Its an 1 45. Good morning everybody. Welcome to the cato institute. Im an immigration policy analyst here. If anything, this 2016 election will turn more on the candidates respected immigration positions than on any other. Donald trump, the republican nominee won his primary, primarily based on this topic for his support to reducing Legal Immigration, building a wall and deporting unlawful immigrants. On wednesday night, he delivered a major address on this topic in phoenix, very basically double downed on his positions, squashing the rumors that he was softening on this issue. A topic of immigration has produced the loudest and most disagreements of this election season so far. Polling show that immigration is not a top issue this year in the minds of voters, the electorate certainly hasnt been then interested in a topic in about a century. Unfortunately, though, that popular interest is combined with a healthy dose of misperception of both the public and policy makers who attempt to remedy that misperception and influence the debate. Cato has put together this conference and invited some of the best researchers and top minds and academics around the country who work on this topic to present their findings to go over forming this debate. Well publish an edited valium of contributions written by our distinguished panelists. Todays conference will consist of five panels. The first panel this one sitting will talk about how they effect wages. The second will discuss unlawful immigration. The third will discuss how the often overlooked topic of how immigrants effect the real estate market, particular importance to americans since the housing crisis and great recession, the fourth will examine and the last panel will delve into the most recent Frontier Research on how immigrants Effect National institutions and destination countries. Each of these topics individually is worthy of their own conference by themselves, but one panel on each will have to do this time. Without any further adieu, allow me to welcome david, who will be moderating our first panel on immigration into labor market. David. Good morning. Primarily financed by facebook cofounder and his wife who serves as our president. The subject of todays conference could hardly received more attention in the president ial campaign, the subject of this panel could hardly have received less attention, which is what is the evidence actually say about the impacts of immigration on the receiving economy. Up until a few years ago i was a senior fellow at the center for Global Development and my former colleague, Michael Clemens who i believe will be speaking this afternoon. That openness to immigration is one of the most powerful ways to reduce poverty in the world. And also the lack of openness represents one of the greatest market failures in the global economy. That argument is part of the reason that i am here, part of the reason that my organization is interested in this issue as a grant maker before as an employee i was a consultant to them and i did an evidence review on exactly this question. I am pleased to present to you some of the leading producers of research in this area as one of the leading consumers of it. Without further adieu, i would like to introduce. Profes sore of economics and uc davis. In journals and books about the questions well hear about and received grants from the world bank and the National Science foundation and so on. He also has an internet presence, so if you want to read more about him, you can. Thank you for inviting me. This is a great conference. And im going to jump right into my topic of only 20 minutes. I tried to squeeze in the information here in 20 minutes. I hope you can read more or less, the question im going to try to address thats been done in the last 15 years. Can we consider immigration is one as one of the cause especially for a low educated skilled worker that has happened in the last 35 years. I will show you how one reason we have too many immigrants that are taking jobs so the simple supply store is they flood our labor market and push down the wage. I will give you a couple of hints then im going to ask, okay, if we dont find much at the National Level. It can it be something specifically for market which has been inundated by that. One question will come by because the evidence seems to me to have pile up against this idea that many in reality there are some channel and some good reason to think in some cases immigration can actually boost wages and i will go through this. So i think there are two facts that people put together and they say immigration hurt wages. And there are two simple fact. In the last im going to take the period of 198 02014 this is when the wage between low educated low skilled and high skilled opened up. Im going to show a figure about this. 34 years have been a period in which immigration has increased. College educated have done much better. This is the growth rate of wages between 198 0 and 2014 dividing the labor market for people who do not enter high school. And more than a bachelor. And this is the percentage change of their wages so their weekly wages over this 34 years and you see that one college and college kated have done well, their wages have grown between 20 and 30 . Almost 1 per year in real terms. The high school very done very badly, but the High School Diploma not well, their wages have gone down. If you. Noncollege have done quite poorly. Could it be that migration is responsible for these diver gents. Now, just stating this fact and stating if you are just willing to do the next step, there are some other changes that have changed, of the labor market, displayed a role in changing the demand for International Trade of shoring that unionization minimum wage have gone down. Theres no clear implication. Its supply store that imgrantds that came in that low then there should be, a very high supply then the inflow of lesson must be much much larger thans educated. The truth looking at the picture of 34 years and comparing the immigrants have been much more it has increased the supply of very highly educated much more. That it goes exactly the opposite way for what you will need to generate depression of this the High School Drop out has been a little larger than the high school inflow. They have used a lot, a simple model of supply and demand. Its a quiet broadly used model in the field. I have criticized this model because consider Everything Else fixed productivity fixed and well talk more about this. Let me just show you how even taking the model that people who assume depressive effect of immigration can use, how far do we get in explaining this changes and wages through immigration. So im going to show you, taking a simple demand supply model and increasing the supply of immigrants, sorry and leading Everything Else fixed, how far do we get in explaining the way to change. And im going to focus on, can we explain the bad performance of nonCollege Educated through this inflow of immigrant and among the College Educated, can we explain the particularly bad performance of least educated of all which are the group of High School Drop outs. This didnt come out the way it was written thats way i gave the pdf and the power point. Let me just point out here, that the wage of group, to college and noncollege, get to the green no point on it. The relative wage of college and noncollege worker. And on the relative supply. If you increase and change then youll depress the and strongly depress the wage of the group. Elasticity of supply. We can actually take these, these simple formula. And see if the change of college that is being generated by immigration how much would it predict and depression of the noncollege wage relative to the college. We can do the same exercise, if the supply generated by the drop out versus high school, we can and how much generate this depression of our wage of High School Drop out. This is important barometer here that we needed to estimate in order to get to the exercise which is, as i said, elasticity. In this exercise im actually going to take the most the parameter that people who are in favor of finding negative impact on immigration are going to argue is the correct one. I am going to buy us this model as much myself as you saw in favor of finding a negative effect of immigration. In this position, the elasticity so that equation for college and no college has been estimated many times and theres a certain consensus, is it between 1. 5 and 2 or 1. 5 and 3 or are we going to take 5. 75. The estimate of how substitutability or how different with high school and people who are drop outs are more contentious, theyre very similar. Others say theyre dissimilar. Im going to take, in these exercise, the most negative potential elasticity or the smallest, the most negative scenario with the smallest elasticity saying with People High School degree are not very substitutable and they have the same. There is different as college and college, think of this as a measure of how different they are, how theyre relative supply effect their relative wage. If i do this, again, there are huge. I want to point ut. If i do this, i can show you for each decade, the main conservative of 2014 which has the latest data, by how much this immigrant increase the group and how much in percentage point it increased and by how much they changed their relative supply which is directly effecting it. The first problem here are the effect that i attribute to immigration with this relative simple model. Remember, the way of decreased inequality nonCollege Educated but starting from 1980 College Educated started doing much better and high School Started falling behind. This is the relative change for College Educated. Minus 13. 7 mean in the 80s the college increased increased their relative wage by 13. 7 and in the 90s by 3. 7 and 2006. 6. The College Going to do better. However, if you look at what part of this number can be explained by the immigration part, you see, this is the part that can be explained by this model. In fact in two of these decade of the three decades, immigration is the opposite way. Immigration by itself actually reduced because there are so many College Educated as percentage of the group that come in. And only in the 90s does a little bit. This is the college noncollege group. If you take the High School Drop out versus high school graduate, again, this is the percentage of drop outs. Theyre due toim grags. This will be the wage and, again, you see that, again, in two of the three decades between the 80s and 2000, immigration either has the wrong sign so it wouldnt explain the increase in drop out r you would explain variable, only in the 199 0s explain some of them. And either very recent is no explanatory. So this exercise done and i think summarized a little bit and going to say, even if you take the most sort of negative estimates here of this effect, for the college noncollege, you simply dont have the number to generate the negative effect, because a lot of immigration was college intense. So it worked the other way to reduce this. So for drop out high school graduates, you have some action, but only in the 90s. The 90s look different. Its to say that really where a lot of this researchers pointed out. Theyve been shape that are effected by the 90s which are different from the 80s and 2000. This is a way at looking from the 90s are different from the 2000 and the 80s. This is the growth of percentage of the group connected. This is the growth in each step. This is what happens in the 2000, 2010. This is what happened in the 2000. This is what happened in 1970. If you see an upwards curve, it means the change of immigration is the nonskilled group is smaller. You see every decade. This is almost look meaning that immigration increased these High School Group relatively more or almost the same. The 90s, only in the 90s low skilled was increasing this group relative to the other, but in the other periods he was not very much. So, again, if were sitting in 2000 and if this is the election of the george bush versus al gore, that may be in mind this second in which immigration getting there. Were seeking in 2015, which immigration is actually going back to be quite skill intensive and gone back to be very, actually, if anything helping a little bit. This is how the wages get in that period. If you think of immigration hurting the hurting the group that where they go in the largest part, the 990s look a little bit like that, in terms of that. But in all the other period, wage work increased in spite of the fact that immigrant ahead and upwards in the skill spectrum increase, so immigrant by themselves should have had the lesson you gave relative supply. Okay. I think at the National Level i think the numbers are explain. What about at the local level. Its certainly true that immigrants are distributely distributing among different regions. It is a long tradition that local immigration flaws and they have not found much of in effect and its on the wage at the local level. Thats why people moved at the National Level they said, yeah, local region are not close. The economy people move. Now, there is a lot of research on this. Let me summarize from the recent paper that we need to wrap it up. If immigration has local effect and important insignificant local effect in the pricing wages and the very first, you should find a negative correlation between places of large immigration and wages of low skilled or wages of general. Theyre going to play and they depress the wage in that place. You have to establish this correlation is caused, a little problematic. But even on a surface, on the surface, the correlation is interesting. If theres no correlation, it means that at least that there are other forces that offset this effective immigrant when they in a place or maybe immigrant flowing different places. This is a scattered of the change in the wage of the worker, its a percentage of the labor force. For all the u. S. Labor market, which are 722 commuting zones, broken down by decade in 1970, 2010. For more immigration, you will observe the negatives, here what youre observing is a positive correlation. These are the labor markets that were using. If you do the same for the changing and deployment rather than in wages of native worker. This will be the change in immigrant you get, again, if it is crowding out, if one more immigrant or percentage less later then you will serve, so in terms of correlation, you have zero correlation with employment and a little positive correlation with average wages. And even if you break wages into wage less educated nd you look at this, immigration, you will then see in much positive relationship and this is a correlation, but you find the relatively positive correlation with the wage of the it looks like, to some rising story, labor market where a lot of immigrant went. Wages of natives, for us andless the same that they grew even a is there a cause correlation where you start a lot of regression controlling and, again, just want to give you an idea, even if you that im pointing is that correlation. The correlation of lowskilled wages relative on the immigration is positive but sometimes not significant. Well do it at the local level or if you look at it state. If you try to address this point given constraint of time, economist to try to isolate the part of immigration flows just by preferences of immigrant based on that they prefer going whether other previous went and not attracted by jobs, even if you isolate that socalled supply push and you reestimated on average wage on wage skill, you find essentially a small not significant on the wage, very noisy on average a little bit positive effect on the wage of College Educated. This is that. So in a sense, this evidence altogether seems to say that just relative number are not there to create an eflkt and at the local level, the evidence that has been accumulated is that there is little effect, particularly little on the wages of low educated and may be a little bit positive effect. But let me spend the last 1 s 3 minutes saying can there be some challenge. The researchers have studied a lot of mechanism that makes plausible that effect positively of making it. The first is that even when you look at similar type of skills theyre not the same type, they take some occupation and jobs that are moving out. They specialize in more outdoor at the low end of the spectrum that immigrant, the native are meeting. So they have to say, they have a component of helping the productivity of native worker. This should be a channel will boost the wages. First, the efforts, sometimes just the type of technique and technology they use when there are a lot of imgrnts which do modeling job very well, they tend not to use what we call more and they tend to use the technology that uses more intensively and make them increase the productivity. And attracted. And where there are immigrant. Finally the and to generate potentially higher variety of food and services which are supplied and this could also be a channel of increased in productivity. High skill immigration is the large in percent ang term a part of immigration. Its crucial to theyre crucial to enhancing as some work on this. And i will add a couple of words on this particularly science technology, engineering the worker which by now seems to boost productivity. In fact, if you look at the distribution of immigrant, theres a share of the pob la population, you go from people with some college all the way with the stamina. You see that immigrants are really concentrated in the very high end of the education spectrum. About 30 with a degree in the u. S. Are immigrant. And you can calculate the effect of this, assuming that there is a positive productive external in this group, this group adopt better technology, this group increases increases the, you know, innovation. And so here just youre just referring to a paper, we thought increasing due to the increased cap of the visa that was passed in the 90s and withdrawn in the 2000, maybe its local increase due to that by about 5 for College Educated and 2 of nonCollege Educated because of immigrants. And there is Something Else well talk about. Im going to wrap it up. I think that this is thing that i showed, shows if immigrants had any part of any wage distribution in the 90s and it was relative to the other and local level to the contrary there seems to be some positive association that there are reasons to think if there was a little bit of negative effect, it will upset or generate positive effect at the local level and in particular high skilled immigration could have been important contributor to that. [ applause ] next, we have ethan lewis, who like my father long ago, is a professor, in particular, associate professor of economics and hes also affiliated with the National Bureau of Economic Research. As weve heard, it sound like hes particularly expert, shall we say, in the impacts of immigration of entrepreneurship. Although im sure lots of other related ideas and hes also published in lots of top journals on these and im sure other topics. Look forward to hearing what you have to say. Hi, so, thank you for having me hear today to talk about immigration. So as alex eluded to in the introduction, immigration has been in the news a lot, lately. Theres a lot of competing claims out there about what the impact, exactly, of immigration is. So what ill do is give you kind of an overview of how economists think about the immigration and what impact it has on workers and unfortunately, a lot of the things you see in the press and in the Public Policy debates, stem from misconceptions about what exactly that impact is. So i want to try to get past those, first, and get to the what we understand through all our evidence and work studying immigration that impact to be and so i called my two wrong and one right models of the impact of immigration. Thats three models ive got to get to in 20 minutes, i better jump right in here. Often the way these kind of talks about immigration or even in the Public Policy debates, the concern about immigration stems basically from looking at like the large numbers, right, like the baseline is some eight digit number that represents the number of immigrants living in the u. S. So this must have some huge impact on our workers. Im going to basically argue today, the absolute number of immigrants is the wrong way to think about immigration. So often in the kind of debates, theres kind of an explicit comparison to the another large number which is like the number of unemployed people. You see headlines like this, you see it in the policy briefs. Here is an example, number of Illegal Immigrants out number the number of unemployed. Its kind of an article here and, you know, if you want to call it that. And this, particular article was drawing on two separate unrelated sources, thats kind of recent estimates of the number of Illegal Immigrants and the bls report on the number of unemploy unemployed. Its kind of often implicit. But the argument is clear, right, so if somehow we got rid of the pesky immigrants, there will be all these job openings and the Unemployment Rate will go down. Is that really true. Well, that brings me to that my raw model number one where the logic of this breaks down. It comes from thinking theres a fixed number of jobs, right. If theres a fixed number of jobs, one immigrant comes in, some native got to lose their job for the immigrant to get a job. So, i just want to point that out that economist have been a arguing against this point of view for a very long time, you know, there was an economist in 189 the called david shlaus. Maybe youve heard this before. Its used to argue forcing workers to have shorter hours will generate jobs, same kind of argument. So economist have been arguing against this for a very long time and its just this wacky idea that wont go away. So let me just start by showing you theres not a fixed number of jobs in the u. S. That the number of jobs in the u. S. Has gone from 60 million to 140 million in the past years. Its more than doubled. It doesnt always go up. Theres recession where its not growing. In the past decade up in that corner there hasnt been that great as we all know and its these periods of slow growth where we start to get the feeling, not the reality, but the feeling that the economy is zero sum. But it isnt. More to the point. Theres a huge body of research and giovani showed the evidence. Its exactly about this exact question. This has been studied and studied and studied. Do immigrants displace zsh theyre not just showing up and sucking money out of the u. S. Economy, by virtue of being here, they are consumers, right, they demand all the things they need to live, housing, clothing, entertainment, et cetera. And as a result, that by itself is almost enough to generate a enough jobs to employ themselves, but on top of that, theres all the other mechanisms that were alluded to, that on average specialize in different kind of jobs and im getting to the end of my talk when i say things like that. There is supply and demand channels by which they raise Product Diversity thats been one of the most, you know, visibly obvious, which has been studied is that immigrants raised the diversity of restaurants in an area and this, by itself, generates implace of employment. They tend to start a lot of businesses. Theres many channels by which they can actually raise the number of jobs rather than decrease it. Notice that simple logic fixed number of jobs leads to the wrong policy conclusion. If we rounded up all the immigrants and sent them home, this will most likely lead to an increase native, not a decrease, right, like the number of jobs would be destroyed by this. So i hope, the audience looked pretty receptive to the idea that this fixed number of jobs idea is pretty wacky. Its so preva zif, i thought i had to beat it down. We all know supply and demand, more workers must mean lower wages. And youve probably seen a picture like this where youve got some kind of wage on the y axis, youve got the number of workers on the x axis, youve got the downward sloping line. They dont usually tell you where that comes from. It turns out thats very important, ill come back to that in a minute. You can draw it up towards if you want. Ive simplified it thats that red line there. So in a market economy in this the title with the labor demand curve is downward sloping. As and came up with this estimate that immigration is ksting us all 2,500 a year. Its put out something called the negative population growth. Theres information in the name of that organization that will come back to in a minute. So is this really true. Is it really costing us all 2,500 a year . That would be an enormous cost of immigration if it were true. We dont always believe what the harvard guys say. This one turns out to be nonsense and its, you know, ill explain very carefully what i mean by that. So that brings me to my model number two. It turns out that the labor demand curve. It comes from the assumption that the stock of capitol is fixed. If you havent heard the term Capital Stock the way economist use it. This refers to all the other input from the economy besides worker that produce out put. This is the buildings, lovely auditory yum, the computer, microphone, et cetera that we use. This is a substantial part of the economy, the tools we work with are responsible for at least a third of the of gdb. So the entire reason for that picture, that downward sloping demand curve. The assumption that that stock of capital can never change. So what happens in this kind of over simplified view of the world is when immigrants come in. The amount of tools people had to work with goes down. Theres less capital for each worker to work with. So maybe youre already also to my next point, economist dont believe this at all, i can start again with the simple fact that the Capital Stock is not fixed the Capital Stock of the u. S. Per worker has also more than doubled in the past let me explain why this doesnt really happen, why immigration doesnt really delete the amount dilute the amount of capital per worker. I can demonstrate it, so suppose giavani immigrants to the u. S. In the fixed Capital Stock view of the world, what would happen is because the stock capital is fixed, hes going to have to share an office with somebody and share a computer, you cant get another computer. Suppose him sharing with me, ill be less productive. But that kind of overly simple example illustrates why its not going to happen. Hes going to realize that he can be a lot more productive if he had his own machine and its not that expensive for him to get his own computer. Its enormous return on capital for him to buy a computer or get an office, that sort of thing. Thats why it doesnt really happen. Adjusts to bring back capital per worker to what it was before immigration. Economists get made fun of for saying this tonight. It turns out were not talking about the span of a human lifetime. Were talking about basically right away. In contrast with the big numbers, the annual rivals of immigrants to the u. S. Are quite small. Theyre, you know, less than 12 hundredth of the existing work force. Just to give you perspective. Thats less than the amount of workers that are at it each year just from basically kids growing up and entering the labor force. Unless you work at the, youre probably okay with that idea, right. Then youre going to capitol stock 2,500 the reason its absurd and nonsense, it comes from a very particular way of calculating the way of impact, it assumes that all 42 imgrabts arrived yesterday and they all and the capitol stock had no time to adjust. And thats absurd, right, like in fact, what happened they dribbled in over the past 50 years and the Capital Stock had plenty of time to adjust. Those are my two wrong models i promise you the right model theyre wrong in exactly the same way. Theyre focused on the number of workers coming into the u. S. What maers instead, not the absolute, but the relative number of different kind of workers janitors do not compete with engineers for jobs, right. It makes a lot more sense to think about workers of different types or workers of the same type competing with each other. So simplify things let me just imagine theres actually two types of workers unskilled and skilled and in fact that sounds simple and it is over simplified, as ill come back to in a minute. That turns out to be a pretty good approximation if you define unskilled as noncollege and skilled as college, what matters in this set up is how much immigration effects the ratio of unskilled to skilled workers. Thats when you can have an impact on the labor market. And it turns out, theres a formula for that and sorry for doing math in this Pretty Simple formula. What you do is take the number of unskilled immigrants to unskilled natives and subtract off the same ratio for skilled native and skilled native in this set up. Each of those ratios represents how much immigrants grow the work force of that type. Let me illustrate this formula with three examples. Imagine all immigrants were unskilled and that first ratio would be a large number, right. That second ratio would be potentially zero and so this would have a large positive impact on the unskilled, skilled ratio, make sense, theyre all unskilled so they push up the unskilled, skilled ratio. Suppose the first number were large so all immigrants were skilled and it will have exactly the opposite case. Theres one more example that turns out to be highly imagine if theyre roughly equally skilled and skilled. That first ratio will be the same as the second and the impact on the skill ratio will be zero. And in that case, the labor Market Impact of immigration will be zero despite the fact that immigration was a large share of the work force over all. More generally, what you would do is you ult ply this with the slope of a relative demand curve to get the wage impact. Its very similar to that picture i showed you earlier. The problem with the picture earlier, it was over simplified. To point out nobody is rejecting supply and demand, its a core product of economics. Im not going to reject that. That very simple model with one type of worker is not adequately describe the market of impact immigration. That will tell us what the labor Market Impact is. And thats shown in this figure here. Which is similar to the numbers that were just shown. The darker bars here, the noncollege immigrants. Thats basically in the each of the decades 90s and 2000, they raised it by roughly 6 . If thats all that happened, there would, potentially, be a modest decline in the wages of noncollege workers, in this very over simplified model. But notice, as was pointed out, thats balanced out by inflows of highly educated. Not all are unskilled. A lot of them are highly skilled, as we know. That balances out. Another way to put this, a lot of the popular focus was on the absolute height of those bars. That immigrants is a large share of the work force. But really what im telling you today is what you should be focused on is the difference in the height of those bars and thats small. Thats teeny tiny theres not much for immigration to effect the landmark. That model is completely over simplified. Thats not two types of workers and immigrants differ and large number of ways from natives even among the unskilled natives. Not as. Its a little more complicated than that formula i gave you. And he showed you the results of that and basically what you end up with when you do all the calculation, is the vast majority of native workers see wage on top of that, theres other mechanisms by which the labor market adjusts, so even that kind of relative demand curve that i ashrewded to that assumes a fixed Production Technology that they cannot adapt at all worker available, in practice firms do adapt and that has the effect of flattening out the labor demand curve and, essentially, making the harm, if there is harm, which its very concentrated on immigrants themselves, is foreign is reduced. So to sum up, i showed you two wrong and one right model of the impact of immigration. The wrong model number one theres a fixed number of jobs this is completely wrong and so preva zif and leads to exactly the wrong policy, so if we actually rounded up all the Illegal Immigrants and sent them home, this will probably lead to increase in Unemployment Rate of native born americans because it would destroy jobs, not create them. Model number two, is implausible on its face, but in practice because capital actually adjusts, theres basically no long the u. S. Economy is scaling good at scaling up. Scale is the adjustment happens in the time frame that immigration is actually occurring. Its kind of short run calculations are not even very relevant. The right model is one in which immigration effects the labor market when it effects the relative supplies of different types of workers, thats why the right focus of policy is filling shortages and that sort of thing. Thats the right way to think about immigration and not the absolute number of workers. In practice as it turns out, as i said, immigration does very little wage harm. If there is any harm it seems to be concentrated on the immigrants themselves. The vast majority of immigrants, native born workers, were basically all better off from immigration. Thats really hardening in this kind of time of rhetoric about immigration. Thank you very much. [ applause ] thank you. Our next and final speaker is Research Fellow in the markets, trade and Institution Division of the International Food policy Research Institute and he has a lot of other important titles. His research has focused on the effects of migration on source households. Were focusing on the receding country and the role of women in Rural Economy developing countries, ive used some of that work before and understanding the impacts of agri cultural intervention. Covers a broad range and i look forward to your comments. To the extent it doesnt cramp your style they ask you to speak in the microphone. No problem. Thank you. Thank you for having me. Im pleased to be here. Im going to attack a little bit the elephant in the room, which is the famous paper written in 2003, entitled the labor demand curve is downward sloping. Before i do that i want to say im a Development Economist i take this from a little bit different perspective. So im thinking about the migrant in mexico, actually written on migrants in el salvador, internal migrants in china, migration within that kind of thing. So thats a very different type of, its a little bit of a different you can think about it as International Migration with free borders, effectively, although thats not the case in china. Thats a way of thinking of migration that im kind of use to. The other thing Development Economist that we concern ourselves with is omitted variable bias. Labor economist concern themselves, we came a little bit late to the game. What i mean by omitted variable bias when we try to find relationship between two different variables we worry a lot that theres actually Something Else out there thats going on i actually saw when i was a professor of economics. Basically he made the claim that the labor demand curve is downward sloping, everybody is estimating it wrong, effectively. Its not accounting for adjustments to local labor supplies, et cetera that the wage of elasticity is minus. 3 or minus. 4 rather than these guys argument. Okay. What i want to do is say lets not believe them for a moment here. Rather than the. What does that mean, lets be careful. It means if we raise the relative supply of one part of the of the labor force. The wages in that group will decrease by a certain percentage. If we raise the relative supply of 10 , theyll decrease by 2 was the systematic reviewed estimate and last year he used that minus if we have any kind of increase in migration, theres basically effectively potential costs to the north. In terms of gdp, were not talking about costs in terms of wages, were talking about real costs to the economies. Gdp per capita will actually decline with large Scale Migration from southern countries or further countries to the north. So one thing i want to note is that and weve seen this a little bit and talked about it a little bit. You know, if we look at historical migration, i took the first three bars from one of the papers, actually, and then the fourth these five bars from 1960 to his data and then 2010 and 2015 i added. You know, migration immigration to the u. S. Was or immigrants in the u. S. Labor force were huge component of the labor force until basically the Great Depression. Wages decline and employment declines, actually theres a lot going on in that period of time. Regardless, 1960 and 1970 when he starts his study, theyre the lowest point in migration. Its increased but it hasnt nearly increased as much as in the 1910s and 1920s. It looks like it leveled off with a big recession that weve had makes sense. So im im going to focus i focused on admitted variable bias when i saw this paper. I focused on the wrong variable that im going to explain. Theres a bigger labor force shock that takes place between 1960 and 2010, which is the entry of women into the labor force, which many of the audience are female and you all know that the labor force changed dramatically in that period of time. Now in 1960 and 2010. So what my coauthor joe russell and i did. We tried to replicate the paper and add women. Lets see if we can get a similar we thought maybe thats the omitted variable by us, naively. Im going to explain why its not. But why it tells you a story about why we shunts believe. Not maybe, why we really shouldnt believe this model. Well go back to believing giovanni and ethan by the end of my talk. The share of women should act the same way, if the labor demand curve is downward sloping and we have the right way to estimate the labor demand curve, if we look at women, women should decrease mens wages. Were not going to look at overall wages, were going to look at mens wages and see what happens to mens wages when we add women to the variable. Let me preview my results in case you get bored and stop listening to me. First were going to replicate borhaus and just doing that gets me back to minus. 2 which is the standard estimate in the literature. That gets us right back to a normal normally estimated wage elasticity of immigration. Im going to find no correlation between womens entry and immigrant entry, which is not a surprise, because theyre entering different parts of the labor force, although, to some extent that might not be quite right, based on what i heard earlier, different parts of the way that borhaus defines the labor force. Im going to get a positive co efficient of the entry of women into the labor force on mens wages, using this model. If i uzan youllized wages, the effect is significant aeb positive. Im not using a labor demand curve. What are we going to do, were going to exactly replicate borhausz, split the labor force into 32 education cells. The education levels are more desegregated than what ethan just used. Were going to go less than high school, high school education, some college and more. One to five, up to 36 to 40 years of experience. Were going to measure wages as the average for that census year among all members all male members of that education experience. Were going to drop women, there are a number of other things that we tried to replicate. Were going to consider men aged 16 to 64. Its effectively difficult to do that. He argued it was difficult to compute experience for women. What weve done is tried to adjust for the fact that women drop out of the labor force for short periods of time to have children. Were going to adjust the experience. We used the calculation by two labor economists to do that. Were going to end up with the share of women womens experiences is going to be slightly lower for a specific age than men. Then we estimate an equation that doesnt look very good also. Lets skip it, really, what matters is p in the first equation, thats the borjas equation. And p and f are going to be the share of women. Essentially the share of immigrants is the number of immigrants over the number of men, and the f is going to be the number of women divided by the number of men plus the number of women in the cell. Okay . So first i want to show you that we replicated borjas data pretty well. He published a table in his paper, we came really we have a we did it ourselves to make sure that we we did it ourselves and we find a correlation thats. 994 i think. This is the actual data. We came really close, we didnt get his exact numbers, that probably has to do with slight differences with coding, we should always worry about that when you see work by one of us, were always we try to be as careful as we can. Youre always making assumptions which david has written about. Thats probably why im saying something. To show you some of the changes in shares and defined here, if we look at so breaking up the labor force, actually the share of immigrants in this less than High School Cell goes way up, right . But that could have to do with the fact that people in the u. S. Are graduating from high school in larger numbers than ever. And so thats the cells that we see growing the fastest. Where as other cells, we saw giovanni was talking about how fast the share of immigrants is growing in terms of College Educated people, its a small component of the College Educated labor force. Thats an interesting. We have to be careful about the way were defining things. Im defining them differently. We look at the change in the share of women by education level. Note that the bars are again increasing, but which bar is increasing the fastest . The bar on the way on the right. Women with a bachelors degree or greater are entering the labor force in much greater numbers. So if we put these things together. And put them on a graph, if we look at the share of imgrabts in a cell, versus the share of women in a cell, what we see is that we see this big bunching of blue dots, the blue dots are less than high school. So thats where the migrants are, and so they have all sorts of different experience levels, thats what the difference is here, the varying experience levels. If we look at the share of women, theyre spread out along that line. For some reason we get a regression line that means something, it really doesnt. Because the slope is insignificant. Because of that big bunch. You get the xs over here, are women, and theyre spread out among the cells, and i can tell you tell you for sure, what are these why are there so few women over here . Well, theres so few women on the why are there any cells over here at all . Is what i wanted to say . Why are there any cells over there at all . Older women dont exist older women with College Degrees are a very small share of the labor force. These are younger women, thats what were seeing as the difference. We look at the correlation between weekly wages and immigration. This is the regression line. These are all of the data, this is what we get. This is like the borjas co efficient, essentially. Thats a picture of what the co efficient looks like, essentially. We replicate this with the share of women, what do we get . We get an upward sloping line. It has to do with i want to be really clear, this is mens wages that im using on the y. Is. This is not womens wages, this is mens wages. If we look at mens wages, using that same model, we get and you see the xs in particular, being the College Educated women, when there are more College Educated women in a cell, mens wages tend to increase, okay . So that has that probably has to do with the capital story or that we were hearing earlier. Or it has to do with a complimentary story. When women enter the labor force, there are new ideas, in particular, the parts of the economy that the women are entering. After i show you the co efficients and the elasticities. So heres our replication of borjas on the left. If you believe that model, even after all of my nice pictures. What you see is that the co efficient drops from minus. 4 to 53 to. 348. All we have to do is add a little more data, ten more years of data, and we can get back to the standard systematically reviewed co efficient in the literature. If we add women to the if we just use women in the model, the second row, we dont see anything. If we add women, we dont get any effect, theyre entering these totally different education cells. Thats where my original hypothesis was completely wrong, and im admitting to that. If we look at the impacts on annual earnings, we get an insignificant effect on immigrants that may have to do with the labor supply story. I dont want to get too far into that. The point i wanted to make, on the share of women, we see this positive co efficient that i was explaining before. That to me means that this model cant really be estimating labor demand. We need a different model of labor demand thats more sophisticated. Let me give you one suggestive explanation that hasnt been described yet. And that has to do with not only have we had a huge increase in capital, but a tremendous structural change that took place in the economy between 1960 and 2010 in the u. S. So just looking at the share what ive done is use the value added shares. The green piece here,the darker green, sort of lime, services. Services are always the largest parts of the economy. The green is manufacturing and what we see is this huge decrease in the share of gdp thats coming from manufacturing. That shouldnt be a surprise to anybody in the room. None of this should be much of a surprise. The blue and orange are where were seeing the most growth in value added in gdp. That is professional and business services. And finance. So were becoming a Service Based economy and knowledge economy. And this is reflected directly in the gdp shares. By value added. Its not surprising we need a lot more workers in those shares. Thats why we see immigrants coming in with ph. D. s and we see women entering these labor forces and having a strong correlation with mens wages. The purple is government and its almost exactly the same size, 50 years on. So to conclude, by adding 2000 if you dont believe anything else ive said, take that if we add 2010 to borjas data, the immigration no longer is no longer there, and if we plug that elasticity, the minus. 22 into borjas own spreadsheet, which he put in his journal, put online as part of his journal of economics literature paper, into that simulation, we get the trillion dollar bills in the sidewalk, being gains to migration. In the world if people move from the south to the north in terms of gdp. All we have to do is move some people from the south to the north and we would see gdp increase. That said, the co efficient on the share of women suggests a structural flaw in those estimates. And its weve seen a lot of evidence from ethan and giovanni that there really isnt that much competition between immigrants and natives for jobs, et cetera. So if you want to i just want to leave you with the takeaway, that we should be really cautious about believing any estimates that the wage impacts of immigration are that high. Thank you. Im going to stoke the discussion first with a couple discussions for our panelist, we have about a half hour. If i may presume to sum up a couple things here, theres a strong message here that we shouldnt be so worried about the impacts of immigration on the Domestic Labor market. And i think theres sort of three planks to that. One is that not only do people who move here work, but they eat. That is to say they produce so they compete in product factor markets, such as the labor market. They consume, they have to buy food, housing, et cetera. Theyre stimulating demand at the same time, the other is the Capital Stock adjusts, right . And it can do so instantaneously. Not only because the labor inflow is relatively small, but its relatively stable. If youre extrapolating from past trends, you look at whats going to happen make your best guess of whats going to happen last year and you already factored in the expansion of the labor supply. The third is the notion of low skill immigrants. I always use the restaurant example. If you have a bigger supply available to you of people who can work in the kitchen washing dishes and such, that improves the economics of the Restaurant Business and allows it to expand and allows you to hire people that work at the retail interface and have to have more cultural savvy. These are all sort of dampeners on what we might otherwise expect to be a harmful effect. So my first question. It was inspired by ethans passion about rebutting these models. Its for any of you. I am a child of divorce, it was a pretty you know, it was a pretty angry divorce, i grew up living with contradictory world views, somehow reconcile them. So to avoid group think here, i want to ask the panelists, whats the strongest contrary argument you can make. Who is the most likely to lose if only in the short term economically. And, you know, im imagining somebody working in a mining town where theres there really seems to be a fixed supply of jobs. You tell that person, an immigrant comes and is not going to hurt you, not going to compete for your jobs, that may not be very reassuring. Whats the strongest contrary argument you can form. Ill take the first crack at that. I mean, you know, theoretically immigration could do harm to those mining workers, to any number of low skilled workers or high skilled workers if they were that concentrated. It turns out that immigration. Immigrants tend not to locate in the places where they would compete the most strongly. Thats part of whats going on. Another feature that we havent talked about, is the fact that theyre responsive, theres a lot of research on that as well, immigrants are the first ones who kind of move out of a market when the economy is not doing well, that has a benefit for the u. S. Labor market as well, it diffuses recessions in local labor markets and makes it the concentrated effects of a recession, less harmful to a particular location. That would be what i would say. I dont know if theres anything else. Yeah, so, maybe another couple points through this there are some types of jobs that have suffered a lot in the last 20 years for a number of reasons. The example you give which is mining, mining is disappearing, because we buy the same product cheaper somewhere else. Buying jobs are disappearing. There are some jobs, if people are stuck in those jobs, theyre going to have a little bit from migration. The worker may be older, they have a very specific skill. Those are the candidates to be more hurt. However, you know, our economy has dozens of other forces which are sort of endangering the same type of jobs. Clearly, if even admitting a little bit of hurt from immigration, the solution to the fact that there are no more manufacturing or mining job, but is clearly not stopping migration. You go to a policy pint, this is a smaller part of the problem. I would say economists, who should look at cost and benefit should strongly say no. Look at other type of solution and i mean, in in a positive way, course education, and training, and more mobility, and increasing the ability of people to move across labor market. This seems to be one order of magnitude solution first for those type of worker that would be heard relative to immigration. Do you have anything you want to add . Yeah, i do. I want to pick up on the cost. I think lets pretend were in a world where the three of us are listened to on the National Stage and they increase. We reduce migration barriers and let in more migrants. So what we have to do then is also think not just about the costs of retraining workers which we neglect all the time thats something that economists theorize about all the time, it never really happens in policy. The other thing the other cost is actually in thinking about education and health policy. I think more about education because i have kids in Primary School and my local Primary School has a transient population, its very migrant and the kids move in and out of the grades and they become very difficult to educate, because theyre moving from one apartment building, which is one district to another, and the or they move from Prince Georges County to Montgomery County or vice versa, or into d. C. Or what not, some of those kids to get here have been through a lot, and they need psychological services, and they need help with english, and they need or they and i think we need to think about those costs im happy to i believe there are no wage impacts, but there are impacts we need to think about as a society that get missed. My other question may be particular for giovanni. We talked a lot about the work of george borjas, an economist at harvard, he has a new book coming out next month, i believe, not coincidentally, i think time for the election. And my best understanding is its meant to be a popular Book Building on his past work rather than presenting new research. One piece of that that hes pushing very hard, is an analysis that he did last year of the impacts on wages in miami, of the mario boat lift, a major influx of immigrants from cuba into florida in particular miami. I think it was 1980, right . April 1980. And giovanni has written a response he may not be the only one, but hes one im aware of. I think it to the extent this book gets coverage, that is going to be something that gets a lot of attention. The mild controversy back again its not that i enjoy fighting in life. I would like to research and argue why its good research. But so let me i mean, being honest of course people make choices in their research of design. There are important choices, but i think that ultimately two things, how relevant is this debate to the overall immigration this is refugee. Its particularly relevant for, lets consider an evaluation of the sudden flow of refugees that come into account. Which is different from 90 of the immigration in the u. S. Which is for economic reason. Lets say, were talking about refugees, and were using this as a relatively things that approximate a little bit scenario in which doomsday, in terms of wages, they come a lot, they are concentrated, how much effect do we find . This debate started because the first paper didnt find any effect. Theres this new paper that finds some effect. I would say that summarizing in two lines is we have very sparse data on that period, and sometimes we calculate average wages with not only 10 observe. We also have a choice. There are some deficit which are a little better than others. Using the bigger deficit, which allow us a few hundreds of observation for that period, and using the market that should have been effected, people with low education, both hispanic and nonhispanic was born, and men and women back then, i simply cannot find any effect whatsoever, if i find something a positive belief of wages around them. There are some deficits which are much smaller, which are out there, and there are some that you can construct. Theyre eliminating a lot of them, which give which can give some negative effect. There are other samples you can construct which gives you opposite effect, so theres so much noise in this data, taking everything on and saying, this one specific sample is the only right one, ignoring what others tell you, seems to me you have to quiet a lot of faith in what the answer is, and what you give, rather than letting the data speak. After all torturing of the datd, we dont find very much effect there. I think that shouldnt be. Because the immigration debate is about in large part, in the u. S. , economic driven and much more distributed. Giovanni is right. Its an extreme case, and unfortunately, theres not a lot of Labor Force Data on miami in that period. But theres better data, theres better data on firms, and i have found that firms adjusted quite well to this. They shifted to producing things that could take advantage of the in flow of low skilled workers, this probably accounts for the fact that you find nothing the way giovanni described it. Id like to open it it up, there should be a microphone that will be passed around. I hope you have lots of questions. Please wait until you get the mike, and identify yourselves. Hi, im rachel i can hear you, im not sure everyone else can. Rachel with welcoming america, i want to thank you for your research and your remarks today. One of the things that were seeing as a trend is that despite sort of the National Rhetoric around immigration, when you get down to a local level, its a much more pragmatic view. Were seeing many cities across the country who are trying to attract and retain immigrants as part of their strategies. Part of that rests in rust belt cities or cities like st. Louis, which again in contrast to this narrative of expelling, st. Louis has a plan to become the Fastest Growing immigrant population by 2020. And the reason for that is because they like many cities have a declining population, and a declining workforce, i wonder if you could speak in your own research around i mean, theres a variable there around the aging population, and aging workforce, and how that may impact some of the numbers. I have two considerations on this, this is an interesting topic, so i think one of the difference between immigration and trade, so that i think is emerging a little bit in the debate and in the research, is that trade is also eliminating some local jobs, because some sectors are shrinking significantly. And maybe is not of course is decreasing the prices of goods, as many other economic benefits, at the local level, maybe it does not generate any positive multiplier once those jobs are gone, there are economies that have found some negative effect of trade on jobs. Immigration is different, there is this one person more, which is in there which also consumes, which adds to the local normally, this is a relatively young person. Its a combination of people who do minor jobs that are needed. Its an Interesting Group of people to attract in a location, in order to revitalize it, because and i think here economies do this interesting job on the job Multiplier Effect of attracting some jobs, if you attract some jobs, you create other connected jobs, which are sort of supplying, and i think immigration has some of that flavor. And in the aggregate definitely, the fact again that there are age distribution prevents some of the shrinkage of our labor force is interesting and in some low calty more than in others. I really do think that immigration is a more effective way of generating local jobs, than other type of mechanisms, which are also at work, and i totally see why immigrants are easier to attract to positive they respond more to positive economic conditions. How some cities have tried and they come back on revitalizing the economy. Its interesting here and there, some research. I think more is needed and more is interested. Ill just add one brief thing, i talked about the adjustment of the Capital Stock, theres a little bit of asymmetry, its easier to adjust up than adjust down. What happens in some of these locations, the schools are kind of decaying, and its hard to adjust the Capital Stock downwards, one of the things immigrants do, they support the housing market, the public sector. A question back in the center. Hi, im from the reason foundation, you mentioned professor borjas new book, i just recently finished a print debate with him on this book. Which will appear by the way in the issue that will hit the stands in october. One of the claims that he makes in his book, that im interested to hear, you take on and respond to is, he says theres an efficient gain of the economy. Essentially represents a decline in wages. And what happens is native wages decline, theres a redistribute of effect for the owners of capital. What immigration does is, it does act to the economy, but it redistributes wealth from workers to owners of businesses. Theres a redistribute of effect. So what do you make of that, and how do you respond to that. This is the second model ethan put up, exactly the effect. This is the consequence of the assumption that the labor demand comes down, because capital is fixed, and is increase. That is calculated out of the exact graph. Efficiency, gain in economics. That is a model that we have left for 20 years. We have showed you mickey mouse theyre much more sophisticated, capital cannot be considered fixed in this calculation. Much better approximation is that capital adjusts the same way, if you assume that. If the labor demand curve is supplied, the efficiency demand is zero in effect. But then you need to look at the empirical evidence on the productivity of immigrants. This assumes zero productivity effects on a return plus fixed capital. I would say its not even a starting point for this conversation, the way i see it. Im disappointed to hear he went with that. I was willing to give him credit that that was the Washington Times or examiner. Thats just wrong. Thanks for the presentations. Ethan you concluded your comments with i dont know if it was a throwaway, or id like you to say mother about the idea of filling shortages, it seems to me the shortage analysis competes with the idea of an adjustable Capital Stock and capital i mean, it seems that using immigrants to fill shortages is an alternative to forcing capital to fill those shortages. Im wondering what effect that has versus allowing the Capital Stock, and doesnt that change the ability of workers to bid up wages. Let me take a couple questions since i saw several hands here. Im going to talk about the elephant in the room for a moment. Undocumented immigrant workers, versus native workers. And a situation that is probably changing in the next few years, the increase in the minimum wage. What will happen if Many Employers find that the requirement for increasing the minimum wage affects them adversely and they begin to hire more undocumented workers to replace the workers that they have, because they cant afford to pay them the salaries that they have been paying. Will that affect both the wage level for native workers and a income of unemployed. I heard the term used here today, not very much of an effect. Im not sure whether that can be quantified particularly in the micro situation like this. My first thought on that is that my economics models are good for increasing, looking at small changes in wages. So the literature that im a consumer of this literature, not a producer. The producer of minimum wage is that it really doesnt affect employment. That said, we dont know what happens when you go from the minimum wages we have now to 15, its a huge jump. And so even the simulation models i dont think any of us would believe what that would do. In terms of what youre what you think from the employers perspective. The employer whos an optimizer, because im an economist, i think hes optimizing, hes going to look at the probability of trying to shift to more capital intensive production relative to the cost of hiring, people below the minimum wage and the risk of getting caught and fined. My sense is that if im a risk averse capital owner, im not going to want to take that many risks, i would believe more in the shift of capital replacing labor than the immigrant the undocumented immigrant taking the place of labor. Thats my immediate thought about that, im talking about ilLegal Immigration in the next session. I think we have a lot of evidence that theres more changes that dont seem to affect employment very much. And the idea that people use more efficient what they have, theres a push toward productivity, a distribution of the costs. 50 of increase of wages, would be a little bit out. We love to see. One thing about undocumented. All paid minimum wage, firms dont want to attract attention on themselves for sure people say they are paid less in under our scrutiny, they were all paid minimum wage which is low right now. If its 15 a day undocumented immigrant right now, are kind of like other workers in many respects, they are on the books, they are paid as the other, they just dont have a Social Security number that matches anything. And its not that theyre in the gray economy like people claim in europe. Theyre not selling drugs, theyre doing jobs. On one hand it will help a lot, the wage of that group. The difference is that theyre paid 10, 20 less. But still currently, minimum wage is so low, so few jobs go close to paying that ourly wage. There is a question what minimum wage will do. What if we also regularize its not going to happen in one ear. And it will be an interesting time to see what would happen for a policy response that yeah, it would be plausible if some election goes one way and not the other. Its a possibility. The question is, was it something more along the lines of, is it better to wait for innovations to fill the kind of shortages in the economy or is it better immigration the solution to that. I dont have an answer to that. Those are two sides of the same coin in a way, theres an interesting example from history when we shut down the borders in 1925 theres a lot of concern about whats manufacturing going to do without all this workforce. There was a big analysis of this. And then the Great Depression happened, and then the same analysis which was ongoing switched to, is automation taking all our jobs. Similar to the debates today, it illustrates those are two sides of the same coin. Thank you. Can you verify that all of your numbers on immigration include both Legal Immigration that people hold valid, including people who overstay their visas. And secondly, maybe somewhat anecdotal, if i look at washington, d. C. , where i live, i observe that an awful lot of the construction jobs office buildings, walking on roads, renovating houses, appear to be held by immigrants whether legal or undocumented, i have no idea. While theres still a fairly large unemploimt of africanamerican males in the city. There appears to be a fairly substantial impact in this city that i cant reconcile with theoretical studies that youve done, i wonder if you can speak to that. The numbers that we use are from sense us, that i try to reach every Single Person who is residing in the united states. There have been these are the estimates from which people start to calculate the number of undocumented. You subject how many entered legally, this is the best way. There are some estimates that say people move around, we dont have 100 undocumented the numbers we use are the best we have to include both documented and undocumented. About the construction jobs and the high unemployment demand in the black community. Youre hitting the short circuit that a lot of people sight. They are unemployed. Kick out of those people that get employment. There are many reasons why some jobs pay relatively legal. And some of these undocumented do not have access. They are the push to work, which is stronger. The employment rate of people with no High School Degree is about 70 . Maybe about 30 . Because they have access to a lot of other benefits. Its not so painful. They have benefits if they are poor, they have medical insurance now. If they are out of their job. In a sense, the incentive is different. And you can argue that taking away the jobs some company will have to shrink. Keeping in mind a lot of those people may have some issues with with alcohol abuse, drug abuse, and all of these seem to be smaller in the immigrant Community Whos working at high levels. Its hard to tell, i would say that the best we can do is trying to look systematic over the nation, to see if places where jobs of immigrants have grown a lot, have corresponded to shrinking of jobs of natives. And as we were showing up there, the opposite is a little true. Cities which are is riving are creating both types of jobs. I know the appeal of saying, kick them out and see what happens if the africanamericans get the job. I dont believe this is the way the causation goes. But im sure some people would be tempted by your type of thinking that, you know, they may try some stuff on this way, i think thats exactly the appeal of putting this together to say one would solve the other. Maybe we this is the top, of course. And i understand, you just put two facts together and then s e some i didnt mean you were saying that, i meant this is the argument by saying, the undocumented should be created nor the natives. Why dont we collect a few more questions and have short responses, is that okay . Up there. This is more to hear the questions than to get full answers. My name is richard, i currently work with the american continental group. Im here on a oneyear visa in ireland. And in 12 days i have to leave because my visa will run out. How do people get vise is as over here, i have come over here, worked every day for three different i worked for government, the ted cruz campaign, the international zairery association, but it was when im looking to stay, i would have to convince a company they have to pay 5,000 and convince the Immigration Office that an american cant do my job. How do people who are these unskilled people and how are they allowed in here, when people who are coming over and very willing to work are told they have to go home after a year . Another question . Any last questions . My names dee, im a lawyer, have you looked at it internationally, your research compared to other countries. Can we learn something from other countries and the way they do their Economic Research . This woman up here . Foundation for empowerment. I have a question, i tried to take a lower level position. You talked about the southern refugee from cuba in miami, that has the overall economy in miami in germany, and you know how many refugees are coming. Im aware you are working on labor markets. What do you think about the impact of the refugees in germany or other. Particularly in germany. And my second question is, i heard that the Hillary Clinton is just like another in america. When you show the number of refugees immigrants in 1930s and 40s, the lowest point of immigrants in america refugees from europe. What should be american policy for refugees in these situations. Thank you very much. Stop it there, i want to give panelists 30 seconds each to very incompletely respond to these questions or offer final thoughts. Id be happy to start. We cant give the time to your questions they deserve. Richard, just to on your point, i think a lot of unskills workers get here through visa lottery. I empathize with your plight. And a lot of people overstay their visas which is how we have Illegal Immigrants in our country, not to give you any ideas. Dee, im going to focus on these two we do work on. I work on other countries and not much on u. S. Immigration. Just to give you the example of gains to migration. We tracked migrants who were moving five years later, people who had moved, made basically their consumption went up by controlling for all kinds of other factors. Controlling for their Human Capital basically. Similar to other people. I was going to respond. We made a policy decision in the 1960s, to favor family members of existing immigrants, and thats where a lot of the unskilled immigration comes in. You know, unfortunately, its very hard to change policies, its not necessarily the best poli policy. A lot more countries have more control over the flows of immigrants, its easier to admit immigrants and change policy to fill shortage and that sort of thing. Theyre all family members is there research on other countries. . Yes. It is very useful. Each country thinks theyre very unique, there are some general consideration. Such as europe has an issue, a problem right now, i think is more linked to the fact that they have ignored this problem, and theres no planning on how to respond to migrant flows. Its been clear for four or five years, that this was going to be an issue, but europe has some difference. With a much more generous welfare system. This makes european immigrants much harder to fined it a job. Much more on welfare states, i go there and talk to them, i say the american system seems to be better people are here, and the best welfare you give to immigrants is a job, thats whats been working in the u. S. Both for the economy and their integration with a lot of issues. We need to continue understanding how this work and plays across other countries. I want to thank all of our panelists. Forgive a very short plug. If you go to david road