Jason Richwine, PhD, is an independent public policy analyst based in Washington, D.C., and a contributing writer at National Review.
Highly educated immigrants, meaning those who arrive with a college degree or more, often find that their skills do not fully transfer to the U.S. labor market. Many end up holding jobs for which they are overqualified based on their paper credentials. This "occupational mismatch" among legal highly educated immigrants is the subject of this report. By a moderate but significant margin, legal immigrants with at least a college degree are more likely than natives to take low-skill jobs. More importantly, the severity of the mismatch varies widely across sending regions. Legal immigrants from some regions of the world struggle much more than others to utilize their college or advanced degrees. As a consequence, policy-makers should be cautious when designing an immigration system that selects for education.
Some notable findings:
- Among immigrants with a college degree, 20 percent have a low-skill (bottom third) occupation, compared to 7 percent of natives.
- Nearly 30 percent of Mexican immigrants with a college degree have a low-skill occupation, as do 35 percent of Central American immigrants.
- About 85 percent of Canadian immigrants with at least a college degree have a high-skill (top third) occupation, compared to 73 percent of natives and 53 percent of Mexican immigrants.
- Among immigrants with an advanced degree, 37 percent have an elite-skill (top tenth) occupation, compared to 50 percent of natives.
- Length of U.S. residency is not strongly correlated with occupational skill level.
Much of the debate over immigration in the past several decades has focused on the low end of the skill distribution. Can a postindustrial nation such as the United States manage the economic, fiscal, and social impact of taking in immigrants with low education? Although that question remains important, some attention has shifted in recent years toward the impact of higher-skill immigration, for two main reasons. First, proposals to raise the average skill level of legal immigrants have gained traction in policy circles and on Capitol Hill. The Cotton-Perdue RAISE Act, for example, would abolish extended family preferences and reallocate employment-based visas toward a points system. Second, immigrants have become more educated even in the absence of a policy change. The percentage of new working-age immigrants with at least a college degree increased from 30 percent in 2000 to 49 percent in 2016.1
Highly educated immigrants do not always secure high-skill jobs, however. Many experience "occupational mismatch", holding jobs for which they are overqualified according to their paper credentials. Perhaps the most obvious reason is that some immigrants lack the legal right to work in the United States.2 However, this report limits its analysis to legal immigrants.
Occupational mismatch can still occur among legal immigrants for a variety of reasons. Cultural obstacles include a lack of English literacy and an unfamiliarity with regulations, networking, and licensing requirements in the U.S. labor market.3 In addition, some skilled immigrants arrive on temporary visas that restrict their job options.4 Beyond labor market obstacles, different countries may simply have different standards for earning degrees, meaning that the actual skills of educated immigrants can vary considerably across sending nations.5
This report offers new data on occupational mismatch among highly educated legal immigrants. It demonstrates that immigrant groups differ, on average, in how they utilize their college or advanced degrees in the U.S. labor market. Some highly educated immigrants, such as those from Canada and Australia, generally take jobs in the United States that require more skill than jobs held by comparably educated natives. By contrast, highly educated immigrants from places such as Mexico and Central America tend to take lower-skill jobs despite their credentials. The occupational skills gap does not disappear as immigrants spend more time in the United States. These results suggest a need for caution in designing a high-skill immigration system.
Data and Methods
Economists have built a voluminous literature on occupational mismatch since the mid-1970s, with several studies focusing on immigrants specifically. The general finding from Western Europe and the Anglosphere is that immigrants are more likely to be mismatched than natives.6
In the context of high-skill immigration to the United States, the most relevant study comes from Barry Chiswick and Paul Miller.7 Chiswick and Miller defined the skill level of each Census-designated occupation using the "realized matches" technique, meaning they determined the most common level of educational attainment held by natives in each occupation. The authors found that highly educated immigrants were more likely than natives to hold jobs below their skill level. They also found that the difference was smaller for immigrants with stays of 20 years or more in the United States. Because they used data from the single year of 2000, however, it is unclear whether the progress is due to acculturation or to a change in the characteristics of immigrants who arrived over time.
Using data from the five-year sample of the 2016 American Community Survey (ACS), this report starts with a version of the "realized matches" technique. It defines the skill level of an occupation as the average years of education held by natives who work in that occupation. Occupations are then ranked by their skill level and put into four categories: bottom third ("low skill"), middle third ("middle skill"), top third ("high skill"), and top tenth ("elite skill"). For example, if we take all the natives who work as receptionists and average their years of education together, we get 13.1 years. Therefore, the occupation of receptionist has a skill level of 13.1, placing it in the "middle skill" category as indicated by Table 1 below.
The advantage of this method is the rich amount of information it provides about the skill level of occupations held by both immigrants and natives. Rather than dichotomously classifying each worker as mismatched or not, we can see the degree to which highly educated immigrants differ from highly educated natives — both in terms of the average years of education required for their jobs, and in terms of the occupational skill categories (low, middle, high, elite) into which they fall.
All results are based on ACS respondents ages 25 to 64 who are currently employed.8 Immigrants must have first arrived at age 25 or older to ensure that most received their education abroad. Finally, immigrants likely to be illegal are removed from the data.9
This section presents occupational skill levels by immigrants' region of origin, then by country of origin, then by length of residence in the United States.
Occupational Skill Level by Sending Region. Table 2 gives results for all workers with a college degree or more. The bottom two rows show that natives have an occupational skill level that is a moderate 0.2 years greater than immigrants, and that natives are more likely to participate in high- and elite-skill jobs. More important than the overall immigrant-native comparison, however, is how much the occupational skill levels vary among highly educated immigrant groups. Roughly one quarter of highly educated immigrants from Mexico and Central America are in low-skill occupations, compared to less than 5 percent of immigrants from Canada and Australia.
The next two tables split highly educated immigrants into college and advanced degree categories. Again, there is considerable variation in occupational skill across immigrant-sending regions. About 77 percent of college graduates from Canada have a high-skill occupation, compared to just 32 percent of college-educated Central American immigrants. In fact, Central American immigrants with a college degree are marginally more likely to be in low-skill occupations than in high-skill ones.
Advanced degree holders, both immigrant and native alike, have substantially higher occupational skill levels than college graduates. Nevertheless, a non-negligible proportion of Latin American postgraduates are profoundly mismatched, holding low-skill jobs despite having at least a master's degree.
Notably, only 50 percent of natives with advanced degrees hold an elite job. Although that is 13 percentage points higher than the figure for comparably educated immigrants, advanced degrees clearly do not guarantee elite occupational status for any group.
Occupational Skill Level by Sending Country. Occupational skill levels are even more disparate across individual countries, with nearly two years separating the college-educated immigrants with the highest occupation skill levels from those with the lowest. Interestingly, although Western Europe and the Anglosphere predominate at the top of the college table, several Middle Eastern countries head up the advanced degree table. Country-to-country rankings should be regarded as approximate, however, as the sample sizes can be small.10
Occupational Skill Level by Length of Residency in the United States. Although occupational mismatch should lessen as immigrants acculturate, the effect is difficult to discern. Recall from the methodology section that Chiswick and Miller found less mismatch among immigrants who had arrived at least 20 years earlier — which would be 1980, in their study — but they were unable to determine whether the difference reflects acculturation or the different characteristics of newer arrivals. The data are similarly ambiguous today. Table 7 shows that highly educated immigrants with fewer than 10 years in the United States appear to have a slightly greater skill level than more experienced cohorts, with higher percentages in both the low-skill and elite categories. No strong pattern emerges across time, however, and no residency group matches native skill levels.11
Lawmakers should approach a high-skill immigration policy with caution. Although highly educated immigrants certainly offer more economic benefits than less-educated immigrants, paper credentials are an inconsistent predictor of success. This report has shown that highly educated legal immigrants tend to take low-skill jobs more often than comparably educated natives do. More importantly, the average value of higher education varies depending on the sending country. Immigrants with at least a college degree from places such as Canada and Australia have an occupational skill level that is higher than their native counterparts. By contrast, highly educated immigrants from Latin America have a substantially lower occupational skill level than natives. To illustrate, 85 percent of Canadian immigrants with at least a college degree have a high-skill occupation, compared to 73 percent of natives and 53 percent of Mexican immigrants.
More broadly, policymakers should remember that "higher skill" does not mean "Einstein". Few college-educated immigrants are Silicon Valley entrepreneurs or Ivy League research scientists. In fact, only 22 percent of immigrants with at least a college degree have an elite-skill occupation. Most are ordinary people working ordinary jobs, just as natives are. If our goal is to recruit Einsteins, selection criteria must be more discriminating and objective than educational credentials alone.12
For space and readability, the tables in the main text report means only. Standard errors (derived from the Census Bureau's replicate weights) are given below.
1 Part of the shift is due to Asia overtaking Latin America as the most common source of new immigration, but even Latin American immigrants have become more educated in recent years.
2 Among the nation's 44 million foreign-born individuals, an estimated 11 million are illegally present in the United States. Some illegal immigrants can have work permits, however. Examples include recipients of Deferred Action for Childhood Arrivals (DACA) and those who have Temporary Protected Status (TPS).
3 Matloob Piracha and Florin Vadean, "Migrant Educational Mismatch and the Labour Market", in International Handbook on the Economics of Migration, Amelie F. Constant and Klaus F. Zimmermann, Eds., Northampton, MA: Edward Elgar Publishing, 2013.
4 This mobility problem may be overstated, however. A recent study finds that immigrants on temporary visas have similar job-changing rates to permanent residents and natives. Mobility is typically lower only when temporary immigrants are using their current employer as a green-card sponsor. See Jennifer Hunt, "How Restricted Is the Job Mobility of Skilled Temporary Work Visa Holders?", NBER Working Paper 23529, June 2017.
5 This is perhaps one reason that U.S. employers devalue foreign credentials. For more on this topic, see Ana Ferrer and W. Craig Riddell, "Education, Credentials, and Immigrant Earnings", Canadian Journal of Economics, Vol. 41, No. 1, February 2008, pp. 186-216.
6 Piracha and Vadean, "Migrant Educational Mismatch and the Labour Market", Table 1.
7 Barry R. Chiswick and Paul W. Miller, "Educational Mismatch: Are High-Skilled Immigrants Really Working at High-Skilled Jobs, and What Price Do They Pay If They Aren't?", High Skilled Immigration in a Global Labor Market, Barry R. Chiswick, Ed., Washington, D.C.: American Enterprise Institute Press, 2010, pp. 111-154.
8 The Census Bureau asks for each respondent's most recent occupation within the last five years. Therefore, it is possible to identify the occupation of some non-working respondents. However, keeping unemployed respondents in the dataset opens the possibility that recently arrived immigrants will give the occupations they had in their native countries. Therefore, non-working respondents are excluded from the analysis. This decision has no practical effect on the results.
9 Illegal immigrants are present in Census data but never explicitly identified by the Bureau. To determine which respondents are most likely to be illegal, CIS follows a methodology similar to the one used by the Pew Research Center and the Center for Migration Studies. We start by eliminating immigrant respondents who are almost certainly not illegal — for example, spouses of natural-born citizens, people who receive direct welfare payments, people who have government jobs, Cubans (because of special rules for that country), immigrants who arrived before 1980 (because the 1986 amnesty should have already covered them), people in certain occupations requiring licensing or a government background check, people likely to be on student visas, and so on. Then the remaining candidates are selected to match known characteristics of illegals (age, gender, country of origin, and state of residence) determined by the Department of Homeland Security. The result is a close match to other published estimates. Of course, these are still merely estimates, and any analysis comparing legals and illegals should be interpreted accordingly.
10 For standard errors, see the Appendix.
11 Complicating the analysis is the effect of age. Younger immigrants tend to be better matched, possibly because they are more likely to arrive on employment visas. Another possibility is that they are more likely to have received their education in the United States, even though this report was designed to capture immigrants who bring their education with them from abroad. To the extent that some immigrants in the data are U.S.-educated, the results would be biased toward finding less mismatch.
12 To its credit, the proposed RAISE Act takes steps in that direction, as it would give extra points for English fluency, STEM specialties, and pre-arranged employment. Direct testing for skills (other than English ability) may work even better, but that is beyond the scope of this report.