The undocumented immigrant population in the United States grew from approximately four million at the time of the last major immigration reform in 1986 to an estimated 12 million in 2008. In the absence of new immigration reform at the federal level, some states have created their own interior enforcement mechanisms to deter the employment of undocumented immigrants. Arizona’s Legal Arizona Workers Act (LAWA) is perhaps the most restrictive of those state efforts. The law, in effect since January of 2008, prohibits a person or entity from “knowingly hiring, recruiting or referring for a fee an unauthorized immigrant.” In addition, the law classifies as a felony the taking of another person’s identity for employment and requires the suspension or revocation of the license of any business found to have knowingly hired an unauthorized immigrant after December 31, 2009. Does state-level legislation against the hiring of undocumented workers reduce the number of undocumented immigrants? What has been the impact of this kind of legislation on labor market outcomes? And at what cost to the states’ economies? To date, there has not been any systematic research on those issues that would inform debates on whether or not state-level legislation targeted at employers effectively curbs the hiring of undocumented workers.
With an award from Russell Sage, Magnus Lofstrom, Sarah Bohn (both of the Public Policy Institute of California), and Steven Raphael (University of California, Berkeley) propose to use individual-level data from the Current Population Survey (CPS) for the period from 2006 to 2009 as well as data from the 2006 to 2008 American Community Survey (ACS) to analyze the labor market impacts of state-level legislation, such as LAWA in Arizona. Specifically, they propose to assess whether such legislation reduces employment levels, individual employment probabilities, and wages for foreign-born non-citizens most likely to be undocumented. Because neither the CPS nor the ACS allows for identification of undocumented workers at the individual level, they define the undocumented as members of groups likely to include high proportions of undocumented workers, i.e., younger, recently arrived, non-citizen immigrants of Mexican origin, with less than a high school degree residing in metropolitan areas. The investigators will also assess whether the labor-market outcomes of documented workers, such as naturalized immigrants or the native-born, are impacted by these laws.
Lofstrom, Bohn, and Raphael will use a simple but powerful econometric technique, the difference-in-differences approach (DiD), to model the effects of the new law. Specifically, they will contrast the employment outcomes for different groups of adults (defined by education, race/ethnicity, nativity, and legal immigrant status) within the state of Arizona. The investigators will also run a series of comparisons across states. The advantage of this strategy is that it assumes that the Arizona legislation is unlikely to have a direct impact on the labor market outcomes of comparable groups of workers in neighboring states. The disadvantage is that the comparison groups, being residents in different states, may be subject to different economic conditions, compared to workers in Arizona. Lofstrom, Bohn, and Raphael argue that the results of the two strategies (within-state and across-states comparisons), when presented side-by-side, will allow them to determine whether LAWA had a measureable impact on the Arizona labor market. Publications of the results include a report for the Public Policy Institute of California, policy briefs, and journal articles.