Skip to main content
Log in

The Impact of Legal Status on Immigrants’ Earnings and Human Capital: Evidence from the IRCA 1986

  • Published:
Journal of Labor Research Aims and scope Submit manuscript

Abstract

The Immigration Reform and Control Act (IRCA), the largest amnesty in U.S. history, took effect in 1986 and legalized all immigrants who arrived before 1982. The IRCA creates a discontinuity, according to the year of entry, in the probability of having legal status. Therefore, I use the regression discontinuity approach to study the impact of legality on immigrants’ labor market outcomes and human capital. Using Californian Latino immigrants from Census 1990, I find that the 1975–81 arrivals, on average, outperform the 1982–86 arrivals in male wages, female employment probability, and male English-speaking ability. These findings are not due to a general trend in U.S. labor market conditions because the same analysis, using refugees, Puerto Rican migrants and U.S.-born Latinos—three comparison groups without legality issues—indicates no difference in outcomes between the 1975–81 and 1982–86 cohorts. However, the advantage of Latino immigrants of the earlier cohort over the later cohort diminishes in Census 2000.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3

Similar content being viewed by others

Notes

  1. Detailed analysis of this point is in Pan (2009).

  2. Two recent working papers (Barcellos 2010; Lozano and Sorenson 2011) also exploit the 1982-cutoff in IRCA eligibility to study the labor market value of legal status. However, both papers were dated after the earlier draft of this paper (Pan 2010); Lozano and Sorenson (2011) cited Pan (2010). As stated in the following text, I not only present the impact of legality but also investigate the mechanism through which the impact takes place; and I find that the mechanism is different between men and women. The gender difference in the mechanism is never documented in other papers and I consider this an important contribution to the literature. In addition, Lozano and Sorenson (2011) use data from the Mexican Migrant Project; Barcellos (2010) takes nationwide samples from the Census. However, this paper restricts the Census sample to the immigrants who migrated as adults and who reside in California. The estimated impact of the IRCA on immigrants’ wages in this paper is smaller than that in Lozano and Sorenson (2011) and much larger than that in Barcellos (2010). Further discussion on this is in Appendix A and footnote 18.

  3. However, Rivera-Batiz (1999) does not account for the selection of women’s labor force participation when estimating women’s wages.

  4. To identify the LAW beneficiaries, one need only know an immigrant’s year of entry. However, to identify the SAW beneficiaries, one must know an immigrant’s occupation in the 1980s.

  5. I use Census 2000, not Census 1990, to examine U.S. citizenship because immigrants are eligible to be naturalized after they have Legal Permanent Resident (LPR) status for 5 years. Most of the amnestied immigrants did not obtain their LPR status until 1990. Hence, they were not yet naturalized in 1990.

  6. One concern about restricting the sample to California residents arises from the potential selective migration issue. Successful immigrants may eventually move out of California to other states. However, the pre-1982 immigrants have legal status and should be, on average, more likely than the post-1982 immigrants to be successful and to move to other states. This suggests that pre-1982 immigrants who stay in California may have lower unobservable attributes than post-1982 immigrants who do so. Therefore, the bias caused by immigrants leaving California, if there is any, should work against my finding.

  7. Appendix Table 1 presents the distribution of sampled immigrants by cities. Among the 22 California cities identified in the Census 2000 5% sample, Los Angeles accounts for 55% of the immigrant population. Therefore, the results of this paper are derived mainly from the residents of this city.

  8. Appendix B contains the definition of wages, employment probability and English-speaking ability.

  9. I also applied a two-stage Heckman selection model to both male and female immigrants, using marital status and number of children as the excluded variable in the labor force participation equation. Though the results are not presented, they are consistent with the OLS estimations that are presented in the paper.

  10. Table 3 presents the breakdown of the refugee population living in Californian cities in 1990 by countries of origin.

  11. Appendix B explains the method that I use to define the year of entry and duration of work for U.S.-born workers.

  12. Native-born Latinos have, on average, five more years of schooling than Latino immigrants. One concern is that high-skilled workers may have different experiences from low-skilled workers during economic recessions. For example, the post-1982 Latino immigrants may have suffered more from the 1980 recession than the post-1982 native-born Latinos did. Hence, I did a robustness check using the native-born Latinos whose years of schooling were 12 or less. The regression results using equation (1) are similar to the results that are derived from the full sample of native-born Latinos.

  13. Though the occupational categories are defined broadly, Latino immigrants actually concentrate in only a few occupations in each category. For example, male Laborers are mainly delivery drivers, construction laborers and machine operators; female Laborers are mainly seamstresses and assemblers of electrical equipment. Male Service workers are mainly cooks and janitors; female Service workers are mainly housekeepers and servants.

  14. Source: http://www.numbersusa.com/text?ID=1049.

  15. It may be the other way around, considering that legal immigrants can return to the U.S. quite easily if they change their minds but illegal immigrants cannot. However, unsuccessful illegal immigrants may temporarily delay their return trip until they achieve their target earnings (Dustmann 2003) rather than become permanent migrants. Lindstrom (1996) documented that the duration of the first-trip of return migrants on average is 22 months and the standard deviation is 28 months. His finding suggests that immigrants who arrived in the early or mid-1980s and planned to return probably had already done so before 1990.

  16. For example, Barcellos (2010) uses the sample of all foreign-borns in the U.S. Censuses and find that arriving before 1982 increases all immigrants’ wages by only 0.6%, as opposed to 4.5%, which is the estimation in this paper.

References

  • Amuedo-Dorantes C, Bansak C, Raphael S (2007) Gender differences in the labor market: impact of IRCA's amnesty provisions. Am Econ Rev 97(2):412–416

    Article  Google Scholar 

  • Barcellos SH (2010) Legalization and the economic status of immigrants, RAND Working Papers 754

  • Bean FD, Lowell BL, Taylor LJ (1988) Undocumented Mexican immigrants and the earnings of other workers in the United States. Demography 25:35–52

    Article  Google Scholar 

  • Briggs VM (2004) Guestworker programs, lessons from the past and warnings for the future. Center for Immigration Studies, Washington D.C

    Google Scholar 

  • Cortes K (2004) Are refugees different from economic immigrants? Some empirical evidence on the heterogeneity of immigrant groups in the United States. Rev Econ Stat 86(2):465–480

    Article  Google Scholar 

  • Dustmann Christian (2003) Return migration, wage differentials and the optimal migration duration. Eur Econ Rev 47:353–369

    Article  Google Scholar 

  • Fortuny K, Capps R, Passel JS (2007) The Characteristics of Unauthorized Immigrants in California, Los Angeles County, and the United States. The Urban Institute, Washington, DC

  • Furtado D, Theodoropoulos N (2010) Why does intermarriage increase immigrant employment? The role of Networks, The B.E. Journal of Economic Analysis & Policy, Berkeley Electronic Press, 10:1

  • Kossoudji SA, Cobb-Clark DA (2002) Coming out of the shadows: learning about legal status and wages from the legalized population. J Labor Econ 20(3):598–628

    Article  Google Scholar 

  • Kossoudji SA, Cobb-Clark DA (2000) IRCA’s impact on the occupational concentration and mobility of newly-legalized Mexican men. J Popul Econ 13(1):81–98

    Article  Google Scholar 

  • Hill LE, Lofstrom M, Hayes JM (2010) Immigrant legalization: assessing the labor market effects, Public Policy Institute of California

  • Lindstrom D (1996) Economic opportunity in Mexico and return migration from the United States. Demography 33(3):357–374

    Article  Google Scholar 

  • Lozano FA, Sorensen T (2011) The Labor Market Value to Legal Status,” IZA Discussion Papers 5492, Institute for the Study of Labor

  • Massey D (1987) Understanding Mexican migration to the United States. Am J Sociol 92:1372–1403

    Article  Google Scholar 

  • Oreopoulos P, von Wachter T, Heisz A (2006) The short- and long-term career effects of graduating in a recession: hysteresis and heterogeneity in the market for college graduates. Columbia University, Mimeo

    Google Scholar 

  • Orrenius PM, Zavodny M (2003) Do amnesty programs encourage illegal immigration? Evidence from IRCA. Demography 40(3):437–450

    Article  Google Scholar 

  • Passel JS (2005) Unauthorized migrants: numbers and characteristics. Pew Hispanic Center, Washington, DC

    Google Scholar 

  • Pan Y (2009) Gains from Green Cards, Immigrant Parents’ Legal Status and Children’s Scholastic Achievement, LSU working paper

  • Pan Y (2010) The Impact of Legal Status on Immigrants’ Earnings and Human Capital: Evidence from the IRCA 1986, LSU working paper

  • Rivera-Batiz FL (1999) Undocumented workers in the labor market: an analysis of the earnings of legal and illegal immigrants in the U.S. J Popul Econ 12(1):91–116

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Ying Pan.

Additional information

I thank Vernon Henderson, Nathaniel Baum-Snow, Rachel Friedberg and Naci Mocan. All errors are mine.

Appendix

Appendix

A. Explanation of the Sample Restriction

As I stated in Section “Pre-1982 and Post-1982 Latino Immigrants”, the working sample consists of the Latino immigrants who came to the U.S. as adults (at least 15 years old) and reside in California cities. In order to explain why these restrictions are necessary, I discuss the cases if the restrictions are relaxed. I plot citizen ratio against years of entry for the following four samples: (a) European and Asian immigrants in the U.S.; (b) Latino immigrants live in states other than California; (c) Latino immigrants who entered the U.S. at the age of 14 or younger. (d) Latino immigrants living in rural California. The plots of the above four cases are presented in Appendix Fig. 5, panels A-D. Each case is presented in contrast with the figure that is plotted using the working sample.

In panels A-C, the three figures plotted using the samples (a)-(c) all present a continuous trend (linear or quadratic) with no obvious discontinuity in 1982. In other words, conditioning on duration of stay in the U.S., all the post-1982 cohorts have achieved the same level of citizenship probability as the pre-1982 cohorts. This implies that many post-1982 immigrants in samples (a)-(c) have adjusted to Legal Permanent Resident (LPR) status (and eventually citizens) without being eligible for the IRCA. They either came legally or have LPR or citizen family members. Therefore, these people make the discontinuity in 1982 undetectable in panels A-C. The figure plotted using sample (d) in Panel D does not have a clear trend. Yet, one can see that the citizen ratios of the 1978–1985 cohorts are flat and have no discontinuity in 1982. This reflects the impact of the IRCA’s SAW program, which legalized post-1982 agricultural workers.

Therefore, the contrast between the working sample and samples (a)-(d) indicates that it is necessary to have the restrictions in order to exclude a lot of immigrants who could have legality without the IRCA’s LAW program. Having these people in the analysis sample can dissipate the estimated impact of the IRCA. Footnote 16 As a placebo test, I use the empirical model in equation (1) to estimat the impact of arriving before 1982 on the natural log of male wage using sample a-d. The results, presented in Appendix Table 9, show that none of the samples shows that pre-1982 male immigrants have higher wages than post-1982 male immigrants.

B. Explanation of the Variables in Use

  1. 1.

    Definitions of Wage and Employment Probability

    Wage is defined as wage income per year divided by hours of work per year. Employment Probability is a binary indicator, which is one if women’s wage income is positive and zero otherwise.

  2. 2.

    Measurement of English-speaking Ability

    In the Census data, an individual’s English-speaking ability is defined by one of the following five cases: (1) speaks only English; (2) speaks very well; (3) speaks well; (4) speaks English but not well; or (5) does not speak English. I construct an indicator for speaking English well, which is one if an individual is in case (1), (2), or (3), and zero otherwise.

  3. 3.

    The Definitions of the Year of Entry and Duration of Work for U.S.-born People

    For U.S.-born people who have at least 8 years of schooling, the year of entry is the year in which they finished schooling, and their duration of work can be calculated by (age - 6 - years of schooling). For those whose education is fewer than 8 years, considering that in the U.S., the minimum age for employment is 14, the year of entry is set to be the year that they reach 14, and duration is (age - 14).

Fig. 4
figure 4

Ratio of Legal Immigrant Women to Total Immigrant Women by Entry Year. Source: Pan (2009)

Fig. 5
figure 5

Citizen Ratio by Years of Entry Using Various Samples. Note: Sample (a) consists of European and Asian immigrants in the U.S. Sample (b) consists of Latino immigrants live in states other than California. Sample (c) consists of Latino immigrants who entered the U.S. at age of 14 or younger. Sample (d) consists of Latino immigrants living in rural California. The working sample consists of the Latino immigrants who entered the U.S. during 1975–1986, who were at least 15 years old at entry but younger than 46 in 1990, and who resided in California cities. Source: Census 2000 Public Use 5% Micro Sample (PUMS)

Table 8 Distribution of sampled Latino immigrants by cities
Table 9 A placebo test of IRCA impact on male wages
Table 10 Summary statistics of pre/post-1982 immigrants

Rights and permissions

Reprints and permissions

About this article

Cite this article

Pan, Y. The Impact of Legal Status on Immigrants’ Earnings and Human Capital: Evidence from the IRCA 1986. J Labor Res 33, 119–142 (2012). https://doi.org/10.1007/s12122-012-9134-0

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12122-012-9134-0

Keywords

Navigation