Skip navigation

The Price of Freedom - Bail and Pretrial Detention of Low Income Nonefelony Defendants in NYC, Human Rights Watch, 2010

Download original document:
Brief thumbnail
This text is machine-read, and may contain errors. Check the original document to verify accuracy.
United States

The Price of Freedom
Bail and Pretrial Detention of Low Income
Nonfelony Defendants in New York City

H U M A N
R I G H T S
W A T C H

The Price of Freedom
Bail and Pretrial Detention of Low Income
Nonfelony Defendants in New York City

Copyright © 2010 Human Rights Watch
All rights reserved.
Printed in the United States of America
ISBN: 1-56432-718-3
Cover design by Rafael Jimenez
Human Rights Watch
350 Fifth Avenue, 34th floor
New York, NY 10118-3299 USA
Tel: +1 212 290 4700, Fax: +1 212 736 1300
hrwnyc@hrw.org
Poststraße 4-5
10178 Berlin, Germany
Tel: +49 30 2593 06-10, Fax: +49 30 2593 0629
berlin@hrw.org
Avenue des Gaulois, 7
1040 Brussels, Belgium
Tel: + 32 (2) 732 2009, Fax: + 32 (2) 732 0471
hrwbe@hrw.org
64-66 Rue de Lausanne
1202 Geneva, Switzerland
Tel: +41 22 738 0481, Fax: +41 22 738 1791
hrwgva@hrw.org
2-12 Pentonville Road, 2nd Floor
London N1 9HF, UK
Tel: +44 20 7713 1995, Fax: +44 20 7713 1800
hrwuk@hrw.org
27 Rue de Lisbonne
75008 Paris, France
Tel: +33 (1)43 59 55 35, Fax: +33 (1) 43 59 55 22
paris@hrw.org
1630 Connecticut Avenue, N.W., Suite 500
Washington, DC 20009 USA
Tel: +1 202 612 4321, Fax: +1 202 612 4333
hrwdc@hrw.org
Web Site Address: http://www.hrw.org

December 2010

ISBN: 1-56432-718-3

The Price of Freedom
Bail and Pretrial Detention of Low Income Nonfelony Defendants in New York City
Summary ........................................................................................................................................... 1
Recommendations ............................................................................................................................. 7
I. The Bail Process............................................................................................................................ 10
Arrest through Arraignment ........................................................................................................ 10
Misdemeanor Arrests .......................................................................................................... 10
Arraignment ........................................................................................................................ 11
The Statutory Framework ........................................................................................................... 13
II. Who Suffers?: Bail and Pretrial Detention of Low Income Defendants .......................................... 20
For the Poor, Bail Means Jail...................................................................................................... 20
Sub Rosa Preventive Detention ................................................................................................. 26
Pretrial Punishment ................................................................................................................... 27
Plea Bargains ............................................................................................................................ 31
III. Factors Influencing Judges’ Bail Decisions .................................................................................. 35
The Exercise of Judicial Discretion .............................................................................................. 35
The New York City Criminal Justice Agency’s Release Recommendations .....................................39
Role of Prosecutor...................................................................................................................... 41
Public Safety ............................................................................................................................ 46
Race .......................................................................................................................................... 47
The Myth that Released Defendants Evade Justice ...................................................................... 50
Judicial Training ......................................................................................................................... 53
IV. An Alternative to Pretrial Detention: Pretrial Supervision ............................................................ 54
Pretrial Supervision: Avoiding the Expense of Detention ............................................................. 56
V. Applicable Constitutional and Human Rights Law ....................................................................... 58
Constitutional law...................................................................................................................... 58
Right to Liberty .................................................................................................................... 58
Equal Protection ................................................................................................................. 60
International Human Rights Law.................................................................................................63
Acknowledgments .......................................................................................................................... 69

Summary
“[Bail] is hostile to the poor and favorable only to the rich. The poor man has
not always a security to pledge…”
—Alexis de Tocqueville, Democracy in America, 1835
Decades ago New York City pioneered bail reforms that have been enormously successful in
reducing the number of people detained in jail while awaiting trial: each year judges order
the pretrial release without bail of tens of thousands of men and women accused of crimes
who are then able to remain in their homes and communities pending conclusion of their
cases. When judges believe defendants might not otherwise return to court, however, they
set money bail as a condition of release. Unfortunately, today as years ago, bail is usually
out of reach for poor defendants. For people scrambling to pay the rent each month or who
live on the streets, a bail of US$1,000 or even $500 can be as impossible to make as one of
$100,000. Unable to post bail, they are sent to jail for pretrial detention.
Pretrial detention may be appropriate for dangerous defendants charged with violent crimes.
But the preponderance of criminal defendants in New York City are accused of low level
offenses such as smoking marijuana in public, turnstile jumping, or shoplifting. Sending
them to jail for want of a few hundred dollars cannot be squared with basic notions of
fairness, human rights, or fiscal common sense.
In the eyes of New York law, pretrial detention is not punishment but a precautionary
measure to ensure defendants show up in court for their cases. From the perspective of
those enduring days and nights behind bars, this is a distinction without a difference.
People should not have to endure jail simply because they are too poor to buy their way out,
particularly when there are other ways of ensuring that such defendants make their
scheduled court appearances.
This report addresses the pretrial incarceration of New York City defendants accused of
nonfelony crimes, mostly misdemeanors. Previously unpublished data provided to Human
Rights Watch by the New York City Criminal Justice Agency (CJA)—covering all cases (117,064)
of nonfelony defendants arrested in New York City in 2008 that proceeded past arraignment—
suggests the extent of the problem: In slightly more than three-quarters (90,605) of the cases
defendants were released pending trial on their own recognizance (i.e., without money bail). In
most of the cases where bail was set (19,137 of 26,459 cases), the bail amount was $1,000 or

1

Human Rights Watch | December 2010

less. Nevertheless, despite the relatively low bail amount, the overwhelming preponderance of
defendants required to post that bail amount were jailed because they could not do so.
•

In 87 percent of the cases (16,649) in which the defendants arrested in 2008 had
bail set at $1,000 or less, the defendants were not able to post bail at their
arraignment and were incarcerated pending trial; and
o
o

Their average length of pretrial detention was 15.7 days;
Almost three out of four (71.1 percent) were accused of nonviolent, nonweapons related crimes.

At any given moment, 39 percent of the city’s jail population consists of felony and
nonfelony pretrial detainees who are in jail because they have not posted bail. The equitable
and human rights concerns about conditioning pretrial freedom on financial wherewithal is,
of course, present in felony as well as nonfelony cases. In this report, however, we focus
solely on the pretrial detention of persons accused of nonfelonies (primarily misdemeanors,
but also including violations and infractions), because their incarceration is uniquely
difficult to reconcile with the fundamental notions of fairness and equality that should be
the cornerstones of criminal justice. Pretrial detention is a disproportionate curtailment of
liberty in light of the non-threatening, petty nature of most of the alleged nonfelony crimes.
Time in jail before one has had one’s day in court is particularly troubling for the one in five
detained nonfelony defendants who, according to the CJA, will not be convicted. It is also
disproportionate in light of sentences typically imposed when there is a nonfelony
conviction: data from the New York State Division of Criminal Justice Services, for example,
indicates that eight out of ten convicted misdemeanor arrestees receive sentences that do
not include jail time.
Monetizing pretrial freedom is inherently disturbing, but more than principle is involved. Jail is
unpleasant and dehumanizing in the best of circumstances; it can be violent and degrading as
well. In addition to the stress of incarceration itself, pretrial detention can harm individuals
and their families in countless ways: for example, detained men and women lose income they
and their families need, and even their jobs; they cannot care for children or ailing relatives
who depend on them; they miss school and exams; they cannot attend substance abuse and
mental health treatment programs; and they can lose their places in homeless shelters.
Pretrial confinement also increases the likelihood of conviction. Pretrial confinement—or just
the threat of confinement—prompts defendants to plead guilty and give up their right to trial.

The Price of Freedom

2

Most persons accused of low level offenses when faced with a bail amount they cannot
make will accept a guilty plea; if they do not plea at arraignment, they will do so after having
been in detention a week or two. Guilty pleas account for 99.6 percent of all convictions of
New York City misdemeanor defendants.
Under New York law, defendants accused of nonfelony
offenses have the right to pretrial release on recognizance or
bail. The theory of money bail is that it provides a financial
incentive for released defendants to attend scheduled court
proceedings; the threat of losing the funds used to post bail
will deter or inhibit the temptation to avoid court. While neither
the US nor New York state constitutions guarantee that bail
shall be made available to defendants, they do prohibit
“excessive bail,” that is, conditions of bail that are greater
than reasonably necessary to achieve the goal of ensuring
court appearances.

Deprivation of liberty pending
trial is harsh and oppressive,
subjects defendants to
economic and psychological
hardship, interferes with their
ability to defend themselves,
and, in many instances,
deprives their families of
support.
—National Association of
Pretrial Services Agencies,
“Standards on Pretrial
Release,” 2004

In theory, if the arraignment judge making the initial bail
determination decides to impose money bail, he or she
should tailor the bail amount to the defendant’s financial resources, setting it only as high
as necessary to reasonably assure the defendant will return to court. The amount that would
provide a meaningful incentive to return to court for someone who lives on $600 a month is,
obviously, different than it is for someone who lives on twice or ten times that much. In
practice, however, judges lack the time, information, and perhaps sometimes even the
inclination, to make careful, individualized bail determinations. In what has not unfairly
been called “assembly line” justice, the entire arraignment process for a New York City
nonfelony defendant is a rushed affair of no more than a few minutes.
Judges have broad discretion to make release and bail decisions, and they exercise that
discretion according to their individual concerns, prior experience, and predilections, with
the prosecutors’ positions having a major influence. Although the legal purpose of bail in
New York is solely to encourage the defendant to return to court, judges in an unknown and
unknowable number of cases take other factors into consideration in any given case. For
example, judges may decide to set bail with an eye to protecting public safety (the
incarceration following inability to make bail serving as a form of sub rosa preventive
detention), to encourage the defendant to plead guilty, or to impose preemptive punishment
on defendants they assume, based on their prior criminal records, to be guilty. These

3

Human Rights Watch | December 2010

reasons are not authorized by law—and they may violate the right to be presumed
innocent—but judges do not have to explain their bail decisions.
Whether deliberately, inadvertently, or carelessly, judges usually set money bail at an
amount the defendant cannot afford, as evidenced by the fact that defendants in only 10
percent of all criminal cases in which bail is set are able to post it at arraignment. Bail set at
under $500 is rare. Judges also almost invariably set bail in the form of cash or secured bond
(a commercial bail bond or other bond secured by a deposit of money or lien upon property),
even though New York law provides alternative financial forms of bail that are less onerous
for low income or indigent defendants. For example, the law permits unsecured bonds. For
reasons that are obscure, however, alternatives to cash bail or secured bonds are simply not
part of customary court practice in New York City. Defendants who do not have the resources
at arraignment to pay cash bail and who cannot obtain a bond from commercial bondsmen
find themselves in pretrial detention.
The American Bar Association (ABA) has long criticized money bail, pointing out that it
undermines the integrity of the criminal justice system, is unfair to poor defendants, and is
ineffective in achieving key objectives of the pretrial release/detention decision. The ABA’s
Standards for Pretrial Release mandate that financial conditions should be used only when
no other conditions will provide reasonable assurance a defendant will appear for future
court appearances. If financial conditions are imposed, the court should first consider
releasing the defendant on an unsecured bond, and if that is deemed an insufficient
condition of release, bail should be set at the lowest level necessary to ensure the
defendant’s appearance and with regard to his financial ability. Significantly, ABA Standard
10-1.4 prohibits bail that results in pretrial detention: “The judicial officer should not impose
a financial condition of release that results in the pretrial detention of a defendant solely
due to the defendant’s inability to pay.”
According to the Pretrial Justice Institute of Washington, D.C., many states have laws that
establish a presumption or preference for nonfinancial conditions of release or unsecured
bonds. The federal government and the District of Columbia prohibit courts from imposing
money bail that defendants cannot meet and which therefore results in their pretrial detention.
Judges may set bail at levels they know are likely to result in defendants’ pretrial detention
because they believe such detention is necessary to ensure the defendants appear in court.
But failure to appear is in fact relatively infrequent: most New York City defendants who are
released pretrial show up for the court proceedings in their cases. Sixteen percent will miss a

The Price of Freedom

4

scheduled court appearance prior to disposition of the case, but most of those who miss an
appearance will return to court voluntarily within 30 days.
Our interviews show that defendants released pending trial come to court for multiple
appearances even though they have to miss work or school, arrange for childcare or care of
ailing relatives, miss drug treatment programs or medical appointments, have to find the
money for transportation to and from the court building, and even though they may have to
spend the entire day sitting in the courtroom waiting for their case to be called. When they
miss a proceeding, it is often because they simply could not manage to make it, their lives
too chaotic or stressful. No doubt some defendants are irresponsible. There is also little
doubt, as many prosecutors point out, that some defendants miss court appearances
because they do not want to risk adjudication and they know there is little likelihood of the
police picking them up on the bench warrant for their arrest that the court issues when they
do not show up. But a large majority of those who miss an appearance show up for
subsequent proceedings. According to the CJA, only 6 percent of released defendants miss a
court appearance and do not return to court within 30 days: the pool of defendants who
deliberately seek to evade justice is quite small.
The criminal justice system has a compelling interest in making sure all criminal defendants
appear in court for their cases. But bail or jail need not be the only means to that end. Pilot
initiatives in New York City and the experience of pretrial services agencies across the
country have shown that there are other ways to secure defendants’ appearance in court.
Systems that employ court date notification and pretrial monitoring and supervision, graded
in intensity as warranted by the circumstances, all promote a defendant’s return to court
while respecting the right to liberty and the presumption of innocence.
Incarcerating defendants who cannot post bail is not a
problem unique to New York. According to the Bureau of
Justice Statistics of the US Department of Justice, 62 percent of
the nation’s jail population consists of detainees awaiting trial.
The Pretrial Justice Institute estimates it costs $9 billion
annually to incarcerate defendants held on bail nationwide.
That the problem is widespread, however, is no excuse.

“In the pretrial context, money
does one thing, and only one
thing, well: separate those
who have it from those who
don’t.”
—Timothy J. Murray, Executive
Director, Pretrial Justice
Institute, 2010

Human Rights Watch believes New York City can go a long way
toward addressing the economic inequities of the present bail system by providing supervised
pretrial release programs for defendants who are not suitable for release on their own
recognizance and who cannot afford bail. Such pretrial supervision would not only honor the

5

Human Rights Watch | December 2010

presumption of innocence, but would save the city tens of millions of dollars in jail costs.
According to the New York City Department of Correction, the amount that would be saved by
reducing the jail population by 800 or more inmates is $161 per inmate a day. Applying that
figure to the data above on nonfelony defendants arrested in 2008—16,649 nonfelony
defendants unable to post bail of $1,000 or less held in pretrial detention an average of 15.7
days—it is easy to see how significant the savings could be. If the city had not incarcerated any
of those nonfelony defendants, it would have saved more than $42 million; if it had
incarcerated only a third as many, it would have saved more than $28 million; and so on.
Any form of pretrial supervision established to replace jail for defendants who cannot afford
bail would surely cost far less than jail. The cost of probation services, for example, is a
fraction of the expense of institutional confinement: in New York state the annual cost of
probation services per probationer is approximately $4,000; the average annual cost per
inmate in New York City jail is $76,229.
Even if it would not yield cost savings for New York City, action should be taken to end the
pretrial detention of nonfelony defendants who cannot afford bail. Although it is routine,
happens every day, and has gone on for decades, confining people in jail simply because
they are too poor to post bail when charged with a low level offense is a serious inequity in
the city’s criminal justice system. Poverty should not be an impediment to pretrial freedom.

The Price of Freedom

6

Recommendations
No individual government agency controls the flow of individuals into pretrial detention
because they cannot afford bail. But the problem is not insoluble, and concrete, effective
measures could accomplish much should the key stakeholders—the mayor, city council,
prosecutors, courts, defense attorneys, and civil society representatives—decide to work
together for needed reforms. Limiting the use of pretrial detention would not increase crime
nor harm the interests of justice. Indeed, a far more parsimonious use of pretrial detention is
consistent with the integrity of the judicial system, human rights, public safety, and fiscal
prudence.
Even absent more systematic reform or the institution of pretrial supervision programs for
nonfelony defendants, much could be done under existing law to limit to a far greater extent
than today the pretrial detention of nonfelony defendants who lack the financial resources to
post bail (and who are not granted release on their own recognizance). Judges, prosecutors,
and defense attorneys should not accept such pretrial detention as inevitable, much less
desirable. They should vigorously, consistently, and where necessary, creatively exercise
their respective authority under the law to try to ensure nonfelony defendants are not
incarcerated pretrial for want of the funds to buy their freedom. We are asking in short, for a
change in the way business is done with regard to bail setting and its outcomes. It may not
be easy, but it is much needed and long overdue.
We hope that this report raises awareness about the serious problems in New York’s bail
system. We also note below a few key recommendations.
I.

The New York State Legislature should amend the New York Criminal Procedure Law
to incorporate provisions that will prevent the incarceration of misdemeanor
defendants solely because they cannot afford financial conditions of release
imposed to secure their appearance in court. Human Rights Watch strongly
recommends the legislature carefully consider amending the law to incorporate the
American Bar Association’s Standards for Pretrial Release. At the very least, we urge
the legislature to amend the law to incorporate the following provisions:
•

When financial conditions of release are imposed on defendants accused of
nonfelony crimes, the court should first consider releasing the defendant on an
unsecured bond. If an unsecured bond is not deemed a sufficient condition of

7

Human Rights Watch | December 2010

release, bail should be set at the lowest level necessary to ensure the
defendant’s appearance with due regard to the defendant’s financial ability.
•

Courts should not impose bail in a form or amount that results in the pretrial
detention of a defendant charged with a nonfelony offense solely due to the
defendant’s inability to pay.

II. New York City judges should use their discretionary authority under current law to set
bail in forms and amounts that misdemeanor defendants can afford. In particular:
•

Judges should make greater use of unsecured bonds and other bail forms
currently authorized under New York law that are less financially onerous than
cash bail and secured bonds.

•

Judges should make greater effort at arraignment to tie bail amounts more
directly to a defendant’s financial resources and should not set bail at amounts
they know or believe the defendant is likely not to be able to pay.

•

Administrative judges should take steps to inform and educate judges about the
nature and significance of the different forms of bail currently authorized under
New York law.

III. In its training program for new judges, the New York State Court Judicial Institute
should ensure judges understand the nature and significance of the different forms
of bail permitted under New York law and the importance of making bail decisions
that respect the right to liberty and to equality under the law.
IV. Defense counsel representing low income and indigent defendants should gather as
much information as they can regarding the financial condition of their clients and
should advocate vigorously for bail to be set in a form and amount that the
defendant can afford, including by advocating for judges to use unsecured or
partially secured bonds. In each case, however unlikely a good outcome may seem,
they should take every opportunity provided by law and procedure to keep the
defendant out of pretrial detention.
V. Prosecutors should not request bail at an amount greater than reasonably necessary
to ensure the defendant’s future court appearances taking into account what is
known about the defendant’s financial resources, even if the information is
unverified. They should not press for bail in a form or amount that is likely to result in
the defendant’s pretrial detention and they should support bail being set in
unsecured or partially secured bonds.

The Price of Freedom

8

VI. New York City should establish supervised pretrial release programs for defendants
accused of nonfelony crimes who are not released on their own recognizance and who
do not have the financial resources to post bail. Once such a program is operational,
the cost savings from greatly reduced use of city jails to confine pretrial defendants
should generate ample funds to pay for it. The city should consider whether current
pilot initiatives in pretrial release for certain felony defendants provide a suitable
model for nonfelony defendants and, if so, should scale-up those initiatives for such
nonfelony defendants. A mayoral task force or commission of key stakeholders should
be created to review the experience of the pilot programs, and to develop the
appropriate input and support for expanded supervised release programs.

9

Human Rights Watch | December 2010

I. The Bail Process
Arrest through Arraignment
Misdemeanor Arrests
New York City police made 340,859 adult arrests in 2009, up from 282,958 arrests five years
earlier.1 Most of the defendants in New York City criminal cases are men, in half of the cases
the defendant was under the age of 30, and most of those aged 29 and under are involved in
a full-time activity: school or work.2
Most of the arrests are for nonfelony offenses, the vast bulk of which are misdemeanors.4
The number and proportion of felony arrests has declined since the city adopted an
aggressive law enforcement strategy targeting low level “quality of life” offenses. In 1989
half the arrests were for felonies. Twenty years later in 2009,
almost three-quarters (72 percent) of the arrests were for
Anthony C., arrested for selling
handbags on the street without
misdemeanors.5 Most of those arrested for misdemeanors,
license, said he was worried
82.4 percent, are black or Hispanic.6 According to the New
about losing his temporary job:
York State Division of Criminal Justice Services, a substantial
“I’ve got $100 in my pocket.
proportion of those arrested have no prior convictions: 60.4
That’s it. No bank account, no
percent of persons convicted of a misdemeanor in fiscal year
nothing. Without my job I have
2009 had no prior felony convictions; and 47.4 percent had
nothing.”3
no prior misdemeanor convictions. Some, however, have

1

New York State Division of Criminal Justice Services, Adult Arrests: 2000-2009,
http://criminaljustice.state.ny.us/crimnet/ojsa/arrests/nyc.htm (accessed October 25, 2010).
2
New York City Criminal Justice Agency, “2008 Annual Report,” December 2009,
http://www.cjareports.org/reports/annual08.pdf (accessed October 25, 2010), pp. 8-9, Exhibits 2 and 3. The figures are for
both felony and nonfelony defendants.
3
Human Rights Watch interview with Anthony C., New York City, March 9, 2010.
4
Misdemeanors are offenses, “other than a ‘traffic infraction,’ for which a sentence to a term of imprisonment in excess of
fifteen days may be imposed, but for which a sentence to a term of imprisonment in excess of one year cannot be imposed.”
New York Penal Law, sec. 10.00(4). Also included among nonfelony offenses are violations, which are offense, other than
traffic infractions, for which the maximum sentence is fifteen days. New York Penal Law, sec. 10.00(3).
5
There were 245,121 misdemeanor arrests and 97,738 felony arrests in 2009. Reflecting the ongoing priority given to drug law
enforcement in New York City, drug offenses accounted for 33 percent of all misdemeanor arrests; indeed, misdemeanor drug
arrests accounted for one-quarter of all criminal arrests. New York State Division of Criminal Justice Services, Adult Arrests:
2000-2009. In fiscal year 2009, there were 43,787 arrests just for possession of marijuana in public view. Unpublished data
on NYC misdemeanor arrests by charge in fiscal year 2009 provided to Human Rights Watch by New York State Division of
Criminal Justice Services. Data on file at Human Rights Watch. For an illuminating study of changes in New York City nonfelony
defendants, cases, and outcomes between 1989 and 1998, see Freda F. Solomon, New York City Criminal Justice Agency, “The
Impact of Quality-of-Life Policing,” Research Brief No. 3, August 2003, http://www.cjareports.org/reports/brief3.pdf
(accessed October 26, 2010).
6
Unpublished data on the criminal records of persons convicted on misdemeanor charges in New York City in fiscal year 2009
provided to Human Rights Watch by New York State Division of Criminal Justice Services. Data on file at Human Rights Watch.

The Price of Freedom

10

substantial prior records: 13.5 percent had three or more prior felony convictions, and 36.8
percent had three or more prior misdemeanor convictions.7

Arraignment
After a criminal arrest, New York City defendants will be brought to criminal court8 for
arraignment, usually within 24 hours of the arrest.9 In 2009, there were 375,837
arraignments citywide, 82 percent of which were for nonfelony offenses.10 The five most
frequently charged offenses at arraignment in 2009 were misdemeanors: criminal
possession of marijuana (smoking or possessing marijuana in public11), assault in the third
degree, possession of drugs other than marijuana, petit larceny (shoplifting), and theft of
services (most typically turnstile jumping).12
Half of all criminal cases end at arraignment,13 either because the defendant pleads guilty to a
charge or because the case is dismissed or adjourned in contemplation of dismissal.14 The
disposition rate at arraignment varies by charge severity, with misdemeanor and lesser
offenses having higher disposition rates citywide (excluding the Bronx) at arraignment than
felonies.15 If the case does not end, then the judge must make a decision to release the
7
Ibid. Reflecting histories of drug addiction and drug law enforcement, persons convicted of misdemeanor drug offenses tend
to have more prior convictions. For example, among those convicted in fiscal year 2009 of misdemeanor criminal possession
of a controlled substance, 23 percent had 3 or more prior felony convictions, and 61.4 percent had three or more prior
misdemeanor convictions.
8
New York City’s criminal courts handle arraignments for all arrests, from turnstile jumping to murder. They have trial
jurisdiction over all misdemeanors and criminal violations in Kings, New York, Queens, and Richmond Counties from the initial
court appearance until final disposition. In Bronx County, misdemeanors that survive criminal court arraignment are
transferred to the Criminal Division of Bronx Supreme Court.
9
Criminal Court of the City of New York, “Annual Report 2009,” p. 21. The average time from arrest to arraignment citywide is
25.39 hours, up from 21.65 in 1999. Instead of holding a suspect after arrest until the arraignment, police may issue a Desk
Appearance Ticket (DAT), which requires the arrested person to return to court at a specified date and time for arraignment.
DATs may be issued for any nonfelony offenses as well as some nonviolent Class E felony arrest charges. New York Criminal
Procedure Law, sec. 150.20.
10
Among arraignments in 2009, 54,970 were on felony charges, 276,112 on misdemeanor charges, 31,853 on
infraction/violation charges, and 12,902 other. Criminal Court of the City of New York, “Annual Report 2009,” p. 26. The total
number of arraignments has grown since 2001, when the total was 339,993. Ibid.
11
New York Criminal Procedure Law, sec. 221.10. Although possession of small amounts of marijuana for personal use is a
violation in New York, holding or smoking marijuana in public is a misdemeanor.
12
Criminal Court of the City of New York, “Annual Report 2009,” p. 31.
13
Ibid., p. 29.
14
Among the total of 300,319 criminal case dispositions in New York City in 2009, 151,094 (50.3 percent) cases ended with
guilty pleas, 44,988 (15 percent) were dismissed, and 75,530 (25.1 percent) were adjourned in contemplation of dismissal.
Criminal Court of the City of New York, “Annual Report 2009,” p. 16. When a case is adjourned in contemplation of dismissal,
the court will “in furtherance of justice” dismiss the case after the applicable period (either six months or one year) if the
defendant has complied with the conditions of the adjournment, such as performance of community service, and there have
not been any other developments such as a new arrest that might change the court’s decision. New York Criminal Procedure
Law, sec. 170.55 and sec. 170.56.
15
Bronx rates differ significantly from the other boroughs because they transfer most criminal cases to the Supreme Court at
arraignment and cases that are equivalent to Criminal Court cases in other boroughs are categorized as continued rather than
disposed. New York City Criminal Justice Agency, “2008 Annual Report,” p. 18, Exhibit 10.

11

Human Rights Watch | December 2010

defendant on his promise to return (release on recognizance),16 set bail, or remand (which is
not permitted for nonfelony defendants). In 65 percent of the cases—felony and nonfelony—
that continued past arraignment in 2008, the defendant was released on recognizance, and
bail was set in 34 percent. The more serious the charge, the less likely the defendant was to be
released on recognizance. Bail was set in 67 percent of the cases of defendants charged with
the highest severity felonies, 21 percent of cases of defendants charged with misdemeanors,
and in 13 percent of cases with defendants charged with violations or infractions.17 The
defendant was remanded, i.e., ordered detained without bail, in the remaining 1 percent of the
continued cases.18 In 2008 the median amount of bail set at criminal court arraignments
citywide was $2,000.19 In 84 percent of the cases, bail was set at $7,500 or less. In about 2
percent of the cases, bail was set at $100,000 or more.20
The New York City Criminal Justice Agency provided Human Rights Watch with new,
unpublished data on the cases of individuals arrested in 2008 on nonfelony charges whose
cases did not end at arraignment. According to that data, as shown in Table 1, 90,605 or 77.4
percent were released on their own recognizance; 19,137 were given bail equal to or less
than $1,000; and another 7,322 received bail of more than $1,000. In 58 percent of the
cases in which bail was set, the amount was $1,000 or higher. It was almost as likely to be at
an amount greater than $1,000 (28 percent of the bail cases) as at $1,000 (30 percent).
When asked what would happen if her son had to post bail, the
mother of a 17-year-old who was being arraigned on his second
arrest told us: “Oh, he’ll sit there [in jail]…Some people just can’t
afford it.”21

16
“A court releases [a defendant in criminal action or proceeding] on his own recognizance when, having acquired control over
his person, it permits him to be at liberty during the pendency of the criminal action or proceeding involved upon condition
that he will appear thereat whenever his attendance may be required and will at all times render himself amenable to the
orders and processes of the court.” New York Criminal Procedure Law, sec. 500.10(2).
17
Misdemeanor rates calculated from rates for all level misdemeanors combined. New York City Criminal Justice Agency,
“2008 Annual Report,” p. 18, Exhibit 11.
18
New York City Criminal Justice Agency, “2008 Annual Report,” p. 16, Exhibit 9.
19
Ibid. Where bond is ordered along with a cash bail alternative for a lower amount, the CJA uses the lower amount in its bail
calculations.
20
Ibid., p. 21.
21
Human Rights Watch interview with defendant’s mother (name withheld), New York City, March 8, 2010.

The Price of Freedom

12

Table 1
Release on Recognizance and Bail Amount at Arraignment of Nonfelony Cases
Percent of Total
Release/Bail Category
Number of Cases
90,605

77.4%

739

0.6%

Bail $500 to $749

7,178

6.1%

Bail $750 to $999

3,204

2.7%

Bail equal to $1000

8,016

6.8%

Bail greater than $1000

7,322

6.3%

117,064

100%

Release on Recognizance
Bail below $500

Total

Source: New York City Criminal Justice Agency
Note: Includes cases of New York City defendants arrested in 2008 on nonfelony charges. Cases with
bail set at $1 excluded.

The Statutory Framework
Under the New York Criminal Procedure Law, persons arrested for nonfelony offenses have a
right to release on their own recognizance, or bail.22 Although the statute does not express a
presumption in favor of release on recognizance, it does indicate a legislative intent that
defendants accused of nonfelonies remain free pending conclusion of their cases, while
recognizing that in some cases, pretrial release should be subject to conditions.
The New York bail statute enumerates the factors for judges to consider in making bail
decisions:
i)
ii)
iii)
iv)
v)
vi)

The defendant’s character, reputation, habits and mental condition;
His employment and financial resources;
His family ties and the length of his residence if any in the community;
His criminal record, if any;
His record of previous adjudication as a juvenile delinquent, if any;
His previous record in responding to court appearances when required or with
respect to flight to avoid criminal prosecution, if any;
vii) The weight of the evidence against him in the pending criminal action and any other
factor indicating probability or improbability of conviction; and
viii) The sentence which may be or has been imposed upon conviction.23
22
New York Criminal Procedure Law, sec. 530.20(1). “When the defendant is charged…with an offense or offenses of less than
felony grade only, the court must order recognizance or bail.” In felony cases, the statute grants the court discretion to order
recognizance or bail, subject to certain limitations. Ibid., sec. 530.20(2)(a).
23
Ibid., sec. 510.30(2)(a). Although the statutory language limits these factors to cases in which the judge’s release or bail
order is discretionary, that is, in felony cases, judges apparently consider these factors in misdemeanor cases as well.

13

Human Rights Watch | December 2010

These factors “are not a catechism or checklist…they are objective indicia of responsibility
which, if favorable, generally correlate with a likelihood that the defendant will reappear in
court.”24 Each factor should be looked at to “correlate the likelihood that a defendant will
reappear in court.”25 The seriousness of the crime charged and severity of possible
punishment are relevant, for example, “but only to the extent that they demonstrate a
defendant’s propensity to flee.…The nature of the case against the defendant and possible
punishment are only some of the numerous factors to be weighed…”26
In simply enumerating a laundry list of factors, New York law provides little guidance to
judges in how to actually make sound and fair release or bail decisions. Neither the
legislation nor the relatively few court decisions interpreting it indicate how the different
factors should be weighed and balanced against each other, nor does the law even set a
target level of risk of nonappearance that judges should use as a guidepost. As a practical
matter, judges’ discretion in bail decisions is extremely broad, although not completely
unfettered.27 They have wide latitude in deciding which of the enumerated factors to
consider, how much weight to give them, and even what conclusions to draw from them.
If the judge decides not to release on recognizance a defendant charged with a nonfelony
offense, then he must set bail. The court may choose to set bail in any two or more of the
authorized forms of bail, designating one as an alternative.28 New York bail law was
overhauled in 1970 as part of the new Criminal Procedure Law, and, among other bail
provisions, the new law increased the permissible forms of bail with the goal of permitting
more defendants to be released on bail.29 Currently, the following forms of bail are
authorized30:
a. cash bail
b. an insurance company bail bond

24

People v. Torres, 446 N.Y.S.2d 969, 973 (1981).

25

Ibid.

People ex rel. Benton v. Warden, 499 N.Y.S. 2d 738, 740 (1986)(court found an abuse of discretion with regard to the bail
amount when the lower court’s “determination as to the amount of petitioner’s bail seems to have been based exclusively on
its conviction that petitioner would be found guilty of serious crimes entailing a lengthy term of imprisonment, and that there
was no discernable attempt to assess whether petitioner’s presence in court would accordingly be rendered uncertain (a
conclusion that must be considered improbable in light of petitioner’s unblemished record of court attendance).”).
27
People ex rel. Klein v. Krueger, 25 N.Y.2d 497 (1969) (internal citations omitted)(the court’s exercise of its discretion must
not be “improvident….The determination of the bail-fixing court will not be overturned unless there is the ‘invasion of
constitutional right,’ and not a ‘difference of opinion.’”).
28
New York Criminal Procedure Law, sec. 520.10(2)(b).
29
The New York Code of Criminal Procedure, repealed in 1970, authorized bail only through cash, surety, or insurance
company bond.
30
See New York Criminal Procedure Law, sec. 500.10, for definitions of the different types of authorized bail.
26

The Price of Freedom

14

c.
d.
e.
f.
g.
h.
i.

a secured surety bond
a secured appearance bond
a partially secured surety bond
a partially secured appearance bond
an unsecured surety bond
an unsecured appearance bond
a credit card31

In essence, the statute permits bail for criminal defendants in the form of cash or bonds.32
The bonds may be secured, unsecured, or partially secured by the defendant (appearance
bonds) or secured, unsecured, or partially secured by someone other than the defendant,
(e.g., commercial bail bonds).
New York law does not direct judges to give preference to release on recognizance or priority
in bail decisions to unsecured bonds or other less financially onerous forms of bail.33 But the
legislature was well aware of the importance to poor defendants of alternatives to cash or
secured bonds. Unsecured or partially secured bonds “were innovations initiated by the
[new Code of Criminal Procedure Law] and represent less burdensome forms of bail than
those previously available. They were added to vest the court with the utmost flexibility,
including the ability to designate alternative forms with alternative amounts.”34
For example, the statute authorizes bail in the form of an unsecured appearance bond,
which permits the defendant to secure her freedom pretrial with a written promise to pay the
specified bail amount to the court should she fail to appear for the proceedings.35 The
statute also authorizes a partially secured appearance bond, by which the defendant would
deposit a fraction of the bail amount (which could be set at whatever percentage the court
wanted but no more than 10 percent).36 While requiring the defendant to come up with some
cash, the financial burden is still far less than cash bail. For a $500 partially secured
appearance bond, the defendant would have to place no more than $50 with the court
(which would be refunded at conclusion of case), a far cry from $500. She would be liable for

31
New York Criminal Procedure Law, sec. 520.10(1). Effective August 2010, the law only authorizes credit cards to be used to
post bail when the charge is a traffic or vehicle violation.
32
Whatever the forms of bail specified in the court’s order, the defendant may always post cash bail in the amount designated
in the order setting bail. New York Criminal Procedure Law, sec. 520.15.
33
Preiser’s Practice Commentary to Criminal Procedure Law sec. 520.10, McKinney’s Consolidated Law of New York
(annotated), 2009.
34
Ibid.
35
New York Criminal Procedure Law, sec. 520.10(1)(h).
36
Ibid., sec. 520.10(1)(f).

15

Human Rights Watch | December 2010

the full $500 if she did not make all her scheduled court appearances. The statute also
authorizes unsecured surety bonds, in which someone other than the defendant, such as a
parent or grandparent, promises to pay the specified sum of money should the defendant
fail to appear in court; that promise is not accompanied by any deposit or lien upon property.
As originally enacted in 1970, the statute provided that if the court set a bail amount but did
not designate the form of bail, the defendant had to post bail as cash or a secured bond. In
1972, the law was amended to permit defendants to post unsecured bonds in the event the
judge failed to designate the form.37 Once again, the legislature did so with an eye to
addressing the plight of defendants too poor to post bail:
In authorizing a wide variety of bail alternatives, the legislature intended to
rationalize and make more flexible an admittedly archaic system.
Unfortunately, the courts have tended not to focus on the various
alternatives available to them with the result that many defendants are
incarcerated prior to trial for lack of collateral, even though the court may
have been inclined to and under the impression that release of such
defendants following[sic] the fixing of relatively low bail.38
While not wanting to interfere with the court’s discretion to fix bail in the amount and form it
chooses, the legislature made the change in the law to promote the use of bail forms that
were “less onerous” for defendants.39 The legislative history to this amendment includes
several letters supporting it (only the District Attorneys Association of the State of New York
opposed it). For example, the Legal Aid Society wrote in 1972 (in words that unfortunately
remain descriptive of bail setting in 2010):
The present procedure is much too casual. $500 bail is set frequently without
any thought of the form it should take or its effect on the defendant. A recent
study we conducted in New York City showed that 61 percent of the
defendants with $500 or less bail fail to make it. This seriously affects the
outcome of the defendant’s case.40
37
Preiser’s Practice Commentary to Criminal Procedure Law sec. 520.10, McKinney’s Consolidated Law of New York
(annotated), 2009.
38
Memorandum from John Haggerty, 45th District chairman, Committee on the Judiciary, to the Honorable Michael Whitman,
state senator, in support of L. 1972, c.784.
39
Preiser’s Practice Commentary to Criminal Procedure Law sec. 520.10, McKinney’s Consolidated Law of New York
(annotated), 2009.
40
Letter from James T. Prendergast, director, Research and Legislation Project, The Legal Aid Society, to Nelson A. Rockefeller,
governor of New York, June 1, 1972.

The Price of Freedom

16

Human Rights Watch does not have statistics to indicate how often judges set a bail amount
and fail to designate the form. Our interviews and observations, however, suggest that they
usually do specify the form and that the 1972 amendment has thus not had the intended effect.
New York City judges almost invariably designate bail in the most financially onerous forms,
cash and a secured surety bond, with the latter being set at a greater amount than the
former.41 For example, the bail may be set as $1,000 cash bail or a $5,000 secured bond.
Cash bail requires a defendant (drawing on her own resources or that of her family and
friends) to post the entire amount of bail in full and in cash. Most of those accused of
misdemeanors who flow through New York City’s courts do not have the financial resources
to post $1,000 cash bail or offer the court a secured bond.42 Commercial surety bonds are of
limited help: while commercial bail bondsmen do work in New York City, they rarely take on
bail involving $1,000 or less and their services are too expensive for low income offenders.43
It is not clear why New York City judges rarely set bail in the form of unsecured or partially
secured appearance or surety bonds. During our research several reasons were suggested.
First, it may be judges believe that defendants not appropriate for release on recognizance
(ROR) are such a flight risk that they should be incarcerated pending trial; not being able to
order pretrial detention of misdemeanor defendants directly, they secure the detention by
making bail unattainable. Second, judges may ignore alternative forms of bail because they
simply are not part of the common, traditional judicial practice. In the pressured
environment in which arraignment decisions are made, judges follow well-established
patterns of decision making and have scant incentive to do something different. Some
judges suggested to Human Rights Watch that many of their colleagues on the bench are
simply unaware of the options; if they did know them once, they have essentially forgotten
about them from disuse.
Third, some judges (as well as prosecutors) place responsibility on defense counsel for the
infrequency with which these alternative forms of bail are used. This is somewhat
disingenuous, as judges have the authority to choose the form of bail regardless of what

41
The statute is generally interpreted as prohibiting “cash only bail.” See, for example, New York State Commission on
Judicial Conduct, “Annual Report 2009,” pp. 20-21. Judicial education and training programs have stressed that “setting bail
in one form only, typically announcing ‘cash only,’ is contrary to CPL 520.10.” Whatever forms of bail a court designates, the
defendant also always has the option to post cash bail. New York Criminal Procedure Law, sec. 520.15.
42
New York law authorizes the use of secured bail bonds secured by personal and real property; the surety may be provided
by the defendant himself or someone other than the defendant.
43
Bond agents charge a 10 percent fee for the first $3,000, 8 percent for the next 7,000, and 6 percent for amounts over
$10,000. The fee is not refunded. Bondsmen also require collateral, typically cash, which is refunded unless bail is forfeited
for failure to appear. Mary T. Phillips, New York City Criminal Justice Agency, “Making Bail in New York City: Commercial Bonds
and Cash Bail,” March 2010.

17

Human Rights Watch | December 2010

defense counsel request. But it is true that defense counsel rarely request a judge to set bail
as an unsecured bond or a form of bail other than cash or secured bonds; it is not part of
their standard operating procedure. They believe judges will deny the request for a different
bail form were they to do so, and pressed for time as everyone is during arraignments, they
feel they cannot afford to make arguments for anything unusual that slows up the process,
particularly if the request will be futile. By not requesting judges to set bail in alternative
forms, however, defense counsel contribute to their disuse. If the defense bar consistently
pressed for unsecured bonds, for example, judicial practice might change.
Some professional participants in and observers of New York City’s criminal justice system
question whether unsecured bonds are appropriate for poor defendants given the likelihood
that their poverty would make it difficult for them to ever pay the bond amount should they fail
to show up in court. But this simply raises the inherent equitable dilemma of conditioning
pretrial freedom on financial resources. In addition, the fact that the bond amount may not be
paid does not leave the criminal justice system without recourse. Defendants who do not show
up for scheduled court appearances have violated court orders and judges will issue bench
warrants for their arrest. As a practical matter in New York City police do not typically seek to
arrest defendants who fail to show up in court (most of whom, as discussed below, will in any
event return to court voluntarily). But if persons with a bench warrant are subsequently
arrested for a new offense, the criminal record check will uncover the bench warrant and they
can then be prosecuted and held accountable for violating the court order.
Like many observers of or participants in the US criminal justice system, the American Bar
Association (ABA), as noted above, has long criticized financial bail, pointing out that it
undermines the integrity of the criminal justice system, is unfair to poor defendants, and is
ineffective in achieving key objectives of the release/detention decision.44 The ABA’s
standards for pretrial release provide that financial conditions should be used only when no
other conditions will provide reasonable assurance a defendant will appear for future court
appearance. If financial conditions are imposed, the court should first consider releasing the
defendant on an unsecured bond, and if that is deemed an insufficient condition of release,
bail should be set at the lowest level necessary to ensure the defendant’s appearance and
with regard to his financial ability. According to the ABA, when financial conditions are
imposed to secure a defendant’s appearance in court, they should not be set at an amount
that results in a defendant’s incarceration solely because he could not post the designated
amount.45
American Bar Association, ABA Criminal Justice Standards: Pretrial Release, 3rd ed. (Washington, DC: American Bar
Association, 2007), Introduction.
45
Ibid., Standard 10-1.4(e).
44

The Price of Freedom

18

After arrest, the accused who is poor must often await the
disposition of his case in jail because of his inability to raise
bail, while the accused who can afford bail is free to return to
his family and job….This is an example of justice denied, of a
man imprisoned for no reason other than his poverty. Think of
the needless waste – to the individual, the family, and the
community – every time a responsible person presumed by a
law to be innocent is kept in jail awaiting trial solely because he
is unable to raise bail money.46
—Supreme Court Justice Arthur Goldberg

46

Arthur Goldberg, “Equality and Governmental Action,” New York University Law Review, vol. 39 (1964), 218-222.

19

Human Rights Watch | December 2010

II. Who Suffers?:
Bail and Pretrial Detention of Low Income Defendants
For the Poor, Bail Means Jail
When bail is ordered at arraignment and defendants cannot post it (provide the court with the
designated bail amount), they are incarcerated.47 Pretrial incarceration, including the
incarceration of defendants who cannot post bail, is not a problem unique to New York.
According to the Bureau of Justice Statistics of the US Department of Justice, 62 percent of the
nation’s jail population consists of unconvicted detainees, and in the nation’s 75 largest
counties, 37 percent of felony defendants are held on bail until case disposition.48 The Pretrial
Justice Institute estimates it costs $9 billion annually to incarcerate defendants held on bail.49
It is rare to learn of a wealthy defendant who is unable to post bail and is consequently
incarcerated, even when the bail amounts are extremely high.50 In the past twelve months,
for example, John “Junior” Gotti, accused of racketeering, was able to post a $2 million bail,
and Sonny Franzese, accused of extortion, posted a $1 million bail.51 But inability to post bail
and subsequent incarceration is all too common for people of scant financial means. Even
when they are accused of low level offenses, such as smoking marijuana in public, the lack
of money to pay bail results in the pretrial loss of liberty.52
The incarceration of pretrial detainees has a significant impact on New York City jails.
According to the New York City Department of Correction, at any given moment, 39 percent of
New York City’s jail population consists of inmates who are in jail pretrial solely because

47

New York Criminal Procedure Law, sec. 510.40(3). “If the bail fixed is not posted, or is not approved after being posted, the
court must order that the [defendant] be committed to the custody of the sheriff.”
48
Todd D. Minton, Bureau of Justice Statistics, “Jail Inmates at Midyear 2009-Statistical Tables,” June 2010; Thomas H. Cohen
and Tracey Kyckelhahn, Bureau of Justice Statistics, “Felony Defendants in Large Urban Counties, 2006,” May 2010.
49
Human Rights Watch telephone interview with Timothy Murray, executive director, Pretrial Justice Institute, Washington,
D.C., August 11, 2010.
50
According to unpublished data provided to Human Rights Watch by the New York City Department of Correction, in 2009
there were 817 admissions of inmates held on bail of more than $100,000, including 34 on bail of $1 million or more. Data on
file at Human Rights Watch. We have no information concerning the inmates’ financial resources.
51
Bruce Golding, Doug Montero, Murray Weiss and Kati Cornell, “John ‘Junior’ Gotti case ends in mistrial, again,” New York
Post, December 1, 2009. John Marzulli, “Legendary mobster Sonny Franzese, 93, indicted for shaking down Hustler,
Penthouse strip joints,” Daily News, May 4, 2010.
52
For an excellent overview of the connection between poverty, bail and pretrial detention, see Jarrett Murphy, “Awaiting
Justice: The punishing price of NYC’s bail system,” City Limits, vol. 31 no. 3 (2007).

The Price of Freedom

20

they have not posted bail.53 In 2009, there were 98,980 total
admissions to the city’s jails, a little more than half of which
(51,047) were pretrial detainees incarcerated solely because
they had not posted bail. Pretrial detainees charged with
misdemeanors who had not posted bail constituted 22,846
admissions, or 23 percent of all admissions.54

Twenty-three year old John F., a
black man, was arrested in
Brooklyn for walking between
subway train cars. He had no
prior arrests, but had six prior
summonses for having an
unleashed dog. The judge set
bail at $500. John F. was lucky;
his mother was able to bail him
out.55

Among defendants arrested in 2008 on nonfelony charges
and given bail of $1,000 or less, only 13 percent were able to
post bail at arraignment (See Figure 1).56 Not surprisingly, the
higher the amount at which bail was set, the less likely the defendants would be able to post
it. In cases with bail set at $1,000, only 11.3 percent of defendants were able to post bail,
compared to 17.6 percent in cases in which bail was under $500.
Figure 1

Source: New York City Criminal Justice
Agency
Note: Includes cases of New York City
defendants arrested in 2008 on nonfelony
charges with bail set equal to or less than
$1,000. Cases with bail set at $1 excluded.

53

Unpublished data provided to Human Rights Watch by New York City Department of Correction. Data on file at Human Rights
Watch. The figure, based on a snapshot of the city jail population on October 18, 2010, excludes pretrial detainees for whom
bail has been set but who might be detained in custody even if they posted bail because of restraints on them. Such
restraints might include, for example, a remand placed on the defendant by other New York courts or warrants from other
jurisdictions of agencies, e.g. federal courts or Immigration and Customs Enforcement.
54
Ibid. The data does not reflect unique defendants, but admissions. If a defendant were sent to jail for pretrial detention
twice in a year, that would be counted as two admissions.
55
Human Rights Watch interview with Brooklyn Defender Services defense counsel for John F. (not his real name), New York
City, March 2, 2010.
56
Unpublished data provided to Human Rights Watch by New York City Criminal Justice Agency, June 2010. When felony and
nonfelony cases are combined, defendants in only 10 percent of the cases are able to post the designated bail at arraignment.
New York City Criminal Justice Agency, “2008 Annual Report,” p. 22, Exhibit 14.

21

Human Rights Watch | December 2010

Some of the nonfelony defendants sent to jail for pretrial detention because they cannot post
bail will subsequently be released before their case ends, because they were subsequently
able to put together the bail amount, for example, or because they are released on their own
recognizance after the prosecutor has failed to “convert” the misdemeanor complaint within
the prescribed time period (See Figure 2).57 But many will remain in pretrial detention for the
duration of their cases.58
Figure 2

Source: Mary T. Phillips, New York City Criminal
Justice Agency, “Bail, Detention, & Nonfelony
Case Outcomes,” Research Brief No. 14, May
2007.

Although some nonfelony defendants who are not able to post bail at arraignment are
released from pretrial detention within a day, most are held in jail for considerable periods

57

Mary T. Phillips, New York City Criminal Justice Agency, “Bail, Detention, & Nonfelony Case Outcomes,” Research Brief No.
14, May 2007, http://www.cjareports.org/reports/brief14.pdf (accessed October 26, 2010), p. 2. The dataset was taken from
the period October 1, 2003 to January 21, 2004. In figure 2, subsequent mandatory ROR includes cases in which defendants
benefited from New York Criminal Procedure Law sec. 170.70 which requires the release of a detained nonfelony defendant
after five days (six if over a Sunday) if the prosecutor has not replaced the complaint with formal charges signed by someone
with firsthand knowledge of the offense. In 45 percent of cases in Criminal Court with bail set at arraignment, the defendant
was not released before the case ended. New York City Criminal Justice Agency, “Annual Report 2008,” p. 24, Exhibit 15.
58
New York City Criminal Justice Agency, “Annual Report 2008,” p. 24, Exhibit 15. Low bail did not significantly increase the
likelihood of release: among defendants whose bail was set at $500 or less, 40 percent were not released prior to case
disposition.

The Price of Freedom

22

of time even when their bail is relatively low. As shown in Figure 3, the defendants spent two
to seven days in pretrial detention in almost half the cases of nonfelony defendants who
were not able to make bail of $1,000 or less at arraignment, and nearly one in four spent
more than 15 days.59 The mean length of pretrial detention was 15.7 days, the median 5 days.
In 3,848 cases, the defendants spent more than two weeks behind bars.
Figure 3

Source: New York City Criminal Justice Agency
Note: Includes cases of New York City defendants
arrested in 2008 on nonfelony charges with bail set
equal to or less than $1,000. Cases with bail met at
arraignment , or with bail set at $1 excluded.

The New York City Criminal Justice Agency (CJA) calculated the proportion of detained
nonfelony defendants who remained in pretrial detention at different times following their
arraignment.60 According to its calculations, 85 percent were in detention after one day,
nearly two-thirds (64 percent) remained incarcerated after the fourth day, and 43 percent
were still in detention at the end of the fifth day. By the 18th day, 25 percent remained in
detention, and 10 percent remained in detention after 50 days.
The greater the bail amount, the longer the pretrial detention. Among defendants arrested in
2008 on nonfelony charges and given bail of $1,000 or less, the mean length of pretrial

59

Percentage calculation excludes 110 cases in which length of detention was unknown.
The analysis was based on a dataset including all arrests from October 1, 2003 through January 31, 2004.The length of
detention is measured by the number of days the defendant spent in jail prior to release on bail or ROR or prior to the
disposition of the case. Phillips, “Bail, Detention, & Nonfelony Case Outcomes,” p. 3, Figure 2.

60

23

Human Rights Watch | December 2010

detention was 13.25 days in cases in which the bail was less than $500, compared to a
mean of 16.76 days when the bail was at $1,000 (See Figure 4).61
Figure 4

Source: New York City Criminal Justice
Agency
Note: Includes cases of New York City
defendants arrested in 2008 on nonfelony
charges with bail set equal to or less than
$1,000. Cases with bail met at
arraignment , or with bail set at $1
excluded.

Misdemeanors and lesser offenses by definition are not very serious offenses, and some of
them include conduct that might be (and in some other jurisdictions is) discouraged without
recourse to criminal arrests, such as smoking marijuana in public, turnstile jumping, or
trespassing on public housing property. Nevertheless, in New York City, persons charged
even with such petty offenses will be held in pretrial detention if bail is set and they cannot
post it at arraignment. As shown in Table 2, the average length of detention for defendants
accused of possession of marijuana in public view who were not able to post bail of $1,000
or less was 10.4 days. For defendants arrested on theft of services charges—typically
turnstile jumping62—the mean length of pretrial detention was 15.35 days. Defendants
arrested for physically injurious nonfelony offenses (assault, domestic violence) and for
property crimes (e.g. shoplifting) had the longest average periods of pretrial detention, 17.25
and 19.01 days, respectively.
61

See also, Phillips, “Bail, Detention, & Nonfelony Case Outcomes,” p. 4. The median number of days in detention increased
from 4 days for cases with bail of $500 or less to 9 days for cases with bail above $5,000.
62
Although there are no definitive statistics, it is generally agreed that most defendants prosecuted for misdemeanor theft of
services are accused of having used the subway without paying the fare, by “jumping” over or ducking under the subway turnstile.

The Price of Freedom

24

Table

Made bail at arraignment I
Released within 1 day
Detained 2-7 days

25

Detained 8-14 days
Detained 15+ days
Number of cases

Human Rights Watch | December 2010

Mean days detained

2

".

7·0%

I
I
I

376

'4'

17·7%

17·3%

10.8%

8·5%

53
7·2%

339

n

32·9%

17·7%

335
32.6%

25·1%

,
5·6%

'9'

1,03B

49'

3'3

",

20.6%

20·5%

14·3%

11·4%

15,]%

2,290

6',

1,B37

1,469

1.152

'99

57·7%

43.6%

36 .3%

42.2%

40.6%

4 0 .7%

21.6%

37.2%

46·7%

'9'

9.

,,6

4"

30 3

,,6

'9

5·5%

11·5%

10·7%

15·B%

4· B%

35
6.1%

7.8%

20.6%

'"

,6
'49

"

.',

24-4%

2.488
13. 1%
3.°36
16.0%
B.075
42-4%
1,5Bo

7·5%

6·7%

"4

,6,

1,034

73'

.,.

19·8%

20·4%

21.2%

28.8%

],967

11·4%
1,411

5,059

2,838

73'

1.029

4"

100%

100%

100%

3.4 8 4
100%

100%

100%

100%

100%

15·35

10.62

,

14·93

,

8.82

15·7

333

,8.

"9

",

48'

9°

19. 137

,

14. 28

Median days detained
Maximum days detained

'74

8.0%

3"

I

"0

3"

Length of detention
unknown
Total
3.9 81
Source: New York City Criminal Justice Agency

'.

,

,

,

10.42

17·25

15·97

'9'

,6o

37°

,
,.,

19·01

'"

,

84

64

'4

3,848

8.2%

13·9%

15·6%

20.2%

'0

19.027

100%

100%

,
1,418

8.J%

,

,

"0

2.858

Note: InclUdes cases of New York City defendanlS a.rested in 2008 on nonfelony chafl\es with bail set equal to or less than $1,000. Cases with bail set at $1 excluded.

Sub Rosa Preventive Detention
When bail is set at amounts that make it impossible for defendants to obtain pretrial release,
the bail decision operates to secure a “form of sub rosa preventive detention.”63 No doubt
some judges set bail at unaffordable levels not realizing they are doing so. But our research
suggests that judges also set bail assuming—and sometimes intending—that pretrial
detention will be the result. As one judge told Human Rights Watch, if judges see a
defendant who “is a junkie, with no job and no home, [they] set bail.”64 That is, in some
cases judges believe jail is the only way to secure a defendant’s future appearance in court;
since the statute does not permit sending the defendant to jail directly, that goal is
accomplished indirectly through unaffordable bail. Judges know that many, if not most,
misdemeanor defendants are not going to be able to post bail and that incarceration will be
the consequence of their bail decision. Judge Emily Jane Goodman told Human Rights Watch:
“Bail equals jail as a practical matter. Everyone knows that’s what’s going to happen. Judges
know that $1,000 will do the trick of keeping them in jail.”65 As another New York City judge
told us, “setting bail of $500 is the equivalent of remand [ordering detention] for most
people, even though the law doesn’t permit remand in misdemeanor cases.”66
New York law contains no provision such as that recommended by the American Bar
Association to prevent bail from becoming an indirect and unacknowledged mechanism for
pretrial detention. Standard 10-1.4(e) of the ABA Pretrial Release Standards states: “The
judicial officer should not impose a financial condition of release that results in the pretrial
detention of a defendant solely due to the defendant’s inability to pay.”67 As the ABA
commentary to this standard explains:
[The] intent behind this limitation is to ensure that financial bail serves only
as an incentive for released defendants to appear in court and not as a
subterfuge for detaining defendants. Detention should only result from an
explicit detention decision, at a hearing specifically designed to decide that
question, not from the defendant’s inability to afford the assigned bail.68

63

National Association of Pretrial Services Agencies, “Standards on Pretrial Release, Third Edition,” October 2004, Commentary
to Standard 1.5 B.
64
Human Rights Watch interview with Judge Michael J. Obus, Supreme Court, New York County, New York, New York, April 3, 2010.
65
Human Rights Watch interview with Judge Emily Jane Goodman, Supreme Court, New York county, New York, New York, May 26,
2010.
66
Human Rights Watch interview with New York City judge (name withheld), New York, New York, June 18, 2010.
67
American Bar Association, ABA Criminal Justice Standards: Pretrial Release, Standard 10-1.4(e), Conditions of release.
68
Ibid., Commentary to Standard 10-1.4(e), p. 44. Federal law contains just such a limitation, see, 18.U.S.C. §3142(c)(2) ( “the
judicial officer may not impose a financial condition that results in the pretrial detention of the person.”).

The Price of Freedom

26

Federal law and the law of the District of Columbia contain provisions similar to that called
for by the ABA.69
Human Rights Watch agrees with the ABA that a decision to detain a defendant prior to trial
should only be made at a hearing in which the government proves by clear and convincing
evidence that no condition or combination of conditions of release will provide reasonable
assurance that the defendant will appear for court proceedings or protect the safety of the
community or any person.70 Unfortunately, in New York City arraignments, there is no need
for the prosecutor to present convincing evidence concerning the necessity of detention.
Such limited argument as takes place during arraignment is ostensibly focused on financial
conditions of release, but it is a largely a charade because everyone knows detention is the
most likely outcome if bail is not set at an amount that the defendant has acknowledged he
or his family can pay.
Human Rights Watch also agrees with the ABA that eligibility for pretrial detention should be
limited to defendants charged with violent or dangerous crimes, or charged with serious
offenses who pose a substantial risk of failure to appear in court.71 Defendants charged with
non-serious offenses should not be eligible for detention before trial unless there is a
substantial risk the defendant “will obstruct or attempt to obstruct justice, or threaten, injure
or intimidate a prospective witness or juror.”72 The ABA appropriately calls for a degree of
proportionality between the seriousness of loss of liberty and the seriousness of the risk
posed by the defendant. As discussed below, there are ways short of detention to mitigate
the risk that a nonfelony defendant will not appear for court proceedings.

Pretrial Punishment
For many people charged with misdemeanor crimes in New York City, the pretrial “process is
the punishment.”73 As Rick Jones, executive director of Neighborhood Defender Service of
69
“Judicial officer may not impose a financial condition that results in the pretrial detention of the person.” 18 U.S.C. sec.
3142(c)(2). Judicial officer “may impose such a financial condition to reasonably assure the defendant’s presence at all court
proceedings that does not result in the preventive detention f the person.” D.C. Code Ann. sec. 23-1321(c)(1)(3).
70
American Bar Association, ABA Criminal Justice Standards: Pretrial Release, Standard 10-5.8(a). See also, National
Association of Pretrial Services Agencies, “Standards on Pretrial Release,” Commentary to Standard 2.8(a). The “clear and
convincing evidence” required is a deliberately high standard that “reflects the high value placed on individual liberty in the
American legal system.”
71
The definitions of dangerous, violent, and serious offense for purposes of this standard is left to individual jurisdictions to
determine. The ABA commentary suggests a defendant charged with criminal fraud who has access to large amounts of money
as an example of a defendant charged with a “serious” offense who may pose a substantial risk of failing to appear in court.
72
American Bar Association, ABA Criminal Justice Standards: Pretrial Release, Standard 10-59(a). See also National
Association of Pretrial Services Agencies, “Standards on Pretrial Release,” Standard 2.9(a).
73
Thirty years ago, Malcolm Feeley observed the functioning and impact of lower courts in New Haven handling misdemeanor
defendants. The thesis and title of his justly famous book, The Process is the Punishment, is all too apt a description of the

27

Human Rights Watch | December 2010

Harlem, points out: “It is a legal fiction that the system will vigorously prosecute those who
are arrested. Prosecutors know they cannot and will not prosecute all the cases. For many
defendants, arrest and pretrial detention is the punishment.”74 In the eyes of the law, pretrial
detention is not punishment. But for those who are in jail because they cannot post bail, the
experience scarcely differs from being in jail because of a sentence of guilt. They endure the
same loss of liberty, the same stress and turmoil of incarceration, and the same loss of
income and ability to care for their families.
The judges we interviewed denied any punitive motivation in their bail decisions.
Nevertheless, defense attorneys we interviewed believe that at least in some cases judges
intend pretrial detention to serve as de facto “preemptive punishment “ for misdemeanor
offenders who—legal niceties about presumed innocence aside—are often assumed to have
done something wrong to have landed in court, especially if they have a criminal record.75 In
addition, as one New York City judge told us, “many judges have what must be considered
contempt for the street people parading before them…[they have] no problem sending them
to jail—better than having them hanging around the neighborhoods.”76
Some criminal justice practitioners insist that the mere fact of a prior criminal record
predisposes judges to set bail, even though prior convictions do not predict failure to appear
(see discussion below). Certainly, as shown in Table 3, about three-quarters of those who
were given bail at or under $1,000 had prior convictions. We do not have the data that would
tell us, however, whether those defendants with prior convictions also had backgrounds
(such as prior bench warrants or weak community ties) that have been demonstrated to
predict failure to appear.

Criminal Court in New York City today. Malcolm M. Feeley, The Process is the Punishment: Handling Cases in a Lower Criminal
Court (New York: Russell Sage Foundation, 1979). The book was republished with a new foreword and preface in 1992.
74

Human Rights Watch interview with Rick Jones, executive director, Neighborhood Defender Service, New York, New York,
April 20, 2010.
75
Many of those arrested for misdemeanors do have records: 39.6 percent of New York City arrestees in 2009 on nonfelony
charges have prior felony convictions, 52.6 percent have prior misdemeanor convictions, and 36.8 percent have three or more
prior misdemeanor convictions. Unpublished data for fiscal year 2009 provided to Human Rights Watch by New York State
Division of Criminal Justice Services. Data on file at Human Rights Watch.
76
Human Rights Watch interview with New York City judge (name witheld), New York, New York, June 7, 2010.

The Price of Freedom

28

Table 3
Criminal History of Nonfelony Defendants Released on Recognizance (ROR) or with
Bail of $1,000 or Less

Defendant’s Criminal History
No prior convictions

Percentage
of ROR

ROR

Bail at or
under
$1,000

Total

52,943

58.4%

4,579

57,522

Prior misdemeanor conviction only

9,042

10.0%

4,448

13,490

Prior felony conviction only

3,854

4.3%

939

4,793

Both misdemeanor and felony
priors
data not available

9,123

10.1%

8,203

17,326

15,643
90,605

17.3%

968
19,137

16,611
109,742

Total

100.0%

Source: New York City Criminal Justice Agency
Note: Includes cases of New York City defendants arrested in 2008 on nonfelony charges with bail set equal to
or less than $1,000. Cases with bail set at $1 excluded.

The punishing experience of pretrial detention is endured by thousands of nonfelony
defendants who are never convicted. According to CJA:
In 22 percent of nonfelony cases with a detained defendant, the defendant
was ultimately acquitted or the case was dismissed. In an additional 24
percent of cases with detention, the defendant was convicted but the
sentence did not include any jail (not even time served). This means that
nearly half of detained [nonfelony] defendants served time in jail only
because they were unable to post bail—often a very small amount.77
Among the defendants arrested in 2008 on nonfelony charges and given bail under
$1,000, 23.8 percent were not convicted.78
Pretrial detention is also disproportionate to the punishment that can be expected for minor
crimes. As shown in Table 479, more than half of the persons arrested on misdemeanor

77

Phillips, “Pretrial Detention and Case Outcomes, Part 1: Nonfelony Cases,” p. 59.
Unpublished data provided to Human Rights Watch by the New York City Criminal Justice Agency, June 2010.Data on file at
Human Rights Watch.
79
The data in Table 4 reflects the case outcomes in 2009 of arrests in which most serious charge was a misdemeanor,
regardless of when the arrest occurred. Convictions include convictions for offenses other than those charged at arrest, e.g., a
misdemeanor arrest may result in a conviction for a violation. Indeed, 56.1 percent of misdemeanor arrests lead to convictions
for non-criminal offenses. For example, a quite common scenario is that a person arrested for a misdemeanor of criminal
possession of marijuana in public view, pleads guilty to the non-criminal offense of disorderly conduct. Only 490
misdemeanor arrests (0.4 percent) ended with felony convictions. New York State Division of Criminal Justice Services,
78

29

Human Rights Watch | December 2010

charges and subsequently convicted are sentenced to fines, probation, or conditional
discharge.80 The time in pretrial detention (as well as in police lockup pre-arraignment) is
quite literally punishment paid in advance for the additional 19.5 percent of such defendants
who received sentences of time served.81
Table 4
Dispositions of Misdemeanor Arrests in 2009
Total
N

Total Convictions
Sentences to:

%

128,351

100.0%

177

.1%

Jail

26,475

20.6%

Time Served

25,009

19.5%

Jail with Probation

211

.2%

Probation

724

.6%

Prison*

Fine

19,631

15.3%

Conditional Discharge

55,746

43.4%

378

.3%

Other

Source: New York State Division of Criminal Justice Services, New York City Adult Arrests
Disposed—Misdemeanors, http://criminaljustice.state.ny.us/crimnet/ojsa/dispos/nyc.pdf
(accessed November 15, 2010).
*Prison sentences given for convictions in cases in which the charge was increased from
misdemeanor to felony following arrest.

When a jail sentence is imposed, some misdemeanants receive sentences shorter than the
length of time in pretrial detention. The median length of pretrial incarceration for
misdemeanor defendants arrested in 2008 is five days, the average is 1582; yet according to
the New York State Division of Criminal Justice Services, in 48 percent of cases in which
people arrested on misdemeanor charges are convicted and sentenced to jail, the sentence
is less than 15 days; in 9 percent of the cases, it is less than five days. 83
Disposition of Adult Arrests 2009, http://criminaljustice.state.ny.us/crimnet/ojsa/dispos/nyc.pdf (accessed November 1,
2010).
80
New York Penal Law, sec. 65.05. A court may impose a sentence of conditional discharge for an offense if it is “of the
opinion that neither the public interest nor the ends of justice would be served by a sentence of imprisonment and that
probation supervision is not appropriate.” Conditional discharge sentences, which can last for one year for misdemeanor
convictions, typically require the defendant to comply with specified treatment programs or engage in community service.
81
The percentage of misdemeanor arrests ending with sentences to time served varies somewhat according to the arrest
offense. According to unpublished data for fiscal year 2009 provided to Human Rights Watch by the New York State Division of
Criminal Justice Services, for example, 27.8 percent of convictions for criminal possession of marijuana resulted in sentences
to time served.
82
See Table 2 above. Among the nonfelony cases in Table 2, the great preponderance are misdemeanors. As discussed above,
the cases of most lesser offenses end at arraignment.
83
Unpublished data on the sentences for NYC misdemeanor defendants in fiscal year 2009 provided to Human Rights Watch
by New York State Division of Criminal Justice Services. Data on file at Human Rights Watch.

The Price of Freedom

30

Plea Bargains
The desire to end the ordeal of the pretrial process—
particularly pretrial detention—pressures defendants to plead
guilty and give up their right to trial.84 According to the New
York State Division of Criminal Justice Services, about 60
percent of all misdemeanor arrests result in guilty pleas.85 In
99.6 percent of the cases in which misdemeanor arrestees are
convicted, the convictions are secured through guilty pleas.86

The unfortunate reality is that
many clients in poor
communities of color like the
Bronx are too poor to post
even modest bail of $500,
$1,000 or $1,500. Forced to
remain behind bars, their lives
destabilize: They lose their
jobs; their physical and mental
health deteriorates; and their
families’ social and economic
networks fall apart. In the face
of these consequences and
under the threat of continued
incarceration, many
defendants, whether guilty or
innocent, plead guilty simply
to get out of jail.
—“The Bronx Freedom Fund
Report,” Fall 2009, on file with
Human Rights Watch.

Defense attorneys believe prosecutors ask for bail, including
in cases they know are likely to be dismissed—as 34.5
percent of misdemeanor arrest cases are—simply to get
pleas from defendants who will not want to languish in
pretrial detention. Prosecutors deny this. But absent
considerations of plea bargaining, it is difficult to fathom
why prosecutors seek bail in the countless cases of petty
crime which are extremely unlikely ever to be tried. Timothy
Murray, executive director of the Pretrial Justice Institute,
believes that prosecutors across the country, including in
New York City, truly believe they are doing the best that can
be done when they make their bail requests. “But woven into
their mindset is the idea you should somehow ‘pay’ from the moment of arrest, that you owe
the system something just by virtue of being accused…because they implicitly believe—and
must believe—that people who are arrested are guilty.”87
Judges, prosecutors, and defense counsel all know that defendants at arraignments who
face the prospect of pretrial detention because they cannot post bail are likely to agree to
plea bargains. If a defendant does not accept the plea, she can spend weeks, months, and
84

Hardly any misdemeanor cases are actually tried. For example, in 2009, there were only 425 misdemeanor trials in the
Criminal Court citywide. Criminal Court of the City of New York, “Annual Report 2009,” p. 52. Although 39 percent of the 425
defendants who went to trial were acquitted, it took over a year on average for them to be vindicated in their contention of
innocence. Ibid., p. 53.
85
Of 205,068 New York City misdemeanor arrests in which the criminal proceedings ended in fiscal year 2009, 123,670
resulted in a conviction secured through a guilty plea. Unpublished data on the disposition of NYC misdemeanor arrests in
fiscal year 2009 provided to Human Rights Watch by New York State Division of Criminal Justice Services. Data on file at
Human Rights Watch.
86
In addition to convictions, 34.5 percent of misdemeanor arrests in which the cases ended in fiscal year 2009 resulted in
dismissals, and 10 percent resulted in other dispositions, e.g., transfers to other jurisdictions or prosecution declined. New
York State Division of Criminal Justice Services, Disposition of Adult Arrests 2009,
http://criminaljustice.state.ny.us/crimnet/ojsa/dispos/nyc.pdf (accessed November 1, 2010).
87
Human Rights Watch interview with Timothy Murray, Washington, D.C., May 4, 2010

31

Human Rights Watch | December 2010

even longer behind bars before the case ends. She can go home immediately, however, if
she pleads guilty and receives the typical sentences of conditional discharge, a fine, or time
served. Defense attorney Mark Loudon-Brown explained the choice facing many defendants:
“Do you want to plea and go home, or do you want to fight the case for a year?”88 Or as
defense attorney Leah Horowitz said: “If you can’t [post bail], the easiest way to get out of
jail is to take a plea…People don’t want to sit in jail to get their day in court and be
vindicated. They would rather have their freedom.”89
Even defendants who are not in pretrial detention will plead guilty. As one judge explained
to Human Rights Watch, “Even innocent people plead just to get it over with. They do not
want to come back for court appearances that will continue for months, they want to move
on with their lives.”90
While the plea may prevent or end pretrial detention, it also may get him a criminal record.91
A New York City judge told Human Rights Watch the criminal justice system “is a selffulfilling system; defendants have to plea, and end up with a record” which brands them for
life as a criminal and influences judges in a future case if arrested again.92 The convictions
can also carry numerous collateral consequences, such as precluding the individuals from
obtaining public housing.93
Whether or not a defendant is in pretrial detention affects the terms of the plea bargain. As
Robin Steinberg, executive director of The Bronx Defenders, told Human Rights Watch, “if
you are in jail [pretrial] your ultimate sentence will be higher because your bargaining
position is weaker.”94 Conversely, when nonfelony defendants are free pretrial—either
because of release on recognizance or bail—they have little incentive to accept a plea that
involves a sentence of additional jail time. As one New York City prosecutor explained to
Human Rights Watch, “If a defendant is not in jail, he is not going to plea to jail time…Time is

88

Human Rights Watch interview with Mark Loudon Brown, The Bronx Defenders, New York City, March 9, 2010.
Human Rights Watch interview with Leah Horowtiz, The Bronx Defenders, New York City, March 9, 2010.
90
Human Rights Watch interview with New York City judge (name withheld), New York City, March 8, 2010. Defendants may be
called to court five to fifteen times in a case before it is finally disposed of, and each court appearance may consume an entire
day as they must arrive in the morning and their case may not be called until the afternoon.
91
Defendants charged with misdemeanors may be offered a plea to a violation, which is not a crime. For example, a defendant
charged with possession of marijuana in public view, a misdemeanor, may take a plea to disorderly conduct, which is a
violation. But that conviction is still on his record, and carries collateral consequences.
92
Human Rights Watch interview with New York City judge (name withheld), New York City, June 18, 2010.
93
Human Rights Watch, No Second Chance: People with Criminal Records Denied Access to Public Housing (New York: Human
Rights Watch, 2004); Legal Action Center, “After Prison: Roadblocks to Reentry,” 2004, http://www.lac.org/roadblocks-toreentry; Legal Action Center, “After Prison Report: 2009 Update,” 2009, http://www.lac.org/roadblocks-to-reentry.
94
Human Rights Watch interview with Robin Steinberg, executive director, The Bronx Defenders, New York City, January 12,
2010.
89

The Price of Freedom

32

defense’s friend when the defendant is out of jail. Cases weaken over time, witnesses fall
away, and evidence gets stale. The defendant can hope the case will fall apart and will hold
off agreeing to plea.”95 Prosecutors make plea offers based on the charges, the strength of
the evidence, and the defendant’s record. “But you negotiate from strength. If you know the
defendant is in jail, you have more strength to drive a better bargain.”96
The desire to secure a plea can also influence prosecutor bail requests. Another prosecutor
acknowledged: “You assume the defendant will not be able to make bail. You want them
behind bars. Because chances of disposing of the case is easier. Defendants are more likely
to plead guilty once in jail. They have a strong incentive to plead. If not in jail they will never
agree to a plea that requires jail time.”97 Timothy Rountree, attorney-in-charge at The Legal
Aid Society in Queens, described what he considers a common scenario: “Prosecutor will
make a plea offer and defendant rejects it. Judge sets bail to force a person to rethink the
offer. If defendant cannot make bail, prosecutor’s offer will look better than sitting in
jail…Prosecutor requests bail to coerce plea.”98
The prevalence of plea bargains, many triggered by the desire to avoid or end pretrial
detention, helps explain the connection between pretrial detention and adverse case
outcomes that has long been reported.99 In a sophisticated and exhaustive analysis of the
link between pretrial detention and case dispositions,100 the New York City Criminal Justice
Agency found that cases with a nonfelony defendant who was released until the case ended
had a 50 percent conviction rate and cases with a nonfelony defendant who was detained to
disposition had a 92 percent conviction rate. The CJA found that “pretrial detention had an
effect on conviction after controlling statistically for the number and severity of arrest
charges, the offense type of the arraignment charge, the defendant’s criminal history,
demographic characteristics, borough, and length of case processing, among other
factors.”101 The CJA concluded, that:

95

Human Rights Watch interview with New York City prosecutor (name withheld), New York City, June 21, 2010.
Ibid.
97
Human Rights Watch interview with New York City prosecutor (name withheld), New York City, April 16, 2010.
98
Human Rights Watch interview with Timothy B. Rountree, attorney-in-charge, Criminal Defense Practice, The Legal Aid
Society, New York City, March 23, 2010.
99
For an overview of the research addressing the link between pretrial detention and case disposition, see Phillips, “Pretrial
Detention and Case Outcomes, Part 1: Nonfelony Cases,” pp. 2-7.
100
Phillips, “Pretrial Detention and Case Outcomes, Part 1: Nonfelony Cases.” The study, using a dataset of arrests in New York
City from 2003 and 2004, sought to determine whether 1) detention itself caused the outcomes (because jailed defendants are
likely to plead guilty to gain release and are at a disadvantage participating in their defense); or 2) judges adjusted their
release and bail decisions according to the probable outcome (setting high bail to keep defendants in custody who are likely
to be convicted and given jail or prison sentences).
101
Phillips, “Bail, Detention, & Nonfelony Case Outcomes,” p. 5. CJA’s research also showed that detention has a small
independent effect on the likelihood of incarceration following conviction, but only has a trivial effect on sentence length.
96

33

Human Rights Watch | December 2010

[The research] supports the hypothesis that pretrial detention has an adverse effect on case
outcomes, especially the likelihood of conviction. The hypothesis is impossible to prove
because some factor or factors for which data are unavailable—the strength of the evidence,
for instance—could be the reason for both higher bail (resulting in detention) and for the
conviction. However, we were able to control for a wide range of case and defendant
characteristics. None, either singly or in combination, completely explained away the
relationship between detention and likelihood of conviction in nonfelony cases. 102

102

Phillips, “Bail, Detention, & Nonfelony Case Outcomes,” p.7.

The Price of Freedom

34

III. Factors Influencing Judges’ Bail Decisions
The Exercise of Judicial Discretion
Judicial discretion regarding release and bail decisions is vast. As a judge explained to
Human Rights Watch, “there are no written guidelines for bail; no one tells a judge what to
do. Nothing is written. It’s a matter of practice.”103 Judges can and do make profoundly
different decisions regarding release and bail for similarly situated defendants. If faced with
the identical defendant, one judge might set bail at $500, another judge might set bail of
$1,500, and another might release the defendant on his own recognizance. The decisions
regarding whether to release a defendant on his own recognizance or to set bail, and if bail
is set, at what level, are supposed to be made solely with an eye to ensuring the defendant’s
appearance in court. But it is difficult to discern how judges in fact assess the risk of failure
to appear.
Understanding why judges exercise their discretion as they do is difficult because while the
outcomes of their release and bail decisions are public (announced in open court), the
reasons for those decisions are not. Judges are not required to explain orally at the
arraignment or in writing the basis for their decisions.104 They are not, for example, required
to explain why they have chosen bail of $500 instead of $150, or why they require cash bail
instead of an unsecured appearance bond. The lack of transparency, and hence of
accountability, in release and bail decisions is particularly troubling when the potential
consequence—loss of liberty—is so serious.
Judges we interviewed believe they and most of their colleagues are fair and reasonable in
their arraignment decisions.105 They suggest it is difficult, if not impossible, to identify patterns
in the decisions because each case is unique, presenting endlessly varied individual factors.
They also acknowledge the role of personal background, experience, predilections, and
temperament in arraignment decisions. Judges who are former prosecutors may, for example,
103

Human Rights Watch interview with New York City judge (name withheld), New York City, March 18, 2010.
New York practice thus differs from that recommended by the American Bar Association that when release on personal
recognizance is denied, the judicial officer “should include in the record a statement, written or oral, of the reasons for this
decision.” American Bar Association, ABA Criminal Justice Standards: Pretrial Release, Standard 10-5(c). The Commentary to
the Standard states the “purposes of this requirement are to encourage rational and fair decision-making, foster
accountability for release/detention decisions made, and provide a record for review of the decision at later stages of the
case.” According to one judge, given the volume of cases, as a practical matter arraignment judges would not have the time to
spell out their reasoning in release and bail decisions. Interview with New York City Judge (name withheld), New York City,
August 30, 2010.
105
We were disappointed by the number of judges who refused to be interviewed, typically citing concerns about the propriety
of commenting, even generally, about judicial bail practices. Among those who did agree to be interviewed, all but two
requested that the interviews be anonymous.
104

35

Human Rights Watch | December 2010

approach arraignment with a different perspective on defendants than judges with no criminal
justice experience. Defense attorneys say certain judges have reputations for being more likely
to release defendants on their own recognizance than others, or for setting high bail. Judges’
decisions can become quite idiosyncratic: one judge, for example, always releases women
charged with prostitution; another is particularly prone to setting bail in domestic violence
cases; another will usually set bail when the offender is charged with breaking into a car.
Research by the CJA confirms that the release decision and amount of bail depends on the
individual judges, and cannot be predicted simply on the basis of the charge or the particulars
of a defendant’s criminal history.106
Judges are able to move quickly through the scores of cases in each shift because they know
how they are likely to rule within seconds of reviewing the documents in front of them and
identifying what for them are the salient features. They set bail in amounts that are familiar
and entrenched, and not closely tailored to the individual’s particular resources. Bail
amounts tend to fall into categories, e.g., $500, $1,000, or $1,500. That is, it is rare for a
judge to set bail at, say, $125 or $325 (let alone $124 or $322).
Although each case is unique, they also fall into familiar categories and over time judges
establish fairly predictable responses to cases in those categories. In fact, judges, prosecutors,
and defense attorneys interviewed by Human Rights Watch agreed that there is typically a
shared understanding, based on experience with countless cases, regarding the likely release
and bail outcome in any given case. It would be considered unusual, for example, for a
defendant not be released on her own recognizance if she were picked up smoking marijuana
in a park, had no prior arrests, had a full time job, and had lived in the community all her life.
In the cohort of defendants arrested in 2008 on nonfelony charges who either received ROR or
bail under $1,000, judges in fact released defendants on their own recognizance in 98 percent
of the cases in which the CJA had recommended release because the defendant had strong
community ties and no prior bench warrants, there were no prior convictions, and there was no
open (or pending) case.107 Most cases, however, are not so clear cut, and hence the great
variability among judges as they respond to the factors in each case.

106
The CJA concluded that “it was impossible to characterize judges overall as ‘lenient’ or ‘strict.’ Judges with higher than
average ROR rates for misdemeanor defendants did not necessarily have the same tendency in felony cases, and leniency in
setting low bail amounts did not carry over to ROR.” Mary T. Phillips, New York City Criminal Justice Agency, “Factors
Influencing Release and Bail Decisions in New York City: Part 3. Cross-Borough Analysis,” July 2004, p. 39. Using data from
Manhattan and Brooklyn, the CJA sought to identify and assess the specific factors influencing judicial release and bail
decisions.
107
Unpublished data provided to Human Rights Watch by New York City Criminal Justice Agency, June 2010. Data on file at
Human Rights Watch.

The Price of Freedom

36

Defense counsel have learned from experience to foresee what particular judges are likely to
do in a given case. Human Rights Watch watched defense counsel meet with over a dozen
defendants before arraignment and each time the lawyers explained what they thought the
plea offer from the prosecutor would be, whether the judge would grant release on
recognizance or set bail, and what the likely bail amount would be. In the subsequent
arraignments, the defense counsel’s predictions proved remarkably accurate. Similarly,
prosecutors have a sense of whether a defendant is likely to be released or face bail, and if
bail is set, what the basic range is within which the judge is likely to operate. They adjust
their bail requests accordingly, knowing if they ask for bail amounts that are wildly out of
line, they may annoy the judge and lose credibility.
In trying to understand release and bail decisions, one cannot ignore the deep impact of the
objective strictures under which the Criminal Court operates. The volume of cases presented
for arraignment, two-thirds of them for misdemeanors, has grown markedly and the number
of prosecutors, defense attorneys, judges, and court personnel has not kept pace. As a
result, arraignments are brief and hurried affairs, typically lasting no more than a few
minutes. During that time, the judge hears from prosecutors and defense attorneys
regarding disposition of the case (e.g., dismissal or a plea bargain) and makes decisions
regarding release or bail and whether other orders are required (for example, orders of
protection in domestic violence cases). The sheer number of cases forces all the
professionals involved—judges, prosecutors, defense attorneys—to value speed and
efficiency over individualized justice, giving arraignments a “conveyor belt quality.”108 A
substantial proportion of the judges interviewed by Human Rights Watch said they felt a
strong institutional pressure to move cases rapidly; if more time were taken in each case,
the courts might not be able to arraign every defendant within 24 hours, as they try to do.
Judges—as well as prosecutors and defense counsel—do the best they can under difficult
circumstances. As New York City Judge Obus told Human Rights Watch, the “hardest thing is
to sit in arraignments and make decisions regarding bail after just a couple of minutes.”109
Careful consideration of the nature and significance of factors enumerated in the bail statute
in the thousands of run-of-the-mill misdemeanor cases is not possible.
Judges have limited information available to them at arraignment: the defendant’s criminal
record (rap sheet), the complaint, the results of the pre-arraignment interview by the New
York Criminal Justice Agency and its release recommendation based on its assessment of
Martha Rayner, “Conference report: New York City's Criminal Courts: Are We Achieving Justice?” Fordham Urban Law Journal,
vol. 31 issue 4 (2004), p. 1055.
109
Human Rights Watch interview with Judge Michael J. Obus, New York City, April 13, 2010.
108

37

Human Rights Watch | December 2010

the statistical likelihood the defendant will fail to appear at future court proceedings and any
additional information that may be provided by defense counsel and prosecutors during the
brief hearing. Defense counsel may not have much to add as they typically meet with their
clients for the first time just moments before standing up before the judge; prosecutors
similarly may be reviewing their files for the first time moments before the case is called.
Some of the information that might be relevant to a bail decision may not be available, for
example, regarding the defendant’s mental health history. Information regarding
defendants’ financial resources is limited to what the defendant says, and is rarely verified.
Although the CJA will attempt before arraignment to verify a defendant’s residence and
employment, this is not always possible.
Lack of information alone, however, cannot be blamed for all the setting bail in a form and at
amounts that defendants cannot provide. Judges usually have enough information in
misdemeanor cases to make an educated guess: they can safely assume, for example, that
someone unemployed and accused of stealing food will not be able to make bail set as
$1,000 cash or a $5,000 bond. It is also striking, as noted above, that judges typically set
bail in fixed amounts, e.g., $500 or $1,000, which suggests the judges have not tried to
tailor bail amounts to defendants’ actual resources. Although it happens, it is rare for a
judge to discuss finances with a defendant and then set bail at, for example, $180, because
that is an amount he could afford and can get to the court immediately to prevent detention.
Some judges may recognize a defendant is poor and will not set bail at, say, $10,000, but
they may not realize—or they may be indifferent to—the difficulty poor defendants have in
coming up with $500 bail , an amount that may to the judges seem quite affordable. As
attorney Leah Horowitz of the Bronx Defenders explained, “people struggling to find and
keep employment, housing, simply don’t have $500 in available funds…even working
families live paycheck to paycheck and don’t have $500 free with which to post bail.”110 A
New York City judge told Human Rights Watch that he realized that even $250 or $500 bail
could be too much for some defendants. As he pointed out, “they can’t just go to an ATM
machine and get the money.”111 For low income or indigent defendants, there is a huge
difference in affordability between bail of $250 and $1,000, yet judges are more likely to set
bail at the latter than the former, and they almost never use unsecured bonds that do not
condition pretrial freedom on financial resources. As shown in Table 1, among the 19,137
cases of defendants arrested in 2008 on nonfelony charges with bail set at $1,000 or less,

110
111

Human Rights Watch interview with Leah Horowtiz, New York City, Feburary 1, 2010.
Human Rights Watch interview with New York City judge (name withheld), New York City, March 8, 2010.

The Price of Freedom

38

bail was set at less than $500 in only 739 cases, compared to 8,016 cases with bail at
$1,000.112
Of course, some judges do make bail decisions taking into account what the defendant can
pay. As one New York City judge told Human Rights Watch, “I will not keep someone in jail
because he doesn’t have $500. That’s just penalizing someone for being poor.”113
Defense attorneys rarely appeal the amount of bail set at arraignment. The public defenders
who typically represent misdemeanor defendants lack the staff resources to seek review of
thousands of cases of unaffordable bail. Even when review is sought, judges are reluctant to
overturn bail decisions absent markedly changed circumstances following the initial bail
decision,114 and it is extremely difficult to prevail in a habeas corpus proceeding arguing the
judge abused his or her discretion in setting bail.115

The New York City Criminal Justice Agency’s Release Recommendations
If a defendant is free in the community pending trial, there is a risk that he may fail to return
to court. The New York City Criminal Justice Agency assesses that risk using information
which its research has shown to have a strong empirical relationship with the likelihood that
defendants will appear for scheduled court dates. It classifies defendants into one of four
categories: recommended for release on recognizance (ROR); moderate risk for ROR; not
recommended for ROR; and no recommendation. The CJA makes no recommendations
regarding bail amount.
The CJA classification is based on information collected in the pre-arraignment interview CJA
personnel conduct with defendants who after arrest have been held in custody until
arraignment 116 and information the CJA obtains from the defendants’ official criminal history.

112
Unpublished data provided to Human Rights Watch by New York City Criminal Justice Agency, June 2010. Data on file at
Human Rights Watch.
113
Human Rights Watch interview with New York City judge (name withheld), New York City, March 14, 2010.
114
Under New York Criminal Procedure Law, sec. 530.30, if bail was denied or, if the bail was excessive, the superior court
judge “may vacate the order of [the] local criminal court and release the defendant on his own recognizance or fix bail in a
lesser amount or in a less burdensome form.”
115
A defendant can bring a petition for a writ of habeas corpus seeking a judgment granting him bail—if bail had been denied—
or reducing bail, but the scope of review of an order denying or fixing bail is extremely narrow, and is limited to determining
whether “it appears that the constitutional statutory standards inhibiting excessive bail or the arbitrary refusal of bail [have
been] violated.” People ex rel. Klein v. Krueger, 25 N.Y.2d 497, 499 (1969).
116
CJA does not interview defendants who were not subjected to custodial arrests, such as those who are arrested and given
desk appearance tickets notifying them that they must appear in court on a future date in connection with the arrest charges.

39

Human Rights Watch | December 2010

The current recommendation system, which began in 2003, includes four measures of
community ties: 1) does the defendant report a NYC address; 2) does the defendant have a
working telephone or cell phone; 3) is the defendant employed, in school, or in a training
program full time; and 4) does the defendant expect someone, such as a friend or relative, to
attend the arraignment. CJA obtains this information from defendants and it tries to verify
their phone numbers, addresses, and work/employment. The CJA also looks at two aspects
of the defendant’s criminal history: has the defendant ever had a prior bench warrant and
does he or she have any current open cases. Points are assigned or deducted according to
the responses, with positive points assigned, for example, for evidence of community ties
and negative points assigned, for example, for the absence of such ties and the existence of
a bench warrant or another pending case. The total score, which can range from -12 to + 12,
is used to determine the CJA’s release recommendation. In 2008, the CJA recommended 32
percent of all adult defendants citywide for release on recognizance; 18 percent were
categorized as a moderate flight risk; 46 percent were not recommended for release, and 4
percent had no recommendation.117
CJA research shows that defendants it has recommended for release and who are in fact
released have lower failure-to-appear rates than those released despite a negative CJA
recommendation. For example, in 2008 released defendants who had been recommended
for release by CJA had a failure-to-appear rate of 9 percent, compared to a 24 percent failure
to appear rate of defendants who had not been recommended.118
Judges take CJA recommendations into consideration, but they are not the decisive factor in
judicial decision-making.119 Some judges told Human Rights Watch that they pay more
attention to the facts ascertained by the CJA (such as whether the defendant is employed)
than to its release recommendations. Although defendants recommended for release by CJA
are more likely to be released than those CJA did not recommend,120 judges also grant
release on recognizance far more frequently than CJA recommends. Judges released
defendants in 73 percent of misdemeanor cases citywide in 2008 that continued past
arraignment even though CJA had recommended release in only 35 percent of those cases.121

117

New York City Criminal Justice Agency, “2008 Annual Report,” December 2009, p. 14, Exhibit 7; Qudsia Siddiqi, New York
City Criminal Justice Agency, “CJA’s New Release-Recommendation System,” Research Brief No. 5, April 2004,
http://www.cjareports.org/reports/brief5.pdf (accessed October 26, 2010).
118
New York City Criminal Justice Agency, “2008 Annual Report,” p. 27, Exhibit 18.
119
Phillips, “Factors Influencing Release and Bail Decisions in New York City,” p. 2, p. 19, pp. 24-32.
120
New York City Criminal Justice Agency, “Annual Report 2008,” p. 19, Exhibit 12. ROR was granted in 81 percent of the cases
in which the CJA had recommended it, compared to 43 percent of the cases in which it had not recommended it. The cases
include defendants with felony and nonfelony charges.
121
Percentages extrapolated from data contained in ibid., p. 19, Exhibit 12; and p. 18, Exhibit 11.

The Price of Freedom

40

Some judges suggested to Human Rights Watch that release on recognizance is granted
more frequently than recommended by CJA because the CJA criteria are too stringent. For
example, the CJA will not recommend a defendant for release if he has ever received a bench
warrant for failure to appear in court in a previous case, regardless of how long ago that
happened, what the reasons were for the prior failure to appear, or whether he voluntarily
returned to court the next day after missing a scheduled proceeding.122 Judges also noted
that the CJA’s point system penalizes poor people, particularly those who do not have full
time jobs or go to school full time and those who do not have working telephones. While the
CJA’s statistical research demonstrates such factors are empirically correlated with risk of
failure to appear, it appears that in many cases judges are reluctant to let mere poverty—in
the absence of other factors—be the deciding factor in release decisions.

Role of Prosecutor
According to research by the CJA, the prosecutor’s bail request is the strongest predictor of
judicial decisions on both release on recognizance and bail amount, outweighing the CJA
recommendations, the nature of the offense, and the defendant’s criminal history.123 The CJA
found it very unusual for a judge to set bail if the prosecutor consented to release on
recognizance (i.e., indicated that release was acceptable). Judges can and do order release
on recognizance when the prosecutor has asked for bail to be set (see box below), but the
higher the bail amount requested by the prosecutors, the lower the likelihood the judge will
grant release.124 With regard to bail, the amount requested by the prosecutor is “the only
important factor” in the amount set by judges.125 Prosecutor requests “establish parameters”
for the ultimate bail decision: judges rarely set bail higher than that requested by the
prosecutor and although they often set it lower, the amount they set rises “in tandem” with
prosecutors requests, i.e., the higher the request by the prosecutor, the higher the judge’s
bail amount is likely to be.126

122

The information CJA has on prior bench warrants is taken from the official criminal history of each defendant prepared by
the New York State Department of Criminal Justice Services. That history will not contain the reasons for the prior failure to
appear and may not reflect the fact that the defendant subsequently returned to court.
123
Phillips, “Factors Influencing Release and Bail Decisions in New York City,” pp. 39-41.
124
Phillips, “Factors Influencing Release and Bail Decisions in New York City,” p. 23.
125
Phillips, “Factors Influencing Release and Bail Decisions in New York City,” p. 39.
126
Ibid., p. 40. According to the CJA analysis, the median amount of bail set rose as the amount of bail requested by
prosecutors rose, although the difference between the prosecutor’s request and the bail set increased as the bail amount
requested increased. The median bail set was $2,500 in both Manhattan and Brooklyn when bail requests were between
$5,000 and $9,999; when requested bail amounts were $50,000 or more, the median bail amount set was $38,000 in
Brooklyn and $35,000 in Manhattan. Ibid., p. 23.

41

Human Rights Watch | December 2010

Sam D., age 26, was arraigned
on charges of stealing $22 of
food (some sugar, a piece of
chicken, and a cake) which he
said he stole because he had
no money and it was his
girlfriend’s birthday. He lives
with her and their one-monthold child on unemployment
insurance of $140 a week. He
had no prior convictions
although another shoplifting
case was pending. Although
the prosecutor asked for $500
bail, the judge released Sam
on his own recognizance.127

Although prosecutors have great influence on judicial release
and bail decisions, they are in fact relatively poor at predicting
failure to appear by defendants. When the CJA undertook an
analysis of prosecutor’s bail requests in light of released
defendant’s actual record with regard to returning to court, it
could find no consistent pattern. Prosecutors had not
requested higher bail for defendants who did not return to
court and they had not consented to release for defendants
who did return to court. “Among released defendants, those
for whom prosecutors had consented to [release on
recognizance] failed to appear at the same rate as those for
whom the prosecutor had requested over $5,000 bail.”128 CJA
believes that “arraignment judges would have a better chance
of making an accurate assessment of risk of flight if they gave
more weight to the CJA recommendation than to prosecutors’
requests.”129

If judges are heavily influenced in their release and bail decisions by what the prosecutor
requests, what influences the prosecutors? According to the CJA’s analyses, prosecutors’
stances on bail or release are influenced by charge severity, criminal history, and whether
the offense was violent or involved a weapon. The existence of a prior bench warrant and the
existence of open cases strongly increased the likelihood that prosecutors would not
consent to ROR.130 However, the CJA’s analyses only went so far, accounting for less than half
the variance in the prosecutor’s consent to ROR.131 Charge severity was the predominant
factor influencing the bail amount requested by prosecutors.132
Prosecutors we interviewed were unwilling to say whether they had formal or informal
guidelines regarding when to consent to release on recognizance. They insisted their
positions depended on the specifics of each case, taking into account the defendant’s
criminal history, strength and severity of the case, and the CJA’s recommendation, among
many other factors. But they acknowledged they lack the time and information to make truly
127

Human Rights Watch attended the arraignment of Sam D. (not his real name) in Manhattan on May 5, 2010.
Mary T. Phillips, New York City Criminal Justice Agency, “Prosecutors’ Bail Requests and the CJA Release
Recommendation,” Research Brief No. 9, August 2005, http://www.cjareports.org/reports/brief9.pdf (accessed October 26,
2010), p. 7.
129
Ibid, p. 2.
130
Phillips, “Factors Influencing Release and Bail Decisions in New York City,” p. 28.
131
Ibid, p.6.
132
Ibid., p. 31.
128

The Price of Freedom

42

individualized decisions. As noted above, prosecutors’ positions on release and bail are
poorly correlated with whether defendants fail to appear in court. There may be an inherent
predisposition among prosecutors against consenting to release, as reflected in the fact that
prosecutors consent to release on recognizance in only about 21 percent of criminal cases.133
As one judge explained to Human Rights Watch, even though prosecutors have become
more used to release without bail (they no longer “choke” on it as they used to), many still
have a bias against it.134
Prosecutors may know from past experience that a judge is likely to grant release in a
particular case, but will ask for bail anyway to send a message to the defendant and his
counsel that they take his case seriously and will be prosecuting it vigorously. One New York
City prosecutor told Human Rights Watch, “I was told always to ask for at least $750…because
less suggests you really don’t care if [release on recognizance] is granted.”135 A judge with
many years on the bench agreed that a prosecutor who requests a low bail amount is signaling
to the court that release is okay, although the prosecutor is not willing to acknowledge that
publicly.136 Because there is always a risk that a defendant released pretrial will commit a
crime that lands the case on the front page, prosecutors find it safer “politically” to ask for bail,
and let the judges take the heat for having released the defendant without bail.
A New York City prosecutor told Human Rights Watch that the most important factors
influencing prosecutors’ consent to release are “whether there is a victim at risk and whether
there is an existing bench warrant.”137 This prosecutor also emphasized the importance of
proportionality between the likely sentence for the crime and pretrial detention: “if we’re not
offering a plea bargain that includes incarceration, then we are not going to ask for bail.”
Prosecutors will also typically ask for bail for defendants who fall into Operation Spotlight, a
citywide program of the mayor and the New York City Office of the Criminal Justice Coordinator
that was initiated in 2002 to target persistent misdemeanor recidivists who engage in qualityof-life offenses.138 According to a New York City prosecutor: “These [Operation Spotlight]
133
The CJA study was done in Brooklyn and Manhattan only, not citywide. Results for prosecutors in other boroughs might be
very different. Mary T. Phillips, New York City Criminal Justice Agency, “Release and Bail Decisions in New York City,” Research
Brief No. 6, August 2004, http://www.cjareports.org/reports/brief6.pdf (accessed October 26, 2010), p. 6.
134
Human Rights Watch interview with New York City judge (name withheld), New York City, March 8, 2010.
135
Human Rights Watch interview with New York City prosecutor (name withheld), New York City, April 16, 2010.
136
Human Rights Watch interview with New York City judge (name withheld), New York City, March 18, 2010.
137
Human Rights Watch telephone interview with New York City prosecutor (name withheld), New York City, January 8, 2010.
138
Freda F. Solomon, New York City Criminal Justice Agency, “Operation Spotlight: Year Four Program Report,” November 2007.
In the program’s fourth full year, there were 20,149 docketed Spotlight program cases. Defendants who are targeted by
Operation Spotlight have been arrested for a misdemeanor, have at least two other arrests within the previous twelve months,
at least one of which must had a misdemeanor as the top arrest charge, and must have two or more misdemeanor convictions,
at least one of which must have occurred within the last 12 months. Defendants meeting Operation Spotlight criteria are

43

Human Rights Watch | December 2010

defendants face one year of incarceration, therefore we usually recommend bail at
arraignment because of all appropriate factors connected with a chronic recidivist and
generally one not compliant with a previous sentence obligation or other court order.”139
Release on recognizance is granted in less than a quarter of Operation Spotlight cases, a low
rate for defendants charged with misdemeanors.140
There is little statistical data on the amount of bail prosecutors request in misdemeanor
cases. Using a sample of felony and nonfelony cases from 2002 and 2003, CJA found that
prosecutors in Manhattan and Brooklyn asked for bail of $500 or less in 12.4 percent of the
cases, $501 to $1,000 in 16.1 percent, $1,001 to $5,000 in 26.8 percent, and over $5,000 in
18 percent.141 In the CJA’s analysis of the factors influencing bail amounts requested by
prosecutors, charge severity was far and away the most important factor, with the nature of
offense (i.e., whether violent) and criminal history considerably less important.142
In interviews with Human Rights Watch, prosecutors shed little light on how they come up
with the dollar amounts they request as bail. Robert Johnson, the Bronx district attorney,
insisted to Human Rights Watch that bail is “not a science…there is no formula for
determining bail requests. It is a human system…subjective…there are so many variables.”143
A New York City prosecutor told Human Rights Watch that bail requests in misdemeanor
cases typically range from $500 to $1,000. “When we ask for $1,000 it’s a case where we
believe the defendant will do jail time if convicted and therefore is more likely to flee.”144 He
said that in his office they never ask for bail of $250, but would just consent to release in
such a case. But when pressed about the importance of bail being affordable and tailored to
defendant’s financial means, the prosecutor pointed out that prosecutors lack verified
automatically identified on rap sheets produced by New York State’s Division of Criminal Justice Services. As the Bronx District
Attorney’s office notes, all such flagged cases receive “enhanced prosecutorial attention” (see
http://bronxda.nyc.gov/fcrime/initiatives.htm#spotlight). Most of the Spotlight defendants are charged with nonviolent
crimes. For example, in the period October 1, 2005 through September 30, 2006, 32.7 percent were charged with drug
possession, 17.8 percent with petty larceny, and 8.2 percent with theft of services (turnstile jumping). Only 4.1 percent were
charged with crimes involving harm to persons, e.g. assault. Solomon, “Operation Spotlight,” p. 11.
139
E-mail communication from a New York City prosecutor (name withheld) to Human Rights Watch, August 25, 2010.
140
Solomon, “Operation Spotlight: Year Four Program Report,” November 2007, p. 10, Table 4A. According to CJA’s analysis, 65
percent of Operation Spotlight defendants were a high risk of failing to appear because of their lack of community ties and
another 12.5 percent had prior bench warrants. Ibid., p. 5.
141
Percentages calculated from Phillips, “Prosecutors’ Bail Requests and the CJA Release Recommendation,” Figure 2. Data
based on defendants arraigned in Criminal Court in Brooklyn and Manhattan between September, 2002 and March, 2003.
142
Phillips, “Factors Influencing Release and Bail Decisions in New York City,” p. 30. Other than charge severity, the other
factors that had a significant impact on the amount of bail prosecutors requested were mostly related to the nature of the
offense, and, to a lesser extent, to criminal history.
143
Human Rights Watch telephone interview with Robert T. Johnson, district attorney, Bronx County, New York City, February
24, 2010.
144
Human Rights Watch interview with New York City prosecutor (name withheld), New York City, May 25, 2010.

The Price of Freedom

44

information on which to assess defendant’s ability to afford bail. “You never really know if a
person can make bail…except if obviously homeless.” He continued, “How should a
prosecutor decide how much? How do we find out a defendant’s income and assets? There
are no verified sources available when we make the bail request. We take into account what
we can, but info, even from the CJA, is limited and not verified.” District Attorney Johnson
insisted it is “very subjective how to gauge economic means.”145
Given the influence prosecutors bail requests have on judges’ decisions, prosecutors have a
responsibility to try to tailor their bail requests to the defendants’ resources, even if they do
so with admittedly limited and mostly unverified information. If they have enough
information to request $500 or $1,000 bail rather than $10,000, they have enough to know
the defendant has scant financial resources and to request less, or to suggest unsecured
bonds rather than cash bail. According to the National Prosecution Standards of the National
District Attorneys Association, prosecutors should “take steps to gather adequate
information about the defendant’s circumstances and history to request an appropriate bail
amount” that will ensure the defendant appears at all required court proceedings, and,
where allowed by law, does not pose a danger to others or to the community.146 The amount
requested should not be “greater than necessary to ensure the safety of others and the
community and to ensure the appearance of the defendant at trial.”147 In addition,
prosecutors “should recommend bail decisions that facilitate pretrial release rather than
detention to the extent such release is consistent with the prosecutor’s responsibilities…”148
Judges and defense attorneys interviewed by Human Rights Watch did not believe that
prosecutors are greatly concerned with making sure the amount of bail they request is no
more than necessary to ensure defendants return to court. Several prosecutors interviewed
by Human Rights Watch did not seem particularly disturbed that nonfelony defendants were
being held in pretrial detention solely because they could not afford bail. Nor did they see it
as their responsibility to help avoid such detention. In their view, it is for the defense
attorneys to gather and present to the judge information about defendants’ finances and it is
the judges’ responsibility to make sure bail is set at amounts that do not force defendants
into pretrial detention just because they are poor. Quoting District Attorney Johnson again:
“It’s the defense attorney’s job to get defendants out of jail or to get a low enough bail that

145

Human Rights Watch telephone interview with Robert T. Johnson, New York City, February 24, 2010.
National District Attorneys Association, National Prosecution Standards 3rd edition, 2010, Standard 4-4.12 and 4-4.2.
147
Ibid., Standard 4-4.2.
148
Ibid., Standard 4-4.4.
146

45

Human Rights Watch | December 2010

they can meet.”149 Kings County District Attorney Charles Hynes told Human Rights Watch
that in theory he agrees that defendants should not be incarcerated because they are too
poor to make bail, but he insisted that defense counsel should provide the zealous
representation of their clients necessary to make sure that does not happen.150

Public Safety
Federal law and the law in many states permit pretrial detention of criminal defendants in
order to protect the safety of the community or specific individuals as well as to secure their
appearance in court.151 While the New York State legislature has not passed a law that
authorizes pretrial detention on grounds of public safety, there is little doubt that
considerations of public safety influence release and bail decisions in New York City.
Bail in New York City is sometimes deliberately set at a level that will ensure the pretrial
detention of defendants deemed dangerous.152 Considerations of dangerousness most
typically and obviously arise in cases in which the defendant is charged with a violent
crime—particularly if the defendant has a prior record of violence—and/or in cases in which
the defendant has demonstrated through his actions a likelihood of injuring a possible
witness against him. During the CJA’s research on factors influencing judges’ release and
bail decisions in Manhattan and Brooklyn, judges acknowledged public safety does play a
role in some cases. But because the law does not authorize detention on safety grounds,
there is no open and full evaluation at the arraignment of the strength of the evidence
supporting or undercutting public safety concerns. A defendant may lose his liberty because
the judge fears he may be dangerous, but that loss of liberty is not the result of a hearing
devoted expressly to the question of whether public safety requires his detention.
Judges are also well aware of the risks to themselves professionally from releasing without
bail a defendant who might prove dangerous. As one judge said, “judges set bail knowing
they will never be criticized publicly for putting someone in jail, only for letting someone out
without bail who then commits a crime.”153 The judicial nightmare, acknowledged by almost

149
Human Rights Watch telephone interview with Robert T. Johnson, district attorney, Bronx County, New York City, February
24, 2010.
150
Human Rights Watch interview with Charles J. Hynes, district attorney, Kings County, New York City, May 10, 2010.
151
United States v. Salerno, 481 U.S. 739, 750 (1987). The U.S. Supreme Court has upheld the constitutionality of such
“preventive detention” on public safety grounds.
152
American Bar Association, ABA Criminal Justice Standards: Pretrial Release, Standard 10-1.4(e), Commentary. Financial bail
should serve “only as an incentive for released defendants to appear in court and not as a subterfuge for detaining
defendants.”
153
Human Rights Watch interview with New York City judge (name withheld), New York City, March 14, 2010.

The Price of Freedom

46

every judge we interviewed for this report, is to end up on the cover of the New York Post for
releasing without bail a defendant who then murders someone.
Prosecutors acknowledged to Human Rights Watch that they are likely to ask for bail when
the charged misdemeanor offense involves domestic violence or in other assault cases in
which the complaining witness might be in danger if the defendant is not detained pending
trial.154 As one prosecutor explained, in cases involving violence, the prosecutor cannot and
should not ignore the importance of protecting a complaining witness. Asking for bail in
such cases is often a deliberate effort to secure the defendant’s detention. “Pretrial
detention is warranted when the defendant is not going to return otherwise or when the case
involves violence and the vulnerability of a witness.”155 Other prosecutors expressed same
view: “Any case where there is a civilian victim, we take a hard look, even if not in the statute,
because we want to protect the victim. We cannot ignore safety issues.”156
Human rights law accepts the legitimacy in principle of pretrial detention of criminal
suspects who are considered too dangerous to be released (see discussion below). If the
New York legislature decided to incorporate public safety as a basis for pretrial detention, a
full hearing consistent with due process guarantees that determines whether public safety in
fact requires pretrial detention in a particular case would be consistent with the right to
liberty and the presumption of innocence. It would also be consistent with the pretrial
standards of the American Bar Association. Under current law and practice, however, New
York City nonfelony defendants lose their liberty because of bail decisions that ostensibly
are made only to secure their future appearance in court.

Race
Pretrial punishment in the form of detention because of inability to post bail is endured
primarily by blacks and Hispanics. The disproportionate pretrial detention of minorities does
not appear to arise from the influence of race in bail setting so much as from the
disproportionate rates of minority arrests. The high percentage of “quality-of-life”
misdemeanor arrests (along with “stop and frisks” that do not lead to arrests) that occur in
heavily minority and poor neighborhoods are themselves cause for great concern, raising

154

Felony charges that involve violence or mention of a weapon significantly affect prosecutorial willingness to consent to
release, in Brooklyn tripling the odds against consent to release and in Manhattan the odds against consent were 17 times
greater. Phillips, “Factors Influencing Release and Bail Decisions in New York City,” p. 29.
155
Human Rights Watch interview with New York City prosecutor (name withheld), New York City, May 25, 2010.
156
Human Rights Watch interview with New York City prosecutor (name withheld), New York City, June 22, 2010.

47

Human Rights Watch | December 2010

questions about the fairness and legitimacy of New York City’s racially disparate policing
strategies.157
Although blacks and Hispanics combined constitute only 51 percent of the New York City
population, they comprise 82.4 percent of all misdemeanor arrestees.158 Many of the
misdemeanor arrests are not based on victim complaints and because of the nature of the
offenses are discretionary, the police could for example, give a warning to someone seen
smoking marijuana rather than making an arrest. Because there is no citywide data on
misdemeanor offense rates we do not know whether higher black and Hispanic rates of
arrests reflect higher rates of engaging in offending conduct. Such data as is available
suggests maybe not. As noted above, a large percentage of arrests and arraignments in New
York City are for smoking marijuana in public. Blacks and Hispanics account for 87.3 percent
of arrests for misdemeanor possession of marijuana in public view,159 even though drug use
surveys indicate that whites use marijuana as much if not more than minorities.160 Given the
high correlation between minority status and income in New York City, it is perhaps not
surprising that, as shown in Figure 5, blacks and hispanics constitute 89 percent of all
pretrial detainees held on bail of $1,000 or less (blacks 58 percent and Hispanics 31
percent).

157
There is extensive literature on law enforcement in New York City minority neighborhoods, particularly since the New York
City police initiated “zero tolerance” and “quality of life” policing. The striking racial disparities in the “stop and frisks” as
well as quality of life arrests have been well documented and are subject of much controversy. Do they have unwarranted
disparate impact on communities of color from racially disparate policing that targets poor people in poor neighborhoods or
are they a legitimate and necessary means of keeping crime rates down and protecting minority as well as other communities?
See, for example, Delores Jones-Brown, Jaspreet Gill, and Jennifer Trone, Center on Race, Crime and Justice, John Jay College of
Criminal Justice, “Stop, Question & Frisk Policing Practices in New York City: A Primer,” March 2010; Jeffrey Fagan and Garth
Davies, “Street Stops and Broken Windows: Terry, Race, and Disorder in New York City,” Fordham Urban Law Journal, vol. 28
(2000), p. 457; Jeffrey Fagan et al, “Street Stops and Broken Windows Revisited: The Demography and Logic of Proactive
Policing in a Safe and Changing City,” in Stephen K. Rice and Michael D. White, eds., Race, Ethnicity and Policing: New and
Essential Readings (New York: New York University Press, 2009); K. Babe Howell, “Broken Lives from Broken Windows: The
Hidden Costs of Aggressive Order-Maintenance Policing,” New York University Review of Law and Social Change, vol. 33
(2009), p. 271; Andrew Golub, Bruce D. Johnson and Eloise Dunlap, “The Race/Ethnicity Disparity in Misdemeanor Marijuana
Arrests in New York City,” Criminology and Public Policy, vol. 6 (2006), p. 131; Harry G. Levine and Deborah Peterson Small,
New York Civil Liberties Union, “Marijuana Arrest Crusade: Racial Bias and Police Policy in New York City, 1997-2007,” 2008,
http://www.nyclu.org/files/publications/nyclu_pub_marijuana_arrest_crusade.pdf (accessed October 25, 2010).
158
According to unpublished data on NYC misdemeanor arrests in fiscal year 2009 provided to Human Rights Watch by the
New York State Division of Criminal Justice Services, whites constituted 12.9 percent of the misdemeanor arrestees, blacks
constituted 48.9 percent, Hispanics constituted 33.6 percent, and other/unknown/missing persons constituted 4.6 percent.
Data on file at Human Rights Watch.
159
Unpublished data on the racial composition of New York City persons arrested for misdemeanor possession of marijuana in
public view in fiscal year 2009 provided to Human Rights Watch by New York State Division of Criminal Justice Services. Data
on file at Human Rights Watch.
160
Levine and Small, New York Civil Liberties Union, “Marijuana Arrest Crusade”; Golub, Johnson, and Dunlap, “The
Race/Ethnicity Disparity in Misdemeanor Marijuana Arrests in New York City,” Criminology and Public Policy.

The Price of Freedom

48

Figure 5

In an analysis of factors influencing bail decisions in Manhattan and Brooklyn, the CJA found
that race did not have a statistically significant effect on release decisions for misdemeanor
defendants in either borough.161 Among nonfelony cases in which the defendant was arrested
in 2008 and either given bail under $1,000 or released on their own recognizance, black
defendants were somewhat less likely to be released on recognizance than white (86 percent
versus 93 percent), as shown in Table 5. Among misdemeanor defendants recommended for
release by the CJA, however, this racial disparity essentially disappears (see Table 6).

161

Phillips, “Factors Influencing Release and Bail Decisions in New York City,” Appendix B. When the results of the logistic
regression models for misdemeanor and felony cases are combined, ethnicity had a statistically significant impact in
Manhattan (being white increased the likelihood of ROR), but not in Brooklyn.

49

Human Rights Watch | December 2010

Table 5
Release on Recognizance & Bail under $1,000 for Nonfelony Defendants, by
Race/Ethnicity

Defendant’s Race/Ethnicity
Black

ROR

Bail <
$1,000

Total

40,009

6,394

46,403

Hispanic

30,911

3,403

34,314

White

13,480

1,054

14,534

Other

5,688

260

5,948

517
90,605

10
11,121

527
101,726

*data not available
Total

Percentage
Receiving
ROR

86.2%
90.1%
92.7%
95.6%
98.1%
89.1%

Source: New York City Criminal Justice Agency
Note: Includes cases of New York City defendants arrested in 2008 on nonfelony charges with bail set less
than $1,000. Cases with bail set at $1 excluded.

Table 6
Release on Recognizance and Bail Amount for Nonfelony Defendants By
Race/Ethnicity with Defendant Recommended by New York City Criminal Justice
Agency for Release on Recognizance
Category
ROR
Bail < $1,000
Total

Black

Hispanic

White

Other
Ethnicity

Ethnicity
Not Known

Total

13,823

11,608

5,303

2,519

5

33,258

96.4%

96.5%

97.2%

97.6%

100.0%

96.7%

518

416

150

62

0

1146

3.6%

3.5%

2.8%

2.4%

0.0%

3.3%

14,341

12,024

5,453

2,581

5

34,404

Source: New York City Criminal Justice Agency
Note: Includes cases of New York City defendants arrested in 2008 on nonfelony charges with bail set less
than $1,000. Cases with bail set at $1 excluded.

The Myth that Released Defendants Evade Justice
Judges who set bail at levels that defendants will not be able to make in order to ensure their
appearance at subsequent court proceedings may not be aware that it is relatively rare for
released defendants to miss scheduled court proceedings.
In New York City, 84 percent of criminal defendants who are not in pretrial detention attend
all their scheduled court proceedings.162 Sixteen percent miss an appearance, but most of
162

New York City Criminal Justice Agency, “2008 Annual Report,” p. 26, Exhibit 16.

The Price of Freedom

50

those return to court voluntarily within 30 days.163 It is unlikely that defendants who return to
court within 30 days of missing a scheduled appearance are trying to avoid justice.164 Only 6
percent miss an appearance and do not return to court within 30 days.
Many prosecutors and judges assume that misdemeanor defendants are likely to avoid
court—even if they have never done so in the past—if they have prior criminal convictions
and hence are likely to face longer jail sentences if convicted again. This belief is not new:
charge severity and extensiveness of prior criminal record were historically the principal
determinants of release decisions by judges. It may be this historical legacy remains
embedded in current judicial culture. While the belief has a certain logic to it, extensive
research by the CJA has demonstrated that far better predictors of the risk of failure to
appear are lack of community ties, prior failure to appear, and the existence of another open
case.165 The empirical evidence suggests that seriousness of charge or prior criminal history
may not significantly affect showing up in court. For example:
•

•

Felonies are more serious crimes than misdemeanors and typically are punished
with harsher sentences. Following the logic outlined above, one would assume that
people accused of felonies would be more likely to flee than those accused of
misdemeanors. But in New York City, defendants charged with felonies have
essentially the same failure-to-appear rates (16 percent) than those charged with
misdemeanors (17 percent).166 That is, charge severity does not seem to have much
of an impact on failure to appear.
CJA analyses demonstrate that prior failure to appear and open cases are strong
predictors of the failure of released New York City defendants to appear in court,
overshadowing other aspects of criminal history including prior convictions. Indeed,
prior convictions are not significant predictors of failure to appear when prior failure
to appear and open cases are taken into account.167

163

Ibid., p. 28, Exhibit 20.
It is notable that almost all (94 percent) defendants successfully contacted by CJA’s Failure-to-Appear Units because they
missed a scheduled court hearing voluntarily returned to court within 29 days. New York City Criminal Justice Agency, “2008
Annual Report,” pp. 33-34, Exhibit 25. The Failure- to-Appear Units, which operate in Brooklyn and Queens, attempt to contact
defendants who missed scheduled court appearances to persuade them to return to court voluntarily.
165
CJA’s research underlying its release recommendation system and reasons for the changes in it are discussed in Qudsia
Siddiqi, New York City Criminal Justice Agency, “Assessing Risk of Pretrial Failure to Appear in New York City,” November 1999;
Siddiqi, New York City Criminal Justice Agency, “CJA’s New Release-Recommendation System”; Qudsia Siddiqi, New York City
Criminal Justice Agency, “Pretrial Failure Among New York City Defendants, ” Research Brief No. 19, January 2009,
http://www.cjareports.org/reports/brief19.pdf (accessed October 26, 2010).
166
New York City Criminal Justice Agency, “2008 Annual Report,” p. 27, Exhibit 17; and p. 40, Exhibit 30.
167
If prior convictions had an independent influence on the likelihood a defendant would fail to appear at subsequent
proceedings (above and beyond the influence of prior instances of failure to appear and open cases), then prior conviction
variables would have been statistically significant in CJA’s research, but they were not. See Qudsia Siddiqi, New York City
Criminal Justice Agency, “Predicting the Likelihood of Pretrial Re-Arrest Among New York City Defendants: an Analysis of the
2001 Dataset,” June 2003, revised October 2005, Table 11. Prior research regarding the impact of criminal histories on pretrial
164

51

Human Rights Watch | December 2010

•

A recent statistical analysis of pretrial misconduct nationwide found that whether or
not felony defendants have prior criminal convictions makes scarcely any difference
in the probability they will appear in court after being released pending trial.168

Our interviews suggest that defendants’ failure to return to court for a scheduled appearance
is more typically the result of difficult, stressful, and disorganized lives or of irresponsible
behavior rather than an intentional effort to avoid adjudication.
Defendants, lawyers, judges, prosecutors, and staff of CJA Failure-to-Appear Units told us
that defendants miss court proceedings for countless reasons: they lose the piece of paper
with the date, they cannot afford to forgo earnings from missed work (defendants may have
to spend better part of a day waiting in court before their case is called), they cannot find
care for the children at home, they do not have the money for bus or subway fare, they have
an exam, they do not understand they cannot just go on a different day, they show up in
court on the scheduled day but leave before their case is called. Many defendants lead
chaotic, unstructured lives in which keeping track of commitments is difficult. For some,
mental illness or drug addiction cuts into their willingness and ability to show up in court on
specified days. Some of the defendants who do not show up simply cannot be bothered to
do so; they do not “respect the system.” Some are just young and irresponsible.
No doubt some misdemeanor defendants who do not make court appointments have
engaged in a calculated effort to avoid the possibility of conviction and punishment, and
they are most likely included among the 6 percent of defendants who miss a court
proceeding and do not return to court within 30 days of doing so. They may know there is
little likelihood of the police picking them up on the bench warrant for their arrest the court
issues when they do not show up. The failure to appear in court may be prosecuted, however,
if the defendant is subsequently arrested for another offense and the criminal records check
uncovers the prior bench warrant.
Research by the CJA indicates that defendants who have previously missed court
proceedings are more likely to miss future ones. It may well be that the reason or reasons
that led to the first missed appearance persist in defendants’ lives. Lives made difficult by

failure to appear in other jurisdictions has yielded mixed results. See Qudsia Siddiqi, New York City Criminal Justice Agency,
“Predicting the Likelihood of Pretrial Re-Arrest for Violent Felony Offenses and Examining the Risk of Pretrial Failure among
New York City Defendants: an Analysis of the 2001 Dataset,” November 2006, pp. 20-21 for a brief review. An earlier CJA study
in 1999, based on data from 1989, found that prior violent felony convictions were a significant predictor of failure to appear
among at-risk defendants whose cases were prosecuted in criminal court.
168
Thomas H. Cohen and Brian A. Reaves, Bureau of Justice Statistics, “Pretrial Release of Felony Defendants in State Courts,”
November 2007, p. 10. The predicted probability of a released felony defendant being charged with failure to appear is 22
percent with no prior convictions, 21 percent with prior misdemeanors, and 23 percent with prior felony convictions.

The Price of Freedom

52

poverty and mental illness, for example, do not suddenly change. People who have no
respect for the legal system or who believe the benefits of not showing up in court exceed
the risks may continue to ignore court orders.
When defendants fail to show up, hearings for which defense counsel, prosecutors, and the
courts have prepared must be postponed, putting burdens on them and reducing the
efficiency of the already strained court system.169 The risk assessments conducted by the
New York City Criminal Justice Agency help the courts identify objectively which defendants
are statistically more likely to fail to appear in future court hearings if released on their own
recognizance. Whether judges follow the CJA recommendation or rely on other factors to
determine risk of failure to appear for court proceedings, the question is what to do with
nonfelony defendants at the high end of the risk spectrum who cannot afford bail.

Judicial Training
New judges in New York state, including in New York City, receive two weeks of training from
the New York State Judicial Institute. While that training includes some material on release
and bail decision-making, relatively little attention is paid to the nature and significance of
the differing bail forms authorized by statute, how to reasonably calculate bail amounts, or
how to assess the likelihood that a defendant will not make scheduled court appearances.
According to Judge Juanita Bing Newton, dean of the Judicial Institute, in addition to classroom
type learning, training for new judges includes a week at the bench watching a sitting judge
handle arraignments.170 New judges are thus exposed to the long-ingrained practice of relying
on cash bail and commercial bond. It is not surprising then that they do the same.171 Judges
interviewed by Human Rights Watch also suggested that their training by the Judicial Institute
provided them with little understanding of how to assess a defendant’s risk of failure to
appear or to calibrate bail amounts; they learn what amounts of bail may be appropriate to
secure a defendant’s appearance in particular cases primarily through experience. One judge
told us that the only guidance regarding bail she recalled from her judicial training a couple of
years earlier was that bail should never be set lower than $500.172 If her recollection was
accurate, such guidance would be contrary to New York and human rights law.

169
A separate question is whether the system could be modified to reduce the number of hearings defendants have to attend
while their cases proceed, or to reduce the amount of time they spend in court waiting for their hearing.
170
Human Rights Watch interview with Judge Juanita Bing Newton, dean, New York State Judicial Institute, White Plains, New
York, August 3, 2010.
171
Training for new judges at the Judicial Institute includes training on bail, but little attention is paid to the different possible
forms of bail or their significance for low income defendants.
172
Human Rights Watch interview with New York City judge (name withheld), New York City, June 7, 2010.

53

Human Rights Watch | December 2010

IV. An Alternative to Pretrial Detention: Pretrial Supervision
New York City can and should develop a way to avoid sending indigent defendants to jail
who cannot post bail and who are deemed too much of a flight risk to be released on their
own recognizance. Judges should have an alternative to the existing choice of release on
recognizance or money bail. That alternative should be supervised pretrial release.
Although New York City had the first pretrial services program in the country—the Manhattan
Bail Project, created by the Vera Institute of Justice in 1961, which subsequently spun off the
New York City Criminal Justice Agency—the focus from the start has been assisting judges
make informed release decisions based on objectively assessed risk factors. New York City
has lagged behind other jurisdictions in developing supervised pretrial release programs
that would enable criminal defendants ineligible for release on their own recognizance to
remain free in the community on non-financial conditions until the conclusion of their case.
According to the Pretrial Justice Institute, 97 percent of surveyed pretrial services programs173
that exist nationwide provide supervision of defendants released pending adjudication.174
Defendants in these programs are typically released on their promise to adhere to certain
court-ordered, non-financial conditions, such as reporting in-person on a regular basis.
Pretrial services or other criminal justice staff supervise the release of the defendant and
enforce compliance with release conditions through methods such as telephone calls or inperson meetings, referrals to substance abuse treatment and/or mental health treatment
programs, drug testing, electronic bracelets, reminding defendants of court dates, and
reporting to the court.
Pretrial services in New York City have traditionally been limited to CJA interviews with
persons subjected to custodial arrests and the presentation of results of the interview with

173

Pretrial services programs are operated by a range of different entities, including courts, jails, probation departments,
independent government agencies, and non-profit organizations providing services under a contract. See generally, Pretrial
Justice Institute, “2009 Survey of Pretrial Services Programs,” August 11, 2009,
http://www.pretrial.org/Docs/Documents/PJI's%20Survey%20of%20Pretrial%20Programs%202009.2.pdf (accessed October
25, 2010).
174
Ibid., p. 46. For a review of efforts to identify evidence-based practices and factors promoting success, see generally, Marie
VanNostrand, Crime and Justice Institute and National Institute of Corrections, “Legal and Evidence-Based Practices:
Application of Legal Principles, Laws, and Research to the Field of Pretrial Services,” (2009), http://nicic.org/Library/022398
(accessed October 25, 2010); National Institute of Corrections, “Topics in Community Corrections 2008: Applying EvidenceBased Practices in Pretrial Services,” http://nicic.gov/Library/022904 (accessed October 25, 2010); David Levin, “Examining
the Efficacy of Pretrial Release Conditions, Sanctions and Screening with the State Court Processing Statistics Dataseries,”
Paper Prepared for the Presentation of the Annual Meeting of the American Society of Criminology, Atlanta, GA, November 1416, www.pretrial.org/Docs/Documents/ASC07.pdf (accessed October 25, 2010).

The Price of Freedom

54

the agency’s release recommendations to the judge, prosecutor, and defense counsel at the
defendant’s arraignment. The CJA also attempts by telephone or letter to notify all released
defendants of their scheduled court appearances in an effort to reduce failure to appear
rates.175 In a promising and important expansion of its services, the CJA recently initiated a
pilot supervised pretrial release program in Queens for certain nonviolent felony defendants.
Participants are monitored through frequent face-to-face and telephone contact with
program staff and may engage in outside services according to an individualized plan.176 The
program has been extremely successful from a flight risk perspective; although CJA had
categorized two-thirds of the more than 200 participants to date as moderate or high risk for
failure to appear in court, there have been only a handful of cases in which courts have
ordered defendants expelled from the program for failure to appear at scheduled court dates.
The nation’s experience with pretrial supervision programs indicates that when pretrial
supervision is performed effectively, unnecessary pretrial detention is minimized, costly jail
services are avoided, public safety is increased, and the equity of the pretrial release
process is enhanced because there is less discrimination on the basis of income. All the
judges, defense counsel, and prosecutors as well as other criminal justice experts
interviewed by Human Rights Watch for this report supported the concept of supervised
pretrial release for misdemeanor defendants who are not granted release on recognizance
and who cannot afford bail. While most had questions concerning the details of such a
program (e.g., what types of supervision services would be available, how courts would
avoid expanding supervised release to those who would otherwise be released on
recognizance,177 and how communication between courts and pretrial release program
personnel would be ensured they all recognized that supervised release program would
provide an important alternative to the current choice of release without financial conditions
or jail secured through the indirect mechanism of bail. Where financial conditions would
truly not be enough to ensure the defendant’s appearance, or where defendants are too poor
for bail to have any meaning other than as a ticket to jail, judges would have the option of
ensuring defendants return to court by placing them in a supervised release program.

175

The CJA also operates a bail expediting program to assist certain defendants with raising money needed to post bail.
Human Rights Watch telephone interview with Mari Curbelo, associate director of Court Programs, New York City Criminal
Justice Agency, New York City, June 29, 2010. At its most intensive, the supervision consists of two meetings a week and one
phone call; the frequency diminishes over time when such intensive supervision is not deemed necessary any longer. Program
staff tries to be flexible setting the schedules so that timing works for the defendants.
177
The concern regarding net widening is that judges may be less likely to grant ROR if they know there is a supervised release
program, thus subjecting defendants who do not pose serious risk of failure to appear to unnecessary restrictions on their liberty.
176

55

Human Rights Watch | December 2010

“[Pretrial defendants] are not proven guilty. They may be innocent.
They may be no more likely to flee than you or I. But they must
stay in jail because, bluntly, they cannot afford to pay for the
freedom….[Pretrial services programs] have generated new
techniques for releasing accused persons prior to trial, without
hampering law enforcement, without increasing crime, and
without prompting defendants to flee. These techniques have
fiscal value. They can help to increase the efficiency of police
forces and they can save communities from the substantial costs
of unnecessary detention….
But even more significant, in a land which has put the quality of
justice ahead of the cost of justice, these techniques have social
value. They can enable courts to tailor bail decisions to the
individual. They can enable lawyers to do a better job of
representing their clients. And, most important of all, they can
save countless citizens from needlessly or unjustly spending days
or weeks or even months in jail.”
—Address by Attorney General Robert F. Kennedy, National
Conference on Bail and Criminal Justice, May 29, 1964

Pretrial Supervision: Avoiding the Expense of Detention
Pretrial detention is an expensive practice. According to the New York City Department of
Correction, the average daily cost per inmate in New York City jails is $202.65, based on
expenditures that are part of the department’s operating budget.178 If the full costs
associated with New York City jails are used, the average cost is $399.61 per inmate per
day.179 It defies common sense, much less fiscal prudence, to incarcerate someone at an
average daily cost of nearly $400 because they cannot afford bail of $500 or $1,000.
Of course, the average daily cost of incarceration is not the cost that would be avoided if the
jail population were reduced by one inmate. Many of the costs remain—for buildings, heat,
staff—until there is a significant inmate reduction. The Department of Correction estimates

178

E-mail communication from Robert Maruca, Department of Correction, to Human Rights Watch, May 11, 2010. The figures
are for fiscal year 2009.
179
Ibid. In addition to direct operating costs of the Department of Correction (DOC), costs associated with incarceration
include debt service on DOC capital projects, DOC pension and fringe benefits, legal services performed on behalf of DOC by
the city law department, legal lawsuits, judgments and claims related to DOC operations paid by the city, healthcare costs for
inmates that are borne by the Department of Health and Mental Hygiene and the New York City Health and Hospital
Corporation, the expenses of the New York City Board of Correction, and the costs expended by the City Department of
Education for adolescents and other inmates entitled to educational services.

The Price of Freedom

56

that if the jail population were reduced by 800 or more inmates, the cost savings would be
$161 per inmate per day.180 If misdemeanor defendants who could not make bail were kept
out of jail, the cost savings would obviously be quite significant. For example, as discussed
above, 16,649 misdemeanor defendants arrested in 2008 and unable to post bail of $1,000
or less spent an average of 15.7 days in pretrial detention. Using the $161 per inmate per day
figure, it cost the city an average of $2,527 to incarcerate each of these pretrial defendants. If
the city had not incarcerated any of them it would have saved more than $42 million.
Pretrial community-based supervision of men and women accused of crimes would
unquestionably be far less expensive than housing, feeding, and caring for them round-theclock in jail. In the CJA supervised release pilot program in Queens for certain nonviolent felony
defendants, the approximate cost is $23 to $31 per day per defendant, far less than the cost of
jail.181 According to the Pew Center on the States, nationwide one day in prison costs more
than 10 days on parole or 22 days on probation. Among probation agencies in the US, the
average daily cost per offender to provide probation services is $3.42, with the high figure at
$7.89.182 In New York state, probation services cost an average $4,000 per probationer per
year,183 a fraction of the $76,229 annual cost per inmate per year in New York City jails.184

180

Ibid.
Telephone interview with Jerome McElroy, executive director, New York City Criminal Justice Agency, New York City, August
10, 2010. The total cost per defendant for a median participation period of three to four months is $2,864.
182
Pew Center on the States, “One in 31: The Long Reach of American Corrections,” March 2009, p. 13.
183
Task Force on the Future of Probation in New York State, “Report to the Chief Judge of the State of New York,” February 2007,
p. 1.
184
City of New York, “Mayor’s Management Report, September 2010,” p. 150,
http://www.nyc.gov/html/ops/downloads/pdf/2010_mmr/0910_mmr.pdf (accessed October 25, 2010).
181

57

Human Rights Watch | December 2010

V. Applicable Constitutional and Human Rights Law
The pretrial incarceration of defendants who cannot afford bail implicates the
constitutionally protected fundamental right to pretrial freedom as well as the constitutional
guarantee of equal protection of the laws. Pretrial detention because of poverty also violates
internationally recognized human rights to liberty and equality under the law.

Constitutional law
Right to Liberty
Criminal defendants in the US are considered innocent until proven guilty; this presumption of
innocence is intimately tied to the “fundamental right” of pretrial release under reasonable
conditions. The Eighth Amendment to the US Constitution, which states that “excessive bail
shall not be required,” does not establish a constitutional guarantee that bail in some form or
amount will be available to all defendants.185 However, that amendment and the Due Process
clause of the Fifth Amendment,186 under which a defendant has a liberty interest in pretrial
release,187 give constitutional undergirding to the “tradition in this country…that one charged
with a crime is not, in ordinary circumstances, imprisoned until after a judgment of guilt.”188
The US Supreme Court has recognized the harms of pretrial detention:
The wrong done by denying release is not limited to the denial of freedom
alone….In case of reversal, [the defendant] will have served all or part of a
sentence under an erroneous judgment. Imprisoned, a man may have no
opportunity to investigate his case, to cooperate with his counsel, to earn the
money that is still necessary for the fullest use of this right to appeal.189

United States v. Salerno, 481 U.S. 739, 750 (1987)( the Eighth Amendment “says nothing about whether bail shall be
available at all.”). In Salerno the Supreme Court ruled that the Federal Bail Reform Act of 1984, which authorized pretrial
detention for dangerous defendants, did not violate the Eighth Amendment nor the Due Process Clause of the Fifth
Amendment to the US Constitution.
186
The Fifth Amendment of the U.S. Constitution provides: “No person shall…be deprived of life, liberty, or property, without
due process of law.” The government may not interfere with rights that are “implicit in the concept of ordered liberty,” that is,
rights, that while not expressly affirmed in the Constitution, are “so rooted in the traditions and conscience of our people as
to be ranked as fundamental.” Ibid., p. 751.
187
Ibid., p. 750.
188
Bandy v. United States, 81 S. Ct. 197, 197-98 (1960).
189
Ibid.
185

The Price of Freedom

58

Although pretrial release is a fundamental right, it can be restricted by a compelling
government interest.190
The constitutional prohibition on “excessive” bail does not mean financial conditions for
release must be affordable. According to the Supreme Court, “the only arguable substantive
limitation of the Bail Clause is that the Government’s proposed conditions of release or
detention not be ‘excessive’ in light of the perceived evil,” such as the risk the defendant will
not return to court.191 Bail should not be more than necessary to achieve the government’s
interests. Bail may be set at a figure a defendant cannot pay, but it will not be constitutionally
“excessive” as long as the amount is “reasonably calculated” to achieve its purpose.192
Like the US Constitution, the New York State Constitution does not establish a right to bail,
but protects defendants against “excessive bail” when it is provided.193 Excessiveness is not
determined in relation to the defendant’s financial means, but by whether the bail amount
exceeds that necessary to achieve its purpose, which under New York law is limited to
ensuring the defendant’s appearance in court.194 As New York courts have noted, “the
presumption of innocence accorded every criminal defendant militates strongly against
incarceration in advance of a determination as to guilt. For this reason, bail may not be set in
an amount greater than necessary to ensure court attendance.”195 Another court noted,
“Since bail is a security device, it must be fixed only in an amount that is both sufficient and
necessary to guarantee the defendant’s appearance.”196The amount of bail is to be
determined on a case by case basis, balancing the factors that suggest likelihood of failure
to appear with the “right to freedom from unnecessary restraint before conviction.”197

190
United States v. Salerno, 481 U.S. 739, 750-751 (1987) (while acknowledging the “fundamental nature” of the pretrial right
to liberty, “this right may, in circumstances where the government’s interest is sufficiently weighty, be subordinate to the
greater needs of society.”).
191
Ibid, p. 754.
192
Stack v. Boyle, 342 U.S. 1, 5 (1951)(bail is not excessive if “reasonably calculated” to assure defendant’s presence.).
193
The Constitution of the State of New York, art. I, sec. 5 states: “Excessive bail shall not be required nor excessive fines
imposed, nor shall cruel and unusual punishments be inflicted, nor shall witnesses be unreasonably detained.”
194
People v. Torres, 446 N.Y.S.2d 969, 972 (1981)(bail should be fixed in an amount “sufficient and necessary to guarantee the
defendant’s appearance.”). People v. Mohammed, 653 N.Y.S.2d 492 (1996)(a defendant’s financial capacity is not a factor in
deciding excessiveness of bail.). People v. Saulnier, 492 N.Y.S.2d 897 (1985). People ex rel. Klein v. Krueger, 307 N.Y.S.2d 207
(1969).
195
People ex. rel. Benton v. Warden, 499 N.Y.S.2d 738, 740 (1986). People ex rel. Lobell v. McDonnell, 296 NY 109, 111 (1947).
196
People v. Torres, 446 N.Y.S.2d 969,972 (1981).
197
People v. Maldonado, 407 N.Y.S.2d 393, 394 (1978); People ex rel. Lobell v. McDonnell, 71 N.E. 2d 423, 425 (1947)(because
reasonable judges can and will vary widely with bail amounts, it is rare for their decisions to be overturned as
unconstitutionally excessive.). See In the matter of Henry r. Bauer, 818 N.E.2d 1113 (NY 2004) for an extreme and unusual case
in which the State Commission on Judicial Conduct decided to remove a judge from office for a range of misconduct, including
setting excessive bail amounts, e.g., 26 instances of bail set between $10,000 and $50,000 for defendants charged with petty
crimes or violations. In one case, bail was set at $25,000 for a defendant who was charged with riding a bike at night on a
sidewalk without appropriate lights.

59

Human Rights Watch | December 2010

Equal Protection
The Eighth Amendment may not require affordable bail, but does the Equal Protection
Clause? Does conditioning pretrial freedom on one’s financial resources violate
constitutional guarantees of equal protection? Years ago, the criminal law scholar Caleb
Foote wrote about the “incredible failure of the Supreme Court, courts in general and lawyers,
to do anything about what has become the most pervasive denial of equal justice in the
entire criminal justice system,” the setting of money bail for indigent defendants.198
Low socioeconomic status is not generally considered a class meriting special protection
under the Equal Protection Clause of the US Constitution. Nevertheless, the courts have
recognized that in the criminal justice arena, access to fundamental rights should not be
conditioned on the basis of an individual’s ability to pay. Thus, the Supreme Court has held
that although the constitution does not guarantee defendants a right to appeal, the denial of
trial transcripts for appellate purposes to indigent defendants who could not pay for them
was “invidious discrimination” that should have no place “in a country dedicated to
affording equal justice to all and special privileges to none in the administration of its
criminal law. There can be no equal justice where the kind of trial a man gets depends on the
amount of money he has.”199
The Supreme Court has found equal protection violated by laws that require the imprisonment
of defendants too poor to pay a fine. In one case, the Supreme Court confronted a state law
under which indigent defendants unable to pay state fines would be imprisoned beyond the
statutory maximum to “work off” the fines. The court ruled that statute constituted “invidious
discrimination” in violation of the Equal Protection Clause because:
[The statute] exposes only indigents to the risk of imprisonment beyond the
statutory maximum. By making the maximum confinement contingent upon
one’s ability to pay, the State has visited different consequences on two
categories of persons since the result is to make incarceration in excess of
the statutory maximum applicable only to those without the requisite
resources to satisfy the money portion of the judgment.200

Pugh v. Rainwater, 572 F.2d 1053, 1068 (5th Cir. 1978)(Simpson, J. dissenting).
Griffin v. Illinois, 351 U.S. 12, 17-19 (1956).
200
Williams v. Illinois, 399 U.S. 235, 242 (1970).
198
199

The Price of Freedom

60

In another case, the court ruled that it was a denial of equal protection to limit punishment
for traffic offenses to payment of a fine for those who are able to pay it, but to convert the
fine to imprisonment for those who are unable to pay it.201
In these two cases, the court rejected imprisonment solely because of indigency, finding that
the “Equal Protection Clause of the Fourteenth Amendment requires that the statutory ceiling
placed on imprisonment for any substantive offense be the same for all defendants
irrespective of their economic status.”202 The court has continued to look askance at laws
that lead to incarceration for poor but not wealthy defendants, finding equal protection
violations where the law “punish[es] a person for his poverty.”203 As one commentator has
recently concluded:
Together, these cases stand for the proposition that unequal access to a
fundamental right in the criminal justice system violates the Equal Protection
Clause if such discrimination is invidious, squalid, and unreasonable. These
ambiguous terms, read in the context of the cases, express the Court’s sense
of moral revulsion when the criminal justice system egregiously
discriminates on the basis of wealth. In particular, the Court is disinclined to
allow defendants to be split into two categories, such that poorer defendants
are incarcerated while wealthier defendants are released.204
The US Supreme Court has never ruled on whether incarceration because of inability to post
bail violates equal protection. “The entire concept of bail, it could be argued, conditions the
fundamental right of pretrial release upon a defendant’s wealth.”205 But different bail
amounts in different cases does not necessarily mean illegitimate discrimination: if bail
determinations are tailored on a case by case basis to a defendant’s resources and closely
tied to the purpose for which bail is sought (e.g. securing the defendant’s appearance in
court or protecting public safety), defendants “are treated equally with regard to their
relative ability to pay and are thus charged a relatively equal price for their liberty.”206 But
because the poor may have no money, they “may have nothing to offer as security for their
presence at trial other than their personal liberty, [they] may in effect have to pay a higher
201

Tate v. Short, 401 U.S. 395 (1971).

202

Ibid., at 399.

203

Bearden v. Georgia, 461 U.S. 660, 671 (1983).

204
Jonathan Zweig, “Extraordinary Conditions of Release under the Bail Reform Act ,” Harvard Journal on Legislation, vol. 44,
pp. 570-71. Zweig’s note addresses the inequity of wealthy defendants being able to buy pretrial release by creating, in
essence, private jails in their own homes.
205
Ibid., p. 574.
206
Laurence Tribe, American Constitutional Law (Mineola: The Foundation Press, 1988, 2nd ed.), p. 1634.

61

Human Rights Watch | December 2010

price than their solvent counterparts.”207 Money bail creates in practice two classes of
defendants: those who can buy their pretrial freedom and those who cannot.
The US Court of Appeals for the Fifth Circuit directly confronted the constitutionality of
imprisoning an indigent defendant prior to trial solely because he could not afford to pay
money bail.208 A panel of the court ruled that the Equal Protection Clause of the Fourteenth
Amendment was violated when an indigent defendant was condemned to pretrial
imprisonment for no other reason than his poverty and the bail system did not contain a
presumption against money bail in the case of indigents. The court noted that while the
Florida law at issue did not discriminate on its face, in practical operation it discriminated
impermissibly because a “man of means can secure his pretrial freedom while the indigent
has no choice but to remain in jail. A basic principle of equal protection is that ‘a law
nondiscriminatory on its face may be grossly discriminatory in its operation.’”209 To satisfy
equal protection a judge must consider less financially onerous forms of pretrial release
before imposing money bail.210 Although the full Court of Appeals en banc overturned the
panel decision on technical grounds,211 it agreed that “incarceration of those who cannot
[afford money bail], without meaningful consideration of other possible alternatives,
infringes on both due process and equal protection requirements.”212
New York courts have never closely examined whether jailing indigent defendants because
they are too poor to afford bail violates state equal protection guarantees. In the one class
action case raising an equal protection challenge to New York’s bail system, the court held
that challenges to the bail statute could not be made by class action lawsuit because bail
decisions are inherently case-specific. The court brushed aside plaintiffs’ equal protection
concerns without directly confronting whether imprisonment because of inability to meet
bail violated the Equal Protection Clause.213

207

Ibid., p. 1634, fn.7.

Pugh v. Rainwater, 557 F.2d 1189 (5th Cir. 1977), rev’d en banc, 572 F.2d 1053 (5th Cir. 1978).
209
Ibid., 557 F.2d at 1196, quoting from Williams v. Illinois, 399 U.S. 235, 242 (1970).
208

210
“We hold that equal protection standards are not satisfied unless the judge is required to consider less financially onerous
forms of release before he imposes money bail. Requiring a presumption in favor of non-money bail accommodates the
State’s interest in assuring the defendant’s appearance at trial as well as the defendant’s right to be free pending trial,
regardless of his financial status. Ibid, 557 F.2d at 1201, rev’d en banc, 572 F.2d 1053 (1978).
211
It did so primarily on grounds that revision of the bail rule that took place during pendency of appeal mooted the district
court’s decision and requires abstention from consideration of the rule of mootness and belief that federal court should
abstain from ruling on a new law that came into force during pendency of case.
212
Pugh v. Rainwater, 572 F.2d. 1053, 1057 (5th Cir. 1978). The court also noted, “We have no doubt that in the case of an
indigent, whose appearance at trial could reasonably be assured by one of the alternate forms of release, pretrial confinement
for inability to post money bail would constitute imposition of an excessive restraint.” Ibid.
213
Bellamy et al, v. Judges, 342 N.Y.S.2d 137 (1973), aff’d 346 N.Y.S.2d 812 (1973).

The Price of Freedom

62

International Human Rights Law
In any given year, about 10 million people worldwide will spend time in pretrial detention.
Indeed, one in three people behind bars is in pretrial detention.214 In the United States, 62
percent of the nation’s jail population are pretrial detainees.215 Outside the United States,
the proportion of the incarcerated population that consists of pretrial detainees is
considerably lower, varying from 47.8 percent in Asia to 20.5 percent in Europe.216
International human rights law permits the use of bail and other conditions of pretrial
release and it also permits pretrial detention. But any pretrial restrictions must be consistent
with the right to liberty, the presumption of innocence, and the right to equality under the
law.217 Pretrial detention imposed on criminal defendants accused of low level offenses
solely because they cannot afford bail is inconsistent with those rights.
Article 9(1) of the International Covenant on Civil and Political Rights (ICCPR), which the
United States ratified in 1992, codifies the right to liberty: “Everyone has the right to liberty
and security of person.”218 A person’s liberty may not be curtailed arbitrarily, either through
arbitrary laws or through the arbitrary enforcement of the law in a given case.219 To comply
with Article 9, “deprivation of liberty must be authorized by law” and “must not be
manifestly unproportional, unjust or unpredictable.”220 In reviewing the case of a Dutch
solicitor held in custody for nine weeks in the course of a criminal investigation, the United
Nations Human Rights Committee stated “‘arbitrariness’ is not to be equated with ‘against
the law,’ but must be interpreted more broadly to include elements of inappropriateness,
injustice and lack of predictability. This means that remand in custody pursuant to lawful
arrest must not only be lawful but also reasonable in the circumstances. Further, remand in

214

Martin Schönteich, “The Scale and Consequences of Pretrial Detention around the World,” Justice Initiatives: Pretrial

Detention (Open Society Justice Initiative), Spring 2008, pp. 11-14.

215
Todd D. Minton, Bureau of Justice Statistics, “Jail Inmates at Midyear 2009-Statistical Tables,” June 2010; Thomas H. Cohen
and Tracey Kyckelhahn, Bureau of Justice Statistics, “Felony Defendants in Large Urban Counties, 2006,” May 2010.
216
Ibid.
217
International human rights standards regarding pretrial detention are predicated not only on the fundamental rights, they
are also grounded in recognition of the stress and suffering detainees may endure from being confined in jail. In addition to
the emotional impact of confinement, detainees are separated from family, friends, and community, are uncertain as to their
future, worry about their legal position, and are anxious about their economic future and that of their family. See Centre for
Human Rights, Crime Prevention and Criminal Justice Branch, Human Rights and Pre-trial Detention: A Handbook of
International Standards relating to Pre-trial Detention (New York and Geneva: United Nations, 1994), p.8.
218
Article 9(1), International Covenant on Civil and Political Rights (ICCPR), adopted December 16, 1966, G.A. Res. 2200A (XXI),
21 U.N. GAOR Supp. (No. 16) at 52, U.N. Doc. A/6316 (1966), 999 U.N.T.S. 171, entered into force March 23, 1976, ratified by the
United States on June 8, 1992.
219
ICCPR, art 9(1). Manfred Nowak, UN Covenant on Civil and Political Rights: CCPR Commentary (Arlington: N P Engel
Publisher, 1993), p. 172.
220
Ibid., p. 173.

63

Human Rights Watch | December 2010

custody must be necessary in all the circumstances, for example, to prevent flight,
interference with evidence or the recurrence of crime.”221
Pretrial detention also implicates the presumption of innocence, affirmed in Article 14 of the
ICCPR as one of the necessary guarantees for a fair trial: “Everyone charged with a criminal
offence shall have the right to be presumed innocent until proved guilty according to law.”222
According to the United Nations Human Rights Committee, the presumption of innocence is
fundamental to the protection of human rights.223 No one should be prejudged or treated as
if assumed to be guilty, regardless of the likelihood of conviction.
Article 9(3) of the ICCPR explicitly addresses pretrial detention:
It shall not be the general rule that persons awaiting trail shall be detained in
custody, but release may be subject to guarantees to appear for trial, at any
other stage of the judicial proceedings, and, should occasion arise, for
execution of the judgment.
Article 9(3) authorizes pretrial release dependent on guarantees, which may be in the form
of money bail or other assurances. Indeed, Art. 9(3) has been interpreted as establishing a
presumption that detainees not be held when they could be released on bail.224 According to
the United Nations Centre for Human Rights:
Offenders should be released with the minimum controls necessary to
ensure their return to stand trial. Factors which indicate that a person is likely
to return even when released on his own recognizance are stable family and
social circumstances, current employment, and past conduct, including lack
of a criminal record or a history of complying with conditions in past criminal

221

Van Alphen v. the Netherlands, Human Rights Committee, No. 305/1988, §5.8, U.N. Doc. CCPR/C/39/D/305/1988 (1990).

222

ICCPR, art. 14(2).
“[B]y reason of the presumption of innocence, the burden of proof of the charge is on the prosecution and the accused has
the benefit of doubt. No guilt can be presumed until the charge has been proved beyond reasonable doubt. Further, the
presumption of innocence implies a right to be treated in accordance with this principle. It is, therefore, a duty for all public
authorities to refrain from prejudging the outcome of a trial.” UN Human Rights Committee, General Comment No. 13, Equality
before the courts and the right to a fair and public hearing by an independent court established by law (Twenty-first session,
1984), Compilation of General Comments and General Recommendations Adopted by Human Rights Treaty Bodies, UN Doc.
HRI/GEN/1/Rev.9 (vol. I)(2008), p. 185, para. 7.
224
Sarah Joseph, Jenny Schultz and Melissa Castan, ed., The International Covenant on Civil and Political Rights: Cases,
Materials and Commentary, 2nd ed. (New York: Oxford University Press, 2005), p. 330.
223

The Price of Freedom

64

proceedings. When these factors are present to a lesser extent…supervised
release is appropriate.225
International treaty bodies and authoritative interpretations of Article 9(3) are uniform in the
view that while pretrial detention may be permissible under certain circumstances, it should
be an exception and as short as possible.226 The maximum length of pretrial detention should
be proportionate to the maximum potential sentence.227 Of critical importance is the limitation
on imposing pretrial detention for offenses which are not punished with custodial sentences.
“If imprisonment is not to be expected as punishment for a crime, every effort should be made
to avoid pre-trial detention.”228 Indeed, according to the United Nation’s Centre for Human
Rights, certain crimes may be “so lacking in severity that pre-trial detention may be
inappropriate.”229 Pretrial detention may not be imposed arbitrarily and must be based on
grounds and procedure established by law. It should be limited “to essential reasons, such as
danger of suppression of evidence, repetition of the offence and absconding, and should be
as short as possible.”230 Seriousness of a crime is not in and of itself justification for pretrial
detention.231 Moreover, when concerns about flight risk or safety require some conditions on
pretrial release, to the extent possible non-custodial measures should be used rather than
pretrial detention; pretrial detention should be “a means of last resort.”232
The jurisprudence of the European Court of Human Rights with regard to Article 5.3 of the
European Convention on Human Rights, which parallels ICCPR Art 9(3), is instructive. The
level of bail set should not be set too high and should be aimed to ensure the presence of
the accused.233 The amount of the guarantee to be furnished by the detained person must be
assessed principally in reference to him and his assets,234 and a court’s failure to assess the
Centre for Human Rights, Human Rights and Pre-trial Detention, p. 15.
See for, example, UN Human Rights Committee, General Comment No. 8, Right to liberty and security of persons (Sixteenth
session, 1982), Compilation of General Comments and General Recommendations Adopted by Human Rights Treaty Bodies,
UN Doc. HRI/GEN/1/Rev.9 (vol. I)(2008), p. 179, para. 3: “Pre-trial detention should be an exception and as short as possible.”
United Nations Standard Minimum rules for Non-custodial Measures (The Tokyo Rules), G.A. res. 45/110. “Pre-trial detention
shall be used as a means of last resort in criminal proceedings…alternatives to pre-trial detention shall be employed at as
early a state as possible. Pretrial detention shall last no longer than necessary.”
227
Centre for Human Rights, Human Rights and Pre-trial Detention, p. 18
228
Centre for Human Rights, Human Rights and Pre-trial Detention, p. 16.
229
Ibid. p. 79.
230
Nowak, UN Covenant on Civil and Political Rights, p. 177.
231
Centre for Human Rights, Human Rights and Pre-trial Detention, p. 15.
232
United Nations General Assembly, Standard Minimum Rules for Non-custodial Measures (The Tokyo Rules), resolution
45/110, U.N. Doc. A/45/49 (1990), rule 6.1. See generally, Centre for Human Rights, Human Rights and Pre-trial Detention, p.
14 (Substitutes for Confinement).
233
Karen Reid, A Practitioner’s Guide to the European Convention on Human Rights, 3rd ed. (London: Sweet & Maxwell, 2008),
p. 460.
234
Neumeister v Austria, European Court of Human Rights, 1936/63 [1974] ECHR 1, May 7, 1974, para. 14; Can v Austria,
European Court of Human Rights, 9300/81 [1985] ECHR 10, September 30 1985, para. 69.
225

226

65

Human Rights Watch | December 2010

detainee’s ability to provide bail constitutes a violation of the right to pretrial release.235 As
the fundamental right to liberty is at stake, the authorities must take as much care in fixing
appropriate bail as in deciding whether or not the accused’s continued detention is
indispensable.236
Equality among all people has been deemed “the most important principle imbuing and
inspiring the concept of human rights.”237 The affirmation of equality and the prohibition of
discrimination are set forth in two provisions of the ICCPR.238 Article 26 of the ICCPR
establishes that all persons are equal before the law and entitled to equal protection of the
law. ICCPR Article 2(1) prohibits discrimination in the context of the rights and freedoms
enumerated in the convention, such as the right to liberty, and specifically identifies
“property” as prohibited grounds for discrimination.239
Concern abounds in human rights commentary and analysis about discrimination against
members of low income groups that prevents them from enjoying their fundamental rights.240
Particular attention has been paid to discrimination in the criminal justice system, and to the
fact that such discrimination frequently targets groups characterized not only by poverty but
also by being members of racial or ethnic minorities.241 Whatever the formal equality in
criminal justice system, “in practice access to the law and access to justice are fundamentally

235
236

Toshev v Bulgaria , European Court of Human Rights, 56308/00 [2006] ECHR 723, August 10, 2006, para. 68.
IsIwańczuk v Poland; Bojilov v Bulgaria, European Court of Human Rights, Commission’s report 1980, Decisions and

Reports 23, p. 196, para 171.
Nowak, UN Covenant on Civil and Political Rights, p. 458.
238
Equality and nondiscrimination operate as positive and negative statements of the same concept. Anne F. Bayefsky, “The
Principle of Equality of Non-Discrimination in International Law,” Human Rights Law Journal, vol. 11 no. 1 (1990).
239
ICCPR, Article 2(1) states: “Each State Party to the present Covenant undertakes to respect and to ensure to all individuals
within its territory and subject to its jurisdiction the rights recognized in the present Covenant, without distinction of any kind,
such as race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other
status.” See also, Article 2(2) of the International Covenant on Economic, Social and Cultural Rights (ICESCR), adopted
December 16, 1966, G.A. Res. 2200A (XXI), 21 U.N. GAOR Supp. (No. 16) at 49, U.N. Doc. A/6316 (1966), 993 U.N.T.S. 3, entered
into force January 3, 1976; General Conference of the United Nations Educational, Scientific and Cultural Organization,
Convention against Discrimination in Education, 429 U.N.T.S. 93, Art. 1. “For the purpose of this convention, the term
‘discrimination’ includes any distinction, exclusion, limitation or preference which, being based on race, colour, sex, language,
religion, political or other opinion, national or social origin, economic condition or birth, has the purpose or effect of nullifying
or impairing equality of treatment in education…” Joseph, Schultz and Castan, ed., The International Covenant on Civil and
Political Rights, p. 532.
240
See, for example, UN Committee on Economic, Social and Cultural Rights, “Non-Discrimination in Economic, Social and
Cultural Rights,” General Comment No. 20, UN Doc. E/C.12/GC/20, para. 35. “Individuals and groups of individuals must not
be arbitrarily treated on account of belonging to a certain economic or social group or strata within society. A person’s social
and economic situation when living in poverty or being homeless may result in pervasive discrimination, stigmatization and
negative stereotyping which can lead to the refusal or unequal access to the same quality of education and health care as
others as well as the denial or unequal access to public places.”
241
See, for example, UN Committee on the Elimination of Racial Discrimination, General Recommendation XXXI on the
Prevention of Racial Discrimination in the Administration and Functioning of the Criminal Justice System, UN Doc. A/60/18
(2005); Jamie Fellner, “Race, Drugs, and Law Enforcement in the United States,” Stanford Law and Policy Review, vol. 20 issue
2 (2009).
237

The Price of Freedom

66

unequal. The reasons may be social, economic or cultural, or the persons concerned may
suffer from social segregation or deep-seated discrimination.”242 The monitoring committee of
the International Convention on the Elimination of All Forms of Racial Discrimination urges
state parties to ensure that the “requirement to deposit a guarantee or financial security in
order to obtain release pending trial is applied in a manner appropriate to the situation of
persons in vulnerable groups, who are often in straitened economic circumstances, so as to
prevent the requirement from leading to discrimination against such persons.”243
Discrimination can occur even in the absence of overtly discriminatory laws; the test is
whether the nondiscrimination and equality guaranteed by human rights law are actually
enjoyed in practice. Prohibited discrimination can also occur in the absence of the intent to
discriminate. A law or practice may not have a discriminatory purpose, but will run afoul of
human rights law if it has an unjustifiable disparate impact adverse to the protected group.
Prohibited discrimination includes any distinction, exclusion, restriction, or preference
which has the purpose or effect of nullifying or impairing the recognition, enjoyment or
exercise by all persons, on an equal footing, of all rights and freedoms.244
Not every difference in treatment will constitute prohibited discrimination, and the
enjoyment of rights and freedoms on an equal footing does not mean identical treatment in
every instance.245 A distinction among various persons or groups of persons is permissible
when the criteria for differentiation are reasonable and objective, and where the aim is to
achieve a legitimate purpose under human rights law.246 International and regional human
rights bodies have also suggested that there must be a reasonable relationship of
proportionality between the legitimate aim and the means employed to attain it.247
Due regard for rights of liberty and to be free of discrimination mandate that the poor should
not be uniquely vulnerable to loss of liberty simply because they cannot afford bail that is
set without due regard to their financial ability to pay. The poor, like the rich, have a right to
liberty, and pretrial release should be equally available to both. The equal right of the

242

Leila Zerrougui, special rapporteur, “Administration of Justice, Rule of Law and Democracy: Discrimination in the Criminal
Justice System, Interim Report to the Commission on Human Rights,” UN Doc. E/CN.4/Sub.2/2005/7, para. 18.
243
Committee on the Elimination of Racial Discrimination, General Recommendation XXXI, para. 26.
244
UN Human Rights Committee, General Comment No. 18, Non-discrimination (Thirty-seventh session, 1989), Compilation of
General Comments and General Recommendations Adopted by Human Rights Treaty Bodies, UN Doc. HRI/GEN/1/Rev.9 (vol.
I)(2008), p. 197, para. 7.
245
Ibid., para. 6.
246
Ibid., para. 13.
247
See, for example, Relating To Certain Aspects Of Laws On Use Of Languages In Education In Belgium - 1474/62; 1677/62;
1691/62 [1967] ECHR 1 (9 February 1967), paras. 11-12; Proposed Amendments to the Naturalization Provisions of the
Constitution of Costa Rica, Advisory Opinion OC-4/84, January 19, 1984, Inter-Am. Ct. H.R. (Ser. A) No. 4 (1984), p. 172, para. 56;
Bayefsky, “The Principle of Equality or Non-Discrimination in International Law,” p. 14.

67

Human Rights Watch | December 2010

indigent to pretrial liberty should be upheld by ensuring bail is not be set at an excessively
high figure which might preclude a detainee from being able to raise it.248
If New York City had no other options than cash bail or secured bonds for ensuring the
appearance of poor people (who are primarily black and Hispanic) at court proceedings, the
human rights considerations might be different. But judges have the authority and tools for
setting bail that is within reach of persons of limited means and consistent with the goal of
ensuring their appearance in court. Supervised release would protect the liberty of poor
defendants while ensuring their appearance in court. It would be a far more proportionate
response than pretrial detention to the risk of flight for offenses which are rarely punished
with imprisonment.
In 1689, the English Bill of Rights prohibited excessive bail, understood as an amount that
was more than the accused could pay, or deployed to impose indefinite imprisonment.249
The legal conception of “excessive bail” in many common law jurisdictions today remains
true to that definition. Bail must not be fixed at a figure so large as to lead to inevitable
imprisonment.250 New York City should return to that original and still compelling, even if not
constitutionally required, notion of “excessive.”

248
Yoram Dinstein, “The right to life, physical integrity, and liberty,” in Louis Henkin, ed., The International Bill of Rights: The
Covenant on Civil and Political Rights (New York: Columbia University Press, 1981), p. 134. See also, Karen Reid, A
Practitioner’s Guide to the European Convention on Human Rights, p. 460 (“The level of bail should not be set too high,

interpreting Art. 5.3 of the European Convention on Human Rights, entitling detainee to release pending trial and release may
be conditioned by guarantees to appear for trial.”).
249
Laurence Claus, “Methodology, Proportionality, Equality: Which Moral Question does the Eighth Amendment Pose?”
Harvard Journal of Law and Public Policy, vol. 31, p. 37.
250
The People v. Roger O’Callaghan, Supreme Court of Ireland [1966] 1 I.R., p.518; DPP v. Broderick, Supreme Court of Ireland
[2006] IESC 34.

The Price of Freedom

68

Acknowledgments
This report was researched and written by Jamie Fellner, senior counsel in the US Program of
Human Rights Watch. The report was edited by Alison Parker, US Program director; Dinah
Pokempner, general counsel; and Joe Saunders, deputy program director. Abigail Marshak,
former associate in the US Program, Vikram Shah, associate in the US Program and two US
Program interns, Lendon Ebbels and Chelsea Rinnig, provided research assistance. Naureen
Shameem, legal intern, undertook international legal research for the report. Brian Root,
consultant, reviewed the statistical data.
Vikram Shah, associate in the US Program provided editing and production assistance. Anna
Lopriore, creative manager, Grace Choi, publications director, and Fitzroy Hepkins, mail
manager, provided production assistance. We also wish to thank Marcia Allina, Marina
Kaufman, David Nachman, and Ricki Roer for their help with interviews for this report.
This report draws on previously unpublished statistical data generously provided to Human
Rights Watch the New York City Department of Correction, the New York State Division of
Criminal Justice Services, and the New York City Criminal Justice Agency. From December
2009 and continuing through mid-October 2010, Jamie Fellner submitted various requests
for data to these agencies and discussed the requests with them to ensure that
appropriately defined, and statistically significant data could be included in this report. The
data they provided is on file at Human Rights Watch. Any errors in the reproduction or
interpretation of the data are the sole responsibility of Human Rights Watch.
This report benefitted immeasurably from the assistance provided by the New York City
Criminal Justice Agency, whose executive director, Jerome E. McElroy, provided strong
support for our research. We wish to single out for special thanks Mary T. Phillips, deputy
director of the CJA Research Department, who not only provided new data and analysis to
Human Rights Watch for this project but who also answered our many questions with infinite
patience. We are also extremely grateful to the New York State Division of Criminal Justice
Services and Marge Cohen, program research specialist, who was also extremely helpful
compiling for us and explaining previously unpublished data on the case outcomes of New
York City misdemeanor arrests. We also wish to thank Eric Sorenson, director of Population
Research for the New York City Department of Correction for providing us with new statistics
on the city’s jail population and admissions.

69

Human Rights Watch | December 2010

This report reflects interviews with scores of New York City judges (most of who preferred to
remain anonymous), prosecutors (who also preferred anonymity), defense attorneys,
defendants and their families, advocates, and academics. We are indebted to the numerous
attorneys with The Bronx Defenders, Bronx Freedom Fund, Neighborhood Defender Service of
Harlem, The Legal Aid Society, and Brooklyn Defender Services who took time to meet with
us and let us sit with them during arraignments. We appreciate the willingness of Kings
County District Attorney Charles Hynes, Bronx County District Attorney Robert Johnson,
Manhattan District Attorney Cyrus R. Vance, Jr., and former Manhattan District Attorney
Robert Morgenthau for being willing to talk with us. We also received many valuable insights
from interviews with Martin Horn, distinguished lecturer at John Jay College of Criminal
Justice; Michael Jacobson, executive director of the Vera Institute of Justice; Harry G. Levine,
professor of sociology at Queens College; Timothy J. Murray, executive director of The Pretrial
Justice Institute; Judge Juanita Bing Newton, dean of the New York State Judicial Institute;
Susan Shaffer, director, and other senior staff at the District of Columbia Pretrial Services
Agency; Vincent Schiraldi, commissioner of the New York City Department of Probation;
Herbert Sturz, senior advisor at Open Society Institute; and Jeremy Travis, president of John
Jay College of Criminal Justice.
Human Rights Watch would like to thank Peter B. Lewis for generously supporting the work
of the US Program.

The Price of Freedom

70

H UMA N R I G H TS WATCH
350 Fifth Avenue, 34 th Floor
New York, NY 10118-3299

H U M A N

www.hrw.org

W A T C H

The Price of Freedom
Bail and Pretrial Detention of Low Income Nonfelony Defendants in New York City
Thousands of defendants in New York City accused of minor crimes are held in pretrial detention each year solely
because they cannot afford to pay even small amounts of bail. The Price of Freedom—based on scores of
interviews with defendants, family members, judges, prosecutors, and defense attorneys, and a trove of new
data—analyzes why this is happening and what can be done to ensure greater equity in the bail process.
Previously unpublished data made available to Human Rights Watch by the NYC Criminal Justice Agency (CJA)
shows that in 87 percent of cases of nonfelony defendants arrested in 2008 in which bail was set at $1,000 or
less (the most recent year for which such data is available), the defendants were not able to post bail at
arraignment. On average, such individuals spent some 16 days in pretrial detention. Almost three out of four such
individuals were accused of nonviolent, non-weapons related crimes such as shoplifting, turnstile jumping,
smoking marijuana in public, or trespassing.
The Price of Freedom recommends that New York City develop a pretrial supervised release program to allow more
nonfelony defendants to remain free while awaiting trial. This approach would honor the presumption of
innocence but cost far less than housing, feeding, guarding, and providing medical care to inmates confined
round the clock in jail. The report also calls for reforms requiring judges to more carefully tailor their bail decisions
to defendants’ financial resources, including wider use of unsecured appearance bonds for those accused of
misdemeanors.

Otis Bantum Correctional Center,
a New York City Department of
Correction facility on Rikers Island
that houses detained adult men.
© 2002/2010 Andrew Garn

R I G H T S