In an interesting new paper two doctoral candidates at the University of Maryland identify some of the political and economic factors that push some states to enforce immigration laws through 287(g) memoranda of understanding—a Republican governor, local government adoption of 287(g) agreements, an increase in the Latin@ population, and state expenditures on public welfare and police and corrections. Heather Creek & Stephen Yoder, With a Little Help From Our Feds: Understanding State/Federal Cooperation on Immigration Enforcement (2010). Heather Creek and Stephen Yoder, students in UMD’s Department of Government and Politics, presented this paper at the 2010 Annual Meeting of the American Political Science Association and posted it recently on the Social Sciences Research Network (SSRN).
Section 287(g) agreements allow local law enforcement officials, trained by ICE, to perform immigration policing functions. Pro-immigrant advocates, including the Immigration Policy Center, harshly criticize the 287(g) program.
Using a social science research method known as an event history model, the authors found that a major political factor that increased the likelihood of a state entering into a 287(g) MOA was having a Republican governor. A Republican-led state is 3.75 times more likely than a state without a Republican governor to enter into a 287(g) agreement Creek and Yoder explain. Creek & Yoder at 20. “This is likely because the governor is the only politician who is always involved in the adoption process, whether the state adopts a MOA through legislative action (governor must sign the bill), executive order (the governor issues the policy directly), or through bureaucratic means (as head of the executive branch the governor oversees the state agencies).” Creek & Yoder at 23.
An additional factor that contributes to 287(g) adoption, the pair explain, is the number of local governments in the state that enter into a 287(g) agreement independent of the state. Only Tennessee adopted a 287(g) MOA after one of its cities had already entered into an agreement with the federal government. “This finding supports the theory of ‘steam valve’ federalism at the local level: states with many localities that have already established MOAs with the federal government will be less likely to adopt state-level agreements with federal law enforcement than states whose localities do not have their own agreements with the federal government.” Creek & Yoder at 23-24.
A third and rather concerning factor that Creek and Yoder identified is an increased percentage of Latin@s in the state. “A one percentage point increase in a state’s Hispanic population is associated with a five percentage point increase in the hazard ratio.” Creek & Yoder at 21. “If the value of the hazard ratio is greater than 1, then the probability of adoption is increasing; if the hazard ratio is less than 1, then the probability of adoption is decreasing.” Creek & Yoder at 20. According to the authors, this finding may reflect “conflict theory, that holds that as an outgroup increases its numbers, the majority ingroup acts in opposition.” Creek & Yoder at 24.
On the economic front, the pair found that the amount of money a state spends on public welfare and police and corrections are the principal economic variables that influence 287(g) adoption. Creek & Yoder at 24. The authors do not conclusively determine why this is. Interestingly, a state’s unemployment rate is not a statistically significant factor. Creek & Yoder at 22.
Though Creek and Yoder are not able to explain why each of these factors does or does not impact a state’s adoption of a 287(g) agreement, this paper is a useful read in identifying the critical factors that portend local attempts to police immigration laws.