DHS OIG, OIG-24-31, ICE's Risk Classification Assessment Process Was Not Consistently Used to Prevent the Release of High-Risk Individuals (2024)

OIG-24-31 June 12, 2024 FINAL REPORT ICE’s Risk Classification Assessment Process Was Not Consistently Used to Prevent the Release of High-Risk Individuals OFFICE OF INSPECTOR GENERAL U.S. Department of Homeland Security Washington, DC 20528 | www.oig.dhs.gov June 12, 2024 The Honorable Patrick J. Lechleitner Deputy Director and Senior Official Performing the Duties of the Director U.S. Immigration and Customs Enforcement Joseph V. Cuffari, Ph.D. Inspector General JOSEPH V CUFFARI ICE’s Risk Classification Assessment Process Was Not Consistently Used to Prevent the Release of High-Risk Individuals Digitally signed by JOSEPH V CUFFARI Date: 2024.06.12 15:51:25 -04'00' MEMORANDUM FOR: FROM: SUBJECT: Attached for your action is our final report, ICE’s Risk Classification Assessment Process Was Not Consistently Used to Prevent the Release of High-Risk Individuals. We incorporated the formal comments provided by your office. The report contains two recommendations to improve the risk classification assessment process. Your office concurred with both recommendations.

Based on information provided in your response to the draft report, we consider recommendations 1 and 2 open and resolved. Once your office has fully implemented the recommendations, please submit a formal closeout letter to us within 30 days so that we may close the recommendations. The memorandum should be accompanied by evidence of completion of agreed-upon corrective actions and of the disposition of any monetary amounts. Please send your response or closure request to [email protected].

Consistent with our responsibility under the Inspector General Act, we will provide copies of our report to congressional committees with oversight and appropriation responsibility over the Department of Homeland Security. We will post the report on our website for public dissemination. Please contact me with any questions, or your staff may contact Kristen Bernard, Deputy Inspector General for Audits, at (202) 981-6000. Attachment OIG Project No. 23-034-AUD-ICE DHS OIG HIGHLIGHTS ICE’s Risk Classification Assessment Process Was Not Consistently Used to Prevent the Release of High-Risk Individuals June 12, 2024 Why We Did This Audit U.S. Immigrations and Customs Enforcement’s (ICE) Office of Enforcement and Removal Operations processed 273,220 noncitizens in fiscal year 2023.

To help process noncitizens, ICE uses a risk classification assessment function to analyze an individual’s criminal history, family ties, and other data to identify those who present a risk to public safety or who are a flight risk. The risk classification assessment helps ICE determine whether to detain or release a noncitizen in its custody. We conducted this audit to determine the extent to which ICE uses its risk classification assessment process to ensure individuals are appropriately detained or released. What We Recommend We made two recommendations to improve ICE’s oversight of the Risk Classification Assessment process.

For Further Information: Contact our Office of Public Affairs at (202) 981-6000, or email us at: [email protected]. What We Found ICE officers did not always perform risk classification assessments to ensure they appropriately detained or released noncitizens held in ICE custody. According to a 2012 email message from the Executive Associate Director, ICE officers are responsible for completing the risk classification assessment for detained noncitizens unless they are subject to mandatory detention or removal from the United States within 5 days. However, in FY 2023, ICE officers did not conduct a risk assessment for 79,977 (33 percent) of 244,376 detained noncitizens.

This noncompletion rate was higher in FY 2022, when ICE did not conduct risk classification assessment for 129,250 (43 percent) of 302,758 detained noncitizens. Further, ICE officials did not always provide detailed justifications when deviating from the risk classification assessment’s recommendation to detain noncitizens. We determined ICE released 11,754 (3 percent) of 339,478 noncitizens who received a risk assessment recommendation to detain. Our statistical sample of this population identified that for 190 (71 percent) of 266 records, ICE officers and supervisors did not provide sufficient justification when their final decision to release a noncitizen differed from the risk classification assessment’s recommendation to detain.

These conditions occurred because ICE does not have oversight to ensure consistent application of the risk classification assessment process. ICE previously assigned an office to oversee the process, but it was disbanded in January 2017, and ICE did not reassign the responsibility. Additionally, ICE has not developed a policy to prompt officers to complete risk classification assessments in accordance with the Executive Associate Director’s message. As a result, ICE cannot ensure its custody decisions are informed, consistent, and transparent, which could result in releasing noncitizens who pose a potential risk to public safety.

ICE Response ICE concurred with all recommendations. Appendix B contains ICE’s management response in its entirety. www.oig.dhs.gov OIG-24-31 OFFICE OF INSPECTOR GENERAL U.S. Department of Homeland Security Background The Immigration and Nationality Act (INA), as amended, grants U.S. Immigration and Customs Enforcement (ICE) administrative authority to detain noncitizens in certain circumstances, including pending removal from the United States. ICE’s Office of Enforcement and Removal Operations (ERO) manages and oversees immigration detention. ERO is responsible for enforcing immigration laws, including arresting, detaining, and releasing or removing noncitizens.

ERO manages more than 8,600 employees located at 25 field offices, called areas of responsibility, throughout the country. According to ICE, in fiscal year 2023, it processed a total of 273,220 noncitizens and received funding to support 25,000 detention beds, on average, each day. Risk Classification Assessment Process In FY 2012, ICE developed the risk classification assessment (RCA) module in its Enforce Alien Removal Module. The RCA is a computerized tool that gathers a noncitizen’s biographical information and uses factors such as special vulnerabilities,1 risk of harm to public safety,2 and risk of flight3 to assess the noncitizen and ultimately recommend a decision to detain or release them.

Noncitizens come into ICE custody either through referrals from U.S. Customs and Border Protection, or from apprehensions made by ICE agents and other law enforcement units. Once a noncitizen is in custody, ICE collects and reviews their background information to determine whether they are subject to mandatory detention.4 Noncitizens who are not subject to mandatory detention or removal within 5 days are subject to the RCA process, and the ICE officer submits the noncitizen’s information through the RCA module. The module assesses risks to public safety and risk of flight, then produces a risk rating of high, medium, or low for each person. Using these risk ratings, the RCA process then generates one of three standardized recommendations for each noncitizen: 1 Special vulnerabilities include noncitizens with a condition, physical or mental illness, or other factor (e.g., victim status) that may affect a custody determination.

2 ICE bases a noncitizen's threat to public safety on an assessment of the noncitizen's criminal history, driving-under- the-influence record, pattern of violence, outstanding active warrants, supervision history, and association to any security-threat group(s). 3 Risk of flight includes a noncitizen’s threat of absconding or failing to appear for future immigration hearings based on ties to the local community, history of absconding, and substance-abuse history. 4 Under INA § 236 (c)(1)(A) – (D), a noncitizen may be subject to mandatory detention if they have committed specified crimes, including criminal offenses involving moral turpitude, a controlled substance, human trafficking, firearms, espionage, counterfeiting, terrorism, or an aggravated felony. www.oig.dhs.gov 1 OIG-24-31 OFFICE OF INSPECTOR GENERAL U.S. Department of Homeland Security • detain in ICE custody; • detain, eligible for bond; or • release on community supervision. The ICE officer reviews the RCA recommendation, either agrees or disagrees with the recommended action, and submits the decision to their supervisory officer for review.

The supervisory officer then renders a final decision to detain or release the noncitizen after considering the RCA and officer-recommended actions. When the RCA recommends detaining the noncitizen, it also generates a custody-classification recommendation. The custody classification designates detained noncitizens as either low, medium, or high risk5 and assigns housing for detained noncitizens according to their classification. For example, detained noncitizens with a high risk rating are held in separate accommodations and are subject to greater oversight than detained noncitizens with lower risk ratings.

Appendix C details a flow chart of the RCA process. For noncitizens who will not be held in an ICE detention facility, ICE issues an order of release on recognizance. The order informs the noncitizen of the conditions for their release. If the noncitizen fails to comply with these conditions, ICE may revoke their release, resulting in arrest.

We conducted this audit to determine the extent to which ICE uses its RCA process to ensure individuals are appropriately detained or released. Results of Audit ICE Did Not Always Perform Risk Classification Assessments In 2012, the Executive Associate Director sent an email message to all ERO officers informing them that they need to complete the risk assessment “as early as possible” when processing noncitizens. The message instructed officers to conduct an RCA on all noncitizens that come into their custody except for those who are subject to mandatory detention, removal from the United States within 5 days, or detention on behalf of another agency. However, ICE did not always perform RCAs on detained noncitizens whose detention was not already mandatory.

Specifically, we identified that, in FY 2023, ICE officers did not perform an 5 The RCA generates a custody classification of high for noncitizens with a history of violent or assaultive charges, convictions, institutional misconduct, or gang affiliation. The RCA generates a custody classification of medium for noncitizens with no recent history of violent or assaultive charges or convictions, no institutional misconduct, and no gang affiliation. The RCA generates a custody classification of low for noncitizens with minor criminal histories and nonviolent felony charges and convictions. www.oig.dhs.gov 2 OIG-24-31 OFFICE OF INSPECTOR GENERAL U.S. Department of Homeland Security RCA for 79,977 (33 percent) of 244,376 detained noncitizens. This noncompletion rate was higher in FY 2022, when ICE did not conduct RCAs for 129,250 (43 percent) of 302,758 detained noncitizens.

Additionally, we determined RCA completion rates varied significantly by ERO area of responsibility. In FY 2023, 9 of the 25 areas of responsibility completed less than 75 percent of the required RCAs, whereas 10 completed more than 90 percent of the required RCAs. Figure 1 shows the FY 2023 RCA completion rate for each ERO area of responsibility. Figure 1.

FY 2023 RCA Completion Rates by ERO Area of Responsibility New York, NY Newark, NJ Miami, FL Boston, MA San Francisco, CA Detroit, MI Chicago, IL Baltimore, MD Salt Lake City, UT El Paso, TX St. Paul, MN Dallas, TX Philadelphia, PA San Antonio, TX Phoenix, AZ Harlingen, TX Buffalo, NY Washington, DC Atlanta, GA Los Angeles, CA San Diego, CA New Orleans, LA Houston, TX Seattle, WA Denver, Co 100.0% 100.0% 99.7% 99.5% 96.7% 96.7% 96.3% 94.4% 93.2% 90.4% 89.7% 87.0% 83.0% 80.8% 79.8% 75.3% 73.4% 64.1% 61.7% 59.7% 43.9% 31.9% 31.1% 29.4% 24.7% 0.0% 20.0% 40.0% 60.0% 80.0% 100.0% 120.0% Source: Department of Homeland Security Office of Inspector General analysis of ICE’s 2023 RCA compliance-rate report www.oig.dhs.gov 3 OIG-24-31 OFFICE OF INSPECTOR GENERAL U.S. Department of Homeland Security According to ICE officers in areas with a higher RCA completion rate, such as San Antonio, their understanding was all noncitizens in their custody must have an RCA. In contrast, officers in areas with lower RCA completion rates, such as Houston, stated each field office has the authority to “do things their way.” ICE Did Not Always Justify Deviations from RCA-Generated Decisions DHS Instruction 141-01-001 Rev 00.1 (September 2019) requires DHS employees to “adequately document the organization, functions, policies, decisions, procedures, and essential transactions of the agency and retain the records.” Further, according to the Risk Classification Assessment – Quick Reference Guide,6 one of the benefits of an RCA is that it “increases standardization but also increases transparency for detention” decisions. Written explanations of deviations from recommended actions support ICE’s goal of standardization and transparency. From FY 2022 through FY 2023, ICE released 11,754 (3 percent) of the 339,478 noncitizens that the RCA recommended be detained.

The RCA process assessed 328 of these 11,754 released noncitizens as being a high risk to public safety. Additionally, the RCA process assessed 11,608 of the 11,754 released noncitizens as being a high flight risk. Table 1 shows the total number of noncitizens ICE released, by risk factor, according to RCA-assigned risk levels. Table 1.

Risk to Public Safety and Risk of Flight for Noncitizens Released in FYs 2022 and 2023 with an RCA Recommendation to Detain Risk Assessment Risk to Public Safety Risk of Flight High Medium Low 328 813 10,613 Totals Source: DHS OIG analysis of FY 2022 and FY 2023 RCA data 11,754 11,608 69 77 11,754 Although ICE released noncitizens contrary to the RCA-recommended actions to detain, ICE officials did not always sufficiently document the rationale for these decisions. From our statistical sample of the 11,754 records for noncitizens who were released counter to the RCA- recommendation actions, we identified 190 (71 percent) of 266 records that did not contain sufficient information to determine why ICE officers released the noncitizen. Rather, we found 6 The Risk Classification Assessment – Quick Reference Guide (June 2022) provides ICE officers step-by-step instructions on how to use the RCA module. www.oig.dhs.gov 4 OIG-24-31 OFFICE OF INSPECTOR GENERAL U.S. Department of Homeland Security the officers and supervisors provided short, broad, or general responses in the RCA module’s comment field for decision justifications. These responses included entries such as: • “release” without additional details; • “OREC [order of release on recognizance]” without additional details; • “ROR [released on recognizance]” without additional details; • “Subject will be released on OREC” without additional details; or • a series of numbers without an explanation.

In contrast, 76 records (29 percent) did provide sufficient details to support the decision. For example, when releasing noncitizens and deviating from the RCA recommendation to detain, ICE officers wrote: • “Subject reported as indicated and has no criminal history. Recommend subject be released.” • “No significant criminal history. Subject does claim poor health.” • “Subject, while a recent border violator, has no apparent criminal history and did appear as directed to be processed for removal proceedings.” • “Subject is part of a family unit from the SWB [Southwest border].

Subject will be placed on ATD [alternative to detention] Smartlink service and release on ROR.” Inferring the sample results to the total population of 11,754 records, we estimate that between 7,860 and 8,931 RCA records did not contain sufficient information to determine why ICE officers released the noncitizen. Without sufficiently documenting justifications for deviating from RCA recommendations, ICE cannot ensure transparency in its detention decision-making process — especially when releasing noncitizens whom the RCA determines may pose a threat to public safety or might be a flight risk. ICE Does Not Have Centralized Management for the Risk Classification Assessment Process The uneven application of the RCA process across ICE’s areas of responsibility — and the lack of justification when deviating from RCA recommendations — are attributable to ICE having neither oversight of the RCA process nor a formal policy. ICE officials stated the Office of Detention Policy and Planning previously provided oversight of the RCA process.

This office had conducted monthly meetings with the ERO areas of responsibility to review RCA completion rates. However, ICE officials explained that this office was disbanded in 2017; since then, no other office has been assigned accountability for ensuring ERO areas of responsibility use the RCA module. We requested foundational documents (such as policies and procedures, standard operating procedures, and manuals) that the disbanded Office www.oig.dhs.gov 5 OIG-24-31 OFFICE OF INSPECTOR GENERAL U.S. Department of Homeland Security of Detention Policy and Planning may have developed with respect to the RCA process, but ICE could not provide them. Unable to locate policies and procedures that the Office of Detention Policy and Planning may have used while conducting its oversight, we asked ICE for existing policy with respect to the use of the RCA module.

ICE officials told us that, aside from the 2012 email from the Executive Associate Director and the Risk Classification Assessment – Quick Reference Guide, they were not aware of any policy that clearly establishes a requirement to use the RCA module, or any additional guidance on when officers should submit information into the RCA module or how they should document their decisions when deviating from actions the RCA recommends. Conclusion The RCA process was designed to help ICE officers make informed, consistent, and transparent custody decisions. However, ICE officials did not consistently use the RCA process or act according to its recommendations. Without a policy and appropriate oversight for the RCA process, ICE cannot ensure its officers will make informed, consistent, and transparent custody decisions that prevent the release of noncitizens who pose a potential risk to public safety.

Recommendations Recommendation 1: We recommend the Executive Associate Director of Enforcement and Removal Operations assign an office responsible for managing and overseeing the risk classification assessment process. Recommendation 2: We recommend the Executive Associate Director of Enforcement and Removal Operations establish and implement a formal policy and procedure for using the risk classification assessment process. The policy should include requirements for when officers should apply the risk classification assessment process and how officers should sufficiently document justifications to ensure the rationales for their decisions are clear, especially when deviating from the risk classification assessment’s recommended decision. Management Comments and OIG Analysis ICE provided management comments in response to a draft of this report.

We included the comments in their entirety in Appendix B. We also received technical comments and revised the report as appropriate. A summary of ICE’s response to each recommendation and our analysis follows. ICE Response to Recommendation 1: Concur.

ICE ERO

Headquarters Field Operations is responsible for ensuring field offices adhere to policy. This office and ICE ERO Law Enforcement www.oig.dhs.gov 6 OIG-24-31 OFFICE OF INSPECTOR GENERAL U.S. Department of Homeland Security Systems and Analysis will review existing guidance and practices, and determine areas needing additional resources, oversight, and updating. Once this review is complete, ICE ERO will implement more formal RCAs and reviews, as appropriate. Estimated Completion Date: November 29, 2024.

OIG Analysis of ICE’s Response: These actions are generally responsive to the recommendation, which we consider open and resolved. We will close the recommendation when ICE provides documentation that it completed reviews of existing guidance, implemented subsequent actions, and formally assigned an office responsibility for overseeing the RCA process. ICE Response to Recommendation 2: Concur.

ICE ERO

Headquarters Field Operations will create formal guidance and procedures to ensure officers consistency apply the RCA process and include strengthened justifications for decisions. Estimated Completion Date: November 29, 2024. OIG Analysis of ICE’s Response: These actions are responsive to the recommendation, which we consider open and resolved. We will close the recommendation when ICE provides documentation that it developed and implemented formal guidance and procedures that include requirements for when officers should apply the RCA process and how officers should sufficiently document justifications. www.oig.dhs.gov 7 OIG-24-31 OFFICE OF INSPECTOR GENERAL U.S. Department of Homeland Security Appendix A: Objective, Scope, and Methodology The Department of Homeland Security Office of Inspector General was established by the Homeland Security Act of 2002 (Pub.

L. No. 107−296) by amendment to the Inspector General Act of 1978. The objective of this audit was to determine the extent to which ICE uses RCA decisions to ensure individuals are appropriately detained or released. To achieve our objective, we reviewed ICE policies, procedures, and other documents related to the application of the RCA process. We also reviewed prior audits and reports, media articles, and congressional hearings pertaining to the audit objective.

We conducted site visits to processing centers within the El Paso, Texas, area. We also interviewed multiple ICE officials from the following offices to gain an understanding of the RCA process: • Office of Regulatory Affairs and Policy • Office of Immigration Program Evaluation • ERO o Custody Management Division o Enforcement Division o Executive Information Unit o Field Operations o Law Enforcement Systems and Analysis Division o Non-Detained Management Division To assess whether ICE uses the RCA process, we analyzed the FY 2022 to FY 2023 RCA compliance-rate reports provided by ICE’s Law Enforcement Systems and Analysis Division for each of the 25 ERO areas of responsibility. To assess the data reliability of the RCA compliance reports, we validated the date range and compared the RCA detain-or-release decision totals to the source data. To assess whether ICE officers follow RCA recommendations when making decisions to detain or release, we obtained 681,183 records from FY 2022 to FY 2023 RCA decision-history data from ICE’s Law Enforcement Systems and Analysis Division.

Of the 681,183 records, we identified 339,478 records that provided an RCA decision type “detain/release” and an RCA recommendation to “detain in the custody of this service.” We compared the initial RCA recommendation to the final RCA decision and identified 11,754 records where the RCA www.oig.dhs.gov 8 OIG-24-31 OFFICE OF INSPECTOR GENERAL U.S. Department of Homeland Security recommendation was to detain, but ICE released the noncitizen. To assess officers’ documentation when deviating from the recommended RCA decision, we reviewed the comment fields within the RCA module. Given the population of 11,754 RCA records, the statistically valid sample size is 266 with 90 percent confidence level, 5 percent sampling error, and 50 percent population proportion. To assess data reliability, we worked with DHS OIG’s Data Services Division to obtain, review, and verify ICE’s FY 2022 to FY 2023 RCA decision-history data, including validating the date range, removing circular references in the spreadsheet, and testing the data for duplicate entries.

Our assessments determined the RCA compliance-rate reports and RCA decision-history data were sufficiently complete to satisfy our audit objective. As such, we determined the data was sufficiently reliable to support the findings, recommendations, and conclusions in the report. We assessed ICE’s internal controls related to our audit objective. We limited our review to specific internal control components and underlying principles that were significant to ICE’s RCA process.

We assessed the design, implementation, and operating effectiveness of the controls in place to determine the extent to which ICE uses RCA decisions to ensure it appropriately detains or releases noncitizens. Our assessment of the significant internal controls disclosed that the overall internal control risk was high. We discussed specific weaknesses in the body of this report. However, because we limited our review to the RCA process, the report may not have disclosed all internal control deficiencies that existed at the time of this audit.

We conducted this audit from June 2023 through February 2024, pursuant to the Inspector General Act of 1978, 5 United States Code §§ 401–424, and according to generally accepted government auditing standards. Those standards require us to plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. DHS OIG’s Access to DHS Information During this audit, ICE provided timely responses to DHS OIG’s requests for information and did not delay or deny access to information we requested. www.oig.dhs.gov 9 OIG-24-31 OFFICE OF INSPECTOR GENERAL U.S. Department of Homeland Security Appendix B: ICE Comments on the Draft Report www.oig.dhs.gov 10 OIG-24-31 OFFICE OF INSPECTOR GENERAL U.S. Department of Homeland Security www.oig.dhs.gov 11 OIG-24-31 OFFICE OF INSPECTOR GENERAL U.S. Department of Homeland Security Appendix C: RCA Process Flow Chart Source: DHS OIG analysis of ICE’s RCA process www.oig.dhs.gov 12 OIG-24-31 ICE officer determines if the individual is subject to mandatory detention based on immigration laws.Individual is subject to mandatory detention.Individual is not subject to mandatory detention.ICE officer completes an RCA to assess the individuals risk factors such as risk to public safety and flight.Detained or detained eligible for bondReleased to Community SupervisionReleased on own recognizanceIndividual is sent to and held at detention center.ICE officer completes the custody classification portion of the RCA.RCA assigns recommended custody classifcation of low, medium or high.The RCA will assign a high, medium, or low risk for the risk factors.

Based on these factors, the tool will recommend detention or release .

OFFICE OF INSPECTOR GENERAL U.S.

Department of Homeland Security Appendix D: Report Distribution Department of Homeland Security Secretary Deputy Secretary Chief of Staff Deputy Chiefs of Staff General Counsel Executive Secretary Director, GAO/OIG Liaison Office Under Secretary, Office of Strategy, Policy, and Plans Assistant Secretary for Office of Public Affairs Assistant Secretary for Office of Legislative Affairs DHS ICE Liaison ICE Executive Associate Director for Enforcement and Removal Operations Office of Management and Budget Chief, Homeland Security Branch DHS OIG Budget Examiner Congress Congressional Oversight and Appropriations Committees www.oig.dhs.gov 13 OIG-24-31 To view this and any other DHS OIG reports, Please visit our website: www.oig.dhs.gov Additional Information For further information or questions, please contact the DHS OIG Office of Public Affairs via email: [email protected] DHS OIG Hotline To report fraud, waste, abuse, or criminal misconduct involving U.S. Department of Homeland Security programs, personnel, and funds, please visit: www.oig.dhs.gov/hotline If you cannot access our website, please contact the hotline by phone or mail: Call: 1-800-323-8603 U.S. Mail: Department of Homeland Security Office of Inspector General, Mail Stop 0305 Attention: Hotline 245 Murray Drive SW Washington, DC 20528-0305