RESEARCH Report | October 2011

BerkeleyLaw

UNIVERSITY OF CALIFORNIA

Chief Justice Earl Warren
Institute on Law and
Social Policy

Secure Communities by the Numbers:
An Analysis of Demographics and Due Process
By Aarti Kohli, Peter L. Markowitz and Lisa Chavez*

Berkeley Law Center for
Research and Administration
2850 Telegraph Avenue
Suite 500
Berkeley, CA 94705

Introduction

Like earlier programs such as the 287(g) pro-

The United States will deport a record num-

Phone: (510) 642-8568

ber of individuals this year, due in large part

Fax: (510) 643-7095

to rapidly expanding federal immigration pro-

www.warreninstitute.org

grams that rely on local law enforcement. The
numbers are sobering: annual deportations
have increased over 400% since 1996 and more

About the
Warren Institute
The Chief Justice Earl
Warren Institute on
Law and Social Policy
is a multidisciplinary,
collaborative venture
to produce research,
research-based policy
prescriptions and
curricular innovation
on the most challenging
civil rights, education,
criminal justice, family
and economic security,
immigration and
healthcare issues
facing California
and the Nation.

than a million people have been removed from
this country since the beginning of the Obama
administration.1 Almost 300,000 individuals
are currently in deportation proceedings but
have not yet been removed.2 The newest and
most controversial immigration enforcement
program partnering with local law enforcement is Secure Communities.
Secure Communities was introduced by the
Bush administration in March 2008 and piloted
in 14 jurisdictions beginning in October 2008.3
Under President Obama, the program has
expanded dramatically. As of the drafting of
this report, Secure Communities is active in
1,595 jurisdictions in 44 states and territories,
a 65% increase since the beginning of
this year.4 The Immigration and Customs
Enforcement (ICE) agency of the Department
of Homeland Security (DHS) has stated that
it plans to have the program active in all
jurisdictions in the United States by 2013.5

gram and the Criminal Alien Program (CAP),
Secure Communities mobilizes local law
enforcement agencies’ resources to enforce
federal civil immigration laws.6 Whereas earlier programs such as 287(g) trained law
enforcement agents to assist with immigration enforcement, Secure Communities relies
heavily on almost instantaneous electronic data
sharing.7 This data sharing has transformed
the landscape of immigration enforcement by
allowing ICE to effectively run federal immigration checks on every individual booked
into a local county jail, usually while still in
pre-trial custody.
It has long been the case that local law
enforcement agencies electronically share
fingerprint data from the people they arrest
with the Federal Bureau of Investigation (FBI).
If that data comes from a Secure Communities
jurisdiction, however, the FBI now forwards
the fingerprints to the DHS.8 DHS checks the
fingerprints against the Automated Biometric
Identification System, also known as IDENT,
a fingerprint repository containing information on over 91 million individuals, including
travelers, applicants for immigration benefits,
and immigrants who have previously violated
immigration laws.9 When a match is detected,

* Aarti Kohli is Director of Immigration Policy at the Warren Institute at the University of California, Berkeley
Law. Peter L. Markowitz is Associate Clinical Professor of Law at the Benjamin N. Cardozo School of Law where
he directs the Kathryn O. Greenberg Immigration Justice Clinic. Lisa Chavez is a Senior Research Associate at
the Warren Institute.
Secure Communites by the Numbers

|

O c t o b e r 2 0 11

1

ICE reportedly examines its records to determine whether

Key findings include:

the person is deportable. If ICE believes an individual may
be deportable, or if ICE wishes to further investigate an

•	 Approximately 3,600 United States citizens have

individual’s immigration status, then ICE issues a detainer.

been arrested by ICE through the Secure Communities

The detainer is a request to the local police to notify immi-

program;

gration authorities when the individual is going to be

•	 More than one-third (39%) of individuals arrested

released from criminal custody and to hold the individual

through Secure Communities report that they have

for up to two days for transfer to ICE.

a U.S. citizen spouse or child, meaning that approxi-

10

Despite the scrutiny that the program has generated in the
public sphere,11 the federal government has conducted limited systematic analysis of its own data on individuals who

have been impacted by Secure Communities;
•	 Latinos

comprise

93%

of

individuals

arrested

are arrested under Secure Communities. To address this

through Secure Communities though they only

gap in knowledge, the Chief Justice Earl Warren Institute

comprise 77% of the undocumented population

on Law and Social Policy at UC Berkeley School of Law has

in the United States;

12

undertaken a comprehensive study of data provided by the

•	 Only 52% of individuals arrested through Secure

federal government to the National Day Labor Organizing

Communities are slated to have a hearing before

Network (NDLON), the Center for Constitutional Rights,

an immigration judge;

and the Kathryn O. Greenberg Immigration Justice Clinic
at the Benjamin N. Cardozo School of Law pursuant to a
partial settlement in NDLON v. ICE.13
This initial report is the first in a series based on that data.

•	 Only 24% of individuals arrested through Secure
Communities and who had immigration hearings had
an attorney compared to 40% of all immigration court
respondents who have counsel;

In this report, we attempt to better understand the profile

•	 Only 2% of non-citizens arrested through Secure

of individuals who have been apprehended through Secure

Communities are granted relief from deportation

Communities and the process they have encountered

by an immigration judge as compared to 14% of all

as they are funneled through the system.14 Overall, the

immigration court respondents who are granted relief;

findings point to a system in which individuals are

•	 A large majority (83%) of people arrested through

pushed through rapidly, without appropriate checks or
opportunities to challenge their detention and/or deportation. This conclusion is particularly concerning given

Secure Communities is placed in ICE detention as
compared with an overall DHS immigration detention rate of 62%, and ICE does not appear to be

that the findings also reveal that people are being appre-

exercising discretion based on its own prioritiza-

hended who should never have been placed in immigration

tion system when deciding whether or not to detain

custody, and that certain groups are over-represented in

an individual.

our sample population.

2

mately 88,000 families with U.S. citizen members

O c t o b e r 2 0 11

|

Secure Communites by the Numbers

The Troubled History of Secure Communities
The Secure Communities program was designed by DHS,
together with the FBI, and touted as a tool to improve
public safety by deporting dangerous “criminal aliens.”15
ICE first introduced Secure Communities on a countyby-county basis, beginning with Harris County, Texas,16
and suggested that participation would be voluntary for
localities, as CAP and 287(g) had always been.17 However,
after several counties expressed a desire not to participate
in Secure Communities, the federal government reversed
its position. DHS announced that since memorandums of
agreement (MOAs) were signed at the state-level,18 once
a state had signed an MOA, then DHS maintained that
localities in that state would not be able to “opt-out” of
Secure Communities.19
The role of states in Secure Communities was tested
recently, when Illinois, Massachusetts, and New York
sought to withdraw from the program or declined to sign
MOAs authorizing initiation of the program in 2011.20
In response, DHS declared that participation is compulsory for all jurisdictions nationwide; states and localities
do not have the option to opt-out.21 Some counties
have now begun using other mechanisms to limit ICE’s
jail based enforcement programs, such as refusing to
hold certain individuals on ICE detainers for transfer into
immigration detention.22

officials and advocates have critiqued it on a number of
fronts.24 First, the program is criticized for its spillover
effects on local and community policing. ICE has maintained that Secure Communities does not impact local
policing because it is merely a data-sharing program
related to routine fingerprint checks.25 However, there is
a strong perception in immigrant communities that local
police are acting as ICE agents.26 Advocates maintain
that this perception results in victims and witnesses not
coming forward to police in fear of deportation.27 In addition, community and advocacy groups have asserted that
Secure Communities is creating an incentive for some
local law enforcement agencies to engage in racial profiling through the targeting of Latinos for minor violations
or pretextual arrests.28 Demographic data in our research
provides some support for these assertions but further
research needs to be conducted on this issue.
A second major critique is that the program has not
stayed true to its stated goal of removing only those serious offenders who pose a threat to public safety. Instead
it has led to the mass deportation of low-level offenders,
such as people who violate traffic laws and people without
criminal histories at all.29 Secure Communities was introduced administratively, rather than through Congressional
mandate, although Congress did comply with the
administration’s request for funding by appropriating $200 million in each of FY 2009 and 2010 to fund

Following the opt-out requests by states and mounting

Secure Communities.30 The Secure Communities program

criticism from advocates and government officials, the

funding was part of a larger appropriation for identify-

federal government established a Task Force on Secure

ing and removing non-citizens convicted of crimes.31 Yet,

Communities as a subcommittee of the Homeland Security

according to ICE’s own figures, well over half of those

Advisory Council (HSAC). The task force included state

deported through Secure Communities had either no

and local government officials, as well as representatives

criminal convictions or had been convicted only of very

of law enforcement, the private sector and academia, and

minor offenses, including traffic offenses.32 This picture

was charged with identifying ways to reduce the impact

has remained relatively constant in the year since this

of Secure Communities on local community policing.

initial data became available. An examination of the offenses

The task force released a draft report in September 2011,

leading to a deportation through Secure Communities

which recommended more transparency, accountability

will be contained in our next report.

and a stronger focus on individuals who pose a threat to
public safety.23 Some task force members resigned because
they did not feel the recommendations went far enough,
instead, they urged that the program should be ended.

A final set of concerns which has received somewhat less
attention relate to a lack of due process for the individuals identified for removal by Secure Communities. These
concerns reflect the challenges for all immigrants facing

The HSAC report acknowledged that while ICE has

deportation since the passage of the Illegal Immigrant

hailed Secure Communities as the new face of immi-

Reform and Immigrant Responsibility Act (IIRIRA) and

gration enforcement, policy-makers, law enforcement

Anti-Terrorism and Effective Death Penalty Act (AEDPA)

Secure Communites by the Numbers

|

O c t o b e r 2 0 11

3

custody.

One of the most disturbing findings

Although the level of detail varies, this form

describes the current charge and the criminal and immigration history of the person. The I-213s are not the main focus

in our research is that 1.6% of cases

of this research, they will instead be the subject of our next

we analyzed were U.S. citizens... We find

report examining the criminal and immigration history of

that approximately 3,600 U.S. citizens
have been apprehended by ICE.

those individuals processed through Secure Communities.

Who is Affected by Secure Communities?
While some general characteristics of deportees in the
United States are known, such as their countries of

in 1996. Together these pieces of legislation created a

origin,

much harsher framework by broadly expanding the types

population. As Secure Communities expands, it is impor-

of crime that can lead to deportation and by establish-

tant to understand information such as age, gender,

ing a mandatory detention system for certain criminal

and family characteristics to provide a fuller picture of

offenses.33 IIRIRA also instituted certain administrative

who is being placed in deportation proceedings. Family

removal mechanisms that allow immigration officials to

characteristics in particular are important to gauge

summarily remove noncitizens without a hearing before

whether children and/or spouses39 are being impacted

an immigration judge.34 Further, IIRIRA constrained the

by deportations.

ability of immigration judges to grant relief even in com-

U.S. Citizens Apprehended

pelling deportation cases.35 Advocates and immigration

On its Secure Communities website, ICE acknowledges

lawyers have argued that individuals who enter deportation

that there might be IDENT matches, or hits, for U.S. citi-

proceedings through Secure Communities and similar

zens for a number of reasons, including that naturalization

programs experience even more limited due process rights

data has not been updated in its databases.40 ICE has never

and protections than the already very limited protections

published any data indicating the number or percentage

in place since IIRIRA.36 These claims are substantiated by

of citizens who have been apprehended through Secure

the data in this report.

Communities. If Secure Communities were working

Data and Analysis

remains

unknown

regarding

this

properly, U.S. citizen hits should never result in the apprehension of such individuals for deportation because U.S.

The findings in this report are based on a random national

citizens cannot be deported. One of the most disturbing

sample of 375 individuals who were identified as “IDENT-

findings in our research is that 1.6% of cases we analyzed

Matches” by the Secure Communities Program and were

were U.S. citizens and all were apprehended by ICE.41 If we

apprehended37 by ICE after October 1, 2008, the pro-

extrapolate that number to the 226,694 cumulative admin-

gram’s formal start date. As noted above, this sample was

istrative arrests and/or bookings into ICE custody from

obtained pursuant to a partial settlement of a Freedom of

Secure Communities’ inception, then we find that approxi-

Information Act lawsuit against the federal government

mately 3,600 US citizens have been apprehended by ICE

by NDLON, the Center for Constitutional Rights, and the

from the inception of the program through April 2011.42

Kathryn O. Greenberg Immigration Justice Clinic at the

What do these apprehensions indicate? Our dataset con-

Benjamin N. Cardozo School of Law. The ICE data consists
of demographic information, detentions and enforcement
actions. We also received data on immigration court proceedings from the Executive Office of Immigration Review
(EOIR) regarding bond and court proceedings.38 Pursuant
to the stipulation, ICE provided the narrative portion of the
I-213, a form filled out by the ICE official who administratively arrests and questions the individual in immigration

4

much

O c t o b e r 2 0 11

|

Secure Communites by the Numbers

tains limited information, but we know that the U.S.
citizens were not officially booked into an ICE detention
facility, but were arrested, held in custody for some period
and presumably subject to questioning regarding their
immigration status. No data is available on their length of
time in ICE custody. One of the U.S. citizens in our dataset appears to have been arrested on a criminal not on an
immigration charge, which indicates that all U.S. citizen

apprehensions are not necessarily unlawful. However,

Demographic Data

the best available data on the remaining five U.S. citizens

There is no definitive dataset on the characteristics of

apprehended in our dataset suggests that they were wrong-

individuals who are placed in deportation proceedings in

fully apprehended for civil deportation.

the United States. To see whether any patterns emerge in

Recent litigation on wrongful deportations and detentions

this report, we compared the age, gender and country of

indicates that the consequences for U.S. citizens appre-

origin of people in our sample population with existing

hended by ICE may be severe. If Secure Communities is

datasets on noncitizens and on unauthorized immigrants.48

to continue, at the very least ICE must improve its record-

Age & Gender49

keeping to indicate the circumstances under which U.S.

Previous research indicates that 43% of the population of

citizens are apprehended by ICE and to indicate if and

undocumented residents in the U.S. are women and 57%

for how long, the person is held for questioning by an

are men.50 By contrast, 93% of individuals in our Secure

ICE official.

Communities sample were categorized as males. The

U.S. Citizen Family Members

median age of individuals in our data is 29-years-old with

Notably 39% of the people identified for deportation by

11% age 21 or younger. Since our data does not contain

43

ICE in our study reported having a U.S. citizen family member. Thirty-seven percent reported a U.S. citizen child
and 5% reported a U.S. citizen spouse. This may reflect
an undercount as immigrants may fear disclosing personal
information to immigration authorities, particularly if
they live in mixed-status families and fear negative consequences for family members.
As the number of individuals who are placed in deportation proceedings grows, so does the impact on their
families and communities. Demographers have noted that
over 4 million children have undocumented parents in the
United States.44 A recent research report in the Harvard
Educational Review highlights the negative impacts of
the fear of deportation on both U.S. citizen children
and undocumented children of unauthorized parents.
Researchers note, “[t]he implications of growing up in
an unauthorized family span a variety of developmental
contexts . . . including psychological well-being, mental
health, physical health, education, and employment.”45
These findings are particularly worrisome as the number

information on length of residence in the United States,
it is difficult to ascertain whether these individuals are
recent migrants or were brought to the U.S. as children.
The federal government’s data indicates that 61% of the
current unauthorized population entered before the year
2000.51 If that percentage is applied to our population then
a significant portion can be assumed to have entered the
U.S. as children. These figures indicate that individuals
processed through Secure Communities are, on average,
younger than the general noncitizen population, which
has a median age of 40.52
Since Secure Communities is intended to target criminal aliens, these data might be interpreted to reinforce
the notion that men are more likely to commit crime than
women. While that is true generally, FBI data indicates
that females represented 25% of arrests nationwide in 2009.53
In contrast, the share of females in our sample population
is 7%. Without more information, it is difficult to reach
conclusions as to the significance of the over-representation of males who are being processed through

of deportations increase. No data exists as to what per-

Secure Communities.

centage of children and spouses leave the United States

Ethnicity

along with their deported family members but anecdotal

There are different points of comparison for national-

evidence points to increasing family separations.

ity data. Data on all foreign-born persons54 in the United

46

Extrapolating out to the total cumulative administrative
arrests and/or bookings into ICE custody since the inception of Secure Communities we find that approximately
88,000 families with US citizen members were affected
by Secure Communities from its inception through April
2011.47

States indicate that 53% are from Latin America, 28%
from Asia and 13% from Europe.55 The unauthorized
population has a different composition, with 77% from
Latin America, 13% from Asia and 6% from Europe and
Canada.56 By either measure, however, Latinos57 are disproportionately impacted by Secure Communities. The data
indicate that 93% of the people identified for deportation

Secure Communites by the Numbers

|

O c t o b e r 2 0 11

5

charged with removal based on a criminal conviction while
others were charged with removal for a civil immigration

Over 8 in 10 (83%) of the individuals

violation or were issued no charge at all.62 Moreover, previous research studies have demonstrated that noncitizens

in our sample were booked into an

are less likely, not more, to engage in criminal behavior.

ICE detention facility as compared to

representation in state and local jails and prisons thereby

Research has shown that immigrants have lower rates of
questioning the perception of immigrant criminality.63

a 6 in 10 (62%) detention rate for all

Our next report will examine more closely the criminal his-

DHS immigration apprehensions.

tory of the individuals in our Secure Communities sample
but these initial data raise further questions as to why young
Latino men are disproportionately represented in Secure
Communities enforcement actions.

through Secure Communities are from Latin American
countries, while 2% are from Asia and 1% are from Europe
and Canada.58 The overwhelmingly large percentage of
Latinos among those identified for deportation by Secure

The Secure Communities Program technically should not

Communities raises serious questions.

affect the due process rights and protections for individu-

ICE has consistently maintained that Secure Communities
does not impact local policing because it is merely a fingerprint check and that it actually protects against racial
profiling because the fingerprint checks are run on all persons arrested or booked into local facilities. However, as
the Homeland Security Advisory Council found, there is a
strong perception in immigrant communities that the local
police are acting as ICE agents.59 Community and advocacy
groups have also asserted that Secure Communities is, in
some jurisdictions, masking local law enforcement agen-

als in immigration proceedings, given that it is supposed
to be a simple data sharing mechanism. Nevertheless, the
rollout of the program has resulted in dramatic increases
in the number of people entering deportation proceedings,64 and little is known about what happens to these
individuals once they are in ICE custody. In the following section, we examine four key issues: the availability
of hearings before an immigration judge, detention
pending removal, legal representation during the process,
and immigration arrest outcomes.65

cies’ practice of racial profiling.60 These jurisdictions are

Hearings Before an Immigration Judge

criticized for targeting Latinos for minor violations and

Given that the “severe penalty”66 of deportation is at issue

pre-textual arrests with the actual goal of initiating immigration checks through the Secure Communities system.61

6

Due Process in Secure Communities
Proceedings

for everyone who is apprehended and detained because
of Secure Communities we might assume that individu-

As outlined above, the data in our sample indicate that

als processed through Secure Communities have access

noncitizens who are arrested and brought to a local jail,

to an immigration judge. Unfortunately, not every type of

and thus subject to Secure Communities, fit a particular

removal process entitles the person to appear before an

profile, namely a young Latino male. Some might assert

immigration judge. These categories include administrative

that since Secure Communities targets criminal aliens, this

and expedited removals. Through administrative remov-

profile makes sense because they presume that this popu-

als, ICE officers can deport individuals with an aggravated

lation (young immigrant Latino males) is more likely to

felony conviction who are not lawful permanent residents

engage in criminal behavior than the average U.S. citizen.

or conditional permanent residents when removal pro-

This assertion is, however, not supported by the data. When

ceedings begin. Expedited removals involve the removal of

ICE begins deportation proceedings against an individual,

certain individuals who have not been formally admitted

it issues a removal charge indicating the reason for depor-

to the United States or who arrive in the United States.67

tation. As explained further below, only approximately

In addition, ICE can simply have orders of removal rein-

one quarter (27%) of the individuals in our sample were

stated without a further hearing for individuals who have

O c t o b e r 2 0 11

|

Secure Communites by the Numbers

Detention Pending Removal

	 Figure 1 | 	Type of Proceeding Initiated

Whether or not they are deported, immigrants identified
through the Secure Communities program face potential

Administrative
Removal
4%
Voluntary
Return
11%

due process infringements arising from their detention

Expedited
Removal
1%

while they await the disposition of their cases. Whether a person is in or out of custody during their removal proceedings
has a tremendous impact on their ability to obtain counsel and on their ultimate chances of prevailing and being
allowed to remain in the United States.72 Over 8 in 10 (83%)

Unknown
6%

of the individuals in our sample were booked into an ICE
detention facility as compared to a 6 in 10 (62%) detention
rate for all DHS immigration apprehensions.73 Detention
for immigrants facing possible deportation is, by DHS’
own assessment, the equivalent of criminal incarceration.
Hearing
52%

Reinstatement
26%

Notably unlike criminal defendants, however, immigration
detainees are not afforded the basic procedural protections
that come with criminal proceedings. Immigration detainees are not provided with attorneys; large categories are not

previously been ordered removed—even if such previous

afforded the right to bond due to harsh mandatory deten-

orders were entered in absentia. In some limited circum-

tion laws; and are not guaranteed a trial in the venue where

stances such as a claim of fear of persecution upon return,

they were arrested, but rather are routinely transferred

these categories of persons may be eligible for a hearing in

thousands of miles away to remote detention facilities in far

front of an immigration judge.

off jurisdictions. In a 2009 audit of the immigration deten-

68

The data reveals that slightly more than half (52%) of people identified through the Secure Communities program
had the opportunity to appear before an immigration judge
following their Secure Communities apprehension (see
Figure 1).69 One percent of our population was returned
through expedited removal, and only 4% through administrative removal. Eleven percent accepted voluntary return,
a discretionary grant by ICE officials that allows an individual to leave without a formal order of removal. The last
26% had their removal order reinstated and thus were only
entitled to a hearing if they claimed a fear of persecution or
torture if returned to their country of origin. For a significant number of people in the sample (6%), the data does
not indicate the type of removal process initiated. Notably
our sample does not indicate whether anyone was subject to
stipulated removal—an increasingly common process where
individuals agree to be deported and waive their right to a
hearing. Recent research suggests stipulated removals may
70

be counted in ICE data as hearings in immigration court
because a judge signs the removal order.71 Thus, we may be
overestimating the share of our population who receive a
hearing before an immigration judge.

tion system, Dr. Schriro, former DHS Special Advisor on
ICE Detention and Removal, recommended that detention
should be used only when necessary and that ICE consider
implementing a risk-based approach to curb punitive detention since immigration is supposed to be a civil system.
Although the current administration has made a public
commitment to reform the immigration detention system,
progress has been slow and widely criticized.74

Length of Detention We found that individuals in our sample population spent an average of 28 days in detention and
28% spent more than one month in detention. One person
spent over 500 days in detention. Examining those in our
sample with ICE detention book-out dates and departure
dates,75 we found that only 10% were released prior to their
departure. A substantial share of those who left the country,
90%, were in detention until the date of their departure
(see Figure 2).76 These data indicate that the vast majority
of those who are detained and subsequently removed do
not have the opportunity to return to their homes to gather
their belongings, get their affairs in order or say good-bye
to family members once they enter detention.

Secure Communites by the Numbers

|

O c t o b e r 2 0 11

7

The critical questions are who gets bond hearings, who

	 Figure 2 | 	ICE Detention Release Date Compared
		
with Departure Date

gets bond, and who gets out? Partially as a result of the
policy and judicial determinations described above, and
partially as a result of the discretion (or lack of discretion)

Released from
ICE Detention
Before Departure
10%

exercised by ICE, our dataset revealed that only 2% of individuals booked into detention under Secure Communities
were given bond by ICE, and only 6% got bond redetermination hearings before immigration judges. Our data
does not reflect whether individuals were able to post bond

Released from
ICE Detention
Same Day as
Departure
90%

in order to be released from custody. The average initial
bond amount set by ICE was $7,000 in our sample population and the average amount after a redetermination
hearing was $5,000, which is slightly below the national
average of $5,941.80 From these data, we conclude that ICE

Bond Individuals in the standard criminal process are
entitled to a bail hearing which normally focuses on an

opportunity to be released on bond.

individualized determination of whether their detention is

Discretion in Detention As mentioned above, a significant

necessary to secure their presence at trial (flight risk) or to

percentage of our population (83%) was booked into an

prevent them from doing some harm (danger to the com-

ICE detention facility. In order to determine whether ICE

munity).

In the immigration system, bonds provide the

was exercising discretion in detention determinations we

same function as bail but not all immigrants receive a bond

analyzed detention rates by both removal charges and by

hearing to determine if they can be released from an immi-

ICE’s own Secure Communities prioritization categories.

gration jail. Congress has deprived some immigrants of

In the first analysis we looked at detention rates by ICE

the right to such an individualized determination and the

removal charges, which indicate the reason ICE is seeking

Supreme Court has upheld Congress’s determination.78

to deport an individual. The Immigration and Nationality

Generally, when ICE apprehends an individual for poten-

Act (INA) contains many grounds for removal. We have

tial deportation, an ICE deportation officer makes an initial

placed the charges in our data in the following categories

determination about the individual’s detention. The officer

according to INA charges: Aggravated Felony,81 Other

may make one of four decisions: release on own recogni-

Criminal Grounds, Other Immigration Grounds (indi-

zance, release subject to supervision, release on bond, or

cating various civil immigration violations, including for

detain. If the officer determines that the person can be

example overstaying one’s visa), and Present Without

released on bond, ICE can either exercise its discretion

Admission (indicating someone who entered the country

and set the bond amount or decide not to set a bond. If

without authorization). Approximately 93% of cases we

an individual is unhappy with ICE’s custody determination,

analyzed were issued at least one removal charge and the

she can, in some instances, ask an immigration judge to

distribution of type of removal charges are shown in Figure

hold a custody redetermination hearing. However, as noted

3.82 Nearly half (45%) of the cases we analyzed were solely

above, in certain cases immigration judges are deprived of

charged with being Present without Admission (PWA)--a

jurisdiction to set bond. For example, immigration judges

charge that does not indicate any criminal history. Only

are completely deprived of jurisdiction to review ICE’s

8% of individuals were charged with being removable fol-

custody determination for many people with even minor

lowing an aggravated felony conviction and 7% were not

criminal convictions and people deemed “arriving aliens.”

79

issued a removal charge at all.83 We found high rates of

This means that immigration judges—presumably neutral

detention for all categories of removal charges. Individuals

fact-finders—cannot balance factors in favor of releasing a

whose removal charges were solely immigration-related or

person from custody, including serious medical conditions,

PWA have similar detention rates to people who are likely

a child’s dependence on the person, or community ties.

subject to mandatory detention based on their Aggravated

77

8

is providing a small share of people in its custody with the

O c t o b e r 2 0 11

|

Secure Communites by the Numbers

two or more other felonies;86 Level 2 includes individuals

	 Figure 3 | 	Removal Charge Issued

convicted of one felony or three or more misdemeanors;
No Removal
Charged Issued
7%

Aggravated
Felony
8%

and Level 3 consists of individuals convicted solely of misdemeanors, including minor traffic offense.87
ICE records the offense levels of individuals placed in
deportation proceedings due to Secure Communities.
These data are reported in the interoperability reports ICE
updates on a regular basis and appear to be one method
the agency uses to measure its success in deportations of

Other Criminal
Grounds
19%

criminal aliens.88 Although the accuracy of the offense
levels is questionable and will be examined further in our
next report, for the purposes of this analysis, we analyze
Present
Without
Admission
(PWA)
45%

Other Immigration
Grounds
21%

detention rates based upon ICE’s own classification system.
We found that 63% of individuals in our sample population were issued at least one Secure Communities offense
level and the distribution of Secure Communities Levels
are shown in Figure 5. Over one-third of our population
(37%) was not issued a Secure Communities offense level,

	 Figure 4 | 	ICE Facility Detention Rates
		
by Removal Charge Issued

suggesting that those individuals were the non-criminals
who are being ensnared through Secure Communities.
Approximately a quarter (28%) of those in our sample

100%

were issued the highest level (Level 1). Figure 6 also shows
75%
50%

detention rates by Secure Communities offense level;
82%

80%

86%

87%

62%

25%
0%

again, there is little difference among those who vary by
Secure Communities offense level. Even for those individuals not assigned an offense level, the large majority were

Aggravated
Felony

Other
Criminal
Grounds

Other
Immigration
Grounds

Present
Without
Admissions
(WPA)

No Removal
Charge
Issued

detained. Accordingly, the data suggests that the government’s determinations about whether or not to detain
individuals are arbitrary and do not follow the risk-based
recommendations of Dr. Schriro.89 Moreover, it suggests

Felony charges or Other Criminal Ground (see Figure 4).
Even for people with no removal charge, 62% were booked

ICE is not appropriately using its discretion when non-citizens are placed in detention.

into an ICE detention facility.84
In our second analysis, we examined detention by the
Secure Communities offense level assigned by ICE. When

Even for those individuals not

Secure Communities began in 2008, it contained a con-

assigned an offense level, the

troversial three-tier system to determine threat levels of
various “criminal aliens” based on whether an individual

large majority were detained.

had been convicted of or charged with particular crimes.85

Accordingly, the data suggests that

Level 1 crimes were ostensibly the most violent, dangerous crimes, although they did include some nonviolent

the government’s determinations

misdemeanor offenses. A memo issued in March 2010 by

about whether or not to detain

ICE Assistant Secretary John Morton changed this definition of Level 1 offenses to refer to individuals convicted of

individuals are arbitrar y.

“aggravated felonies” under § 101(a) (43) of the INA, or
Secure Communites by the Numbers

|

O c t o b e r 2 0 11

9

	 Figure 5 | 	Secure Communities Offense Level Issued

transferred two or more times (see Table 1). One person
was transferred 10 times in less than five months. Of those
transferred, 28% were sent out-of-state. We also found that

NO SC Offense
Level Issued
37%

Level 1
28%

the number of days individuals spent in detention correlated with the number of times they were transferred (see
Figure 7). These data indicate that individuals who wait
an average of three months for an immigration court
hearing91 are more likely to be moved away from their place
of initial encounter. Therefore, they are more likely to be

Level 2
12%

away from family and friends who could potentially help
them get representation.

Access to Legal Advice and Representation
The practice of immigration law has become a highly speLevel 3
23%

cialized area because of the complexity of the laws and
regulations. Despite the high stakes, the government does
not provide counsel to individuals appearing before an

	 Figure 6 | ICE Facility Detention Rates by Secure
		
Communities Offense Level Issued

immigration judge because deportation is not considered
to be a criminal proceeding. Thus, only those individuals
who can pay for an attorney or who find a pro bono attor-

100%

ney have legal assistance during their proceedings. Even
then, however, non-citizens in deportation proceedings face

75%
50%

many barriers to accessing counsel, including locating coun86%

88%

87%

77%

25%
0%

sel while in detention, collecting the needed paperwork for
their case, and staying in contact with their attorney while
they are in custody and being moved by ICE.92 Scholars
and advocates have raised concerns that as the numbers of

SC Level 1

SC Level 2

SC Level 3

No SC Offense
Level Issued

immigrants in deportation proceedings increase because of
programs like 287(g), the Criminal Alien Program (CAP),
and Secure Communities, fewer people will have access to a

10

Transfers We analyzed the number of times individuals

lawyer.93 These concerns are borne out by our analysis.

were transferred during their detention period. The num-

Nearly half (46%) the people in our sample had an

ber of times a person is transferred and the distance of

immigration court proceeding. The Executive Office of

their transfer is crucial for individuals having access to

Immigration Review (EOIR) indicates in its annual report

legal representation and to maintaining contact with fam-

that approximately 41% of individuals in immigration

ilies. In a review of all transfers between 1999 and 2008

court from 2008 through 2010 had an attorney.94 The data

and interviews with immigrants and their families, Human

reveals that individuals who are processed through Secure

Rights Watch found that transfers were increasing and that

Communities, however, have far lower rates of represen-

“the impact on detainees and their families is profound.”

tation (see Figure 8).95 Among those individuals in our

Out of state transfers are particularly onerous personally

sample with an EOIR proceeding in front of an immigra-

and legally; moving a person to a different state can lead to

tion judge just 24% of those individuals were represented

changes in the way the law is applied to their case, and “can

by a lawyer unlike 41% of those in immigration proceed-

ultimately lead to wrongful deportations.”90

ings nationwide.

Analysis of our data found that nearly half (47%) of

The number of individuals who were granted relief or

individuals placed in deportation proceedings were trans-

who appealed their decision to the Board of Immigration

ferred to a different ICE facility at least once; 22% were

Appeals is small (8 and 5, respectively). As a result, we are

O c t o b e r 2 0 11

|

Secure Communites by the Numbers

	 TABLE 1 | Patterns of Detention Among Individuals 	
	
Booked into an ICE Facility
Number of Days
in Detention

Number of Times
Transferred to
a Different Detention
Facility

In-Out
Same Day

27%

Never

53%

1-30 Days

45%

Once

25%

31+ Days

28%

Two or More

22%

TOTAL

100%

TOTAL

100%

	 Figure 7 | Average Number of Days in ICE Detention
		
by Number of Times Transferred to
		
a Different Facility
67 days

32 days

28 days

10 days

All ICE
Detentions

Never
Transferred

Transferred
1 Time

The Value of Legal Representation –
Insight from ICE narratives
Very few of the ICE I-213 narratives mention legal
representation, but those that do paint a picture of
individuals who are much more empowered in their
decision-making when they have an attorney.
SC-LA-1867 was a Mexican man arrested for driving
under the influence in San Bernardino County in
August 2009. He had no criminal history and no immigration history in the United States, and when he was handed
to ICE in September 2009, they offered him a voluntary return. The respondent was one of the few to have a lawyer,
and after obtaining legal advice over the phone, made a
conscious decision to stay in custody and await a hearing
before an immigration judge.
SC-4888 was a woman from an unnamed country who
was arrested for handling of a controlled substance
in August 2008 and was identified as a deportable noncitizen by a 287(g) officer (local police officer trained by
ICE). The charges against the respondent were dropped
and she had no prior criminal or immigration history, but she was nevertheless immediately turned over
to ICE instead of being released. The respondent was
able to contact her lawyer, and on her lawyer’s advice
refused to sign any paperwork.

Transferred
2 or More Times

unable to draw statistically significant conclusions about
	 Figure 8 | Legal Represention Status
		
in Immigration Court

the legal representation status of these two small groups
as subsamples. However, it is noteworthy that of the small
share of people who were granted relief from deportation,
nearly two-thirds (63%) had an attorney of record and
5 out of 5 of those who filed an appeal had legal representation.96 This suggests that having access to counsel makes

59%

76%

a difference.

Immigration Arrest Outcomes
Even those noncitizens entitled to have a hearing may

41%
24%
All Completed EOIR Proceedings
(2008-2010)

Did Not Have Legal
Representation

Secure Communities Apprehensions
with EOIR Proceedings

Had Legal Representation

not always appear before a judge because of the enormous backlog in the system. Currently, approximately
285,000 cases are waiting to be heard in the immigration
courts.97 Recent research suggests that as a result of lengthy
wait times, more immigrant detainees give up valid claims
for relief or agree to waive their rights and sign stipulated
orders of removal, presumably to get out of custody sooner.98
Although the dataset for this report did not record when
there was a stipulated order of removal, we do know that

Secure Communites by the Numbers

|

O c t o b e r 2 0 11

11

A meager 2% of individuals in our

	 Figure 9 | 	Outcomes Among All Individuals
		
Arrested Under Secure Communities

dataset were granted relief from

Outcome is
Unknown
20%

deportation, while the majority
were removed or ordered removed.
Yet, nationwide 14% of proceedings

EOIR Case was
Terminated
0.3%

Was Removed
or Was Issued
Removal Order
63%

in immigration courts resulted in
grants of relief.

Granted
Relief
2%

individuals detained through Secure Communities wait an
average of 88 days, nearly three months, for an immigration judge decision.

99

Although immigration judges hear

cases of those detained faster than those out of custody
three months can be a long time, particularly for those
who have never been in jail before. For those who are not
detained in our sample population, the average wait time
for a decision from an immigration judge is 208 days.
Finally, even if an individual does appear before a judge,
their chances of relief may be slim. IIRIRA eliminated the
212(c) waiver, which allowed immigration judges to grant
relief from deportation for lawful permanent residents
convicted of aggravated felonies.100 Immigration judges
can no longer take mitigating factors, including rehabilitation, length of residence and the impact of deportation
on lawfully residing family members, into consideration in
removal proceedings for aggravated felonies; these individuals are subject to mandatory deportation.

101

IIRIRA also

limited avenues for relief for long-term undocumented
residents with family ties in the U.S. Prior to IIRIRA, judges
could suspend the deportation of undocumented immigrants who had lived in the U.S. for 7 consecutive years

12

O c t o b e r 2 0 11

|

Secure Communites by the Numbers

Voluntary
Return/
Departure
15%

and could show extreme hardship to a US citizen or LPR
family member. IIRIRA narrowed the criteria to require
10 years of residence and a showing of ‘exceptional and
extremely unusual hardship’ in order for an immigrant to
receive what is now called ‘cancellation of removal’.102
A meager 2% of individuals in our dataset were granted
relief from deportation, while the majority were removed
or ordered removed (see Figure 9). Yet, nationwide 14%
of proceedings in immigration courts resulted in grants
of relief.103 EOIR reports that 75% of those with immigration proceedings in 2010 resulted in a removal. Based on
EOIR’s data, it is likely that a significant percentage of our
currently unknown cases will also result in removal. These
might include cases that have not yet been adjudicated and
other cases where the results were not recorded.

Conclusion
Programs like Secure Communities are understudied,

These disturbing findings indicate that our system of jus-

largely because of their rapid implementation and expan-

tice is not working well for those who are being detained

sion and because the data has been kept, in large part,

and deported under Secure Communities. If Secure

confidential. Given that its expansion does not appear to

Communities is to continue, we recommend the following

be slowing down, it is all the more imperative that research

to address the concerns raised by this analysis:

on the impact of Secure Communities continues.
Our analysis provides a fuller picture of the population
being processed through Secure Communities and raises
serious concerns about the level of screening and potential targeting of certain social groups. In particular, we find
that US citizens are significantly impacted by the Secure
Communities program, both through their own apprehension, and through the impact on UC citizen family
members. We are also concerned that Latinos appear to

•	Implement improved safeguards to protect US
citizens from wrongful arrests and to provide further transparency about the impact of Secure
Communities on US citizen populations;
•	Conduct a thorough investigation into the potential racial profiling of Latinos as a result of Secure
Communities and implement safeguards to protect
against such abuses;

be disproportionately impacted by Secure Communities.

•	Overhaul the mechanism and guidelines by which

Further, we have noted that those who are identified

detention decisions are made by ICE to ensure that

through Secure Communities enter a complex legal pro-

only individuals who pose a significant flight risk or

cess, often while in detention, and rarely with legal advice

genuine danger to the community are detained;

about their rights and options. The adjudication process

•	 Provide access to government-appointed counsel,

for those processed through Secure Communities points

particularly for detained populations. In the alterna-

to minimal procedural and due process protections. Thus,

tive, improve access to pro-bono representation.

individuals who are not meant to be in the system, may

•	Improve transparency by providing regular public

have little opportunity to get out.

reporting on the topics covered in this report
•	Halt the expansion of Secure Communities until the
recommendations above can be implemented and
further research can be done to determine whether
due process can be adequately protected for individuals affected by Secure Communities.

Secure Communites by the Numbers

|

O c t o b e r 2 0 11

13

APPENDIX A
The findings in this report are based on a data obtained pur- Pursuant to the stipulation, ICE also provided the narrasuant to a partial settlement of a Freedom of Information tive portion of the I-213, a form generally filled out by the
Act lawsuit brought by the National Day Labor Organizing ICE official who administratively arrests a person to be
Network, the Center for Constitutional Rights, and the deported. Although the majority of key data fields were
Kathryn O. Greenberg Immigration Justice Clinic at the populated, we did not have complete data on every indiBenjamin N. Cardozo School of Law (collectively “the vidual in the sample.
plaintiffs”) against several federal agencies involved in A total of 502 individuals (out of the 1650) had data availadministering Secure Communities, most significantly the able in ICE databases (with the majority having an I-213
U.S Immigration Customs and Enforcement agency.
narrative) and 359 had EOIR data. Presumably the IDENT
Per the settlement, data fields were extracted from vari- matches for whom no ICE or EOIR records exist were not
ous federal government databases in two steps. First, an subject to deportation; they could, for example, be U.S. citinitial random sample of 1,650 individuals was drawn from izens, or lawful immigrants who have committed no crimes.
all IDENT matches (Secure Communities fingerprint que- Among the 502 individuals for whom data was available,
ries that resulted in “hits” in the Department of Homeland 127 were excluded from the analysis for three reasons:
Security’s databases) between October 1, 2008 and January 1) they had enforcement actions that pre-dated Secure
31, 2010. An IDENT match indicates previous contact with Communities, and/or 2) they were issued detainers but
DHS officials and does not necessarily indicate that the per- were not apprehended by ICE, and/or 3) we were unable
son is subject to deportation. IDENT matches may occur to draw reasonable conclusions about them given the forfrom any contact ranging from obtaining a tourist visa, to mat of data received. The ICE data fields were provided
becoming a naturalized U.S. citizen, to being previously at the individual level rather than the encounter level and
ordered deported. Random identification numbers were all data fields were provided in separate files that were to
assigned to the initial sample with identifying information be merged using individual identification codes. Given this
structure, we were unable to determine which data fields
redacted.
Next, information was extracted from two data sources for
each identification number where available: ICE databases
(ENFORCE, EID, IIDS, and GEMS) and the Executive
Office of Immigration Review (EOIR) Case Access System
(CASES). The ICE data consists of demographic data
fields such as age, gender, country of citizenship, relatives and their countries of citizenship, and marital status,
and data fields describing enforcement actions including
status at entry, detainer date, apprehension date, Secure
Communities offense level, Notice to Appear date, bookin and book-out of ICE detention facilities dates, facility
name(s), type of proceeding initiated, removal charge(s),
and date of departure from the U.S. The EOIR data fields
pertained to immigration court proceedings include:
information on custody redetermination, representation
status, date of court proceedings, immigration judge decision, and date of appeal to Board of Immigration Appeals.

14

O c t o b e r 2 0 11

|

Secure Communites by the Numbers

corresponded to which ICE apprehension event for individuals with multiple ICE apprehensions. As a result, the
analyses in this report are based on 375 individual cases
that had an ICE apprehension date that occurred after
October 1, 2008 when Secure Communities formally began.
The majority (over 72%) had apprehension dates in 2009
while 22% had these enforcement dates in 2010. A small
share took place in 2008 and 2011. The last apprehension
date was in April 2011. Of the 375 apprehended individuals in our sample, 46% had EOIR data with a post-Oct 1,
2008 proceeding and 91% had an I-213 narrative. We conducted one-sample t-tests on all sample-based percentages
discussed in this report and all are statistically significant
at the 95% confidence level unless specifically noted. We
confirmed that there were no differences in detention rates
by removal categories (Figure 4) and Secure Communities
offense levels (Figure 6) by performing Chi-Square tests of
significance.

Endnotes
1. Office of I mmigration Statistics, U.S. Dep’t of Homeland Sec .,

7. Although the centerpiece of Secure Communities is a data-

2010 Yearbook of I mmigration Statistics (2011), 15, 94 available at

sharing program, local officials and governments have been

http://www.dhs.gov/xlibrary/assets/statistics/yearbook/2010/

concerned that police are now being perceived as a “gateway”

ois_yb_2010.pdf. The administration has indicated that it will

to immigration enforcement. Homeland Sec . A dvisory C ouncil ,

deport over 400,000 people this year.

U.S. Dep’t
ties

2. Transactional Records Access Clearinghouse, Immigration,
Rising Immigration Backlog At All-time High Yet Criminal, National
Security, and Terrorism Cases Fall, (Sept. 14, 2011), available at
http://trac.syr.edu/immigration/reports/261/.
3. U.S. I mmigration & C ustoms Enforcement, U.S. Dep’t of Home land

Sec ., Activated Jurisdictions (2011), available at http://www.

ice.gov/doclib/secure-communities/pdf/sc-activated.pdf (statis-

of

F indings

Homeland Sec ., Task Force

and

on

Secure C ommuni -

R ecommendations 16 (2011), available at http://

www.dhs.gov/xlibrary/assets/hsac-task-force-on-secure-communities-findings-and-recommendations-report.pdf

(hereinafter

“HSAC Task Force Report”).
8. DHS is the umbrella agency which houses Immigration and
Customs Enforcement (ICE) and other immigration enforcement agencies.
9. U.S. Dep’t Of Homeland Security, DHS E xhibit 300

tics as of September 27, 2011).
4. Id. In January 2011, after over two years of implementation, only 969 jurisdictions were active. See also http://www.
newyorkimmigrationlawyersblog.com/2011/01/secure-commu-

P ublic R elease B y10/Nppd —US-Visit—Automated Biometric
I dentification S ystem (IDENT) (2010), http://www.exercise.dhs.
gov/xlibrary/assets/mgmt/e300-nppd-usvisit-ident2010.pdf.
10. U.S. I mmigr ation & C ustoms Enforcement, supra note 5. See in

nities-activation.html.
5. U.S. I mmigration & C ustoms Enforcement, U.S. Dep’t of Home -

particular “The Secure Communities Process”.

Sec ., Secure C ommunities F requently A sked Q uestions,

11. See, e.g., Julia Preston, States Resisting Program Central to Obama’s

http://www.ice.gov/secure_communities/ (last visited Oct. 12,

Immigration Strategy, N.Y. Times, May 6, 2011, at A18, available at

2011).

http://www.nytimes.com/2011/05/06/us/06immigration.html;

land

6. The 287(g) program and CAP are two other ICE administered programs that have the same goal of screening inmates
in local jails and state prisons for the purpose of identifying
deportable non-citizens. The 287(g) program, introduced in
1996, allows local police to be trained to act as immigration
enforcement officers in their jurisdictions. U.S. I mmigr ation &

and

24052748704681904576321404203166580.html; Editorial, Resistance Grows: Massachusetts, New York and Illinois Reject the Obama
Enforcement-Only Way, N.Y. Times, June 8, 2011, at A22, available
at http://www.nytimes.com/2011/06/08/opinion/08wed1.html.

Homeland Sec ., Fact Sheet :

12. Transactional Records Access Clearinghouse, Immigration,
Current ICE Removals of Noncitizens Exceed Numbers Under Bush

Nationality Act, http://www.ice.gov/news/library/

Administration (Aug. 2, 2010), http://trac.syr.edu/immigration/

of

of

factsheets/287g.htm (last visited Oct. 6, 2011). CAP, which has
existed in various forms since the mid-1980’s, involves police
interviews of new inmates. U.S. I mmigration & C ustoms Enforce ment,

2011, at A3, available at http://online.wsj.com/article/SB100014

I mmigration Authority Section 287(G) I mmigr a-

C ustoms Enforcement, U.S. Dep’t
Delegation
tion

Miriam Jordan, States Rebel Over Deportations, Wall St. J., May 14,

U.S. Dep’t

of

Homeland Sec ., C riminal A lien P rogr am,

http://www.ice.gov/criminal-alien-program/ (last visited Oct.
6), http://www.immigrationpolicy.org/sites/default/files/docs/
Criminal_Alien_Program_021710.pdf. Secure Communities is

reports/234/.
13. NDLON v. ICE, 10-cv-3488 Dkt. #84 (S.D.N.Y. July 12,
2011) (stipulation and order), available at http://ccrjustice.
org/files/4-8-11%20Sampling%20Stipulation.pdf.

For more

information about the case, visit http://www.ccrjustice.org/
secure-communities.

the third technology-based arm of this strategy.

Secure Communites by the Numbers

|

O c t o b e r 2 0 11

15

14. An “apprehension” in this context is an administrative arrest

22. Jen Sabella, Cook County Votes To Free Immigrant Inmates

of an individual believed to be present in the United States in

Wanted By Feds, Huffington Post C hicago, Sept. 9, 2011, http://

violation of immigration laws. See 8 U.S.C. § 287(a)(2), 8 C.F.R.

w w w.huffingtonpost.com/2011/09/10/cook-county-votes-to-

§ 287.8(c). We use apprehension and arrest interchangeably in

free_n_954158.html

this report.

23. HSAC Task Force Report, supra note 7, at 25-27.

15. Supra note 5.

24. Id. at 25-27.

16. Under President George W. Bush, the program was piloted
in 14 counties in 2008: Harris County, Texas (October 27), Suffolk County, Massachusetts (November 5), Wake County, North
Carolina and Dallas County, Texas (November 12), Buncombe,

25. Id. at 11.
26. Id. at 12.

Gaston and Henderson Counties, North Carolina (November

27. A merica’s Voice Educ . F und, P ublic Safety

17), Maverick and Val Verde Counties, Texas (December 9),

You Police A C ommunity That Won’t Talk To You ? (2011),

Bucks and Montgomery Counties, Pennsylvania and Kinney and

available

Real Counties, Texas (December 16), and Pinal County, Arizona

bdm6iy3u2.pdf.

at

on

Ice : How Do

http://amvoice.3cdn.net/73093e2eeae748d60d_

(December 23). See ACTIVATED JURISDICTIONS, supra note 3.
28. C enter For C onstitutional R ights, National Day L aborer
17. See Dep’t of Homeland Sec. Appropriations for 2010: Hr’g

Organizing Network and C ardozo S chool of L aw I mmigr ation Jus -

on Priorities Enforcing Immigration Law Before the H. Appro-

tice

priations Comm. Subcommittee on Homeland Sec., 11th Cong.

available at http://uncoverthetruth.org/wp-content/uploads/

917-1280, at 1238 (2009), see also Uncover the Truth, Newly

Secure-Communities-Fact-Sheet-Briefing-guide-8-2-2010-Pro-

Released Secure Communities Documents Signal Opening

duction.pdf.; R ights Working G roup, Faces of Racial Profiling:

for

http://uncoverthetruth.org/featured/

A Report From Communities Across America 7 (2010), available at

newly-released-secure-communities-documents-signal-opening-

http://w w w.rightsworkinggroup.org/sites/default/files/rwg-

for-local-opt-out/

report-web.pdf.

18. See Shankar Vedantam, Wash. Post, No Opt-Out for Immigration

29. Restoring Community supra note 19 and see A arti K ohli & Deepa

Enforcement, Oct. 1, 2010, available at http://www.washingtonpost.

Varma , C hief Justice E arl Warren I nstitute

Local

Opt-Out,

on

R ace, Ethnicity

com/wp-dyn/content/article/2010/09/30/AR2010093007268.

& Diversity Borders, Jails And Jobsites: An Overview Of Federal

html.

Immigration Enforcement In The U.S. (Feb. 2011) at 23.

19. For a more detailed exploration of the opt-out controversy

30. U.S. I mmigr ation

and

C ustoms Enforcement, Secure

see Restoring Community, A National Community Advisory Report

C ommunities : Q uarterly R eport, F iscal Year 2010, Report to

on ICE’s Failed ‘Secure Communities’ Program, 25-28, August 2011,

Congress (Aug. 11, 2011), http://www.ice.gov/doclib/foia/

available at http://altopolimigra.com/s-comm-shadow-report.

secure_communities/r_congressionalfy10rdquarterreport.pdf.

20. New York State, Illinois and Massachusetts have all sought to

31. Id.

leave the program. See e.g., Resistance Grows, supra note 9.
21. Kirk Semple & Julia Preston, Deal to Share Fingerprints Is

32. Secure Communities: IDENT/IAFIS Interoperability Monthly
Statistics through April 30, 2011, U.S. I mmigr ation

and

C ustoms

Dropped, Not Program, N.Y. Times, Aug. 6, 2011, at A11, available

Enforcement, retrieved from http://www.ice.gov/doclib/foia/

at

sc-stats/nationwide_interoperability_stats-fy2011-feb28.pdf;

http://www.nytimes.com/2011/08/06/us/06immig.html;

Elise Foley, Secure Communities Agreements Canceled, Participation Still Required, Huffington Post, Aug. 5, 2011, http://www.
huffingtonpost.com/2011/08/05/secure-communities-updatedepartment-of-homeland-security_n_919651.html.

16

C linic Briefing Guide to Secure Communities, 3 (Aug. 2, 2010),

O c t o b e r 2 0 11

|

Secure Communites by the Numbers

See also C enter For C onstitutional R ights, supra note 28.
33. Immigration and Nationality Act of 1952, 8 U.S.C. (hereinafter 8 U.S.C.) § 236(c) (2011).

34. Illegal Immigration Reform and Immigrant Responsibility

43. In many of the litigated cases the U.S. citizens concerned had

Act of 1996, Pub. L. 104-208, § 321, 110 Stat. 3009, 3009-627.

mental disabilities and were left in highly vulnerable situations
after deportation. See Ian James, Wrongly Deported, American Citi-

35. Id.

zen Sues INS for $8 million, A ssoc . P ress, Sept. 3, 2000; Guzman

36. See e.g., A merican C ivil L iberties Union
The Policies A nd Politics
tion

of

L aws : 287(G) P rogram

in

of

North C arolina ,

L ocal Enforcement

of

I mmigr a-

North C arolina , 16, Feb. 2009,

available at http://acluofnc.org/files/287gpolicyreview_0.pdf
(noting that 287(g) leads to increased due process violations
because of inadequate training and unfamiliarity of local officials with the complexities of immigration law and procedure);
Teresa A. Miller, Lessons Learned, Lessons Lost: :Immigration
Enforcement’s Failed Experiment, 38 FORDHAM URB. L. J. 217,
238-241 Nov. 2010 (arguing that immigration detainers, which
are the lynchpin of programs like CAP and Secure Communities, lead to increased due process violations).
37. Supra note 14.
38. For a complete list of data fields see Appendix A.
39. We do not have data on non-traditional family structures
such as domestic partnerships.

v. Chertoff, No. 08-01327 (C.D. Cal. filed Feb. 27, 2008) (detention and deportation of USC with mental disabilities, settled in
December 2009); Elise Foley, Lawsuit Claims ICE Deported Mentally Ill U.S. Citizen, Wash. I ndependent, Oct. 14, 2010, discussing
Lyttle v U.S.A, .http://www.aclu.org/immigrants-rights/lyttlev-united-states-america-et-al filed by the ACLU on behalf of
Hispanic USC with mental disabilities who was detained and
deported).
44. Jeffrey S. Passel & Paul Taylor , Unauthorized I mmigr ants
and

Their U.S.-B orn C hildren, P ew H ispanic C tr ., 1 (2010),

available at http://pewhispanic.org/files/reports/125.pdf.
45. CAROLA SUÁREZ-OROZCO, ET AL., Growing Up in the
Shadows: The Developmental Implications of Unauthorized
Status, 81 HARV. EDUC. REV. 438, 462 (2011).
46. Id. and Damien Cave, Crossing Over, and Over, N.Y. Times,
Oct. 3, 2011, at A1, available at www.nytimes.com/2011/10/03/
world/americas/mexican-immigrants-repeatedly-brave-risks-

40. Secure Communities supra note 32 at 50.

to-resume-lives-in-united-states.html.

41. None were issued ICE detainers.

47. We generated this estimate by following the same methodol-

42. The cumulative number of administrated arrests and/or
bookings was retrieved here: Secure Communities: IDENT/
IAFIS Interoperability Monthly Statistics supra note 32 at 2. We
extrapolated the number of U.S. citizens apprehended by ICE
by multiplying 226,694 by 0.016 (the estimated proportion of
U.S. citizens in the sample). The proportion of 0.016 is statis-

ogy used to extrapolate to apprehended U.S. citizen, see supra
at note 42, except we multiplied 226,694 by 0.39. Based on the
upper and the lower boundary of the 95% confidence interval,
we estimate the maximum and minimum number of families
with U.S. citizens affected by Secure Communities to be 99,498
and 77,021, respectively.

tically significant at the 95% confidence level according to a

48. Data on the undocumented population in the U.S. is an apt

one-sample t-test against zero. We also used the upper and the

point of comparison because at least 80% of our sample con-

lower boundary of the 95% confidence interval of the estimated

sists of individuals whose status upon entry was Present without

proportion of U.S. citizens among our sample to estimate the

Admission (PWA) indicating they are likely to be undocumented.

maximum and the minimum numbers of U.S. citizens being
apprehended by ICE, which are 6,574 (226,694 x .003) and 680
(226,694 x .029). The federal government no longer reports the
number of administrative arrests and/or ICE bookings in their
interoperability reports; as such, it is impossible to calculate
the number of U.S. citizens apprehended after April 2011 using
the same methodology.

49. ICE maintains data for males and females. Therefore, the
number of transgendered persons impacted by Secure Communities is unknown at this time.
50. M ichael Hoefer , Nancy R ytina , & Bryan C. Baker , Office
I mmigr ation Statistics, U.S. Dep’t
of

of

of

Homeland Sec ., E stimates

The Unauthorized I mmigr ant Population R esiding in the U.S.:

January 2010, 5 (February 2011). Although women comprise just
7% of arrests under Secure Communities, as noted above they
are impacted by the deportation of their partners, siblings and
fathers. Although our population consists of immigrants in all
statuses (unauthorized, visitors, lawful permanent residents) the
majority are unauthorized.

Secure Communites by the Numbers

|

O c t o b e r 2 0 11

17

51. Id. at 3.
52. William A. K andel , C ong. R esearch Serv.,

the

U.S. Foreign -

B orn Population : Trends and Selected C haracteristics 15 (2011),
available at http://assets.opencrs.com/rpts/R41592_20110118.
pdf.

64. Janet Napolitano, U.S. Secretary

of

Homeland Secu -

rity,

on

Smart E ffective

R emarks

B order Security

A merican University
and

I mmigr ation Enforcement (Oct. 5,

2011), (transcript available at http://www.dhs.gov/ynews/
speeches/20111005-napolitano-remarks-border-strategy-andimmigration-enforcement.shtm.)

53. F ederal Bureau of I nvestigation, C rime in the United States,
2009, (2010).

naturalized citizens and undocumented residents.

of

Birth

of the

Foreign -B orn Population : 2009

2 (2010), available at http://www.census.gov/prod/2010pubs/
acsbr09-15.pdf.

alty.”) (citing Fong Yue Ting v. United States, 149 U.S. 698, 740
(1893).
67. 8 U.S.C. § 1225(b)(1)(A)(i), (iii) (ordering removal, without
further hearing or review, of arriving aliens who are inadmissi-

56. Jeffrey S. Passel, Pew Hispanic Ctr., The Size
teristics of the

66. Padilla v. Kentucky, 130 S. Ct. 1473, 1481 (2010) (“We have
long recognized that deportation is a particularly severe pen-

55. E lizabeth M. G rieco & Edward N. Trevelyan, U.S. C ensus
Bureau, P lace

65. The remaining analyses in this report are based on the 369
non-U.S. citizens who were arrested by ICE.

54. Foreign-born persons include lawful permanent residents,

Unauthorized Migrant Population

and

Charac -

in the

U.S. 5

(2006), available at http://pewhispanic.org/files/reports/61.pdf.
57. For the purposes of this analysis we categorize all individuals
who claim a Latin American country as their country of citizenship as Latino or Hispanic, which are interchangeable terms.
We use Latino in this report.
58. The remaining 3% of the population is from Africa and
other countries. PASSEL supra note 56.
59. HSAC Task Force Report, supra note 7, at 12.

ble because they lack proper documentation or have committed
fraud or willful misrepresentation to procure an immigration
benefit and persons who have entered the United States without
inspection within the prior two years).
68. 8 U.S.C. § 1225(b)(1)(B)(iii)(IV).
69. Two individuals had a “voluntary departure” proceeding
initiated and we coded these as having a “Hearing” because voluntary departure can only be granted by an immigration judge.
70. For a critical analysis of the stipulated removal process see
Jennifer L ee K oh, Jayashri Srik antiah, K aren C. Tumlin, Depor tation

60. Briefing Guide, supra note 28.
61. Restoring Community supra note 19, at 10.
62. See Figure 3. We see a similar demographic pattern in the
portion of our sample that did not have criminal removal
charges: 95% are Latino, 92% are men, and the median age is 29.
63. P ublic Policy I nstitute of C alifornia , I mmigrants and C rime,
(2008),

available

at

http://www.ppic.org/content/pubs/jtf/

JTF_ImmigrantsCrimeJTF.pdf. See also Brad Myrstol, Noncitizens
Among Anchorage Arrestees, 20 A lask a Just. F. 1, 7 (2003), available
at

at

http://justice.uaa.alaska.edu/forum/20/1spring2003/201.

spring2003.pdf. Immigrants do have higher rates of representation in federal prisons but that is mainly attributed to
immigration-related offenses such as misdemeanor unlawful
entry (8 USC 1325) or felon re-entry (8 USC 1326).

Without Due P rocess 5 (Sept. 2011), available at: http://

www.nilc.org/immlawpolicy/arrestdet/Deportation-WithoutDue-Process-2011-09.pdf.
71. “ [O]ther records suggest that immigration judges are given
“case completion” credit for stipulated removals as if they had
completed a regular court hearing, thereby providing an incentive for them to sign stipulated removals as quickly as possible in
order to claim higher individual case closures and manage their
extraordinarily high caseloads.” fn20 [Email from S. McDaniel
to S. Griswold and S. Rosen, re: Stips (Apr. 25, 2008) (EOIR2008-5140(5)-000039-42) (noting that a San Francisco court was
able to meet “case completion goals” due to the large number of
stipulated removals in the jurisdiction).] Id. at 5.
72. K atzmann Immigrant R epresentation Study Group and the Vera
Institute

of

Justice, The New York Immigrant Representation Study

Preliminary Findings (May 3, 2011), available at http://graphics8.
nytimes.com/packages/pdf/nyregion/050411immigrant.pdf.

18

O c t o b e r 2 0 11

|

Secure Communites by the Numbers

73. We use DHS’s 2009 statistics here for comparison since the

80. A mnesty I nternational , Jailed Without Justice : I mmigr ation

large majority of our sample are 2009 apprehensions. In 2009

Detention

DHS had approximately 613,000 immigration apprehension and

www.amnestyusa.org/pdfs/JailedWithoutJustice.pdf.

detained approximately 383,000 of them, for a 62% detention
rate. U.S. Dep’t

of

Homeland Sec ., I mmigration Enforcement

Actions : 2009, ANNUAL REPORT (Aug. 2010), available at
http://w w w.dhs.gov/xlibrary/assets/statistics/publications/
enforcement_ar_2009.pdf. This likely understates the disparity
between the detention rate for Secure Communities versus other
ICE arrestees since that large majority of DHS apprehensions in
2009 were Customs and Border Patrol (CBP) not ICE apprehensions and CBP is likely to have a higher detention rate. See Dr .
Dora S chriro, Dep’t of Homeland Sec ., I mmigration and C ustoms
Enforcement, I mmigration Detention O verview and R ecommendations

2, 4 (Oct.6, 2009), available at http://www.ice.gov/doclib/

about/offices/odpp/pdf/ice-detention-rpt.pdf (last visited Oct.
12, 2011); Anil Kalhan, Rethinking Immigration Detention, 110
Colum. L. Rev. Sidebar 42 ( July 21, 2010); and Ellis Johnston,
Once a Criminal,Always a Criminal? Unconstitutional Presumptions
for Mandatory Detention of Criminal Aliens, 89 Geo. L.J. 2593, 2596
(August 2001).
74. See e.g., Human R ights F irst, Jails
ing the

and

Jumpsuits Tr ansform -

U.S. I mmigration Detention S ystem —A Two -Year R eview,

in the

USA, (March 2009) 18, available at: http://

81. Aggravated felonies are defined in section 101(a)(43) of the
Immigration and Nationality Act. The term is often called a
misnomer. For a criminal offense to be deemed an aggravated
felony, it need not be a felony nor involve aggravating circumstances. For example, shoplifting with a one-year suspended
sentence can be considered an aggravated felony under immigration law.
82. Twenty-two percent of those with at least one removal
code had more than one code (and most had just two codes).
For these cases, we categorized them in the following priority
order:

Aggravated Felony, Other Criminal Grounds, Other

Immigration Grounds, Present Without Admission. For example, if someone had two removal codes and one was for Other
Criminal Grounds and the other was for Other Immigration
Grounds, they were categorized as the former for purposes of
this analysis.
83. Almost half of the 7% without removal charges were slated to
have an immigration hearing.

2011, available at http://www.humanrightsfirst.org/wp-content/

84. While the data provides no explanation for the detention

uploads/pdf/HRF-Jails-and-Jumpsuits-report.pdf (last visited

of individuals without removal charges, it does raise serious

Oct. 12, 2011).

questions about whether individuals not subject to removal were

75. We assume that a departure date indicates that the person
physically left the U.S.
76. Individuals with detention book-out dates and departure
dates comprised 65% of our sample.
77. Bail Reform Act of 1984, 18 U.S.C. § 3142.
78. 8 U.S.C. § 236(c); Demore v. Hyung Joon Kim, 538 U.S. 510

wrongfully detained.
85. This classification raised serious due process issues since not
everyone charged with an offense is ultimately prosecuted or
found guilty.
86. M emor andum F rom John Morton, Director , Dep’t
Homeland Sec ., I mmigr ation

and

of

C ustoms Enforcement Civil

Immigration Enforcement: Priorities for the Apprehension,

(2003).

Detention, and Removal of Aliens (Mar. 2, 2011), available at

79. See 8 U.S.C. § 1226(c) (mandating detention of certain

tondc.pdf.

criminal aliens); 8 U.S.C. § 1226(e) (denying judicial review of
discretionary decisions regarding application of apprehension
and detention of aliens); ); 8 U.S.C. § 1225(b)(1)(A)(i), (B)(iii)
(IV) (ordering removed arriving aliens who are inadmissible
because inter alia they lack a valid entry document and providing

http://www.ice.gov/doclib/news/releases/2011/110302washing

87. Id.
88. See note 32, supra.
89. See note 73, supra.

mandatory detention for such individuals who indicate a credible fear of persecution or an intention to seek asylum); 8 C.F.R.

90. Human R ights Watch, L ocked Up Far Away : The Tr ansfer

§ 1003.19(h)(2)(i) (denying immigration judges with ability to

I mmigr ants

re-determine conditions of custody for inter alia arriving aliens

2 (2009), available at http://www.hrw.org/sites/default/files/

in removal proceedings).

reports/us1209web.pdf.

to

R emote Detention C enters

Secure Communites by the Numbers

in the

|

of

United States,

O c t o b e r 2 0 11

19

91. See discussion infra Immigrant Arrest Outcomes.
92. See e.g., A mnesty I ntl ., Jailed Without Justice, supra note

99. For this analysis we compared the Notice to Appear date
with the immigration judge’s decision date.

80; and Peter L. Markowitz, Barriers to Representation for Detained

100. 8 U.S.C. § 1182(c) (1988). (repealed by IIRIRA). However,

Immigrants Facing Deportation: Varick Street Detention Facility, A Case

in Immigration and Naturalization Service v. St. Cyr, the Court ruled

Study, 78 Fordham L. R ev. 541 (2009).

that certain categories of non-citizens with convictions pre-dat-

93. See Mark Hamblett, Forum Stress Plight of New York’s Unrepresented Immigrants, New York L. J., (May 6, 2011), available at

ing IIRIRA may be eligible for 212(c) relief. 533 U.S. 289, 326
(2001).

http://www.law.com/jsp/law/international/LawArticleIntl.jsp?

101. See 8 U.S.C. § 1229b(a) (2006). Cancellation of Removal is

id=1202493130604&slreturn=1.

not available to noncitizens convicted of aggravated felonies.

94. U.S. Dep’t of Justice, E xecutive Office of I mmigration R eview,
FY 2010 Statistical Year B ook , G1 ( Jan. 2011), available at http://
www.justice.gov/eoir/statspub/fy10syb.pdf.
95. These were identified as those having a date of E-28, Notice
of Entry of Appearance as Attorney or Representative.
96. Also, 70% of those who had a new bond hearing had lawyers.
97. Rising Immigration Backlog supra note 2.

Even for LPRs who are eligible for Cancellation do not always
prevail. In FY 2010, approximately 3700 cases of Cancellation
of Removal were granted. E xecutive Office For I mmigr ation
R eview, U.S. Dep’t

of

Justice, F y 2010 Statistical Year B ook , at

R3 (table 15), available at http://www.justice.gov/eoir/statspub/
fy10syb.pdf.
102. 8 U.S.C. § 240A.
103. U.S. Dep’t of Justice supra note 94, at D2.

98. Jennifer L ee K oh, Et A l ., supra note 70.

This report was made possible by generous grants from Unbound Philanthropy, Ford Foundation, Walter and Evelyn Haas Jr. Foundation and
the Akonadi Foundation. The authors thank Bridget Kessler for her invaluable contributions during the initial stage of the data analysis. We
are also indebted to Eleanor Taylor-Nicholson, Su Li, Rebecca Silbert, Elaine Mui, and Janet Velazquez for their feedback and support. We are
grateful to Catherine Barry, Amy Barsky, Margaret Chen, Sharada Jambapulati, Taneisha Means and Jennifer Park for their expert research
assistance. We would not have been able to conduct this research without the efforts of the National Day Labor Organizing Network, the Center
for Constitutional Rights and the Cardozo Immigration Justice Clinic who obtained federal Secure Communities data through litigation and
made it available to the Warren Institute for analysis. In particular, we are grateful to Caroline Glickler, James Horton, Jessica Karp, Chris
Newman, Sunita Patel, Sarahi Uribe, and Hannah Weinstein. The conclusions contained in this report are those of the authors and should
not be attributed to our funders.

20

O c t o b e r 2 0 11

|

Secure Communites by the Numbers