A program in flux: New priorities and implementation challenges for 287 (g

Tags: MOA, enforcement, noncitizens, immigration, immigration enforcement, US Immigration and Customs Enforcement, Task Force, County Sheriff's Office, Florida Department of Law Enforcement, template, law enforcement, the program, Immigration Authority, Paul M. Kilcoyne, State Law Enforcement Agency, Maricopa County Sheriff's Office, Integration, Immigration and Nationality Act, LEAs, illegal immigration, enforcement priorities, House Committee on Homeland Security, Migration Policy Institute, supervision, Office Georgia Department of Public Safety Gwinnett County Sheriff, Office Task Force Jail Enforcement Jail Enforcement, Jail & Task Force Task Force, Office Jail Enforcement Jail Enforcement Jail Enforcement, Office Pinal County Sheriff, Arizona Department of Corrections, York County Sheriff, Gaston County Sheriff's Office Guilford County Sheriff, Regional Domestic Security Task Force Supervisor, Immigration Law, Prince William County Sheriff's Office, Alabama Department of Public Safety, Enforcement Task Force, House Committee, Delegation of Immigration Authority, state and local law enforcement agencies, Regional Domestic Security Task Force, Federal Immigration Law Enforcement, voluntary agreements, Immigration and Nationality, law enforcement officials, Domestic Security, Los Angeles County Police Department, George Washington University, Office City of Springdale Police Department Rogers Police Department Washington County Sheriff
Content: A Program in Flux: New Priorities and Implementation Challenges for 287(g) Cristina Rodrнguez, Muzaffar Chishti, Randy Capps, and Laura St. John March 2010
Acknowledgments This report and the research underlying it were supported by the Carnegie Corporation of New York. The authors would like to thank Doris Meissner, Donald Kerwin, Laureen Laglagaron, and Margot Mendelson of the Migration Policy Institute for their guidance in the writing and editing of this report, as well as Justin Cox, formerly of Casa de Maryland. Nicole Svajlenka, a graduate student at George Washington University, developed the map included in the report. The authors would also like to thank staff from US Immigration and Customs Enforcement, especially the Office of State and Local Coordination, as well as several state and national civil-rights organizations for providing their important perspectives. © 2010 Migration Policy Institute. All Rights Reserved. No part of this publication may be reproduced or transmitted in any form by any means, electronic or mechanical, including photocopy, or any information storage and retrieval system, without permission from the Migration Policy Institute. A full-text PDF of this document is available for free download from www.migrationpolicy.org. Permission for reproducing excerpts from this report should be directed to: Permissions Department, Migration Policy Institute, 1400 16th Street, NW, Suite 300, Washington, DC 20036, or by contacting [email protected] Suggested citation: Rodrнguez, Cristina, Muzaffar Chishti, Randy Capps, and Laura St. John. 2010. A Program in Flux: New Priorities and Implementation Challenges for 287(g). Washington, DC: Migration Policy Institute.
Table of Contents Executive Summary ................................................................................................................1 I. Introduction .....................................................................................................................3 II. The History and Context of 287(g) ............................................................................5 A. The Evolution of Enforcement Priorities....................................................................................6 B. Public Critiques and Analyses of 287(g)......................................................................................8 C. 287(g) in the Larger Enforcement Context ............................................................................ 10 III. The Terms of the 2009 Template and Questions Related to its Implementation .....................................................................................................................11 A. Enforcement Priorities and Objectives .................................................................................... 12 B. Scope of Authority 14 C. Personnel and Training................................................................................................................. 16 D. Federal Supervision....................................................................................................................... 18 E. Data Collection and Reporting Requirements ....................................................................... 19 F. Mechanisms of Accountability .................................................................................................... 21 IV. Conclusion......................................................................................................................23 Appendices ............................................................................................................................24 About the Authors...............................................................................................................27 Works Cited .........................................................................................................................28
Executive Summary State and local enforcement of federal immigration laws has generated considerable controversy in public policy circles. The controversy has focused, in particular, on the Section 287(g) program. In 1996, Congress added section 287(g) to the Immigration and Nationality Act (INA), authorizing state, county, and local law enforcement agencies (LEAs) to enforce federal immigration law pursuant to agreements signed with US Immigration and Customs Enforcement (ICE). The 287(g) program aims to expand the federal government's enforcement capacities while enabling LEAs to respond directly to popular concerns regarding illegal immigration. At the same time, critics have charged that policing practices associated with the program have undermined community trust in law enforcement, led to arrests based on minor infractions, and given rise to racial profiling. Some immigrants-rights groups and law enforcement organizations have called for an end to the program. The US Government Accountability Office (GAO) also has criticized the program for its lack of proper federal oversight, inadequate data collection, and failure to articulate clear priorities. In July 2009, Homeland Security Secretary Janet Napolitano announced that the Obama administration would continue but reform the program, which had grown rapidly during the second term of the Bush administration. At the heart of the Obama administration's approach to the program is a new standardized memorandum of agreement (MOA) that will govern all future Section 287(g) collaborations. The new agreement makes a number of changes to the formal terms of the program and reflects two overarching objectives. First, it articulates a set of enforcement priorities that emphasize the identification, detention, and removal of immigrants who have been convicted of serious offenses, pose a threat to public safety, or have been ordered removed. The MOA makes clear that ICE expects LEAs to establish and follow the same priorities. Second, the new MOA seeks greater federal control over the operations of jurisdictions in which police patrolling the streets possess the authority to perform immigration functions. According to ICE, by January 2010, 66 jurisdictions had signed onto the new MOA, an additional five agreements were still active while being renegotiated, and two new agreements were being negotiated, with the outcome uncertain. Six previously participating jurisdictions had declined to sign the new MOA or had withdrawn from negotiations, and a number of applications to enroll in the program remained pending. This report compares the pre-2009 MOAs with the new standardized agreement, referred to here as a template, and identifies the crucial factors that must be considered as ICE and the LEAs implement the new MOA. The template's emphasis on targeting noncitizens who have committed serious offenses may well reflect the administration's intent to return the 287(g) program to its original purpose of focusing on noncitizen criminals who pose a threat to national security and public safety, rather than on unauthorized immigrants generally, which increasingly had become a focus of the program. The changes in the terms of the agreement, primarily the requirements for greater federal supervision, may also help address the criticisms of the program highlighted above. At the same time, the new MOA introduces features into the program that could complicate its implementation. The new agreement does not set a minimum time limit on the training received by LEAs, and it eliminates the requirement that officers nominated for the program have two years of 1
experience, which could result in undertrained officers implementing the program. The new template also prohibits LEAs from releasing any data they collect during enforcement operations, potentially conflicting with state and local public records laws and undermining transparency. Whether the new policies embodied in the 2009 MOA will lead to changes in practice ultimately will depend heavily on how the agreements are implemented -- a development that can only be answered through an investigation of how ICE and the LEAs work together on the ground. To that end, the Migration Policy Institute (MPI) and the New York University School of Law will conduct an in-depth and onsite research study in six to nine jurisdictions across the country to evaluate the effectiveness of the MOAs in advancing federal immigration enforcement goals and the impact of the MOAs on local communities. The important questions about the new agreement that remain to be answered include: How can ICE ensure that resource allocation by the LEAs matches the stated priorities? How will ICE enforce the new requirement that LEAs obtain ICE approval before interrogating or processing immigration violators during operations in the community? How will ICE use its authority to approve or deny LEA actions that the MOA indicates can be reviewed on a case-by-case basis? How will ICE hold accountable those LEAs that do not abide by the terms of the new MOA? Will the changes to the program address the concerns of the communities where it has been controversial? This report provides an analytical framework for determining whether the 287(g) program generates greater benefits than costs and is worth maintaining. In addition, the questions posed can help federal and local officials shape the program to promote efficiency, accountability, and basic human rights, and assist community leaders in monitoring the program. 2
I. Introduction The participation of state and local officials in immigration policing has emerged in recent years as a significant and controversial trend in immigration enforcement policy. In 1996, Congress added section 287(g) to the Immigration and Nationality Act,1 permitting state, county, and local law enforcement agencies (LEAs) to enter into agreements with US Immigration and Customs Enforcement (ICE) to enable LEAs to perform certain immigration functions. These functions include screening inmates at local jails and state prisons for immigration status, arresting and detaining individuals for immigration violations, investigating immigration cases, and working with ICE on task forces to address immigration-related crimes. Though Congress first authorized collaboration in 1996 as part of the Illegal Immigration Reform and Immigrant Responsibility Act,2 ICE did not sign the first Memorandum of Agreement (MOA) with a state or local agency until 2002,3 when it entered into an agreement with the Florida Department of Law Enforcement. By the end of 2006, ICE had signed only eight agreements. But in 2007, state and local interest in the 287(g) program increased significantly: ICE signed 26 new agreements in 2007 alone, followed by another 28 in 2008. By May 2009, 66 active MOAs had been signed.4 (See Appendix 1 for a list of active and pending MOAs.) In July 2009, Homeland Security Secretary Janet Napolitano announced that ICE would sign 11 new agreements, signaling the Obama administration's intention to continue the program, notwithstanding demands by immigrantrights advocates and some law enforcement organizations that the program be shelved. Secretary Napolitano also announced that the program would be governed by a more standardized agreement, referred to here as a template, designed to provide stricter federal oversight and focus the program on the detention and removal of serious criminals. As of January 8, 2010, 66 "mutually signed" agreements were in place, five agreements in place before July 2009 were being renegotiated, and two agreements were being negotiated for the first time.5 With 71 agreements currently in operation, it is an opportune moment to reflect on the significance of the changes the Obama administration has made to the 287(g) program. 1 Immigration and Nationality Act (INA) §287(g), 8 U.S.C. § 1357(g) (2000). 2 Public Law 104-208. 3 City and county police departments in Salt Lake City, Utah explored the possibility of signing an agreement in 1998, citing the shortage of Immigration and Naturalization Service (INS) officials in the area to transport unauthorized immigrants identified for removal to federal facilities, but did not follow through with the plan because of public opposition. See Shawn Foster, "SLC Council Says No to Cross-Deputization," Salt Lake Tribune, September 2, 1998. 4 US Immigration and Customs Enforcement (ICE), The ICE 287(g) Program: A Law Enforcement Partnership (Washington, DC: ICE, 2009), www.ice.gov/doclib/pi/news/factsheets/Section287_g.pdf. (ICE regularly updates the Fact Sheet, so older versions may no longer be available). The number of MOAs actually peaked at 67; Barnstable County, Massachusetts, signed an MOA, but prior to May 2009, ICE terminated the agreement. See also Maria Sacchetti, "Framingham, Barnstable No Longer Enforcing U.S. Immigration Laws," Boston Globe, October 1, 2009, www.boston.com/news/local/breaking_news/2009/10/framingham_barn.html. 5 ICE, Delegation of Immigration Authority Section 287(g) Immigration and Nationality Act (Washington, DC: ICE, January 8, 2010), www.ice.gov/pi/news/factsheets/section287_g.htm. In August 2008, 80 jurisdictions had program applications pending. See ACLU Immigrants Rights Project, Catalogue of 287(g) Agreements/MOAs (New York: ACLU, 2008), on file with the Migration Policy Institute (MPI). Many of these jurisdictions dropped out, however, due to several factors, including: misunderstandings between ICE and the LEAs concerning the nature of the 3
By authorizing LEAs to perform immigration functions, the 287(g) program creates a role for state and local officials within an enforcement regime traditionally understood to be primarily, if not exclusively, federal. As explained in more detail below, ICE maintains a number of programs that provide for collaboration between federal and local officials in the enforcement of immigration law, some of which overlap geographically and substantively with the 287(g) program. The 287(g) program in particular, however, has attracted a great deal of comment and criticism because of its unique enlistment in immigration enforcement of officials engaged in day-to-day policing of communities. The program's value to ICE is that it expands the agency's enforcement capacities. At the same time, critics have charged that the involvement of state and local police leads to low-priority arrests and detentions,6 and the program arguably enables state and local actors to drive ICE's enforcement priorities and workload. The program's value to LEAs conceivably includes the power it gives them to respond directly to public safety and other social concerns produced by immigration, particularly illegal immigration. Critics -- particularly civil-rights agencies and law enforcement associations -- have suggested that, in addition to leading to arrests on minor charges, the 287(g) program may also increase the risk of racial profiling and could undermine community trust in local law enforcement, thus compromising public safety. The Obama administration's decision to continue the program makes it important to understand the objectives reflected in the new template, as well as to study the implementation of the agreement over time. Overall, the 2009 template reflects two broad priorities: setting enforcement priorities that target the identification, arrest, and removal of noncitizens who have committed serious crimes, and ensuring LEA adherence to these objectives under the program; and exercising close control of the enforcement operations of jurisdictions in which police patrolling the streets are authorized to perform immigration functions. Such changes represent significant new policy positions. Whether these changes will improve the 287(g) program's effectiveness and address criticisms of the program, including that it encourages racial profiling and enables local jurisdictions to depart from ICE enforcement priorities. These questions can only be answered through careful analysis of how the 287(g) program is being implemented on the ground under the new template. This report explores the new focus of the 287(g) program by comparing the standardized MOA template announced by Secretary Napolitano in July 2009 with past agreements.7 The report identifies: program, concern by both ICE and the LEAs about the costs of implementing the program, and the availability of other enforcement partnership programs, such as Secure Communities and the Criminal Alien Program. 6 US Government Accountability Office (GAO), Immigration Enforcement: Better Controls Needed over Program Authorizing State and Local Enforcement of Federal Immigration Laws (Washington, DC: GAO, 2009), 3, www.gao.gov/new.items/d09109.pdf. 7 Throughout this report, references are made to the text of the old MOAs. The MOAs, along with the 2009 template, are on file with MPI. 4
The salient terms of the new agreement The key changes made by the Obama administration The ways the new agreement addresses previous criticisms of the program Key questions concerning the continued implementation of the program, particularly its community impacts and whether it should expand to encompass a greater number of jurisdictions This analysis is the prelude to an in-depth field study being conducted by researchers at the Migration Policy Institute (MPI) and the New York University School of Law. The field research will involve six to nine jurisdictions across the country and will evaluate the effectiveness and impacts of the program in those locales. That research could provide a useful roadmap for others engaged in monitoring the program and assessing its ongoing viability. II. The History and Context of 287(g) As a general matter, a 287(g) agreement has always followed one of three basic models: The "jail model" cross-deputizes local incarceration officers to perform immigration functions with respect to individuals already detained on state criminal charges.8 The "task force" model authorizes officers to perform immigration-related enforcement in the community. The "hybrid" or "jail and task force" model allows jurisdictions to combine the jail and task force functions.9 Although the formal structure of the 287(g) agreement has remained substantially the same since the program's inception, the purposes for which these models have been used have evolved over time. 8 As the texts of the old MOAs reveal, though the jail agreements were initially confined to the post-conviction stages of criminal proceedings, as in Los Angeles County, newer pre-2009 MOAs did not similarly constrain jail officials. Most LEAs screened all inmates booked into their facilities, including during time of initial intake, and including persons who had been detained for traffic or other minor offenses. 9 The 66 MOAs in place in May 2009, prior to the DHS announcement of the new policy, consisted of 30 jail models, 24 task force models, and 12 hybrid models; see ICE, The ICE 287(g) Program. Before the July 2009 changes, state and city law enforcement agencies predominantly used the task force model, while nearly all of the counties that had signed agreements followed the jail or hybrid models. This difference may be explained by the fact that counties typically have primary responsibility for incarcerating convicted criminals at the local level, whereas city police departments arrest individuals, hold them for short periods of time, and then move them to county facilities for long-term detention. Following the July 2009 changes to the program, the mix of models did not change substantially: In January 2010, of the 71 agreements approved or active but in the process of renegotiation, 32 were jail models, 27 were task force models, and 12 were hybrid jail and task force models. See Appendix 1 and ICE, Delegation of Immigration Authority. 5
Figure 1. Types of 287(g) MOAs Signed by Year, 2002-2009
Year Jail Agreements
Task Force
Agreements
2002
--
1
2003
--
1
2004
--
--
2005
3
--
2006
3
--
2007
11
7
2008
11
13
2009
4
5
Total
32
27
Source: ICE, Delegation of Immigration Authority.
Hybrid Jail/Task Force Agreements -----6 6 -12
Total 1 1 0 3 3 24 30 9 71
A. The Evolution of Enforcement Priorities The program originally focused on identifying and detaining a narrow class of high-value noncitizens. The legislative history of 287(g), though limited, suggests that Congress intended the program to focus on the identification and apprehension of noncitizens then deemed "absconders," now called "fugitive aliens."10 In a 2007 Fact Sheet, ICE described the program as targeting noncitizens accused of "violent crimes, human smuggling, gang/organized crime activity, sexual offenses, narcotics smuggling, and money laundering."11 The 2007 ICE Fact Sheet further emphasized that the program was not intended to authorize LEAs to "perform random street operations" or to address issues such as "day laborer activities" or "excessive occupancy."12 Though authorized in 1996, it was not until after the terrorist attacks of September 11, 2001, that ICE signed the first 287(g) agreement. During the subsequent four years, only a handful of jurisdictions sought to participate in the program, and the agreements ICE signed in this period were mostly narrowly focused. Under the first agreement signed with Florida, for example, the state officers were trained as members of the Regional Domestic Security Task Force, which was created to address perceived shortcomings in the state's ability to combat terrorism.13 The MOA's wording addressed counterterrorism and domestic security needs. In 2003, the State of Alabama entered into an MOA in response to increases in the presentation of fraudulent documents by applicants for drivers' licenses, though the scope of the Alabama MOA itself was not limited to particular types of 10 Floor statement of Representative Tom Latham, (R-IA) on March 20, 1996, during consideration of H.R. 2202 to amend the Immigration and Nationality Act, Congressional Record 142, H2475-01, *H2477, (observing that the program was intended to "allow state and local law enforcement agencies to enter into voluntary agreements with the Justice Department to give them the authority to seek, apprehend and detain those illegal aliens who are subject to an order of deportation"), http://thomas.loc.gov/cgi-bin/query/C?r104:./temp/~r104XdMS3A. 11 ICE, Delegation of Immigration Authority. 12 "Excessive occupancy" refers to the number of occupants in dwellings. Local zoning ordinances targeted at crowded dwellings have been adopted in some localities to attempt to control the congregation of large numbers of imMigrant Workers in a dwelling. 13 Testimony of Mark F. Dubina, Special Agent Supervisor, Tampa Bay Regional Operations Center, Florida Department of Law Enforcement, and Regional Domestic Security Task Force Supervisor, "The 287(g) Program: Ensuring the Integrity of Americas Border Security System Through Federal-State Partnerships," before the House Committee on Homeland Security, Subcommittee on Management, Integration, and Oversight, 109th Cong., 1st sess., July 27, 2005, 15, 17. 6
violations.14 In 2005, the Los Angeles County Police Department entered into a narrowly targeted MOA that authorized jail officials to screen inmates after their convictions for serious crimes. In July 2005, with these three agreements in place, the House Homeland Security Committee's Subcommittee on Management, Integration, and Oversight held a hearing on the program. Paul M. Kilcoyne, Deputy Assistant Director of ICE's Office of Investigations, testified that the program would remain "focused on criminal organizations, those individuals who pose a threat to the border security." When asked if the program would target the "landscape architect [who] gets pulled over for an expired tag," Kilcoyne assured the subcommittee that "what we are trying to do is go after criminal aliens . . . not the landscaping type of individuals."15 Indeed, early arrest numbers confirm that 287(g) agreements did focus on a small number of high-priority targets.16 Kilcoyne emphasized the importance of maintaining "a very focused approach" to the program, in order to make the best use of limited fiscal and managerial resources.17 By September 2006, the program had expanded into four additional jails, bringing the total number of agreements to seven. By the end of 2006, however, the nature of the program began to change, as law enforcement officials in an increasing number of jurisdictions sought broader immigration enforcement authority. Some law enforcement officials expressed concern that unauthorized immigrants were more likely to commit crimes and impose fiscal burdens on public services, thus warranting broader participation by LEAs in the enforcement of immigration law.18 Some LEAs sought the authority to apprehend as 14 Testimony of Major Charles E. Andrews, Chief, Administrative Division, Alabama Department of Public Safety, before the House Committee on Homeland Security, Subcommittee on Management, Integration, and Oversight, 109th Cong., 1st sess., July 27, 2005, 20. 15 Testimony of Paul M. Kilcoyne, Deputy Assistant Director, Office of Investigations, US Immigration and Customs Enforcement, before the House Committee on Homeland Security, Subcommittee on Management, Integration, and Oversight, 109th Cong., 1st sess., July 27, 2005, 34. 16 Ibid., 28. Early arrest numbers also suggest that the first 287(g) programs focused on a small number of highpriority targets. Under the first MOA signed, for example, the State of Florida devoted 35 officers to high-value targets, particularly aliens who posed threats to national security. See Dubina testimony before the House Committee on Homeland Security, Subcommittee on Management, Integration, and Oversight, July 27, 2005, 15. In the first 36 months of operation, those officers made approximately 165 arrests -- fewer than five arrests a month, or slightly fewer than one arrest per seven deputized officers per month. See Tanya Weinberg, "1 Year, 35 Agents, 165 Arrests," The Sun Sentinel, Aug. 2, 2003, http://articles.sun-sentinel.com/2003-0802/news/0308020099_1_immigration-law-clear-law-enforcement-florida-immigrant-coalition. Under the second MOA, 21 members of the Alabama State Police were trained; see Andrews testimony before the House Committee on Homeland Security, Subcommittee on Management, Integration, and Oversight, July 27, 2005, 21. 17 Kilcoyne testimony before the House Committee on Homeland Security, Subcommittee on Management, Integration, and Oversight, 26. 18 See Aarti Shahani and Judith Greene, Local Democracy on ICE: Why State and Local Governments Have No Business in Federal Immigration Law Enforcement (Brooklyn, NY: Justice Strategies, 2009), 18-19, http://www.justicestrategies.org/sites/default/files/JS-Democracy-On-Ice.pdf (discussing statements of Sheriff Richard Jones of Butler County, Ohio, criticizing use of taxpayer dollars to support unauthorized immigrants and pursuit by Sheriff Jones of 287(g) authority to issue civil detainers in jails and perform civil arrests on the streets to assist in the deportation of unauthorized immigrants). See also testimony of Jim Pendergraph, Sheriff of Mecklenburg County, North Carolina, "Empowering Local Law Enforcement to Combat Illegal Immigration," before the House Committee on Government Reform, Subcommittee on Criminal Justice, Drug Policy and Human Resources, 109th Cong., 2nd sess., August 25, 2006. In 2006, Sheriff Pendergraph reported that the agreement had permitted his officers to identify a large number of civil immigration law violators -- in other words, unauthorized immigrants without serious criminal charges against them. 7
many unauthorized immigrants as possible, rather than focus on criminal aliens.19 The program thus evolved into a more general enforcement program through which some local officials arrested and detained noncitizens for drunken driving, traffic violations, and minor infractions such as carrying an open container of alcohol.20 This turn to 287(g) for broader enforcement purposes was arguably motivated, in part, by local political pressures to address illegal immigration, as well as local frustration at the failure of Congress to enact immigration reform legislation in 2006 and 2007. Seeing no near-term likelihood of a federal solution to the illegal immigration problem, some local leaders and constituencies sought direct participation in immigration enforcement and found a willing partner in the Bush administration.21 Between 2007 and July 2009, ICE signed approximately 54 MOAs, all of which followed a similar template that contemplated broader enforcement authority. The increasingly standardized MOAs included a general mandate establishing that the agreement authorized "the LEA to identify and process immigration violators," and, in some cases, "to conduct criminal investigations," as well. By 2008, the ICE Fact Sheet on the program, continually updated by the agency, no longer indicated that the program was limited to the pursuit of fugitive aliens.22 At least publicly, ICE had moved away from its original focus on the removal of serious criminals. B. Public Critiques and Analyses of 287(g) Immigrants' rights advocates and some law enforcement associations have criticized the 287(g) program, with many advocates calling for its termination.23 The chief criticisms include that the program creates incentives for police to engage in racial profiling and that collaboration between LEAs and federal immigration authorities diminishes trust in the police among immigrant communities, therefore undermining the primary mission of local law enforcement: ensuring public safety.24 19 See Jerry Seper, "Ohio Deputies Get Federal Training," Washington Times, January 9, 2008; William Hershey, "Illegal Alien Home from Jail, Lawyer Says Armando Mondragons Arrest Could Help Him," Dayton Daily News, August 16, 2004; Clint Bolick, Mission Accomplished: The Misplaced Priorities of the Maricopa County Sheriff's Office (Phoenix, AZ: Goldwater Institute, 2008), www.goldwaterinstitute.org/Common/Img/Mission%20Unaccomplished.pdf; Shahani and Greene, Local Democracy on ICE. 20 GAO, Immigration Enforcement: Better Controls Needed Over Program Authorizing State and Local Enforcement of Federal Immigration Laws. 21 Audrey Singer, Jill H. Wilson, and Brooke DeRenzis, Immigrants, Politics, and Local Response in Suburban Washington (Washington, DC: Brookings Institution, 2009), 18, www.brookings.edu/~/media/Files/rc/reports/2009/0225_immigration_singer/0225_immigration_singer.pdf. 22 ICE, The ICE 287(g) Program: A Law Enforcement Partnership "Frequently Asked Questions" (Washington, DC: ICE, 2009), www.ice.gov/partners/287g/Section287g_faq.htm. 23 See the letter organized by the National Immigration Law Center and signed by 521 national and local organizations to President Barack Obama calling on him to terminate 287(g), August 25, 2009, www.nilc.org/immlawpolicy/LocalLaw/287g-Letter-2009-08-25.pdf. 24 International Association of Chiefs of Police (IACP), Police Chiefs Guide to Immigration Issues (Alexandria, VA: IACP, 2007), www.theiacp.org/Portals/0/pdfs/Publications/PoliceChiefsGuidetoImmigration.pdf (emphasizing effects of cooperation with federal immigration authorities on community trust); Shahani and Greene, Local Democracy on ICE: Why State and Local Governments Have No Business in Federal Immigration Law Enforcement, 3-5; Deborah Weissman, Rebecca C. Headen, and Katherine Lewis Parker, The Policies and Politics of Local Immigration Enforcement Laws (Chapel Hill, NC: American Civil Liberties Union and University of North Carolina at Chapel Hill School of Law, 2009), 32-35, 8
In January 2009, the US Government Accountability Office (GAO) released a report evaluating 29 of the then-existing agreements. GAO concluded that the program lacked certain controls, such as documented program objectives, an articulation of how LEAs are to use their authority, clear and consistent mechanisms of supervision, and protocols identifying the types of data local officials must collect and report to ICE.25 The absence of such controls, in the GAO's estimation, made it difficult for ICE to determine whether the 287(g) program has advanced the agency's enforcement objectives or served its intended purpose.26 In addition, GAO observed that the lack of controls was related to potential misuses of the program. Some participating agencies, for example, have used their authority to pursue removal of noncitizens who have committed relatively minor offenses, such as speeding. At least one sheriff has expressed the erroneous belief that the program authorized LEAs to question individuals not suspected of criminal activity about their immigration status in their homes.27 In her comprehensive report assessing the state of immigration detention, Dora Schriro, former director of ICE's Office of Detention Policy and Planning, documented the use of 287(g) authority to arrest nonviolent offenders. According to the report, for fiscal year 2008, 57 percent of the noncitizens in detention as the result of the 287(g) program were noncriminal, and 72 percent of the initial bookings were noncriminal.28 For FY 2009, the percentages were 53 percent and 65 percent, respectively.29 The Obama administration's announcement of a new, more standardized template for all 287(g) agreements comes on the heels of these criticisms and reflects an intention to reform and potentially expand the program. The new MOA's emphasis on the targeting of the most dangerous noncitizens could well reflect the fact that the program has come full circle, back to its original purpose. Recent actions by the Obama administration also suggest its willingness to constrain law enforcement officials who misuse their 287(g) authority. In March 2009, for example, the Department of Justice announced that it had opened an investigation into the practices of Sheriff Joseph Arpaio of Maricopa County, Arizona -- perhaps the most high-profile 287(g) jurisdiction. Investigators are considering whether Arpaio and his deputies engaged in "patterns or practices of discriminatory police practices," such as "unconstitutional searches and seizures" and "national- origin discrimination," including failure to provide meaningful access to services for persons of limited English proficiency.30 In addition, in October 2009, ICE announced that it had terminated the task www.law.unc.edu/documents/clinicalprograms/287gpolicyreview.pdf (citing chilling effect on reporting of crimes caused by 287(g) program). 25 GAO, Immigration Enforcement: Better Controls Needed, 4-5. 26 Ibid. 27 Ibid., 11-12. 28 Dr. Dora Schriro, Immigration Detention Overview and Recommendations (Washington, DC: ICE, 2009), www.ice.gov/doclib/091005_ice_detention_report-final.pdf. 29 Ibid. For further discussion of misuse of 287(g) authority, see Shahani and Greene, Local Democracy on ICE, 3641. 30 Letter from Loretta King, Acting Assistant Attorney General, US Department of Justice, to Sheriff Joseph Arpaio, March 10, 2009, regarding investigation of the Maricopa County Sheriffs Office, available at http://ndlon.org/images/documents/usdojlettertoarpaio.pdf. See also Daniel Gonzalez, "Arpaio to Be Investigated over Alleged Violations," Arizona Republic, March 11, 2009, www.azcentral.com/arizonarepublic/news/articles/2009/03/11/20090311investigation0311.html. 9
force authority of officers in Maricopa County,31 though the detention component of the 287(g) agreement remained in place as of January 2010.32 Despite the Maricopa County example, however, the changes ICE has made thus far have been primarily related to the structure of the agreements. The extent to which the new policies embodied in the 2009 template will result in changes in practice, or alter the nature or course of the program, depends in large part on how the new agreements are implemented. Because ICE did not complete the renegotiation of all existing MOAs until October 2009 (and a handful of agreements remained pending at this writing), it remains to be seen whether the changes made in the formal terms of the template will translate into substantive changes that address or exacerbate some of the criticisms outlined above.33 C. 287(g) in the Larger Enforcement Context The 287(g) program represents one of 14 programs that fall under the umbrella of ICE Agreements of Cooperation in Communities to Enhance Safety and Security (ACCESS).34 The 287(g) program provides one of several means through which state and local officials can collaborate with ICE on immigration enforcement. The two programs that resemble 287(g) most closely, in their focus on the removal of criminal aliens, are the Secure Communities Program and the Criminal Alien Program (CAP). Both Secure Communities and CAP have been instituted in jurisdictions that have signed 287(g) agreements, as well as in jurisdictions that are not 287(g) participants. Evaluating the 287(g) program thus presents a complex task, because it requires determining whether and how it duplicates, complements, or complicates these other ICE initiatives. The Secure Communities program, launched in 2008, permits state and local officials in prisons and jails to check the fingerprints of detainees against Federal Bureau of Investigation (FBI) and Department of Homeland Security (DHS) databases, in order to screen for immigration status and prior immigration violations. Participation in Secure Communities is voluntary, and ICE has expanded the program to 116 jurisdictions in 16 states. ICE expects to make the program available to each of the nation's 1,200 state and federal prisons and 3,100 local jails by the end of 2013.35 CAP also focuses on identifying, processing, and removing criminal aliens incarcerated in federal, state, and local detention facilities. Pursuant to the program, ICE works with the detention facilities to identify noncitizens in custody who may be removable prior to the completion of a criminal sentence. The program thus prevents the release of certain removable noncitizens after they have 31 Randal C. Archibold, "Immigration Hard-Liner Has His Wings Clipped," New York Times, October 6, 2009, www.nytimes.com/2009/10/07/us/07arizona.html?_r=1&hpw. 32 ICE, Delegation of Immigration Authority. 33 In addition, the July 2009 template may be further modified on a case-by-case basis during negotiations between ICE and the LEAs, making definitive conclusions about the Obama administrations approach to the program premature. As of this writing, the only significant modification made in the newly negotiated MOAs is that, whereas the template gives LEAs the authority to issue arrest warrants, the newly negotiated agreements only give LEAs the authority to execute warrants. 34 Complete list of ICE ACCESS programs available here: www.ice.gov/osic/iceaccess.htm. 35 ICE, Strategy to Accelerate and Expand Secure Communities (Washington, DC: ICE, March 4, 2010), www.ice.gov/secure_communities/deployment. 10
served their criminal sentences and prior to the initiation of removal proceedings. ICE's Office of Detention and Removal Operations (DRO) administers CAP and assigns federal officers to detention facilities across the country. DRO officials screen inmates to identify those who are foreign born and initiate formal proceedings in immigration court.36 Given the broader context within which the 287(g) program exists, the ongoing MPI study will attempt answers to these broad questions: Does the 287(g) program advance federal immigration enforcement goals and local law enforcement objectives, and why do local jurisdictions elect to participate in the program? What are the social and community impacts and fiscal effects 287(g) agreements have had, and will the 2009 changes, once implemented, alter these impacts? Has the 2009 template enhanced the transparency, accountability, and effectiveness of immigration enforcement policy? Do the recent changes, as implemented, focus the program more narrowly on the arrest, detention, and removal of immigrants who have committed serious offenses or already been ordered removed? III. The Terms of the 2009 Template and Questions Related to its Implementation The statutory authority for the 287(g) program requires that the "specific powers and duties" of each officer or employee of a state or political subdivision authorized to perform immigration functions be "set forth in a written agreement."37 The 2009 template consists of a standard form agreement that, in 12 sections, articulates the purpose of the collaboration and defines the general terms of the agreement.38 The template specifies functions to be performed by the LEAs, supervision requirements, procedures for identifying participating officers, cost allocation, reporting requirements, and mechanisms for protecting civil rights and community interests. Although the 2009 MOA offers a standardized template, the appendices to the MOA appear to be drafted to allow the agreements to be tailored to the circumstances of particular jurisdictions. (Appendix A of the MOA lists points of contact, Appendix B outlines complaint procedures, Appendix C addresses communication with the media, and Appendix D provides a template for standard operating procedures, or SOPs, that can be adjusted depending on the interests of a particular jurisdiction).39 36 ICE, Criminal Alien Program Fact Sheet (Washington, DC: ICE, November 19, 2008), www.ice.gov/pi/news/factsheets/criminal_alien_program.htm. 37 Immigration and Nationality Act (INA) § 287(g), 8 U.S.C. § 1357(g) (2000). 38 Links to the MOAs for all active 287(g) programs can be found at ICE, Delegation of Immigration Authority, www.ice.gov/pi/news/factsheets/section287_g.htm. 39 The pre-2009 agreements varied somewhat in their details, but the bulk of the agreements, particularly after 2006, adopted nearly identical terms. In this report, we focus on the significant changes to pre-2009 practice made by the 2009 template. 11
The July 2009 template embodies potentially significant changes to past practice along six dimensions: The articulation of enforcement priorities The scope of authority given to state and local LEAs The nomination and training of state and local officers Federal supervision of state and local activity Data collection and evaluation of the program Mechanisms of accountability and opportunities for rescission of agreements The changes made in each of these domains raise questions that must be addressed in order to fully evaluate the 287(g) program and the 2009 template. A. Enforcement Priorities and Objectives Perhaps the most important feature of the 2009 template is its articulation of a clear set of enforcement priorities. Section I of the template contains a statement of purpose that makes clear that the 287(g) program is intended to identify and process for removal "criminal aliens who pose a threat to public safety or a danger to the community." More significantly, Appendix D, which outlines the SOPs, asserts that ICE retains "sole discretion" with respect to how to manage its limited resources. Appendix D also directs LEAs to manage their own resource allocations according to a three-tiered set of priorities that reflect ICE's priorities for all of its detention and removal operations. In this priority lexicon, Level 1 includes noncitizens convicted or arrested for major drug offenses and violent crimes. Level 2 pertains to noncitizens arrested or convicted for minor drug offenses and property crimes. Level 3 includes noncitizens convicted or arrested for "other offenses."40 This articulation of enforcement priorities represents a significant departure from past practice, though the specific offenses included in each category and other particularities have yet to be developed fully. As noted above, the earliest MOAs focused on combating terrorism and other serious offenses, while the bulk of 287(g) agreements signed between 2006 and 2008 began with a generalized articulation of purpose: to authorize LEAs to "perform certain functions of an immigration officer."41 Few of the pre-2009 MOAs articulated objectives for the program or provided guidance to LEAs on allocating resources according to enforcement priorities. As noted earlier, the agreements typically stated that the intent of the agreeing parties (ICE and the law enforcement agency) was either that the MOA would "enable [the law enforcement agency] to identify and process immigration violators" or "result in enhanced capacity to deal with immigration violators." 40 Although most of the noncitizens going through the 287(g) program and other ICE enforcement programs have been unauthorized immigrants, some are immigrants with legal status such as lawful permanent residents. 41 MOA of Butler County, Ohio Sheriffs Office (signed February 5, 2008); MOA of Etowah County, Alabama Sheriffs Office (signed July 8, 2008) ("to perform certain immigration enforcement functions"); MOA of City of Manassas, Virginia Police Department (signed March 5, 2008) (same); MOA of Jacksonville, Florida Sheriffs Office (signed July 8, 2008) (same). All of the pre-2009 MOAs are on file with MPI; those linked on ICEs Web page are the currently active versions, reflecting the 2009 changes. 12
Research Questions The 287(g) program has been criticized for undermining the ability of DHS and ICE to set a coherent immigration enforcement agenda, because it devolves control from ICE to its local partners with respect to where, when, and how to deploy enforcement resources. This devolution arguably has permitted local officials, including elected sheriffs, to set enforcement priorities to meet local concerns rather than to contribute to a broader national enforcement agenda. The absence of clearly articulated objectives has led to widely divergent applications of the newly acquired powers by LEAs. In addition, though ICE has not released comprehensive arrest statistics on the 287(g) program,42 data released by local jurisdictions and obtained by journalists and advocates suggest that significant resources have been expended to remove unauthorized immigrants who are guilty only of minor violations. In Frederick County, Maryland, for example, more than half of the first 300 suspected immigration law violators charged under its MOA were arrested for driving without a license.43 Likewise, data from Gaston, North Carolina reveal that, between April 2007 and June 30, 2008, 95 percent of state charges filed against 287(g) arrestees were for misdemeanors, and 60 percent of these were for traffic violations that were not DWIs.44 In Mecklenburg County, 2,321 unauthorized immigrants were placed in removal proceedings in 2007. Data show that less than 5 percent of the charges against these individuals were for the commission of felonies.45Although these data do not come from official ICE sources or represent the program in its entirety, they suggest that a substantial number of immigrants arrested pursuant to 287(g) have not been charged with serious crimes. The explicit articulation of priorities in the 2009 template seems to reflect an effort by ICE to centralize control and more narrowly target the 287(g) program. The template, however, does not specifically require LEAs to focus only on Level 1 or Level 2 offenders, nor does it specify how resources should be distributed, either in the main text or in the agreement's appendix. For example, it does not indicate the percentage of resources that should be devoted to targeting, arresting, detaining, and removing Level 1 criminal aliens. The template also does not prohibit LEAs from using their authority to pursue noncitizens arrested or convicted of minor offenses, including traffic violations. At the same time, LEAs will be required to report all arrests in terms of the priority level when officers enter arrest data into ENFORCE46 -- the central DHS database that records information on immigrants arrested, detained, and removed by ICE and the Border Patrol. Whether the template's articulation of enforcement priorities and its reporting requirements succeed in shaping LEAs' enforcement efforts presents one of the most important questions related to the implementation of 287(g). The crucial question is whether DHS can and will assert sufficient control over resource allocation to ensure that the 287(g) program serves the federal enforcement agenda. The questions to be pursued in order to make this determination include: 42 Schriro, Immigration Detention Overview and Recommendations, 13. 43 Frederick County Sheriffs Office Corrections Bureau, Annual Report 27 (2008) (on file with MPI). 44 Michael Barrett, "Officers Decide When to Arrest, But for Immigrant Community Decision Can Lead to Deportation," Gaston Gazette, July 7, 2008, www.gastongazette.com/news/cloninger-22388-charges-arrested.html; Shahani and Greene, Local Democracy on ICE. 45 Lindsay Haddix, Immigration and Crime in North Carolina: Beyond the Rhetoric (Chapel Hill, NC: University of North Carolina, Department of City and Regional Planning, 2008). 46 In-person communication between staff from ICEs Office of State and Local Coordination and MPI researchers, September 3, 2009. 13
Which crimes will be included in each of the priority levels, and will local jurisdictions be given any flexibility in defining the crimes that fall within each level? Will the listing of priorities act as a restraint on LEAs, and will it lead to the direction of law enforcement resources away from minor traffic violations and the like, and toward violent crimes and other national enforcement priorities? Will ICE and/or the LEAs develop numerical targets, de jure or de facto, with respect to each of the priority levels? What incentives would such targets create for the LEAs, and what unintended consequences might result? Through what means, other than monitoring arrests via data collection, will ICE ensure that LEAs abide by the renewed focus on noncitizens who have committed serious crimes? Will local ICE offices engage in some form of day-to-day supervision? Will ICE put pressure of some kind on LEAs to adhere to these enforcement priorities, either by threatening to or actually rescinding the agreements of jurisdictions that do not comply with the priority scheme? Will any such efforts to secure compliance with the enforcement priorities be made during the annual review process contemplated by the new template? Will ICE exercise its discretion not to remove noncitizens detained by LEAs who fall into Levels 2 or 3, or will ICE direct LEAs to release noncitizens they have detained but who do not pose serious security threats? B. Scope of Authority As noted earlier, 287(g) agreements always have followed one of three models that determine the scope of LEA authority: the jail model, the task force model, and the hybrid model. The pre-2009 MOAs specified the powers and authority of the deputized officers in the "Designation of Authorized Functions" section of the agreement. Though the pre-2009 MOAs differed in their details, they contained all or some of the following designated powers: The 2009 template substantially maintains this framework, with a few salient differences. The list of powers is nearly identical to the most extensive of the pre-2009 MOAs, with a few apparent exceptions. The 2009 template: Includes as part of the task force model the power and authority to execute search warrants. Gives LEAs the power and authority to issue arrest warrants for immigration violations.47 The actual agreements that have been negotiated since the release of the July 2009 template, however, authorize task force officials only to serve arrest warrants for immigration violations, thus narrowing somewhat the scope of authority (though the power to serve arrest warrants existed in many of the pre-2009 agreements).48 Specifies that ICE supervisors must grant approval to task force officers (TFOs) before they interrogate noncitizens about their right to be in the United States or process them solely for immigration violations -- a power the SOP template makes clear will be 47 This power existed pursuant to "Power and Authority to Arrest," Code of Federal Regulations, Title 8, Pt. 287.5(c)(2), 2005 ed., which recognizes the authority of listed officials to issue arrest warrants. 48 This power exists pursuant to "Service of Warrant of Arrests for Immigration Violations," Code of Federal Regulations, Title 8, Pt. 287.5(e)(3), 2005 ed. 14
authorized on a case-by-case basis only. The supervisor "will approve" the exercise of the power only for the purposes of advancing the priorities of "removing serious criminals, gang members, smugglers, and traffickers" and when the TFO has "reasonable suspicion" to believe the noncitizen was involved in criminal activity. In the same vein, the 2009 template specifies that jurisdictions operating under the jail model will "identify and remove" criminal aliens pursuant to the "tiered level of priorities" set out in Appendix D of the MOA. These latter additions fit within ICE's apparent goal of ensuring that the 287(g) program advances national enforcement priorities by leading to the targeted arrest, detention, and removal of noncitizens who pose public safety concerns, rather than of immigration violators generally. Research Questions The above-mentioned changes to the terms of the MOA raise an important question about whether the requirement that TFOs obtain ICE approval before interrogating or processing immigration violators will lead to more targeted use of LEA authority. This change in practice is potentially quite significant, because it opens up the possibility of substantial federal oversight and control of LEA enforcement. What will the approval process mean in practice? Will ICE use this authority to ensure that task force officers act in a targeted fashion on a day-to-day basis, or will such priority setting be devolved to local officials, with ICE approving broader operations plans? Which of the listed powers are most frequently invoked and particularly susceptible to abuse, e.g., could the power to interrogate any person believed to be an alien result in racial profiling, a problem that has been identified as a consequence of 287(g) authority in numerous jurisdictions?49 In addition, the scope of LEA power under the task force and jail models differs significantly. Under the jail model, LEA authority is far more circumscribed because the power applies to noncitizens already in a correctional facility, whereas the task force model permits LEAs who interact with the public on the streets and other settings to wield immigration authority. Are there noticeable differences in the Community Reception and/or effectiveness of the 287(g) program between these two models? In jurisdictions that possess only detention authority, does evidence exist that non-287(g) Police officers who patrol the streets rely on non-immigration-related pretexts, such as violations of traffic laws, to bring noncitizens into custody, in order to trigger the immigration powers under the jail model? Finally, a recent study by the Chief Justice Earl Warren Institute on Race, Ethnicity, and Diversity at the University of California Berkeley School of Law contends that police officers in Irving, Texas arrested Hispanics for minor offenses at much higher rates than whites and blacks in order to funnel 49 See Shahani and Green, Local Democracy on ICE, 40; Weissman, Headen, and Parker, The Policies and Politics of Local Immigration Enforcement Laws, 44-47. 15
Hispanics through CAP, which authorizes local detention facilities to hold arrestees until ICE screens them for their status.50 Does the 287(g) jail model create incentives that lead to similar racial profiling? Do notable differences exist between the operations of detention programs that permit screening only post-conviction and those models that permit screening of any detained alien? What powers does the 287(g) jail model give to LEAs that they would not possess under Secure Communities or CAP? How does that power enhance the federal government's enforcement capacities? C. Personnel and Training The terms governing the nomination of personnel remain substantially similar to past practice. The LEAs nominate candidates for ICE training, and ICE retains the authority to conduct independent background checks and approve the candidates for inclusion in the 287(g) program. The new template, like the old agreements, requires LEAs to participate in the program for a minimum of two years, presumably to ensure a return on the investment of time and resources in the training process. In addition, like its predecessors, the 2009 template requires that officers have experience related to the particular functions they will be performing. Under the 2009 template, task force officers' experience must include interviewing witnesses, interrogating subjects, providing constitutional rights warnings, obtaining statements, and executing search warrants. Officers under the jail model must have had experience supervising inmates and maintaining security in their facilities and must demonstrate an ability to "meet and deal with people of differing backgrounds and behavioral patterns." The 2009 template adjusts the nomination procedure somewhat. For example, most of the pre-2009 MOAs identified the maximum number of officers who could be trained to participate in the program.51 The 2009 template contains no such limitations, though many of the pre-2009 MOAs also contained a provision authorizing future expansion of the program to be based on oral agreement of the parties. The pre-2009 MOAs also typically required that candidates have at least two years of prior experience in law enforcement. The SOP of the new template requires that applicants for task force authority have a minimum of one year of experience in law enforcement, but no such experience requirement is part of the detention model SOP. Finally, the 2009 template specifically requires the LEAs to conduct criminal background checks of officer candidates and to provide ICE with any information gathered, if asked by the agency to do so. Prior agreements contained a more general provision establishing that the LEA "agrees to use due diligence to screen individuals nominated for the program." 50 See Trevor Gardner II and Aarti Kohli, The C.A.P. Effect: Racial Profiling in the ICE Criminal Alien Program, (Berkeley, CA: Chief Justice Earl Warren Institute on Race, Ethnicity, and Diversity, 2009), www.law.berkeley.edu/files/policybrief_irving_FINAL.pdf. 51 The pre-2009 MOAs authorized as few as two officers to be trained, as in Shenandoah County, Virginia, and as many as 200 in Maricopa County, Arizona. The Las Vegas Metropolitan Police Department and the Butler County Sheriffs Office MOAs specified ten officers to be trained. 16
The formal training requirements also remain similar in substance. Training under both the old MOAs and the new template includes discussion of the terms of the MOA, the scope of LEA immigration authority, the relevant immigration and civil-rights laws, ICE's use-of-force policy, the Department of Justice's Guidance Regarding the Use of Race by Federal Law Enforcement Agencies of June 2003, "cross-cultural" issues, and the obligations of LEAs under federal and other laws. Procedurally, however, the 2009 template departs somewhat from past practice. Whereas the pre2009 MOAs mandated a minimum of four weeks of training, the 2009 template does not appear to set such a floor. The 2009 template also specifies that LEAs will be trained according to the Immigration Authority Delegation Program, which will be tailored to the functions to be performed and will include examinations that candidates must pass with a minimum score of 70. The "Certification and Authorization" section of the pre-2009 MOAs referred to "required testing," and thus appeared to contemplate that candidates would be subject to exams during their training, but the 2009 template contains a more explicit mandate and an expectation of a certain level of proficiency. Research Questions Though the 2009 template, like the predecessor agreements, contemplates some screening of candidates for 287(g) authority, the degree of rigor applied during such screening likely depends on particular circumstances. With what frequency does ICE follow up on the information uncovered by the screening, or refuse candidates nominated by local agencies, and for what reasons? As a practical matter, does the elimination of the two-year experience requirement result in less qualified personnel becoming part of the program, as measured by downstream disciplinary consequences or as reflected in the nature of ICE supervision? The formal terms of LEA training give rise to additional concerns. Whereas ICE agents receive five months of training in the intricacies of immigration law, 287(g) officers under the pre-2009 regime received four weeks of ICE training and officers under the new regime may receive even less. Given the complexity of federal immigration law, and the fact that immigration enforcement is not the priMary Job of local law enforcement agents, is the extent of the training provided adequate to prepare 287(g) officers to perform their functions effectively?52 How are changes in federal immigration law communicated to local jurisdictions, for example? To help answer this question, it will be important to determine not only the content of the training programs provided in different jurisdictions, but also to assess the content of the exams the 2009 template contemplates giving, along with the pass rates. 52 Noting that immigration law is intricate, voluminous, and distinct, IACP has expressed concern that local law enforcement agents acting under 287(g) agreements could violate the unique standards and constitutional requirements surrounding immigration enforcement: "What constitutes ,,probable cause in immigration matters may not be easy to discern." IACP, Police Chiefs Guide to Immigration Issues, 13. Indeed, instances of 287(g)authorized officers illegally detaining and even deporting US citizens while purporting to be enforcing immigration laws have been reported. See Paloma Esquivel, "Suit Filed over Disabled U.S. Citizen's Deportation Ordeal," Los Angeles Times, February 28, 2008, http://articles.latimes.com/2008/feb/28/local/me-guzman28. 17
The 2009 template appears to have given the LEA training program a new and official name -- the Immigration Authority Delegation Program (IADP). Has the content of the training changed, and is it more or less rigorous than previous training programs? D. Federal Supervision The pre-2009 MOAs generally contained a section emphasizing that immigration enforcement by LEAs was to be "supervised and directed by ICE supervisory officers," but with limited delineation of what such supervision would entail. The MOAs also established that the default policies and procedures governing LEA activity would be those used by DHS and ICE, and that conflict between federal rules and standards and local policies should be reported to the supervising ICE authority. The 2009 template introduces two new requirements of major potential significance. First, within the task force model, the power and authority to interrogate any person reasonably believed to be an unauthorized immigrant about his or her right to remain in the United States and the power to process a noncitizen solely based on an immigration violation will be delegated on a case-by-case basis.53 To exercise such power, task force officers must obtain prior approval from an ICE supervisor, who will approve the exercise only in furtherance of the enforcement priorities set out in the MOA. Second, before conducting an enforcement operation pursuant to 287(g) authority, LEAs must have their operation plans approved by the ICE supervisor. Requiring advanced approval ensures ongoing federal oversight of the exercise of authority by the LEAs. Within the jail model, the template also introduces the new requirement that LEAs report "all encounters of an individual who claims US citizenship" to the Field Office Director, who must then notify DRO headquarters. The supervision section of the 2009 template's SOP also lists a series of more specific supervisory practices, tailored to either the task force or jail model. For both models, the supervision section reinforces that the 287(g) program is designed to identify and remove "criminal aliens . . . pursuant to the tiered level of priorities set forth in Appendix D." Finally, the 2009 SOP imposes certain responsibilities on ICE, including to request and review noncitizen files and provide the LEA with updated DHS policies with respect to the processing of "illegal aliens," to audit on a regular basis the entries and records made by LEA officers into the federal IDENT/ENFORCE database and to communicate errors to the LEA, and to provide the LEA with guidance for presenting cases referred for federal criminal prosecution. With respect to the jail models, ICE is assigned similar responsibilities. In addition, the LEAs must notify ICE of any detainers within 24 hours of their placement and report immediately all encounters with individuals claiming US citizenship. 53 The 2009 agreement also makes clear that participating LEAs have the power, among others, to arrest without warrant for felonies cognizable under any law regulating the admission, exclusion, expulsion, or removal of aliens if there is a reason to believe that the person arrested has committed such a felony and there is likelihood of the person escaping before a warrant can be obtained; and the power to arrest for any criminal offense against the United States if the offense is committed in the officers presence. 18
Research Questions Understanding the extent and nature of ICE supervision represents perhaps the most pressing component of an assessment of the 287(g) program, both from the perspective of assessing the program's utility to the federal government and of ensuring that LEAs do not abuse their authority. In discussing the 287(g) program, ICE and its partners in the past have emphasized the importance of close supervision over participating 287(g) partners.54 It will be important to document the frequency of communication between the LEAs and ICE, and to determine when and under what circumstances ICE officials exert control over LEAs by directing particular courses of action. For example: Does ICE ever refuse to approve arrest authority? If so, under what circumstances? The new requirement that ICE approve any LEA enforcement operation plan is potentially significant. How does ICE review these plans? Does ICE ever request that an LEA change its operation plan? In what ways? How often do LEAs place ICE detainers on noncitizens in their custody, and what criteria do LEAs use to identify such noncitizens? How often do citizenship notification claims arise? How does ICE respond upon receiving notification that someone in custody claims citizenship? Because of the clustering of 287(g) agreements in a small number of states, some regional ICE offices have a greater supervision responsibility than others. (See Appendix 1.) Does this uneven distribution lead to more limited oversight by ICE officials who oversee more rather than fewer LEAs, or are there no differences in degree and scope of supervision based on the number of LEAs under a given ICE official's umbrella? Does the degree of supervision change over time, as LEAs acquire more experience under the program? What does ICE learn from its audits and what does it do with the information acquired? E. Data Collection and Reporting Requirements The pre-2009 MOAs required LEAs to track and maintain accurate data and statistical information regarding the operation of their 287(g) programs and to provide such data to ICE upon request.55 The agreements did not, however, provide specific guidelines or requirements about the nature of the data that law enforcement agencies should collect. Likewise, ICE has released no information 54 Florida Task Force Supervisor Mark Dubina, for example, testified to a congressional subcommittee in 2005 that, pursuant to that MOA, "[i]n all cases, the ICE team leader to the [task force], the [state law enforcement] special agent supervisor, and the local ICE immigration supervisor must agree on a decision to arrest or detain a person, pursuant to 287(g) authority." See Dubina testimony before the House Committee on Homeland Security, Subcommittee on Management, Integration, and Oversight, 16. 55 The pre-2009 MOAs also contained some language requiring reporting of LEA activities to ICE, providing that participating personnel "shall give notice to the ICE supervisory officer as soon as practicable after, and in all cases within 24 hours of, any detainer issued." That said, only five of the 66 MOAs set forth specific circumstances in which 287(g) partners were required to notify ICE of their actions. 19
about the processes or protocols by which it has audited or reviewed its 287(g) partners and the data they have submitted. The 2009 template departs from the previous broad and open-ended data collection requirement by specifically requiring that LEAs provide the arrest and statistical data they enter into the ENFORCE database. Additionally, in the 2009 template, ICE reserves the right to request specific tracking data or other documents related to a particular arrest. Finally, the 2009 template prohibits LEAs from releasing data they collect without ICE approval, which may give rise to conflicts with state and local laws governing the disclosure of public records. Research Questions Given that the data reporting requirement has been made more specific, it is important to understand what ENFORCE requires and provides. ICE is currently upgrading this database to include not only demographic and immigration-related information, but also to record the priority level of the crimes detained immigrants have committed.56 Eventually, ENFORCE will be linked to the FBI database and include a full criminal history. ENFORCE is thus designed to capture the basic information ICE will need to determine whether LEAs are adhering to the priority levels ICE has set. How accurately will LEAs code data on the types of offenses committed? How will ICE use these data to supervise and hold LEAs accountable for seeking out the most serious criminals? If ICE discovers that an LEA is focusing primarily on Level 3 noncitizens -- the least serious offenders -- will it attempt to redirect the LEA's enforcement activities? It also will be important to identify whether any valuable data related to the operation of the 287(g) program would not appear in the ENFORCE database. To what extent does ICE exercise its discretion to request data beyond what is entered into ENFORCE, and what function do the data ICE collects play in ICE's oversight of the program? Under the old agreements, despite a broad reporting requirement, ICE did not define what data should be tracked or how that data should be reported.57 LEAs expressed confusion regarding reporting requirements.58 As a result, ICE might not have had the information necessary to monitor the program effectively.59 Given this experience, the change in the formal reporting requirements may improve ICE's monitoring of the program. Whether the data collection requirement assists in ensuring that the 287(g) agreements advance program goals depends in large part on the quality and scope of the data collected and on what ICE chooses to do with that information. 56 In-person communication between staff from ICE Office of State and Local Coordination and MPI researchers, September 3, 2009. 57 GAO, Immigration Enforcement: Better Controls Needed, 5. 58 Ibid., 17. 59 Ibid., 18. 20
F. Mechanisms of Accountability The 2009 template, like the pre-2009 MOAs, contains some basic mechanisms for ensuring accountability and transparency in the 287(g) program. The pre-2009 MOAs generally included a section on the duration of the MOA, which provided that the MOA would remain in effect until terminated by either party to the agreement. The 2009 template establishes that the agreements will remain in effect from three years of the date of signing, unless terminated by either party, and provides for a review process at the end of three years to determine whether the MOA should be expanded, modified, or permitted to lapse. The 2009 template also adds that upon a "good-faith" determination by ICE that the LEA is not fulfilling its duties, ICE shall notify the LEA that it has 90 days to demonstrate a continued need for the 287(g) program. Failure to demonstrate such a need could result in termination of the program. In addition to the option given to ICE to rescind the agreement, less direct mechanisms of accountability exist in the agreements. Like some of its predecessors, the 2009 template establishes that LEAs are expected to follow up on criminal charges for every noncitizen arrested pursuant to the program and provides that the state or local entity incurs the cost of incarceration that results from conviction of a state or local offense. This provision, in theory, could ensure that LEAs do not have financial incentives to arrest noncitizens they think can be deported before serving their sentences, unless the arrest actually advances local law enforcements priorities. At the same time, some minor offenses, such as traffic violations, might be sufficiently inexpensive to prosecute that the costs of enforcement would not present a significant deterrent. Most of the pre-2009 MOAs provided for the formation of a steering committee, convened by the ICE Field Office Director and the chief law enforcement officer of the jurisdiction, to meet "periodically" to review the program. The MOAs provided that steering committee members would be given case reviews, copies of filed complaints and media coverage, and statistical information to the extent practical. The 2009 template eliminates the specific reference to a steering committee but incorporates nearly identical language into a "communication" section providing for an annual (as opposed to "periodic") meeting by ICE and the LEA to review the program in consultation with each other. The annual meeting ensures that ICE and local officials meet face-to-face to discuss the status of the program, providing a forum for further internal oversight. The 2009 template provides that, when necessary, ICE and the LEA may limit participation by nonlaw enforcement personnel in these meetings, thereby introducing the possibility that people outside of law enforcement could be included on the committee. The pre-2009 MOAs did not appear to contemplate participation on the steering committee by non-law enforcement personnel. The reservation of the right to exclude such parties in the 2009 template suggests that some community involvement now might be possible, perhaps giving the public the opportunity to promote accountability through participation, though the template does not expressly provide for this possibility. In addition, like the pre-2009 agreements, the new template includes a complaint procedure, in Appendix B of the MOA, that requires LEAs to notify ICE of any complaints, as well as of the results of any internal inquiries related to the complaint. The new template also contemplates that ICE will refer the complaint to the appropriate ICE office for investigation and that any complaint will be resolved within 90 days of being received. 21
The budgeting requirements of the program also provide a potential mechanism of accountability. As with the pre-2009 agreements, the 2009 template establishes that LEAs are responsible for their personnel expenses associated with the program. But whereas previous agreements also specified that performance of designated functions was at the LEA's expense, the 2009 template refers only to personnel expenses, perhaps leaving implicit that the LEAs will bear the cost of program implementation. Under the 2009 template, as was previously the case, ICE is responsible for the salaries and benefits of all of its own personnel and will provide instructors, training manuals, and resources for the "purchase, installation, and maintenance of technology . . . necessary to support the investigative functions of the LEA" (though the LEA is responsible for office supplies and security equipment, such as handcuffs). Because the LEAs will bear a substantial financial burden, including the costs of convicting, detaining, and possibly transporting noncitizens (though presumably ICE will cover the cost of transporting those it has elected to take into custody), LEAs may well have a strong incentive to weigh their pursuit of 287(g) authority against competing local enforcement needs. Finally, the new template contains several civil rights-related provisions that could provide a check on LEA enforcement activity by protecting the rights and interests of individuals subject to LEA authority. Like its predecessor agreements, the template makes clear that LEAs must follow federal civil-rights statutes and regulations.60 The new agreement additionally makes an explicit commitment to providing arrestees with limited English language proficiency the opportunity to request an interpreter. The agreement directs LEAs to maintain a list of qualified interpreters and instruct personnel on how to obtain interpretation services -- requirements present in only a few of the pre2009 MOAs. Research Questions ICE maintains the discretion to terminate a 287(g) agreement in the 2009 template, but neither the template nor its predecessor agreements specify the circumstances that might lead to termination, and ICE has only rescinded 287(g) authority in two cases to date -- the task force authority of Sheriff Joseph Arpaio in Maricopa County, Arizona, and the MOA signed by Barnstable County, Massachusetts. Which criteria will ICE use to determine whether an agreement should be rescinded? Does the language of the 2009 template referencing an LEA's failure to perform its duties include instances of jurisdictions failing to identify, arrest, or detain aliens for removal, or does it also contemplate failures such as violations of civil-rights laws or deviation from the three priority levels for detention spelled out in Appendix D of the new agreement? During the annual meeting between ICE and the LEAs, what will be ICE's goal? Will ICE identify specific measures to evaluate the 287(g) programs, or must rigorous evaluation await consideration during the review process at the end of the three-year MOA period? 60 The pre-2009 MOAs stated generally that LEA personnel were bound by all federal civil-rights statutes and regulations. The 2009 MOA explicitly references Title VI of the Civil Rights Act of 1964, noting that the statute bans discrimination based on race, color, or national origin, including on the basis of limited English proficiency), in any program receiving federal financial assistance. 22
How might arrested noncitizens learn about complaint procedures, and how are complaints fielded? How does ICE respond to complaints? Are complaints effective in changing LEA behavior? Will representatives from outside the law enforcement community have any input into the evaluation of the 287(g) program? Will such representatives be included in steering committees? How will the balance of costs (with respect to factors such as training, supervision, technology, detention, and transportation) between ICE and the LEAs change with the new agreement, if at all? At what level would cost become prohibitive for particular jurisdictions? Will the potential shifting of costs result in LEAs narrowing their focus to serious criminals? If civil-rights violations are alleged, how are they pursued? Are arrested noncitizens informed of the availability of interpretation? Is language interpretation consistently provided? How effective is interpretation? IV. Conclusion The 287(g) program is in a transitional phase. In unveiling the standardized agreement to govern the 287(g) program in 2009, DHS signaled its intent to continue federal-state-local collaboration in immigration enforcement, as well as an apparent willingness to modify the program to better advance ICE's enforcement goals. Perhaps the most important change evident in the new policy is the clear and repeated articulation of a central objective -- of identifying and removing noncitizens according to the three priority levels articulated in the MOA, with a focus on Level 1 noncitizens who have committed violent and serious offenses. The formal changes ICE has made to the template may well bring greater direction and structure to the program. But political pressure from the LEAs, Congress, and the public may make it difficult for ICE to narrow the program's focus to serious criminal aliens entirely. Whether 287(g) serves ICE's objectives without interfering with the goals of local law enforcement or threatening civilrights protections ultimately depends on how the LEAs and ICE operationalize the formal terms of the MOAs -- a question that demands field study. As participating jurisdictions put the new template into place, MPI will assess the extent to which ICE monitors the exercise of 287(g) authority and whether and how the federal government adjusts the program in light of any abuses or inefficiencies that may come to light. By considering whether the mechanisms of supervision, training, and accountability contained in the 2009 template lead in practice to the effective and fair exercise of immigration authority by state and local police, our research aims to provide the information and insight necessary to assess the extent to which the program advances its goals of promoting security, improving immigration enforcement, and protecting civil rights. 23
Appendices
Appendix 1: Active MOAs and Those in Negotiation, January 2010
State
Law Enforcement Agency
Support Type
ALABAMA ARIZONA ARKANSAS CALIFORNIA COLORADO CONNECTICUT DELAWARE FLORIDA GEORGIA MARYLAND MINNESOTA MISSOURI NEVADA NEW HAMPSHIRE
Alabama Department of Public Safety Etowah County Sheriff's Office Arizona Department of Corrections Arizona Department of Public Safety City of Mesa Police Department City of Phoenix Police Department Florence Police Department Maricopa County Sheriff's Office Pima County Sheriff's Office Pinal County Sheriff's Office Yavapai County Sheriff's Office Benton County Sheriff's Office City of Springdale Police Department Rogers Police Department Washington County Sheriff's Office San Bernardino County Sheriff's Office Colorado Department of Public Safety El Paso County Sheriff's Office City of Danbury Police Department** Delaware Department of Corrections** Bay County Sheriff's Office Collier County Sheriff's Office Florida Department of Law Enforcement Jacksonville Sheriff's Office Cobb County Sheriff's Office Georgia Department of Public Safety Gwinnett County Sheriff's Office Hall County Sheriff's Office Whitfield County Sheriff's Office Frederick County Sheriff's Office Minnesota Department of Public Safety Missouri State Highway Patrol Las Vegas Metropolitan Police Department Hudson City Police Department
Task Force Jail Enforcement Jail Enforcement Task Force Task Force Task Force Task Force Jail Enforcement Jail & Task Force Jail & Task Force Jail & Task Force Jail & Task Force Task Force Task Force Jail & Task Force Jail Enforcement Task Force Jail Enforcement Task Force Jail Enforcement Task Force Jail & Task Force Task Force Jail Enforcement Jail Enforcement Task Force Jail Enforcement Jail & Task Force Jail Enforcement Jail & Task Force Task Force Task Force Jail Enforcement Task Force
Dates Signed 9/10/2003 7/8/2008 9/16/2005 4/15/2007 11/19/2009 3/10/2008 10/21/2009 2/7/2007 3/10/2008 3/10/2008 3/10/2008 9/26/2007 9/26/2007 9/25/2007 9/26/2007 11/19/2005 3/29/2007 5/17/2007 10/15/2009 10/15/2009 6/15/2008 8/6/2007 7/2/2002 7/8/2008 2/13/2007 7/27/2007 10/15/2009 2/29/2008 2/4/2008 2/6/2008 9/22/2008 6/25/2008 9/8/2008 5/5/2007
24
NEW JERSEY
Hudson County Department of Corrections Monmouth County Sheriff's Office **
Jail Enforcement Jail Enforcement
8/11/008 10/15/2009
NORTH CAROLINA
Alamance County Sheriff's Office Cabarrus County Sheriff's Office Durham Police Department Gaston County Sheriff's Office Guilford County Sheriff's Office*** Henderson County Sheriff's Office Mecklenburg County Sheriff's Office Wake County Sheriff's Office
Jail Enforcement Jail Enforcement Task Force Jail Enforcement Task Force Jail Enforcement Jail Enforcement Jail Enforcement
1/10/2007 8/2/2007 2/1/2008 2/22/2007 10/15/2009 6/25/2008 2/27/2006 6/25/2008
OHIO
Butler County Sheriff's Office
Jail & Task Force
2/5/2008
OKLAHOMA Tulsa County Sheriff's Office RHODE ISLAND Rhode Island State Police**
Jail & Task Force Task Force
8/6/2007 10/15/2009
SOUTH CAROLINA
Beaufort County Sheriff's Office Charleston County Sheriff's Office** York County Sheriff's Office
Task Force Jail Enforcement Jail Enforcement
6/25/2008 11/9/2009 10/16/2007
TENNESSEE
Davidson County Sheriff's Office Tennessee Highway Patrol / Department of Safety
Jail Enforcement Task Force
2/21/2007 6/25/2008
TEXAS
Carrollton Police Department Farmers Branch Police Department Harris County Sheriff's Office
Jail Enforcement Task Force Jail Enforcement
8/12/2008 7/8/2008 7/20/2008
UTAH VIRGINIA
Washington County Sheriff Office Weber County Sheriff's Office Herndon Police Department Loudoun County Sheriff's Office Manassas Park Police Department Manassas Police Department Prince William County Police Department Prince William County Sheriff's Office Prince William-Manassas Regional Jail Rockingham County Sheriff's Office Shenandoah County Sheriff's Office
Jail Enforcement Jail Enforcement Task Force Task Force Task Force Task Force Task Force Task Force Jail Enforcement Jail & Task Force Jail & Task Force
9/22/2008 9/22/2008 3/21/2007 6/25/2008 3/10/2008 3/5/2008 2/26/2008 2/26/2008 7/9/2007 4/25/2007 5/10/2007
Active MOAs Pending "Good-Faith" Negotiations, January 5, 2010
State
Law Enforcement Agency
Support Type
Dates Signed
CALIFORNIA
Los Angeles County Sheriff's Office Orange County Sheriff's Office Riverside County Sheriff's Office
Jail Enforcement Jail Enforcement Jail Enforcement
2/1/2005 11/2/2006 4/28/2006
MASSACHUSETTS Massachusetts Department of Corrections
Jail Enforcement
3/26/2007
New Mexico New Mexico Department of Corrections
Jail Enforcement
9/17/2007
25
New MOAs Pending "Good-Faith" Negotiations, January 5, 2010
State
Law Enforcement Agency
Support Type
Dates Signed
RHODE ISLAND Rhode Island Department of
Corrections**
Jail Enforcement
--
NEW JERSEY Morristown Police Department**
Task Force
--
Note: **These jurisdictions did not have any trained 287(g) officers as of January 2010. Source: ICE, Delegation of Immigration Authority
Appendix 2. Types of 287(g) MOAs by Jurisdiction Type, Pre-2009
LEA Type
Jailhouse
Task Force Agreements
Agreements
State
4
9
County
25
6
City
3
12
Total
32
27
Source: ICE, Delegation of Immigration Authority
Hybrid Jailhouse/ Task Force Agreements 0 12 0 12
Total 13 43 15 71
Appendix 3: Map of Active MOAs and those in Negotiation, January 2010
Notes: "JEO" = jail model; "TFO" = task force model; "JEO/TFO" = hybrid model. Source: Analysis of data from ICE, Delegation of Immigration Authority, by N. Prchal Svajlenka, February 2010. 26
About the Authors Cristina Rodrнguez is Professor of Law at the New York University (NYU) School of Law and a Nonresident Fellow at the Migration Policy Institute (MPI). Her fields of research and teaching include constitutional law and theory, immigration law and policy, administrative law, language rights and language policy, and citizenship theory. Before arriving at NYU, Professor Rodrнguez served as a law clerk to Justice Sandra Day O'Connor of the US Supreme Court and to Judge David S. Tatel of the US Court of Appeals for the DC Circuit. Professor Rodrнguez earned a bachelor's degree in History from Yale College in 1995, a master of letters in Modern History in 1998 from Oxford University, where she was a Rhodes Scholar, and a J.D. from Yale Law School in 2000. Muzaffar Chishti is Director of MPI's office at NYU School of Law. His work focuses on US immigration policy, the intersection of labor and immigration law, civil liberties, and state and local enforcement of immigration laws. Prior to joining MPI, Mr. Chishti was Director of the Immigration Project of the Union of Needletrades, Industrial & Textile Employees (UNITE). He serves on the Boards of directors of the National Immigration Law Center, the New York Immigration Coalition, and the Asian American Federation of New York. Mr. Chishti was educated at St. Stephen's College, Delhi; the University of Delhi; Cornell Law School; and the Columbia School of International Affairs. Randy Capps is a Senior Policy Analyst at MPI. His research interests include US immigration policy, community impacts of immigration law enforcement, labor market integration of US immigrants, integration of US immigrants' children, and social policies and programs to assist in integration. Prior to joining MPI, Dr. Capps was a Senior Research Associate at the Urban Institute, where he studied immigrant integration and social programs for needy US populations. He has a bachelor's degree in Political Science from Williams College (1987), a master's of Public Affairs from the University of Texas at Austin (1992), and a Ph.D. in Sociology from the University of Texas (1999). Laura St. John graduated from the NYU School of Law in 2009 and currently is clerking for the Honorable Eric Smith on the Alaska Superior Court. 27
Works Cited American Civil Liberties Union Immigrants' Rights Project. 2008. Catalogue of 287(g) Agreements/MOAs. New York: ACLU. Andrews, Major Charles E. 2005. Testimony of the Chief, Administrative Division, Alabama Department of Public Safety before the House Committee on Homeland Security Subcommittee on Management, Integration and Oversight. 109th Cong., 1st sess., July 27, 2005. Archibold, Randal C. 2009. Immigration Hard-Liner Has His Wings Clipped. New York Times, October 6, 2009. www.nytimes.com/2009/10/07/us/07arizona.html?_r=1&hpw. Barrett, Michael. 2008. Officers Decide When to Arrest, But for Immigrant Community Decision Can Lead to Deportation. Gaston Gazette, July 7, 2008. www.gastongazette.com/news/cloninger_22388_article.html/charges_arrested.html. Bolick, Clint. 2008. Mission Accomplished: The Misplaced Priorities of the Maricopa County Sheriff's Office. Phoenix, AZ: Goldwater Institute. www.goldwaterinstitute.org/Common/Img/Mission%20Unaccomplished.pdf. Dubina, Mark F. 2005. The 287(g) Program: Ensuring the Integrity of America's Border Security System Through Federal-State Partnerships. Testimony of the Special Agent Supervisor, Tampa Bay Regional Operations Center, Florida Department of Law Enforcement, and Regional Domestic Security Task Force Supervisor before the House Committee on Homeland Security Subcommittee on Management, Integration and Oversight. 109th Cong., 1st sess., July 27, 2005. Esquivel, Paloma. 2008. Suit Filed Over Disabled U.S. Citizen's Deportation Ordeal. Los Angeles Times, February 28, 2008. http://articles.latimes.com/2008/feb/28/local/me-guzman28. Foster, Shawn. 1998. SLC Council Says No to Cross-Deputization. Salt Lake Tribune, September 2, 1998. Frederick County Sheriff's Office Corrections Bureau. 2008. Annual Report 27. (On file with Migration Policy Institute.) Gardner, Trevor and Aarti Kohli. 2009. The C.A.P. Effect: Racial Profiling in the ICE Criminal Alien Program. Berkeley, CA: Chief Justice Earl Warren Institute on Race, Ethnicity, and Diversity. www.law.berkeley.edu/files/policybrief_irving_FINAL.pdf. Gonzalez, Daniel. 2009. Arpaio to Be Investigated over Alleged Violations. Arizona Republic, March 11, 2009. www.azcentral.com/arizonarepublic/news/articles/2009/03/11/20090311investigation0311.html. Haddix, Lindsay. 2008. Immigration and Crime in North Carolina: Beyond the Rhetoric. Chapel Hill, NC: University of North Carolina, Department of City and Regional Planning. 28
Hershey, William. 2008. Illegal Alien Home from Jail, Lawyer Says Armando Mondragon's Arrest Could Help Him. Dayton Journal News, August 16, 2008. www.journal-news.com/n/content/oh/story/news/local/2008/08/16/ddn081608immigrant.html. International Association of Chiefs of Police. 2007. Police Chiefs Guide to Immigration Issues. Alexandria: IACP. www.theiacp.org/Portals/0/pdfs/Publications/PoliceChiefsGuidetoImmigration.pdf. Latham, Tom. 1996. Statement of US Representative (R-IA) on the floor of the US House of Representatives during consideration of H.R. 2202 to amend the Immigration and Nationality Act. Congressional Record 142, H2475-01, *H2477. March 20, 1996. http://thomas.loc.gov/cgi-bin/query/C?r104:./temp/~r104XdMS3A. National Immigration Law Center. 2009. Letter from 521 national and local organizations to President Barack Obama calling on him to terminate 287(g). www.nilc.org/immlawpolicy/LocalLaw/287g-Letter-2009-08-25.pdf. Pendergraph, Jim. 2006. Empowering Local Law Enforcement to Combat Illegal Immigration. Testimony of the Sheriff of Mecklenburg County, North Carolina before the House Committee on Government Reform Subcommittee on Criminal Justice, Drug Policy and Human Resources. 109th Cong., 2nd sess., August 25, 2006. "Power and Authority to Arrest." Code of Federal Regulations Title 8, Pt. 287.5(c)(2), 2005. Sacchetti, Maria. 2009. Framingham, Barnstable No Longer Enforcing U.S. Immigration Laws. Boston Globe, October 1, 2009. www.boston.com/news/local/breaking_news/2009/10/framingham_barn.html. Schriro, Dora. 2009. Immigration Detention Overview and Recommendations. Washington, DC: US Immigration and Customs Enforcement. www.ice.gov/doclib/091005_ice_detention_report-final.pdf. Seper, Jerry. 2008. Ohio Deputies Get Federal Training. Washington Times, January 9, 2008. "Service of Warrant of Arrests for Immigration Violations." Code of Federal Regulations Title 8, Pt. 287.5(e)(3), 2005. Shahani, Aarti and Judith Greene. 2009. Local Democracy on ICE: Why State and Local Governments Have No Business in Federal Immigration Law Enforcement. Brooklyn, NY: Justice Strategies. http://www.justicestrategies.org/sites/default/files/JS-Democracy-On-Ice.pdf. US Immigration and Nationality Act. § 287(g), 8 § 1357(g) (2000). US Government Accountability Office. 2009. Immigration Enforcement: Better Controls Needed Over Program Authorizing State and Local Enforcement of Federal Immigration Laws. Washington, DC: GAO. www.gao.gov/new.items/d09109.pdf. 29
US Immigration and Customs Enforcement. 2007. Delegation of Immigration Authority, Section 287(g), Immigration and Nationality Act. Washington, DC: ICE. www.ice.gov/pi/news/factsheets/070906factsheet287gprogover.htm. ______. 2009. The ICE 287(g) Program: A Law Enforcement Partnership. Washington, DC: ICE. www.ice.gov/doclib/pi/news/factsheets/Section287_g.pdf. ______. 2010. Delegation of Immigration Authority Section 287(g) Immigration and Nationality Act. Washington, DC: ICE. www.ice.gov/pi/news/factsheets/section287_g.htm. ______. 2010. Strategy to Accelerate and Expand Secure Communities. www.ice.gov/secure_communities/deployment/. ______. Section 287(g), Immigration and Nationality Act, Frequently Asked Questions. www.ice.gov/partners/287g/Section287g_faq.htm. ______. Criminal Alien Program. www.ice.gov/pi/news/factsheets/criminal_alien_program.htm. Weinberg, Tanya. 2003. 1 Year, 35 Agents, 165 Arrests. Sun Sentinel, August 2, 2003. http://articles.sun-sentinel.com/2003-08-02/news/0308020099_1_immigration-law-clear-lawenforcement-florida-immigrant-coalition. Weissman, Deborah, Rebecca C. Headen, and Katherine Lewis Parker. 2009. The Policies and Politics of Local Immigration Enforcement Laws. Chapel Hill, NC: American Civil Liberties Union, and University of North Carolina at Chapel Hill School of Law. www.law.unc.edu/documents/clinicalprograms/287gpolicyreview.pdf. 30

File: a-program-in-flux-new-priorities-and-implementation-challenges.pdf
Author: Cristina Rodriguez
Published: Thu Mar 25 18:14:02 2010
Pages: 34
File size: 0.61 Mb


, pages, 0 Mb

Rome's cultural revolution, 21 pages, 0.24 Mb
Copyright © 2018 doc.uments.com