“Trust between law enforcement agencies and the people they protect and serve is essential in a democracy. It is key to the stability of our communities, the integrity of our criminal justice system, and the safe and effective delivery of policing services.”1
– The President’s Task Force on 21st Century Policing, 2015
Law enforcement personnel across the nation today are facing complex challenges. Whether it is through interacting with individuals who have mental health or substance abuse issues, dealing with gun crimes, or responding to domestic violence, officers must address difficult realities on a daily basis. In the midst of heightened racial and ethnic tensions, they must tackle these challenges with constrained resources, including limited training, inadequate and outdated police equipment, enormous technological and data gaps, uneven recruitment, and reduced support services for officers.2
But on top of law enforcement officers’ primary mission of keeping their communities safe, President Donald Trump has consistently called for the creation of a “deportation force” to maximize the number of immigrants removed from the country and has proposed a range of efforts that would supercharge the role of state and local law enforcement agencies, or LEAs, in federal immigration enforcement.3 From signaling plans to aggressively promote the 287(g) program around the country to withholding federal grants from so-called sanctuary jurisdictions, the Trump administration has made clear that it aims to enlist state and local law enforcement in its civil immigration enforcement efforts through both inducement and coercion.4
State and local elected officials and LEAs are now, or soon will be, faced with a choice: whether and how to assume a greater role in enforcing federal immigration laws. As this issue brief illustrates, exercising that role could lead to significant financial burdens, increased litigation, and diminished public trust—all at the expense of public safety and the general welfare of all members of U.S. communities.
A brief history of the 287(g) program and sanctuary policies
The 287(g) program
Section 287(g) was added to the Immigration and Nationality Act in 1996. It authorizes state, county, and local LEAs to perform federal immigration enforcement duties pursuant to written memorandums of agreement, or MOAs, with U.S. Immigration and Customs Enforcement, or ICE.5
Historically, 287(g) agreements have taken three forms: 1) task force agreements that allow local law enforcement officers to perform immigration enforcement functions in their communities; 2) jail enforcement agreements that allow officers to question and screen individuals already being held in jail; and 3) hybrid agreements that combine the features of task force and jail enforcement agreements.6
The first 287(g) agreement was signed in 2002, shortly after the September 11 terrorist attacks, and the program grew from there. In 2007, ICE signed 26 new MOAs and added another 28 in 2008; by May 2009, there were 66 active MOAs.7
In 2010, the Office of Inspector General at the U.S. Department of Homeland Security, or DHS, identified serious concerns about the program and noted that the lack of training and oversight provided to task force jurisdictions by ICE increased the risk of civil rights violations.8 Partly in response to this and other developments, the Obama administration phased out the use of task force and hybrid agreements at the end of 2012.9 The Trump administration, however, hopes to increase the number of 287(g) agreements signed nationwide, as well as reintroduce the task force and hybrid models.10
While there is no one definition, sanctuary jurisdictions generally are those that have adopted resolutions, ordinances, laws, or policies that limit local law enforcement’s involvement in federal immigration enforcement efforts. These policies are intended to facilitate trust and increase public safety by ensuring that one’s immigration status will not affect the ability to, for example, come forward as a witness or victim of a crime. Overall, more than 600 cities and counties and a few states have adopted sanctuary policies.11
The number of sanctuary jurisdictions—particularly those that limit when people may be held in custody based solely upon an ICE detainer or a request to hold someone past the point at which they would ordinarily be released—increased in response to the Secure Communities program. This program was launched under the George W. Bush administration and expanded nationwide under the Obama administration. Through Secure Communities, federal immigration officials received fingerprint information collected during booking by state and local LEAs and shared it with the Federal Bureau of Investigation for use in criminal background checks. With this information, ICE significantly increased its issuance of detainer requests.12
Other jurisdictions adopted sanctuary policies to reduce their liability exposure after a series of federal court rulings on ICE detainer requests. These rulings call into question whether depriving people of liberty based on an ICE detainer request violates their constitutional rights to due process and to be free of unlawful seizure.13
How local entanglement in federal immigration enforcement harms communities
In the coming months, the Trump administration is expected to conduct significant outreach to states and localities to enter into new 287(g) agreements or to add task forces onto existing jail enforcement agreements. The administration is also expected to make some effort to deny federal grants to jurisdictions that the Secretary of DHS—in his discretion—deems to be sanctuary jurisdictions.14
Such efforts fly in the face of law enforcement best practices, particularly those of community policing. While there is no universal definition of community policing, one common tenet is that “[r]educing crime and disorder requires that the police work cooperatively with people in neighborhoods to (1) identify their concerns, (2) solicit their help, and (3) solve their problems.” Failure to maintain trust and open lines of communication with the public results in an unwillingness to cooperate or share information.15
Research shows that the general principles of community policing apply with equal force when looking at effective policing in immigrant communities. A study by the University of Illinois at Chicago found that Latinos, whether citizens or noncitizens, are “less likely to volunteer information about crimes because they fear getting caught in the web of immigration enforcement themselves or bringing unwanted attention to their family or friends.” The study further found that 70 percent of unauthorized immigrants and 44 percent of Latinos are less likely to communicate with law enforcement if they believe officers will question their immigration status or that of people they know.16 Reviewing several recent instances in which unauthorized immigrants assisted law enforcement as victims or witnesses to crime, Chuck Wexler, executive director of the Police Executive Research Forum, observed: “Had these undocumented people, and countless others in cities across America, not stepped forward to report crime and cooperate with the police, we would have more dangerous offenders committing more crime—and more serious crime—against innocent victims.”17
287(g) agreements are expensive and litigation-prone and undermine public safety and community trust
In deciding how to respond to the Trump administration’s 287(g) efforts, jurisdictions need to consider how limited resources and potential increases in racial profiling and litigation will affect their primary mission of ensuring public safety.
Strains on already limited resources
Under the 287(g) program, participating jurisdictions perform federal immigration enforcement functions largely at their own expense.18 While ICE may provide some reimbursement for detention costs in some cases, it provides minimal funding for training and information technology equipment and services. Local agencies are entirely responsible for the salaries and benefits of their personnel as well as travel costs, housing, and per diem for required training.19
Many jurisdictions that have participated in 287(g) programs have found them to be a raw deal. A 2009 study conducted by the Brookings Institution found that Prince William County, Virginia, had to increase property taxes and pull money from its rainy day fund to implement its 287(g) program. Overall, the program cost the county $6.4 million in its first year and was estimated to cost $26 million over five years.20 An analysis of North Carolina, the state with the highest nationwide number of 287(g) jurisdictions, found that the program cost Mecklenburg and Alamance counties together more than $10 million in the first year alone.21 Sheriff Ed Gonzalez in Harris County, Texas, recently terminated his 287(g) agreement to make better use of the $675,000 in associated salary expenses; according to the sheriff, incorporating ICE-trained deputies elsewhere in the jail complex could reduce the $1 million in overtime costs the county incurs every two weeks managing the overcrowded facility.22
Increased racial profiling and litigation
A frequent complaint against jurisdictions operating 287(g) task force agreements was the broad discretion provided to law enforcement to detain and arrest people believed to be in the country unlawfully created opportunities for abuse, especially when not closely monitored. Racial profiling—the practice of targeting a set of individuals based on their race, ethnicity, religion, or national origin—has long been a troubling policing tactic.23 Even with proper training, law enforcement officers, as all humans, are susceptible to relying on racial or ethnic stereotypes. Because racial profiling worsens already strained relationships with communities of color, it also makes it harder for law enforcement to build lasting trust. Mayors and other elected officials, police chiefs, and residents understand that without that trust, public safety is diminished.
Perhaps the most prominent case of discriminatory policing by a 287(g) jurisdiction involved the Maricopa County Sheriff’s Office, or MCSO, under Sheriff Joe Arpaio. In 2006, the MCSO entered into an MOA to operate a 287(g) task force, which was scaled back to a jail enforcement agreement in 2009. In 2011, the U.S. Department of Justice found that the MCSO engaged in discriminatory policing practices, leading the DHS to terminate the agreement.24 A federal court later ruled that Maricopa County engaged in unconstitutional racial profiling and pretextual stops.25 In addition, a 2008 Goldwater Institute study found that the MCSO’s significant investment in immigration enforcement efforts detracted from the office’s public safety mission and blew an enormous hole in its budget.26
The MCSO’s unlawful and discriminatory conduct was determined to have been intentional.27 But even without meaning to, any LEA performing immigration enforcement functions with insufficient oversight, training, and guidance is at a heightened risk of making unlawful stops of individuals who look or sound foreign. Such conduct also creates a heightened risk of litigation. Furthermore, it increases the likelihood that people of color will stop communicating and working with their local LEAs to solve crimes.
Sanctuary policies reduce litigation risks and are associated with lower crime rates
Faced with the threat of losing important federal funds, some sanctuary jurisdictions may be considering changes to their policies, and other jurisdictions may be reconsidering whether to adopt such policies. In making those choices, jurisdictions should consider the fiscal and public safety costs associated with increased cooperation, as well as the likelihood that the administration will overcome significant legal obstacles to withholding federal funds.
Reduced litigation risks due to unlawful detention
The proliferation of sanctuary policies limiting when local LEAs can hold people pursuant to ICE detainer requests can be traced, in part, to a growing body of lawsuits that have resulted in court judgments and hefty settlements. Many of these cases have concluded that warrantless ICE detainer requests fail to provide the probable cause required under Fourth Amendment jurisprudence to allow a state or locality to deprive a person of liberty.28 The city of Allentown, Pennsylvania, and Lehigh County, Pennsylvania, for example, settled with a U.S. citizen who was detained for three days pursuant to a detainer request that ICE mistakenly issued. The lawsuit cost the city $25,000 and the county $95,000.29 Salt Lake County, Utah, paid $75,000 to settle a case brought by an individual held on an ICE detainer for 46 days after he had posted bail.30 Clackamas County, Oregon, settled a case for $30,100 brought by a woman detained for two weeks based solely on an ICE detainer request.31
In light of the Trump administration’s threat to punish sanctuary jurisdictions by withholding federal funds—discussed in the next section—the constitutional deficiencies with ICE’s current detainer practices place law enforcement agencies in a difficult place. Jurisdictions that respond to the threat by deciding to hold people regularly pursuant to ICE detainers will be more likely to encounter legal challenges.32 These litigation risks should be taken seriously when jurisdictions determine what detainer policy is in the best interest of public safety and overall resource management.
Lower crime rates
The Trump administration has attempted to justify a crackdown on sanctuary policies by claiming that such jurisdictions “have caused immeasurable harm to the American people and to the very fabric of our Republic.”33 Yet research finds that, on average, there are 35.5 fewer crimes per 10,000 people in sanctuary counties than in non-sanctuary counties.34 According to the Major Cities Chiefs Association, or MCCA, entangling local law enforcement with federal immigration enforcement “would result in increased crime against immigrants and in the broader community, create a class of silent victims and eliminate the potential for assistance from immigrants in solving crimes or preventing future terroristic acts.”35 The MCCA has consistently cited five major areas of concern with linking local law enforcement with federal immigration enforcement, including undermined community trust, lack of resources, and overly complex and time-consuming training that detracts from the public safety mission.36
Threats to withhold federal funding from sanctuary cities may be legally impermissible
President Trump’s January 25 executive order purports to allow the government to withhold federal grant money from sanctuary jurisdictions. It also broadly directs the attorney general to “take appropriate enforcement action” against any entity with a policy that “prevents or hinders the enforcement of [f]ederal law.”37 However, the administration’s threats to withhold funding from or otherwise punish sanctuary jurisdictions will be constrained by longstanding 10th Amendment jurisprudence that prevents the federal government from commandeering state and local resources.38 Additionally, Supreme Court precedent under the spending clause limits when, under what circumstances, and to what extent the federal government can impose conditions on the granting of federal funds.39
Although the administration has not yet described the specific funds it would seek to withhold, in recent years, congressional Republicans have attempted to pass legislation that would jeopardize several law enforcement, economic development, and community development grant programs.40 This sledgehammer approach would have broad and dangerous consequences for jurisdictions that are already feeling the pinch of smaller budgets—which in turn would make communities less safe and less prosperous.41
Today, law enforcement is faced with many challenges that are beyond the scope of their primary duties and functions. Efforts to increase police involvement in federal immigration enforcement illustrate the Trump administration’s disconnect with the real challenges facing local law enforcement. Instead of working to meet these growing challenges, the administration is choosing to add to them by pursuing an agenda guided by anti-immigrant ideologues such as Kris Kobach and Stephen Bannon—not by community priorities, public safety data, or research.42
As a candidate and as president, Trump has labeled immigrants as criminals who pose a threat to public safety.43 These statements fly in the face of a range of evidence that immigrants are actually less likely to commit serious crimes than the U.S.-born and that increases in immigration are associated with decreases in crime.44 Although the country badly needs a responsible solution to our outdated immigration system, as former Los Angeles Police Chief William Bratton wrote in 2009, “[T]he solution isn’t turning every local police department into an arm of Immigration and Customs Enforcement.”45 Given the evidence presented in this brief, local law enforcement should think twice about engaging in federal immigration enforcement.
Danyelle Solomon is the Director of Progress 2050 at the Center for American Progress. Tom Jawetz is the Vice President for Immigration Policy at the Center. Sanam Malik is the Research Assistant for Immigration Policy at the Center.