Jump to ContentJump to Main Navigation
Bridging the GapsLinking Research to Public Debates and Policy Making on Migration and Integration$

Martin Ruhs, Kristof Tamas, and Joakim Palme

Print publication date: 2019

Print ISBN-13: 9780198834557

Published to Oxford Scholarship Online: April 2019

DOI: 10.1093/oso/9780198834557.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2019. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see www.oxfordscholarship.com/page/privacy-policy).date: 25 May 2019

Migration Research and Policy in the United States

Migration Research and Policy in the United States

Between Admissionists and Restrictionists

(p.146) 10 Migration Research and Policy in the United States
Bridging the Gaps

Philip Martin

Oxford University Press

Abstract and Keywords

US researchers have reached more consensus on the number and characteristics of migrants than on their socio-economic impacts, especially with regard to unauthorized migrants. When there is consensus among social scientists on some aspect of migration, such as the additional economic output due to the presence of migrants, this consensus result suggests very different policies for advocates. For example, the finding that the US economy may have been up to $10 billion larger in the 1990s due to migrants was touted by advocates of more migrants as proof of their benefits, and criticized by those favouring less migration as the equivalent of two weeks economic growth. President Trump is an example of a policy-maker selectively using migration research to justify restrictionist policies.

Keywords:   immigration, socio-economic impacts, Trump, enforcement, economic growth


Experience in the United States suggests that bridging the gap between research and policy is far easier on demographics (including the number of legal and unauthorized foreigners, and their age structure, fertility, and internal mobility) than on the socio-economic impacts of migrants. This chapter reviews examples of government-funded efforts to develop consensus among researchers on the number and impacts of foreign-born persons in the United States, including the Bi-National Commission on Mexico–US Migration and National Academy of Sciences studies released in 1997 and 2016.

The US experience demonstrates that, even when researchers achieve consensus on the socio-economic impacts of migrants, the results can be interpreted very differently by admissionists (who favour more immigration and the legalization of unauthorized foreigners) and restrictionists (who oppose amnesty and want to reduce immigration). For example, the consensus of social scientists in the 1997 National Academy study was that the 15 million foreign-born workers in the US labour force in 1996 depressed average hourly earnings by 3 per cent and led to a net expansion of US gross domestic product (GDP) of $8 billion. Admissionists touted the $8 billion net gain from immigration, while restrictionists emphasized that the then $8 trillion US economy was growing by 3 per cent ($240 billion) per year, making the net gain due to immigration equivalent to 12 days of US economic growth (Migration News 1997).

(p.147) The effect of economic research on policy-making in the United States is muted because migration’s major economic effects are (re)distributional, with migrants and owners of capital the major winners. Admissionists stress the gains to individual migrants, the minimal costs to US workers, and other benefits of immigration ranging from preserving industries to repopulating cities and increasing diversity. Restrictionists highlight migration as a key reason, along with technology and trade, for depressing wages and hurting low-skilled Americans, increasing inequality, and reducing social trust.

As immigration numbers and impacts rise, the debate over migration policy is increasingly dominated by the most extreme admissionists and restrictionists. Researchers are also tugged toward these ‘no borders’ and ‘no migrants’ extremes by the funders who support and publicize their work. If current trends continue, migration risks joining abortion, guns, and other issues on which Americans are very polarized, and migration research risks joining other issues where links between funders and researchers make research findings suspect, reducing the credibility of experts and research in a wide range of fields.

The chapter begins with discussions of US immigration patterns and research during the periods 1970–2000 and 2000–2016. It then provides a brief reflection on recent developments under the presidency of Donald Trump before concluding with implications for bridging the gaps between research, public debates, and policy-making on immigration.

Immigration Patterns and Research: 1970–2000

The United States is a nation of immigrants whose motto, E pluribus unum (from many, one), reflects openness to newcomers.1 The United States had 42 million foreign-born residents in 2014, almost 20 per cent of the world’s international migrants. Over 50 per cent were from Latin America and the Caribbean, including 28 per cent from Mexico.2 A further 25 per cent were from Asia, the major source countries being China, India, and the Philippines. Almost 50 per cent of all foreign-born residents are naturalized US citizens (Brown and Stepler 2016).

Immigration to the United States occurred in four major waves, beginning with the largely British wave before immigrant admissions began to be recorded in 1820. A second wave was dominated by Irish and German Catholics in the 1840s and 1850s, a third wave included many southern and eastern Europeans between 1880 and 1914, and a fourth wave was set in motion by 1965 laws that switched priority for admission from a migrant’s country of origin to US sponsors requesting the admission of relatives or needed workers. Waves suggest peaks and troughs, with troughs in the aftermath of (p.148) the Civil War in the 1860s and World War I in 1914, followed by legislation in the 1920s that prevented a resumption of large-scale immigration from Europe (Martin and Midgley 2010).

There is no end in sight to the immigration wave launched by the 1965 switch from favouring Europeans to giving priority to foreigners whose US relatives sponsored them for immigrant visas. The change from national origins to family unification was not expected to change immigration patterns, but it did. There was little research to counter the assertion of Senator Edward Kennedy (D-MA) in 1965 that a family unification based selection system would not change ‘the ethnic mix of this country’ (Congressional Digest 1965; CIS 1995).

Kennedy was wrong. During the 1950s, 56 per cent of the 2.5 million immigrants were from Europe; by the 1970s, fewer than 20 per cent of 4.2 million immigrants were from Europe (DHS, Yearbook of Immigration Statistics, table 2).3 Chain migration—as when immigrants and naturalized US citizens sponsor their relatives for visas—was soon apparent, especially because the United States has one of the world’s most expansive definitions of immediate family, including children up to the age of 21 and the parents of US citizens. The United States allows US citizens to sponsor their adult children as well as their adult brothers and sisters for immigrant visas, resulting in sometimes twenty-year waits for visas to become available. The United States offers 50,000 ‘diversity immigrant visas’ awarded by lottery to citizens of countries that sent fewer than 50,000 immigrants to the United States during the previous five years, creating new family networks to sponsor relatives for immigration.

The 10 million foreign-born residents in 1970, the beginning of the fourth and current wave of immigration, were about 5 per cent of US residents at the time. Most immigration research until the 1970s involved historians who explained the integration of third-wave immigrants who arrived before 1914, and debated the effects of very low levels of immigration between the 1920s and 1960s and efforts to ‘Americanize’ southern and eastern European newcomers, many of whom were making a transition from agriculture abroad to cities in the United States.

Researchers debated the factors most important for immigrant integration, including factories and unions, mobilization for war, public schools, and religious and ethnic organizations (Higham 1984).

Two immigration issues drew the attention of social scientists in the 1970s: farm workers and Asians. Between 1942 and 1964, the US government allowed farmers to employ a total of 4.6 million Mexican guest workers under a series of bilateral Bracero agreements. Most of these guest workers returned year after year, but between 1 million and 2 million Mexicans worked in the United States. Bracero admissions peaked in 1956, when 445,000 Braceros (p.149) constituted 20 per cent of US hired farm workers. The number of Braceros fell to fewer than 200,000 after 1962 as the US government tightened enforcement of regulations aimed at protecting US and Bracero workers, and increased the cost of Braceros and spurred labour-saving mechanization. The Bracero programme was ended as a form of ‘civil rights for Hispanics’ in 1964.4

A combination of no new Bracero guest workers, few unauthorized workers, and consumer support for farm worker boycotts during the civil rights era brought about a golden age of rising wages for US farm workers. The United Farm Workers won a 40 per cent wage increase in its first table grape contract in 1966, and represented most Californian table grape and lettuce workers by the early 1970s (Martin 2003). Rising labour costs encouraged labour-saving mechanization, and rising demand for fruits and vegetables kept farm worker employment expanding despite higher wages.

Farm employers knew there were experienced workers in rural Mexico, and some encouraged their legal Mexican-born supervisors to recruit friends and relatives in Mexico and encourage them to enter to the United States illegally. There were no penalties on employers who knowingly hired unauthorized workers until 1986, so rural Mexicans who faced a choice between uncertain incomes in Mexico and a guaranteed job in the United States moved north. Illegal immigration from Mexico surged in the 1980s, especially after a short-lived oil-inspired Mexican government spending boom in the late 1970s collapsed with the price of oil. Farm worker unions protested that ‘illegal aliens’ were undercutting their demands for higher wages and benefits, and demanded that the federal government impose sanctions on employers who hired such workers, but Congress, at the behest of farmers, refused to act.

The research that helped to end the Bracero programme showed that the presence of Braceros depressed the wages of US farm workers, and seemed vindicated by the sharp jump in farm wages in the late 1960s. Clemens et al. (2017) contest this conclusion. They assembled data on farm employment and wages before and after 1964, and concluded that average hourly earnings in states with more Braceros rose at the same pace after 1964 as in states with few or no Braceros, suggesting that removing Braceros did not lead to wage spikes in states that could no longer employ Braceros.

There are several issues with such a before and after comparison as in Clemens et al. (2017), including the fact that the Bracero programme had been shrinking for a decade before it ended, so that state-wide labour markets—in which Braceros constituted less than 5 per cent of farm workers—may be an inappropriate unit with which to measure Bracero effects. Nonetheless, the major response to rising wages demonstrated that flexibility to adjust to fewer workers is mostly a demand side response, not a supply response. Reducing the supply of farm workers does more to accelerate labour-saving substitution (p.150) or to encourage farmers to switch to less labour-intensive crops than to draw US workers into the farm workforce.

During the 1970s and 1980s, research played little role in the debate over unauthorized migrants in agriculture despite case study analyses of how unauthorized workers replaced US citizens and legal immigrants. Farmers turned to contractors to obtain workers, rather than hiring them directly, spurring indirect competition between employers to obtain jobs for crews of legal versus illegal workers, rather than direct competition between legal and illegal workers to be hired (Mines and Martin 1984). Contractors hiring illegal workers contributed to the rising share of unauthorized California crop workers, which jumped from less than 25 per cent in the mid-1980s to 50 per cent a decade later.

The major intervening variable was the Immigration Reform and Control Act (IRCA) of 1986, a compromise between restrictionists (whose priority was to reduce illegal migration) and admissionists (who wanted to legalize the estimated 3–5 million unauthorized foreigners in the United States). The IRCA imposed federal sanctions on employers who knowingly hired unauthorized workers, and allowed unauthorized foreigners who had lived in the United States for at least five years or who had worked in agriculture for at least 90 days to become legal immigrants.

The IRCA proved to be a victory for admissionists. Some 2.7 million unauthorized foreigners, 85 per cent of whom were Mexicans, were legalized, and the widespread use of false documents to obtain legalization under the farm worker programme, which accounted for 40 per cent of all legalizations, taught low-skilled Mexicans that they could continue to get US jobs by providing false documents to their employers (Martin 1994). Legal and unauthorized Mexicans spread throughout the United States, from agriculture to construction, manufacturing and services.

The IRCA unleashed a wave of research. One strand asked how employers adjusted to employer sanctions, and found that labour costs fell because of the upsurge in illegal migration (Martin 1994). However, the National Academies found that the earnings of newly legalized foreigners increased by 10 per cent to 15 per cent, largely because legal status increased their mobility in the US labour market, allowing them to seek jobs with ‘better employers’ (Smith and Edmonston 1997). Farm worker unions shrank due to increased illegal migration, with their problems compounded by internal union problems and the rise of labour contractors and other intermediaries (Martin 2003).

The second major area of research involved the economic progress of, especially, Asian immigrants. Newcomers to the United States typically earn less than similar US-born workers, but the earnings gap narrows over time. Chiswick examined various cohorts of immigrants, such as those arriving in the 1960s and 1970s. He concluded that newcomers arriving in the 1960s (p.151) experienced rapid income gains, catching up to similar US-born workers within thirteen years and then surpassing their US peers, suggesting that average US incomes could be raised via immigration (Chiswick 1978).

Borjas (2016) extended the analysis and concluded that ‘immigrant quality’ as measured by earnings growth in the United States was falling. Chiswick’s data analysis was correct, but reflected a unique and one-time event. Asians found it hard to immigrate until 1965, and those who arrived in the 1960s were especially talented. As immigration from Latin America surged in the 1970s, the initial earnings gap between newcomers and similar US-born workers widened, and immigrant earnings rose much more slowly. Immigrants who arrived in the five years before the 1960 census earned 10 per cent less than US-born workers in 1960, while those who arrived between 1995 and 2000 earned 30 per cent less in 2000 (Borjas 2016).

The legalization of 2.7 million mostly low-skilled Mexicans together with the continued arrival of unauthorized foreigners raised questions about how low-skilled migrants affected similar US workers. Case studies from the 1970s and 1980s suggested that the availability of low-skilled newcomers, legal or illegal, displaced similar US workers and depressed their wages (GAO 1988). However, comparisons of cities with more and fewer immigrants, and studies that compared the wages and unemployment rates of US workers who were assumed to be similar to immigrants in a particular city, could not detect wage depression and displacement, which led to the conclusion that low-skilled migrants do not hurt similar US workers (Smith and Edmonston 1997).

The best-known study of migrant impacts on US workers involved the ‘natural experiment’ of 125,000 Cuban Marielito migrants who arrived in Miami between April and September 1980, increasing Miami’s labour force by 7 per cent. Card (1989) found that the unemployment rate of Blacks in Miami rose more slowly than in several comparison cities that did not receive Cuban migrants, suggesting that the Marielitos benefited, rather than hurt, Blacks in Miami.

Borjas (2016) disagreed with this ‘no-harm-and-perhaps-benefit from migrants’ conclusion. He noted that, when another wave of Cubans tried to reach Florida in 1994, the US Coast Guard intercepted them and sent them to Guantanamo, a US naval base at the eastern end of Cuba. In 1994, the unemployment rate of equal Blacks in Miami rose, while it fell for Blacks in comparison cities. This prompted scepticism of the validity of natural experiments in 1980 and 1994 that failed to find the impacts predicted by economic theory—first, rising unemployment and, then, falling unemployment when migrants did not arrive. One could conclude that there are many factors in addition to migration that affect the unemployment rate of Blacks and other similar US workers.

(p.152) The Mariel controversy has become a symbol of the debate over whether the arrival of low-skilled migrants hurts or helps similar US workers. Card’s conclusions bolstered so-called immigrant supply siders, who believe that migrants add to economic activity, and benefit themselves and other residents, analogous to supply-side economists who believe that tax cuts encourage additional work, benefiting those whose taxes go down and others via the multiplier effects of increased economic activities. Borjas, by contrast, believes that the demand curve for labour is downward sloping; that is, adding to the supply of labour reduces wages, especially in the short run.

Leubsdorf (2017) reviews the arguments over the Marielitos, emphasizing that the comparison group of US workers selected to check for the impacts of migrants affects the conclusions drawn. Analysts who use more inclusive groups of Americans, such as women and teens still in school, are less likely to find negative effects of the Marielitos than those who focus only on adults who did not finish high school. Borjas (2017) dismissed a suggestion that a change in the number of Blacks in the sample led to his conclusion that the arrival of Marielitos reduced the wages of similar Blacks, emphasizing that the wages of Blacks must be constructed, rather than measured by the survey. By this, he meant that analysts are estimating the effects of Marielitos on a constructed, rather than measured, wage of US Blacks. This debate is still ongoing.

The second major focus of research during the 1990s involved the fiscal impacts of immigrants—the question of whether immigrants pay more in taxes than they receive in tax-supported benefits. California Republican Governor Pete Wilson blamed the need to provide services to unauthorized foreigners for the state’s budget deficit in the early 1990s, and won re-election in November 1994 as voters approved Proposition 187 by 59–41 per cent to deny state benefits to unauthorized foreigners, including K-12 education to unauthorized children (Migration News 1994).

Most of Proposition 187 was declared unconstitutional, but law suits demanding that the federal government reimburse states for the cost of providing services to unauthorized foreigners prompted studies of the fiscal impacts of immigrants. The Republican-controlled Congress, in response to Proposition 187, enacted several laws in 1996 to make it more difficult for low-income residents to sponsor their relatives for immigrant visas, and denied federal welfare benefits to legal immigrants arriving after 23 August 1996. At a time when 11 per cent of US residents were foreign-born, 45 per cent of the estimated federal savings from the new welfare system were estimated to come from denying benefits to immigrants until they had worked in the United States for at least 10 years or had become naturalized US citizens after five years (Migration News 1996).

(p.153) In response to Proposition 187 and the 1996 immigration laws, the Commission on Immigration Reform sponsored the first National Academies study. It concluded the US economy was $1 billion to $10 billion larger in 1996 than it would have been with no immigrants, with the best estimate that immigrants were responsible for a net $8 billion gain (Smith and Edmonston 1997).

The model for estimating the net economic gain from immigration assumed that adding immigrants to the labour force reduced average wages by 3 per cent, from an assumed $13 an hour to the actual $12.60 in 1996. The lower wages of all workers expanded the economy and increased the returns to owners of capital, making them and the immigrants who moved to the United States for higher wages and more opportunities the major beneficiaries of immigration.

Estimating the public finance effects of immigrants required more assumptions. The study by the National Academies calculated the net present value (NPV) of the average immigrant in 1996 by assuming that the earnings of immigrants will catch up to those of similar US workers, and that the children and grandchildren of immigrants will have the same average profiles with regard to earnings, taxes paid, and benefits received as the children and grandchildren of native-born children (Smith and Edmonston 1997). The study further assumed that immigration did not raise the cost of public goods such as defence, and that persisting federal government budget deficits would force the government to raise taxes and reduce benefits, rather than continue to borrow to provide benefits for ageing residents, meaning that both young immigrants and young US-born workers would pay more in taxes and receive fewer benefits.

These assumptions helped generate two major findings. First, the average immigrant had a positive NPV of $78,000, meaning that a typical immigrant was expected to pay $78,000 more in federal taxes than they would receive in federal benefits in 1996 dollars over their lifetimes and those of their children and grandchildren. The NPV of immigrants with more than a high school education was plus $198,000, while the NPV of immigrants with less than a high school education was minus $13,000; that is, even assuming that the children of low-educated immigrants have the same average earnings, taxes, and benefits as US-born children, low-skilled immigrants and their children impose a net cost on US taxpayers.

The study led to an obvious conclusion: to generate the maximum economic benefits from immigrants for US-born residents, the selection system should favour young and well-educated newcomers who are most likely to earn higher incomes, pay more in taxes, and consume fewer tax-supported benefits. This recommendation was rejected, as those favouring the current system—including advocates for particular migrant groups, churches, and (p.154) immigration lawyers—argued that levels of immigration should be increased to accommodate more high-skilled foreigners, rather than introduce a points-selection system that would admit more high-skilled and relatively fewer family immigrants. Furthermore, many US employers preferred the current demand-oriented system under which they sponsor foreigners who have temporary work permits for immigrant visas, a system that ties foreigners to a particular employer for years as guest workers. By contrast, a Canadian-style supply-oriented points-selection system would allow newcomers to move from one employer to another.

During the three decades from 1970 to 2000, the share of foreign-born residents in the US population doubled from 5 per cent to 10 per cent. The number of unauthorized foreigners, after dipping briefly with legalization in the late 1980s, more than doubled from 3.5 million in 1990 to 8.6 million in 2000. The effects of legal and unauthorized foreigners were debated. Most economists agreed with Card that, since they could not find the expected negative effects of low-skilled foreigners on similar US workers, there were few or no such effects. There was more consensus among demographers on the number of unauthorized foreigners and more agreement on public finances, since it was easy to visualize how higher levels of education and higher incomes would mean more taxes paid and less reliance on welfare benefits.

Immigration Patterns and Research: 2000–2016

The election of Presidents Vincente Fox in Mexico and George W. Bush in the United States in 2000 was expected to usher in a new era in Mexico–USA migration, marked by cooperation to reduce illegal migration and violence along the Mexico–USA border, legalization of unauthorized foreigners in the United States, and new guest worker programmes. Indeed, just before the terrorist attacks on 11 September 2001, Fox was in Washington, DC, imploring Bush and the US government to enact immigration reforms that would legalize unauthorized foreigners before the end of 2001 (Migration News 2001).

Instead, security took centre stage after the attacks on 11 September 2001, and several laws led to increased monitoring of immigrants and visitors. As the US economy recovered from recession and as illegal immigration rose, there were renewed calls for legalization for unauthorized foreigners. However, there was deadlock in Congress between restrictionists, who emphasized the need for enforcement to deter unauthorized foreigners, and admissionists, who wanted to legalize unauthorized foreigners.

Economists were also deadlocked over the impacts of low-skilled foreign workers on similar US workers. Borjas published an influential article that reinforced economic theory. Given the mobility in the US labour market, (p.155) Borjas argued that the effects of immigrants on US workers must be studied nationally, rather than in any particular city.

Borjas (2003) divided foreign- and native-born workers into age and education cells, so that workers aged 25–30 years with less than a secondary school education were in one cell, and those aged 30–35 years in another; he found up to 10 per cent lower wages for the US-born workers in the young and less-educated cells due to the presence of immigrants. Foged and Peri (2015) disputed Borjas’ conclusion. They suggested that if migrants and natives within each age and education cell were complements, rather than substitutes, playing different labour market roles despite similarities in age and education, and if employers responded to the arrival of migrants by investing more to create jobs for them and US workers, the presence of immigrants raised, rather than lowered, the wages of similar US workers.

The debate over the impacts of low-skilled migrants was mirrored in a similar debate over high-skilled migrants. The United States created the H-1B programme in 1990, at a time when there were believed to be sufficient US workers, as indicated by the unemployment rate of 5.6 per cent, but not enough to fill all of the jobs being created in the rapidly expanding information technology (IT) sector. Some 20,000 temporary foreign workers with college degrees and fashion models were being admitted at the time, and the H-1B programme made it easy for US employers to recruit and employ up to 65,000 per year in the expectation that the number of H-1B visas would begin high and fall over time as US colleges and universities ramped up training and Americans filled more IT jobs.

Instead, the H-1B programme expanded slowly, not reaching the 65,000 cap until 1997 (Martin 2012). At a time of low unemployment and in anticipation of the Y2K problem of computers not adjusting to the year 2000 properly, US employers persuaded Congress to raise the cap, add 20,000 H-1B visas for foreigners who earned Master’s degrees from US universities, and exempt non-profit employers such as universities from the visa cap, allowing over 200,000 H-1B workers per year to enter. Since each H-1B worker can stay up to six years, the United States soon had over a million H-1B visa holders.

Researchers studied the impacts of H-1B workers and reached opposing conclusions. Some found that US employers preferred to hire H-1B workers because they were younger and cheaper than similar US workers. In an IT labour market experiencing considerable mobility, H-1B workers were ‘loyal’ to a particular employer since most wanted to be sponsored for a permanent immigrant visa. Critics called H-1B workers high-tech Braceros, a reference to the by then discredited programme that brought Mexican farm workers to the United States under what are now seen as exploitative circumstances, and found that the presence of H-1B workers reduced wages and the employment of US workers (Bound et al. 2017).

(p.156) Other researchers stressed the spill-over effects of highly skilled foreigners with H-1B visas (Peri et al. 2015). They found, inter alia, that cities with more H-1B foreigners generated more patents and experienced faster wage and job growth—findings that supported employers who wanted to raise the cap on visas (Nell and Sherk 2008). Some researchers echoed employers in arguing that it made no sense for US universities to educate foreigners in STEM-related fields and deny them an opportunity to stay in the United States and work.

Employers have resisted efforts to link more protection for US workers with an increase in the number of H-1B visas available, arguing that requiring employers initially to try to recruit US workers would slow down their need to hire H-1B workers quickly. Instead, they persuaded the DHS, in 2005, to allow foreign students who graduate from US universities with STEM (science, technology, engineering, and mathematics) degrees to remain in the United States and work in jobs related to their degree for up to 30 months, so-called ‘optional practical training’ (OPT), giving these foreign graduates time to find a US employer to offer them H-1B visas valid for six years.5 The Trump administration in 2018 proposed that the period of OPT for all fresh graduates be decreased to twelve months.

By 2005, when Congress began to consider immigration reforms to deal with unauthorized foreigners, most social science researchers agreed that any negative economic effects of low-skilled migrants on similar US workers were small; that high-skilled migrants had positive spill-over economic effects; and that legalization of unauthorized foreigners would increase their mobility and wages, as well as expand the US economy. However, restrictionists in the House of Representatives approved an enforcement-only bill in December 2005 that would have increased enforcement on the Mexico–USA border, required all employers to use the internet-based E-Verify system to check the legal status of new hires, and made illegal presence in the United States a crime, perhaps hindering the ability of unauthorized foreigners to become legal immigrants in the future.

This enforcement-only bill was widely denounced as ignoring the benefits of migration and culminated in a ‘day without migrants’ on 1 May 2006 that involved many businesses closing for the day to highlight the contributions of migrants. In May 2006, the Senate enacted a three-pronged comprehensive immigration bill favoured by President Bush and many social science researchers; in other words, increase enforcement to deter illegal migration, legalize most unauthorized foreigners and put them on a path to US citizenship, and create new guest worker programmes for low-skilled workers. The House refused to act. A similar comprehensive immigration reform bill failed in the Senate in 2007, but was approved in 2013 with the support of President Obama; the House again refused to act and there was no reform.

(p.157) During these immigration reform debates, most social scientists supported the legalization of unauthorized workers and more guest workers. There was very little research on how employers, labour markets, and the economy might adjust to fewer foreign-born workers, since immigration reforms were expected to legalize current workers and admit more.

Trump and Migration

In 2015–2016, Donald Trump campaigned on seven major issues, two of which involved migration: have the United States build and Mexico pay for a wall on the 2,000 mile Mexico–USA border and deport the 11 million unauthorized foreigners in the United States. President Trump issued three executive orders during his first week in office, planning a wall on the Mexico–USA border, increasing deportations and dealing with sanctuary cities, and reducing refugee admissions. Trump said: ‘Beginning today, the United States of America gets back control of its borders.’

Trump launched his bid for the Republican presidential nomination in June 2015 by accusing unauthorized Mexicans of ‘bringing drugs. They are bringing crime. They’re rapists…but some, I assume, are good people’ (Rural Migration News 2015). There was an immediate negative reaction. Most pundits thought that Trump’s inflammatory comments would doom his first campaign for elective office, especially since he was competing with well-known senators and the brother of ex-President George W. Bush.

Trump won the most votes in state-by-state primaries and became the Republican candidate for President in July 2016 with a nationalist platform that centred on the slogan, ‘Make America Great Again’. After a short visit to Mexico, candidate Trump outlined a 10-point immigration plan on 31 August 2016 that began with a wall on the Mexican border and ended in ambiguity about what would happen to unauthorized foreigners in the United States. He said ‘No citizenship. They’ll pay back taxes.…There’s no amnesty, but we will work with them’ (Rural Migration News 2016).

The National Academies released its second consensus report on immigration in September 2016 during the presidential election campaign (Blau and Mackie 2016). The report estimated that immigrants generated up to $54 billion in benefits for Americans in 2015, equivalent to 0.3 per cent of US GDP of $17.5 trillion in 2015. This net benefit reflects a loss in wages to US workers of $494 billion and a gain in profits for the United States of $548 billion. The conclusion: ‘the immigration surplus stems from the increase in the return to capital that results from the increased supply of labor and the subsequent fall in wages’—meaning that the arrival of immigrants depresses wages, expands the economy, and increases profits.

(p.158) The report concluded that immigration bolsters ‘economic growth, innovation and entrepreneurship’, and has ‘little to no’ negative effect on US wages in the long term, largely because the models used to estimate immigration’s impacts on the labour market assume that there will be no long-run impacts on wages.6 The report highlighted negative impacts of newcomers on previous immigrants and US-born workers with little education, including teenagers.

The report found slowing rates of wage convergence, meaning that newcomers to the United States begin their American journeys with a larger earnings gap than previous arrivals, and are slower to close this gap as they integrate into the United States. One reason for this widening earnings gap is that a higher share of newcomers to the United States has low levels of education, and most are learning English more slowly than earlier arrivals.

The National Academies report concluded that immigrants pay less in taxes than the cost of the public services they receive, and their US-born children do not close this gap because the federal government runs a deficit, meaning that all of the taxes paid by all residents do not cover federal government expenditures. At the state and local levels, where governments generally must have balanced budgets, immigrants pay less in taxes than the cost of their tax-supported services, and this immigrant deficit is covered by taxes paid by natives. If the US-born children of immigrants fare as well as other US-born children, over 75 years this immigrant fiscal deficit disappears at the federal level but persists at the state level.

The National Academies report was used selectively in the political debate (Edsall 2016). Restrictionist oriented think tanks such as the Center for Immigration Studies welcomed the report, summarizing it as follows: ‘Immigration is primarily a redistributive policy, transferring income from workers to owners of capital and from taxpayers to low-income immigrant families. The information in the new report will help Americans think about these tradeoffs in a constructive way’ (Camarota 2016).

Admissionist oriented groups such as the American Immigration Lawyers Association said that ‘the study found that immigrant workers expand the size of the U.S. economy by an estimated 11 per cent each year, translating to $2 trillion in 2016 alone. In fact, the children of immigrants are the largest net fiscal contributors among any group, native or foreign-born, creating significant economic benefits for every American’ (American Immigration Lawyers Association 2016). The Migration Policy Institute did not review the report.

Neither candidate Clinton nor candidate Trump responded to the National Academies report when it was released but, in January 2017, President Trump ordered the DHS to plan for construction of a wall on the Mexico–USA border and to beef up interior enforcement by adding 10,000 agents to the current (p.159) 10,000 to detect and remove unauthorized foreigners convicted of US crimes. Trump said that Mexico would pay for the wall.

Trump also reinstated a programme that allows federal immigration agents to train state and local police officers to detect unauthorized foreigners and to hold them for federal agents, or involves state and local police joining task forces with federal enforcement agents to pursue criminal gangs. Trump’s order expanded the definition of criminals who are the highest priorities for deportation to include those charged with, but not necessarily convicted of, US crimes.7 Trump threatened to withhold federal grants from sanctuary cities that ‘willfully refuse’ to cooperate with the DHS, prompting Californian legislators to say they would nonetheless defy Trump and prohibit state and local police from cooperating with federal immigration enforcement agents.8

Trump sought to suspend the admission of refugees for 120 days, block the entry of Syrian refugees indefinitely, reduce planned refugee resettlements in the United States in financial year 17 from 110,000 to 50,000, and ban entries for ninety days from seven countries: Syria, Iran, Iraq, Somalia, Sudan, Libya, and Yemen.9 However, this executive order was blocked by the courts, leading to a revised order issued in March 2017 to block only new entries from six countries, with Iraq removed from the list. This revised order was also blocked by the courts, and re-issued in September 2017; the US Supreme Court is expected to confirm Trump’s authority to regulate the admission of foreign visitors before 1 July 2018.

Trump’s executive orders were widely condemned by most researchers, who emphasize that unauthorized Mexico–USA migration has fallen to historic lows as Mexico completes its fertility transition, and better education and more jobs in Mexico keep most potential migrants at home. Most researchers conclude that migrants are less likely to commit crimes than similar US-born persons, and that efforts to detect and remove unauthorized foreigners would be costly10 and break up mixed families, those in which some members are unauthorized while others are US-born and thus US citizens. Finally, some researchers decried reducing refugee admissions, arguing that the United States has long been a haven for those seeking refuge, and that most refugees integrate successfully and are not terrorist threats.

The United States does not have a governmental migration commission that studies the socio-economic effects of migration on an ongoing basis. Each House of Congress has an immigration sub-committee that conducts oversight hearings on migration-related issues that range from visa issuance to unauthorized migration to guest workers. Researchers are often invited to testify, although the majority party controls most of the witnesses who are allowed to testify, and most witnesses are either employers of or advocates for migrants. Private foundations and the federal government support a wide (p.160) range of migration research, most of which concludes that migrants and their children are integrating successfully, with few adverse and many positive effects on the US economy and society.

Conclusions: Bridging the Gap

Most social science research on migration is optimistic, finding that immigrants help themselves by moving to the United States, and enrich the US economy and society. There are several reasons for this optimism, including the fact that migrants expand the labour force and the economy. Economists find that migrants promote growth without hurting US workers significantly; sociologists find that most newcomers integrate successfully; and political scientists are more likely to celebrate the benefits of diversity, rather than emphasize its potential loss of social capital and trust (Blau and Mackie 2016).

There are three major lessons to be learned from the US experience on bridging the gap between migration research and policy. First, the locus of migration research shifted from history to contemporary migration. When immigration was at a low point in 1970, migration research was dominated by historians who examined how immigrants integrated into and changed US society; these were studies of past events. By 1990, a new generation of non-historians focused on the impacts of contemporary migration. Many were immigrants, including George Borjas (from Cuba), David Card (Canada), Alejandro Portes (Cuba), and Giovanni Peri (Italy).

Second, most government- and foundation-supported migration research concluded that immigration was beneficial for the migrants and the US economy and society, supporting those who wanted to expand immigration and legalize unauthorized foreigners. The Mexican Migration Project (MMP), supported by the US government, obtained work and migration histories from thousands of Mexicans in Mexico who had been in the United States. Analyses using MMP data concluded that Mexico–USA migration had mutually beneficial effects as Mexicans circulated between Mexican homes and US jobs until the US government stepped up border enforcement in the 1990s, ‘trapping’ unauthorized Mexicans in the United States (Massey et al. 2002).

The general theme of social science research today is that immigrants generate more benefits than costs, and that these benefits could be increased if unauthorized foreigners were allowed to legalize their status, which would give them more mobility in the US labour market and provide incentives to learn English. Researchers acknowledge the risk of segmented assimilation, as when frustrated immigrant children or the children of immigrants who identify with US minorities feel unable to get ahead, drop out of school, and, perhaps, join gangs (Blau and Mackie 2016; Meissner et al. 2006).

(p.161) Third, unauthorized migration, especially, has become an increasingly contentious issue. The research and elite consensus was that three-pronged comprehensive immigration reforms—namely, more enforcement, legalization for unauthorized foreigners, and new guest worker programmes—would be enacted after Hillary Clinton was elected president in 2016. In anticipation of Clinton’s election, many of the major foundations supporting migration research shifted funding to implement legalization.11

This meant that there was little research on what is happening as a result of increased enforcement under President Trump. Preliminary research suggests that employers are reacting as economic theory would predict. As wages rise, employers are substituting capital for labour in industries from agriculture and care giving to construction and restaurants (Martin 2017).

Bridging the gap between research and policy is often difficult. For two centuries, economists have preached the virtues of freer trade, arguing that comparative advantage ensures that most people in trading countries are better off because winners win more than losers lose, so that winners can compensate losers and leave everyone better off. Free trade became the mantra of opinion leaders in both major political parties and all significant research institutions; most paid only lip service to the need to compensate the losers from freer trade by retraining displaced workers for new jobs. Opposition to freer trade came largely from unions representing manufacturing workers, who found that displaced manufacturing workers who were forced into service jobs generally had lower wages, even after retraining, belying the promise that trade’s winners would fully compensate losers.

Bridging the gap is similarly difficult in climate change. Even with research agreement that the climate is warming, there is disagreement over how much of this warming is due to human activities and the appropriate policy response in which to invest now to avoid problems in the future. Most researchers conclude that human activities are a major cause of climate change and urge a significant investment now to minimize future adjustment costs. As with migration, the few researchers who disagree on the need for a carbon tax or other investments now to avoid future problems are considered out of the mainstream; their research is often dismissed because some was funded by energy firms that would be adversely affected by carbon taxes.

It is sometimes said that, in economics, it is easy to become famous without ever being right. Many migration researchers reach the same conclusion in each of their papers and some of those who make false predictions seem to incur no penalties. In most cases, one needs to read only the name of the researcher to know the conclusions, as with economists who consistently find that low-skilled migrants help or hurt similar US workers. Second, there appear to be few penalties for false predictions, as illustrated by the careers of those who confidently predicted in the 1970s, when there were fewer than 2 million (p.162) Mexican-born US residents, that Mexicans were sojourners, not settlers, and would not settle in the United States. Forty years later, there were 12 million Mexican-born persons in the United States, plus an additional 18 million children born to them in the United States.

Americans are better educated than ever before, there is more scientific research than ever, and many research-influenced policies are ever more contentious. One reason may be that better educated voters and politicians realize that research generally, and migration research in particular, rarely reaches truly definitive answers. Researchers who oversell their results, and the think tanks, media, and politicians who amplify their message, may wind up reducing the credibility of all research on contentious issues.


Bibliography references:

American Immigration Lawyers Association (AILA). 2016. National Academies Report Underscores Immense Economic Importance of Immigrants to America. www.aila.org/advo-media/press-releases/2016/nas-report-econ-importance-immigrants-to-america

Blau, F., and Mackie, C. 2016. The Economic and Fiscal Consequences of Immigration. National Academies. www.nap.edu/catalog/23550/the-economic-and-fiscal-consequences-of-immigration

Borjas, G. 2003. The Labor Demand Curve is Downward Sloping: Reexamining the Impact of Immigration on the Labor Market. Quarterly Journal of Economics 118(4): 1335–74.

Borjas, G. 2016. We Wanted Workers. Unraveling the Immigration Narrative. New York: Norton.

Borjas, G. 2017. Race and Mariel. https://gborjas.org/2017/06/12/race-and-mariel/

Bound, J., Khanna, G., and Morales, N. 2017. Understanding the Economic Impact of the H-1B Program on the U.S. PSC Research Report No. 16–857. www.psc.isr.umich.edu/pubs/abs/10003

Brown, A., and Stepler, R. 2016. Statistical Portrait of the Foreign-Born Population in the United States. www.pewhispanic.org/2016/04/19/statistical-portrait-of-the-foreign-born-population-in-the-united-states-trends/

Camarota, S. 2016. National Academy of Sciences Study of Immigration: Workers and Taxpayers Lose, Businesses Benefit. CIS. https://cis.org/National-Academy-Sciences-Study-Immigration-Workers-and-Taxpayers-Lose-Businesses-Benefit

Card, D. 1989. The Impact of the Mariel Boatlift on the Miami Labor Market. NBER Working Paper No. 3069. Cambridge, MA: NBER.

Chiswick, B. 1978. The effect of Americanization on the earnings of foreign-born men. Journal of Political Economy 86(5): 897–921.

CIS. 1995. The Legacy of the 1965 Immigration Act, 1 September. https://cis.org/Legacy-1965-Immigration-Act

(p.164) Clemens, M. A., Lewis, E. G., and Postel, H. M. 2017. Immigration Restrictions as Active Labor Market Policy: Evidence from the Mexican Bracero Exclusion. NBER Paper 23125. Cambridge, MA: NBER.

Congressional Digest, May 1965, 152. https://catalog.hathitrust.org/Record/008315096

DHS. Yearbook of Immigration Statistics. www.dhs.gov/immigration-statistics/yearbook (Accessed 15 July 2017).

Edsall, T. 2016. What does immigration actually cost us? New York Times. 29 September.

Foged, M., and Peri, G. 2015. Immigrants’ effect on native workers: New analysis on longitudinal data. American Economic Journal: Applied Economics, 8(2): 1–34. CReAM Discussion Paper Series 1507.

GAO. 1988. Illegal Aliens: Influence of Illegal Workers on Wages and Working Conditions of Legal Workers. www.gao.gov/products/PEMD-88-13BR

Higham, J. 1984. Send These to Me: Immigrants in Urban America. Baltimore, MD: John Hopkins University Press.

Leubsdorf, B. 2017. The great Mariel boatlift debate: Does immigration lower wages? Wall Street Journal, 17 June.

Martin, P. 1994. Good intentions gone awry: IRCA and U.S. agriculture. The Annals of the Academy of Political and Social Science 534(1): 44–57.

Martin, P. 2003. Promise Unfulfilled: Unions, Immigration, and Farm Workers. Ithaca, NY: Cornell University Press.

Martin, P. 2012. High-skilled migrants: S&E workers in the United States. American Behavioral Scientist 56(8): 1058–79.

Martin, P. 2017. Immigration and Farm Labor. Challenges and Issues. Giannini Foundation. http://bit.ly/2tvaUSw

Martin, P., and Midgley, E. 2010. Immigration in America (June). Washington, DC: Population Reference Bureau.

Martin, S. 2011. A Nation of Immigrants. Cambridge: Cambridge University Press.

Massey, D. S., Durand, J., and Malone, N. J. 2002. Beyond Smoke and Mirrors Mexican Immigration in an Era of Economic Integration. New York: Russell Sage.

Meissner, D., Meyers, Deborah W., Papademetriou, Demetrios G., and Fix, Michael. 2006. Immigration and America’s Future: A New Chapter. Washington, DC: Migration Policy Institute.

Migration News. 1994. Prop 187 Approved in California 1(11). https://migration.ucdavis.edu/mn/more.php?id=492 (Accessed 15 July 2017).

Migration News. 1996. Welfare Overhaul and Minimum Wage Changes 3(9). https://migration.ucdavis.edu/mn/more.php?id=1246 (Accessed 15 July 2017).

Migration News. 1997. NRC on Immigration 4(6). https://migration.ucdavis.edu/mn/more.php?id=1246 (Accessed 15 July 2017).

Migration News. 2001. Fox Visits Bush 8(10). https://migration.ucdavis.edu/mn/more.php?id=2463 (Accessed 15 July 2017).

Mines, R., and Martin, P. 1984. Immigrant workers and the California citrus industry. Industrial Relations 23(1): 139–49.

National Academies. 2016. The Economic and Fiscal Consequences of Immigration. Washington, DC: National Academies Press.

(p.165) Nell, G., and Sherk, J. 2008. More H-1B Visas, More American Jobs, A Better Economy. Washington, DC: Heritage Foundation.

Peri, G., Shih, K., and Sparber, C. 2015. STEM Workers, H-1B Visas, and Productivity in US Cities. Journal of Labor Economics 33(S1) (Part 2): S225–S255.

Rural Migration News. 2015.

Rural Migration News. 2016.

Smith, J., and Edmonston, B. (eds). 1997. The New Americans: Economic, Demographic, and Fiscal Effects of Immigration. Washington, DC: National Research Council. (p.166)


(1.) The original meaning of E pluribus unum was that one nation emerged from the thirteen colonies, but the phrase has evolved to symbolize unity from diversity, or the ability of the United States to integrate newcomers (Martin 2011).

(2.) There were 11.7 million Mexican-born residents in 2014: 4 million born in the Caribbean, 3.3 million born in Central America, and 2.8 million born in South America. That is, 21.8 million (52 per cent) of all foreign-born residents were from Latin America.

(3.) Stocks changed slower than flows. In 1960, 85 per cent of the 9.7 million foreign-born residents were from Europe or Canada; by 1980, their share dropped to 43 per cent of 14 million (Brown and Stepler 2016).

(4.) The average employment of hired workers on US farms in the early 1960s was 2.5 million. http://usda.mannlib.cornell.edu/MannUsda/viewDocumentInfo.do?documentID=1063

(5.) Employers do not have to try to recruit US workers before hiring OPT graduates, and there are no special wages that must be paid to OPT employees. www.uscis.gov/working-united-states/students-and-exchange-visitors/students-and-employment/stem-opt

(6.) ‘In the case of structural studies, when capital is assumed to be perfectly flexible, [average] wage effects on natives are zero, although this result is built in by theoretical assumptions’ (National Academies 2016).

(7.) Two-thirds of the 2 million foreigners who were put in removal proceedings after being detected by Secure Communities enforcement had committed only misdemeanour crimes. The Priority Enforcement Program unveiled in November 2014 targeted foreigners convicted of felonies (640,000) and those who arrived in the United States after 1 January 2014 (640,000). In July 2015, the Migration Policy Institute noted that the Priority Enforcement Program made one-eighth of the 11 million unauthorized foreigners in the United States priorities for DHS enforcement.

(8.) Sanctuaries are states, counties, and cities that limit their cooperation with the Immigration and Customs Enforcement (ICE) agency of the DHS. In 2015, there (p.163) were four states, 326 counties, and thirty-two cities that had declared themselves to be sanctuaries for unauthorized foreigners (www.americanimmigrationcouncil.org/research/sanctuary-cities-trust-acts-and-community-policing-explained).

(9.) The United States has admitted 785,000 refugees since 11 September 2001, including twelve who were arrested or removed from the United States due to concerns over terrorism. Some 3.2 million refugees have been admitted since 1975, including 85,000 in financial year 16, of whom 72 per cent were women and children. In financial year 16, 38,900 Muslim and 37,500 Christian refugees were admitted.

(10.) In 2016, ICE estimated that it costs $12,200 to identify and remove each unauthorized foreigner, a cost that could drop if state and local governments cooperated with ICE. See www.politico.com/story/2016/12/is-donald-trump-deportation-plan-impossible-233041