Research spotlight: Local immigration enforcement increases US president’s power

When presidents delegate authority to local governments, they don’t necessarily lose power – they may gain it. But there’s a catch: the local partners most eager to help are often extremist outliers willing to bend or break the rules.

These are the key findings from new research by Asya Magazinnik, Professor of Social Data Science at the Data Science Lab. Magazinnik’s study of US immigration enforcement reveals how federal-local cooperation can simultaneously strengthen presidential authority while attracting problematic local allies. Her paper, “An Agency Perspective on Immigration Federalism”, published in the Journal of Politics, deepens our understanding of power-sharing in federal systems.

Federal-local cooperation in immigration enforcement is nothing new

President Trump’s plan to dramatically expand the policing, detention and deportation of immigrants relies on the cooperation of local law enforcement agencies – in particular, county sheriffs. But the infrastructure required for intergovernmental cooperation in immigration enforcement is not new: its foundations reach back to the post-9/11 national security reforms under President George W. Bush, which every subsequent administration has in some way further developed.

Magazinnik’s research focusses on the 287(g) programme, which allows local law enforcement to be deputised with the powers of federal Immigration and Customs Enforcement (ICE) agents, including the authority to demand immigration documentation in the course of routine policing (including traffic stops), to detain people in local jails to investigate their immigration status, and to initiate deportations.

Involving local law enforcement can strengthen presidential power

To understand how this “cooperative federalism” in immigration enforcement came into being, and how it transformed the policy landscape, Magazinnik develops a game theoretic model that yields predictions about when presidents will decide to involve local governments in policymaking as opposed to unilaterally issuing mandates from above. Her analysis highlights that extending authority downward is by no means in tension with increasing the power and influence of the president: when the incentives of national and subnational government actors are sufficiently aligned, the president can strategically use local participation to promote his own goals.

“Participation in 287(g) led to a 213% increase in immigration detainers.”

But the model also reveals an important trade-off for presidents to manage. While local authorities can usefully contribute resources toward costly policymaking, the jurisdictions that will voluntarily choose to do so are likely to be “preference outliers”, for instance, sheriffs with extremist views who engage in unconstitutional policing while contributing to federal enforcement efforts.

Dramatic increase in detentions, primarily of people with no criminal record

Professor Magazinnik’s empirical analysis relies on a comprehensive dataset of “immigration detainers” – requests issued by ICE for local authorities to hold people in jail in order to investigate their immigration status – as well as the degree to which local agencies cooperated with these requests. The analysis reveals several striking patterns:

• Self-selection of restrictionist jurisdictions: Counties that voluntarily signed 287(g) agreements already demonstrated significantly higher immigration enforcement activity in the years leading up to their entry into the programme.
• Quantifiable enforcement increases: Participation in 287(g) led to a 213% increase in immigration detainers compared with not participating in the programme.
• Competing enforcement priorities: Contrary to federal priorities that emphasised targeting individuals with serious criminal histories, counties participating in 287(g) focussed overwhelmingly on individuals with no criminal record (213% increase) or minor misdemeanour charges (11% increase), with no significant impact on those with felony charges.
• Federal-local tensions: The research documents how the Obama administration struggled to maintain control over local enforcement priorities, particularly in jurisdictions like Arizona’s Maricopa County under Sheriff Joe Arpaio, which engaged in what the Department of Justice described as one of the worst regimes of racial profiling in US history.

Implications beyond immigration policy

Professor Magazinnik’s findings have important implications for understanding policy implementation in federal systems:

• Policy inconsistency: Regimes that rely on voluntary local participation lead to inconsistent application of national laws, creating a patchwork of enforcement that varies dramatically by jurisdiction. As Magazinnik notes, “immigrant communities living in different jurisdictions are policed and prosecuted differently under the same national law”.
• Political polarisation: The research suggests that delegating enforcement authority downward may “polarise and segregate the electorate even further”, as citizens’ preferences are shaped by policies in their local environment.
• Resource allocation challenges: The study highlights how federal executives must balance the resource benefits of local cooperation against the costs of monitoring and controlling local partners with divergent priorities.
• Broader policy implications: Magazinnik’s conceptual framework allows immigration enforcement to be compared against other policy domains that require significant resources, such as environmental regulation and climate change mitigation, where “local initiative is not a perfect substitute for coordinated national action”.

This timely research provides critical insights for policymakers and scholars seeking to understand the complex interplay between federal authority and local implementation in our increasingly polarised political landscape.

You can read Professor Asya Magazinnik’s paper in Journal of Politics here.