Content area
Full text
The record-keeping requirements of the Immigration Reform and Control Act(IRCA), and fines for illegal employment, may induce employers to discriminate against foreign-appearing workers. The General Accounting Office (GAO) reported widespread IRCA-related discrimination but did not link reported discriminatory practices to discriminatory employment behavior. We analyze the GAO's random survey and, controlling for selectivity effects, demonstrate that employers who report discriminatory practices actually employ fewer Hispanics. Although the measured reduction of Hispanic employment due to IRCA is fairly small, this finding parallels research alerting us to adverse consequences of a law that so far has achieved few of its intended effects.
The passage of the Immigration Reform and Control Act of 1986 (IRCA) marked the end of a 34-year anomaly in U.S. law known as the "Texas Proviso," which had exempted employers from legal culpability for employing unauthorized aliens (Papademetriou et al. 1991). The IRCA makes it unlawful to knowingly hire any person who cannot demonstrate authorization to work in the United States, and imposes sanctions for either record-keeping or hiring violations by an employer.
At the same time, the provisions of IRCA may unintentionally lead some employers to discriminate against foreign-appearing and potentially unauthorized persons. Many observers argue that Hispanics in particular are likely to receive fewer job offers or less favorable treatment. This concern prompted Congress to create a fast-track repeal option and to require the General Accounting Office (GAO) to report whether IRCA resulted in "increased" or "widespread" discrimination (U.S. Congress 1986).
On the basis of responses to a mail-out survey the GAO calculated that 10% of U.S. employers, who hired 2.9 million workers, reported changing their treatment of foreign-appearing workers. The GAO (1990:71) determined that it "is more reasonable to conclude that a substantial amount of discriminatory practices resulted from IRCA rather than not." Yet Congressional hearings have failed to repeal IRCA's provisions for sanctions: critics maintain that the GAO's measure of "potentially affected" workers does not demonstrate concrete instances of IRCA-related discrimination (Federal Publications 1992).
In this article we examine IRCA's unintended outcome as revealed in the GAO survey. First we consider critiques of the GAO's methodology and other available research. Then we present our analytic strategy, including a basic model of statistical discrimination and corrections for sample selectivity. Next we...