 I want to thank all the participants for participating and coming. I want to put this a discussion about it's going to focus mostly on immigration policy and I want to put it into a larger context. The immigration issues that people talk about fall into two camps. One is what to do about those who are here undocumented and the issue of a path of citizenship and all that. And the other issue is, and those issues are usually talked about, don't break up the families, about what the contributions of immigrants can make to our country, to issues of democracy by not having second class citizen or second class status and many other issues, all of which are giving opportunities to the dreamers, etc. Those are all important issues. Another part of the debate is about guest workers. How many, how do we decide, what are the rules and regulations that govern them. So I want to bring this discussion to the larger question about wages in America, which is going to be one of the main topics of discussion during the presidential debate. We have, I'm really happy to report, that both the Republican and Democratic candidates are talking about wages and what we can do about them. And you know, that's not something that was true a few years ago. I mean, four years ago we were barely talking about inequality until Occupy Wall Street showed up. That conversation about inequality didn't really get to wages, even when the president started talking about the issue. It eventually moved that way, I think because of the fight for 15, the fast food workers, Walmart workers, and it actually became bipartisan when the Republicans noticed the outcome of the last election, two things. One the unemployment rate had gone down quite a lot, so they couldn't attack Obama about jobs as readily as they used to, but they also noted that people were concerned about growing incomes and especially about wages. So they started talking about wages, which is not actually a very traditional Republican issue. So you know, here we are. We have an EPI, we have formulated an agenda for how to raise wages called raising America's pay, and I think it's been distributed to everybody. One of those items has to do with immigration, which is where I want to make the connection. We are saying that providing a path to citizenship, not just authorization, providing a path to citizenship is a critical part of raising wages in America. We have 5% of the workforce that are unauthorized, undocumented. We know that people in that situation don't have a lot of legal rights. They've actually been blocked by being covered by various laws, and they have obviously the threat of deportation and all of that. They are vulnerable to exploitation, and they are exploited, which is kind of a natural law of capitalism. If you are vulnerable to exploitation, you will be. That's the way employers and markets work. So the papers today provide some insight about that, and I want to point that out because a lot of the discussion of the papers has to do with comparing what happens to temporary workers or people with green cards versus unauthorized workers or temporary status versus unauthorized workers. I want to focus on what having legal rights and especially citizenship has on wages, which is part of the larger topic. So Lauren's paper makes a great contribution to this from a unique data set of Mexican immigrants, which is not all immigrants, but that's a pretty important segment of the immigrant population. Read in table three on the fourth column is an estimate that I can read that not everybody can read, but it says that people with green cards, LPRs in the terminology of the paper, which it doesn't come naturally to me, but make 13% more than those who are unauthorized. That's an important finding. Now that doesn't show what the effect of citizenship is because citizenship, actually, by other studies, shows an even greater wage effect. So there's a wage advantage to being a citizen, being naturalized citizen versus having a green card. Our Manuel Pasteur, a long-time collaborator at EPI and a member of the board, has done research on some special data in California where he argues that the effect of citizenship is far larger than the effect of having a green card. So that would mean the 13% differential could be twice as large for citizenship. Her data doesn't have the identification of citizenship to be able to make that estimate. Tom Hertz's work in agriculture has the disadvantage of being for one sector, but it has a tremendous advantage of having actual data on exactly people's legal status and standing. He does show a wage effect of having a green card and shows that being a citizen has a bigger effect than having a green card. Both effects are relatively modest, but that may be because of the special nature of you're just capturing people within a sector. But it does sort of confirm the idea that there is something different about citizenship for just having legal status. So I just want to then note from this that the implication should be that those people who are arguing for comprehensive immigration reform and those people who are arguing for better wages in America should recognize and articulate that immigration reform can make a great contribution to better wages in America because having 5% of the workforce standing outside of our labor standards can't be good for those workers and it can't be good for the workers in similar fields of work. We don't have specific estimates, which is a frontier of social science that perhaps Lauren can turn to. But how do we estimate what the spillover effect is of undocumented status, the wage penalty and the wages of other people? The undocumented workers are concentrated in particular occupations and in particular regions and that's something that we hope to explore in the future. But I just want to illuminate those issues that come from these papers, thank the authors for their work and now we can have the rest of the discussion. Thank you. Thank you Larry. So since we started a little bit late I'll just give a very brief introduction of Lauren Apgar. She is a PhD candidate in sociology at the University of Indiana, Indiana University sorry, in the Department of Sociology and she's almost done, right, on wood. So her research interests center in the field of immigration and stratification and political sociology and work and as Larry alluded to she's actually done some groundbreaking research using the Mexican migration project dataset. So with that Lauren please go ahead and put up your presentation. Well thank you everyone, it's great to be here today. Thank you for your interest and the guest worker experience. So as you may know temporary foreign worker programs allow migrants to enter the US workforce for a designated period of time after which the workers return to their home countries. And legislators view the program as a way to decrease unauthorized migration especially that from Mexico. So the idea behind the program or one of the ideas is that Mexicans aspiring to work in the US would obtain a guest worker visa instead of crossing the border to look for work without authorization. However I argue that if guest worker programs are really going to be a viable alternative to unauthorized migration they should really reduce the risks that they encounter in the workforce. To that end I wanted to examine guest workers work experiences. So my research question is how do temporary workers economic outcomes such as their hourly wages compare to those of legal permanent residents and unauthorized workers. So just briefly I'd like to give some background on the legal rights of temporary workers. Under the guest worker program Mexicans commonly enter on H2A visas for agricultural work and H2B visas for low skill or seasonal work. All H2 workers are legally authorized to work in the US for one year and up to three and jobs not filled by Americans. And all H2 workers admission to the US is based on a visa that's tied to their employer therefore they cannot change employers or jobs while working in the US. And then H2 workers must be paid at or above the average hourly wage in the occupation of their employment. So these workers are distinct from LPRs or legal permanent residents because LPRs have most of the same rights as native born citizens in terms of employment they can negotiate with their employer they can change jobs and then their temporary workers are also distinct from unauthorized immigrants especially because unauthorized workers face deportation if their presence is detected by immigration officials. So from the rights given to these different groups you know we might think that there is some kind of hierarchy so immigrants with limited rights might have lower wages than those with more rights and there has been research showing that LPRs earn more than unauthorized workers because they have rights unauthorized workers do not. So we might think that since temporary workers rights kind of fall in between these two groups, their economic outcomes might fall in the middle. However, one study that actually compared the work experiences of unauthorized workers with temporary workers found that both groups actually suffered similar workplace abuses such as not receiving pay for all of their work. So that was a study done in the forestry industry and it really kind of suggests that maybe temporary workers and unauthorized workers have similar labor market experiences. So the question really remains how do the wages of these two groups compare? So I use the Mexican Migration Project data to answer this question and this data is a really unique data set. It's a random sample of heads of households from various communities in Mexico and so here's a picture of the various communities where the data have been collected and it's been collected yearly since 1987. Because they survey household heads, it's usually mostly men who have completed the survey and they ask household heads if they had any migration experience in the US and if so they collect detailed information on their most recent migration trip. So the data have experiences in the US from 1987 to 2011. And then the MNP measures hourly wage so it asks the workers the amount they earned per hour in their most recent job on their most recent trip to the US. And legal status was determined through the documents that they reported they used on their most recent trip or the lack thereof. So I performed OLS regression which allows us to see the differences in wages between the legal status groups holding other characteristics at similar levels. So I'm considering things such as years of education, English language proficiency, the occupation sector and the year of migration. So my findings, turning to results, so as Larry mentioned my results replicate other researchers findings that unauthorized workers wages were lower than LPR's wages. And then when we look at temporary workers, I find that temporary workers wages and unauthorized workers wages were about the same. There was like a 1% difference that was not significant. And then so compared to LPR's temporary workers earned about 11% less. And I can use my regression model to predict the wages that these workers receive. So for example, I could compare the average Mexican worker across the legal status groups. And to give you an idea, the average Mexican worker has about seven years of education, has been to the US about four times before. He's proficient in English. And in this data set, he's working in a skilled job. So things like skilled construction or skilled manufacturing. So if an average worker was in LPR, he would earn about $10.34 per hour. This is in the year 2000. That's the midpoint of my data. If he was an unauthorized worker, he would earn a little more than $9 an hour. And temporary workers would earn about $9.17 per hour. So as you can see, despite temporary workers' legal authorization to work in the country, their wages really do not differ from the wage levels of unauthorized workers. And both groups earned less than LPR's. So in that wage analysis, I considered all guest workers. But within this group, H2A workers, or those in agriculture, have legal rights not given to H2B workers. Employers are required to provide H2A workers with housing at no cost to the worker. And so housing could act as a compensation benefit to H2A workers since they do not have to pay for rent. So I wanted to take this into consideration. And I asked, how do H2A workers monthly earnings and the compensation compare to the monthly earnings of unauthorized workers and LPR's? So again, I'm using the same data set, the Mexican Migration Project. I just limited it to the agricultural workers in the data set. And I considered a respondents total monthly earnings. So that was his reported hourly wage times the average number of hours worked per week times four. And then to capture the added compensation covered by H2A employers for housing, I basically added $155 to the monthly earnings of H2A workers who indicated that they did not spend anything on rent. So where is this $155 coming from? This is actually the median amount of rent spent by H2A workers whose housing cost was not covered by their employer in my data set. And just a note, so I'm comparing H2A compensation, this monthly earnings plus $155, to unauthorized workers and LPR's monthly earnings because they do not receive this housing benefit. So I find that, again, unauthorized workers' monthly earnings were much lower than LPR's monthly earnings. They earned about 30% less than LPR's. And again, temporary authorization to work in the country did not really improve H2A workers' economic outcomes. Their monthly earnings are close to those of unauthorized workers. And H2A workers' monthly earnings are 22% lower than those of LPR's. However, when I compared H2A workers' compensation, so again that monthly earnings plus the housing benefit, to unauthorized workers and LPR's monthly earnings, I found that H2A workers were compensated 16% more than unauthorized workers. So the housing covered by employers did increase H2A workers' monthly earnings so that their compensation was above that of unauthorized workers. And H2A workers' compensation is closer to the monthly earnings of LPR's than those of unauthorized workers. So we're seeing that this benefit does help H2A workers economically. So overall, I find that temporary work authorization does not provide better economic outcomes for temporary workers. Temporary workers' wages and monthly earnings are similar to those of unauthorized workers. And both groups' wages and monthly earnings fall below those of LPR's. And these findings suggest that temporary workers' dependence on their employers for their legal presence in the country results in their low wage levels. So as previous studies have shown, both unauthorized workers and temporary workers are dependent on their employer for their presence in the country and often fear deportation if they report some kind of wage theft or abuse or exploitation. And so employers can rely on this fear to offer them the lowest possible wages. The only difference in the economic outcomes between unauthorized workers and H2A workers is the housing allowance for H2As. And so without this benefit, their economic outcomes are virtually identical. And then H2B workers who made up around 37% of H2Vs in 2014, they are not entitled to this benefit. So with policymakers and employers pressuring the government to increase the number of temporary workers admitted into the US, it's imperative that we recognize that the work conditions commonly faced by temporary workers are not much better than those experienced by unauthorized workers. So thank you. Thank you for that, Lauren. And now we have Tom Hertz. Tom is an economist in the Economic Research Service in the US Department of Agriculture. And he's been in the Rural Economy Branch of the Resource and Rural Economics Division since 2009. He's also taught at American University and was a visiting scholar at the Food and Agriculture Organization in Rome. And I'll just let you get right into it. Thanks, Daniel. And thanks, DPI, for hosting. And I'd also like to make a shout out to Daniel Carroll, who is the man behind the NAWS survey, the National Agricultural Workers Survey, which is the source of every scrap of data that I will present today. So what I'm looking at is a couple of questions relating to the current policy context we're in. I'm sure you're all aware of the DAPA and DACA executive orders, which could grant administrative relief from deportation and grant work permits to about half of all currently unauthorized crop farm workers. That's the estimate. Is it roughly half of the unauthorized farm workers would be DAPA or DACA eligible? And those unauthorized workers in turn make up about half of the overall hired crop farm workforce. So it's a non-trivial number of farm workers. And the questions that I want to ask are the extent to which legalization would alter wages, hours of work, total farm and non-farm earnings of the currently unauthorized. And also say what I can, which is not as much as I'd like to be able to say, about how legalization would affect the rate of transition out of agriculture. So it's no surprise that farm worker earnings are low, have been low, numbers from fiscal 2012 for the average unauthorized crop farm worker in the NAWS survey were just around $9. And that was compared to about $9.70 if you were authorized immigrant or citizen. So there is a wage differential there. There's also an earnings gap because of the duration of employment. So average weeks for unauthorized were 39 in 2012. And that's considerably up from the decade of the 1990s when average weeks worked were only about 23. So the good news for earnings for farm workers unauthorized included has been an increase in full year employment, less old fashioned seasonality, more year round employment. So that's been a source of climbing earnings for some farm workers. But some of the disadvantages that crop farm workers face of course, 80% are immigrant and about half of all of them lack immigration status. So advocates have always understood that there's two basic approaches to addressing farm worker well-being. One is to try and raise wages and hours of work on the farm and improve working conditions. And the other is to facilitate the transition out of agriculture into more stable and better paid industries. So some of the interesting initiatives that have come up in the past worth mentioning the Agricultural Labor Relations Act in California where a huge number of hired crop farm workers are located created a very supportive legal environment for union activity in California agriculture. It was exempted from the National Labor Relations Act but California has an act modeled on it which is quite strong. It generated a lot of litigation but has not succeeded in unionizing the industry. In fact unionization declined in the long run. And according to the NAWS survey, less than 1% of farm workers are represented by a contract which seems quite low. United Farm Workers, the main union involved has been more successful via legislation. They were a driving force behind the recent increase in the minimum wage from $8 to $9 and it's going up to 10 in 2016. Maybe with the current momentum it'll go up a little further or a little faster but that much is already set in stone. And then of course in 86 we had the Immigration Reform and Control Act which authorized 2.7 million people including 1.1 million via the Special Agricultural Worker Program. So it had a disproportionate impact in agriculture because of precisely because of growers concerns about what would happen if their farm workers were legalized and left or how would they survive if the new penalties that were imposed by IRCA made them unable to employ the remaining farm workers who were still unauthorized. So there was a special push to authorize agricultural workers. It was known that not all of the workers who were authorized through this program were in fact agricultural workers. But, and it was hoped that with more secure legal status higher earnings would follow. As far as we can tell in agriculture itself IRCA seems to have had little effect on wages and employment. Yes, many people were authorized. I will argue that many of them did leave agriculture as a result. They were replaced by a continuing inflow of unauthorized workers and it sort of was business as usual in agriculture. The enforcement at the workplace turned out to be much less truthful, fearsome than growers had feared. So that, those are two major policy initiatives that didn't really alter economic conditions in agriculture all that much. So I will ask whether IRCA allowed people to get out of agriculture and I will point out that the question was debated but not resolved and I'll present some estimates to suggest that it did accelerate the transition out of agriculture which for many people was the plan. There are many people who work in agriculture for the rest of their lives and we'll tell you so, this is where I'm gonna stay but there's another chunk who are saying I'm here for a few years and I'm trying to get a job in the broader economy. So the current policy environment, we have a slowdown in unauthorized immigration due to economic growth and demographic change in Mexico as well as tougher enforcement both at the border and internally I believe. There are some signs of tightening in farm labor markets but you don't really see it in real wages yet. I mean real wages are slowly climbing for farm workers which is remarkable in itself given that real wages aren't climbing for other workers with low education and other occupations. So farm workers are doing marginally better in wage terms I think than people at similar levels of education in the U.S. economy but not that you could point to any national farm shortage, farm labor shortage or anything. Comprehensive immigration reform appears stalled so DAPA DACA is sort of the policy issue that seems most salient and one of the things that's interesting about it is it's not a comprehensive reform, it's a single intervention. It is authorization of a number of people. It's not combined with a big guest worker program, it's not combined with any particular enforcement changes or e-verify or anything like that. So in principle it's an easier policy to understand what its implications would be than a whole package of policies like comprehensive reform might be. So the methods that I use are very similar to Lauren's. These are the outcomes that I look at. I analyzed them by legal status using standard regression methods quite similar to hers. I don't have a slide listing all the control variables but there's a lot of them. One difference between the way I go about it and the way she did was that I allow the effect to differ in certain ways. So I calculate a different effect of legal status for each time period. The status effect has changed over time and I also calculate a different legal status effect for each level of experience. So once you do that, you don't have a single answer to the question what's the legal status premium. You have an answer that depends on who you are. And once you do that, the most natural way to calculate the answer is to say what if we take the people who are currently unauthorized with their relatively low levels of experience and authorize them. The average effect of doing that will be smaller than the overall average because the overall average would include the effect of authorizing a whole lot of people with higher levels of experience. So you get a smaller answer in my approach than you might in an overall average approach by allowing the effect to be different for different people and focusing on the group that you're concerned about which just happened to be the group for whom the benefits of legalization are smaller because they have less experience, okay? So that may be one reason why I come up with a much smaller green card measurement than you do for your agriculture only subset. The other reason is there's no reason why two different surface have to give the same answer as much as you'd like them to. All right, and another thing I don't do and I don't think you do either is correct for selection bias, unobservable differences by legal status. This is all the rage in the econometrics literature and I'm trying to not go there but I think I'll have to sooner or later. So for the second question where I'm looking at the exit from agriculture, I just do a sort of a spreadsheet analysis where I look at the prima facie evidence of a decline in the number of IRCA, as I call them, IRCA authorized workers, okay? So after IRCA, a bunch of people were authorized. They're found in the agricultural survey and then their numbers dwindle. Well, that could just be the normal attrition of a fixed cohort over time or it could have been accelerated by legalization itself. Were they able to get out of agriculture faster? So that's the question that I answer or try to answer by comparing their employment trends to the people who are similar to them in terms of age and year of entry to the US but remained unauthorized. And what's interesting is that there was so much fraud in the seasonal agricultural worker program, people who got authorized without ever having been farm workers or without proper documentation because it was relatively easy to do. It was designed to be easy to do because the farmers were terrified about losing their workers that there may not be all that much difference between somebody who did authorize and someone who didn't. It may not be that the ones who authorized had a huge amount more farm experience. It could be a fairly random thing. So that's my control group and I'll get to those results in a minute. This here is the decline in employment of the IRCA authorized cohort that I study in the final section of my analysis. The purple section is the rise in the unauthorized employment. And you can see that since all of these people were previously unauthorized as well, if you sort of add them together, you can see that overall unauthorized employment had the IRCA not legalized this group here would still have risen. And the main reason for that is the decline in U.S. born employment, but which then rebounded somewhat. So the unauthorized share in agriculture has been about 50% for the past decade or so. Doesn't seem to be changing all that much. These are the raw wage differentials not adjusted for anything. Green card workers making about 10% more than the unauthorized. Naturalized citizens making about 20% more than the unauthorized in the most recent time period that I focus on. When you apply the regression analysis, much of the difference in earnings between unauthorized legalized permanent residents and naturalized citizens can be explained away by their lower levels of education, their lower levels of experience. Those are the primary things right there. English language proficiency matters as well, but it's primarily experience that does the explanatory work. So you get these relatively modest wage gains of around 4% on average since in the second decade of the survey. That's for green card, which I would argue is the relevant number here because if DAPA DACA grants work authorization, that's neither a green card nor citizenship, but it's probably closer to a green card than anything else in terms of its effect on your labor market options. So this is the similar graph for the naturalized. This is about 8% on average over this interval here, a bigger effect as Larry pointed out. But both of them notice seem to be dropping in the last period. It's a little hard to draw much about the fluctuations. This is a three year average, each one of these is a three year average. There seems to be a slight decline in the wage premium for green cards and for the naturalized in the most recent three year study, but it's hard to know if that's gonna hold up or if it's just a statistical fact. So here are all the outcomes that I looked at. On top is as observed and below is adjusted. So here is the green card, roughly 10% differential again for the second decade before adjustment and after adjustment goes down to 4%. For naturalized, it starts at 23 and goes down to about eight. And then I also looked at all these other things. Days worked, total farm earnings, probability of also working off the farm and an estimate of your total farm and non-farm earnings. And many of them differ in predictable ways before adjustment for covariating factors, but most of those differences go away for the green card category. They make a little bit more per hour. They don't work significantly more days. They actually seem to work a little bit, about 3% fewer days. Their overall earnings are basically identical. They don't seem to have better off-farm opportunities or to take advantage of better off-farm opportunities. And overall earnings are pretty similar. So what this suggests is that legal status does confer an hourly wage advantage, but not much else among people who remain in agriculture. And that's an important caveat. I only see these people if they stay on the farm, otherwise they exit the survey. I can't track them as they go into other industries. And that's the real way that legalization is supposed to raise your earnings. But one would like to know what it does in agriculture. And what we do see is that many more of these outcomes are positive even after adjusting for covariates among the naturalized. The hourly wage differential is larger. Days worked. They seem to be working 22% more days, making 27% or 28% more overall. But they also work off the farm significantly less. So naturalized citizens who are working in agriculture are not transitioning to other things. They are settled in agriculture. They're not spending a whole lot of time working other jobs. They have full-year full-time jobs in agriculture. And that's what they do. So it's not as if, well, I'll leave it at that. OK, finally, I think I'm running a little short on time. Let me zip through this. This was my attrition analysis of the IRCA authorized cohort. The green area is the actual sort of smoothed number of workers that were observed. And the purple is my estimate of the additional losses that were due to legalization. That had legalization not occurred, the cohort would have declined at this rate instead of the faster rate that we observe. So after five years, which is about when the gap is the biggest, I'm saying that of the 47% that disappeared, 26% would have disappeared anyway, and 21% disappeared because they had legal status and that allowed them to move through other industries. But 10 years later, there was no difference. So in other words, it looks like what legalization did was accelerate the exit of the people who plan to exit sooner or later anyway. That's the conclusion that I draw from this. And then after about 10 years, you're left with the people who don't plan to exit. And it's actually quite remarkable statistic, which I come up in the tail end of my paper. I believe the number is 80% who say that they intend to work in agriculture for the rest of their life or as long as they are able to. So when asked, how long do you expect to continue doing farm work in the USA, 80% of unauthorized workers in the most recent few years of the survey respond over five years or as long as I am able. So these are not people who are anticipating leaving agriculture. And it's important to remember that the workforce is really a mix of people who are very committed to agriculture. It's their best earnings opportunity. It's what they do and people who are passing through on the way to better employment in other sectors. So final conclusions. The wage effect appears fairly modest of legalization. Does not appear to be an effect on hours or earnings. These estimates are lower than some other published results that are on the order of 10% or 20%. I've already explained that I think that part of the reason might be the treatment on the untreated approach that I take, which is less than its overall effect. But other people have done that and still gotten bigger numbers than I using the same data. And the difference there is probably the econometrics of selection bias. And real quickly, if the unauthorized exert greater effort, and we can't see that, and we assume that they would still exert greater effort if they were authorized, then that might explain why I'm getting a smaller number than people who are able to control in magical econometric ways for this unobservable effort. Now, it's quite plausible that the unauthorized do exert greater effort because if they don't and they get fired, the consequences of losing that job are much more dire. But then the question becomes, well, if they were authorized, would they still work as hard? Or would that security enable them to relax just a little bit, in which case it may not be appropriate to continue to control for those unobservable factors? The unobservables themselves may be endogenous in the language of my profession. So I'm going to argue that my results are plausible and that some of the other estimates may be inflated by correcting for selection bias in inappropriate ways. But that remains a topic for future research that I won't belabor. Real quickly, the exit rate calculations provide a rough guide to how farm labor supply might respond to DACA or a similar legalization. For example, if half the current workforce is currently unauthorized and half of them become authorized and that has a 21% effect on their proclivity to be in agriculture after five years, that would be half of a half of 21% would be a 5% decline in the overall farm labor supply, which does not seem like a very large number, but it's important to point out that in some crops in some regions, the number unauthorized is much higher. Certain growers, three quarters or more of their workers are unauthorized. If those people are all DACA eligible, they could face more impact on their labor supply. And finally, perhaps the most important outcome of all is the actual outcomes, whether it's the cost of the farmers or the benefit to farm workers, depend crucially on the future strength of the non-farm economy. I mean, all this is historical data and it all depends on the economic conditions. What actually happens whenever we legalize whatever we do, actually depends on the most strongly on the non-farm economies, conditions, what your other employment alternatives are. All right, thanks very much. Thanks, Tom. And finally, we have Charles Kamisaki, who is with the National Council of La Raza, which is the nation's largest civil rights and advocacy organization. He's also a fellow with the Migration Policy Institute. And he's writing a book on the Immigration Reform and Control Act, which I'm dying to read, which Tom talked about. So I have asked Charles to contextualize what these findings mean in terms of the broader debate about immigration reform and legalization. Thanks, Daniel, and thanks to EPI and Larry. There's actually, people don't know it, but there's kind of like a slant to this because of the wire, so you keep wanting to go back. Anyway, sorry. But thanks to EPI, Larry, and Ross, and Daniel for not just this invitation, but for the incredibly important work they do every day. I do need to start with a caveat. So unlike Lauren or Tom, I don't either have a PowerPoint or, for that matter, much new research or technical knowledge to share with you, but I hope I can provide both a little bit of a historical and policy context. So first, Lauren and Tom were not just really substantive but unusually clear about presenting rigorous findings from both the MMP and the NAWS databases, notwithstanding their limitations. And their findings really aren't at all surprising, right? That the fact that temporary workers are either worse off considerably or certainly no better off than unauthorized workers is something that we've known for a long time. And I would argue, in fact, they are both economically and, in other ways, far worse off, in part because temporary workers, neither temporary workers nor unauthorized workers have access to safety net and other supports available to other low-income workers inadequate as they may be, low-income tax credit, child tax credit, and so forth. And secondly, this doesn't take into account the non-economic exploitation that unauthorized workers, and in fact, we believe many temporary workers, or also face the fact that the industries in which they are concentrated tend to have the highest rates of accidents and death on the job. And this is true for at least three reasons, and let me go from the, just to be contrarian, going from this specific to the general. First, that whether you're talking about H2 temporary workers or the unauthorized, the populations themselves face considerable constraints in exercising their rights and protections under the law. It's not just, although it is true, that some may fear and may be deterred from complaining due to fear of deportation or other forms of retribution, it is also true that they lack really the adequate access to information networks and other sources of assistance to effectively assert their rights. I'll talk about the abysmally low enforcement budgets for OSHA and the Wage and Hour Division and State Wage and Hour Divisions in a second, but I was just note that the system places very heavy burdens on individuals to assert their rights. You've got to know who to contact, you probably have to have help in filing a complaint, and the nonprofit or free or limited cost legal services available to support these groups unauthorized or temporary workers is really very, very limited. So you have a individualized environment that is very difficult for temporary workers or unauthorized to assert their rights, but that is reinforced by a profoundly difficult policy structure at the nexus between labor law enforcement and immigration law enforcement, and just to give you one statistic to underscore that nationally, according to a recent report by MPI, we are spending about $18 billion a year for immigration law enforcement. That is essentially to try and address a population of about 11 million unauthorized and to prevent that population from growing, to try and enforce labor rights and standards for well over 100 million full-time workers that are fully covered by the law, perhaps eight or nine million employers, the nation spends just over $1 billion for all labor law enforcement. That's at the federal level, I think it's about 1,000 inspectors and maybe a similar number at the local level. So we got 2,000 inspectors to try and enforce the rights of more than 100 million people at eight million or nine million work sites. It is true that as Lauren has pointed out, the temporary workers cannot move among employers. The movement of unauthorized is not maybe quite as constrained, but similarly constrained because of the lack of work authorization. But what's important is even the upward mobility, even limited upward mobility within, say agricultural sectors, from labor to foreman, from foreman to crew chief or trochero in the vernacular, to eventually maybe running your own crews or running multiple crews, those are all essentially denied to temporary workers and severely constrained for unauthorized workers. And then for every worker in this sector, you also have the problem of the Fair Labor Standards Act exemption for all agricultural workers. So you have a really difficult policy infrastructure to try and enforce what limited rights farm workers already have regardless of immigration status. And finally, one may argue, one might ask, why does this exist? And I don't think I have to tell this audience a lot about it, but it really is representative of a in effect political economy underlying this policy system that is really rigged against the worker from the very beginning. We have a hundred year history in this country. Roots going back literally to World War I and the following placero programs, four repatriation campaigns, repatriation campaigns that included more than a million US citizens deported back to Mexico in the 30s and the 1950s, all based on trying to preserve a very unique system of employment that essentially treats workers as commodities in the context of agriculture. That is, when they're needed, we want as many of them as possible. And when they're not needed, because they are not by and large granted permanent immigration status, they are simply expected to return. So what was in the 1970s, it was very famous in the immigration literature. There was a so-called iron triangle that consisted of agribusiness, a friendly Congress, particularly in the grower friendly Senate, and the INS, the Immigration and Naturalization Service, which wasn't so much lax in enforcement on a consistent basis, but just worked very closely with the agricultural community with respect to when and how it exercised its discretion in enforcement. One may think that is a relic of the last century, but in fact, it is not. Literally every year, every time harvest season comes about, you will see a whole host of bipartisan groups of senators and representatives writing to ICE, or CBP, the enforcement arms of the Department of Homeland Security, calling on them, asking them, urging, demanding, threatening them that they should be exercising their discretion in not enforcing immigration laws during harvest season, so as not to endanger the underlying importance of needing to harvest the crops. So I would argue this is not just a historical relic, but it is some form of iron triangle still in existence today. There is also in fact, I think, a concern about race, which I won't go into in detail, but I don't believe it is especially accidental, that at the time when agricultural workers were largely exempted from labor laws, the itinerant labor force in agriculture was largely African American, now it is largely Latino and is not surprising at least to me, that it is one of those factors that works synergistically to ensure that in good times and in bad times, agricultural workers, regardless of status, are heavily disadvantaged in labor market. So given this big problem, what do we do about it? And I would argue, because it is complex, multi-layered with powerful economic policy and political barriers, we need to move on all three of those fronts. That is to say, on the economic side, we should be trying to use our consumer market power to push for better corporate practices as the Mockley Workers Project has done or the Equitable Food Initiative that the Farm Worker Justice is a big player in trying to force the large retail industries susceptible to consumer pressure to impose better standards on their growers. On the policy and litigation side, as so many people in the room already do, we obviously need to use all of the tools that are disposed to ensure maximum protections in temporary worker programs, deploying non-profit assistance as one strategy to enforce standards and non-profit legal assistance, I'm sorry, and obviously working towards comprehensive immigration reform that would both reduce the size of the inherently exploitable unauthorized population as well as strengthen reforms in temporary worker programs. And finally, I think as part of an effort to politically push for rising tide for all workers that hopefully will lift the boats even in agriculture, we at NCLR believe in working closely with people like EPI, I'm sorry, to try and improve the condition of all workers in this country, so thank you. Testing one, two. So thanks everyone, I think that was very interesting and we don't have a ton of time so I'll maybe ask one question to get us started, but if anyone, everyone in the crowd could start thinking up their questions and we'll call on you. One of the things that wasn't talked about is in agriculture, the H2A adverse effect wage rate. This is essentially a prevailing wage rule that sets what H2A agricultural guest workers will be paid. I'm wondering if anybody has any thoughts, but specifically maybe Tom, maybe can you say something just about how that works, how it's set and what sort of impact do you think it has in agricultural wages? I mean, maybe that's what's keeping H2A wages in Lauren's data set as high as they are. Maybe they'd be lower or should they be, you know, would they be higher or lower, what do you think? I'd be happy to take a stab at that. I should have said right from the very start, of course, that I'm speaking my own opinions. I work for the U.S. government, but I'm not speaking for the USDA or the Economic Research Service. So with that reinforced proviso, let me take a stab at your question. So the adverse effect wage rates are determined by the wage in each region from a survey in the previous year. Department of Labor publishes them and says, okay, if you're gonna hire an H2A worker, this is the wage you gotta pay. It's supposed to be the average for that region so that it sort of keeps things from undercutting that average. They are often, they're on the order of $10, $11, $12 an hour depending on the region. They may be higher than wages paid to unauthorized workers and that can be a source of unhappiness among growers with the H2A program, but there are the housing costs are actually probably the bigger cost item and the bureaucratic difficulties of working within the H2A framework are legendary. So in fact, the wage rates are probably not the real source of trouble. And one of the things that's interesting is one of the best pieces of evidence that the farm labor market is tightening right now is that despite all the complexities and difficulties of working with the H2A program, growers have doubled their interest in it over the past few years. It's gone from about 50,000 certified requests for H2A workers to about 100,000. So that's proof that wages are tightening and they're being driven into this program that they're not generally very happy with. Can you maybe say one follow up word on those employer complaints about the burdensomeness of the program because I actually think it's my own opinion is that it's a bit controversial because if you look at the labor department certification data, I mean, everything is processed timely but 99% of the applications are processed timely. We don't have the DHS data but if you look at the State Department data it's about 90, 90 something percent of the visas are actually issued that get to them from DHS. So I question whether that actually means, whether burdensome means we don't like to pay for stuff and we don't like to have to file forms and pay an immigration attorney. Is it really burdensome or is it burdensome because we have to do something? I've thought about this question a lot and my personal opinion is that the way the program is administered, it's not just the wage rate, it's not just the housing cost though those are important but the actual process is truly part of the problem. That's my personal opinion. Anybody else have any thoughts on H2A or wage rates? No, but I bet Adrienne has a rebuttal back there. She saw me out. Could you please state your name and your affiliation please? Apartment on anything but I appreciated Charles's observation about the Iron Triangle and the notion that the concerns about particularly agriculture program doesn't really lend itself well to a Republican or a Democratic platform. We see it happen both ways. Certainly in wages they are a lot higher than the first year we published wages that in agriculture there were no less than $10 an hour in order to cut them into codas they've been increasing actually a lot faster because of the oil and gas they've been increasing fairly rapidly they're at almost $14 an hour and we get lots of complaints from members of Congress and growers about the cost, the average cost to actually pay. What is, I had my question for the panel would be there's been a lot of good evidence here and I appreciate if you have put in some some very important for me. The sort of discussion political on the Hill we're talking about an A word average spec wage rate that's actually fairly livable at least $10 an hour. The Senate bill that's been passed certainly going to from what I read reduce worker wages by almost $2 per hour create a very large guest worker ag program on the tune of a half a million to the growers coming in. There is some reduction of what you would consider to be we consider to be core benefits to workers. And I'm just sort of curious about what people think might be on the horizon there because that seems to have been a compromise solution from the Senate at least that is supportive of the administration in relation to some of the research evidence presented here about the impact that could have on wages and working conditions of domestic handage to a guest worker ag program. Did you wonder are you deferred or are you okay? I don't know if anyone has any comments on that. Well, I would just say as an organization that did support with some reluctance to the Senate bill there like any bill there are numerous compromises involved in the legislative process. Correct. That's a very diplomatic answer. I think and while there is no capital on this program that is one reason why there are these various which law legal and legal is not being quite as common. Well, I can address your question about the recruitment fees. So unfortunately MMP doesn't ask any question about recruitment fees or costs to move to the U.S. I know other studies have shown that temporary workers often go into debt to get into the U.S. So unfortunately the wage rate really just covers and the monthly earnings cover hourly wages and the number of hours worked. Could I go ahead? Well, I would just add and this was mentioned in Lauren's paper and one needs to be, I know I'm treading dangerously into substance here, but some of these purported benefits come with downsides as well. So indeed housing is a potentially significant benefit but in the hands of some employers that turns into essentially a closed system of isolating the workers, often, I mean this again goes back to previous centuries where they're charged for groceries and other kinds of services or products not unlike the old sort of script systems in the copper mine. So yeah, there's no question that not having to pay rent is an economic benefit but being gouged on your food every week may diminish that benefit. Just to follow up on what Charles said, one thing I also mentioned in the paper is with the employer providing housing, it can in a sense again isolate workers and kind of increase their dependence on the employer. So that just kind of adds to the power dynamic and the relationship. It might make it that much harder for individuals trying to assert their claims. On the issue of, well, one more thing on the H2A program. So the current H2A program is unlimited and there are almost no H2A workers in California where there are 300,000 farm workers. It's easier to hire unauthorized. So it has been estimated by other people, not me, Phil Martin at UC Davis, that the Senate bill temporary worker program would be a temporary program that would suit California growers and it would become the effective labor supply for California. So there must be some difference between the way that, the difference between the way that program is structured and the way the current program is structured according to him was dispositive. It would make enough of a difference that it would go from a program that isn't used to a program that is used. So that sort of bounds the cost of the two programs a little bit, gives you an understanding of how those would shake out. And as for the effect on U.S. workers, the kind of analysis that I run looks at the current labor market in its current equilibrium where we have who we have and looks at the gaps between them to do a more sophisticated analysis of how all these wages would circulate in a new equilibrium where a group was authorized, temporary workers were brought in. That's a bigger question. So it's very important caveat to say that studying current differences between different types of workers, even with all these regression methods, isn't really predicting the future of what will happen to a complicated interconnected series of markets if a policy actually passes. So the cross-sectional results tell you what they tell you, but they're sort of a first cut at what in truth is a much more complicated economic question. This is Ross Eisenberg from EPI. If you're an employer and the current program says you have to pay the A-work and now the new program says you have to pay an A-work that is arbitrarily a dollar or two less, that alone would be reason for you to support the new program. The new program also doesn't have the housing guarantee. So I mean, those parameters, I think, are more than explanatory beyond any, you know. No, the actual upfront costs are, whatever, that's true, yeah. Do you have a question? Yeah, I keep going back and forth. So I have, I'm Mark Rosenblum from Migration Policy Institute. So I have one question for Lauren and then I'm trying to think of how to weigh in on the incentives for employers to use H2A and H2B or not. But so my question for Lauren is, if I understand your findings correctly, when you looked at that temporary workers overall, you found no way to difference between temporary workers and unauthorized workers. Then when you look just at H2As, comparing them to ag workers, you see that there is a benefit mostly in the form of the housing, the housing necessity. So the question is, did you look at non-ag temporary workers, H2Bs, and compare them to unauthorized workers and their industries just to sort of, because that was the exercise you did with the ag workers, and I'm wondering if you did that same industry-specific analysis with the non-ag workers, did you look for a benefit? You know, I actually, I did not look at just the H2Bs, and I think that would be a really interesting analysis that I could definitely add to the paper, yeah. And, you know, obviously, sort of one reason it matters, you know, to go to this conversation about, you know, whether employers would use the new, you know, the S744 guest worker program, and, you know, one way to think about this is, what are the incentives for employers to use a temporary worker program or not? And, you know, from Lauren's findings, if employers aren't paying any extra for a guest worker versus a non-authorized worker, then you would expect employers to use the guest workers, you know, because that's legal, and if there were any disincentive to using an authorized, to hiring an authorized worker, then you would prefer to hire a lot worker if it's not gonna cost you anymore. So either there are some hidden costs in there, the processing costs, or there's some wage that Lauren's not picking up, or, and this may also be true, you know, we're doing nothing to hold employers accountable for hiring non-authorized workers, but, you know, in terms of policy interventions, we can, you know, obviously, doing more effective work-side enforcement can cause employers to use a program that they don't like because it's, you know, because wages are higher or it's burdensome. In the absence of that, it's very difficult to get employers to opt into these programs, but, you know, Lauren's findings that there's no cost difference between an authorized and temporary workers, and yet employers still don't use them, you know, tells us something about, you know, what the incentives are for employers. Yeah, I'll just say, I mean, you guys can answer, but I'll just say, if you look at the H2B occupations, you can go back, you can go back 10, 12 years, wages have been completely flatter, even declining in the top, you know, 20 occupations or so, and the unemployment rates are sky high, I mean, double digits in almost every occupation, so it's not like there's some big national labor shortage, maybe there's some localized ones, but there isn't a big national one, but they are liking to use the H2B program, so that's sort of the other, you know, another wrinkle in it, but. I like to use the H2B program, which, as Lauren's data suggests, has, you know, it's less expensive, you know, compared to authorized workers, and they don't take advantage of the H2A program, which is uncapped, you know, it's more expensive and or more burdensome. Any thoughts? And the H2B program is capped. Yeah, right, right, right. And you have also, you know, in equilibrium, those wages could be the same, but that doesn't mean that if you removed one group or the other, they could just fill that hole with the other group, right? I mean, it may be that there's no wage difference between them, but it doesn't prove that they, that they feel the same about both groups of workers, if you see what I'm saying. I think that in terms of wage gaps, generally looking at racial disparities, and particularly looking at the history of farm work going all the way back, right, is a really important thing for us to take a look at. But also, just a question about state laws, and when we look at policies that may target enforcement on individuals, there are certain states that have certainly moved up, or put more pressure on workers because of the inability of the federal government, you know, the solution that ability to move forward on comprehensive regression reform. I wonder if there are any studies that have looked at sort of a regional, or state-by-state, state laws that may enforce immigration more heavily than other regions or areas in that state. What does that data tell us about policies at a state level that are pushing wages even lower? That would be Lauren's answer. I actually don't know of any studies that have done this yet, but I actually am moving that way for my dissertation. So I'm glad that you're interested in that topic and check back in the year or so, hopefully. Well, couldn't you look at the Mexican Migration Project data and pick out the states where there have been enforcement actions, same with the NAWS? I think with the MMP data, there are not quite, there's not quite enough state-by-state variation for me to do that, but I'm hoping to use ACS or CPS and impute legal status and then look at cross-state differences. Please. I'm just doing it from the foreign capital. I have a question that you sort of touched on, I'm just wondering, this business of sealing the border, which has not been an issue earlier on, where you had migrants coming across and they were working in the back and it wasn't so difficult or so intimidating to go back and forth. And I'm wondering if there is an effect on wages based on the intimidation of sealing the border or what effect does sealing the border have on wages? I don't know, I think the main effect of sealing the border has been to make the unauthorized population a longer-term population than it's been. I think 10 years ago it was about a third, we're here longer than 10 years now, it's more like two thirds, but I don't know if anybody have any thoughts on that. Any other questions? Well, I'll ask, have any of you thought about, I mean, I just one thought on that, in the last decade or so, maybe slightly less. And if one wants to go back to the IRA in 1996, there has been a discernible change and I think the studies you're referring to are from Doug Massey and many of his colleagues. One of the things that's going to be interesting, at least for me, to think about, and I certainly don't know of any studies here, is it has changed something very significantly, which is for agriculture, this assumption that you had an inexhaustible supply of unauthorized as a sort of fallback, that may well be changing. And that may well in fact represent a greater willingness to go through the bureaucratic hassle of utilizing temporary worker programs, but that's just a mild guess. Border security or enforcement in general was cited by I think Pew Hispanic Trust and they're trying to understand why net migration from Mexico had slowed down or maybe even reversed. I think it has reduced inflows. So if it has reduced inflows, it will tend to raise the wages of the people who the inflowers are competing with, which is generally the previous generation of inflow, the previous year's inflows, who are the most similar worker characteristics. So a reduction in migration from Mexico and South Central America should tighten labor markets and raise wages for the people who are most directly in competition with those workers who are no longer coming in. I'll ask one. I have one more question. When you were, Gordon was first presenting the data, two-thirds of the people were documented, one-third were actually temporary, registered temporary. And I'm wondering why that is and is it because of the difficulty of the employer to fill the paperwork and the limits to how many people can sign up and what effect that two-thirds, one-third roughly has on the wages? I mean, I see that there's a difference of the waging, I understand that, but the fact that it's so overwhelmingly the undocumented people who are getting the work. So as to the differences and who would apply for temporary worker versus come as an unauthorized worker, I don't know of any studies that have really examined as to what kind of path is chosen. When people have interviewed temporary workers, they say that they try to obtain a temporary visa because it offers kind of some sort of stability to them. And then in my data, you can also see that unauthorized workers have a lot more social capital in terms of connections to people who have come to the US before. So they have kind of a familiarity with the US that maybe temporary workers don't always have. So I can see that from my data. And then can you repeat the second part of your question? It was how does having. I'm curious how that skews the information that you're collecting because so overwhelmingly reflecting people who are not filling in the paperwork that are coming anyway. So in terms of just, so in terms of the number in my data. Okay, I see what you're saying. So although my temporary worker population is kind of, is smaller, you know, the difference is I can, there's no reason to suggest that this temporary worker population is maybe not a representative sample of the average temporary worker population because. So fewer are going to California because California doesn't, the employers in California aren't using temporary workers, but employers in other regions of the countries are. So you're seeing, you're getting an average overall average comparison across regions. I do control for border state. If you're in a border state and compare, you know, try to make those comparisons for the few that I do have in border state regions. So that makes sense. Okay, so we've reached in our times. I'm just gonna ask one, everybody on the panel, if they can give one quick policy recommendation to raise wages among these populations and you don't have to answer if you can't. But I'd also like, if you have any thoughts on the use of shortage occupation lists, this is something that other countries do, especially for labor migration, where you study to see, you look at wages, turnover rates to see which occupations in the country are actually having a labor shortage. And if there's a labor shortage, then you would give employers streamlined access to say an H2B worker, and then I think it would be smart to put them on a sort of provisional path to legal permanent residents if they wanted. I think that's, from the EPI perspective, sort of the smarter way to go. And I mean, the UK does this, Germany does other countries do this, but get a lot of pushback from the business community that the government isn't as smart as the governments of the UK and Germany and Canada. So if you have any thoughts on that, I'd love to hear them or any other policy recommendations to raise wages for the unauthorized and guest workers, especially considering the fact that there probably isn't political support for legalization between now and 2024 or so. You can go first and... Just work our way down? Yeah. So to me, and maybe because I'm more of an academic than in policy, but to me, one of the roots of the problem with the Tipperary Worker Program is that the workers are tied to their employer. I think that the dependent on their employer really limits them. So I would try to reform the program, at least in a way to where temporary workers have more mobility while they are here in the US. Along with increased regulation and increased oversight, just to make sure that they are receiving the rights that they're entitled to. Thanks. Because I work for the Department of Agriculture for the US government, of course, I'm not allowed to comment on policy recommendations, but I can say that in the absence of any policy change, there seems to be a consensus that the status quo is one in which in the future, farm labor markets will tighten, they already are tightening. The trends, the demographic trends in Mexico are such that it's probably, there's a thesis out there, Ed Taylor and his collaborators and Davis are arguing at the error of labor abundance is maybe over and employers have already responded the way they always do, which is they've mechanized to the extent possible, they've pushed the frontiers of mechanization and of mechanical assists, which are not full mechanization, but which can increase productivity and allow older workers to do the jobs that used to be more physically strenuous. So there's all kinds of tinkering around the edges of the way the agricultural production process works to save on labor and it's already underway and it's likely to continue. Is the IRS ever played around with shortage occupation unless, separate question, or shortage analysis? I don't know, that's a good question. I do know that in the very early days- You're talking about farm labor markets tightening, but it seems like it would be a place that's ripe to- Well, I've tried, I've written a paper where I looked at certain counties in certain industries where there seemed to be both a fall in employment and a spike in wages, which is what you would expect if it was the labor supply problem. And you can spot, you know, farm labor contractors in this particular county saw their wages go up by a lot, maybe something's going on there. So there's some evidence, very crude evidence, for localized shortages here and there. There's a lot of anecdotal evidence for it. In terms of trying to quantify it and build a, we don't have anything in place. They used to do that in the earliest days of the farm labor survey, the NAS farm labor survey, which used to be called something else, the Q something or other. It goes back to the 30s. They would publish an estimate of growers' needs for labor and available supply. How they did it, I don't know, but maybe it's time to dust off some of those older methods. Family Charles, the last word? Well, as I concluded my presentation, I don't believe issues like this are particularly amenable to single solutions. So I'll just repeat, I do believe we need comprehensive immigration reform and I think you would see the results not dissimilar from the previous experience from ERCA, which is some people leaving agriculture on those who remain earning some wage premium. I do like the notion of portability within temporary worker programs. As Lauren just suggested, the problem is once you start introducing those kinds of measures in the temporary worker programs, you lose the business support. I am not convinced that that's actually any more doable or a shortage occupation list or a Ray Marshall-like immigration commission. All of those things, I am not at all convinced or any easier to pass and to enforce and to sustain than Conference of Immigration Reform. Okay, good for thought. Thank you everyone for coming. Really appreciate it.