The Role of ICE Detainers Under Bush and Obama

Over the years, Immigration and Customs Enforcement (ICE) detainers, often called "immigration holds," were thought to be a primary tool that ICE needed to apprehend the suspects it was seeking to deport. However, as data has emerged tracing ICE's actual use of detainers[1], whether they ever played such a necessary or primary role has been called into question. This report extends the body of evidence now available and documents how surprisingly low the number of individuals deported as a result of the use of detainers actually appears to have been.

Large-scale use of ICE detainers is a relatively recent phenomenon. Detainers were infrequently used during the first five and half years of former President George W. Bush's Administration. However, during the last two years under Bush, detainer usage increased rapidly and continued to grow when President Barack Obama assumed office.

Examining what detainers actually achieved and did not achieve during the Obama and Bush years is important because under the Trump Administration's recent flurry of immigration executive orders it appears that the use of detainers is likely to surge[2]. It also should be observed that very recent changes in the agency's Freedom of Information Act (FOIA) policies indicate that transparency about this and other ICE activities has been sharply reduced.

The data indicate that the growth in the use of detainers under Bush and Obama was surprisingly short lived. The preparation of ICE detainers peaked in August 2011 when 27,755 were recorded. And the number of these detainers that were followed by ICE taking the individual into custody peaked even earlier, during March 2010. In that month 16,713 of the detainers, according to ICE records, were followed by the individual being taken into custody. This peak in March of 2010 was barely a year after President Obama assumed office. Detainer usage fell off after this.

Just because a person was taken into ICE custody also didn't automatically mean the individual was ordered deported and removed from the country. Newly available government records suggest that a surprisingly small proportion of those taken into custody ultimately were deported. Figure 1 compares detainer-connected removals to the volume of detainers issued.[3] Note that actual removals are relatively few as compared with the number of ICE detainers.

Figure 1. Comparing ICE Detainers to Detainer-Connected Removals

These new findings are based on a detailed analysis by the Transactional Access Records Clearinghouse (TRAC) at Syracuse University of recently released ICE case-by-case records on removals following the issuance of a detainer. These data were obtained as a result of TRAC's multi-year Freedom of Information Act (FOIA) campaign that included hundreds of FOIA requests, appeals, and a successful lawsuit.

Another way to examine the evidence is to assess just what proportion of removals were of individuals who were previously targeted with a detainer. ICE case-by-case removal records show that during the first three months of FY 2016, only 5 percent of ICE removals from the interior of the U.S. (1,100 cases) were associated with a previously issued detainer. On an annual basis, this means the detainer- connected removals were recently running only 4,400 a year[4]. Here, both ICE detainers, as well as ICE-issued notices, are included in these counts.

When compared against the total of interior and border ICE removals, detainer-connected removals represented only 1.8 percent. And if deportations by Customs and Border Protection (CBP) are added to ICE removals, the percentage is even smaller.

It is true that these numbers are down sharply from the 10,702 detainer connected removals during the first three months of FY 2013 (October-December 2012). But even this higher number is surprisingly modest. Even back then, this represented only around a quarter (28%) of ICE removals from the interior, and less than 12 percent of all ICE removals. See Figures 2 and 3.

Figure 2 plots the proportion of ICE removals from the interior of the United States that followed the issuance of a detainer, while Figure 3 shows these same numbers as a percentage of total ICE removals including those from both the border and the interior.

Figure 2. ICE Deportations from Interior and Detainer Usage

A free user query tool accompanying this report allows the public access to underlying data behind these TRAC findings, along with additional useful information on ICE removals over an even longer period of time.

Figure 3. All ICE Deportations and Detainer Usage

Detainer-Connected Removals: from SCOMM through PEP

Detainer usage not only peaked early in the Obama years, but also was highest when only a relatively small proportion of jurisdictions were covered by Secure Communities (SCOMM). It was only later that most local and state law enforcement agencies had fingerprint records they submitted to the FBI automatically shared with ICE through the SCOMM program. See TRAC April 2014 report.

Even though detainer usage peaked early during the Obama years, removals resulting from detainers would not necessarily see a parallel decline. This is true for several reasons. For some individuals subject to a detainer, it could take many months or even years—given the Immigration Court's backlog—to work their way through civil proceedings and actually be ordered deported. For others, deportation might be delayed until individuals finished serving their prison sentence if they were convicted of committing a serious crime. Further, the expanded coverage of SCOMM gave ICE a greatly increased volume of fingerprint records to review, allowing it to refine and perhaps better target how and when detainers were used. This might lead to a higher proportion that resulted in an actual deportation.

A more detailed comparison - month-by-month—covering October 2012 through December 2015—highlights the role of detainers starting when SCOMM became fully functional in nearly all communities, and ending thirteen months after the announcement that the SCOMM program was being replaced by Homeland Security's Priority Enforcement Program (PEP)[5]. See Figures 4 and 5. Figure 4 plots the actual number of removals, while Figure 5 shows these trends as a proportion of ICE removals from the interior of the country. As before, both detainers and notices are included in these plots.

Figure 4. ICE Deportations With Prior Detainer by Month

Both graphs show a similar pattern of decline, with a modest temporary upsurge starting the month before the official announcement of SCOMM's replacement by PEP. After this upsurge, however, the plots show that once the new priorities PEP ushered in became fully operational, these declines resumed again.

Figure 5. Percent of ICE Interior Deportations With Prior Detainer by Month

Improvements in ICE Transparency Needed

Despite TRAC's new findings, additional unanswered questions remain. First, these findings focus on trends at the national level. There is no geographic breakdown because ICE did not release the needed geographic data. Further, with President Trump's new executive orders, an even wider range of information is now needed to monitor how these policies will be implemented.

However, the outlook for public access to the information required to monitor the detainer program is not promising. A recent and dramatic change in ICE FOIA policies occurred in the waning months of the Obama Administration. These changes in FOIA policies appear to be designed to drastically restrict the already limited flow of data the agency releases to the public.

Fields of information that ICE had routinely provided to TRAC in response to its regular monthly FOIA requests recently started getting left off the files TRAC received without explanation. More and more fields disappeared as new ICE responses arrived. For example, TRAC received 15 separate shipments from ICE one week in January and each of these shipments in response to individual FOIA requests were largely unusable because ICE had stopped providing so many key fields of information.

Omitted, for example, from the files were the year of birth of individuals, details on criminal convictions, whether the agency actually took custody of individuals it had asked state and local law enforcement officials to detain, or what actions ICE took to deport individuals once they were in ICE custody, along with many other data fields. All of this information the agency had previously been releasing to TRAC in response to its monthly FOIA requests, and are essential to the public's understanding of what the agency is actually doing to enforce immigration laws.

Thus far, administrative appeals within the agency have been unavailing. In response to one of TRAC's appeals, the ICE appellate office notified us that many fields the agency previously provided would no longer be provided. ICE claimed that these past releases were discretionary. However, the agency does not claim that this information is any way exempt from disclosure, only that it requires them to carry out "analyses" they are not required to undertake. TRAC, however, does not ask the agency to compile any statistics, it only requests the release of actual copies of their case-by-case records that obviously do exist and contain the information we are seeking. This is a very troubling development. Litigation appears to be needed to challenge these new unlawful practices.

The Inherent Challenge of Connecting ICE Detainers with Actual Removals

There is a further problem which casts doubt on how effective a greatly expanded use of detainers will actually be, under the announced reinstatement of Secure Communities by the Trump Administration. Even these reported figures on detainer-related removals implicitly assume that ICE actually has a reliable system for tracking what happens to cases after a detainer is prepared. Without reliable data, agency managers are without the information they need to effectively administer any detainer program.

By a reliable system, TRAC means one that both records subsequent events that occur after the detainer is prepared, and then allows these events to be linked so that the number of individuals ultimately deported as a result of the use of ICE detainers can be accurately monitored.

However, these assumed linkages are particularly problematic. For example, ICE contends that it only knows when a detainer is prepared and does not track whether it was actually issued. Further, when ICE later assumes custody of an individual we only know that custody occurs after the detainer was prepared. ICE contends it does not track whether ICE assumed custody from the law enforcement agency that issued the detainer. It could have arisen from entirely separate ICE initiatives. Likewise, we do not know that the use of the detainer actually led to the removal, or whether the removal was the result of other unconnected enforcement efforts. Again ICE contends its data systems do not allow it to track this.

Thus, these limitations in the available data may actually over-estimate the actual impact of detainers on removals since enforcement initiatives wholly apart from the preparation of a detainer could have been what led to some of these deportations. On the other hand, to the extent there are other types of data quality problems, these could interfere with ICE's ability to connect the issuance of a detainer to the later deportation even where the detainer was a key factor in securing the individual's removal. This would lead to under counting removals resulting from detainers. Either way, if ICE contentions on the limitations of its data systems are correct, ICE managers currently lack the information they need to effectively administer this program.

Neither the Department of Homeland Security nor Immigration and Customs Enforcement under Presidents Bush or Obama published annual statistics on its detainer program. While the department used to publish monthly statistics under its Secure Communities program, these statistics did not contain any information on the agency's actual use of detainers as part of SCOMM. Nor to our knowledge has the agency released any systematic studies of the reliability of the agency's records that are used by managers to monitor detainer usage and to track cases so that they can determine the effectiveness of its use of detainers in meeting the agency's goals.

Such studies are critically needed.


[1] See side bar for links to earlier reports TRAC has issued tracking detainer usage.

[2] See, for example, Executive Order 13768 of January 25, 2017 on "Enhancing Public Safety in the Interior of the United States." This executive order reinstates the Secure Communities program, and seeks to withdraw federal funds from so-called sanctuary jurisdictions. It also terminates Obama's Priority Enforcement Program (PEP) that placed a number of restrictions on the use of ICE detainers, and sets a target, subject to the availability of appropriations, of hiring 10,000 additional ICE officers.

[3] Figures on detainers and those on which ICE assumes custody are based upon the month the detainer was prepared, while figures on detainer-connected removals are based upon the month the removal occurred. Note that it is important to distinguish between detainer-connected removals shown here, and removals connected to automatic fingerprint matching. Many removals credited to the Secure Communities program did not involve the use of detainers, and detainer usage was never restricted to Secure Communities. A forthcoming TRAC report will examine the use of detainers within the Secure Communities program.

[4] As discussed later in this report, ICE has recently started withholding the data that would allow extending these results to the rest of 2016 and the beginning of FY 2017.

[5] See Department of Homeland Security's Secretary Jeh Johnson's November 20, 2014 directive on "Secure Communities." In replacing SCOMM with PEP, the justifications included citations to a long series of court decisions finding ICE detainers did not comply with probable cause requirements, placing state and local law enforcement agencies in violation of the Fourth Amendment to the U.S. Constitution. The undermining of community trust in local law enforcement as a result of the close cooperation called for in SCOMM was also cited. As the memorandum explained, these twin concerns had led to growing resistance to honoring ICE detainers by governors, mayors, and state and local law enforcement officials from around the country, as well as to the passage of more and more legislation restricting such cooperation.