Panel Transcript: Welfare Use by Immigrant Households with Children

MARK KRIKORIAN: Good morning. My name is Mark Krikorian. I’m executive director for the Center for Immigration Studies, a think tank here in town that examines and critiques the impact of immigration in the United States.

The issue of immigrant use of taxpayer funded services for the poor has been part of immigration policy since colonial times where the colonies established what we later came to call public charge policies – in other words, prohibiting the immigration of people who were likely to become a public charge and would end up on what we would now call welfare.

Obviously that issue became a whole lot more pertinent as the state grew and as government services for the poor expanded into what we would now call the welfare state with a wide variety of cash and non-cash type of services. And then obviously now it’s even more pertinent given the fiscal austerity that all levels of government are now facing – probably most specifically at the state and local level which is where much of the welfare expenditures come from.

And unfortunately I think much of the debate on the intersection of immigration and welfare misses the point. Often the – while for instance in the 1996 welfare reform debate much of it was animated by the idea that you could continue to let in large numbers of low-skilled people into a modern society but somehow wall them off from the welfare state.

And on the flip side immigration expansionists often claim that you can admit huge numbers of low-skilled people into a modern society without any cost to society or to taxpayers as a whole.

And so the report we’re releasing today doesn’t really specifically address those kinds of misconceptions necessarily but it does try to offer a statistical overview of what the situation is, not with immigration and welfare overall but specifically with the welfare that is used by immigrant families with children at home because in a sense that’s what our welfare program is designed to do is to help subsidize the working poor with children. And that’s almost a synonym really for the typical immigrant according to our current policies.

So Steve Camarota, our director of research, is the author of the report and will be presenting the highlights of the findings. The folder out there has the report itself in it as well as contact information.

And then we’ll have two respondents followed by some Q&A. Mickey Kaus, to my left, is author of, one of the first political blogs, covered the welfare reform debate in the ‘90s quite extensive, is author of “The End of Equality” which deals with a variety of issues including the issue of welfare and welfare reform. And actually I didn’t know this, he’s a graduate of Harvard Law though apparently emerged with his common sense intact.

The other respondent at the far end of the table is Iain Murray. He’s vice president for strategy at the Competitive Enterprise Institute, writes mainly on environmental and similar issues, author of “The Really Inconvenient Truth,” a book from I guess a couple of years ago.




MR. KRIKORIAK: Oh, okay.


MR. MURRAY: No, 2008 – (inaudible, off mic).


MR. KRIKORIAN: And he likewise emerged from Oxford with his common sense intact. So each of them will sort of give their reactions and thoughts about both the report and maybe the issue in general and then we’ll take Q&A. So Steve?


STEVEN CAMAROTA: Well, thank you, Mark. Everybody can hear me all right? Maybe I should raise this.


MR. KRIKORIAN: I think everybody can hear you fine.


MR. CAMAROTA: Okay, there. Well, as Mark indicated, the report that we’re releasing today deals with the issue of immigrants and immigration and welfare use. Now, the first question I guess you might want to ask is why study welfare use among the foreign-born or among immigrants. Well, I think that there are two main reasons but there are others and we can discuss them later if you’d like.

The first is it’s one measure – though it’s not the only by any stretch of the imagination. But it is one measure of its impact on American society. If immigrants are accessing these programs at very high rates, then that might be an indication that there is a large fiscal cost associated with current immigration policy.

But the second reason also – there’s a second reason and that is that it’s also a way of measuring how immigrants are doing in American society. If a large fraction are using these social services, then that might be an indication that they’re having trouble in American society – their income is lagging, that they’re turning to the state to support themselves.

On the other hand, if their use rates are very low, then that might be an indication that they’re doing pretty well. Again, it’s not the only measure of these things but it would be an important measure I think. So use of welfare programs can be seen as both a way of looking at an impact on our society as well as what might be generally referred to as immigrant adaptation or assimilation into American society.

Now, we focus here in this study on households with children – that is, households that have at least one person under the age of 18. And why do we do that? Well, because as Mark indicated in his opening remarks, our welfare system is very much geared to that population. Welfare is primarily – though by no means exclusively – designed specifically to help households and families with children. So that’s the target population. So it makes sense to look at that population and examine their use.

And especially since ‘96 when we had extensive reform of some of our welfare programs, we have tried as a society to tailor our efforts particularly towards households with children that have workers in them. So again, households with children have been an important focus of the welfare system and that’s especially true since ‘96.

Now, to look at this question, we are fortunate that there is some data. We don’t just have to guess or try to piece together administrative data. There’s a survey done by the government called the “Current Population Survey.” It’s been around a very long time. In March of each year they oversample minorities and also ask a whole series of questions about use of social services.

And since 1994, they’ve actually asked questions about whether you are foreign born or not. A person is foreign born if they’re not a U.S. citizen at birth. That’s the Census Bureau definition of the foreign born. And in this report we use the term foreign born but we also use it interchangeably with the word immigrant. So when I say immigrant, generally – unless otherwise indicated – we’re talking about people who are not U.S. citizens at birth. That’s the foreign born.

It would include in the survey – in that Current Population Survey it would include people who came to the United States on green cards and then eventually naturalized – that is, became citizens. It would include people who are legal immigrants but on green cards, some modest number of temporary visitors as well as a large number of illegal immigrants are also include in this data.

Most research suggests that about 90 percent of illegal immigrants respond to government surveys like the Current Population Survey. So – and I cite that research. You’re free to dispute that. You might think the population is actually larger or smaller. But we think that most of the illegal immigrants are in the Current Population Survey.

Now, in this report, I examine welfare by immigrant and native households – that is, based on whether the household head is an immigrant or native-born. Later in the report I also try to break it out among the immigrants whether the household head is in the country legally or not. So we look at legal immigrant households and illegal immigrant households separately.

Why look at things by household? Well, it’s a very common practice. The Census Bureau recently put out some new numbers looking at data from the American Community Survey, a much larger survey but doesn’t ask all the detailed questions.

So they have some data on cash assistance, for example, they’ve put out. And they looked at it by household. National Academy of Sciences has done research looking at it by households. Most scholarly journals have looked at immigrant use of welfare by household. So it’s a very common practice. It’s the usual way you do it and there are several reasons for that.

One is that the data is collected by household. In other words, one person in the household fills out the survey. So that’s the way all Census Bureau data is basically collected. So it’s sort of a survey by proxy but the point is the unit that they use to collect the data is the household. So that’s one reason to do it. Obviously another reason when thinking about welfare is benefits flow to everyone in the household often.

So if you buy something with food stamps, everyone in the household can potentially eat that food. If you live in public housing or rent subsidized housing, everyone in the household would benefit from that. So there’s kind of practical reasons why to look at things by household.

In effect what we’re doing here is – and throughout the report – is comparing immigrants and their children because that’s who mostly lives in immigrant households overwhelmingly to native-born people and their children and that’s who mostly lives in native-born households.

Now, one final note about data quality – although the Current Population Survey is considered one of our bets sources of information on the immigrant population and it’s one of our primary sources on welfare use, on welfare use there is a well-documented understatement of program use. So all the numbers I’ll give you are just self-reporting.

They’re not adjusted in any way. If 10 percent of households say they receive cash assistance, that’s just what I report here. But recognize that the actual number is larger. People, particularly on the cash programs, seem to understate significantly their use of services. So it’s a caveat to keep in mind.

But the assumption here we compare immigrants to native is that the two don’t understate their use of such services in significantly different ways. That in general, the immigrants understate it and so do the native-born. There’s a little bit of research to back that up and we’re going to proceed from that assumption. But always remember the actual rates are higher than reported here.

So what did we find? Well, overall we found that 50 percent – 57 percent of households headed by an immigrate – legal and illegal – if there’s a child present, used at least one major welfare program compared to 39 percent of households with children headed by a native-born person.

So one way to sort of – the overall picture, looking at using at least one program is that both immigrant and native households have high use rates. That is, native and immigrant households with children. But that the immigrant household rate is significantly higher.

Now, it’s important to understand – and I have a lot of statistics in the report and we could talk about that later – but immigrant household use of welfare tends to be much higher than native for the food assistance programs – food stamps, WIC and free school lunch – and less for – and also for Medicaid. For the cash assistance programs and for the housing assistance programs, immigrant use seems to be very similar overall to native-born use – not huge differences there.

Now, again, it’s important to understand that a large share of welfare use by immigrant households with children is received on behalf of their U.S.-born children who are U.S. citizens. But even households with children comprised entirely of immigrants – that is, no U.S.-born children – still had an overall welfare use rate of 56 percent based on the 2010 data.

Thus, the presence of citizen children does not explain really the results that we’re reporting here. I have a more detailed discussion of that issue in the report, how it affects the data.

Now, immigrant households with children, their use of welfare has been higher than the native-born for quite some time. The survey was redesigned in 2002 and the survey asks about welfare in the previous year. So I should have said at the beginning, the 2010 data looks at welfare use in 2009.

If we go all the way back to when the survey was last redesigned in 2002 and we look at the 2001 results from the 2002 survey, it shows that 50 percent of all immigrant households with children used at least one major welfare program compared to 32 percent for native-born households.

So what that tells us is that there’s a general upward trend in welfare use for both immigrants and natives but the large differences when we look overall have persisted for quite some time. It’s not the result of the current recession.

Now, other important findings or interesting findings in the report are that among the states where immigrant welfare use tends to be the highest overall – again, looking at households with children – are Arizona, Texas, California, New York State, Pennsylvania, Minnesota, Oregon and Colorado. I think, though, that in the report in some detail I discuss or have statistics for you as well the differences between certain localities and states with others.

And I think that might have some important political implications that we could talk about. But let me just touch on one. In a place like New York City, welfare use is extremely high among both the native-born population and the immigrant population. It’s 58 percent for the native-born. It’s like 68 percent for the foreign-born or immigrant population – so very high use of social services overall. But it’s not that different. The immigrants come in and access these programs when they have children at very high rates.

But it’s not that different than the native-born. Contrast that to a place like Maricopa County, Arizona, which is the main big county around Phoenix. Much of the population of Arizona lives there. There the immigrant use of welfare is pretty high and it’s twice the rate of the native-born population.

And that may help explain why the public might be much more dissatisfied with immigration in Arizona. It’s not that welfare use is so much higher among immigrants in Arizona than New York City, but rather the differences with the native-born population are quite striking. And that may account for why Arizona is one of the places in the country that’s most dissatisfied with immigration.

We also found that households with children – immigrant and native – sorry, immigrant – vary quite a lot by their use rate by country. Those that tend to have the highest use rates are, say, the Dominican Republic, Mexico and Guatemala, Ecuador. Those that tend to have the lowest use rates that we could identify are the United Kingdom, India, Canada and Korea.

Now, when we turn to try to distinguish legal from illegal immigrants and we do this based on the characteristics of individuals in the household, there’s kind of a whole cottage industry of trying to distinguish legal and illegal immigrants in the Current Population Survey – also in the American Community Survey as well and in the decennial census, or at least there used to be. The current census will not allow that kind of analysis anymore.

But we find, or we estimate, that we look at just legal households with children – households headed by a legal immigrant – we find that 52 percent used at least one major welfare program. When we focused on illegal immigrant households, it was 71 percent. So 52 for legal, 71 percent for illegal immigrant households.

But there’s a couple of important points that illegal immigrants generally receive benefits on behalf of their U.S.-born children and it’s also important to note that illegal immigrant households primarily almost exclusively use the food assistance programs and have someone on Medicaid in the household. Their use of the cash assistance programs and the housing assistance is extremely low – virtually zero – whereas among the legal immigrant population, they kind of use these programs at relatively high rates across the board.

So if we look at the share of households using at least one program, we get a very – the illegals seem to be quite high. But it’s very important to note that it’s several specific programs – again the food assistance – food stamps and a lot of WIC and a lot of free school lunch.

Now, welfare use by immigrant households is a complex question. I break it down a lot of different ways in the report. But one of the important findings – though there are others and this one isn’t so surprising – is that it is the relatively high rate is partly explained by the relatively low educational attainment of immigrants in general. Looking at all immigrant households, if we look at those headed by someone who has not graduated high school, we find that 80 percent accessed at least one of the major welfare programs.

If we looked at immigrants who had at least a college degree, it’s 25 percent – again, focusing on households with children, so very large differences by educational attainment. Now, it does not explain all the differences. In fact, a decent amount of it, it doesn’t explain. But nonetheless, it’s important to understand that welfare use rates among immigrants vary very much by educational attainment.

An unwillingness to work is not the reason immigrant welfare use tend to be relatively high. The vast majority – 95 percent of immigrant households with children had at least one worker in 2009. But the relatively large share with low levels of educational attainment means that more than half of these working immigrant households with children still access the welfare system.

So that’s a very important point. People sometimes think that the welfare system is only for people who don’t work, who sit on their couch. But that’s not the way it’s designed. There are a whole series, particularly of non-cash programs and even some cash programs, that go to people who also work. You can work full-time but still live in public housing. You can work full-time but your kids can still qualify for free school lunch.

If you have a new child or are expecting, you can still get WIC. If you work full-time in many states, you can still sign yourself or your children up for Medicaid. That’s how our system works. Work and welfare don’t necessarily conflict. They often go together. And that’s especially true since the ‘96 reforms but it’s always been the case.

Now, although we focus on households – also, I should mention one other point. We don’t find that the inclusion of refugees makes that much difference in the data. We try to take out the primary refugee-sending countries and recalculate but they’re not a big enough share and their use rates are not dramatically different.

So taking out the refugees doesn’t seem to make much difference. I should also point out that if we look at all immigrant and all native households – those with children and those without children – put them together – we get a total of 37 percent using at least one major welfare program for the immigrant households and 22 percent for the native-born households.

So as you can see, the households with children use higher rates but when you look at all households the gap between immigrants and natives doesn’t really narrow in a meaningful way. So what does this tell us?

Well, it’s important to note that although most new legal immigrants are barred from using some welfare programs for the first five years and illegal immigrants aren’t supposed to use the programs at all, these provisions have only a modest impact on use rates because most immigrants have been in the United States for longer than five years.

The ban only applies to some programs. Some states provide welfare to immigrants on their own. By becoming a citizen, immigrants become eligible for all welfare programs and perhaps most important, the U.S.-born children, including those born to illegal immigrants, are all automatically granted American citizenship and are therefore eligible for all programs right at birth.

So what does this mean sort of for policy when we think about it? Well, the first thing to understand is of course hat most immigrants come to the United States to work. In fact, immigrant households with children are somewhat more likely to have a worker in them than our native-born households with children.

However, the relatively low education level of many immigrants means that a large majority of working – a large share, I should say, of working immigrant households still access the welfare system. In other words, they have low levels of educational attainment and as a consequence they tend to earn significantly low wages on average and that’s what drives most of the results.

Just to give you on statistics – about a third of immigrant households are headed by someone who has not graduated high school. For the native-born population, it’s about a tenth – so somewhat over three times as likely to be headed by someone with the least level of education.

But what does this mean? Well, it means one thing I think that when someone like President Bush says – look, when he thinks about the immigration issues, what he’s thinking about is matching a willing worker with a willing employer. What these statistics suggest is that that’s not all there is to think about. You still might favor his policy prescription and they’re very similar to President Obama’s policy prescriptions.

There’s not much meaningful difference between what President Bush wants to do and – wanted to do – and what President Obama would like to do. There’s a lot of agreement there. But what these data suggest is it’s not just between the worker and the employer, that if you allow in lots of unskilled workers, it can create very significant costs for taxpayers, that individuals with relatively little education in the modern American economy are going to have relatively modest wages and thus qualify for social programs.

So the discussion of what to do about this problem, assuming it is seen as a problem, should be conducted with a recognition of its complexity. On the one hand, it’s not enough to say, well, we bar newly arrived immigrants. That does not reflect how the system actually works. I think it’s also a bad idea to point out immigrants came in the past and didn’t need welfare. I think that’s not very helpful either because these programs largely didn’t exist.

So I think it’s a mistake – and I think there’s a lot of people who do this who think about immigrant welfare use as a kind of moral defect. And I think that is a terrible mistake. It does not capture what’s happening. What’s happening is most people come to work. Most people work. But because of their education levels, a very large fraction use the welfare system specifically to support their children.

When thinking about this issue, therefore, it makes more sense to acknowledge it seems to me that spending on welfare is part of every advanced industrial society including our own. Moreover, we have to recognize the limited wages that less educated workers will earn.

Our immigration system needs to reflect this reality and if we’re concerned about welfare use, what these numbers tell us is we need to select immigrants who aren’t likely to need these programs. One of the best way to do that is to select immigrants much more based on their educational levels. Trying to bar the immigrants after they’ve arrived is not likely to work.

It’s also of questionable fairness because it says specifically to legal immigrants, you may come but don’t expect to be treated like one of us. And I think that’s a troubling message as well. And the experience with the ‘96 welfare reform law suggests specifically for the non-cash programs that trying to bar people after they have arrived is not likely to be that effective for alt he reasons that I’ve touched on.

The bar is only temporary. You can become a citizen and avoid it, the presence of citizen children. It seems to me we need to focus if we want – if we’re concerned about these issues on the question of selecting immigrants who are unlikely to need the programs. Thank you.


MR. KRIKORIAN: Thank you, Steve. Mickey?


MICKEY KRAUS: Thank you. I’ll just make a couple of points and try to be brief. There’s a lot of interesting stuff in this survey and it seems honest – the basic conclusion seems unassailable, that immigrants cost – both legal and illegal -illegal and legal – cost a lot more in welfare costs or in the cost of these programs than native-born Americans, if you care about the cost – and we do especially these days – you have to worry about who you’re going to let in.

That said, from the welfare reform debate in ‘96, the problem – what people were so troubled about, about welfare was not the cost so much. It was that welfare seemed to enable a culture of – I don’t know however you want to call it – ghetto poverty, underclass culture, a culture of non-work and single parent families that did not seem – the cost went well beyond the budgetary cost of the programs.

And the startling thing about Steve’s data to me is that – is the low use of the key welfare program that we were debating about in ‘96 which was then called AFDC and is now called TANF. The data seems pretty clear that immigrants, both legal and illegal, don’t use the specific TANF program or AFDC as much as the native-born. They use a lot of other very expensive programs but not that one. It’s not – one question I had is why this is.

There are three possible reasons suggested in here. One is in households headed by workers, they just might not qualify for welfare in some states because there’d be, you know, a requirement that if you work so much you can’t get it. If you’re illegal you might be scared to interact with the bureaucracy because you might be found out. And I guess those would be the two main reasons.

And it’s not – I couldn’t figure out which was more important from the report. Also you mixed in under the rubric “cash welfare” TANF and SSI, as I understand it. And I always thought SSI wasn’t – at least the core of SSI wasn’t really a welfare program at all because it’s a work test. In other words, you only get it if you’re unable to work. Now, I know that there’s a huge exception with disability checks for children. That’s been allegedly abused, so-called crazy checks – that may make SSI more like a welfare program.

I’d be very interested in how that breaks out. In other words, of the cash welfare here, how much is SSI and how much of that is, you know, the SSI that nobody argues with – with these checks going to 80-year-old – you know, people who can’t be expected to work and how much is, you know, the kind of disability part.

But anyway, it’s hard to see that given this pattern of welfare use you’re going to get with 95 percent of the families have somebody working in them, you’re going to get the welfare culture that we worried about in the late ‘90s – in fact, the entire second half of the 20th century and that we’re still worrying about. It’s really just an expense thing, which is a perfectly legitimate reason to worry. But it’s not as big a worry as the traditional worries about, quote, “welfare,” unquote.

The second – my second point is welfare is on both sides. You know, I made the horrible mistake of running for senate in California in last year. And I had to articulate why – I was basically running on an anti-comprehensive immigration reform platform. And when you articulate why I am concerned with unconstrained immigration, it’s not the welfare expense so much as it is driving down wages for the working poor and changing our national culture for the worse.

And I did find that there is a sort of borderless mentality in the people who were arguing against me. People did not understand. They said, I’m in this country. Why don’t you – why don’t you want to give me full dignity? Are you a racist?

That’s basically the level at which the argument takes place in California. And if you do believe that the border doesn’t make sense and you sort of don’t recognize the legitimacy of the border, it does – it also makes sense that there would be no reason why people would want to keep you out of the country. You must be a racist. So there’s that mentality that one is also worrying about, what effect that mentality will have if it takes hold.

But because immigration drives down the wages for the working poor, it increases welfare use among the native-born as well as – as well as the use among the immigrant population, which because it’s unskilled tends to have a lot of working poor people in it. So in a sense, immigration drives both sides of your ledger. It drives – there would be fewer cases among the native-born if there were lower immigration. So I just wanted to make that point.

And the final point is I guess it’s very important that you made the point about the moral – it not being a moral defect because that is – if you make that argument it feeds into this – the borderless antiracist mentality that I talked about, the saying, what – you’re saying we’re bad people, are you a – are you – and I tried with not too much success to argue that no, California lives off the labor of immigrants – legal and illegal.

It’s just that at some point there has to be an end to the process if we’re also going to protect the wages of our native-born people. And the way the immigration advocates outline it, I see no end in sight. You know, what happens to the next wave of immigration and the next wave. So it’s very important that we not fall into the trap of saying these are bad people we’re letting in and that’s why we don’t want to let them in. But if we do care about expense, you’ve clearly documented that the expense is huge.


MR. KRIKORIAN: Thank you, Mickey. Iain?


MR. MURRAY: Thanks, Mark, for asking me to respond to this new study on behalf of the Competitive Enterprise Institute. CEI has been a principle defender of immigration for over 25 years on the basis that immigrants generally provide a net benefit to the economy and society as a whole. So we tend to play the role of Banquo at the feast at events like this.

On the other hand, we generally oppose welfare programs as a burden on society. That makes this study especially interesting from the libertarian point of view. Now, you might now have noticed that I am myself an immigrant – and a legal immigrant, I should stress. I claim no special knowledge of the welfare system from that angle, I am glad to say. But it might also be of interest that my wife’s family has been in the country since 1638.

I would like to start by discussing what the study does show. I’ll then go on to discuss what it doesn’t show and finally discuss what lessons we can draw and what further questions we need to ask. And the first thing I think we should note is that the study strongly suggests that anchor babies are not the problem many claim them to be, at least as far as welfare goes.

There’s been an effort in some quarters to change the basic constitutional right of anyone born in the United States to citizenship on the grounds that such anchor babies entitle their parents to more government largesse. And as Steven says in the report, the presence of U.S.-born children does not explain the high overall welfare use of immigrants – of immigrant households. So I think that particular argument against restricting citizenship falls as a result of this report.

I’m also glad that the report points out that welfare benefits are fungible, or as the report says, if the government provides food or health insurance for children, then their parents will not have to spend money on these things. That is a crucial point in the discussion of welfare. It creates a poverty trap where households cannot afford to earn more money for fear of losing government-provided benefits. That’s quite substantial in some cases.

As one observer recently noted, welfare has created, and I quote, “a cycle of generational poverty, government dependency and economic disparity.” That observer was none other than Marion Barry, the famously radical former mayor of Washington, D.C. As a general rule, we find that more welfare means more poverty.

Now, Mark’s comments that we are not talking about the worst contributing programs are apposite. But the high level of general welfare uptake identified in this report is important. In fact, that high uptake of welfare by immigrants makes me very worried. For immigrants now to be caught in the same welfare trap as other sectors of the nation have been is a vast betrayal.

Immigrants generally come to the United States not for the benefits, like they do in my native Europe, but for the opportunity our free economy provides. That’s demonstrated by the fall-off in unauthorized immigration since the Republicans tanked the economy and the Democrats started kicking it while it was down.

We’re down a million unauthorized immigrants a year since 2007. If immigrants were coming for the benefits, that number would probably not have changed. Now, other studies have shown that immigrants are generally more entrepreneurial than the native-born population, participate more it the labor force, save more, apply more efforts during working hours and are averse to crime and unemployment.

Welfare dependency attacks all these virtues as well as others such as family cohesion. It is in the interest of immigrants as well as natives to get these welfare participation numbers down. It is also interesting, as Steve pointed out, that when the report gets into the weeds and compares like to like, it shows little difference in welfare use between immigrants and the native-born population.

For example, in the less educated households, 80 percent of immigrants use welfare while 76 percent of native households do. This difference is unlikely to be statistically significant. And this demonstrates that the study should have gone farther in its analysis.

The late great Julian Simon, one of the inspirations of CEI, looked at precisely this question – the immigrant use of welfare – in his 1989 work, “The Economic Consequences of Immigration into the United States,” and followed up on these questions until his death in 1998.

He did detailed regression analysis that found comparing like with like, similar education levels, similar income levels, similar age cohorts and so on, immigrant use of welfare was equivalent to or less than native use. So I’d like to have seen a lot more sectoral analysis in this study. I’d like to see the figures for income level just as we saw for education level.

Do poor immigrants use more or less welfare than poor natives? The study doesn’t tell us. Do young immigrant households use more or less welfare than young native households? The study doesn’t tell us. That’s unfortunate.

Moreover, the study doesn’t tell us how much the average immigrant household claims in welfare. That’s an extremely important data point. An immigrant household may use more programs than the equivalent native household but it may still claim less overall in terms of cold, hard cash. Previous data has suggested that immigrants actually claim less and I’m afraid the study doesn’t tell us anything to update our views on that. And I’d like to have seen that data.

The report also covers only a few American welfare programs and omits the two largest – Social Security and Medicare. Medicare and Social Security are together more than four times the size of Medicaid for financial year 2012. All the other programs analyzed by the report are pretty cheap. Please note that this is not an argument for maintaining them. In many ways, these programs are that tail that wags the dog.

All the analyses I’ve seen show the government outlay per average person is much higher for natives than immigrants. Again, I’d to see the updated data. Then when it comes to Social Security, for example, we find that immigration is a net but modest fiscal gain for the Social Security system. The Social Security trustees estimate that increasing net immigration by 300,000 a year would eliminate one-tenth of Social Security’s 75-year deficit.

Undocumented immigration apparently is especially beneficial because they are much less likely to collect these benefits. In 2007 it is estimated they improved Social Security’s cash flow by $12 billion. The Social Security Protection Act of 204 states that non-citizens issued a Social Security card after 2003 can receive Social Security benefits only if they were legally authorized to work.

Now, we all know that Social Security is in crisis because it is at heart an unfunded Ponzi scheme. Restricting immigration would cause it to collapse all the sooner which further suggests that urgent reform of Social Security is needed, not of immigration. I think we also need to address some counterevidence.

The Rand Corporation’s detailed 2006 investigation into the health care use of immigrants in Los Angeles County found some important things. Legal immigrants consume less health care than natives and undocumented immigrants consume even less than legal immigrants. This shouldn’t be surprising. The American welfare state is – as Steve suggests – primarily designed to help the elderly, sick and women. Immigrants are mainly young, healthy and men.

That’s why a study looking at welfare use in households with children can’t make a particularly meaningful statement about immigration in general. That’s not to say that this study isn’t an important contribution to the debate. It is and I commend the Center for Immigration Studies for doing it. I would just like to see it going much further.

Finally, I’d like to say something about the reports’ conclusion that immigration policymakers should consider a system that puts a premium on education. No less a figure than Julian Simon agreed that this was a possibility that should be considered. But he noted several obstacles in designing such a system. And I think we should also note that there are several agricultural regulations that artificially encourage low educated immigration, like the Depression Era raisin price stability program. Reform of these rules could significantly reduce low education immigration and I think that’s something that policymakers have to look at.

Now, education could be a consideration in immigration reform. Welfare reform, however, is clearly the answer to the problem insofar as there is one identified in this report. When I came to the USA in 1996-7 it was stressed to me on more than one occasion that I needed to support myself and my family.

Obviously, as Steve shows, that stricture has not been applied to more recent immigrants. We need to look more closely at that. Steve is right to say that welfare use is not a moral defect. It’s perfectly understandable. But welfare provision can be demoralizing. As Julian Simon said, requiring immigrants to be self-supporting has at least three advantages.

First and most immediate is that it benefits natives through the public coffers. Second, in the longer run, it improves selection of immigrants for the economy and the society. And a third benefit, by reducing the number of egregious cases welfare abuse by immigrants, it diffuses an objection to immigration which is really founded on unrepresented anecdotes of gross abuse drawn from the newspapers and gossip.

That’s why I think that the argument – Steve’s argument that there are ethical concerns surrounding restricting use of welfare needs to be addressed. There are ethical concerns now surrounding immigrant access to welfare. These considerations apply to native children as well. It’s not 1995 anymore. Welfare reform worked and it’s time to apply that lesson again.

For these reasons, I think CEI and CIS should agree that there needs to be serious reform of the rules surrounding immigrant use of welfare and of welfare in general. That is not, however, an argument for restricting immigration beyond perhaps further exploration of the role of education in selecting immigrants.


MR. KRIKORIAN: Thank you. Iain. Steve, did you want to sort of respond and then we’ll open it up to the audience.


MR. CAMAROTA: Yeah, sure, sure. You know, let me think about something Mickey said. On the question of, you know, the concern about cash -that cash was kind of a trap – by the way, it’s about half and half, the SSI versus the – the problem is the question lumps TANF really in with some other state programs.

The other this is remember we’re dealing with a survey – a survey by proxy. And so the person who’s filling out the information may know that he family gets cash but they would have to know explicitly which program if you wanted to break it by those programs. And so there’s some question as to whether or not – you can break it by SSI and TANF. You can. And it may be accurate. The Census Bureau reported separate numbers and I have elsewhere. But there’s an open question as to how right that is just because of the limitation there.

On the larger question, I was thinking about this – you know, when is someone more dependent on the government. On the one hand, Medicaid is the most expensive program. It’s more expensive than all the other programs combined here. But if your child is enrolled in Medicaid, you may not go to the doctor’s for a whole year.

Now, consider a very popular program that’s not that very controversial – free school lunch. On the other hand, if the government’s feeding your child every day at lunch and you’re not making lunch, you’re not – and it’s something I have to do a lot of mornings, so I think about it. Is that a different kind of dependency? I mean, you go to work every day but the government feeds your children every day. I’m not sure what the answer is there.

But that’s a pretty big responsibility to make sure your kids have lunch and make sure that they can afford to buy it and if the government is stepping in, is that more or less a dependency? I don’t know. It certainly seems a more profound dependency than Medicaid, even though the Medicaid is much more expensive because something is very immediate and very every day. But I don’t know, I just thought I’d think about that.

What is a dependency fostered by that? On Iain’s comments, well, one reason I didn’t look at every possible program, I do have some statistics in the report on the earned income tax credit. The reason I didn’t look at Social Security and Medicare is that not only are they generally not considered welfare programs though you could view them that way depending on how you define welfare.

But the fact is they’re use is universal by everyone who reaches a certain age. If you look ta immigrants who reach the age of 65, almost all of them are on Social Security – look at natives, everyone is on Social Security as well. So the programs themselves are universal of their take-up at a certain age. So they’re not something where there’s any meaningful difference. Once people reach that age they use those programs.

On the larger question of payment size, I do have a discussion in the report and it’s not clear the CPS is very good at payment size issues. The 2010 data that I was primarily focusing on doesn’t have a lot of that information in it yet. For other programs like WIC and food stamps and free school lunch, they don’t have any payment information at all.

Even later they don’t include it. Some data they try to go back in and calculate it and put it back in. But it wasn’t yet available. Now, on the larger question of whether immigrants pay enough in taxes to cover their consumption of public services, the general thought on that question is if you looked at immigrant households and if you looked at native households, then the immigrant households are a net fiscal drain.

They don’t pay enough. But one response or one criticism people say is that, well, you know, that’s because you’re counting U.S.-born children who aren’t really immigrants. And other people would say, but those children are here and those parents – the social services the parents receive on behalf of those children should be counted.

The National Academy of Sciences, when it looked at households, came to the conclusion that the net drain from immigrant households was back in the ‘97 study but the data was ‘94 – was $11 to $20 billion a year. That’s the difference between all the taxes immigrant households pay and all the services that they use. But you could criticize that number. Some people think it’s too low. Some people think it’s too high.

In my report –and let me find one page where you can think a little bit about this – page 11, figure four, I do look at both the earned income tax credit and also the share of immigrant households that have no income tax liability. So for example in figure four we see that 33 percent of native households with children have no income tax liability compared to 48 percent. So what this means is everyone paid the taxes they were supposed to.

There’s no control here for actual compliance. It’s just your liability. The immigrants are much more likely not to have to pay federal income tax than are the native-born, at least if we’re looking with households with children and this reflects lower income, somewhat larger household size and so forth. If you looked at all immigrant and native households, you also get a gap that the immigrant households are significantly more likely to have no income tax liability.

Now remember, a large fraction of natives. Here you can see it’s a third of households have no – zero income tax liability based on the Current Population Survey. So but this study is not really about that balance between public services used and taxes paid. It’s really more about the question of what services are the immigrants using and how do they differ from the native-born population.

And I would agree that in general if you controlled for household size and education level, the differences between the natives often aren’t that big at all except oddly enough if you look at – here’s one thing that’s somewhat surprising and it certainly is an area – Iain’s right – to do more work on.

If you look at page eight, table two, you will see – oh I’m sorry – that’s what’s interesting if you look at household size. But look at page 10, table three, by educational attainment – when you look at the least educated, less than high school, the figures aren’t very different for the native-born and the foreign-born overall and that speaks to Iain’s point.

On the other hand, when you lump the education level, you do find that the differences don’t narrow that much. You can do a kind of compositional analysis here and see how much the large concentration of immigrants who have a high school, don’t have a high school education explains the difference. But it only seems to explain maybe 25, 30 percent of the total difference. So for example if you look only at people with a high school education, the overall figure is 65 percent for the immigrants and 52 percent for the native-born.

Or look at the college graduates – it’s 25 percent for the immigrants and 13 percent for the native-born. So even when you look household education or at least the household head’s education, you generally find that once you move out of the least educated, the differences don’t narrow as much as I would expect or expected to find. So it’s complicated.

But, you know, the program use might differ. So you might be getting use of one type of program versus another. So you’d have to do a lot of detail there. So I guess I would say that educational attainment seems to explain a significant share of the findings but it’s by no means the whole story and table three suggests there’s a lot of unexplained here.

Why would immigrant households headed by a college graduate tend to have significantly higher use rates at least in some programs than a native-born with a college graduate? It’s kind of puzzling. You’d guess they’d be about the same. But they’re not. Anyway.


MR. KRIKORIAN: Okay, thanks, Steve. Questions from the audience? Somebody must have provoked somebody to ask a question, so I’ll open it up. Well, let me take – I mean, I’ll ask the first question, I guess of Iain. And I mean, to maybe put your position kind of too pithily, you might – you might put it in this way.

There’s nothing wrong with mass immigration that ending welfare wouldn’t fix. And you can take that position, I suppose. I mean, I think it’s a logically – it could make sense. The problem is we’ve conducted a vast social experiment starting in the ‘90s – actually from long before that even – in trying to wall off lower skilled immigrants from welfare.

And its failed repeatedly, over and over again, because – I mean, my sense is that a modern society like ours is not going to go to the lengths that would be necessary in walling off newcomers from the welfare system that would make something like that position workable.

And so my question is since we’ve tried to keep immigrants from accessing welfare and failed over and over and over again, is it really responsible fiscally or any other way to keep taking in a million people a year until we somehow fix what you would conceive of as a problem that immigrants are able to access welfare? In other words, until we fix that, isn’t it just prudent to reduce intakes until we fix what you would address as the problem of immigrant access to welfare?


MR. MURRAY: I think that’s a very interesting question. I think – as I tried to make clear – that there is a general problem that access to welfare that goes far beyond immigrant status. And I think that’s the thing that we’re trying – that we need to fix. If there is a possibility of doing a quick fix on immigrant welfare, along the lines that Julian Simon described, then I think it’s worth – that’s worth trying. You know, that’s a quick fix that you can do that perhaps will have some benefits when it comes to welfare policy.

You may well be right that attempts to do the quick fix have failed before. But I think that just generally shows that there is a wider problem with welfare than just immigration. We tried to get large numbers of people off welfare in the ‘90s. But as Steve’s data show, there are some states where most households, most native households, are still using welfare. That’s probably a cultural issue.

If immigrants are buying into that culture, then that’s a significant problem. That’s one of the areas where libertarians tend to believe that there is an argument for restricting immigration is if there is going to be a cultural shift as a result of immigration that affects the free enterprise system. And I think, you know, that is something we have to look at.

Is there something going on that is causing problems for our free enterprise system? I happen to think that what’s causing the most problems for the free enterprise system is actually in the native-born population as much as it’s in immigrants. But again, perhaps it can be part of a quick fix.


MR. CAMAROTA: Well, if I could follow up on that though, I guess the question is – and you might – let’s assume that you think that these programs tend to be very popular – things like WIC, right? The Women, Infants and Children nutrition program, this is for low-income women who are pregnant or a person who’s recently had a child – formula and other nutritional supplements. Now, is it your sense – and that programs’ used very heavily by immigrants.

You know, I just say the name of that program and I think I’ve won the debate. Women, infants and children – do you imagine in the America that we live in that that’s a program where we’re going to restrict immigrant access to? You know, one out of every four children born in the United Sates is born to an immigrant woman right now and of low-income children it’s got to be closer to a third.

It’s one out of every three low-income children born to an unemployed is born to an immigrant woman. Immigration has two effects. It dramatically increases the number of people who might need that program, assuming you think it’s a good program. I’m not arguing that you do. But also that it just becomes impractical to talk about not – can you imagine trying to deny free school lunch or health care? See, the cash programs I think you can make that case and kind of did it.

But I think that argument has gone as far as it’s going to go. We’re not going to be restricting WIC. Do you think we would really in the United States?


MR. MURRAY: Well, Paul Ryan is about to present a budget proposal at the AEI in half an hour or so which attempts to reform Medicare – so to take $4 trillion off the cost of it. I think that, you know, what we need is more big thinking like that. You know, so rather looking at each individual triangle of the program and arguing, well, you can’t take that away, you can’t take that away – I think we need to be looking at big picture reform of all these programs. And you know, when it comes to the big picture reform, immigration is a very, very small part of it.


Q: Mr. Murray, the very end of your remarks you talked about the possibility of investing in educational criteria for the issuance of green cards. I’d like to know if you favor that and also very briefly toward the end your remarks you talked about – (inaudible) – the agricultural program – (inaudible) – could you talk about that as well?


MR. MURRAY: To deal with the agricultural program first, we’re still living with a lot of Depression Era policies that have had a significant effect on us ever since they were brought in as part of the New Deal. A very interesting example of that is the raisin price stability program. It has a formal name which I can’t remember at the top of my head.

But what this does is in an effort to bring price stability to the raisin harvest, because in the 1930s with all the climatic issues that were going on then, there was significant volatility in the raisin price. And as a result, raisin farmers were going out of business one year and experience a glut the next and going out of business because they had too much.

So FDR and his people brought in a program whereby the government bought – buys still to this day – between a third and a half of all the raisins grown in the United States. And it distributes some of those to children as part of a school nutrition scheme. Some of them it just throws away. We are growing far too many raisins in the United States to meet market demand.

Now, it just so happens that raisin harvesting is probably the single most labor intensive program – agricultural product – there is. So raisin harvesters in California are using huge numbers of undocumented immigrants to come in and pick these raisins which are then simply thrown away.

If we got rid of that Depression Era program and used free market methods to stabilize the price – the unique climatic factors in the 1930s aren’t an issue anymore – then we will deregulate the industry, take a burden off our raisin farmers, open up raisin farming to competition and at the same time reduce undocumented immigration.


Q: By how much?


MR. MURRAY: Sorry?


Q: By how much?


MR. MURRAY: Unfortunately I can’t remember the figures off the top of my head. If you give me your contact details, I can get those to you.


Q: And then the question about education?


MR. MURRAY: Yes, as I was saying, those raising harvesters that are coming in are normally right at the lowest end of the educational spectrum. So that sort of reform –reform of those programs would significantly reduce the lower educated people, who as Steve identifies are the main – the main takers of welfare.

When it comes to higher educated – trying to put a premium on higher education – countries all over the world attempt this. A lot of countries – Canada, even in my own U.K. at one point – have point systems whereby to qualify for – I wouldn’t say automatic – but to qualify for easier access to the country, you got more points if you had an advanced degree.

I’m not sure of the success of those programs. I need to look into that a bit more. But Julian Simon, when he reviewed the programs in the ‘90s thought that some of them showed progress and some of them had significant problems. I think it would be very worthwhile for somebody to do a study of how those programs have panned out since the ‘90s to identify whether it’s an option that we could successfully pursue in the United States.


MR. KRIKORIAN: Yes, ma’am? You had your hand up.


Q: I was wondering how –


MR. KRIKORIAN: If you could identify yourself too, please?


Q: Oh I’m sorry. I’m Amanda Carroll – (inaudible) – and I was wondering how illegal aliens access, like, welfare and provisions such as that. Like, according to the statute you have to be a legal immigrant or a naturalized citizen or whatnot to access the program. And I understand if you have an American citizen child but by statute then you’d only be allotted for an amount for the citizen child. So then how is the argument coming about that illegal aliens are accessing welfare programs?


MR. CAMAROTA: Well, it depends on how you look at that, right? There are a few cases of fraud. There would be the case of a person who needs emergency medical care, particularly during a pregnancy, can be enrolled in Medicaid.

So there are a couple of exceptions. But mainly we’re looking at households. So the household head is illegal and typically they have a spouse who is also illegal but they have U.S.-born children and the children are enrolled in free school lunch, the children are enrolled in Medicaid and the parents – if they have, say, a young child –are enrolled in WIC.

So that would be an example. Now, there are cases and we have a few of them identifiable actually in the data where you have a child who was born in the United States who’s considered disabled and is getting SSI and you have cases where – and it’s considered a child-only account. Or you have cases where people receive TANF but not the parents. It’s essentially for the children.

There’s not a huge number of those and that’s why the cash assistance shows low. But there’s a very large fraction of illegal immigrants who sign their kids up for free school lunch, who sign their kids up for Medicaid, who sign their kids up for WIC. And there’s also some who get food stamps.


California in particular – you can see it in the data and you hear about it all the time – has a real fraud problem in its food stamps program. So there it does look like illegal immigrants have really penetrated that program. But generally that’s not the case. Generally that’s not what’s happening.


Q: So the amount then – (inaudible) – you would include the number of the household in your study even though it would like be, like, an amount allotted to one?


MR. CAMAROTA: Well, not necessarily just allotted but just enrolment, right, or, you know, in Medicaid. Yes, no and let me make that clear. The estimates and then we break it down by each individual program are based on the characteristics of the household head –whether we think the person is legal or illegal – and then the share who have someone in that household and for the illegals it’s typically – although not always – a U.S.-born child enrolled in these programs and then we have them laid out.


MR. KAUS: Right, and for TANF, the share that goes to the children is the vast bulk of the welfare benefits. The share that the parents would be disqualified for is like 20 percent. It’s not –


MR. CAMAROTA: Yeah, it’s real small.


MR. KAUS: The bulk would go to the children but as this data shows, illegal immigrants don’t sign up for much of TANF.


MR. CAMAROTA: In a few places, that seems to happen. And I’m not sure if it’s a statistical quirk or what. There are people who would be the parent is disqualified but the children get it and they show up in the data and the methodology we use to distinguish illegal from legal just allows for that and we find those people. But it’s not huge.

The big stuff is the food assistance and the Medicaid for the illegals. For the legal immigrants, it’s everything, especially the unskilled legal immigrants. In one of the tables – this is really the toughest challenge for I think a person who makes the case that unskilled legal immigration is a good idea. On page 17 we look at households that are not refugee-sending countries.

And this is table six, where the household head has no more than a high school degree. So either the household head didn’t graduate high school or has only a high school degree and pretty much across the board you’ve got pretty darn high use rates. Now, remember for the cash it shows 11 percent. The actual rate is higher, as I indicated at the outset, because people understate their use.

But even the use of housing assistance is much higher than the native-born – so, somewhat higher and so forth. Those individuals – you now, that is a – that is, I think, really one of the biggest questions. You know, could you change that somehow by reforming the welfare system? It seems unlikely to me. Iain feels that you could change that and not have these folks accessing so many programs in high rates. I’d be very skeptical.


MR. KAUS: If I could just say one thing to argue with Iain, it seems to me that in ‘96 we decided we wanted to shift – you know, we did it very imperfectly and in a halfway sort of way. We wanted to shift from a society that gave welfare benefits to fight poverty. So I guess it would be poor – to a society that wanted to give benefits to the working poor.

In other words, if you worked in our society, we didn’t want you to be poor and so we would – if you worked, you would get all of these benefits ladled on top of your wage. And because incomes – the income spectrum has been stretched and unskilled work doesn’t pay that much, it turns out that it’s a fairly substantial subsidy if you add up the EITC and all the other benefits available to the working poor. I don’t think we’re going to change that.

I think we’ve decided if you work in the society at a minimum wage, you should be able to live a life of dignity and we’re going to give you the EITC and WIC and all the – and medical care on top of your wage. SO I don’t think you can – and if immigrants come in and they get those things, it’s going to be a substantial expense. I don’t think you’re going to get rid of the welfare part because welfare by and large is now – there are exceptions like food stamps directed to working poor people.


MR. MURRAY: Well, I think here is where we can actually learn from what’s going on in the United Kingdom at the moment where the tastefully named Iain Duncan Smith who is the work and pension secretary over there identified that British welfare had created a working poor poverty trend – that the sheer number and variety of benefits that people were – that the people were obtaining were basically keeping people earning very small amounts of money and stopping them from – certainly from becoming entrepreneurs and also from accepting promotions and so on.

So he is moving towards a system which sweeps away this vast number of welfare programs in the U.K. and replacing them with a single universal credit which is to be designed such that simply earning more money does not make you poorer.

And that I think is a problem that we’ve got in the States at the moment. And that’s – so I think we need to look at what the U.K. is doing there to think about whether we should be enacting a similar welfare reform, which would indeed be of the scale of what happened in the ‘90s.


MR. KRIKORIAN: I had kind of a question, I guess directed sort of to Mickey. And I mean, you would acknowledge that welfare – cash or non-cash – is at least a necessary evil. In other words, it would be better if nobody had to collect welfare, right?


MR. KAUS: Right, in other words –


MR. KRIKORIAN: So is there any rationale for an immigration policy that admits anyone who is likely to use welfare? In other words, isn’t every – the admission of any foreign-born person who is going to use welfare at any rate, even if it were lower than that of the native-born – isn’t that in a sense a kind of failure of policy?

In other words, not so much a moral failure because I think we all agree this issue that, you know, my grandma from Sicily didn’t use welfare. These guys today are somehow, you know, worse – that that’s silly. But isn’t a failure of policy to have any immigrants on welfare?


MR. KAUS: Well, if – is it a failure of policy if you let in an immigrant and they work very hard at a low-wage job and as a result they qualify for the earned income tax credit and various things? And I would say a part – if that was the only effect we wouldn’t be worried about immigration. Clearly on the other side it’s good for the immigrants to come to America.




MR. KAUS: Clearly there – if – you know, there are employers that like having that immigrant laborers. So and yes, it costs something if they qualify for the EITC but if that were the only negative aspect of immigration, we wouldn’t be opposing immigration. We’re opposing uncontrolled immigration.

We oppose it because it drives down the wages of the native-born and it’s threatening to create an inegalitarian society on the order of Rio de Janeiro in parts of southern California and it’s also challenging our national culture and national sovereignty. Those are the big issues. This is an issue that the expanse of welfare is something that’s very salient now because we’re in the middle of a big budget crisis. But by itself I don’t think it would be enough to worry that much about immigration.


MR. KRIKORIAN: But as Steve pointed out, it’s not really just the expense. It’s also welfare use as a kind of yardstick for some of the things you are actually talking about. In other words, if we’re turning into a Brazil-kind of society with huge gaps between rich and poor, high welfare use is sort of one of the canaries in the coal mine that points in that direction.

Do you see what I mean? In other words, it’s not just the money, it seems to me. It’s that this is an indicator – a symptom of failed immigration policy which has all kinds of other effects too.


MR. KAUS: It’s a symptom of a failed immigration policy but – I agree with that. But I mean, I guess – you know, if we have a minimum wage and people get the earned income tax credits, yes, we don’t become Brazil.

We become Brazil if, you know, immigrants pour in and they’re working under the table and the people – you know, people who would ordinarily be working for the minimum wage – Americans – can’t find work at all and they end up on welfare. That’s the problem to me.


MR. KRIKORIAN: Anyone else? We can take one last question I think before we break up. No? So I think – I’m not sure about everybody’s schedule. But I’m pretty sure Steve is willing to be accosted afterwards if you have anything you want to say.

So for those of you who didn’t get the report, it’s back there on the table outside. It’s in a folder with the report and the contact information. This and all of the rest of our work is online at And I’d like to thank Mickey and Iain for coming out for this. And thanks to all of you and see you at our next event.