards should be alive today right now it's 11.5 million now right away you can see a little problem with this and the little problem with this is that when the acs or the census goes out they don't count everybody they miss people and it's not impossible to believe that many of the people they miss probably are illegal and therefore to make these numbers come up in a spreadsheet you have to make an assumption about what the under count rate is for illegal limiters and needless to say we have no such data that will be available to anybody so we have to make an assumption and you might agree with assumption or you might disagree with assumption but depending on what you believe is you can either co you can either push those numbers up or those numbers down right if you go into the footnotes of the report that dhs publishes and look very closely in the fine print you will see that they assume a 10 under count rate and i'll leave it at that if you think it's higher uh then blow those numbers up proportionally if you think we count everybody you can blow these numbers down but there's a big debate over these numbers and that's where they come from all right so let me tell you a little about the economics of all this this is really the reason why we're having a debate in the us over immigration this is just raw data from census from every census in 1960 through today basically giving you the percent wage gap between immigrants and natives believe it or not back in 1960 immigrants earned five percent more than natives 1970 there was parity by 2010 it's almost 20 25 below natives so it's a pretty striking graph and the fact that there's been a decline in the economic performance of immigrants in the us really underlies most of the questions at the core of what we care about politically not the core of immigration policy in terms of those two questions but the discussion in the debate really resolves over this fact so let me tell you what the way we think about this there are three issues in immigration paul in economics that i want to discuss today the first is you know who are the immigrants one important thing to remember is not everybody migrates i mean the fact of the matter is as i say in the slide even after a lot of illegal immigration and so on only 10 of mexicans have moved the wage gap between mexico and the us is huge has been huge for the last 300 years probably and still 90 percent of the mexicans choose not to move now you might say well it's because they're not allowed to move but that cannot be the only reason because the second bullet point is look let's look at a place where people can actually move and where the wage gap between the u.s and the place is very very high and that place is puerto rico puerto rico puerto ricans by birth are u.s citizens the transportation costs between puerto rico and the u.s are minimal the adjustment costs are quite low as well because there are huge networks of puerto ricans in many u.s cities with a lot of family connections already and still only 30 percent of puerto ricans have moved so two-thirds of puerto ricans have made a decision and voted with their feet by not moving so the question is what what's economic what what does economics have to say about what selects the people who move from the people who don't move and basically we have a model in economics that says that what motivates people to move are difference in wages and if the difference in wage is really high more people in that group will move if it's not as people fewer people will move but think of it in terms of skills in some countries the rate of return to skills is very very high as compared to the us or as compared to any host country in other countries the very return to skills is very low like to give an example in terms of the us compare sweden and the us sweden for whatever set of reasons tends to have a very compact income distribution the skilled workers in sweden are not much much better off than unskilled workers in sweden either because of the way the labour market works or because of the way the distribution system works well when you look at a country like sweden and we're actually having this debate today over california versus texas when you look at a country like sweden the people who want to move out of sweden will tend to be the most skilled that they have to have the most to gain so we will tend to get a positive selection of immigrants from countries that have a very low return to skills on the other hand if you have a country with a high return to skills which means a lot of income inequality the well-off are really well off and the less well-off are really less well-off it's the last world we want to move and from countries like that you're going to have a negative kind of selection and in fact if you look at the data using world bank data as a matter of fact this is what you tend to get if you look at the earnings of immigrants in the u.s by country of origin and you correlate that with how much income inequality there is in the country of origin you will see that countries like sweden are in fact all the way up to the left and the top in other words they have very high earnings and very little inequality whereas countries that are at the very bottom like haiti and brazil and guatemala and honduras and colombia and so on and so forth they tend to have a lot of inequality and the people who come to the us from those countries don't tend to do so well so there's a correlation clearly between characteristics of the source country in terms of return to skills and how people do in the us when they get here because i think that actually matters a lot is this other correlation again using world bank data between how immigrants in the us and the gdp per capita in this horse country a huge positive correlation you can control for education and that would still be there you can control for many things we observe and clearly people who come from wealthier countries have skills for whatever reason that just tend to be more easily transferable to the us so it's a lot of con the composition of countries of the country mix of immigrants has a lot to do with the average performance of immigrants in the u.s and that's what sort of one of the key results from the economics literature in the last two or three decades the second question i want to address is this question and the question is easy to ask what happens to the source country or to the host country when immigrants arrive in the labor market and you can actually ask that to interns of the sending country right a lot of people believe what happens to the people left behind well let me actually give you a little quote and this quote is from a book uh that many of you in my age group and older will know it's paul samuelson's introductory textbook in economics and actually picked a year specifically on purpose he has many many additions obviously right i picked 1964 because that happened to be the year before 1965 when immigration policy in the u.s changed so he's talking about an immigration regime that is completely foreign to us today but was true back then and therefore what he stole i mean right now the question that i'm asking here is very controversial a big debate in economics over a big debate in public policy over it so i wanted i wanted to remove the an economist answer from the political controversy and just say what would samuelson have said when none of this mattered politically nobody really cared about this so it went back to 1964 and i took the answer out of the book and the answer is look and he's talking about a world that we're no longer living in he said in world war ii world war one laws were passed limited immigration the national origins quota system only circle of americans were admitted since which is certainly true by capable labor supply down immigration policy keep wages high now we can reverse that to today and if you were to ask the same question today in 1965 laws were passed liberalizing immigration many more immigrants have been admitted by allowing people in we're keeping wages down so you know it's just a very simple supply demand kind of argument econ 101 supply demand things we were teach in the first week of economics nevertheless it's been incredibly difficult to prove this little insight empirically okay and there's been a big debate in the literature robert you know since 1982 actually so it's now 31 years and counting that people have disagreed over this so uh i'll be the first let me let me be completely honest what i'm to about to present to you is work i've done and i can more that more than i am more than willing to talk to you about all the kinds of work in the question session if you want or even now if you want me to but there's a big debate over this let me tell you what the data looks like if you do it in a particular way and this is a scatter diagram basically uh suppose you think of skill groups as composed of groups by education and age so for example there's a label market for people who high school dropouts in the early 20s that's one labor market there's another labor market for people who are college graduates in the early 40s that's another label market still another label market for people who are high school graduates in their mid-30s suppose you define each of these education skill groups that way and the way this is done is by defining five education groups so high school dropouts high school graduates some college college graduates and post college so five education groups and eight age groups so you have 40 skill groups let's track them over time so in other words let's just follow young high school dropout since 1960 through 2010. in every decade that group's weight changed in every decade the number of immigrants in that group also changed what would happen if you just graph that what would just happen if you graph the change in wages experienced by a skill group in a decade to the change in the number of immigrants entering that skill group in that decade that's the graph again nothing done to it except just raw data and what you see in the graph is there seems to be a negative correlation and if you draw a little trend line through it like i did there the elasticity will be minus 0.37 which basically means the following if you see a skill group and if you wear to helicopter lift and dump ten percent more people into that skill group that school group's weight will go down by about 3.7 percent which is not a huge number but it's not zero either okay so that is really the best evidence there is without putting any theory or anything like that just raw data for whatever reason over the last 50 years the groups that experience the highest influx of immigrants were the groups that experienced the slowest wage growth in that decade okay a lot of people that i introduced this methodology back in a paper in 2003. since then many people have replicated this and you tend to find it systematically in many different places like for example i've done two replications there are three others there you do it in canada the coefficient the the slope of that line is minus 0.35 you put in 10 more people in the canadian labor market the wage for that group goes down by 3.5 you do it in mexico and it was done by somebody in the world bank actually it's plus 0.31 now why is it plus because in mexico taking people out if you take ten percent of the people out in a skill group in mexico the wage of that school group in mexico goes up by three point one percent you do it in germany it's minus point one three you do it in nor waves minus 0.27 and then i did it for puerto rico five years ago and puerto rico sort of interesting uh first of all because no restrictions of any kind you know limited migration but also because remarkably enough even though i talked before about people leaving puerto rico there's actually a huge number of people moving into puerto rico at the same time it's one of the few places in the world that you have both a large inflow a large outflow and a large inflow and the people who move into puerto rico are mainly children born in the u.s of puerto ricans from the earlier migration so they're returning back and what you can do is that for any school group you now have two shocks people leaving people coming in and you can estimate what happened you can draw those graphs like i just did before you can do the graph and that's what you get so it's a very clear correlation between supply shocks and wages now by any means this is not the first time in in the world that an economist has discovered a correlation between supply shock and wages you know if you go back to for example finance welch's famous baby boom paper as the baby boom come in came into the marketplace the ways of young people declined you can go back to to um the gospel is a well-known paper in the in the in the jpe recently in the aer recently by hornbeck showing that when people moved out of the gospel state something happened to the wage you can go to the work of assemble glue and david otter on what happened in world war two when women came into the market the wage went down so this is just the you know supply and demand works in the immigration context for whatever reason people don't want to accept it so much because of the political implications but nevertheless an increase in supply or shift in supply will affect wages now this is all raw data basically this is that graph with a little fancy kind of metrics right what people have done since it's also putting structure into the raw data and this is something that even though i did myself in that paper originally i am much less enthused about because the minute you're putting structure into the data you're putting in structure into the data it's no longer the data that's talking right you're putting it in structure and the structure could be so limiting that you're more or less torturing the data to say something that might not be there and people i think have not really worked efficiently about that but nevertheless if you put instruction into the data the nice thing about it is you can allow for cross effects that scatter diagram i drew before that's just beyond effect but we know that when low skill immigrants come in that will have an impact on high school workers and vice versa right so by allowing by putting instruction you can more or less predict what the full effect of the weights of the wage impact will be and that's sort of what you get so clearly there's an impact but again the structure and then once you put in destruction you can then play theoretical games this is a short run impact before any adjustment takes place or a long impact after all the adjustments take place now in the media particularly people look at the long run impact number it's 0.0 and they say wow economists have discovered that immigration has no impact in the long run that's completely false that's a theoretical restriction it has to be 0.0 no matter what you do given the way these models are cooked up it has to be 0.0 the assumption that the economy has constipated into scales ensures that on average the wage effect has to be 0.0 in the long run in the short run it's minus 3.4 and again that that problem with this kind of modeling all these papers that do this assume a cup douglas production function well the cup double production function builds in numerically not qualitatively but numerically what the impact has to be because the impact elasticity has to be equal to capital share of income which in the u.s is 0.3 so you're going to go you see 0.3 is a lot and that's how you get the 3 there it's 3 0.3 times 10 roughly speaking so it is something a very even though it's nice to have an elegant model and be able to account for cross effects you have to emphasize as well that you're building in so much structure that a lot of what we think is there is not really there from my point of view and i've been doing this for a while now from my point of view all you can really say is this anything beyond this involves assumptions and involves building restrictions that may or may not be true okay uh all right one important reason for doing all this is that uh not only should we care about the distributional impact of immigration but the fact of the matter is that almost any economic model one can think of implies that the gains from immigration to the receiving country happen to depend on the parameter that measures the distributional impact okay i mean it's an intimate link benefits and costs go together and the same parameter determines both and the remarkable thing is that you hear in the debate from people who argue that immigrants have no impact on the wage nevertheless have created huge benefits or people who argue that immigrants have a terrible impact on the wage and nevertheless create zero benefits both of those statements are contradictory in order to have benefits there must have been a distributional impact if you assume the distribute that the if you assume that the distributional impact is such that the 10 percent increase in supply lowers the wage by three percent these numbers come out of the calculation so where does the net gain of 30 billion dollars annually now this actually is a very interesting number because in the public debate today which is resembling the public debate back in 2006 when president bush wanted to push through his particular program the cea was asked to produce a number about the benefits and the cea came up with a third bullet point the benefits are around 30 billion one interesting lesson from economics is that it is very difficult if not impossible to take the canonical model of a competitive labor market and cook it up in a way that leads to a big number there you just can't you just cannot do it okay so the you you the the the constraints of the model tell you that it's impossible for the benefits to be 300 billion they're going to be in the range of tens of billions in the u.s economy now what the ca report did not mention was the first two bullet points but the 30 billion actually comes around from a huge redistribution the only way to get that 30 billion net gain is by having a loss to workers of 400 billion and again two people who use immigrants of 430 billion so that 30 billion is actually masking you know tremendous flows of wealth around the economy or you know given that we all if you've taken an economics class and you've seen this the way this is done with a little triangle and the rectangles we all know rectangles are way bigger than triangles and that's what this is saying okay the the the flows of wealth are much greater than the net gain okay uh all right let me turn to the last question about immigration policy people worry a lot about which economics has something to say about and that's the whole issue of the fiscal impact of immigration until now i've talked about the labor market let me now talk about the fiscal impact i want to put it in context you know people worry about a lot about it today well people worried about it back in 1645 just to give you an idea okay so the very first restrictions were in massachusetts in 1645 and 1655 where they prohibited the entry of poor immigrants into the colony new york in 1691 must have hired the very first phd economies before adam smith because whoever they hire as a consultant told them let's set up a bonding system and if families want to bring in their relatives they have to put up a bond and that bond will become will revert to the state if the family member they want to bring in happens to need charity or assistance or medical or medical help so this goes back to 1691 okay now in 1876 and this is another thing that reverberates in the debate today in 1876 the supreme court said that all these state regulations of immigration were unconstitutional that's where the notion that only the federal government can regulate immigration policy comes from and if you recall uh from the earlier slides that right around that time right after 1876 congress began to enact all these restrictions as to who should be admissible or not so it's a supreme all states used to do it on their own the supreme court said the states cannot do it congress then stepped in and began to say well asians cannot come in and public charges cannot come in and this and that cannot come in in 1882 congress passed the law which is in the books today banning the entry of anybody who can become a public charge in 1903 congress passed a law which still in the books today saying that if somehow you can get in even though you weren't a missile in the first place and you become a public charge after you enter you're deportable so all these things have been in the books for over a century okay and they have not changed now just to give you an idea of what congress does and what people actually do are two different things the immigration agents in the us the ins didn't get around to defining a public charge officially until after 9 11. so it took 100 years for them to actually write a paragraph about what a public charge meant okay anyway this is what the data looked like for real this is cps data and again no fancy econometrics and the cps you can basically count how many households receive assistance and in the cps you can basically see that immigrants have a pretty high rate of assistance receipt and it's been increasing it declined dramatically actually for a bit for a bit and that's right after the the the the enactment of welfare reform in 1996 by bill clinton and then that was the short run trend and then it's picked up again and you can basically see right now it's something like uh i don't know 33 or something like that as compared to 20 or 23 for natives so that clearly is a physical impact and uh there's again a big debate over all this but i'm just trying to show you the raw number this is what it looks like if you just take the cps data download it you know when you leave the room today download it from the ipoms and within two hours you can generate that that graph really easily okay all right so now let me return to the question in the core of all this which was mentioned at the beginning what does all this mean what should we do what does everything that economists and myself particularly have done over the last three decades of their lives worrying about and studying have to say about immigration policy and i hate to tell you this but the fact of the matter is that okay nothing at all and the reason it implies nothing at all is because there's something left out of every single slide i've put up with all these facts and there's something that's left out is really the core question the debate which is what do we want to be as a country what is it we're trying to accomplish with immigration policy let me give you what i what an example what i mean by by that okay suppose that i put on a hat of being a really nice guy totally humanitarian caring about everybody in the world and i say to myself i want immigration policy to be you know to help poor people all over the world that's what i want well the us right now you can interpret has the biggest anti-poverty program ever run in the entire universe through its immigration policy they've admitted tens of millions of low-skilled poor people from all over the world and the impact is an impact of perhaps depressed wages for low-skilled workers here it's an impact of perhaps higher cost for the taxpayer but because my goal in life is to help poor people i am willing to pay the cost okay so same facts i just presented but the goal being i want to help poor people will need to say i want to do nothing about it that's exactly what i want to accomplish and that's a price i have to play i have to pay to accomplish what i want to accomplish now let me take off that hat and put on the head of a taxpayer who wants to save his money in the bank and spend it on himself in his retirement by that hat i look at what's going on and i say to myself why should i pay all these taxes to support 33 percent as dependency rates i don't want that so from the point of view of the taxpayer who only cares about his income and his retirement what we're doing is awful right and what we should do is change the immigration policy in a different way same facts but different objective functions lead to very different immigration policies and that is why nothing economists can ever do really tells you what the optimal policy should be because an answer to what the optimal policy should be you have to bring something in outside the system what is you're trying to maximize what is the objective function you're trying to maximize and different objective functions will lead to different policies with the same facts you know men cannot live on facts alone there'll be a way of putting it okay you need something beyond the facts to make the the the jump between what we know and what immigration policy should be so the way to think about this is look we have two policies we have two questions that i raised at the beginning how many and which many countries have confront confront this every day and one way to think about the solution that countries impose is that they they have something called the point system now in some countries they actually call it the point system in other countries they don't use those words but nevertheless every country uses the point system the reason being that most countries that are immigrant receiving countries will tend to attract many many more people than what they will be willing to emit so the way countries do this let me give you an example in specifically canada if you want to see the way the canadian immigration policy system works when you leave this room today go to your computer type in canada immigration and that will bring you to the canadian immigration website and act as if you want to move to canada and by clicking on the right things you will be given a test you will literally be given a test online okay and the test will ask you how old are you and if you say i'm 25 they say wow 15 points they say over 50 yeah not so good right yep take minus yes they ask you how much schooling do you have phd wow terrific high school dropout go to the us uh they will ask you um what do you know what do you do for a living a chef lots of points for chefs economists a little doubtful okay yes at the end of all this they will add up the points for you conveniently okay and they will say right now our passing rate is x and that passing grade changes okay from you know not from day to day but changes in australia literally changes from month to month in canada doesn't change all that often but it changes and depending on your score and the passing grade you're either admissible or not okay now that's the way that the canadian system works that's the way the australian system works new zealand many many countries have that the us in fact has that system except we will never admit it in the u.s we basically ask for the most part one question do you have family in the u.s if you say yes you get a ton of points and if you say no you get very few points making it very difficult to enter now believe it or not part of the bush program back in 2006 2007 was to introduce the notions of a point system sneaking into the into the immigration policy structure i mean that wasn't very publicized but among the 10 000 pages of legislation there were a couple of pages where a point system was being introduced for the very first time okay that specifically looked at education and age so a lot of countries do that because they perceive for whatever reason that highly highly skilled workers tend to produce more gains now it's obviously the case that they do when you look at the fiscal impact highly skilled workers are not going to use as much assistance you're going to pay higher taxes there's a big debate still waiting to be settled as to whether highly skilled workers do much better for you in terms of labor market productivity effects there's a whole new literature in economics right now trying to find whether the admission of highly skilled workers makes other workers more productive okay so that that's basically the rationale for the point system now before we set up a point system i mean the variables we pick in the point system it's basically parameters of the objective function you need to specify i mean what is it you want to maximize if what you want to maximize are economic gains the questions you're going to ask in the point system are very different from the questions you would ask if you want to maximize something else and that's why i want to say that you know it really matters what the objective function is you cannot really talk i mean i know we do this all the time in economics i mean believe me i've written enough economics papers to know that format or the template you have an introduction you have a little theory section with a couple of equations to impress people a little regression model a table summary of data a couple of regressions a summary and the very last paragraph and it's the policy implications right we should change the world according to this because of that that that's complete nonsense policy implications do not come from facts alone policy implications can only come from the joining of a fact and a policy objective a goal you know and the question really is whose will being do you want to maximize who what do you care about i mean let me give you a good example of this uh you can think of the world as being composed of say you know several groups right you can think of natives as being a group you want to care about native workers you can think uh or native just natives together native workers and capitalists you can think of the immigrants themselves are people you should worry about you want them to be better off you can also think of the people left behind as people you should worry about it is going to be very very difficult to come up with an immigration policy that makes all three groups better off okay it's to be very very difficult so that means that you're going to have to allocate weight to different groups and the allocation of weight is what the objective function is about and what really in the end determines what immigration policy is about it is not the facts that determines the policy it is the weights you attach to the groups that either benefit or lose that determine the policy and that is unfortunately something that we really won't never discuss because we don't we know we don't want to get into the business of attaching weights because attaching weight is really revealing values a lot of people since i work on immigration all the time a lot of people you know my friends or my friend's brother or whatever whoever me for the first time will say you know what what should we do and my answer which i now know to be true is who are you rooting for you tell me who you're rooting for that's what we should do thank you thanks ladies so rather than trying to provide extensive comments on what george has said i mean as he mentioned at the beginning there is a hidden debate on terms of what is the effect of immigration on wages no i mean there is another school led by david carr and other colleagues running experiments uh he shows that there is no effect on wages but i don't want it you know because as you already mentioned you are writing a book which is summarizing some of this debate so rather than doing that just let me try to mention some of the comments that we are getting through the jobs knowledge platform from several countries in the region there is one from latin america precisely pointing to this issue of several economies have opposed borga's view saying that immigration is not a big deal and that this issue is more cultural and social since u.s has easily absorbed immigrants and that immigration is not taking jobs they are doing jobs that wouldn't have existed other way i mean that's one and then from many region we have um for many years immigration was thought of as a solution but now it has become a cultural problem why do you think this is happening from mongolia we have a business economic development lead some people to leave their countries for all for other markets but discrimination of them makes them willing to leave their home how can countries encourage the most skilled workers to move to new markets while preventing discrimination that's one of the questions but why don't we take some from the floor and then we can please okay professors and let's start with this let me talk about the debate first so thank you okay so um let me put it in context even though the question of what do immigrants do to wages in the receiving country it's as old as there's been immigration itself uh in fact you can actually find this is a little anecdote you can actually find letters from the people who came to settle the us back in the 1600s writing back to england saying please don't come anymore because conditions are really bad and more of you just means it's hard to get farm jobs or whatever so this is very very old question despite that the very first empirical papers of this in economics was only published in 1982 so we're only 30 years away from the beginning of the literature okay uh now beginning with that initial paper by gene baldwin grossman she introduced a methodology that became very influential and the methodology was one in which what you want to do to measure the impact of immigration is compare people living in different cities so for example if you believe that immigrants coming to san diego and they do harm to the labor market in san diego you would expect to find that waste in san diego will be lower than wages in pittsburgh where few immigrants came in i call that this it's like a spatial correlation right you correlate things across space geographic space there's a big literature in economics until what the scatter i just showed you that basically does these spatial correlations remarkable thing about those spatial correlations in the u.s context is that when you correlate immigration and wages you don't tend to find much and that's the source of the debate so when you do it the way i presented it which is by looking over time at skill groups you tend to find this negative slope when you look across cities you tend to find very little slope uh if any okay now the question is why and believe me people have been working on this for you know 20 years now trying to figure out why and the answer is the the notion that motivates the comparison of cities is the notion that somehow each city is its own labor market that somehow immigrants going to san diego something happens there and that doesn't percolate anywhere else but the fact of the matter is that cannot be true for two reasons number one immigrants do go into san diego randomly i mean this must be there must have been a reason why they went to san diego and did not go to pittsburgh and more likely than not is because the wage was higher in san diego so that will tend to contaminate whatever correlation you get on top of that immigrant is going to san diego well that means that if things get really bad there people are going to respond and natives might respond by not moving there or moving out or ferns might move there to use the low wage labor but the point is these flows will tend to diffuse the impact from san diego to other cities okay so let me all i'm saying is there are two schools of thought on this and there's a big debate about how to resolve this to this day and if you want to read more about it wait till my book comes out next year okay uh it's a little plug for a very technical book that will sell like 300 copies and no more um anyway so let me 300 at the world bank that yeah and and 290 of those would be my family purchase of them okay uh all right so uh let me return to another question which is sort of interesting uh because the the way it was phrased is sort of interesting immigration was always thought to be a solution and now it's a problem uh in the u.s context as far back as there are polls immigration has never been seen to be a solution to anything i mean if you ask people consistently should we have more immigrants or should we have fewer immigrants the answer is always the same we have fewer immigrants and that's been historically true and you know never has there been a public perception that some have something good will come out of this immigrant i was talking to you earlier about this immigration is one of those few cases where the political elite the academic elite the washington post and the york times believes one thing you know immigration is great it's a solution everybody else believes the opposite okay it's just a very very strange cognitive dissonance kind of thing and uh it is one of those rare i mean more often than not when policies go through in the u.s you know many of these sectors tend to agree you know one other example i can think of is trade where the academic elite and the public perception of trade are clearly very very different and somehow we still have trade right and we and we liberalize it it's not really clear to me that immigration will be like trade or not so uh the the the perception that it's not a problem it's never been there so in the u.s context people have always pointed out the problems and in fact very few people have actually ever bother estimating the benefits and one one important lesson from economics i think is that what we call a problem which is the wage impact it's actually the source of the benefits and the more of the problem you have the bigger the benefits you have in the end as well so uh there's a there's a relationship there right let me then take the chance uh professor uh you actually said it all depends on which objective function we are maximizing and so on we at the bank have been observing uh on the concrete side of things attempts to find to devise policy schemes that might make possible to compatibilize different objective functions in which to some extent we might find win-win arrangements and that changed a little bit may change the landscape i have in mind the idea that some countries have played with temporary migration schemes i'm not passing any any value of judgment on this but but can different schemes gain innovation in terms of policies facilitate the conciliation between different objective functions okay i think one has to look at it specifically in terms of the actual proposals but temporary migration since you mentioned that i want to talk about that a little bit okay okay so one you could think as a win-win situation temporary migration would be one in which a receiving country emits people temporarily uses the value that they provide in the labor market and then at some point in the end they go back they have no political impact they have no cultural impact in some sense right and that would be a win-win that temporary migrants clearly gained otherwise they would never have come and the country that welcomed them also gained because you get their services for a while and then you don't have to worry about all the consequences of that right uh and theoretically that's perfect the problem is in a country like the u.s that dream hits a wall of reality and the wall of reality and the wall of reality was actually hit by germany back in the 1950s when they saw the same thing and they let in a lot of turkish guest workers right the best line i mean i've read a lot of studies about the german experience but there's one line that stuck in my mind as being the true thing that summarized the the problem and the best line was and the novelist that said this we wanted workers but we got people instead okay and uh that is something that you know you have to remember you know importing a car into the u.s you use it it breaks down you throw it out it goes to the junkyard importing a person is not like that in the u.s specifically once you're here in the mainland you know once you're here i mean once you're here physically all kinds of entitlements constitutional entitlements open up that prevent all kinds of things from happening to you even though we might want to send you back like for example in other countries i don't want to mention names but in other countries when a guest worker uh when a guest workers time is up or the guest worker becomes pregnant a bus is sent to that house where the guest worker works that guest worker is putting to the bus taken to the airport and shipped out that will be inconceivable in the u.s right yeah in any case uh any scheme like this would have to be accompanied by safeguard policies by investments in housing and everything else so as to make and it may be it's only feasible or at least it's more feasible in contiguous space when when we know as your the puerto rican statistics so clearly and the evidence also points out that usually migrant doesn't go because he wants to move cut the fuse because he's enough he or she sees an opportunity so not necessarily he wants to he or she wants to move to change the country anyhow any scheme of this in order to be minimally uh uh compatible with moving temporary people would have to be accompanied by a whole scheme and and then things start to change because then the costs are not the same right get your point yes please identify yourself my name is prince haito um i just wanted to bring in a different dynamics to especially the debate concerning us do you think that um the identity of u.s as a nation of immigrants um it's also like defining the kind of debates that the political elites and the policy makers are having in terms of immigration being a good policy in contrast with what the general public you you are saying that there's a a sharp contrast between what the general public thinks and what like the media and the political elite think about immigration so i wanted to ask you uh in light with the facts and undefined goals of what the immigration policy of the u.s should be working towards do you also think that maybe identity may be playing a role here um or maybe uh that rhetoric of being um the protector of like uh human rights and like you know all these uh grand ideas that the u.s have of itself and secondly i also want to ask don't you think that legalization of the illegal immigrants would also have another implication in terms of like in terms of which because uh now they will no longer be part of the uh black market or like the underground economy and therefore don't you think that they may be pushing for higher wages in that context so what is your thoughts on that great okay thanks yes um louise fox and i work on labor markets in africa um but i want to turn return to the u.s um i understand completely your skepticism about cge models in this case and i share it um having said that i think that the general equilibrium and long-term effects are a little bit more complicated first of all immigrants make a set of goods that are purchased in the united states by different people in different percentages cheaper so there's cheap food but there's also cheaper services especially caring for people services which might be purchased by more upper income groups and low income groups another side of the income distribution is that i agree with you that where was that number from the council of economic advisors that said uh minus 2.8 plus 3 on average 0.2 i mean the question that will definitely not be distributed equally per capita that's for sure and um so again the income distribution effect um the in the us we don't have a history of the gainers compensating the losers uh directly uh and we're having much more trouble collecting taxes from the gainers these days so i wondered if you could address a little bit more the income distribution effects and how you think they are uh maybe affecting the current debate well thank you very much uh for this excellent presentation i have two brief questions one is that given the um uh the number of people who applied to the uh to the uh to the lottery right obviously if walls all those will be turned tore down there will be more immigration but my question is if um suppose that the objective function is to reduce poverty in the world right and all all votes will be turned down tore down everywhere right how much will that three percent increase because you also said that in puerto rico i mean there are no costs of of migrating wage differentials are big and still not all puerto rico moves to brooklyn or to or to manhattan so therefore i mean not everybody will move just to equalize wage differential so how far is that three percent right or what it should be if we want to maximize poverty reduction in the world and the second brief question is you said that uh um here and here in the u.s immigration has never been perceived as a solution my question is is that an american fact or that will be true in australia in germany in sweden all countries that we know that the percentage of people who are foreign born is more or less similar to the u.s my question is basically uh looking at immigration one of the one of the biggest impacts of immigration is also political and social right a lot of groups are marginalized and the way the the immigrants are treated in the host country um you know there are different ways of living and you know those those ways of living clash and now it's becoming even more important in the world everywhere um so immigrants are usually the kind of marginalized groups starting from you know 1800 i uh late 1800s irish in the u.s and now you know now you have a different group so they're they're always treated with sort of you know they're treated as immigrants uh and then they later become part of the population i just want to ask you how this political and social impact of immigration which is becoming even more important in today's politics um you know how is that part of the immigration policy and how you would you integrate those kind of that thinking within uh you know what you know about immigration and how you look at it okay um let me provide some answers okay the ones that i know uh the the ones i don't know are far greater than the ones i know uh related to the last question one thing i think i've learned is that there really are no universal laws of immigration i mean the context where things are being done the political context the economic context and so on really matters a lot so just because in the u.s suppose you believe all the facts that you presented here suppose everything is true for the period of in the last five decades that doesn't mean it was true before then it doesn't mean it's gonna be true in the next 30 years a great example of that which relates to the integration issue is the whole notion of the melting pot people often talk about the melting pot in the u.s in the 20th century as being sort of the prime example of the way a host country could take a group of very diverse groups of immigrants and somehow mold them into americans over the course of you know seven eight nine ten decades right which is not a big amount of time in the scheme of things but don't forget that all happened within a very specific context it happened at a time when the manufacturing site was being built up by the us and in fact the manufacturing sector was built up by immigrants it happened at a time when there were world wars against countries that had sent a lot of immigrants and people had to decide whether germany or italy or the us was really what they're going to be at many states in the u.s believe it or not only 100 years ago during the beginning right after the world war one began made it illegal to speak german in public so that helped assimilation as well so the political context the cultural context of what led to integration then you know there's a lot of little details that are ingredients into the whole process that may or may not be present today so i'll be there even though i've studied this for a long time i'd be the first person to tell you one cannot take any of these numbers and extrapolate because conditions change and what's true in the us might not between germany might not between australia might not be true in canada tomorrow and maybe it was true in canada yesterday we don't really know there really are no universal laws of immigration per se uh going back uh the question about whether immigration is seen as more of a solution in other countries and in the u.s uh it i don't know if seen as a solution but certainly differently and i've been to many of the immigrant receiving countries and talked to the policymakers there they have no qualms whatsoever about telling you what their objective function is and from their point of view their objective function is to make natives better off and everything they've done in terms of designing the immigration policy is precisely for that they want to have a skilled immigrants will pay taxes who will contribute to productivity and they'll want to avoid all the problems that other kinds of immigrants might or might not bring in and they're more than willing to tell you that in the u.s and this return to one of the questions as well because we're a nation of immigrants i do think people tend to perceive things differently like for example one favorite proposal economists make is let's sell visas you know in economics you know you have a market problem create a market for it right you know let's create a market for visas so you see you know proposal of the proposal every once in a while by an economist saying let's sell visas uh you know becker for example has a business week column on this i don't know back in 1990 which you read as an economist and it says of course you know this is it you read it as a human being as an american you say this guy must be insane and and that's precisely the reaction that you hear from policymakers when you tell them let's sell visas you know that's not who we are they will tell you that's not what america is about and so therefore definitely the perception of what we are and where we come from matters a lot and again in the policy debate one of the one of the ways to kill the argument for skilled immigrants is to say we want skilled immigrants because you know they do the following set of good things the best way to kill the argument is would your grandparents gotten in 100 years ago and the answer is probably not and therefore end of argument so in the u.s perceptions that really really matter legalization and the distribution effects on general equilibrium a lot of what economists have done over the last three decades is very partial equilibrium very very partial equilibrium like the whole selection models that we have you know of how people choose to move and not choose to move are based on the notion that the parameters of the income distribution in various countries is fixed well that's fine if only a few percent of people move but if you're gonna have huge flows of people moving those things can no longer be true and we know nothing whatsoever about the general equilibrium model of selection i mean that's actually one that's one of the frontiers in economic research on this we really have to bring in the general equilibrium aspect into into the modeling uh now unfortunately at the same time all this has been happening labor economics as a field which is the people who tend to do this for a living has changed it went from being very theory based to being very very empirically based and i would suspect if you wrote a general equilibrium paper in labor economics it might not have a big audience these days now that will change again because all these fads are fads but it is not in the in the tradition of current research to do things like that as important as they are and um something about the world and the world gdp and the equilibrium of maximizing world gdp look there's no doubt that if you want to maximize world gdp the best way to do it is to open up borscht okay you can construct a very simple minded model that will show you huge gains from maximizing world gdp the problem with those models is twofold most people don't want to move and to maximize volution to increase substantially requires a movement of hundreds of millions if not billions of people because the north advantage in productivity is so large those models don't take the cost of moving into account and it is not that difficult believe me to hook up a little example where the cost of booming could easily swamp the gains to world gdp that's number one number two returning to general equilibrium issue all those models that implies such huge gains basically a very partial equilibrium whatever it is about the north that makes people more productive will still be there even after hundreds of millions of people move there again that might be true for small flows it is hard to believe it could be true for very large flows so if you really want to get those gains you have to look at it from a ge point of view and looking at it from a ge point of view i suspect would tend to attenuate the gains dramatically okay great fantastic uh we uh as we know here at the bank yeah very brief okay sony come in uh we are running against the clock only quickly took two questions quickly you have talked about i'm sonya plaza working in prem but working on the migration issues on the global knowledge partnership nomad with dilip two questions do you talk about the function objective of the of the government of the country but you haven't mentioned about the immigrants what will be the function objectives some of them that are here this morning in the radio there was a debate listening and some of the immigrants that have been here and have put the papers they have to wait 30 years or three years they're complaining against the new immigrants that are going to be put in in the line so i would like to get your views and second the views on what exactly what do you see the immigrants option function will be the ones that are here how do you see that given that you have been talking to different people in the united states you presented the workers and you presented the firms but what are the immigrants second it's a what will be the policy question that the bank will help into this debate from your point of view those are very uh i mean the honest answer is i don't really know okay if you want me to what policy issues should the bank address as a way of thinking about immigration uh i think the way you phrased it earlier which is to find policies that allow some people to gain while either minimizing or perhaps creating an environment where the loss to other people is not that large is really the way to solve it's really the way to advance the debate because look the fact of the matter is most social policies immigration is only one of them creates winners and losers i mean that's just that's just the you know redistribution it's redistribution right and immigration is really a redistribution policy redistributing wealth from particular groups of people to other groups of people people i mean the way to think about it generally is immigration redistributes wealth from people who compete with immigrants in some sense and i mean by in a very broad sense to people who use immigrants in a broad sense and that's where distribution takes place so i think one way to advance the debate is to identify the groups better identify the groups better find out exactly what the nature of the losses are and where they come from and find a way of mitigating that you know yeah okay we fully agree with that uh given the demographic changes we know that migration uh either to cities or among across waters will be among the major challenges that mankind will be facing along the next decades so let me thank you on behalf of michael x for this brilliant presentation and enlighten us and hope we can have you back as soon as possible thank you thank you you