The Costs of Illegal Immigration:
The Impact of Illegal Aliens on the Federal Budget

Panel Discussion Transcript

August 25, 2004
National Press Club
Washington, D.C.

Moderator:
Mark Krikorian, Executive Director, Center for Immigration Studies

Panelists:
Steven Camarota, Director of Research, Center for Immigration Studies

B. Lindsay Lowell, Director of Policy Studies, Institute for the Study of International Migration, Georgetown University

Robert Rector, Senior Research Fellow, The Heritage Foundation

MARK KRIKORIAN: Good morning. My name is Mark Krikorian. I'm executive director of the Center for Immigration Studies, a think tank here in town that examines and critiques the impact of immigration on the United States. Let me get the plug in at the beginning: all of our work, including the report we're releasing today, is online in its entirety at our website.

Last night the draft Republican Party platform was released in preparation for debate and discussion and the general convention -- the Republican Convention next week -- and the platform echoed President Bush's call earlier this year for, among other things, widespread illegal alien amnesty. Senator Kerry, the Democratic candidate, has also called for a different, speedier version of an amnesty for a large number of the 10 million illegal aliens in the United States, and there are a variety of other proposals introduced by Republicans and Democrats that also call for amnesty as the solution to illegal immigration, in a sense to define it away.

None of the discussion, though, of an amnesty in the illegal population has looked at the fiscal fallout -- the budgetary consequences of doing this. The last amnesty, which Congress passed in 1986, had enormous fiscal fallout. In the first 10 years we had published an estimate of direct and indirect costs of something like $70 billion. No one has conducted a similar examination of the current and likely future costs of illegal immigration, and this is what the report we're releasing today does. There are a couple of caveats. Steve, the author, will make them. But let me emphasize, this looked only at the taxes paid and the services used at the federal level, not at the state and local levels. There are a variety of both taxes and services at those levels that would be relevant to a broader look at it. This is looking at -- sort of the first look at the national issue, at the federal issue. And it was actually quite striking to me that no one has even raised the question as to what effect this is going to have. The discussion has been -- even for Washington -- has been remarkably superficial and frivolous. And so to try to remedy that we're releasing today a report that is really the first detailed look at what taxes illegal alien-headed households pay and what services they use, to try to come up with an estimate of what the impact is and what the likely impact would be if illegals were amnestied.

The author of the report is going to speak first. Steve Camarota, the director of research at the center, has become one of the best-know commentators on the effects of immigration in the United States, and has written -- we've published a variety of reports that he's written, all of which are on our website. Then we'll have comment from two experts -- outside experts on this subject. Lindsay Lowell, he is director of policy studies at Georgetown University's Institute for the Study of International Migration. He's written on this subject for years and in fact was research director at the Barbara Jordan Commission, the commission on immigration reform in the 1990s. And the other commentator is Robert Rector, a senior research fellow at the Heritage Foundation who has written widely on issues focusing on welfare policy, low-income populations and the underclass, including on immigration-related topics.

So what we'll do is Steve will make his presentation first. We'll have some comment from the two discussants, and then we'll open it up to Q&A for all of the panelists from the audience.

Steve?

STEVEN CAMAROTA: Thank you, Mark. Let me just reiterate what Mark said, that the report that we are releasing today is available at our website.

Now, obviously, concern over illegal immigration spans a wide range of issues, from national security to the rule of law to the risks illegal aliens take to enter the United States. However, with perhaps 10 million illegal aliens now living in the country, concern about their impact on American taxpayers has become, obviously, a very big issue. Of course, putting aside the fiscal costs, many Americans are understandably outraged at the very idea that millions of people can cross our borders illegally or overstay temporary visas and live here with little difficulty and the federal government doesn't seem to do much to stop them. And while other issues are clearly important, again, the effect illegal aliens have on public coffers is at the center of the ongoing debate about illegal immigration.

Now, of course, almost all observers agree that illegal immigration is a problem, but there's less agreement on the size and scope of the fiscal costs. But this is an important question because if there are large fiscal costs then this should have some impact -- and almost certainly will -- on the kinds of services that can be offered to natives or illegal immigrants, or on the other hand, the tax burden on Americans. Thus, policymakers need to take into account the fiscal impact when formulating their response to illegal immigration.

As Mark also pointed out, surprisingly few studies have attempted to measure the total fiscal impact of illegal immigration. Several studies have focused on all immigrants, making no distinction by legal status. Other research has examined only costs or only tax payments. And most work has focused on the state and local level, giving little attention to the federal government.

We undertook this study to correct some of these previous shortcomings. We attempt to answer in this report two related questions. First, what is the fiscal balance at the federal level? That is, if you take all the taxes illegal aliens pay and all the services that illegal aliens use at the federal level, what is the balance between those two things?

Second -- and a closely related question -- what would happen to that fiscal balance if illegal aliens were given amnesty? Now, again, it's important to note that the study being released today deals only with the impact of illegal immigration on the federal budget. Costs at the state and local level are not considered.

Now, how do we do this? How can we come up with these estimates? Well, to estimate the impact of households headed by illegal aliens, we rely heavily on the approach used by the National Research Council in its 1997 study The New Americans. The panel that conducted that study included most of the top people in the immigration field, including economists and demographers and sociologists.

Now, although that study did not separate out illegal immigrants, it did provide a model on how to approach the impact of immigration on the United States in terms of fiscal effects. Like that study, we mainly used the March Current Population Survey collected by the Census Bureau, and we used the March file from 2003 of that survey, and that survey asks people about their use of welfare and the taxes and their income and so forth in the previous year. So all the estimates I'm going to give you today are from 2002.

Now, like the National Research Council, we assume that illegal aliens have no impact on defense-related expenditures, and therefore those costs are assigned only to native-headed households. We also follow the National Research Council's example and define households as persons living together who are related.

Now, the reason that we use "household" -- we use it for the same reason that the NRC, the National Research Council, did. "A household is the primary unit through which public services are consumed and taxes are paid," in the words of the NRC. Thus it makes perfect sense to estimate the tax payments and services by household, but it should be pointed out that by calculating cost by households we are including the U.S.-born children of illegal aliens who are American citizens. Because the presence of these children and the costs they create are a direct result of their parents having been allowed to enter and remain in the country illegally, counting services used by these children allows for a full accounting of the costs of illegal immigration.

While the Census Bureau does not ask people, of course, whether they are illegal aliens, lots of outside researchers, including the Urban Institute, the Immigration Service, and the Census Bureau itself have actually used the characteristics of people in the survey to identify illegal aliens, and we follow their example. This method of course is based on some very well established facts about the illegal population. I won't go into all the details. It may surprise some people that illegal aliens actually respond to the Census or to government surveys, but it turns out they have such little fear of enforcement that when sent a form they generally send it back, or when someone knocks on the door they answer questions.

Now in some cases we assume that an individual will have zero chance of being an illegal alien, such as someone who says he's a naturalized citizen, someone who says he's a veteran, someone who says that he personally receives Social Security or cash from welfare programs, or that he personally is enrolled in Medicare. However, other members of the household, mainly their U.S.-born children, can and often do receive such benefits.

Now, we estimate that in March of 2003 there were roughly 9 million illegal aliens in the current population survey. By design, our estimates for the size and characteristics of the illegal population are very similar to those prepared by the Census Bureau, the Immigration Service, and the Urban Institute.

Now, in terms of tax payments, we follow the example of previous research and assume that about 55 percent of illegal aliens are paid on the books; that is, that they have income tax withheld. Again, that may surprise people, but you have to understand how lax the system is to understand why that happens. If you go in and just give a made-up Social Security number to an employer, it will take quite some time, if ever, before Social Security and/or the IRS ever get back to the employer. So you can actually go in as long as you've got a drivers license or some other documentation that you can buy on the street that says that's your Social Security number, [and] you can pay taxes on that number.

Now, to estimate the impact of giving illegal aliens amnesty, we assume that any amnesty that passes Congress will have lawful, permanent residence as a component. Lawful permanent resident status is also called a green card. It allows you to live in the United States for as long as you like and get citizenship after five years. Now, to estimate what would happen if illegals got legal status, we assume that illegals, if given this green card, they would begin to pay taxes and use services, like households headed by legal immigrants, with the same characteristics in terms of educational attainment. We also control for whether the household is from Mexico.

So what did we find? Looking first at households headed by illegal aliens, we found that they paid on average $4,200 a year to the federal government for a total of about $16 billion a year in 2002 -- certainly not a trivial sum. However, we also found that they imposed costs on the federal government of nearly $7,000 a year, or $26.3 billion. This means that the net fiscal deficit created by illegal households was $10.4 billion, or $2,700 per household. Put differently, illegal aliens imposed $10.4 billion more in costs than they paid in taxes.

Among the largest costs were Medicaid, $2.5 billion; cost to treat the uninsured, $2.2 billion; food assistance programs, $1.9 billion; the federal prison system and court system, $1.6 billion, and federal aid to local schools, $1.4 billion in costs. And one of the most troubling statistics in the report is the discovery, based on prior research, done actually at the Urban Institute, that roughly 17 percent of the prison population at the federal level is illegal aliens. That's a huge number since illegal aliens only account for about 3 percent of the total population.

Now, why do illegal aliens create a net cost at the federal level? Well, it's really pretty straightforward. With nearly two-thirds of illegal aliens lacking a high school degree, the primary reason they create a fiscal deficit is their low education levels and resulting low incomes and tax payment, not their legal status or even particularly heavy use of most social services. Nor is it caused by an unwillingness to work. The vast majority of adult illegals in fact hold jobs. But their low education levels mean their incomes are very low, and there is no single better predictor of income and economic well being than one's education. And of course, income is the primary determinant of one's tax payments and also one's eligibility for many social services.

Now, when we try to estimate what would happen to this $10 billion deficit, if we gave illegal aliens legal status and they began to pay taxes like similarly skilled legal immigrants, we find the costs go up dramatically. If illegal aliens were given amnesty and began to pay taxes and use services like households headed by legal immigrants with the same education levels, the net fiscal deficit would go up from $2,700 per household to $7,700 a household. And the total net cost would rise from $10.4 billion to $19 billion. In other words, if we gave illegal aliens legal status, they would become basically unskilled, for the most part, legal immigrants. And as a consequence, the costs, the net fiscal deficit, would nearly triple.

Now, how do we know this? Well, we know this because legal immigrants with the same education levels as illegal immigrants make extensive use of public services but still tend to have very modest incomes reflecting their low levels of education. Thus, even though we estimate that average tax payments would rise by 77 percent per household on average, we estimate that costs per household would rise 117 percent.

To understand why costs go up it might be helpful to consider a program like the earned income tax credit, which pays cash to low-income workers. Now, illegals currently account for only 1.5 percent of the credit because they need a valid Social Security number and then they have to file an income tax return, so the vast majority don't. But if they were legalized they would begin to use this program, and in fact, the cost for that program would go up 10-fold.

Now, remember, this dramatic rise in cost is not due to laziness on the part of the illegal aliens. In fact, only those who get the credit -- and to get the credit you have to work. The dramatic rise in costs simply reflects the low education levels of illegal aliens and their resulting low incomes. It must be remembered that legalizing illegal aliens will not change their education levels or the fact that the modern American economy offers very limited opportunities to unskilled workers, whatever their legal status, nor will it change the basic fact that the United States, like all industrialized democracies, has a well-developed welfare state that provides assistance to low-income workers. Large fiscal costs are simply the unavoidable outcome of unskilled immigration, given the economic and fiscal realities of modern America.

Now, of course, many native-born Americans often observe that, hey, my ancestors came to America, they'll say, and they didn't place huge demands on government services. Perhaps this is true, but the size and scope of government was dramatically smaller during the last great wave of immigration, not just welfare programs, which were virtually nonexistent, but expenditures on public schools and even roads were only a fraction of what they are today. Thus, the arrival of unskilled immigrants in the past did not have negative fiscal implications in the way that it does today. Moreover, the American economy has changed dramatically since the last wave of immigration.

Education is now the key determinant of economic success. In the past you could come to America, 100 years ago, with little formal education and still find work in mining, agriculture, or primarily in industry. Put simply, large-scale unskilled immigration is incompatible with the economic conditions in the United States.

What does all this mean? Well, I think it's fair to say that it is -- one of the key conclusions is that it is grossly inadequate for policymakers and politicians in both political parties to say that all that matters is matching a willing worker with a willing employer. The arrival of unskilled workers has enormous implications for taxpayers. While businesses can be counted on to ignore these costs, policymakers must at least consider them. This study makes clear that looking even just at the federal level and ignoring the local level, those costs are considerable.

Now, assuming that the fiscal status quo is unacceptable, one set of options to deal with this problem might be to try to, well, reduce the costs illegal alien families impose, but doing so would be extremely difficult. Illegal alien households already impose only about 46 percent as much in costs on the federal government as other households. The problem is they only pay about 28 percent as much in taxes. But the bottom line is the amount of money that can be saved by curtailing their benefits would probably be quite limited. Moreover, the fact that benefits are often received on behalf of their U.S. citizen children means that it is very difficult to prevent illegal alien households from accessing what programs they do because their citizen children always retain eligibility.

So if you were to say -- and we do -- that illegal aliens can't use Medicaid, but if the illegal alien has three U.S.-born children and his incomes qualifies -- if the income is low enough -- those three children can be enrolled as full members of Medicaid, and there is no way under existing law, which awards citizenship to the children of illegal aliens, to prevent that from happening. So talking about how we would try to cut back illegal aliens' use of public services is likely to have very limited impact. It's not likely to provide large savings.

And many of the programs, it should be pointed out, that illegal aliens use most extensively are politically very difficult to cut. Such programs as, for example, the Women Infants & Children program -- now, that name should tell you right there why there would be an enormous outcry if you tried to cut that program, even though illegal aliens are making very extensive use of it. And other costs, such as incarcerating illegal aliens who have been convicted of crime, seem to be obvious and unavoidable if the illegal aliens are allowed to stay.

Put simply, it seems almost certain that if the illegals are allowed to remain in the country, so will the fiscal costs. If we are serious about avoiding the fiscal costs of illegal immigration, the only real option is to enforce the law and reduce the number of illegal aliens in the country. First, this would involve much greater efforts to police the border, both the land border and the sea borders. At present, fewer than 2,000 agents are on duty at any one time on the Mexican and Canadian borders. Second, much greater effort must be made to ensure that those allowed into the country on a temporary basis such as tourists and guest workers return to their home country, as they agreed to, and not remain in the United States as illegal aliens. Of the roughly 10 million illegal aliens who are now in the United States, probably somewhere around a fourth to a third are visa overstays. That is, they came in on a temporary visa and they just never went home.

Now, the centerpiece of any enforcement effort in terms of illegal immigration must be to enforce the ban on hiring illegals in the United States. At present it is unlawful to employ an illegal alien, but the law remains entirely unenforced. Now, this would require things like using existing databases to make sure that all new hires are authorized to work in the United States, and levying heavy fines on businesses which employ illegal aliens.

Finally I would add that a clear message must come from policymakers in both political parties, especially, though, senior members of whoever is in charge in terms of the administration, that enforcement of the law is valued and vitally important to the nation. This would dramatically increase the extremely low morale of those who currently are involved in enforcement.

Consider the effect on people who work in the immigration services that enforce the law . . . of both the president and the leading challenger, John Kerry, both saying that, well, the way to deal with illegal aliens is we are going to give them legal status. That is like the head of the DEA saying, the way to deal with illegal drugs is we should legalize them. Now, but you go out and enforce the law. That is a critically important component -- doesn't involve spending more money, though that would be needed, but it would have an enormous impact on what would happen.

Now, of course, policing the border, enforcing the ban on hiring illegal aliens, denying temporary visas to those likely to remain permanently, as well as all the other things necessary to reduce illegal immigration, would take both time and money. However, since the costs of illegal immigration are very significant, that also means that significant resources could be devoted to enforcing the law and still might leave taxpayers with a nice net savings. Again, I should add that it must be remembered that the negative -- the net negative impact of illegal aliens on the federal government of over $10 billion does not include costs at the state and local level, where the costs are likely to be even higher.

Now, enforcement not only has the advantage of reducing the cost, it is also very popular with the general public. Nevertheless, policymakers can expect strong opposition from interest groups, especially ethnic advocacy groups and those elements of the business community that do not wish to pay their workers more or invest in labor-saving devices and techniques, and instead just want access to a lot of unskilled workers.
If, instead, we choose not to enforce the law or we decide to grant illegal aliens amnesty, both the public and policymakers have to understand that these latter two choices involve a significant and unavoidable long-term cost for taxpayers.

Thank you.

MARK KRIKORIAN: Thank you, Steve. We will first go to Lindsay and then to Bob.

B. LINDSAY LOWELL: Good morning. I'm going to make some general comments on two issues, one of which is the reliability of the report, and the other is what is the value of the report? On the reliability, to cut to the chase, the precision of the estimate can be debated, but I think most reasonable experts would concur that a net fiscal balance is what happens with this population. As for the value of the report, I think we should call attention to the obvious fact that there is never a free lunch. Amnesties will introduce costs to the U.S. policies that should be considered, but there other aspects of U.S. policy that are important here as well -- not just immigration policy, but our labor market policy has created much of the situation.

So how reliable is the estimate? The report points some things out that are in the favor of unauthorized aliens. The use of cash assistance programs is low. Even compared to otherwise similar natives, we know that immigrants often tend to use welfare at lower rates. This is true of illegal residents as well. And over half of unauthorized aliens pay taxes of at least $4,200 per household. That is a significant tax for these often very poor households. Furthermore, about $7 billion flow into Social Security and Medicare from this population. This reinforces the observation, among other things, that we know these people come to work, as Steve has just said.

The report also mentions some surprising and little-discussed costs: 17 percent, according to best estimate, of those in federal prisons are illegal aliens and they run up a substantial cost. This isn't news at all to the states that seek reimbursement. A roughly similar cost is found in federal courts. And payment for the treatment of the uninsured may run as high as $2.2 billion. Again, this is no surprise to hospitals and states that are trying to get federal reimbursement for these costs. There is a shadow side to the population that needs to be considered. It's real.

Now, as to the line items, it's kind of tedious to go through but it's worthwhile because that's where many of the questions come. Experts will argue about various sides of the ledger but let's talk about some specific things, what I would call uniquely allocated costs. So for example, should the former INS actually be -- should its costs be attributed to the illegal population? Wouldn't, in fact -- suppose all unauthorized aliens were to leave tomorrow, wouldn't you still need border enforcement? I would think so. In fact, it's often the case that it's greater enforcement that leads to the absence of crime, or illegal entry in this case, so might not even costs go up higher? I think that's reasonable to say, but that just argues in fact that maybe this cost item should be put into the "all other expenses" that are also attributed to natives, for example, but if that's done it's unlikely to change the story.

There is what I would say, what's the true amount of something? This is the problem of marginal versus average cost. The marginal cost is -- it doesn't cost as much to add an additional person to a school system as the first person, or an additional person in terms of infrastructure and roads, and that's where the "all other expenses" in this particular estimate really has some issues because we will probably want to see marginal costs here. And this is a big item. In fact, most of the deficit in the illegal population estimate here is this other-expenses item -- line item.

Well, let's assume that the true cost here is not an average cost; it's a little bit lower than what we've got there, so we reduce that estimate. I mean, is it reasonable to keep that estimate? Yes. Illegal residents incur costs and we should allocate costs to them for living in our nation. But even a significant reduction of those costs would still leave us at the same place, with a net fiscal deficit as the likely story that we've got here.

There's a public investment issue. Wouldn't the cost of not educating a child or not providing health care even be greater than providing the costs in the first place? Unhealthy people are a public health hazard. Well, that really is a serious issue and it's not simply philosophical. And there are probably ways to kind of figure out what the net costs are that would, again, reduce these estimates a little bit, but nowhere in the report is there a suggestion that these costs should be eliminated, that these services should not be provided to this population. And at any rate, states are currently suing the federal government for cost sharing and a lot of different things. It seems sensible, once again, in the way we're operating in this nation, to assign these costs to illegal residents.

Well, then there's misplaced costs. The citizen children of illegal parents are eligible for the benefits they receive. They are eligible. Should these items then go on a line-item cost against the illegal population? Well, at the same time, citizen children who have left the household aren't assigned costs, so it really does seem reasonable that as the NRC panel concluded, that the children of the household are a charge against their parents, and so it makes sense in many ways, I think, to go ahead and put the cost for while they're in the household onto the unauthorized household.

Aren't we singling out illegal aliens in many cases? In fact, low-income households, including those of legal immigrants, as we've just heard, and even of the native born, generate a net negative fiscal balance. Well, that's true enough but at the same time, until national boundaries fall we haven't chosen, under the color of law, to have unauthorized aliens. And so in that regard I think it's quite reasonable to separate their cost out as a different kind of cost to the nation.

What's the bottom line? The exact amount of the net negative fiscal balance is something that one should not probably take to the bank, but there is error in the estimate. But economic theory and logic shouldn't make the basic findings surprising. Poorly educated workers don't earn much, they can't pay much in taxes, yet these are precisely the households that often live in poverty and need public benefits. The cost is likely, no matter how calculated, to be negative and somewhat substantial.

Well, having said that, what's the value of the report? As I said, it points out that there is no such thing as a free lunch. In a post-modern industrial economy, this population runs up costs. The earned amnesty and guest worker programs being discussed on the Hill may not come to pass this year, and maybe not even next. So is it a red herring? I don't think so. There has been more interest in this issue expressed in the last three or four years than in the previous decade, and for the first time, many different actors in the debate are willing to entertain the notion, not simply of earned amnesty for sizable populations of 500,000 per million workers, but indeed for the entire population.

As that debate goes forward it's very reasonable to entertain the costs that will arise. And they're not just fiscal costs. More, at the same time I should also point out, are the costs themselves the reason for not taking certain kinds of actions like granting amnesty, but they should enter into the debate and they should give some reason for pause.

From my point of view, what a broader reading of this report should also make clear is that there are some disturbing trends that create the context in which we find ourselves. Low-skilled workers in the United States for the past few decades have been seeing their earnings erode relative to more skilled workers. That paused a little bit during the new economy in the late 1990s, but in fact there is still a lowering of the relative wage of earners on the low-end side.

There seems to be a consensus in the research community that this has to do with an increased demand for knowledgeable workers in a technology-advanced society. But there are also those who point out that there has been a simultaneous reduction, or erosion, in the institutional support for the wages of low-skilled workers. Some of these are fairly easy to see. The minimum wage has kept nowhere near the inflation rate since the 1960s. A living wage is variously calculated to be $12 to $14 an hour. At the same time, we know that newly arrived unauthorized workers generally earn maybe $7 to $8 and hour, max.

We also know that especially since the 1980s -- but it's a part of a longer-term trend -- that both worker and union clout has suffered in the United States. So there is less ability for these workers to stand up for their rights and get paid what we would probably want to pay them, all other things being considered.

U.S. immigration policy, along with these other developments, has, with a wink and a nod, encouraged the growth of a low-rate sector that is supplied to a large degree by unauthorized workers. And in that regard, perhaps employers and consumers benefit, but as citizens we're abetting the growth of an underprivileged class, and as taxpayers, we are subsidizing employers.

MARK KRIKORIAN: Okay, thanks, Lindsay.

And Bob Rector.

ROBERT RECTOR: Thank you very much. I find it to be an excellent report and I think the numbers are quite accurate, particularly with regard to the central issue of the receipt of welfare benefits. I think the underlying principle here is a fairly simple one: that a society which has an advanced welfare state for the economically disadvantaged will incur considerable costs when it imports large numbers of low-skilled individuals, that those individuals will inevitably end up being a substantial and significant fiscal drain on the rest of society as well as possibly creating social problems, which I will get into later.

Just for example, the total cost of means-tested welfare assistance in the United States last year -- means-tested welfare are programs that go only to low-income persons -- was $530 billion. That is over five percent of total personal income. Some people think that we ended welfare in the mid-1990s. It would be interesting to know what this cost would be if we hadn't ended it. In fact, I think we did modestly slow down the growth of these programs, but only modestly, and they continue to absorb a very significant portion of our economy. We spend a considerable amount of time and effort essentially taxing upper middle-class families in order to transfer very, very large amounts of economic resources to those that are less fortunate. The question is, what happens in that type of system when you import people who inherently are going to be less fortunate within that system? Another issue I think that is tied in here is that the welfare system maybe 20 years ago may have significantly involved families where the single mother didn't work at all. It's changing now, and the most expensive cash program is the earned income tax credit, which is designed as a wage supplement for low-income parents.

So whether or not the individual works somewhat during the year is largely irrelevant to the type of welfare costs that are going to come in. In fact, if you look at the 1996 welfare reform, what in effect actually happened in that reform was that we largely curtailed welfare payments that consisted of sending out monthly welfare checks to single mothers who didn't work over long periods of time, and we took virtually all of that money and put it into various types of subsidies for working low-skilled parents, particularly daycare, but there is a huge expansion to the earned income tax credit, creation of state-earned income tax credits and so forth.

So the system is, I think, rationally designed to take parents who have a difficult time supporting families according to our standards by their earnings and to supplement them and to bolster them. And then in addition, these families -- all of these kids would be getting a free education or a highly subsidized education at the state level, which is probably an even greater cost that is not considered in this report. And these -- when you look at large numbers of workers coming in without a high school degree, anyone who understands the nature of welfare, understands the nature of the underclass and the semi-underclass, can see that essentially you are importing future clientele for these programs. It's not surprising that you find that even with an illegal status that they are receiving considerable benefits under these programs, and if legalized, that the receipt of and the cost would go up considerably.

I think it's important to note that this is a static report. It assumes no behavioral response to periodic granting of amnesty. Of course we all know that each amnesty that will be offered is absolutely the last amnesty that will ever be offered, but in fact, if the nation gets into a policy of serial amnesty, which seems to be the direction that we are moving in, it's very difficult to believe that that would not have a significant effect on future behaviors on people seeking to come into the United States specifically to benefit from the next installment of the last amnesty we're ever going to give. And in particular, if you look at visa overstays, it's very difficult to imagine that people wouldn't be inclined to -- more likely to overstay on visas if they felt that they would, some time down the future, benefit from another installment of the last amnesty ever.

Another behavioral response here that I think is very important, and is a little subtle, is the impact of socially marginal families, of families that have a difficulty in educational attainment, families that have difficulty in marital structure and things like that -- the sort of compounding effect that these families have on each other. When you look at studies of the underclass or of the near-underclass, it's a continuum of behaviors. One of the things that we seem to know is that if you compound -- you compound the effect by putting large numbers of similar families together in a single neighborhood, and that seems to intensify the behavior. So that for example, if you add in additional families that have children out of wedlock, that have low educational attainment and other things like that into a neighborhood, I would argue that that probably intensifies those problems and increases a negative behavioral response of everyone in that neighborhood -- that there are compounding effects here, or tipping effects, that the more people that have essentially difficulty in maintaining normal middle-class status in our society, the more you bring in, the more in fact you are likely to intensify those problems within certain sections of the native-born population as well.

I also think that this report doesn't allude to -- but it's very difficult in my mind to believe that the very large number of low-skilled workers here coming in illegally has no significant effect on the wage levels of low-skilled native-born workers. That can be highly debated, but intuitively this would seem to be one effect. And therefore, by depressing -- if you are having the effect of depressing the wages of low-skilled native-born workers, then you are also incurring social and economic and fiscal costs on that population as well. I mean, there is a certain ambiguity in this of saying, ah, well, these workers are taking jobs that no one else will do, but on the other hand it's having no effect on wages. Those two things don't go together. It's one or the other. And if they are in fact taking jobs that no one else will do, then they also are undoubtedly keeping the wages of those jobs down well below where they would be otherwise.

There is also, I think, a political feedback effect to illegal immigration. Basically these figures would indicate, for example, that about one-tenth of the poor people in the United States are illegal non-citizens. Now, tomorrow the Census Bureau is going to be releasing the annual poverty estimate and the annual estimate of the medically uninsured in the United States, and I can assure you that no place in that release is it going go say that, by the way, one-tenth of the poor people we found here happen to be illegal people who weren't supposed to be in this country at all in the first place, or a similar percentage of the uninsured. And the very magnitude, in effect, of saying, ah, we have over 35 million poor people -- some of whom are illegals but that is never acknowledged -- is in fact used as an impetus for increasing spending in these programs in general and therefore has another indirect fiscal consequence as a result of it.

I think that if you look back on the debate about welfare and immigration over the last 10 years, there was a naive expectation among some conservatives that somehow you could do away with welfare, and that is simply not going to happen. For example, if you have individuals who are advocates of an open border, they would recognize that if you have open borders or semi-open borders and that you bring in large numbers of low-skilled people, they will take up welfare. But the response to that was, aha, well, let's do away with welfare. Well, guess what? It didn't happen, it shouldn't have happened, and it's not going to happen in the future. The welfare state is going to remain here. The welfare state can be somewhat more rational than it has been in the past but the fact of the matter is that it seems inevitable that we are going to have a system, and continue to have a system, where if you import large numbers of low-skilled individuals they are going to participate in that system and they are going to impose a direct fiscal cost on the government, and welfare reform will have no impact or very little impact in reducing that effect.

A couple of other points. I mentioned this is a federal-only study. My intuitive sense would be that if you added in state and local [data] that you would get considerably more fiscal -- negative fiscal impacts on that, and considerably larger.

And I also would reference -- I found very striking in here the statistics that close to one out of five individuals in federal prisons are illegal aliens, and as I understand it, they are not there for immigration-related offenses; they are there for some other type of substantive criminal offense. That is a very shocking figure. Now, I don't know whether that figure would correspond to state and local imprisonments, but let's say that at the state and local level the figure was not one in five but, say, one in ten -- we don't know what that number is but to throw a number out to kind of get an assessment of what this might mean. The total cost of crime -- economic costs as well as psychological costs -- is over $450 billion a year in the United States. So one-tenth of those costs, minimum you're talking $40, $50 billion there.

Now, again, we only know the federal imprisonment rate, we don't know the state and local rate, but if there is a disproportionate crime involvement by illegal immigrants and a disproportionate amount of crime being performed, then the social costs of that criminal activity would probably -- would almost certainly dwarf the smaller fiscal governmental impacts that are talked about in this report.

Overall I think this is a very good report. It seems to be quite accurate, and it is a very useful contribution to the public policy discussion.

MR. KRIKORIAN: Thank you, Bob.

Let me have Steve -- give Steve just a minute or two to address a couple of points that were brought up and then we'll throw it to the audience.

Steve?

MR. CAMAROTA: Yeah. On the question of how to assign costs, I should point out that the report tries to be as conservative as possible. Let me give one example. Federal aid to education goes primarily, though not exclusively, to low-income districts and low-income students, but in the report we just assigned costs for federal education based on the number of kids in each household, not controlling for the fact that those kids in illegal households obviously are disproportionately in low-income districts and are themselves very low-income. If we did that, the costs would rise very dramatically. So, in general, we try to have pretty conservative estimates of what -- you know, the costs.

Let me just touch very briefly on two other points. There probably is a significant effect on low-wage workers in the United States. As Mr. Rector pointed out, you can't have it both ways. You can't say that illegal aliens only take jobs Americans don't want but there is no impact on wages because the very implication of that is that the jobs are not allowed to rise in terms of their wages, benefits, and working conditions because you're constantly bringing in more unskilled immigrant workers to fill them. It seems certain that there is an effect, and I think there is a pretty good body of research to show that there is a significant effect. The National Academy of Sciences concluded that there was a significant negative effect; a new paper by George Borjas at Harvard suggests that the effect is even larger. So I think that that's pretty clear.

One final point to emphasize one more time: this study only deals with the federal effects. It's almost 100 percent certain that the fiscal costs at the state and local level are dramatically larger than they are at the federal level. Certainly, the National Academy of Science, when it looked at all immigrants who are unskilled immigrants, in general found that the costs at the state and local level were much larger. So our estimate that illegal aliens impose a net cost of over $10 billion at the federal level, you would have to almost certainly assume that's it much larger than that and add to it at the state and local level.

MR. KRIKORIAN: Thanks, Steve.

If you could please identify yourself and speak up when you ask your question.

Yes, sir.

Q: I'm Al Millikin (sp), affiliated with Washington Writers. Do we know what other countries supply or provide the most illegal immigrants to the U.S.? And when a nation like Mexico has taken the positions they have on our immigration policy, does this report show them acting in their best interest or would anything from this report cause another country to question what they are doing in relation to U.S. immigration, whether legal or illegal?

MR. CAMAROTA: Well, since the costs are primarily being borne by American taxpayers, we can pretty much count on the Mexican government not to put a very high priority on fiscal costs borne by American taxpayers, and that would generally be true of Central American governments and Caribbean governments as well where the most illegal aliens obviously come: Mexico, Central America and the Caribbean.

So in that sense, if they are very concerned about American taxpayers, sure, they might decide that, gee, you know, maybe we want to cooperate more with the United States, but my assumption is that that would not be the case because they have little incentive. These individuals who come to the United States are persons that those governments find it very difficult to provide employment and housing and a decent life for. So the fact that they are able to send a large share of their population to the United States to kind of get them off their hands is generally beneficial to the elites in those countries. That's why the elites in those countries very much like large-scale migration to the United States if for one thing it tends to get rid of people who are dissatisfied with the status quo in the country, and if you are a member of the ruling elite in Mexico or the Dominican Republic, that's a good thing. You want to shed your problems and put them on the U.S.

Now, in fairness, if the United States adopts policies that impose costs on their taxpayers, I'm sure we can be counted on to ignore that as well. Countries don't generally represent the interests of other -- you know, leaders of one country don't care very much about, and that's what we would expect from Mexico and all of their behavior suggests that they are not very concerned, and why would they be?

MR. KRIKORIAN: Yeah, go ahead.

MR. LOWELL: Just a quick addition. These are overwhelmingly poor people and certainly we shouldn't hold that against them, and the Mexican government, quite logically, and other governments from their point of view, aren't primarily concerned about how our social welfare plays out. They do express a lot of concern about the welfare of the population here nonetheless, and that's very reasonable, so the efforts with the matricula, the emphasis on trying to secure the rights of these workers is something where their attention tends to focus, and I don't think we should be surprised by that.

MR. CAMAROTA: But you wouldn't say that we necessarily share the same interests as those governments. Our interests might be quite different.

MR. LOWELL: I would say we should have at least the same interests in terms of the quality of life for these workers. Unfortunately, it's reasonable to say that both governments are a little bit lax in expressing that concern by letting the situation continue the way it is.

MR. KRIKORIAN: Julia?

Q: Two questions. If you could address what some people have said. First of all, in Social Security the report seems to indicate the benefits -- well, the one about the long haul, and another issue that some supporters of broader immigration have is that it's so essential to the economy, so if it essential to the economy then perhaps it's worth the amount we're paying in federal fiscal costs.

MR. CAMAROTA: Yeah. Let me address the Social Security. We estimate that illegal aliens pay $7 billion a year or more in Social Security and Medicare tax than they use in those services. The problem is they are a net benefit for those two programs of $7 billion but they are a net drain on everything else of $17 billion, creating the overall net drain of $10 billion. You can't just consider those two programs in isolation.

One minor point to consider is if we legalized illegal aliens, the vast majority, given their education levels and the kinds of jobs they do, would of course receive the earned income tax credit, but it must be pointed out that the earned income tax credit is specifically designed to cancel out the Social Security payments that people make. It's a way -- it's kind of like giving them their money back. In other words, whatever net benefit we get on Social Security, with legalization that would completely vanish just from the earned income tax credit. But as I said, their benefit right now as illegal aliens, not as amnesty recipients, is clear and unambiguous for Social Security and Medicare -- and those estimates are in the study -- but the net deficit everywhere else completely offsets that.

On the question of the economic impact of illegal aliens, the National Academy of Sciences of course devoted a lot of attention to the overall impact of immigration on the U.S. economy, and their conclusion was that it's very small, one way or the other. It mainly drives down wages for the poorest workers -- it depends on how you feel about that; we've expressed some concern on this panel about that, but putting that aside -- and it generally should help owners of capital and consumers. But the overall effect on the U.S. economy, according to the National Academy of Sciences, was between one and two-tenths of 1 percent of the total of U.S. economy in terms of what they thought the net benefit was. Now, if that's right -- I should tell you that their net benefit figure was $1 to $10 billion from immigration in an $8 trillion economy; so in other words, a very tiny percentage: one-tenth of 1 percent. Now, that was for all immigrants. Illegal aliens only make up about 25 percent of the foreign-born, so whatever that benefit is you'd have to assume it would be a small share of it.

In addition to that, of course, some more recent work done by Weinstein and Davis at Columbia suggests that there is no economic benefit, even the teeny tiny one estimated by the NRC, and their work suggested it's an entirely net loss for natives, somewhere in the tune of like $65 billion a year. So there isn't even any agreement that there is a net benefit. To the extent that people have found one, it seems to be trivial. George Borjas at Harvard, who is considered the top economist in the field, you know, summarizes a lot of this information in a book a few years back, and he basically says, as he has before, that any immigration's affect is trivial, and illegals are even a smaller portion of that, so it's less than trivial, whatever that is.

So I guess that's my answer. And I should point out, when the National Academy of Sciences did their estimates, they didn't even take into account the economic effects because they thought that the effects were so trivial they don't play out in terms of taxes or services.

And I guess one final point. If immigration drives down the wages of the poor, that has additional implications for their use of public services. That is, native-born workers are poorer because of the presence of illegal aliens; surely they must increase in some way their use of public services. We don't take that into account here, we just assume each group as they are given and don't measure the impact on wages on either.

MR. KRIKORIAN: Go ahead, please.

MR. RECTOR: As I understand this argument -- I mean, first of all you have to separate out different streams of immigration here -- very high-skilled individuals, very low-skilled individuals -- those will have different economic effects, they'll have different social and governmental effects. But in terms of low-skilled illegals, to say that they perform a vital function to our society, essentially what people seem to be arguing is, well, who is going to clean up the motel room if these women aren't there, or, who is going to be standing around taking up those day labor jobs? And in effect what you're saying is -- okay, let's take who is going to clean up the motel room? If the illegals weren't here, those jobs would go vacant and one would assume then that if this is such a big deal that you would have to see a very substantial increase in the wages of individuals in those occupations if the illegal immigrants weren't there taking up that slack.

If their economic contribution is so essential, then that would also imply that the wage effect of their presence would also be quite substantial. Most of my job, or half of my job, involves people continually, generally liberals, complaining about the wage levels of native-born Americans precisely in those jobs, you know. I can't go a single day without hearing an endless litany of complaints about welfare mothers getting off and making only $7 an hour and how this is terrible, and we have to do this, that and the other thing. And in that context to say, you know, one of the great benefits here of having illegal immigrants is that they keep the wages of these jobs down is just a bit paradoxical, and I think is a bit cynical.

It seems to me quite likely that the presence of large numbers of low-skilled illegals does keep wages in those industries down, and that has some negative effects on native-born. In particular, if you are a liberal theorist in these matters, you would say that these low wage levels are in fact the primary contributing factor to all the other behaviors that go with the dependent class, that the low wage levels reduce the probability of marriage and low wage levels inducing poverty cause educational failures of the kids and so forth. Now, I actually don't really subscribe to that, but it's a very -- almost prevalent view on the left wing of these debates. So if you were subscribed to those viewpoints then the negative effects -- potential negative effects on wages would be even more significant. I just generally think that saying that one of the good things that we get from this is that we pay low-skilled people less is not a particularly compelling argument.

MR. KRIKORIAN: Quick comment on that. Lindsay?

MR. LOWELL: Yeah, I think the basic question there is very interesting. Other than these direct costs that we know at least are measured to be adverse on the wages of other low-skilled workers, are there other positive things that these workers bring to the game? And I think the answer is clearly yes. On the one hand, also, while I like to look at the economic side of things, it's also the case that a confident and economically successful country should see welfare payments and benefits as part of its obligation to its population, so I don't have any problems with that except for the fact that the unauthorized population is a population that's here outside of the law and outside of what at least our upfront intentions are.

So from that point of view, the costs imposed are an issue that needs to be paid attention to. The economic plus of this population -- are they essential workers? In many sectors where these workers are found, technology could replace them in many sectors in which they are found. Consumers and businessmen could find other kinds of things to do, and raising wages is only one of the possibilities. But as a society we would have to make a choice as to what we want to do. Do we want people come and mow our lawns? Do we want people to keep our price of lettuce down? Would we be happy to live in a more European situation where our food costs were twice the share of our household income than they are now, and those kinds of things? But those choices are something we can make. Those are choices we can make.

MR. CAMAROTA: Well, let me just say -- I have great respect for Lindsay. On the question of food costs being twice, there's actually a fair amount of research on this question.

MR. LOWELL: I'm not going to debate that.

MR. CAMAROTA: Yeah. The impact of illegal aliens in the sector where they make up the largest share of workers is actually even there pretty small because the price of a head of a lettuce is determined by transportation -- packaging, food, herbicides, pesticides -- not the price of the person who picks it. In addition, of course, you could easily, in many cases, substitute technology. Businesses don't want to do that; they'd rather hire lots of unskilled workers and pass the costs onto taxpayers, but there's a lot of reasons why we might not want them to do that. In general, however, the impact of illegal immigration on prices has to be small, even though it could have a significant impact on low-wage workers because low-wage workers overall account for such a tiny fraction of total economic output, even in agriculture.

MR. KRIKORIAN: Jerry? If you could identify yourself.

Q: Yeah, Jerry Kammer with the Copley News Service.

Steve, CIS has long criticized the federal government for lack of interior enforcement of immigration laws, but is it possible that someone could read your report not only as an indictment of federal immigration policy but also as an indictment of federal labor policy, which has created a system of minimum wages and lack of enforcement at the worksite, that according to your report would keep people still among the working poor even after they are legalized? Are you also indicting the wage system and the labor system?

MR. CAMAROTA: Well, I'm pretty sympathetic to arguments about a higher minimum wage and greater regulation, but it isn't going to change the basic fact that in the modern American economy a person with an eighth-grade education whose language skills are limited in English is going to earn very low wages though they might be at a much higher minimum wage.

So I am sympathetic to that argument. It would be -- my sense is if you want to improve the lives of America's working poor or low-income workers in general, what you would want to do is -- two sides of that equation. You probably would want to do things like raise the minimum wage and at the same time reduce the number of unskilled immigrant workers in the United States. The beauty of doing that is that one disadvantage of raising the minimum wage is that it would tend to increase unemployment -- how much is a matter of debate -- because you've raised the costs, but if you reduce the number of workers at the same time by reducing immigration, then it's much easier to raise the minimum wage without fear that you're going to cause some significant number of workers to be laid off.

So I would say that this could really -- if you really wanted to make a difference for unskilled workers in the U.S., you would want to do things both with our immigration policy and with regulation of the labor market.

MR. RECTOR: If I could -- basically workers are paid at their level of marginal productivity. Less than 2 percent of American workers are paid at the minimum wage. And what we have as a national policy is a very clear policy that says if you are a low-skilled worker and you're trying to support a family, we will massively support you in that effort. We will provide you with close to $4,000 a year in the earned income tax credit. That's a refund; it's a check in the mail. We will give you food stamps. We will provide you with Medicaid. In some cases we'll provide you with public housing. That is a policy that we've developed as a nation, recognizing that there isn't any magic pill that suddenly takes low-skilled workers and makes them into highly productive workers.

Now, the issue here is whether the presence of low-skilled immigrants is pushing somewhat more people -- or suppressing the wages of people, say, making between $5 and $8 an hour. We don't really know the answer to that, but the social -- I mean, it certainly isn't having a helpful affect on those wages, and, you know, I don't really think that we have a deficit as a nation because we have recognized that we have low-skilled workers, we do have very extensive policies to try to support them by supplementing their wages, so that's really the core of the welfare state now.

MR. CAMAROTA: Good point.

MR. RECTOR: And the question is, you know, do we have a sort of unlimited obligation to provide those supports to anyone who manages to get in this country legally or illegally? That's really the question here.

MR. KRIKORIAN: Quick follow up?

Q: We're talking about the fiscal burden, but obviously there are winners here, so thousands of employers are clamoring for more immigrants to come up at the low-wage levels. Do we have a system here that privatizes the gain in the hands of the employer class and middle classes and socializes the loss across the American taxpayer base?

MR. CAMAROTA: Oh, absolutely. I mean, just put simply, it's a great deal for employers if the government will provide health care to their workers. It's a great deal to employers if the government will send them a check in the form of the additional child tax credit, which illegals explicitly can get and the EITC, which relatively few get now but would if they got legal status.

It's a great deal for employers if the government will provide workers with housing and food stamps, and that's exactly what we're doing. And one of the problems with illegal immigration is that the benefit to the employer is narrowly focused and obvious, but the cost is diffuse, borne by all taxpayers, and as a consequence, in politics, whenever you have a narrowly focused benefit, it typically trumps the diffuse costs, even though the diffused costs may be much larger than the narrowly focused benefit, and that's what we're seeing. That's why we get so much illegal immigration; that coupled with the fact that the people who are harmed by illegal immigration in the labor market are poor and relatively powerless as well.

MR. KRIKORIAN: Let's take one more question if anyone has one, otherwise -- yes, sir. Yes.

Q: I'm (off mike) Rodriguez, Department of Economics, University of -- (off mike).

Well, for me it's a little bit shocking, all of these results. I see that both these are point estimates that you are doing. I think we have to be very careful about the real cost, the real benefits, especially when talking about immigration because the demographic (off mike) of immigration is very different to the locals. They start working younger. They die younger. Maybe there is some benefit from Social Security that we're not taking into account. (Off mike.)

About local labor markets, we have to be very careful because the labor market is segmented. We're talking about illegal and legal, and we're talking about how the illegal immigrants are going to affect the poor people and those guys who are earning low wages. And it's not very clear for me; it's not straightforward to see that the effect, low-wage people, because we're talking about segmented markets.

MR. KRIKORIAN: Okay, let me get some reaction, because we're running over.

Go ahead, Steve.

MR. CAMAROTA: Yeah, real quick. I mean, in terms of long term, the National Academy of Science did do a whole series of estimates on long term and found that unskilled workers, in the course of their lifetime, impose so many fiscal costs that even over multiple generations they never pay that back. Now, again, the National Research Council is an important organization, but you can make of that what you will.

On the question of job competition, my own research -- I have tried to look at unskilled natives and compare them to illegal aliens, and obviously most Americans are not in competition with illegal aliens, but the roughly 10 million native-born Americans who lack a high school education and the roughly 2 or 3 million legal immigrants who lack a high school education, they do seem to do exactly the same kinds of jobs, which are, you know, they clean hotel rooms, they clean pools, they are nannies, they are construction laborers. And of course, what else would they be?

So my sense is they are in competition, not with most Americans but unfortunately they are in competition with the poorest Americans, but that's also a reason why it's so appealing to the middle class because they can say, look, we're helping this guy out, we're letting him here, he's working, and he's taking a job nobody wants. What the college-educated American really means is, he's taking a job I don't want, and that's always nice when you can give somebody else's job away.

MR. KRIKORIAN: Let's wrap it up here. The panelists may well be willing to be accosted afterwards by anybody who has further questions. Let me reiterate, the report is online in its entirety at our website. And thanks for coming.