Skip navigation

Shutting Down the Profiteers - Why and How the Department of Homeland Security Should Stop Using Private Prisons, ACLU, 2016

Download original document:
Brief thumbnail
This text is machine-read, and may contain errors. Check the original document to verify accuracy.
Tribute: The Legacy of Ruth Bader Ginsburg and WRP Staff

SHUTTING DOWN THE PROFITEERS:
Why and How the Department of Homeland Security
Should Stop Using Private Prisons

September 2016

1

Shutting Down The Profiteers:
Why and How the Department of Homeland Security
Should Stop Using Private Prisons
Authors: Carl Takei, Michael Tan, Joanne Lin

© 2016 ACLU Foundation

American Civil Liberties Union
125 Broad Street
New York, NY 10004
www.aclu.org

Cover photo: Wikimedia Commons, Jon ShakataGaNai Davis

EXECUTIVE SUMMARY
On August 18, 2016, the Deputy Attorney General directed the Bureau of Prisons
(BOP) to take steps to reduce, and ultimately end, the use of privately-operated
prisons. In undertaking these reforms, the Justice Department (DOJ) embarked on a
course to ensure that all federal prisoners are eventually incarcerated at BOP facilities,
not contracted facilities.
Shortly after the DOJ announcement, the Homeland Security (DHS) Secretary tasked
a subcommittee of the Homeland Security Advisory Council (HSAC) with evaluating
whether Immigration and Customs Enforcement (ICE) should follow DOJ’s lead and
eliminate use of private immigration detention. This task cannot be competently
undertaken without a comprehensive assessment of the immigration detention system
as well as of the laws, policies, and practices that shape the current detention
population. To be meaningful, the HSAC review must put all options on the table.
For the past three decades, the immigration detention population has increased
dramatically, reaching historic highs under the Obama administration. To manage the
growing detainee population, ICE has increasingly turned to for-profit prison
corporations and to county jails. By summer 2016, both the ICE detainee daily
population and the proportion of detainees in privately run facilities reached record
high levels, with the average daily detainee population exceeding 37,000. Of this
population, approximately 73 percent are detained in privately run facilities, about 15
percent in county jails, and only 12 percent are in federally-owned facilities. Thus, on
any given day some 24,500 detainees are locked up in private prisons, with most of
the remainder detained in county jails.
Irrespective of the type of detention facility, ICE has never been able to ensure safe
humane conditions that comport with a civil detention model. Detainee deaths,
suicides, sexual abuse, and denial of medical care have been documented at all types
of facilities. Private prison companies, which are incentivized to cut medical staffing
and deny care to maximize shareholder return, have a particularly grisly track record.
In February 2016, the ACLU, Detention Watch Network, and the National Immigrant
Justice Center examined each of the deaths in custody over a two-year period in which
ICE’s own death review concluded that non-compliance with ICE medical standards
contributed to the person’s death. Six of these eight deaths took place in private
prisons. For example, when Evalin-Ali Mandza had a heart attack at GEO Group,
Inc.’s Denver Contract Detention Facility in Aurora, Colorado, the company’s medical
staff did not even call 911 until nearly an hour after a code-blue emergency was called,
in part because a nurse prioritized filling out transfer paperwork over calling an
ambulance. This delay, along with staff not being familiar with the relevant medical
protocols and failing to administer appropriate cardiac medication, were identified as
potential contributing factors in Mr. Mandza’s death.
The immigration detention system is devoid of rigorous oversight and transparency.
The Corrections Corporation of America’s Eloy Detention Center in Arizona
exemplifies the impunity these companies enjoy. Since 2003, fourteen detainees have
died at Eloy—which makes it the deadliest immigration detention facility in the
nation. At least five of these deaths were suicides, yet Eloy still had not adopted a
suicide prevention plan when José de Jesús Deniz Sahagun killed himself in his cell in
May 2015. Eloy still holds nearly 1,500 immigrants today. The most horrific example
of sexual abuse in ICE custody also comes from a CCA facility. Between 2009 and

AMERICAN CIVIL LIBERTIES UNION

2010, a CCA guard responsible for transporting detainees from CCA’s Hutto detention
facility to the airport for their deportation flights sexually assaulted multiple women
while en route, typically at night. His modus operandi was to stop his van on the way
to the airport, order each woman outside of the van, and sexually assault them on the
side of the road. He was able to do this with impunity because, in violation of CCA’s
contract with ICE, the company permitted him to drive the women without a fellow
guard in the van. The only reason why his sexual abuse came to light was that one of
the women reported the abuse to an airport employee before boarding her deportation
flight.
Unlike federally-run facilities, contract facilities are not obligated to provide to DHS
or Congress critical information about detention operations and conditions.
Therefore, these contract facilities—including those run by for-profit prison
corporations—operate outside the purview of public oversight and accountability.
At the same time, for-profit prison companies have played a central role in the
expansion of the immigration detention system and the implementation of harsh
detention policies and practices. Prior to June 2014, ICE detained fewer than 100
children and mothers at one facility in Pennsylvania. However, starting in mid-to-late
2014, ICE swiftly began detaining unprecedented numbers of families in Karnes,
Texas and Dilley, Texas—together operating over 3,500 beds. These facilities were not
erected by ICE, but by the nation’s two largest private prison corporations. Those
companies were able to swiftly create this horrific new form of mass detention
because they stood at the ready—primed to create massive new jails as quickly as ICE
requested them.
Similarly, under the Obama administration ICE has greatly increased the detention of
asylum seekers, from some 10,000 asylum seekers detained in Fiscal Year (FY) 2009
to over 44,000 asylum seekers detained in FY 2014. These asylum seekers have fled
brutal violence and persecution, and the vast majority do not have a criminal record.
Most asylum seekers are locked up in privately run facilities or county jails, and many
are detained for over a year while they pursue their claims in immigration court.
These two disturbing detention trends—mass family detention and unprecedented
rates of detention of asylum seekers—have all developed under the Obama
administration. As the government has implemented punitive policies against the
most vulnerable immigrants, ICE has increasingly turned to for-profit, prison
corporations to expand its detention capacity.
As the HSAC Subcommittee undertakes its review, we urge the Subcommittee to
consider:
First and foremost, profiteering should have no place in any detention system. This
applies to profits reaped by private prison companies and to profits made by county
jails that contract with ICE. Profiteering should never be part of the equation when
deprivation of physical liberty is at stake, especially the liberty and well-being of
children, mothers, and asylum seekers.
Second, the private prison companies running many of ICE’s largest detention
facilities are the same prison companies that the BOP is in the process of severing ties
with. In both the BOP and ICE contexts, these prison companies operate with
essentially the same business model. The DOJ’s Inspector General found that the
companies’ methods of operation put prisoners’ rights and needs at risk. And in the
ICE context, this fundamentally penal model is wholly inconsistent with what is
2

AMERICAN CIVIL LIBERTIES UNION

supposed to be a civil detention model. As described by ICE, “ICE detention is solely
for the purpose of either awaiting the resolution of an individual's immigration case or
to carry out a removal order. ICE does not detain for punitive reasons.”
Third, some ICE officials have suggested that the only way to end ICE’s reliance on
private prisons would be to massively expand its reliance on county jails. That is the
wrong path. Any long-term strategy to assure safe, humane, and truly civil conditions
cannot rely on ICE’s past pattern of outsourcing. Instead, ICE must ensure the federal
government maintains direct operational control over nearly all ICE detention beds.
Moreover, ICE must adopt true civil immigration detention standards along the lines
of those proposed by the American Bar Association, and must require all federal
facilities to implement such standards.
Finally, DHS should look to the DOJ experience and specifically to the underlying
policies that have fueled record-level detention rates. The August 2016 DOJ
announcement to phase out BOP’s use of private prisons was made possible by a
series of decisions to shorten certain types of nonviolent drug sentences, which
together reduced the BOP population by approximately 25,000 prisoners from its
2013 peak. While DOJ has implemented common-sense criminal justice reform
policies, exemplified by its 2013 “Smart on Crime” initiative to ensure more
proportional sentences for low-level and nonviolent drug offenses, DHS has
undertaken no comparable course of detention reform. Rather, DHS and ICE have
pursued extremely aggressive policies which have produced unprecedented detention
levels by all metrics: total numbers of people detained, vulnerable immigrants
detained, duration of detention, and revenue for for-profit prison corporations.
In order to curtail and eventually end the use of private prisons, DHS will need to
implement policies aimed at reducing unnecessary detention.
This white paper explains how to carry out that reduction.
Specifically, ICE should generally stop using detention as a means to ensure
appearance for court proceedings. Instead, as the Vera Institute of Justice
recommended to ICE’s predecessor agency more than 15 years ago, detention should
be reserved chiefly for effectuating removal orders.
Fully committing to this shift could completely eliminate ICE’s need for reliance on
private prisons. In place of its current mass detention apparatus, the government
should rely on release on recognizance, bond, and a range of community-based
Alternatives to Detention (ATDs) with case management services. ATDs have long
been proven effective at ensuring appearance, at a far lower cost to taxpayers than
detention, and relying on them would be consistent with best practices from the
criminal justice system.
If the next president implements these recommendations, DHS will finally be able to
end profiteering by private prisons in the immigration detention system. This will
make the detention system more accountable, transparent, and better able to assure
safe, humane, and truly civil detention conditions.
The key elements of the transition from a detention-based model to a communitybased model of compliance include the following policy changes. We have included
estimates of the likely detention population reductions associated with each policy
change, as follows:
3

AMERICAN CIVIL LIBERTIES UNION

Policy change

Estimated impact1 on
the average daily
detention population

End family detention and detention of asylum seekers:
Close all three family detention centers, utilize immigration
court (Immigration and Nationality Act § 240) removal
proceedings rather than expedited removal for family units, and
release asylum-seeking families to sponsors in the community.

11,000 to 15,000
people

Clarify to all ICE field offices that the 2009 Directive—Parole of
Arriving Aliens Found to Have a Credible Fear of Persecution or
Torture—remains in full force and has not been rescinded by the
DHS 2014 Enforcement priorities memorandum.
End prolonged detention without bond hearings: DHS
and DOJ should construe the general immigration detention
statutes to require a bond hearing before an immigration judge
for all individuals detained more than six months, where the
government must justify continued detention.

At least 4,500
people

Interpret the mandatory custody statute to permit a
range of custodial options, and apply it only to
immigrants recently convicted of serious crimes who
do not have meritorious immigration cases: Recognize
that electronic monitoring, house arrest, and other coercive
liberty restrictions satisfy INA § 236(c)’s “custody” requirement.

5,000 to 10,000
people

Narrow application of mandatory custody to exclude those with
substantial challenges to removal.
Properly construe INA § 236(c) to apply only to individuals who
are taken into ICE custody at the time of their release from
criminal custody.
Stop imposing exorbitant, unaffordable bonds: DHS and
DOJ should require that immigration officials use the least
restrictive conditions necessary to ensure the individual’s
appearance for court proceedings, including alternatives to
detention, when determining such conditions under INA §
236(a) or 212(d)(5). Moreover, where ICE or an Immigration
Judge sets a cash bond, they must consider the individual’s
ability to pay and impose no bond amount greater than
necessary to ensure appearance. Finally, DHS should make a
more flexible range of secured and unsecured bonds available,
rather than require individuals to post the full cash amount to be
released.

At least 1,300
people

1 Note that because a single individual may fall into multiple categories, the total impact on the detention
population may be smaller than the sum of all individual categories.

4

AMERICAN CIVIL LIBERTIES UNION

TABLE OF CONTENTS
Executive Summary ......................................................................................................... 1	
Table of Contents ............................................................................................................. 5	
I.	 Introduction: Why ICE Must End its Reliance on Private Prisons .......................... 6	
1.	 The detention system ............................................................................................ 7	
2.	 Over-detention, privatization, and the need for less detention ........................... 9	
3.	 Human consequences of over-detention and privatization ............................... 12	
II.	 How to Reduce the Immigration Detention Population ....................................... 15	
III.	 ICE Must Replace Detention with a Range of Calibrated Alternatives ............... 17	
1.	 Adopt best practices from the criminal justice system by reserving coercive
measures and intensive supervision for those who pose the greatest risk of noncompliance ................................................................................................................. 18	
2.	 Utilize more intensive supervision as needed .................................................... 19	
3.	 Secure compliance of families through attorneys ..............................................20	
IV.	 Avoid Detaining People Who Pose Little Danger or Flight Risk, and Use
Supervision to Mitigate Risks for Others ...................................................................... 21	
1.	 End family detention and the mass detention of asylum seekers ...................... 21	
2.	 End prolonged detention without bond hearings .............................................. 23	
3.	 Interpret the mandatory custody statute to permit a range of custodial options,
and apply it only to immigrants recently convicted of serious crimes who do not
have meritorious immigration cases ......................................................................... 25	
A.	 Recognize that electronic monitoring, house arrest, and other coercive
liberty restrictions satisfy 236(c)’s “custody” requirement .................................. 26	
B.	 Narrowly apply mandatory custody to exclude those with substantial
challenges to removal ............................................................................................ 27	
C.	 Properly construe the mandatory custody statute to apply only to individuals
taken into ICE custody at the time of their release from criminal custody ..........28	
4.	 Stop imposing exorbitant, unaffordable bonds.................................................. 29	
V.	 Conclusion ..............................................................................................................30	

5

AMERICAN CIVIL LIBERTIES UNION

I.

INTRODUCTION: WHY ICE MUST
END ITS RELIANCE ON PRIVATE
PRISONS

This policy white paper is a roadmap for how ICE can phase out its reliance on private
prison contractors. For decades, it has been clear that handing immigrants in federal
custody over to private prison companies is a recipe for abuse, neglect, and
misconduct. But thanks to decades of increasing use of detention being implemented
through increased outsourcing, immigration enforcement officials have become
heavily dependent on this predatory industry.
Some ICE officials have suggested that the only way to end ICE’s reliance on private
prisons would be to massively expand its reliance on county jails, which are the other
major form of outsourced detention.2 That is the wrong path. For ICE, any long-term
strategy to assure safe, humane, and truly civil conditions cannot rely on the same
failed pattern of outsourcing this core federal responsibility. Instead, ICE must ensure
the federal government maintains direct operational control over nearly all ICE
detention beds. That in turn will require a significant reduction in ICE’s detention
population.
This white paper therefore describes how to achieve that population reduction.
Specifically, to facilitate its transition away from private prisons and avoid
unnecessary detention, ICE should generally stop using detention as a means to
ensure appearance for court proceedings. Instead, detention should be reserved
chiefly for effectuating removal orders.
Fully committing to this shift could completely eliminate ICE’s reliance on private
prisons. As of June 2016, more than 25,000 of the 37,000 people in ICE detention did
not have a final order of removal. Depending on the outcome of their immigration or
expedited removal proceedings, these individuals may end up either being deported
or granted permission to remain in the United States. This is an extraordinarily high
use of detention for these civil proceedings, and is out of step with best practices for
pretrial detention and supervision in the criminal justice system. In place of its
current mass detention apparatus, the government should utilize release on
recognizance, bond, and a range of community-based ATDs with case management
services. ATDs have long proven effective at ensuring appearance, at a far lower cost
to U.S. taxpayers than detention, and relying on them would be consistent with best
practices from the criminal justice system.
Just as a 25,000-person decline in the federal prison population enabled the Justice
Department to phase out its private prison contracts without needing to build new
federal prisons, the population reductions described in this white paper will provide
DHS the ability to safely and humanely phase out most or all of its private prison
contracts without a major investment in building new federal detention facilities or

See Brian Bennett, White House Considers Ending For-Profit Immigrant Detainee Centers, But Critics
Say It Could Add Billions To The Cost, L.A. TIMES (Sept. 6, 2016, 3:00 AM),
http://www.latimes.com/nation/la-na-immigration-detention-20160906-snap-story.html.
2

6

AMERICAN CIVIL LIBERTIES UNION

increased reliance on other forms of detention outsourcing, such as county jail
contracts.

1.

THE DETENTION SYSTEM

The practice of detaining massive numbers of immigrants in hundreds of jails and jaillike facilities across the country is relatively new. In 1980, fewer than 2,000 people
were held in immigration detention nationwide. Between 1980 and 1990, the system
more than tripled in size, to nearly 7,000 beds. And in the last two decades, it has
exploded. Between FY 1995 and FY 2016, the average daily immigration detention
population grew from 7,475 to 32,985.3

4

3 Doris Meissner, et al., Immigration Enforcement in the United States: The Rise of a Formidable
Machinery 126 (Jan. 2013), http://www.migrationpolicy.org/pubs/enforcementpillars.pdf; See also
Transactional Records Access Clearinghouse, New Data on 637 Detention Facilities Used by ICE in FY
2015, Syracuse University (Apr. 12, 2016), http://trac.syr.edu/immigration/reports/422/ (stating that a
total of 39,083 individuals were still in ICE custody at the end of FY 2015).
4 Chart provided courtesy of César Cuauhtémoc García Hernández, Assistant Professor, University of
Denver Sturm College of Law, and updated with FY 2015 and FY 2016 data by the ACLU. Prof. García
Hernández generated the chart using data from the following sources: DHS, Office of Insp. Gen., ICE’s
Release of Immigration Detainees (Aug. 2014) http://www.oig.dhs.gov/assets/Mgmt/2014/OIG_14116_Aug14.pdf (FY 2013 data); ICE, ICE Detainee Population Statistics By Fiscal Year,
http://www.ice.gov/doclib/foia/reports/ero-facts-and-statistics.pdf (FY 2001-12 data); Alison Siskin,
Congressional Res. Serv., Immigration-Related Detention: Current Legislative Issues (Jan. 12, 2012);
Alison Siskin, Congressional Res. Serv., Immigration-Related Detention: Current Legislative Issues
(April 28, 2004) (FY 1994-2000 data); Timothy J. Dunn, The Militarization of the U.S.-Mexico Border
1978-1992 (1996); ICE, Fiscal Year 2014 ICE Enforcement and Removal Operations Report Draft 9 (n.d.),
https://assets.documentcloud.org/documents/1375456/ice-draft-report.pdf. The chart was updated with
FY 2015 and FY 2016-to-date data using statistics from ICE, Weekly Departures and Detention Report,
at 9 (June 20, 2016) (“ICE WRD Report”), https://oversight.house.gov/wpcontent/uploads/2016/07/ICE-Weekly-Departures-and-Detention-Report1.pdf.

7

AMERICAN CIVIL LIBERTIES UNION

In recent months, the average daily detention population has grown to record levels,
hitting nearly 37,000 people in June 2016. This is fueled by a sharp increase in the
number of immigrants and asylum-seekers who are being detained despite not having
any history of criminal convictions. As of June 2016, the majority of all immigration
detainees—more than 20,000 people—have no criminal record.5
Immigration detention is intended to be civil and non-punitive.6 As ICE recently
stated to reporters, “ICE detention is solely for the purpose of either awaiting the
resolution of an individual's immigration case or to carry out a removal order. ICE
does not detain for punitive reasons.”7 Nevertheless, ICE detention facilities—both
those run by public officials and by private companies—overwhelmingly consist of
jails and jail-like facilities.8 This state of affairs has persisted for decades,
notwithstanding the Obama administration’s oft-cited 2009 pledge to transform ICE
detention into a “truly civil detention system.”9 In 2009, correctional expert Dr. Dora
Schriro, who served as Department of Homeland Security (DHS) Secretary Janet
Napolitano’s Special Advisor on ICE Detention and Removal, noted:
With only a few exceptions, the facilities that ICE uses to detain aliens were
built, and operate, as jails and prisons to confine pre-trial and sentenced
felons. ICE relies primarily on correctional incarceration standards designed
for pre-trial felons and on correctional principles of care, custody, and control.
These standards impose more restrictions and carry more costs than are
necessary to effectively manage the majority of the detained population.10
Dr. Schriro’s observation holds true today. Although ICE promulgated new detention
standards in 2011,11 these standards are still derived from prison and jail standards
and are less protective of the rights of individuals in detention than the American Bar

ICE WRD Report, supra note 4, at 6.
Dora Schriro, Immigration Detention Overview and Recommendations 2-3 (Oct. 6, 2009) (“Schriro
Report”), available at http://www.ice.gov/doclib/about/offices/odpp/pdf/ice-detention-rpt.pdf. See
also Zadvydas v. Davis, 533 U.S. 678, 690 (2001) (“The proceedings at issue here are civil, not criminal,
and we assume that they are non-punitive in purpose and effect.”); Wong Wing v. United States, 163 U.S.
228, 237 (1896) (permitting immigration commissioner to detain immigrants but not impose a sentence
of hard labor).
7 Carrie Johnson, Justice Department Will Phase Out Its Use Of Private Prisons, NAT’L PUBLIC RADIO
(Aug. 18, 2016, 1:15 pm ET), http://www.npr.org/sections/thetwo-way/2016/08/18/490498158/justicedepartment-will-phase-out-its-use-of-private-prisons.
8 Human Rights First, Jails and Jumpsuits: Transforming the U.S. Immigration Detention System—A
Two-Year Review 7-12 (Oct. 6, 2011), http://www.humanrightsfirst.org/wp-content/uploads/pdf/HRFJails-and-Jumpsuits-report.pdf.
9 See, e.g., Nina Bernstein, U.S. to Reform Policy on Detention for Immigrants, N.Y. TIMES, (Aug. 5,
2009),
http://www.nytimes.com/2009/08/06/us/politics/06detain.html?pagewanted=all; See also Human
Rights First, Jails and Jumpsuits, supra note 8 at 4-6.
10 Schriro Report, supra note 6, at 2-3.
11 See ICE, Performance-Based National Detention Standards 2011,
https://www.ice.gov/doclib/detention-standards/2011/pbnds2011.pdf. However, the 2011 standards
represent a major improvement over ICE’s previous standards, which were deficient even in comparison
to prison and jail standards. See also Written Statement of The American Civil Liberties Union: Hearing
on Holiday on ICE: The U.S. Department of Homeland Security’s New Immigration Detention
Standards, Before the H. Judiciary Subcommittee on Immigration Policy and Enforcement (Mar. 28,
2012), https://www.aclu.org/files/assets/aclu_detention_standards_hearing_statement_final_2.pdf.
5

6

8

AMERICAN CIVIL LIBERTIES UNION

Association’s model standards for civil immigration detention.12 In 2013, the U.S.
Commission on International Religious Freedom (USCIRF) concluded that “all
detainees, including asylum seekers, continue to be detained under inappropriately
penal conditions.”13 USCIRF found that even the facilities that ICE described as “civil
detention” facilities imposed restrictions on detainee movement, often do not permit
detainees to wear their own regular clothes, retain penal practices such as frequent
headcounts (sometimes as often as eight times daily), and maintain penal
architectural elements such as “perimeter fences, razor wire, barbed wire, or
concertina coils, and locked entry doors as well as an extensive use of video and sound
monitoring throughout the facilities.”14 In September 2015, the U.S. Commission on
Civil Rights reached the same conclusion, finding that “DHS and its component
agencies and contractees detain undocumented immigrants in a manner inconsistent
with civil detention and instead detain many undocumented immigrants like their
criminal counterparts in violation of a detained immigrant’s Fifth Amendment
Rights.”15

2.

OVER-DETENTION, PRIVATIZATION, AND THE
NEED FOR LESS DETENTION

The growth of immigration detention is closely tied to the growth of the private prison
industry. The reason is that as Immigration and Naturalization Service (INS), and
later Immigration and Customs Enforcement (ICE), rapidly expanded the detention of
immigrants in the 1990s and 2000s, the agency repeatedly chose to further outsource
detention rather than developing in-house detention capacity. (Indeed, the
Corrections Corporation of America’s inaugural contract was with INS in 1983, to
detain immigrants in Texas.16)
As a result, the overwhelming majority of people detained in ICE custody are now
held in non-federal facilities. While ICE does utilize a handful of federally-owned
detention facilities—known as Service Processing Centers—around the country, these
facilities hold fewer than 4,000 detainees, or about 12 percent of ICE’s detention
population.17 All other ICE detainees are held in a mixture of private prisons and local
jails—with a substantial majority in private prisons. According to ICE, 73 percent of
people detained by the agency are now held in facilities operated by private prison
12 See A.B.A., ABA Civil Immigration Detention Standards (2012),
http://www.americanbar.org/content/dam/aba/administrative/immigration/abaimmdetstds.authcheck
dam.pdf.
13 See U.S. Commission on International Religious Freedom, Assessing the U.S. Government’s Detention
of Asylum Seekers: Further Action Needed to Fully Implement Reforms 4 (Apr. 2013),
http://www.uscirf.gov/sites/default/files/resources/ERSdetention%20reforms%20report%20April%202013.pdf
14 Id. at 4-5.
15 U.S. Commission on Civil Rights, Statutory Enforcement Report: The State of Civil Rights at
Immigration Detention Facilities 97-106 (Sept. 2015),
http://www.usccr.gov/pubs/Statutory_Enforcement_Report2015.pdf.
16 Grassroots Leadership, The Dirty Thirty: Nothing to Celebrate About 30 Years of Corrections
Corporation of America 1 (June 2013),
http://grassrootsleadership.org/sites/default/files/uploads/GRL_Dirty_Thirty_formatted_for_web.pdf
; See also Lee Fang, How Private Prisons Game the Immigration System, THE NATION (Feb. 27, 2013),
https://www.thenation.com/article/how-private-prisons-game-immigration-system/
17 National Immigrant Justice Center, The Immigration Detention Transparency & Human Rights
Project: August 2015 Report 5 (Aug. 2015),
http://immigrantjustice.org/sites/immigrantjustice.org/files/images/NIJC%20Transparency%20and%2
0Human%20Rights%20Project%20August%202015%20Report%20FINAL3.pdf.

9

AMERICAN CIVIL LIBERTIES UNION

companies.18 This represents a significant increase over 2009, when 49 percent of ICE
detention beds were run by private prison companies.19

These trends have enriched private prison investors. In 2008, the two biggest private
prison companies—Corrections Corporation of America (CCA) and GEO Group
(GEO)—received a combined $307 million in revenue from ICE detention contracts.
By 2015, that number had more than doubled, to more than $765 million.
Most recently, the Obama administration’s 2014 abrupt policy shift toward mass
family detention—detaining thousands of young children and their mothers to “send a
message” against others who might seek to enter the United States20—has created an
enormous windfall for the private prison industry. Last year, just one family detention
facility in Dilley, Texas accounted for 14 percent of CCA’s revenue.21 Indeed, as CCA
noted in its annual report, the company’s increased federal revenues in 2015
“primarily resulted from” the Dilley family detention contract.22

Steven Nelson, Private Prison Companies, Punched in the Gut, Will Keep Most Federal Business, U.S.
NEWS & WORLD REPORT (Aug. 18, 2016), http://www.usnews.com/news/articles/2016-08-18/privateprison-companies-punched-in-the-gut-will-keep-most-federal-business.
19 Grassroots Leadership, Payoff: How Congress Ensures Private Prison Profit with an Immigration
Detention Quota 6 (Apr. 2015),
http://grassrootsleadership.org/sites/default/files/reports/quota_report_final_digital.pdf
20 Julia Preston, Detention Center Presented as Deterrent to Border Crossings, N.Y. TIMES (Dec. 15,
2014), http://www.nytimes.com/2014/12/16/us/homeland-security-chief-opens-largest-immigrationdetention-center-in-us.html.
21 Chico Harlan, Inside the Administration’s $1 billion Deal to Detain Central American Asylum Seekers,
WASH. POST (Aug. 14, 2016), https://www.washingtonpost.com/business/economy/inside-theadministrations-1-billion-deal-to-detain-central-american-asylum-seekers/2016/08/14/e47f1960-581911e6-9aee-8075993d73a2_story.html.
22 Corrections Corp. of America, Annual Report (Form 10-K) for the fiscal year ended December 31, 2015,
at 62 (Feb. 25, 2016).
18

10

AMERICAN CIVIL LIBERTIES UNION

Private prison companies have also engaged in extensive lobbying and other
influence-peddling that appears to be aimed at maintaining ICE’s heavy reliance on
their industry. A recent report by the nonprofit Grassroots Leadership examined CCA
and GEO’s lobbying disclosure documents, and found that both companies repeatedly
disclosed that they engaged in direct lobbying on “Issues related to comprehensive
immigration reform,” “Issues relating to housing of ICE prison inmates,” and DHS
appropriations for ICE detention facilities. The vast majority of both companies’
lobbying expenditures occurred in quarters when they were lobbying the DHS
Appropriations Subcommittee.23 The revolving door between ICE and the private
prison industry also raises serious concerns about agency capture. David Venturella
left a position as an assistant director at ICE (where he pushed to apprehend more
immigrants to boost deportation numbers) to become Executive Vice President for
Corporate Development at GEO.24 Julie Myers Wood, formerly the DHS Assistant
Secretary in charge of ICE, now serves on GEO’s board of directors.25
However, the harms of over-detention are not limited to private prisons. The other
form of outsourced detention—ICE contracts with state and local facilities—also
expanded significantly in the 1990s and 2000s. As INS and then ICE expanded these
county jail contracts, the agency failed to keep many of these contracts updated to the
agency’s latest detention standards. In particular, both a 2014 Government
Accountability Office (GAO) report and a 2015 National Immigrant Justice Center
(NIJC) report found that ICE continues to permit many county jails to operate under
23 Grassroots Leadership, Payoff: How Congress Ensures Private Prison Profit with an Immigration
Detention Quota, at 11-13 (Apr. 2015),
http://grassrootsleadership.org/sites/default/files/reports/quota_report_final_digital.pdf
24 Andrea Jones, The For-Profit Immigration Imprisonment Racket, ROLLING STONE (Feb. 22, 2013),
http://www.rollingstone.com/politics/news/the-for-profit-immigration-imprisonment-racket20130222; See also Center for Responsive Politics, David Venturella Employment Timeline (last visited
Sept. 12, 2016), https://www.opensecrets.org/revolving/rev_summary.php?id=78082.
25 GEO Group, Board of Directors, Julie M. Wood (last visited Sept. 12, 2016),
http://www.geogroup.com/Julie_M__Wood

11

AMERICAN CIVIL LIBERTIES UNION

the oldest version of its detention standards, which are more than fifteen years old
and predate not only ICE’s 2009 detention reform initiative, but the creation of ICE as
a separate agency.26 NIJC, Detention Watch Network, and the ACLU have also
identified serious, life-threatening deficiencies in ICE’s oversight of both private
prison contracts and county jail contracts.27 Moreover, because the purpose of jails is
to hold people in criminal custody, they operate in a manner inconsistent with civil
detention principles. Indeed, some jails actually intermingle ICE detainees in housing
units together with prisoners held on criminal charges.28
The longstanding failure of ICE to ensure safe, humane conditions of civil detention in
its outsourced detention system—both county jails and private prisons—underscores
the difficult, important oversight task that a federal agency faces whenever it hands
people in federal custody over to non-federal entities. And it suggests that ICE should
heed the Justice Department’s recent conclusion that private prisons “compare
poorly” to federally-run prisons.29 That poor performance carries over to county jails
as well. Thus, a long-term strategy to assure safe, humane, and truly civil conditions
cannot rely on ICE’s default strategy of outsourcing to private prisons or county jails.
Instead, ICE must ensure the federal government maintains direct operational control
over all but a handful of ICE detention beds.

3.

HUMAN CONSEQUENCES OF OVERDETENTION AND PRIVATIZATION

ICE’s ever-expanding reliance on private prisons has taken a terrible human toll.
Although all three major private prison companies have agreed to abide by ICE’s most
recent detention standards, there is no guarantee that the companies will actually
follow through on that promise. And their record of abuse, neglect, and misconduct
gives little reason to believe that they will.
Take, for example, CCA’s Eloy Detention Center in Arizona. Since 2003, fourteen
detainees have died at Eloy—which makes it the deadliest immigration detention
facility in the nation.30 In 2012, a routine annual inspection evaluated Eloy’s suicide

U.S. Government Accountability Office, Immigration Detention: Additional Actions Needed to
Strengthen Management and Oversight of Facility Costs and Standards, GAO-15-153, 28-35 (Oct.
2014), http://www.gao.gov/assets/670/666467.pdf; See also National Immigrant Justice Center, The
Immigration Detention Transparency & Human Rights Project: August 2015 Report 5-7 (Aug. 2015),
http://immigrantjustice.org/sites/immigrantjustice.org/files/images/NIJC%20Transparency%20and%2
0Human%20Rights%20Project%20August%202015%20Report%20FINAL3.pdf.
27 Detention Watch Network & National Immigrant Justice Center, Lives in Peril: How Ineffective
Inspections Make ICE Complicit in Immigration Detention Abuse (Oct. 2015),
http://immigrantjustice.org/sites/immigrantjustice.org/files/THR-Inspections-FOIA-Report-October2015-FINAL.pdf; See also American Civil Liberties Union, Detention Watch Network & National
Immigrant Justice Center, Fatal Neglect: How Ice Ignores Deaths in Detention (Feb. 2016),
https://www.aclu.org/sites/default/files/field_document/fatal_neglect_acludwnnijc.pdf.
28 U.S. Government Accountability Office, Immigration Detention: Additional Actions Needed to
Strengthen Management and Oversight of Facility Costs and Standards, GAO-15-153, 9 (Oct. 2014),
http://www.gao.gov/assets/670/666467.pdf; Schriro Report, supra note 6 at 21.
29 Memorandum from Sally Yates, Deputy Attorney General, U.S. Department of Justice, regarding
Reducing Our Use of Private Prisons (Aug. 18, 2016),
https://www.justice.gov/opa/file/886311/download
30 Megan Jula & Daniel Gonzalez, Eloy Detention Center: Why so many suicides?, ARIZONA REPUBLIC
(July 29, 2015), http://www.azcentral.com/story/news/arizona/investigations/2015/07/28/eloydetention-center-immigrant-suicides/30760545/; ICE, List of Deaths in ICE Custody, 10/1/2003 to
26

12

AMERICAN CIVIL LIBERTIES UNION

prevention policies. The inspectors found that Eloy’s suicide watch room—the place
where people at the most acute risk of suicide are supposed to be housed and whose
chief purpose is to deny them the means to kill themselves—contained “structures or
smaller objects that could be used in a suicide attempt.”31 The following year, Elsa
Guadalupe-Gonzalez hanged herself in one of Eloy’s general population units. Two
days later, Jorge Garcia-Mejia hanged himself in a different general population unit.
ICE conducted death reviews afterward, which found that “confusion as to who has
the authority to call for local emergency medical assistance” led to delays in CCA staff
calling 911 after each suicide. The reviews also found that CCA and ICE staff failed to
conduct an appropriate debriefing of medical and security staff after the two suicides,
and that Eloy lacked a formal suicide prevention plan.32 Just over two years later, on
May 20, 2015, José de Jesús Deniz Sahagun committed suicide in his cell just hours
after a doctor had removed him from suicide watch. ICE’s death review found that
Eloy still had not adopted a suicide prevention plan at the time of Mr. Deniz
Sahagun’s death.33
The most horrific example of sexual abuse in ICE custody also comes from a CCA
facility. Between 2009 and 2010, a CCA guard responsible for transporting detainees
from CCA’s Hutto detention facility to the airport for their deportation flights sexually
assaulted multiple women while en route, typically at night. His modus operandi was
to stop his van on the way to the airport, order each woman outside of the van, and
sexually assault them on the side of the road. He was able to do this with impunity
because, in violation of CCA’s contract with ICE, the company permitted him to drive
the women without a fellow guard in the van. The only reason why his sexual abuse
came to light was that one of the women reported the abuse to an airport employee
before boarding her deportation flight.34
A number of detainees have tried to protest their mistreatment and unnecessary
detention by engaging in hunger strikes, especially as the number of asylum seekers in
detention has risen in recent years.35 However, private prison companies are willing to
7/28/2016 (Sept. 8, 2016),
https://www.ice.gov/sites/default/files/documents/FOIA/2016/detaineeDeaths_07_28_2016.pdf
31 NIJC, In Focus: ICE Inspections at Eloy Federal Contract Facility, Arizona,
http://immigrantjustice.org/focus-ice-inspections-eloy-federal-contract-facility-arizona (last visited
Sept. 8, 2016). Inexplicably, ICE’s inspectors nevertheless gave Eloy a passing rating on suicide
prevention. Id.
32 Human Rights Watch, US: Deaths in Immigration Detention (July 7, 2016)
https://www.hrw.org/news/2016/07/07/us-deaths-immigration-detention; https://perma.cc/H7CDCESP; https://perma.cc/KB73-R7E5j.
33 Id.
34 ACLU, ACLU of Texas Today Files Federal Lawsuit on Behalf of Women Assaulted at T. Don Hutto
Detention Center (Oct. 19, 2011), https://www.aclu.org/news/documents-obtained-aclu-show-sexualabuse-immigration-detainees-widespread-national-problem; ACLU of Texas, Their Stories In Their Own
Words (Oct. 19, 2011), https://www.aclutx.org/en/press-releases/hutto-sexual-abuse-case-background;
Renee Feltz, Guard’s arrest highlights sexual assault of immigrant detainees, PBS (Aug. 27, 2010),
http://www.pbs.org/wnet/need-to-know/security/lax-regulations-lead-to-sexual-assault-of-immigrantdetainees/3149/.
35 See, e.g., Julián Aguilar, For Detained Immigrants on Hunger Strikes, Results Aren’t Guaranteed,
TEX. TRIB. (Nov. 28, 2015), https://www.texastribune.org/2015/11/28/detained-immigrants-hungerstrikes-could-be-gamble/; Associated Press, Judge Approves Force-Feeding of Detainee on Hunger Strike
(Dec. 8, 2015), https://www.yahoo.com/news/judge-approves-force-feeding-detainee-hunger-strike234845109.html?ref=gs Jennifer Chowdhury, South Asian Detainees Seeking Asylum, NBC NEWS (Oct.
21, 2015, 3:29 PM), http://www.nbcnews.com/news/asian-america/south-asian-detainees-seekingasylum-hold-hunger-strike-n448681; Elise Foley, Detained Immigrant Women Claim Retaliation for
Hunger Strike, HUFFINGTON POST POL. (Nov. 3, 2015, 5:41 PM),
http://www.huffingtonpost.com/entry/immigrant-detention-hunger-

13

AMERICAN CIVIL LIBERTIES UNION

go to great lengths to squelch such protests. When detainees went on hunger strike at
GEO’s Northwest Detention Center in Tacoma, Washington to protest conditions of
confinement and lack of access to bond hearings, GEO staff invited them to meet with
an assistant warden to discuss their grievances—and then escorted them directly into
solitary confinement cells without the promised meeting. They were released from
solitary only after the ACLU of Washington and Columbia Legal Services filed a
lawsuit alleging that these actions violated the First Amendment.36 At CCA’s Hutto
detention facility in Texas, women who had been detained after seeking asylum began
refusing to eat last November in protest of their continued detention; in news reports,
ICE denied the existence of a hunger strike even as the women described CCA
retaliation.37 It was later revealed that CCA had adopted a written policy explicitly
instructing its staff how to retaliate against detainees who are refusing to eat, which
the policy described as “protesting in a passive aggressive manner.” The policy
authorizes CCA staff to put the facility on lock-down in response to a food strike,
rescind commissary-purchasing privileges for the entire facility, and rescind TV,
radio, visitation, and telephone access for those participating in the protest.38
Once detained, people are entirely dependent on ICE and its contractors for medical
care, because of barriers to getting care from an outside doctor while detained. But
private prison companies, which are incentivized to cut medical staffing and deny care
to maximize shareholder return, have a grisly track record in this area. In February
2016, the ACLU, Detention Watch Network, and the National Immigrant Justice
Center examined each of the deaths in custody over a two-year period in which ICE’s
own death review concluded that non-compliance with ICE medical standards
contributed to the person’s death. Six of these eight deaths took place in private
prisons.39 For example, when Evalin-Ali Mandza had a heart attack at GEO’s Denver
Contract Detention Facility in Aurora, Colorado, medical staff did not even call 911
until nearly an hour after a code-blue emergency was called. At one point during this
hour-long delay, a nurse attempted to perform an electrocardiogram (EKG) on
Mandza but was unable to get a reading, then performed the wrong test, and then was
unable to interpret the results because she had not received the necessary training.
Another nurse prioritized filling out transfer paperwork over calling 911, even after a
strike_563918c3e4b0307f2caaf975; Kate Linthicum, Why 26 Asylum-Seekers in a California Detention
Center Have Stopped Eating, L.A. TIMES (Nov. 11, 2015, 5:00 AM),
http://www.latimes.com/local/lanow/la-me-ln-in-adelanto-hunger-strike-20151110-story.html; Aviva
Stahl, More than 100 US Immigrant Detainees Are on Hunger Strike this Thanksgiving, VICE NEWS
(Nov. 26, 2015, 5:10 PM), https://news.vice.com/article/more-than-100-us-immigrant-detainees-areon-hunger-strike-this-thanksgiving; Why Immigrant Detainees In California Just Launched A Hunger
Strike, THINK PROGRESS (Nov. 2, 2015),
http://thinkprogress.org/immigration/2015/11/02/3718005/adelanto-hunger-strike/; Roque Planas,
Immigration Detainees Begin Hunger Strike at Georgia Detention Center, HUFFINGTON POST (Apr. 21,
2016, 7:18 pm ET) http://www.huffingtonpost.com/entry/immigration-detainees-hungerstrike_us_57194fd3e4b0d0042da8bfb1.
36 Steven Hsieh, After Lawsuit, ICE Releases Hunger Strikers From Solitary Confinement, THE NATION
(Apr. 4, 2014), https://www.thenation.com/article/after-lawsuit-ice-releases-hunger-strikers-solitaryconfinement/.
37 Alexis Garcia-Ditta, Hutto Hunger Strike Reportedly Growing Despite ICE Denials, TEXAS OBSERVER
(Nov. 5, 2015), https://www.texasobserver.org/hutto-hunger-strike-reportedly-growing-despite-icedenials/.
38 A Shocking Glimpse Inside America’s Privatized Detention Facilities for Immigrants, THINKPROGRESS
(Mar. 10, 2016) http://thinkprogress.org/justice/2016/03/10/3757575/secretive-world-of-privatizedimmigrant-detention/.
39 American Civil Liberties Union, Detention Watch Network & National Immigrant Justice Center,
Fatal Neglect: How Ice Ignores Deaths in Detention 6 (Feb. 2016),
https://www.aclu.org/sites/default/files/field_document/fatal_neglect_acludwnnijc.pdf.

14

AMERICAN CIVIL LIBERTIES UNION

doctor had ordered the call to be made.40 In July 2016, Human Rights Watch analyzed
ICE death reviews from eighteen subsequent deaths and found similar failures. For
example, before his death at CCA’s Houston Contract Detention Facility, Peter George
Carlysle Rockwell complained of blurred vision to medical staff. Although a
physician’s assistant determined he should be seen by a higher-level provider within a
day, that appointment was never scheduled. Ten days later, Mr. Rockwell collapsed of
a hemorrhagic stroke, but staff failed to recognized the emergency, delayed calling
911, delayed starting CPR, and did not apply an automated external defibrillator
(AED) until 13 minutes after his collapse. By the time he was transported to the
hospital, Mr. Rockwell was already nonresponsive and had to be placed on a
ventilator; he died seven days later.41

II. HOW TO REDUCE THE
IMMIGRATION DETENTION
POPULATION
According to ICE, the agency keeps approximately 24,500 people in private prisons on
any given day.42 Accordingly, to reduce the immigration detention population enough
to cancel all of these contracts without major investments in new federal detention
facilities and without replacing private prisons with county jail beds, the government
must shift approximately that number of people from detention to release on
recognizance, release on bond, or release on various forms of supervision.
To carry out that shift, DHS should generally stop relying on detention as a means of
ensuring court appearance, and chiefly reserve detention for effectuating removal
orders. In place of its current mass detention apparatus, the government should rely
on community-based compliance mechanisms: release on recognizance, bond, and
community-based ATDs with case management services.
As of June 2016, more than 25,000 of the 37,000 people in ICE detention did not
have a final order of removal;43 depending on the outcome of their immigration or
expedited removal proceedings, these individuals may end up either being deported
or granted permission to remain in the United States. Replacing detention with
community-based compliance mechanisms for those with pending immigration
proceedings is a proven technique, is consistent with the best practices for pretrial
detention and supervision in the criminal justice system, and would dramatically
reduce the number of people held in detention.
This policy shift would not create significant public safety risks. Approximately 15,000
of the 25,000 detainees without final orders have no criminal record whatsoever. For
the 10,000 with criminal records, many could be safely supervised in the
community—including those subject to INA § 236(c)’s mandatory custody
Id. at 7-8.
Human Rights Watch, US: Deaths in Immigration Detention (July 7, 2016),
https://www.hrw.org/news/2016/07/07/us-deaths-immigration-detention.
42 Steven Nelson, Private Prison Companies, Punched in the Gut, Will Keep Most Federal Business, U.S.
NEWS & WORLD REPORT (Aug. 18, 2016), http://www.usnews.com/news/articles/2016-08-18/privateprison-companies-punched-in-the-gut-will-keep-most-federal-business.
43 ICE WRD Report, supra note 4, at 6.
40
41

15

AMERICAN CIVIL LIBERTIES UNION

requirements. A recent study found that half of those detained under INA § 236(c)’s
mandatory custody requirement would be candidates for release even using ICE’s
current, release-averse risk assessment.44 Moreover, ICE’s current risk assessment
system does not distinguish between people who recently committed serious offenses
and those who have been rehabilitated or “aged out” of crime.45 These distinctions are
important, because there are a range of tools available to secure the compliance of
people on supervision, depending on the severity and recency of their criminal history
and other risk factors. Indeed, there are many models that ICE can draw upon from
best practices in the criminal justice system.
Although shifting to a community-based model of compliance would represent a
major change in strategy for ICE, the idea is hardly unprecedented. The Vera Institute
of Justice recommended exactly this path for ICE’s predecessor agency in 2000:
How can the INS most effectively assure compliance with the law and, at the
same time, treat noncitizens in a just and humane manner? The research
suggests the answer is for INS to release to alternatives (such as community
supervision) as many people as it can, as quickly as it can, while they complete
their immigration court hearings. The alternatives would include not only
community supervision, at varying levels of intensity, but also the existing . . .
alternatives of bond, parole, and recognizance. Once individuals have
completed their hearings, the INS would reassign those required to leave the
country to the more rigorous alternatives—either detention or intensive
supervision. In this way, detention would be reserved for those who cannot be
released and, along with the most intense supervision, for people who are at
the stage of the removal process when they are most likely to abscond.46
Indeed, in the 16 years since the Vera Institute made its recommendations to INS, the
state of the art in criminal justice supervision has been further advanced by more
research and application in the field.
To carry out this reduction in the detention population, the government must commit
itself to two key tasks:
First, the government should embrace best practices from the criminal justice system
and replace most detention with a range of calibrated, community-based alternatives.
See Point III, infra.
Second, both DHS and DOJ should fully exercise their authority to release individuals
on parole and bond so that ICE no longer detains individuals who pose no significant
flight risk or danger to the community. See Point IV, infra.
Only both of these changes—a commitment to using appropriate and calibrated
community-based compliance tools, and the expansive use of release authority—will
allow the government to reduce the detained population enough so that DHS can end
its reliance on private prisons. Notably, the executive branch can carry out this shift
44 See Robert Koulish, Using Risk to Assess the Legal Violence of Mandatory Detention, 5:30 LAWS, at 20
(July 5, 2016), http://www.mdpi.com/2075-471X/5/3/30.
45 Id. at 11.
46 Vera Institute of Justice, Testing Community Supervision for the INA: An Evaluation of the
Appearance Assistance Program, Vol. 1, at 70 (Aug. 2000), available at
http://archive.vera.org/sites/default/files/resources/downloads/INS_finalreport.pdf.

16

AMERICAN CIVIL LIBERTIES UNION

from detention to community-based compliance mechanisms without any legislative
changes.
Separate and apart from the administrative policy reforms that DHS/ICE should
undertake, both Congress and the executive must scrap the 34,000-bed ICE detention
quota. This quota (an annual requirement that ICE “maintain” at least 34,000
detention beds regardless of operational needs) is arbitrary, irrational, wasteful, and
an extreme outlier. No corrections system in America operates under a similar
quota.47 At a time when DOJ and states across the country are instituting commonsense criminal justice reforms to reduce incarceration levels, it is unacceptable for
Congress to legislatively guarantee unnecessary profits for the private prison
companies and county jails that contract with ICE. Congress should repeal the bed
quota, and ICE should not permit the quota to drive any detention policies or
practices. Instead of filling an arbitrary number of beds each night, DHS should
concentrate its detention resources and operations on those immigrants with final
removal orders who are likely to be deported in the near future. Even with the quota
in place, ICE can begin a shift to community-based compliance mechanisms, since
ICE has enough flexibility to cancel a substantial number of contracts without
dropping below the arbitrary minimum number of beds.48

III. ICE MUST REPLACE DETENTION
WITH A RANGE OF CALIBRATED
ALTERNATIVES
This section describes how ICE can ensure immigration respondents comply with
immigration proceedings through various community-based mechanisms, in place of
detention.

47 Written Statement of the American Civil Liberties Union: Hearing on the Release of Criminal
Detainees by U.S. Immigration and Customs Enforcement: Policy or Politics?, Before the H. Comm. on
the Judiciary 5-6 (Mar. 19, 2013), available at
https://www.aclu.org/files/assets/aclu_statement_for_3_19_house_judiciary_committee_hearing_on
_immigration_enforcement_final_3_18_13.pdf; National Immigrant Justice Center, Eliminate the
Detention Bed Quota, http://www.immigrantjustice.org/eliminate-detention-bed-quota (last visited
Sept. 22, 2016).
48 ICE currently has contracts for a total detention capacity of approximately 44,453 beds, more than
10,000 beds above the congressional quota. See ICE ERO Custody Management Division, Authorized
DMCP Facilities (Dec. 8, 2015) (adult detention contracts have a defined capacity of 39,609 beds, plus at
least 1,190 “as-needed” beds potentially available); GEO Group, 2015 Annual Report at 10 (2016) (Karnes
has 1,158-bed family detention capacity); CCA, Annual Report (Form 10-K) for the fiscal year ended
December 31, 2015, at 16 (Feb. 25, 2016) (Dilley has 2,400-bed family detention capacity); A.B.A.
Commission on Immigration, Family Immigration Detention: Why the Past Cannot Be Prologue, at 19
(July 31, 2015),
https://www.americanbar.org/content/dam/aba/publications/commission_on_immigration/FINAL%2
0ABA%20Family%20Detention%20Report%208-19-15.authcheckdam.pdf (Berks has 96-bed family
detention capacity).

17

AMERICAN CIVIL LIBERTIES UNION

1.

ADOPT BEST PRACTICES FROM THE
CRIMINAL JUSTICE SYSTEM BY RESERVING
COERCIVE MEASURES AND INTENSIVE
SUPERVISION FOR THOSE WHO POSE THE
GREATEST RISK OF NON-COMPLIANCE

Alternatives to incarceration have widespread support in the pretrial criminal justice
context. They are recommended as cost-savers by the American Jail Association,
American Probation and Parole Association, American Bar Association, Association of
Prosecuting Attorneys, Heritage Foundation, International Association of Chiefs of
Police, National Conference of Chief Justices, National Sheriffs’ Association, Pretrial
Justice Institute, Texas Public Policy Foundation, and the Council on Foreign
Relations’ Independent Task Force on U.S. Immigration Policy.
This does not mean, however, that detention should simply be replaced with acrossthe-board intensive supervision. Criminologists widely agree that when designing
community correctional programs, it is important to calibrate the intrusiveness and
intensity of supervision to the individual, with frequent check-ins and monitoring
reserved for those who pose the highest risks of non-compliance.49 Directing too many
supervision resources and interventions at supervisees who are already low-risk can
actually have the counterintuitive effect of making them more likely to fail by
imposing multiple unnecessary burdens on them.50
In line with this principle, many detainees pose little risk of flight or danger to the
community and could be released on bond or their own recognizance, costing the
government nothing for supervision and monitoring.
Criminal justice research shows that inexpensive and nonintrusive interventions can
significantly enhance compliance for people released on recognizance or bond.
Reminding defendants of their court appearances—known as court date notification—
is a pretrial release intervention designed to reduce failures to appear (FTAs) and
associated costs. Numerous studies have demonstrated the effectiveness of court date
notification. With funding from the U.S. Department of Justice’s Bureau of Justice
Assistance, the Pretrial Justice Institute (PJI) conducted an extensive literature
review and identified court notification programs in six states that have been
rigorously evaluated. The target populations ranged from defendants issued a
citation/summons for minor offenses to those charged with felonies. All of the studies
concluded that court date notifications are effective at reducing FTA in court.51 For
example, a Colorado pilot project designed to measure the impact of telephone calls to
defendants concluded that live-telephone callers either reminding defendants to come
See, e.g., Christopher T. Lowenkamp & Edward J. Latessa, Understanding the Risk Principle: How
and Why Correctional Interventions Can Harm Low-Risk Offenders, TOPICS IN COMMUNITY CORRECTIONS
(2004); Pew Center on the States, Risk/Needs Assessment 101: Science Reveals New Tools to Manage
Offenders (Sept. 2011),
http://www.pewtrusts.org/~/media/legacy/uploadedfiles/pcs_assets/2011/pewriskassessmentbriefpdf.
pdf
50 See Lowenkamp & Latessa, supra note 49.
51 Pretrial Justice Institute, State of the Science of Pretrial Release Recommendations and Supervision
15-20 (June 2011),
https://www.pretrial.org/download/research/PJI%20State%20of%20the%20Science%20Pretrial%20Re
commendations%20and%20Supervision%20(2011).pdf.
49

18

AMERICAN CIVIL LIBERTIES UNION

to court or notifying them of their warrant status after the FTA could reduce FTAs by
more than half.52

2.

UTILIZE MORE INTENSIVE SUPERVISION AS
NEEDED

More intense forms of supervision and monitoring, such as enrollment in an ATD
program, carry higher costs than release on recognizance or bond, but those costs still
pale in comparison to the cost of detention. They are thus well-suited for mitigating
the risks associated with individuals who cannot be released on recognizance or a
simple bond. Current ATDs range in cost from pennies,53 to $8.49 per person per
day54 depending on the type of monitoring.
The most restrictive alternatives, like ankle-attached GPS monitors, are very effective
for higher-risk individuals but also the most costly. Ankle monitors require
confinement in a specific space for many hours per day to charge the device. Not only
is this a substantial restriction on liberty, but it is more expensive for ICE and is rarely
necessary for an asylum-seeker who has every incentive to attend her hearing. DHS
should only use GPS devices when no other conditions could reasonably ensure public
safety and compliance with the immigration process. When ICE has used intensive
supervision, however, it has achieved impressive compliance. A recent GAO report
found that over 95 percent of those on “full-service” ATDs (which include both case
management and technology-based tracking) appeared for their final hearings.55
For the vast majority of immigrants who require supervision, community-support
ATD models are far more appropriate than electronic monitoring. These models are
particularly well-suited for asylum seekers, torture survivors, the elderly, individuals
with medical and mental health needs, and other vulnerable groups. To be most
effective, ATD programs must be operated by community-based nonprofit
organizations, as these organizations are trained and equipped to identify the needs of
immigrants and to build trust with immigrant communities.
Holistic programs that offer case management services and facilitate access to legal
counsel as well as safe and affordable housing have been shown to substantially
increase program compliance without the extensive use of electronic monitoring.
Previous pilots have shown excellent results:
•

95.6 percent appearance rate: In 2013 Lutheran Immigration and
Refugee Services (LIRS) entered into a memorandum of understanding with

52 Timothy Schnacke et. al., Increasing Court Appearance Rates and Other Benefits of Live-Caller
Telephone Court Date Reminders: The Jefferson County, Colorado, FTA Pilot Project and Court Date
Notification Program 15 (May 2011),
https://www.pretrial.org/download/research/Jefferson%20County%20CO%20Increasing%20Court%20
Appearance%20Rates%202011.pdf.
53 S. Comm. on Appropriations, Summary of Emergency Supplemental Funding Bill,
http://www.appropriations.senate.gov/news/chairwoman-mikulski-releases-summary-emergencysupplemental-funding-bill (Jul. 2014).
54 ICE FY 2016 Congressional Budget Justification,
http://www.dhs.gov/sites/default/files/publications/DHS_FY2016_Congressional_Budget_Justificatio
n.pdf.
55 Government Accountability Office, Alternatives to Detention: Improved Data Collection and Analyses
Needed to Better Assess Program Effectiveness (Nov. 2014),
http://www.gao.gov/assets/670/666911.pdf.

19

AMERICAN CIVIL LIBERTIES UNION

ICE to screen vulnerable immigrants for release and enrollment in LIRS’
Community Support Initiative. Between June 2013 and Nov. 2014, 44 out of
46 formal referrals were in full compliance.56
•

96 percent appearance rate at 3 percent the cost of detention: In
1999 the Immigration and Naturalization Service (INS) partnered with
Lutheran Immigration and Refugee Service to assist 25 Chinese asylum
seekers released from detention. INS released the asylum seekers into open
shelters around the country, where they received housing, food, medical care,
and continuous case management. Participants had a 96 percent appearance
rate and the annual program costs were just 3 percent of what it would have
cost to detain them.57

•

97 percent appearance rate: Between 1999 and 2002 INS collaborated
with Catholic Charities of New Orleans to work with 39 asylum seekers
released from detention and 64 “indefinite detainees” who could not be
removed from the United States. The court appearance rate for participants
was 97 percent and the program cost $1,430 per year per client, a fraction of
the cost of detaining them.58

3.

SECURE COMPLIANCE OF FAMILIES
THROUGH ATTORNEYS

It is well-documented that being represented by an attorney is a strong indicator that
a family will appear for their immigration court hearings. According to an analysis of
2015 data from the Executive Office of Immigration Review (EOIR) by Human Rights
First, 98 percent of families with legal counsel are in compliance with their obligations
to appear for court hearings.59
Accordingly, if DHS and DOJ simply ensure that all family units in immigration
proceedings receive representation by attorneys, there will be no reason to keep these
families detained except on the rare occasions when a particular parent or child poses
an unacceptable risk to public safety.

Women’s Refugee Commission et. al., The Real Alternatives to Family Detention,
https://www.womensrefugeecommission.org/images/zdocs/Real-Alternatives-to-Family-Detention.pdf
57 Lutheran Immigration and Refugee Service, Unlocking Liberty: A Way Forward for U.S. Immigration
Detention Policy (Oct. 27, 2011), www.lirs.org/dignity.
58 Sue Weishar, A More Human System: Community-Based Alternatives to Immigration Detention
(Part 2), Just South Quarterly (2010).
59 Human Rights First, Myth vs. Fact: Immigrant Families’ Appearance Rates in
Immigration Court (July 2015), http://www.humanrightsfirst.org/sites/default/files/MythvFactImmigrant-Families.pdf.
56

20

AMERICAN CIVIL LIBERTIES UNION

IV. AVOID DETAINING PEOPLE WHO
POSE LITTLE DANGER OR FLIGHT
RISK, AND USE SUPERVISION TO
MITIGATE RISKS FOR OTHERS
Currently, ICE severely underutilizes its existing statutory authority to release
individuals on parole and bond. Through the measures described below, ICE could
significantly reduce the number of people who remain unnecessarily detained. We
have estimated the likely impact of each measure based on publicly available data.

1.

END FAMILY DETENTION AND THE MASS
DETENTION OF ASYLUM SEEKERS

Recommendations: Close all three family detention centers, utilize immigration
court (Immigration and Nationality Act § 240) removal proceedings rather than
expedited removal for family units, and release asylum-seeking families to sponsors in
the community.
Clarify to all ICE field offices that the 2009 Directive—Parole of Arriving Aliens Found
to Have a Credible Fear of Persecution or Torture—remains in full force and has not
been rescinded by the DHS 2014 Enforcement Priorities Memorandum.
Estimated Reduction in Detained Population: 11,000 to 15,000 people60
Asylum seekers are people who are entering the United States in order to seek
protection from harm in their home countries—a status that is protected under
international law. Under the 1951 Convention relating to the Status of Refugee and its
1961 Protocol, asylum seekers are not to be penalized for arriving at a country of
refuge without immigration documentation, and the detention of asylum seekers is
subject to narrow limits.61
Asylum seekers are generally good candidates to be released while their immigration
cases are pending. Empirical research has found that asylum seekers fleeing
60 See Dep’t of Homeland Security, Detained Asylum Seekers, Fiscal Year 2014 Report to Congress, at 8
T.1, 23 T. 7 (Sept. 9, 2015) (FY 14 asylum detention data obtained by Human Rights First via FOIA);
discussion supra note 48 (family detention capacity). The average daily population (ADP) of detained
asylum-seekers was approximately 8,170 in FY 2014 (the most recent statistics available). This ADP was
calculated using FY 14 statistics on the number and average lengths of stay for detained affirmative
asylum applicants, credible fear asylum applicants, and defensive asylum applicants, using the following
formula: ADP = Admissions x ALOS / 365. Due to the dramatic increase in use of family detention and
the 86% increase in findings of credible fear between FY 2014 and FY 2016, however, the FY 2014 asylum
statistics likely represent a significant undercount. To account for this, the ACLU added a range of
between 3,000 and 7,000 additional people to the estimate to account for increased detention of adult
asylum-seekers and changes in family detention (in particular, the activation and full ramp-up of the
Dilley and Karnes family detention facilities) during this time period.
61 Human Rights First, Lifeline on Lockdown: Increased U.S. Detention of Asylum Seekers 8-9 (July
2016), http://www.humanrightsfirst.org/sites/default/files/Lifeline-on-Lockdown_0.pdf.

21

AMERICAN CIVIL LIBERTIES UNION

persecution are predisposed to comply with legal processes, even if they might lose
their cases, and can be effectively supervised in the community. Releasing asylum
seekers (on alternatives as needed) and affording legal assistance can protect the
rights of asylum seekers and facilitate compliance with proceedings and legitimate
removals, at far less human and financial cost than detention. 62 But in recent years,
the U.S. has significantly increased its detention of asylum seekers. The shift follows—
and appear to be influenced by—two major policy changes announced by the Obama
administration in 2014: a policy of deterring Central American families from seeking
asylum by detaining them en masse, and a November 2014 immigration enforcement
priorities memorandum that deems people “apprehended at the border or at ports of
entry attempting to unlawfully enter the United States” as top enforcement
priorities.63
Government data show a sharp decrease in parole grants in the wake of these policy
changes, even though they did not actually rescind the existing 2009 Asylum Parole
Directive.64 In 2012, before these changes, ICE granted parole to 80 percent of
arriving asylum seekers who established a credible fear.65 By contrast, in the first nine
months of 2015, ICE granted parole to only 47 percent of arriving asylum seekers who
established a credible fear. In one ICE district, the grant rate was as low as 12
percent.66 A recent report by Human Rights First documented numerous denials of
parole even when asylum seekers meet the Parole Directive’s criteria.67
Decreased parole grants translate into increased detention of asylum seekers. In FY
2014, ICE held 44,270 asylum seekers in immigration detention facilities—a threefold increase from 2010, when the agency detained 15,769 asylum seekers.68
Additionally, under the administration’s family detention policy, approximately 3,600
mothers and children seeking asylum are held in family detention centers at any given
time. This is the federal government’s largest project of detaining families since the
mass incarceration of Japanese Americans during World War II. The policy
overwhelmingly affects families with legitimate asylum cases; in Q2 2016 (the most
62 See generally Mark Noferi, A Humane Approach Can Work: The Effectiveness of Alternatives to
Detention for Asylum Seekers (July 2015) available at http://cmsny.org/publications/noferi-detentionasylum/.
63 See Memorandum from Jeh Johnson, Secretary of Homeland Security, regarding Policies for the
Apprehension, Detention, and Removal of Undocumented Immigrants (Nov. 20, 2014),
https://www.dhs.gov/sites/default/files/publications/14_1120_memo_prosecutorial_discretion.pdf;
DHS, Statement by Secretary Johnson Regarding Today’s Trip to Texas (Dec. 15, 2014),
https://www.dhs.gov/news/2014/12/15/statement-secretary-johnson-regarding-today%E2%80%99strip-texas.
64 The 2009 Asylum Parole Directive provides that an arriving asylum seeker determined to have a
“credible fear” of persecution should generally be paroled from detention if his or her “identity is
sufficiently established, the alien poses neither a flight risk nor a danger to the community, and no
additional factors weigh against release.” ICE Directive 11002.1, Parole of Arriving Aliens Found to Have
A Credible Fear of Persecution or Torture (2009), available at
https://www.ice.gov/doclib/dro/pdf/11002.1-hd-parole_of_arriving_aliens_found_credible_fear.pdf.).
65 USCIRF, Assessing the U.S. Government’s Detention of Asylum Seekers, supra note 13, at 9-10.
66 ICE data obtained through a Freedom of Information Act Request by the ACLU and the Center for
Gender and Refugee Studies at the UC Hastings School of Law.
67 HRF, Lifeline on Lockdown, supra note 61, at 17-22. These include denials based on the November
2014 enforcement priorities; denials based on unexplained assertions of flight risk; denials based on the
purported failure to establish identity even where asylum seekers have submitted extensive
documentation; and denials without the interview required by the Parole Directive or any explanation of
the reasons for the denial.
68 See id. at 11.

22

AMERICAN CIVIL LIBERTIES UNION

recent quarter that is publicly available), 89.5% of detained families established
credible fear.69 And this detention is being carried out primarily by private prison
companies, which manage the vast majority of the family detention beds now in
operation.70 The detention of adult asylum seekers is similarly reliant on private
prisons. Five detention facilities—all of which are privately operated—held nearly a
third of all detained asylum seekers nationwide in 2014.71
The government’s mass detention of asylum seekers raises serious human rights and
civil rights concerns, imposes harms that are wildly out of proportion to the
government’s interest in assuring people’s presence at court hearings, and is simply
the wrong policy for a country that should act as a place of welcome and beacon of
hope in an unstable world.

2.

END PROLONGED DETENTION WITHOUT
BOND HEARINGS

Recommendation: DHS and DOJ should construe the general immigration
detention statutes to require a bond hearing before an immigration judge for all
individuals detained more than six months, where the government must justify
continued detention.
Estimated Reduction in Detained Population: At least 4,500 people72
As a result of court backlogs, people who challenge their deportation in immigration
court are subjected to the longest average detention times.73 Many immigrants—
including asylum seekers and longtime lawful permanent residents—are incarcerated
for months or even years while the immigration courts and federal courts resolve their
immigration cases. Moreover, many detainees never receive the basic due process of a
bond hearing to determine whether they can be released while their case is pending.
As a result, many detainees are subjected to prolonged detention even though they
have substantial challenges to removal and pose no flight risk or significant danger to
public safety. Indeed, perversely, individuals who are likely to be legally entitled to
remain in the United States are especially vulnerable to prolonged detention, as they
have the strongest incentives to fight their cases despite being detained.74
69 See USCIS Asylum Division, Family Facilities Credible Fear (May 16, 2016),
https://www.uscis.gov/sites/default/files/USCIS/Outreach/Upcoming%20National%20Engagements/P
ED_CF_RF_FamilyFacilitiesFY14_16Q2.pdf.
70 See CCA, South Texas Family Residential Center, http://www.cca.com/facilities/south-texas-familyresidential-center; GEO Group, Karnes County Residential Center,
http://www.geogroup.com/maps/locationdetails/23.
71 HRF, Lifeline on Lockdown, supra note 61, at 12-13. The five facilities are CCA’s T. Don Hutto
Residential Center in Taylor, Texas; CCA’s South Texas Detention Complex in Pearsall, Texas; GEO’s
Coastal Bend Detention Facility in Robstown, Texas; CCA’s Eloy Detention Center in Eloy, Arizona; and
GEO’s LaSalle Detention Facility in Jena, Louisiana.
72 This estimate is derived from an analysis of EOIR data obtained through the Freedom of Information
Act of detained respondents whose removal cases were pending before EOIR for at least six months.
73 Transactional Records Access Clearinghouse, Legal Noncitizens Receive Longest ICE Detention,
Syracuse University (June 3, 2013), http://trac.syr.edu/immigration/reports/321/.
74 ACLU, Prolonged Detention Fact Sheet,
https://www.aclu.org/sites/default/files/assets/prolonged_detention_fact_sheet.pdf; TRAC
Immigration, Legal Noncitizens Receive Longest ICE Detention,
http://trac.syr.edu/immigration/reports/321/ (“In a perverse way, individuals who were legally entitled
to remain in the United States typically experienced the longest detention times”).

23

AMERICAN CIVIL LIBERTIES UNION

Such unnecessary detention causes harm to individuals and their families and
imposes a significant financial burden on U.S. taxpayers, at an average cost of $186.68
per detainee per day.75
This prolonged lack of access to bond hearings is also wildly out of step with practices
in the criminal justice system, where judicial decisions regarding bond and release
conditions are typically made within hours or days of arrest. In federal criminal cases,
for example, these decisions are made at the defendant’s initial appearance, which
must be held “without unnecessary delay” before the nearest available magistrate or
judicial officer.76
Six Courts of Appeals have held that prolonged immigration detention without
constitutionally adequate review raises serious due process concerns, and the Second
Circuit and Ninth Circuit have specifically identified six months as the presumptive
point in time after which a bond hearing is required.77 By adopting a presumptive sixmonth limit on detention without a bond hearing, after which time the government
must justify further detention at a bond hearing before an immigration judge, DHS
and DOJ would reconcile agency practice with these six appellate court rulings and
significantly reduce unnecessary detention.78

75 This was calculated by dividing ICE’s FY 2016 custody operations cost ($2,316,744,000) by 365 days to
obtain a daily custody operations cost ($6,347,243.84) and dividing that by the number of FY 2016
authorized detention beds (34,000). See Dep’t of Homeland Security FY 2017 Congressional Budget
Justification, Vol. 2, at 3, 5, 6,
https://www.dhs.gov/sites/default/files/publications/FY%202017%20Congressional%20Budget%20Jus
tification%20-%20Volume%202_1.pdf. This is the same methodology that the National Immigration
Forum used to calculate the average per-detainee, per-day cost of detention using FY 2013 statistics. See
National Immigration Forum, The Math of Immigration Detention, at 2 (Aug. 2013),
https://immigrationforum.org/wp-content/uploads/2014/10/Math-of-Immigation-Detention-August2013-FINAL.pdf. The substantial cost increase from FY 2013 appears to be driven by the rise of family
detention beginning in FY 2014.
76 See 18 U.S.C. § 3142(f) (directing that decisions regarding conditions of release or detention be made at
the initial appearance unless a continuation is granted, and limiting the duration of such continuances);
Fed. R. Crim. P. 5(a) (requiring person arrested to be taken for initial appearance “without unnecessary
delay”); Mallory v. United States, 354 U.S. 449, 455 (1957) (where defendant “was arrested in the early
afternoon and was detained at headquarters within the vicinity of numerous committing magistrates,”
police decision to delay arraignment by six hours to interrogate him violated Fed. R. Crim. P. 5(a)).;
Corley v. United States, 556 U.S. 303, 320 (2009) (Rule 5 presentment requirement “stretches back to
the common law”).
77 See Lora v. Shanahan, 804 F.3d 601 (2d Cir. 2015); Rodriguez v. Robbins, 804 F.3d 1060 (9th Cir.
2015), cert. granted Jennings v. Rodriguez, 15-1204 (June 20, 2016). Other circuits have adopted a caseby-case analysis for when mandatory detention is unreasonably prolonged. See Reid v. Donelan, 819 F.3d
486 (1st Cir. 2016); Diop v. ICE/Homeland Security, 656 F.3d 221 (3d Cir. 2011); Sopo v. Attorney
General, 825 F.3d 1199 (11th Cir. 2016); Ly v. Hansen, 351 F.3d 263 (6th Cir. 2003) (requiring release
when mandatory detention exceeds a reasonable period of time).
78 This rule would apply to the general immigration detention statutes, 8 U.S.C. §§ 1225, 1226, and 1231,
and not the detention statutes that apply specifically to individuals detained for terrorism or other
national security reasons, 8 U.S.C. §§ 1226a, 1531–1537. A six-month period would also be consistent
with the Supreme Court’s ruling in Demore v. Kim, which upheld the constitutionality of mandatory
detention under 8 U.S.C. § 1226(c) for the “brief period of time necessary” for removal proceedings—a
period that it understood to average between a month and a half and five months—as well as the Court’s
holding in Zadvydas v. Davis that the detention of individuals with final orders of removal beyond six
months is presumptively unreasonable. 538 U.S. 510, 513 (2003); 533 U.S. 678, 701 (2001).

24

AMERICAN CIVIL LIBERTIES UNION

3.

INTERPRET THE MANDATORY CUSTODY
STATUTE TO PERMIT A RANGE OF CUSTODIAL
O P T I O N S , A N D A P P L Y IT O N L Y T O
IMMIGRANTS RECENTLY CONVICTED OF
SERIOUS CRIMES WHO DO NOT HAVE
MERITORIOUS IMMIGRATION CASES

Recommendations: Recognize that electronic monitoring, house arrest, and other
coercive liberty restrictions satisfy INA § 236(c)’s “custody” requirement. Narrow
application of mandatory custody to exclude those with substantial challenges to
removal. Properly construe INA § 236(c) to apply only to individuals who are taken
into ICE custody at or near the time of their release from criminal custody.
Estimated Reduction in Detained Population: 5,000 to 10,000 people79
INA § 236(c), codified at 8 U.S.C. § 1226(c), requires that noncitizens—including
many longtime lawful permanent residents–with certain criminal convictions be
mandatorily kept in ICE custody for the “brief period” necessary for removal
proceedings.80 In many cases, the convictions at issue are very minor, such as drug
possession or shoplifting. INA § 236(c) is routinely applied to require detention even
when the person poses no flight risk or danger and is highly likely to be eligible for
immigration relief that would allow them to remain in the country.
Former government officials and legal scholars have long criticized the mandatory
detention statute as imposing an overly restrictive constraint on the agency’s
discretion to make optimal use of its limited detention resources.81 The government
has made a bad situation even worse by adopting an expansive interpretation of INA §
236(c) that is not required by the statute and, in many cases, applies mandatory
detention in ways that raise serious constitutional concerns. By limiting INA § 236(c)
to its proper scope, DHS will gain flexibility to ensure that detention is used only
where an individual poses a danger or flight risk that requires confinement, and not
imposed where an individual may be safely supervised in the community. The
government should do this in several ways.

The lower bound of 5,000 people was calculated by analyzing the number of detained respondents with
at least one INA § 236(c) predicate charge whose case was pending before EOIR in June 2016. The upper
bound of 10,000 people is based on the number of detainees with criminal records as of June 2016. See
ICE WRD Report, supra note 4, at 6. Although not all detainees with criminal records are subject to INA
§ 236(c), the mandatory custody requirement encompasses many low-level offenses, making it difficult to
identify which detainees would be subject to this requirement without knowing their actual offense
histories.
80 See Demore, 538 U.S. at 513.
81 See, e.g., Margaret Taylor, Demore v. Kim: Judicial Deference to Congressional Folly, Immigration
Stories (2005), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=739231.
79

25

AMERICAN CIVIL LIBERTIES UNION

A.

RECOGNIZE THAT ELECTRONIC
MONITORING, HOUSE ARREST, AND
OTHER COERCIVE LIBERTY
RESTRICTIONS SATISFY 236(C)’S
“CUSTODY” REQUIREMENT

For those individuals who are properly subject to INA § 236(c), ICE should frequently
use electronic monitoring, house arrest, or other restrictive forms of custody short of
costly detention. This approach would address ICE’s legitimate interest in ensuring
court appearance and community safety, but at far lesser humanitarian and fiscal
costs.
INA § 236(c) provides that the Attorney General “shall take into custody any alien
who” is “deportable” or “inadmissible” for a qualifying crime, and prohibits that
person’s “release . . . from custody” except for purposes of the federal Witness
Protection Program. INA § 1226(c) (emphasis added). However, INA § 236(c) does
not place any textual limitation on the term “custody”, and the term is not defined
anywhere in the INA or immigration regulations.82
DHS should interpret the term “custody” in INA § 236(c) as including not only
physical detention but alternative forms of custody such as electronic monitoring,
curfews, and home detention. This definition of “custody” is consistent with federal
law, which defines custody broadly. Federal habeas law, for example, defines custody
as restraints on a person’s liberty that are “not shared by the public generally.”83
Courts have interpreted custody as including probation, release on recognizance with
conditions of supervision, mandatory attendance in a rehabilitative program, and
even the intensive supervision requirements under ICE’s ISAP program.84 In the
criminal context, the Supreme Court has clarified in Reno v. Koray that the
distinction between custody and release turns on whether a defendant is “subject to
the control” of the Bureau of Prisons, not whether he or she is physically
incarcerated.85 Federal court precedent thus suggests that the term “custody” has a
broad meaning that includes alternatives such as home arrest and electronic
monitoring. It is well within DHS’s discretion—as well as sound policy—to adopt a
broader interpretation of “custody” that includes restrictive forms of supervision short
of costly detention.
By contrast, 8 U.S.C. § 1226(a), which provides general detention authority to the government in
immigration cases, is worded in terms of “detention” and nowhere mentions custody.
83 Jones v. Cunningham, 371 U.S. 236, 240 (1963).
84 In the federal habeas context, custody has been found to include probation, Olson v. Hart, 965 F.2d
940 (10th Cir. 1993), parole, Jones v. Cunningham, 371 U.S. 236 (1963); DePompei v. Ohio Adult Parole
Authority, 999 F.2d 138 (6th Cir. 1993), conditions placed on a defendant released on recognizance while
his case was on appeal, Hensley v. Municipal Court, 411 U.S. 345, 351 (1973), and mandatory attendance
at rehabilitative program, Dow v. Circuit Court of the First Circuit, 995 F.2d 922, 923 (9th Cir. 1993). By
rendering its decision in Nguyen v. B.I., Inc., 435 F. Supp. 2d 1109 (D. O.R 2006), the district court de
facto recognized that supervision under ISAP, including wearing an ankle bracelet and 12-hour curfews,
constituted custody. In a separate context, under the federal escape statute, 18 U.S.C. § 751(a), escape or
attempted escape from custody may occur from a pre-release guidance center, work release program, or
halfway house. See United States v. Rudinsky, 439 F.2d 1074 (6th Cir. 1971); Perez-Calo v. United States,
757 F. Supp. 1 (D. Puerto Rico 1991).
85 515 U.S. 50, 62-63 (1995) (holding, in sentencing context, that whether an individual is “released”
depends on if he remains “subject to [the custodian’s] control,” and not whether he is still subject to “jaillike conditions”).
82

26

AMERICAN CIVIL LIBERTIES UNION

Instead, however, DHS narrowly interprets “custody” to require the physical detention
of individuals. This interpretation is inconsistent with the use of the term “custody” in
federal law generally, and results in the unnecessary detention of everyone subject to
INA § 236(c). Just as the Bureau of Prisons exercises broad discretion to determine
which form of custody to use in individual cases, ICE should be making discretionary
decisions about which forms of custody to use for people subject to INA § 236(c).

B.

NARROWLY APPLY MANDATORY
CUSTODY TO EXCLUDE THOSE WITH
SUBSTANTIAL CHALLENGES TO
REMOVAL

INA § 236(c) requires ICE to maintain in its custody noncitizens who are “deportable”
or “inadmissible” on designated criminal grounds. In Matter of Joseph,86 the BIA
broadly construed INA § 236(c) to apply to any noncitizen who is charged with
deportability or inadmissibility under one of the designated grounds, unless the
individual can show that the government is “substantially unlikely” to establish the
charges. This highly expansive standard, however, is not required by the statute or
regulations, and—in combination with DHS’s narrow construction of the term
“custody”—results in the across-the-board detention of many individuals who have
substantial challenges to removal, including claims to relief that would permanently
entitle a noncitizen to remain in the United States, such as cancellation of removal.87
For example, in nearly 20% of cases that were decided in FY 2015, the Immigration
Court ruled in favor of immigrants held in mandatory custody.88
DHS and DOJ should reject the overbroad interpretation of the statute set forth in
Joseph. 89 Rather, where an individual has a substantial challenge to removal,
including a claim to permanent immigration relief, the government should not subject
them to mandatory detention, but provide a bond hearing to determine if s/he poses a
danger or flight risk that can only be mitigated with actual detention.
22 I & N Dec. 799 (BIA 1999).
Indeed, under the Joseph standard, even an individual whose proceedings are terminated by an
immigration judge on the grounds that he or she is not removable is subject to mandatory detention if
ICE chooses to appeal that decision to the BIA, unless the individual can demonstrate that the
government is “substantially unlikely to prevail” on its appeal. See Joseph, 22 I&N Dec. at 665.
88 EOIR, Certain Criminal Charge Completion Statistics 2 (Aug. 2016),
https://www.justice.gov/sites/default/files/pages/attachments/2016/08/25/criminal-chargecompletion-statistics-201608.pdf.
89 Notably, the Supreme Court in Demore upheld only the mandatory detention, for a brief period of
time, of an individual who had conceded deportability, because such detention was reasonably related to
the purpose of INA § 236(c). 538 U.S. at 527-28. In contrast, the mandatory detention of individuals with
substantial challenges to removal raises serious due process concerns because such individuals do not
pose the categorical flight risk or danger to public safety that warrants their mandatory imprisonment.
See, e.g., Gonzalez v. O’Connell, 355 F.3d 1010 (7th Cir. 2004) (noting that Supreme Court specifically
left open this question in Demore); Demore, 538 U.S. at 577 (Breyer, J., dissenting) (concluding that the
constitutional claim to bail where an individual raises a substantial challenge to removal is “strong”);
Tijani, 430 F.3d at 1246-47 (Tashima, J. concurring) (concluding that Joseph standard is “egregiously”
unconstitutional in case of LPR challenging deportability); Casas v. Devane, No. 15-cv-8112, 2015 WL
7293598 (N.D. Ill. Nov. 19, 2015) (holding mandatory detention of person with good faith challenge to
removal unconstitutional; petitioner sought post-conviction relief from guilty plea due to ineffective
assistance of counsel); Papazoglou v. Napolitano, 2012 WL 1570778, at *5 (N.D. Ill. May 3, 2012)
(mandatory detention of individual whom the immigration judge had granted new adjustment of status
to lawful permanent residence “present[ed] a question of constitutionality”).
86
87

27

AMERICAN CIVIL LIBERTIES UNION

C.

PROPERLY CONSTRUE THE MANDATORY
CUSTODY STATUTE TO APPLY ONLY TO
INDIVIDUALS WHO ARE TAKEN INTO ICE
CUSTODY AT THE TIME OF THEIR
RELEASE FROM CRIMINAL CUSTODY

Finally, the government should stop misapplying mandatory custody to individuals
who are taken into ICE custody months or years after serving their related criminal
sentence. Congress envisioned mandatory custody to require ICE to take custody of
noncitizens “when the alien is released” from serving their criminal sentences for
certain designated crimes,90 meaning at the time of such release. Nonetheless, under
the BIA’s decision in Matter of Rojas,91 the government applies INA § 236(c) to
individuals it takes into custody any time after their release from criminal custody,
even if the release occurred nearly 20 years ago, when the statute went into effect. As
a result, instead of using the mandatory custody statute to ensure a continuous chain
of custody between the criminal justice and immigration enforcement systems, ICE is
applying mandatory custody to individuals who have been at liberty for months or
years and leading productive lives in their communities.
The Ninth Circuit has rejected Matter of Rojas, holding that “[u]nder the plain
language of 8 U.S.C. § 1226(c), the government may detain without a bond hearing
only those criminal aliens it takes into immigration custody promptly upon their
release” from criminal custody for an offense referenced in the mandatory detention
statute.92 Rojas also has been rejected by most district courts to consider the question
and by three of six judges sitting en banc in the First Circuit.93 By contrast, four other
circuits have agreed with the government’s position, albeit on different grounds.94
Nothing precludes the government from adopting a narrower construction of the
statute that conforms to congressional intent and makes more efficient use of the
government’s limited resources. DHS and DOJ should abandon the overbroad
interpretation set forth in Matter of Rojas and apply mandatory custody only to
individuals ICE apprehends at the time of their release from criminal custody.

8 U.S.C. § 1226(c).
23 I. & N. Dec. 117 (BIA 2001).
92 Preap v. Johnson, --- F.3d ---, 2016 WL 4136983, at *11 (9th Cir. 2016). The Ninth Circuit specifically
affirmed district court orders requiring bond hearings for detainees in California and Washington State
who were not immediately detained upon their release from relevant criminal custody); See also Khoury
v. Asher, No. 14- 35482, 2016 WL 4137642, at *1 (9th Cir. Aug. 4, 2016) (unpublished).
93 See Castañeda v. Souza, 810 F.3d 15, 18–43 (1st Cir. 2015) (en banc) (Barron, J.).
94 See Lora v. Shanahan, 804 F.3d 601 (2d Cir. 2015); Sylvain v. Attorney General, 714 F.3d 150 (3d Cir.
2013); Hosh v. Lucero, 680 F.3d 375 (4th Cir. 2012); Olmos v. Holder, 780 F.3d 1313 (10th Cir. 2015).
90
91

28

AMERICAN CIVIL LIBERTIES UNION

4.

STOP IMPOSING EXORBITANT,
UNAFFORDABLE BONDS

Recommendation: DHS and DOJ should require that immigration officials use the
least restrictive conditions necessary to ensure the individual’s appearance for court
proceedings, including alternatives to detention, when determining such conditions
under INA § 236(a) or 212(d)(5). Moreover, where ICE or an Immigration Judge sets
a cash bond, they must consider the individual’s ability to pay and impose no bond
amount greater than necessary to ensure appearance. Finally, DHS should make a
more flexible range of secured and unsecured bonds available, rather than require
individuals to post the full cash amount to be released.
Estimated Reduction in Detained Population: At least 1,300 people95
Both the federal government and a growing consensus of federal courts have
recognized that incarcerating criminal defendants solely because they cannot afford to
pay bond is unconstitutional.96 As the U.S. Department of Justice recently explained
in a Georgia bail case, “a bail scheme that imposes financial conditions, without
individualized consideration of ability to pay and whether such conditions are
necessary to assure appearance at trial, violates the [due process and equal protection
guarantees of the] Fourteenth Amendment.”97 “[A] jurisdiction may not use a bail
system that incarcerates indigent individuals without meaningful consideration of
their indigence and alternative methods of assuring their appearance at trial.”98
The same concerns apply to immigrants kept in detention solely because they cannot
afford to pay an ICE or immigration court bond. Nonetheless, neither ICE officials nor
Immigration Judges routinely consider a person’s ability to pay when setting bond or
alternative conditions of supervision. Consequently, immigration detainees are
routinely locked up solely due to their lack of money, without any finding that they
pose a danger or flight risk that cannot be mitigated through other means.

95 This is a very conservative estimate based on a national extrapolation of the average daily population of
people detained for nontrivial periods of time on bonds they cannot afford to pay in the Central District
of California. The Central District of California data are derived from Hernandez v. Lynch, No. 5:16-cv00620 (C.D. Cal.).
96 See, e.g., Jones v. City of Clanton, No. 2:15-cv-00034, 2015 WL 5387219, at *2-3 (M.D. Ala. Sept. 14,
2015); Walker v. City of Calhoun, GA, No. 4:15–CV–0170–HLM, 2016 WL 361612, at *10-11, 14 (N.D.
Ga. Jan. 28, 2016) (granting class-wide preliminary injunction); Pierce v. City of Velda City, MO, No.
4:15-cv-00570-HEA, 2015 WL 10013006, at *1 (E.D. Mo. June 3, 2015) (Dkt. 16) (issuing declaratory
judgment); Thompson v. Moss Point, MS, No. 1:15-cv-00182-LG-RHW, 2015 WL 10322003, at
*1 (S.D. Miss. Nov. 16, 2015) (Dkt. 18) (same); Cooper v. City of Dothan, 1:15-cv-00425-WKW-TFM,
2015 WL 10013003, at *3-4 (M.D. Ala. June 18, 2015) (Dkt. 7) (granting temporary restraining order);
see also Rodriguez v. Providence Cmty. Corr., Inc., No. 3:15-cv-01048, 2015 WL 9239821, at *1, 6-7
(M.D. Tenn. Dec. 17, 2015) (holding that detaining misdemeanor probationers on money bail without an
inquiry into their ability to pay was unconstitutional and granting class-wide preliminary injunction).
97 Br. for the U.S. as Amicus Curiae, p.13, Walker v. City of Calhoun, 2016 WL 4417419 (2016), available
at https://www.justice.gov/crt/file/887436/download; accord Statement of Interest of the United States,
p. 1, Jones v. City of Clanton (hereinafter “DOJ Statement”), 2015 WL 5387219 (2015); U.S. Dep’t of
Justice, Dear Colleague Letter, Mar. 14, 2016, at 7, available at
https://www.justice.gov/crt/file/832461/download (“courts must not employ bail or bond practices that
cause indigent defendants to remain incarcerated solely because they cannot afford to pay for their
release.”).
98 Id. DOJ Statement at 21.

29

AMERICAN CIVIL LIBERTIES UNION

As in the criminal justice context, the government should adopt bond-setting
procedures to ensure that people in civil immigration proceedings are not
incarcerated due to their poverty.
First, when setting conditions of release, the government should apply a standard at
least as protective as the one that the federal Bail Reform Act applies to criminal
defendants in federal court: use only the least restrictive means required to ensure
appearance to guarantee that noncitizens are not unnecessarily deprived of their
liberty. Under this standard, ICE and the IJ should first consider whether an
individual can be released on his own recognizance and, if not, apply only those
conditions of release (including bond and conditions of supervision) that are
reasonably necessary to ensure his appearance.99
Second, the government should end its policy and practice of requiring immigration
detainees to post the full cash value of the immigration bond to obtain release. The
government’s reliance on full cash bonds—without consideration of an individual’s
financial ability to pay the bond—is out of step with well-established procedures in
federal and state courts, results in the detention of individuals based solely on their
poverty, and is not required by the governing statute.100 Bond systems for criminal
detainees routinely accept the posting of deposit or property bonds, rather than
insisting on payment of a full cash bond, and often rely on fully unsecured bonds.101
Indeed, recent research indicates that for criminal defendants, unsecured bonds are
just as effective at achieving public safety and court appearance as secured bonds.102
DHS should bring its bond practices in line with the criminal justice system and
permit individuals a more flexible range of secured and unsecured bonds, as
appropriate to ensure their appearance.

V. CONCLUSION
For decades, ICE and its predecessor agency have relied on outsourced detention—
both county jails and private prisons—to fuel the expansion of the immigration
detention system. As a result, ICE has never been able to ensure safe, humane
conditions of civil detention, and the private prison industry has developed an
alarmingly cozy relationship with the agency.
As described in this white paper, ICE needs to end its relationship with the private
prison industry by significantly reducing its reliance on detention. Specifically, ICE
should generally stop using detention as a means to ensure appearance for court
proceedings. Instead, as the Vera Institute of Justice recommended to ICE’s
predecessor agency more than fifteen years ago, detention should be reserved chiefly
for effectuating removal orders. This white paper identifies specific recommendations
See 18 U.S.C. § 3142(a)-(c).
8 U.S.C. § 1226(a), the general immigration detention statute, refers broadly to “security approved by .
. . the Attorney General” and does not require full cash bonds.
101 Deposit bonds permit a defendant to post a percentage, such as 10%, of the bond as security, and the
total bond amount becomes due only if he fails to appear. Property bonds allow the defendant to post
property as security, which would be forfeited if the person fails to appear.
102 See Pretrial Justice Institute, Unsecured Bonds: The As Effective and Most Efficient Pretrial Release
Option 3 (Oct. 2013),
http://www.pretrial.org/download/research/Unsecured%20Bonds,%20The%20As%20Effective%20and
%20Most%20Efficient%20Pretrial%20Release%20Option%20-%20Jones%202013.pdf.
99

100

30

AMERICAN CIVIL LIBERTIES UNION

for how to carry out that shift. The ACLU urges the HSAC Subcommittee and DHS to
adopt these recommendations.

31