Amnesty and Continued Low Skill Immigration
Will Substantially Raise Welfare Costs and Poverty

By Robert Rector
Posted May 29, 2006

Originally published by the Heritage Foundation.
Republished with permission.

Original URL:

Backgrounder #1936

May 12, 2006

This paper focuses on the net fiscal effects of immigration with particular emphasis on the fiscal effects of low skill immigration. The fiscal effects of immigration are only one aspect of the impact of immigration. Immigration also has social, political, and economic effects. In particular, the economic effects of immigration have been heavily researched with differing results. These economic effects lie beyond the scope of this paper.

Overall, immigration is a net fiscal positive to the government’s budget in the long run: the taxes immigrants pay exceed the costs of the services they receive. However, the fiscal impact of immigrants varies strongly according to immigrants’ education level. College-educated immigrants are likely to be strong contributors to the government’s finances, with their taxes exceeding the government’s costs. By contrast, immigrants with low education levels are likely to be a fiscal drain on other taxpayers. This is important because half of all adult illegal immigrants in the U.S. have less than a high school education. In addition, recent immigrants have high levels of out-of-wedlock childbearing, which increases welfare costs and poverty.

An immigration plan proposed by Senators Mel Martinez (R-FL) and Chuck Hagel (R-NE), the Comprehensive Immigration Reform Act (CIRA, S. 2611) would provide amnesty to 9 to 10 million illegal immigrants and put them on a path to citizenship. Once these individuals become citizens, the net addi­tional cost to the federal government of benefits for these individuals will be around $16 billion per year. Further, once an illegal immigrant becomes a citizen, he has the right to bring his parents to live in the U.S. The parents, in turn, may become citi­zens. The long-term cost of government benefits to the parents of 10 million recipients of amnesty could be $30 billion per year or more. In the long run, S. 2611, if enacted, would be the largest expansion of the welfare state in 35 years.

Current Trends in Immigration

Over the last 40 years, immigration into the United States has surged. Our nation is now expe­riencing a second “great migration” similar to the great waves of immigrants that transformed America in the late 19th and early 20th centuries. In 2004, an estimated 35.7 million foreign-born per­sons lived in the U.S. While in 1970 one person in twenty was foreign born, by 2004 the number had risen to one in eight.

About one-third of all foreign-born persons in the U.S. are illegal aliens. There are between 10 and 12 million illegal aliens currently living in the U.S.[1] Illegal aliens now comprise 3 to 4 percent of the total U.S. population. Each year approximately 1.3 million new immigrants enter the U.S.[2] Some 700,000 of these entrants are illegal.[3]

One-third of all foreign-born persons in the U.S. are Mexican. Overall, the number of Mexicans in the U.S. has increased from 760,000 in 1970 to 10.6 million in 2004. Nine percent of all Mexicans now reside in the U.S.[4] Over half of all Mexicans in the U.S. are illegal immigrants,[5] and in the last decade 80 to 85 percent of the inflow of Mexicans into the U.S. has been illegal.[6]

The public generally perceives illegals to be unattached single men. This is, in fact, not the case. Some 44 percent of adult illegals are women. While illegal men work slightly more than native-born men, illegal women work less. Among female illegals, some 56 percent work, compared to 73 per­cent among native-born women of comparable age.[7] As well, Mexican women emigrating to the U.S. have a considerably higher fertility rate than women remaining in Mexico.[8]

Immigrants and Education

On average, immigrants have low education lev­els relative to native-born U.S. citizens. One-quarter of legal adult immigrants lack a high school degree, compared to 9 percent among the native-born population. However, there is a well educated sub-group within the legal immigrant population. Some 32 percent of legal immigrant adults have a college degree, compared to 30 percent of native-born adults.[9]

The education levels of illegal aliens are lower than those of legal immigrants. Half of all adult ille­gal immigrants lack a high school degree.[10] Among Latin American and Mexican immigrants, 60 per­cent lack a high school degree and only 7 percent have a high school diploma. By contrast, among native-born workers in the U.S., only 6 percent have failed to complete high school degrees and nearly a third have a college degree.[11]

Decline in Immigrant Wages

Over the last 40 years the education level of new immigrants has fallen relative to the native popula­tion. As the relative education levels of immigrants have declined, so has their earning capacity com­pared to the general U.S. population. Immigrants arriving in the U.S. around 1960 had wages, at the time of entry, that were just 13 percent less than natives’. In 1965, the nation’s immigration law was dramatically changed, and from 1990 on, illegal immigration surged. The result was a decline in the relative skill levels of new immigrants. By 1998, new immigrants had an average entry wage that was 34 percent less than natives’.[12] Because of their lower education levels, illegal immigrants’ wages would have been even lower.

The low-wage status of recent illegal immigrants can be illustrated by the wages of recent immi­grants from Mexico, a majority of whom have entered the U.S. illegally. In 2000, the median weekly wage of a first-generation Mexican immi­grant was $323. This was 54 percent of the corre­sponding wage for non-Hispanic whites in the general population.[13]

Historically, the relative wages of recent immi­grants have risen after entry as immigrants gained experience in the labor market. For example, immigrants who arrived in the U.S. in the 1960s and 1970s saw their relative wages rise by 10 per­centage points compared to natives’ wages during their first 20 years in the country. But in recent years, this modest catch up effect has diminished. Immigrants who arrived in the late 1980s actually saw their relative wages shrink in the 1990s.[14]

Immigration and Welfare Dependence

Welfare may be defined as means-tested aid pro­grams: these programs provide cash, non-cash, and social service assistance that is limited to low-income households. The major means-tested pro­grams include Food Stamps, Temporary Assistance to Needy Families, public housing, the earned income credit, and Medicaid. Historically, recent immigrants were less likely to receive welfare than native-born Americans. But over the last thirty years, this historic pattern has reversed. As the rel­ative education levels of immigrants fell, their ten­dency to receive welfare benefits increased. By the late 1990s immigrant households were fifty percent more likely to receive means-tested aid than native-born households.[15] Moreover, immigrants appear to assimilate into welfare use. The longer immi­grants live in the U.S., the more likely they are to use welfare.[16]

A large part, but not all, of immigrants’ higher welfare use is explained by their low education lev­els. Welfare use also varies by immigrants’ national origin. For example, in the late 1990s, 5.6 percent of immigrants from India received means-tested benefits; among Mexican immigrants the figure was 34.1 percent; and for immigrants from the Domin­ican Republic the figure was 54.9 percent.[17] Ethnic differences in the propensity to receive welfare that appear among first-generation immigrants persist strongly in the second generation.[18] The relatively high use of welfare among Mexicans has significant implications for current proposals to grant amnesty to illegal immigrants.

Chart 1

Some 80 percent of illegal immigrants come from Mexico and Latin America.[19] (See Chart 1) Histori­cally, Hispanics in America have had very high levels of welfare use. Chart 2 shows receipt of aid from major welfare programs by different ethnic groups in 1999; the programs covered are Medicaid, Food Stamps, public housing, Temporary Assistance to Needy Families, General Assistance, and Supple­mental Security Income.[20] As the chart shows, His­panics were almost three times more likely to receive welfare than non-Hispanic whites. In addition, among families that received aid, the cost of the aid received was 40 percent higher for Hispanics than for non-Hispanic whites.[21] Putting together the greater probability of receiving welfare with the greater cost of welfare per family means that, on average, Hispanic families received four times more welfare per family than white non-Hispanics.

Part, but not all, of this high level of welfare use by Hispanics can be explained by background fac­tors such as family structure. [22] It seems likely that, if Hispanic illegal immigrants are given permanent residence and citizenship, they and their children will likely assimilate into the culture of high welfare use that characterizes Hispanics in the U.S. This would impose significant costs on taxpayers and society as a whole.

Chart 2

Welfare use can also be measured by immigra­tion status. In general, immigrant households are about fifty percent more likely to use welfare than native-born households.[23] Immigrants with less education are more likely to use welfare.

Chart 3

The potential welfare costs of low-skill immigra­tion and amnesty for current illegal immigrants can be assessed by looking at the welfare utiliza­tion rates for current low-skill immigrants. As Chart 3 shows, immigrants without a high school degree (both lawful and unlawful) are two-and-a-half times more likely to use welfare than native-born individuals.[24] This underscores the high potential welfare costs of giving amnesty to illegal immigrants.

All categories of high school dropouts have a high utilization of welfare. Immigrants who have less than a high school degree are slightly more likely to use welfare than native-born dropouts. Legal immi­grants who are high school dropouts are slightly more likely to use welfare than native-born drop­outs.[25] Illegal immigrant dropouts, however, are less likely to use welfare than native-born dropouts mainly because they are ineligible for many welfare programs. With amnesty, current illegal immigrants’ welfare use would likely rise to the level of lawful immigrants with similar education levels.

Illegal Immigration and Poverty

According to the Pew Hispanic Center, 4.7 mil­lion children of illegal immigrant parents currently live in the U.S.[26] Some 37 percent of these children are poor.[27] While children of illegal immigrant par­ents comprise around 6 percent of all children in the U.S., they are 11.8 percent of all poor children.[28]

This high level of child poverty among illegal immigrants in the U.S. is, in part, due to low edu­cation levels and low wages. It is also linked to the decline in marriage among Hispanics in the U.S. Within this group, 45 percent of children are born out of wedlock.[29] (See Table 1.) Among foreign-born Hispanics the rate is 42.3 percent.[30] By con­trast, the out-of-wedlock birth rate for non-His­panic whites is 23.4 percent.[31] The birth rate for Hispanic teens is higher than for black teens.[32] While the out-of-wedlock birth rate for blacks has remained flat for the last decade, it has risen steadily for Hispanics.[33] These figures are important because, as noted, some 80 percent of illegal aliens come from Mexico and Latin America.[34]

Table 1

In general, children born and raised outside of marriage are seven times more likely to live in poverty than chil­dren born and raised by married cou­ples. Children born out of wedlock are also more likely to be on welfare, to have lower educational achievement, to have emotional problems, to abuse drugs and alcohol, and to become involved in crime.[35]

Poverty is also more common among adult illegal immigrants, who are twice as likely to be poor as are native-born adults. Some 27 percent of all adult ille­gal immigrants are poor, compared to 13 percent of native-born adults.[36]

Economic and Social Assimilation of Illegal Immigrant Offspring

One important question is the future economic status of the children and grandchildren of current illegal immigrants, assuming those offspring remain in the U.S. While we obviously do not have data on future economic status, we may obtain a strong indi­cation of future outcomes by examining the educa­tional attainment of offspring of recent Mexican immigrants. Some 57 percent of current illegal immigrants come from Mexico, and about half of Mexicans currently in the U.S. are here illegally.[37]

First-generation Mexican immigrants are individ­uals born in Mexico who have entered the U.S. In 2000, some 70 percent of first-generation Mexican immigrants (both legal and illegal) lacked a high school degree. Second-generation Mexicans may be defined as individuals born in the U.S. who have at least one parent born in Mexico. Second-generation Mexican immigrants (individuals born in the U.S. who have at least one parent born in Mexico) have greatly improved educational outcomes but still fall well short of the general U.S. population. Some 25 percent of second-generation Mexicans in the U.S. fail to complete high school. By contrast, the high school drop out rate is 8.6 percent among non-His­panic whites and 17.2 percent among blacks. Criti­cally, the educational attainment of third-generation Mexicans (those of Mexican ancestry with both par­ents born in the U.S.) improves little relative to the second generation. Some 21 percent of third-gener­ation Mexicans are high school dropouts.[38] Simi­larly, the rate of college attendance among second-generation Mexicans is lower than for black Ameri­cans and about two-thirds of the level for non-His­panic whites; moreover, college attendance does not improve in the third generation.[39]

These data indicate that the offspring of illegal Hispanic immigrants are likely to have lower rates of educational attainment and higher rates of school failure compared to the non-Hispanic U.S. population. High rates of school failure coupled with high rates of out-of-wedlock childbearing are strong predictors of future poverty and welfare dependence.

Immigration and Crime

Historically, immigrant populations have had lower crime rates than native-born populations. For example, in 1991, the overall crime and incarcera­tion rate for non-citizens was slightly lower than for citizens. Strikingly, imprisonment for violent crime was half as likely for non-citizens as for citizens.[40]

On the other hand, the crime rate among His­panics in the U.S. is high. Age-specific incarcera­tion rates (prisoners per 100,000 residents in the same age group in the general population) among Hispanics in federal and state prisons are two to two-and-a-half times higher than among non-His­panic whites.[41] Relatively little of this difference appears to be due to immigration violations.[42]

Illegal immigrants are overwhelmingly Hispanic. It is possible that, over time, Hispanic immigrants and their children may assimilate the higher crime rates that characterize the low-income Hispanic popula­tion in the U.S. as a whole.[43] If this were to occur, then policies that would give illegal immigrants per­manent residence through amnesty, as well as poli­cies which would permit a continuing influx of hundreds of thousands of illegal immigrants each year, would increase crime in the long term.

The Fiscal Impact of Immigration

One important question is the fiscal impact of immigration (both legal and illegal). Policymakers must ensure that the interaction of welfare and immi­gration policy does not expand the welfare-depen­dent population, which would hinder rather than help immigrants and impose large costs on American society. This means that immigrants should be net contributors to government: the taxes they pay should exceed the cost of the benefits they receive.

In calculating the fiscal impact of an individual or family, it is necessary to distinguish between public goods and private goods. Public goods do not require additional spending to accommodate new residents.[44] The clearest examples of government public goods are national defense and medical and scientific research. The entry of millions of immi­grants will not raise costs or diminish the value of these public goods to the general population.

Other government services are private goods; use of these by one person precludes or limits use by another. Government private goods include direct personal benefits such as welfare, Social Security benefits, Medicare, and education. Other govern­ment private goods are “congestible” goods.[45] These are services that must be expanded in proportion to the population. Government congestible goods include police and fire protection, roads and sewers, parks, libraries, and courts. If these services do not expand as the population expands, there will be a decrease in the quality of service.

An individual makes a positive fiscal contribu­tion when his total taxes paid exceed the direct benefits and congestible goods received by himself and his family.[46]

The Fiscal Impact of Low Skill Immigration

The 1997 New Americans study by the National Academy of Sciences (NAS) examined the fiscal impact of immigration.[47] It found that, within in a single year, the fiscal impact of foreign-born house­holds was negative in the two states studied, New Jersey and California.[48] Measured over the course of a lifetime, the fiscal impact of first-generation immigrants nationwide was also slightly nega­tive.[49] However, when the future earnings and taxes paid by the offspring of the immigrant were counted, the long-term fiscal impact was positive. One commonly cited figure from the report is that the net present value (NPV) of the fiscal impact of the average recent immigrant and his descendents is $83,000.[50]

There are five important caveats about the NAS longitudinal study and its conclusion that in the long term the fiscal impact of immigration is posi­tive. First, the study applies to all recent immigra­tion, not just illegal immigration. Second, the finding that the long-term fiscal impact of immigra­tion is positive applies to the population of immi­grants as a whole, not to low-skill immigrants alone. Third, the $83,000 figure is based on the predicted earnings, tax payments, and benefits of an immigrant’s descendents over the next 300 years.[51] Fourth, the study does not take into account the growth in out-of-wedlock childbearing among the foreign-born population, which will increase future welfare costs and limit the upward mobility of future generations. Fifth, the assumed educational attainment of the children, grandchil­dren, and great grandchildren of immigrants who are high school dropouts or high school graduates seems unreasonably high given the actual attain­ment of the offspring of recent Mexican and His­panic immigrants.[52]

The NAS study’s 300-year time horizon is highly problematic. Three hundred years ago, the United States did not even exist and British colonists had barely reached the Appalachian Mountains. We cannot reasonably estimate what taxes and benefits will be even 30 years from now, let alone 300.

The NAS study assumes that most people’s descendents will eventually regress to the social and economic mean, and thus may make a positive fiscal contribution, if the time horizon is long enough. With similar methods, it seems likely that out-of-wedlock childbearing could be found to have a net positive fiscal value as long as assumed future earnings are projected out 500 or 600 years.

Slight variations to NAS’s assumptions greatly affect the projected outcomes. For example, limiting the time horizon to 50 years and raising the assumed interest rate from 3 percent to 4 percent drops the NPV of the average immigrant from around $80,000 to $8,000.[53] Critically, the NAS projections assumed very large tax increases and benefits cuts would begin in 2016 to prevent the federal deficit from rising further relative to GDP. This assumption makes it far easier for future gen­erations to be scored as fiscal contributors. If these large tax hikes and benefit cuts do not occur, then the long-term positive fiscal value of immigration evaporates.[54] Moreover, if future tax hikes and ben­efit cuts do occur, the exact nature of those changes would likely have a large impact on the findings; this issue is not explored in the NAS study.

Critically, the estimated net fiscal impact of the whole immigrant population has little bearing on the fiscal impact of illegal immigrants, who are pri­marily low-skilled. As noted, at least 50 percent of illegal immigrants do not have a high school degree. As the NAS report states, “[S]ome groups of immigrants bring net fiscal benefits to natives and others impose net fiscal costs… [I]mmigrants with certain characteristics, such as the elderly and those with little education, may be quite costly.”[55]

The NAS report shows that the long-term fiscal impact of immigrants varies dramatically accord­ing to the education level of the immigrant. The fiscal impact of immigrants with some college edu­cation is positive. The fiscal impact of immigrants with a high school degree varies according to the time horizon used. The fiscal impact of immigrants without a high school degree is negative: benefits received will exceed taxes paid. The net present value of the future fiscal impact of immigrants without a high school degree is negative even when the assumed earnings and taxes of descen­dents over the next 300 years are included in the calculation.[56]

A final point is that the NAS study’s estimates assume that low skill immigration does not reduce the wages of native-born low-skill workers. If low-skill immigration does, in fact, reduce the wages of native-born labor, this would reduce taxes paid and increase welfare expenditures for that group. The fiscal, social, and political implications could be quite large.

The Cost of Amnesty

Federal and state governments currently spend over $500 billion per year on means-tested welfare benefits.[57] Illegal aliens are ineligible for most fed­eral welfare benefits but can receive some assis­tance through programs such as Medicaid, In addition, native-born children of illegal immigrant parents are citizens and are eligible for all relevant federal welfare benefits.

Granting amnesty to illegal aliens would have two opposing fiscal effects. On the one hand, it may raise wages and taxes paid by broadening the labor market individuals compete in; it would also increase tax compliance and tax receipts as more work would be performed “on the books,”[58] On the other hand, amnesty would greatly increase the receipt of welfare, government benefits, and social services. Because illegal immigrant households tend to be low-skill and low-wage, the cost to gov­ernment could be considerable.

The Center for Immigration Studies (CIS) has performed a thorough study of the federal fiscal impacts of amnesty.[59] This study found that illegal immigrant households have low education levels and low wages and currently pay little in taxes. Ille­gal immigrant households also receive lower levels of federal government benefits. Nonetheless, the study also found that, on average, illegal immigrant families received more in federal benefits than they paid in taxes. [60]

Granting amnesty would render illegal immi­grants eligible for federal benefit programs. The CIS study estimated the additional taxes that would be paid and the additional government costs that would occur as a result of amnesty. It assumed that welfare utilization and tax payments among current illegal immigrants would rise to equal the levels among legally-admitted immigrants of similar national, educational, and demographic back­grounds. If all illegal immigrants were granted amnesty, federal tax payments would increase by some $3,000 per household, but federal benefits and social services would increase by $8,000 per household. Total federal welfare benefits would reach around $9,500 per household, or $35 billion per year total. The study estimates that the net cost to the federal government of granting amnesty to some 3.8 million illegal alien households would be around $5,000 per household, for a total federal fiscal cost of $19 billion per year. [61]

Amnesty and the Hagel/Martinez Bill

Senators Mel Martinez (R-FL) and Chuck Hagel (R-NE) have proposed the Comprehensive Immi­gration Reform Act (CIRA) to offer amnesty and cit­izenship to current illegal aliens (S. 2611). This plan would offer amnesty and citizenship to between 60 and 85 percent of the nation’s current 11.9 million illegal immigrants.

Under the plan, illegal immigrants who have been in the U.S. five years or more (60 percent of the total) would be granted immediate amnesty. Illegal immigrants who have been in the country between two and five years (25 percent of the total) would travel to one of 16 “ports of entry” where they would receive work permits that would bestow permanent residence and allow the bearers to become citizens. Overall, the plan is likely to grant citizenship to 85 percent of the cur­rent illegal alien population, or some 9 to 10 mil­lion individuals.

As noted, illegal aliens in the U.S. have very low education levels: at least half lack a high school education and a third have less than a ninth grade education. Illegal immigrants earn low wages simi­lar to the wages of other low-skill workers in the economy. This means they are prone to poverty and welfare dependence.

Illegal immigrants are currently ineligible for most federal welfare benefits. Granting citizenship would provide eligibility to welfare programs such as the Earned Income Credit, Food Stamps, Medicaid, and Temporary Assistance to Needy Families. This would greatly increase welfare costs. The added government costs can be esti­mated by assessing government benefits and tax payments among current illegal immigrants com­pared to government benefits and tax payments among legal immigrants of similar national and educational backgrounds. This comparison shows that granting citizenship to 85 percent of current illegal immigrants would increase net federal fis­cal costs by some $16 billion per year. Granting citizenship to 60 percent of current illegal immi­grants would increase welfare costs by some $11.4 billion per year.[62]

These costs would not occur immediately. The CIRA plan imposes a prospective six-year waiting period prior to granting legal permanent residence to illegal immigrants. Individuals would wait another five years after receiving permanent resi­dence before becoming citizens. Thus, much of the cost of the plan might be delayed; however, once millions of individuals are put on the path to citi­zenship there would be enormous (and probably irresistible) political pressure to grant them the same benefits that are available to the general pop­ulation quickly, rather than enforce a long delay.

In addition, the cost estimates presented above are based on a static analysis that assumes that amnesty will not alter behavior. In reality, illegal immigrants are likely to have significantly more children once they are permanently settled in the U.S. These children will increase welfare costs and child poverty further..

Family Chain Migration

The impact and cost of CIRA would extend well beyond the ten million or so individuals initially granted amnesty. When an individual is granted citi­zenship, he is given the unrestricted right to bring his spouse, minor children, and parents into the country. Each of these individuals would have the right to become a citizen after he or she has lived in the coun­try five years. Thus, each individual granted amnesty under CIRA could bring five or more additional immigrants, all of whom could become citizens.

As noted, many of the individuals who would be granted amnesty under the amendment have fami­lies abroad. Illegal immigrants granted permanent residence would have the immediate right to bring spouses and minor children into the country. Once here, the spouses and children would receive gov­ernment services and have the right to become cit­izens. The total number of foreign-born persons who would ultimately be granted citizenship under S. 2611 could be far more than 10 million, and if so, government costs would swell far above the $16 billion figure given above.

But the fiscal problem gets worse; when an illegal immigrant has obtained citizenship through the amnesty process, he or she would have the right to bring his or her parents in the U.S. as permanent lawful residents. (Currently one-tenth of the annual flow of legal immigrants to the U.S. are par­ents of recent immigrants who have naturalized.) If ten million current illegal immigrants were granted amnesty and citizenship under CIRA, as many as twenty million foreign born parents would be given the right to immigrate to the U.S. Once in the U.S., the immigrant parents would receive social services and government funded medical care, much of it paid for through the Medicaid disproportionate share program.

These immigrant parents coming to the U.S. would also be eligible to apply for citizenship themselves. On attaining citizenship, most would become eligible for benefits from the Supplemental Security Income (SSI) and Medicaid programs, at an average cost of over $18,000 per person per year. While it is true that the language require­ments of the citizenship test would serve as a bar­rier to immigrant parents becoming citizens, the tests are not very difficult and the financial rewards of citizenship would be very great. If only ten per­cent of the parents of those receiving amnesty under CIRA became citizens and enrolled in SSI and Medicaid, the extra costs to government would be over $30 billion per year.

Obviously, these costs would not begin for some time, but the long-term potential of amnesty to raise government spending is quite real.

While no one can predict how many spouses, children, and parents of the beneficiaries of amnesty would enter the country, the pool of those who could enter is enormous, and the potential long-term government costs would be staggering.

Granting Amnesty is Likely to Further Increase Illegal Immigration

The Immigration Reform and Control Act (IRCA) of 1986 granted amnesty to 2.7 million ille­gal aliens. The primary purpose of the act was to decrease the number of illegal immigrants by limit­ing their inflow and by legalizing the status of ille­gal immigrants already here.[63] In fact, the act did nothing to stem the tide of illegal entry. The num­ber of illegal aliens entering the country increased fivefold from around 140,000 per year in the 1980s to 700,000 per year today.

Illegal entries increased dramatically shortly after IRCA went into effect. It seems plausible that the prospect of future amnesty and citizenship served as a magnet to draw even more illegal immi­grants into the country. After all, if the nation granted amnesty once why wouldn’t it do so again?

S. 2611 would repeat IRCA on a much larger scale. This time, nine to ten million illegal immi­grants would be granted amnesty. As with IRCA, the bill promises to reduce future illegal entry but contains little policy that would actually accom­plish this. The granting of amnesty to 10 million illegal immigrants is likely to serve as a magnet pulling even greater numbers of aliens into the country in the future.

If enacted, the legislation would spur further increases in the future flow of low-skill migrants. This in turn would increase poverty in America, enlarge the welfare state, and increase social and political tensions.

Permanent “Guest Worker” Program

Finally, CIRA would issue 325,000 new visas per year to “guest workers.” The number of visas available could increase by 20 percent annually, reaching two million per year within ten years. By 2017, the guest worker program would have admitted some eight million new workers. Illegal aliens who have been in the country for less than two years would be eligible to become guest workers and would probably be the primary recipients of these supposedly temporary (H2C) visas. Recipients of these visas could bring spouses and children into the country immedi­ately, increasing the number of entrants over ten years well above eight million. Because nearly all of the guest workers and their families would within a few years become eligible for government welfare and other services, the fiscal costs from the program could rival those stemming from the direct amnesty provisions of the bill.

On the surface, individuals in the guest worker program would be limited to a six-year stay in the U.S. But they would have the option to convert to legal permanent residence (LPR) after four years. This would make them permanent residents with the right to naturalize. In addition, all children born to guest workers would automatically become U.S. citizens. This would make it very unlikely that the parent would ever be forced to leave the country.

As structured, the CIRA guest worker program could, within a decade, double the inflow of legal permanent immigrants into the U.S. Many or most of these immigrants would be low-skill and would thus impose fiscal costs on U.S. taxpayers. It is true that many employers would benefit from addi­tional low-skill laborers; however, if such laborers are granted citizenship and permanent residence, their employment is likely to generate negative externalities that impose costs on the rest of society. A guest worker program that, in fact, provides per­manent residence and citizenship would not be beneficial to the nation’s finances.

Policy Implications

Immigration to the U.S. is a privilege, not a right. Immigrants should be net contributors to the gov­ernment and society and should not be a fiscal bur­den on American society. While highly educated immigrants, on average, make positive fiscal contri­butions, the overall fiscal impact of low-skill immi­grants is negative.

Over the last 20 years, around 10 million indi­viduals without a high school degree have entered the United States. Many of these also have a high probability of out-of-wedlock childbearing, a key predictor of poverty and welfare dependence. Unless U.S. immigration policy is changed, these trends are likely to continue. Granting amnesty to current illegal immigrants exacerbates the problem.

Sound immigration policy should be based on two principles. The first is respect for the rule of law. American citizens should determine who is allowed to enter the country, to become a citizen, and to vote in our elections. Lax border enforce­ment and the non-enforcement of laws against employing illegal immigrants have encouraged over 10 million individuals to enter the country unlawfully. Past and pending amnesties reward this behavior. Under the current system, decisions about who will live in the U.S. and who will become a citizen tend to be made unilaterally by foreigners. S. 2611 would further undermine the rule of law and put the U.S. on the path of uncon­trolled immigration punctuated by recurring amnesties.

Second, recognizing the fact that low-skill immi­grants are likely to be a fiscal burden on society, gov­ernment should increase the average skill and education levels of incoming immigrants. Currently, the average skill level of immigrants is significantly reduced by two factors: largely uncontrolled border crossings and the high priority on kinship ties in the issuance of permanent residence visas. Only 7.6 per­cent of individuals granted visas for permanent entry into the U.S. are selected on the basis of their educa­tional attainment and skills.[64] To the increase the skill levels of future immigrants, the U.S. should stop the inflow of illegal immigrants, reduce the number of family reunification visas, and increase the number of employment- and skill-based visas.

Five specific policies follow from these principles:

  1. The influx of illegal immigrants should be stopped by rigorous border security programs and strong programs to prevent employers from employing illegals.

  2. Amnesty and citizenship should not be given to current illegal immigrants. Amnesty has nega­tive fiscal consequences and is manifestly unfair to those who have waited for years to enter the country lawfully. Amnesty would also serve as a magnet, drawing even more future illegal immigration.

  3. Any guest worker program should grant tem­porary, not permanent, residence and should not be a pathway to citizenship. A guest worker program should not disproportionately swell the ranks of low-skill workers.

  4. Children born to parents who are illegal immi­grants or to future guest workers should not be given citizenship status. Granting citizen­ship automatically confers welfare eligibility and makes it unlikely the parent will ever leave the U.S.[65]

  5. The legal immigration system grants lawful permanent residence to some 950,000 per­sons each year. This system should be altered to substantially increase the proportion of new entrants with high levels of education and skills in demand by U.S. firms. Under current law, foreign-born parents and siblings of natu­ralized citizens are given preference for entry visas. The current visa allotments for family members (other than spouses and minor chil­dren) should be eliminated, and quotas for employment- and skill-based entry increased proportionately.

Robert Rector is Senior Research Fellow in Domestic Policy Studies at The Heritage Foundation.


[1]Jeffrey Passel, Unauthorized Migrants: Numbers and Characteristics, Pew Hispanic Center, Washington, D.C., June 14, 2005, p. 6.



[4]Ibid., p. 36.

[5]All figures are from Passel, op. cit.

[6]Passel, op. cit.,p. 16.

[7]Ibid., pp. 18 and 25.

[8]Ibid., p. 38. Passel asserts this is due to the socioeconomic characteristics and region of origin of the emigrant women.


[10]Ibid., p. 23. By contrast, the Center for Immigration Studies estimates that two thirds of illegal immigrants lack a high school degree. Steven A. Camarota, The High Cost of Cheap Labor: Illegal Immigration and the Federal Budget, Center for Immigration Studies, Washington D.C., August 2004, p. 5.

[11]Robert J. Samuelson, “We Don’t Need ‘Guest Workers,’” Washington Post, March 22, 2006, p. A21.

[12]George J. Borjas, Heaven’s Door: Immigration Policy and the American Economy, Princeton University Press, 1999, p. 28.

[13]Richard Fry and B. Lindsay Lowell, Work or Study: Different Fortunes of U.S. Latino Generations, Pew Hispanic Center, Wash­ington, D.C., May 28, 2002, tables B1 and B2. The figures in the text refer to individuals aged 25 to 44.

[14]Borjas, op. cit., p. 30.

[15]Ibid.,p. 109.

[16]Ibid., pp. 105, 106. Borjas examined a cohort of immigrants aged 18 to 34 who arrived in the U.S. from 1965 to 1969. Over the next three decades, cash welfare use for this cohort of immigrants rose sharply. Welfare use among a similar native cohort rose as well but not as steeply.

[17]Ibid., p. 110.

[18]A 10 percentage point difference in receipt of welfare in the first generation leads to an 8.2 percentage point difference between groups in the second generation. Borjas,. op. cit., p. 143

[19]Passel, op. cit.,p. 4.

[20]Gordon H. Lester and Jan Tin, “Dynamics of Economic Well Being: Program Participation, 1996 to 1999, Who Gets Assis­tance,” Current Population Reports, P70-94, U.S. Census Bureau, Washington, D.C., January 2004, p. 9.

[21]Ibid., p. 27.

[22]Robert A. Moffit and Peter T. Gottschalk, “Ethnic and Racial Differences in Welfare Receipt in the United States,” in Neil Smelser, William Julius Wilson and Faith Mitchell, eds., America Becoming: Racial Trends and Their Consequences, Volume II, National Academies Press, Washington, D.C. 2001, pp. 156-158.

[23]Steven A. Camarota, “Back Where We Started: An Examination of Trends in Immigrant Welfare Use Since Welfare Reform,” Center for Immigration Studies, March 2003. Welfare use in this study is defined as receipt of Temporary Assistance to Needy Families, Food Stamps, Supplemental Security Income, or Medicaid.



[26]Passel, op.cit. p. 20.

[27]Information provided by Jeffrey Passel.


[29]Joyce A. Martin, Births: Final Data for 2003, National Vital Statistics Reports, Volume 54, Number 2, National Center for Health Statistics, U.S. Department of Health and Human Services, September 8, 2005, p. 49.


[31]Martin, op.cit., p. 49.

[32]Ibid., p. 54, 55.

[33]Center for National Health Statistics, Department of Health and Human Services, Vital Statistics of the United States, 2001, Volume One, Natality, table 1-17, at Among Hispanics, 40.8 percent of births were out-of-wedlock in 1993, compared to 45 percent in 2003. Among blacks, 68.9 percent of births were out-of-wedlock in 1993, compared to 68.5 percent in 2003.

[34]Passel, op. cit., p. 4.

[35]Patrick Fagan, Robert Rector, Kirk Johnson, and America Peterson, The Positive Effects of Marriage: A Book of Charts, The Heritage Foundation, Washington, D.C., April 2002.

[36]Passel, op. cit., p. 34.

[37]Passel, op. cit., pp. 4 and 36.

[38]This category of third-generation Mexicans includes all individuals of Mexican ancestry who have parents born in the U.S.; thus this group also includes the fourth, fifth, and further generations.

[39]All figures are from Richard Fry and B. Lindsay Lowell, op. cit. All figures in the text refer to individuals aged 20 to 24. The inter-group differences for individuals aged 25 to 44 are very similar to those of individuals aged 20 to 44. The attainments for Mexicans in the U.S. are similar to those of Hispanics in the U.S.

[40]National Research Council, The New Americans: Economic, Demographic and Fiscal Effects of Immigration, National Academy Press, Washington, D.C. 1997, p. 388. Figures refer to males aged 18 to 54.

[41]Paige M. Harrison, and Allen J. Beck, “Prisoners in 2003,” Bureau of Justice Statistics Special Report, NCJ 205335, Office of Justice Programs, U.S. Department of Justice, Washington, D.C. November 2004, table 12. See also Thomas P. Bonczar, “The Prevalence of Imprisonment in the U.S. Population, 1974-2001,” Bureau of Justice Statistics Special Report, NCJ197976, August 2003.

[42]The Immigration and Naturalization Service deports hundreds of thousands of illegal immigrants each year. Few of these deportations involve imprisonment in federal or state prisons. In 2000, some 15,000 individuals were in federal prisons due to immigration violations. Most of these were Hispanic. These individuals comprised about 8 percent of the total His­panic population in federal and state prisons. The most common violation leading to imprisonment was reentry following prior deportation. Half of those imprisoned for immigration offenses had prior felony convictions. See John Scalia and Marika F.X. Litras, “Immigration Offenders in the Federal Criminal Justice System, 2000,” Bureau of Justice Statistics Special Report, NCJ-191745, Office of Justice Programs, U.S. Department of Justice, Washington D.C, August 2002.

[43]If the crime rate for Hispanic non-citizens was lower than the rate for non-Hispanics in the general population, then that the crime rate for Hispanic citizens was even higher than the three-to-one ratio (of Hispanic to non-Hispanic whites) sug­gests. It would be higher because low crime Hispanic non-citizens would be included in the denominator used to calculate the overall Hispanic crime rate.

[44]Nation Research Council, op. cit., p. 256.


[46]This is the basic methodology employed by the National Research Council to assess the fiscal gains and losses presented by immigration. National Research Council, op. cit., chapters 6 and 7.


[48]Ibid., Chapter 6.

[49]Ibid., p. 334, table 7.5


[51]Ibid., p. 342. According to net present value calculations the impact of much later years have much lower value than those of the near future; nonetheless, the extension of the NAS projections to 300 years has a definite effect on their conclusions.

[52]The projected educational attainments of the children, grandchildren, and great grandchildren of immigrants who are high school drop outs or high school graduates are presented on page 357 of the National Academy study. (National Research Council, op. cit.) The actual attainments of the descendents of recent Hispanic immigrants are provided in Fry, op. cit.

[53]National Research Council, op. cit., table 7.6 on p.337 and table 7.8 on p. 343.

[54]Ibid., table 7.6 on p. 337.

[55]Ibid. pp. 352 and 353.

[56]Ibid., table 7.5 on p. 324 and figure 7.10 on p. 332.

[57]The total cost of federal and state means-tested welfare benefits spread across all U.S. household is about $5,000 per household. In practice welfare benefits are largely limited to households in the bottom one half of the income distribution and are rarely received by households above the median income level of $44,000 per year. If all benefits were spread equally among the lowest income half of households, the value would be about $10,000 per household.

[58]Steven A. Camarota, The High Cost of Cheap Labor: Illegal Immigration and the Federal Budget, Center for Immigration Stud­ies, Washington D.C., August 2004. This study estimates that 45 percent of illegal immigrant wages are “off the books”.

[59]Ibid. The estimated number and characteristics of the illegal population in this study are very similar to the estimates in most other research. An important element of this study is that, like the National Research Council study, it adjusts for under-reporting of welfare benefits in the Census Bureau’s Current Population Survey. The cost of welfare benefits is adjusted to equal actual government expenditures. The study allocates government private goods and public goods in a manner very similar to the National Academy of Sciences study.

[60]Thus the cost of amnesty is mitigated, somewhat, by the fact that illegals already receive some welfare and social services. This fact, however, only underscores the overall fiscal cost of illegal immigration to society.

[61]The average cost of federal means-tested welfare spending amounts to around $4,000 per household if spread evenly among all U.S. households (including upper- and middle-income households that, in fact, receive little welfare). The Camarota 2004 study assumes that, after amnesty, illegal households would, on average, receive some $9,400 in federal welfare spending, or about 1.3 times more than the artificial average for all households. This seems reasonable given the characteristics of the illegal population.

[62]Calculation based on Steven A. Camarota, 2004, op. cit.

[63]National Research Council, op. cit. p. 29.

[64]Barry R. Chiswick, “Written Testimony” at the hearing on “Immigration: The Economic Impact,” Committee on the Judi­ciary, United States Senate, April 25, 2006, p. 8.

[65]John C. Eastman, “From Feudalism to Consent: Rethinking Birthright Citizenship,” Legal Memorandum No. 18, The Heri­tage Foundation, Washington, D.C., March 30, 2006.
Return to DNI Home