The 287(g) Program: An Overview

The 287(g) program is named for Section 287(g) of the Immigration and Nationality Act (INA) and became law as part of the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (IIRAIRA). Through the 287(g) program, state and local law enforcement officers collaborate with the federal government to enforce federal immigration laws. In the past, the 287(g) program has been costly for localities, has historically targeted individuals with little or no criminal history, and has harmed the relationship between police and local communities. This fact sheet provides an overview of how the 287(g) program works and discusses some of the problems associated with its operation.  

How does the 287(g) program work?

Section 287(g) of the INA allows the Department of Homeland Security (DHS) to enter into formal written agreements (Memoranda of Agreement or MOAs) with state or local law enforcement agencies and deputize selected state and local law enforcement officers to perform certain functions of federal immigration agents. The MOAs are negotiated between DHS and the local authorities and include delegation of authority to a limited number of state and local officers. All of this must be done under the supervision of Immigration and Customs Enforcement (ICE). Deputized officers are required to abide by federal civil rights laws and regulations. In general, deputized officers are authorized to:

• interview individuals to ascertain their immigration status;
• check DHS databases for information on individuals;
• issue immigration detainers to hold individuals until ICE takes custody;
• enter data into ICE’s database and case management system;
• issue a Notice to Appear (NTA), which is the official charging document that begins the removal process;
• make recommendations for voluntary departure in place of formal removal proceedings;
• make recommendations for detention and immigration bond; and
• transfer noncitizens into ICE custody.

After a 287(g) agreement expires, DHS is not legally obligated to renew it. However, not all 287(g) agreements include a specific expiration date.Once a 287(g) agreement is entered into, it may be terminated at any time by either party.

What types of 287(g) agreements exist?

Currently, only two types of 287(g) agreements are employed in the field: the “jail enforcement” model and the “warrant service officer” model.

• Under the jail enforcement model, deputized officers may interrogate suspected noncitizens who have been arrested on state or local charges regarding their immigration status and may place immigration detainers on those thought to be subject to removal.
• Under the warrant service officer (WSO) model, ICE trains, certifies, and authorizes selected state and local law enforcement officers to execute ICE administrative warrants. These officers are permitted to perform the arrest functions of an immigration officer within the law enforcement agency’s jails and/or correctional facilities. The WSO model differs from the jail enforcement model in that ICE does not authorize the local law enforcement officer to interrogate alleged noncitizens about their immigration status.

Two additional types of 287(g) agreements—the “task force” model and the “hybrid” model—have been used in the past. However, they were both discontinued following an ICE policy memo in 2012 that stated other enforcement programs were a “more efficient use of resources for focusing on priority cases.”

• Under the task force model, deputized officers who encountered alleged noncitizens during the course of daily activities could question and arrest individuals they believed had violated federal immigration laws.
• The hybrid model combined functions of the task force model and the jail enforcement model. Task force officers could initiate immigration processing and transfer individuals thought to be subject to removal to 287(g) jail officers who completed immigration screening and ICE paperwork requirements.

How has the 287(g) program evolved over time?

In 2009, ICE renegotiated all existing MOAs using a new, standardized template. In 2013, the MOA was revised based on a recommendation from the Office of Inspector General to update language that did not clearly specify program requirements or provide a measurable standard for assessing compliance. In 2020, 75 jail enforcement MOAs were renewed by ICE with revised language that removed end dates from the 287(g) agreements and reduced requirements for designated officers.

Due to mounting concern over the impact of the 287(g) program on immigrants and police-community relations, and in response to the growth of the Secure Communities program, the task force and hybrid models were discontinued. As of June 2021, ICE had 146 contracts with law enforcement agencies: 76 warrant service model MOAs in 11 states and 70 of the jail enforcement variety in 20 states.  

Federal funding for the 287(g) program hit a high in Fiscal Years (FY) 2010–2013 (Table 1). Annual appropriations cover the cost of training 287(g) officers and fund the program’s management and oversight. Given the subsequent scaling back of the program, the request for appropriations decreased in FY 2014.

Table 1: Federal Funding for the 287(g) Program

Sources: William A. Kandel, Interior Immigration Enforcement: Criminal Alien Programs
(Washington, DC: Congressional Research Service, 2016), 17, https://fas.org/sgp/crs/homesec/R44627.pdf; U.S. Congress, Congressional Record, 115th Congress, 2nd session, 2018, vol. 164, pt. 508; U.S. Congress, House of Representatives, Department of Homeland Security Appropriations Bill, Report (to accompany H.R. 5634), 114th Congress, 2nd session, 2017, H. Rep. 114-668, 36, https://www.congress.gov/114/crpt/hrpt668/CRPT-114hrpt668.pdf; U.S. Department of Homeland Security, 287(g) End-of-Year Report: Fiscal Year 2019 Report to Congress, https://www.dhs.gov/sites/default/files/publications/ice_-_287g_end-of-year_report.pdf.

On the campaign trail, President Biden’s Agenda for the Latino Community included a commitment to void all 287(g) agreements executed by the Trump administration and to minimize the role of previous 287(g) agreements and all similar programs that enable or require law enforcement officials to carry out the duties of immigration officers.As of June 2021, only one 287(g) agreement (with the Bristol County Sheriff’s Office in Massachusetts) had been terminated by DHS. The Biden administration’s FY 2022 budget request for DHS also did not seek any cuts to the 287(g) program.

What training do deputized officers receive?

Under the jail enforcement MOAs renewed by ICE in 2020, deputized 287(g) officers must be U.S. citizens and pass a background check conducted by ICE. Any officer deputized under the program must complete a four-week Immigration Authority Delegation Program at the Federal Law Enforcement Training Center (FLETC) ICE Academy (ICEA) in Charleston, South Carolina. According to ICE, the training program teaches officers about “immigration law, the use of ICE databases, multi-cultural communication and the avoidance of racial profiling.”  At the end of the training program, officers must pass examinations with a minimum score of 70% in order to become deputized.

Deputized officers can complete refresher training “as needed” with no mandatory minimum requirement and a maximum frequency of once every two years. In addition, they are not required to make a two-year commitment to the position, which was a requirement prior to the 2020 MOA revisions, allowing law enforcement agencies to move officers in and out of the 287(g) program with greater flexibility.

Training requirements are further reduced for officers who participate in the WSO Model, which requires only an 8-hour training on legal authorities and enforcement protocols facilitated by an ICE Field Office.

Are localities responsible for 287(g) program costs?

ICE covers the cost of training deputized officers in the 287(g) program, including travel and expenses. In addition, ICE is responsible for the installation and maintenance of the Information Technology (IT) infrastructure at partner law enforcement agency sites. State and local governments are responsible for all personnel costs, including salaries, benefits, and overtime, as well as all administrative supplies. Some of the costs of detention may be reimbursed by the federal government through the State Criminal Alien Assistance Program (SCAAP). However, the federal government has never fully funded SCAAP, and reimbursements only cover a fraction of the costs incurred by states and localities.

How are jurisdictions chosen for the 287(g) program?

Interested law enforcement agencies can send an inquiry about participation in the 287(g) program to ICE via email. According to ICE, the agency will then begin the process of evaluating the interested agency under its standards, including whether it has the capability to act as an ICE force multiplier, and determine whether to sign a 287(g) agreement with the law enforcement agency. However, a January 2021 Government Accountability Office (GAO) report found that ICE was not considering the optimal use of resources and program benefits in the selection of future 287(g) participants, instead focusing primarily on signing as many agreements as possible regardless of the benefits.

Amid growing concerns about how ICE vetted and chose local law enforcement agencies for participation in the 287(g) program, in 2009 ICE established an Internal Advisory Committee to review and assess ICE field office recommendations about pending 287(g) applications. The committee includes representatives from DHS’ Office for Civil Rights and Civil Liberties (CRCL) and ICE’s Office of Investigations, Office of Detention and Removal Operations, Office of Principle Legal Advisor, Office of Congressional Relations, and Office of Public Affairs. The presence of CRCL is intended to ensure that DHS takes into account a local jurisdiction’s civil rights and civil liberties history before signing a 287(g) MOA. However, based on the widespread instances of racial profiling discussed below, this has not proven effective at ensuring that local law enforcement agencies abide by civil rights law while acting as federal immigration agents.

How does the 287(g) program work in conjunction with other federal-local programs?

DHS maintains several different partnerships with state and local police agencies aimed at identifying deportable individuals. In addition to the 287(g) program, DHS administers the following programs:

• The Criminal Alien Program (CAP) is a program that gives ICE access to jails and prisons where ICE officers screen and interview individuals in order to identify noncitizens for deportation. In 2020, ICE made 42,796 CAP removals.
• Secure Communities is a program through which the fingerprints of all persons arrested by state or local police are submitted to DHS and checked against immigration databases. If ICE believes an individual identified through Secure Communities is deportable, it can take custody of them or detain them directly from the law enforcement agency’s custody.

When a person is identified and taken into ICE custody, it can be difficult to determine which program was responsible. For example, when there is a Secure Communities “hit,” a CAP officer may go to the prison or jail and conduct an interview with the individual to determine whether he or she is removable. Or if the jail has a 287(g) program, a deputized officer may perform the same function.

What problems have the 287(g) program caused?

287(g) agreements have resulted in widespread racial profiling

• A 2011 investigation by the Department of Justice concluded that the Maricopa County Sheriff’s Office in Arizona engaged in a pattern and practice of constitutional violations, including racial profiling of Latinos, after entering a 287(g) agreement. For example, the investigation found that deputies of former Sheriff Joe Arpaio routinely conducted “sweeps” in Latino neighborhoods, and that Latino drivers in certain parts of Maricopa County were up to nine times more likely to be stopped than non-Latino drivers. Following the investigation, the Obama administration terminated Maricopa County’s 287(g) agreement.

• A separate Justice Department investigation concluded in 2012 that the Alamance County Sheriff’s Office in North Carolina engaged in a pattern and practice of constitutional violations by unlawfully detaining and arresting Latinos. The investigation found that the sheriff’s deputies set up checkpoints at entrances to Latino neighborhoods; that Latino drivers were up to 10 times more likely to be stopped than non-Latino drivers; and that Latino drivers were often arrested for traffic violations for which non-Latino drivers received only citations. Following the investigation, the Department of Justice brought a lawsuit against the department for racial profiling, DHS terminated Alamance County’s participation in the 287(g) agreement, and the county has never rejoined the 287(g) program.

287(g) partnerships vary by local jurisdiction and primarily lead to the arrest of individuals with minor criminal histories 

• The Migration Policy Institute (MPI) conducted a comprehensive analysis of the 287(g) program in 2011, at a time when the task force model was in more common use, and found that it did not target individuals with serious criminal convictions. Half of all detainers issued through the program were for people who had committed misdemeanors and traffic offenses.
• MPI found that some jurisdictions “target” their programs to identify individuals with serious criminal convictions. Other jurisdictions operate a “universal” model, designed to identify as many undocumented immigrants as possible, regardless of criminal history. These universal models were concentrated in the Southeast of the United States. 
• MPI also found that the 287(g) program can be a tool for localities pursuing anti-immigrant agendas. Sheriffs, who are generally elected officials, can use the program to meet political goals and respond to community pressure to “crack down” on immigration. In southeastern states, where the “universal” model is prevalent, the immigrant population grew rapidly in the 1990s and 2000s, causing backlash and putting pressure on elected officials to reduce undocumented immigration. 
• Studies in 2009 and 2010 by the University of North Carolina at Chapel Hill found that 287(g) agreements under the task force model in the state were primarily used to target offenders who posed no threat to public safety or individuals with no criminal record. In that state, a large share of individuals arrested by 287(g) officers in Gaston, Mecklenburg, and Alamance counties were arrested for traffic violations, such as driving without a license and speeding. Overall, 33 percent of individuals detained through the 287(g) program were charged with traffic violations; in Gaston County the figure rose to 57 percent.
• The availability of resources and detention space also determines how the 287(g) program is utilized. A large detention capacity allows for more resources to be spent on individuals with little or no criminal history.

287(g) agreements can be expensive for localities 

• In February 2017, Sheriff Ed Gonzalez announced that Harris County, Texas, would terminate its 287(g) agreement, saying that the decision was a resource allocation issue. The sheriff said he would put the $675,000 the county spent on the program toward improving clearance rates of major crimes and other priorities.
• Waukesha County, Wisconsin, Sheriff Eric Severson testified at a February 2017 Senate hearing that he did not have the resources for 287(g). When Senator Claire McCaskill asked him whether he planned to apply for the 287(g) program, Sheriff Severson replied: 

At this time, I don't have the resources to participate with that nor is our community structure such that I don't know if that is necessarily a high priority for us right now and again I am fortunate enough to report to you today that the instances of criminal activity of illegally present immigrants beyond their status is relatively uncommon in my County and generally in Wisconsin, it is less common than some other communities.

• A 2010 report by the University of North Carolina at Chapel Hill found that the first year of operating the 287(g) program in Mecklenburg County, North Carolina, cost a total of $5.3 million. Meanwhile, the first full year of operation in the state’s Alamance County cost $4.8 million. 
• A 2016 study by the Brookings Institution found that Prince William County, Virginia, had to raise property taxes and take money from its “rainy day” fund to implement its 287(g) program. The report found the program cost $6.4 million in its first year and would cost $26 million over five years. To cut costs, the county slashed $3.1 million from its budget—money that was intended to buy video cameras for police cars to protect against allegations of racial profiling.
• Before DHS revoked its 287(g) agreement with Maricopa County, Arizona, in 2011, Sheriff Joe Arpaio’s office created a $1.3 million deficit in just three months, much of it due to overtime. 

ICE does not provide sufficient guidance, direction, or supervision

• A January 2021 GAO report found that ICE had failed to establish performance goals for the 287(g) program and did not have measures in place to systematically assess its effectiveness. Without a set of performance measures that are actively tracked and monitored, the report concluded that ICE would be unable to provide effective oversight over law enforcement agency partners since there is no standard to determine their compliance with the MOA.
• A March 2010 report by the DHS Office of Inspector General (OIG) found that ICE and its local law enforcement partners had not complied with the terms of their 287(g) agreements; that the standards by which deputized officers were evaluated contradicted the stated objectives of the 287(g) program; that the program was poorly supervised by ICE; and that additional oversight was necessary.
• According to the DHS OIG, ICE has not planned adequately before approving new 287(g) participants. In September 2018, the OIG reported that law enforcement trainings were inefficient, and that ICE was not monitoring officers through the completion of the required training. Although the participating local enforcement agencies were expected to function under ICE supervision, the program managers were stretched thin while assigned to oversee several jurisdictions spread over thousands of miles. This poor planning hindered ICE’s ability to adequately manage, oversee, and educate participating agencies in enforcing immigration laws correctly.
• DHS’s budget request for Fiscal Year 2022 included funding requests for the 287(g) program but acknowledged its shortcomings. It stated:

While the 287(g) program has yielded successes, ICE recognizes the program is not universally regarded as the most effective or appropriate model for all stakeholders or in every jurisdiction.

287(g) agreements threaten community safety and hinder community policing

• In 2010, the Police Executive Research Forum interviewed Maricopa County law enforcement executives who stated that by enforcing federal immigration law, the sheriff’s office damaged the relationship between law enforcement and the Latino community. 
• A 2009 study by the University of North Carolina School of Law and the American Civil Liberties Union of North Carolina (ACLU-NC) found that “287(g) encourages, or at the very least tolerates, racial profiling and baseless stereotyping, resulting in the harassment of local residents and the isolation of an increasingly marginalized community.” 
• In 2005, the International Association of Chiefs of Police (IACP), the nation’s premier law enforcement association, stated: 

Local police agencies depend on the cooperation of immigrants, legal and [otherwise], in solving all sorts of crimes and in the maintenance of public order. Without assurances that they will not be subject to an immigration investigation and possible deportation, many immigrants with critical information would not come forward, even when heinous crimes are committed against them or their families. 

• In 2019, the Major Cities Chiefs Association (MCCA), a group of police chiefs from the 78 largest police departments in the United States and Canada, found that “without assurances that contact with the police would not result in purely civil immigration enforcement action, the hard-won trust, communication and cooperation from the immigrant community would disappear.”