Gatekeepers: The Role of Police in
Ending Mass Incarceration
S. Rebecca Neusteter, Ram Subramanian, Jennifer Trone,
Mawia Khogali, and Cindy Reed
August 2019

From the Director
Reforming the criminal justice system has become
a bipartisan priority and a topic of intense public
interest. Much of the focus is on reversing mass
incarceration—lowering the numbers of people
in prison and jail, creating constructive pathways
for people returning to their communities, and
addressing the stark racial and ethnic disparities
that have been a primary feature of the American
criminal justice system. This work remains essential:
in the United States, half of all adults have an
immediate family member who is or has been
incarcerated. Moreover, black adults are 50 percent
more likely than their white counterparts to have
had an incarcerated immediate family member.
But ending the practice of mass incarceration and
repairing its extensive collateral consequences
must begin by focusing on the front end of the
system: police work. A police officer’s encounter
with a civilian, if it ends in custodial arrest, is where
mass incarceration begins. And the enforcement
numbers—most notably arrests—are staggering.
According to the Federal Bureau of Investigation
(FBI), as of 2017, some 70 million Americans had
been arrested. If everyone with an arrest record
held hands, they would circle the globe three times.
	
Researchers looking at arrest data for the years
1997 to 2008 found that one out of three young
adults—and nearly half of all black men—had an
arrest record by age 23. It wasn’t always this way;

young adults today are 36 percent more likely to be
arrested than were their parents at the same age.
And something else has changed: a person arrested
today is far more likely to land in jail.
Rewinding to the moment of arrest, several striking
patterns emerge. First, police disproportionately use
enforcement against black people. Second, mental
illness, drug and alcohol abuse, homelessness, and
poverty are frequently the subtexts of encounters
with police—in fact, police have become the default
first responders to numerous social issues that
they are neither trained nor equipped to properly
handle. And third, officers—flooded with 911 calls
for things that might not constitute true public
safety emergencies—often respond with the tool
most expedient and available to them by arresting
or citing people for low-level offenses.
Beyond unleashing the steep human and financial
costs of jail time, punitive actions by police often
saddle already poor people, and a disproportionate
number of people of color, with onerous fines and
fees and limit their life opportunities by branding
them as criminals. Moreover, a punitive approach
to law enforcement makes policing vastly more
dangerous for both members of the public and
officers, creates ripples of social and emotional
harm that burden entire communities, and drives
a wedge between officers and the people they are
sworn to serve and protect.

All of these dynamics are well known. Yet efforts
to address them at their point of origin, when an
officer is deciding whether to arrest someone, have
been far more limited compared with interventions
at points downstream—from bail reform, to the
election of progressive district attorneys, to
campaigns to remove criminal record information
from employment criteria. Some police departments
are experimenting with alternatives to arrest,
but those efforts are largely ad hoc, with little
systematic examination of what works and why.
That’s not enough to change the front end of the
criminal justice system in ways that address racial
disparities in criminal justice contact and reduce
mass incarceration.
We expect police to support community safety
and security, a job that often has negative
consequences for the officers’ mental and physical
health. Yet because of insufficient training and
resources, perverse organizational incentives,

individual and systemic biases, and other drivers,
police often default to enforcement in lieu of less
intrusive interventions. Recognizing the roughly
18,000 police agencies around the country as
gatekeepers of the system, this report explores
the factors driving mass enforcement, particularly
of low-level offenses; what police agencies could
do instead with the right community investment,
national and local leadership, and officer training,
incentives, and support; and policies that could shift
the policing paradigm away from the reflexive use
of enforcement, which unnecessarily criminalizes
people and leads directly to the jailhouse door.

S. Rebecca Neusteter
Policing Program Director
Vera Institute of Justice

About this report
This report is one of a series that the Vera Institute of Justice (Vera) is
releasing with the Safety and Justice Challenge—the John D. and Catherine
T. MacArthur Foundation’s initiative to reduce overincarceration by
changing the way America thinks about and uses jails. The initiative
is supporting a network of competitively selected local jurisdictions
committed to finding ways to safely reduce jail incarceration. Other
publications in the series to date include:
››
››
››
››
››
››
››
››

Incarceration’s Front Door: The Misuse of Jails in America;
The Price of Jails: Measuring the Taxpayer Cost of Local Incarceration;
Overlooked: Women and Jails in an Era of Reform;
Out of Sight: The Growth of Jails in Rural America;
Divided Justice: Trends in Black and White Incarceration 1990–2013;
The New Dynamics of Mass Incarceration;
Justice Denied: The Harmful and Lasting Effects of Pretrial Detention; and
the multimedia storytelling project, The Human Toll of Jail.

Through the Safety and Justice Challenge, Vera’s office in New Orleans,
and direct partnerships with jurisdictions nationwide, Vera is providing
expert information and technical assistance to support local efforts to
stem the flow of people into jail, including using alternatives to arrest and
prosecution for minor offenses, recalibrating the use of bail, and addressing
fines and fees that trap people in jail. For more information about Vera’s
work to reduce the use of jails, contact Nancy Fishman, project director,
Center on Sentencing and Corrections, at nfishman@vera.org. For more
information about Vera’s policing work, contact Rebecca Neusteter, policing
program director, at RNeusteter@vera.org. For more information about the
Safety and Justice Challenge, visit www.safetyandjusticechallenge.org.

Gatekeepers: The Role of Police in Ending Mass Incarceration

v

Contents
1	

Introduction

9	

Mass enforcement by the numbers

	

10	

Volume of arrests nationally

	

12	

Serious or not?

	

15	

Lesser forms of enforcement

	

16	

Disparate racial impact

	

18	

The revolving door

20	

What’s driving the overreliance on punitive enforcement?

	

22	

Person-to-person factors

	

23	

Messages from above

	

27	

Perverse incentives

	

27	

Lack of alternatives

36	

Shifting the paradigm

	
37	
		

Identify, promote, and invest in alternatives to enforcement 	
that don’t involve the criminal justice system

	

Reengineer the 911 system

42	

	
44	
		

Increase the number and types of offenses that don’t 		
require punitive enforcement

	

48	

Expand the reach of alternatives to arrest

	

49	

Create structural incentives for police to use alternatives

	
50	
		

Find out what works best through experimentation, 		
research, and analysis

52 	 Conclusion
55	Endnotes

Introduction

E

veryone is familiar with the image of a police car driving away
with a person in handcuffs—but to where? In the United States,
some people wind up at a police station, from which they are later
released. More often, however, people in police custody eventually land in
local jails—county or municipal detention facilities that primarily house
people who have been charged but not yet convicted of a crime. In fact, the
likelihood that arrest will lead to jail incarceration has increased steadily

Growth in jail admissions as crime and arrest
rates have fallen to lows not experienced
since 1970 and 1980, respectively, strongly
suggests that police enforcement has
become an expressway to jail.

over the years. For every 100 arrests police officers made nationwide
in 2016—the most recent year for which data is available—there were
99 jail admissions.1 Twenty-five years ago, when crime rates and arrest
volume overall were higher, the ratio of arrests to jail admissions was
much lower—there were 70 jail admissions for every 100 arrests.2 While
not all jail admissions stem from arrests—people suspected of violations
of probation or parole can end up in jail too, for example—the growth in
admissions as crime and arrest rates have fallen to lows not experienced
since 1970 and 1980, respectively, strongly suggests that between then and
now, police enforcement has become an expressway to jail.3 (See Figure 1
on page 2.)

Gatekeepers: The Role of Police in Ending Mass Incarceration

1

Figure 1

Jail admissions and arrest volume, 1980–2017

18M
Arrests

10M

A decline
in arrest
volume...

ns
dmissio

Jail a

...together
with an
increase
in jail
admissions...

...suggests that
police enforcement
has become an
expressway to jail.

17
20

04
20

92
19

19

80

0

Sources: For arrests from 1980–2016, see Vera Institute of Justice, “Arrest Trends," https://arresttrends.vera.org/arrests?unit=rate#bar-chart and for arrests in 2017, see
Federal Bureau of Investigation, “2017 Crime in the United States: Persons Arrested,” https://perma.cc/W7GJ-GPDV. For jail admissions from 1980–2015, see Vera Institute
of Justice, “Incarceration Trends”; for jail admissions in 2016, see Zhen Zeng, Jail Inmates in 2016 (Washington, DC: Bureau of Justice Statistics, 2018), https://perma.cc/
HN73-Q8FY; and for jail admissions in 2017, see Zhen Zeng, Jail Inmates in 2017 (Washington, DC: Bureau of Justice Statistics, 2019), https://perma.cc/FD5J-7ZUC.

With a growing consensus that local jails are a primary locus of mass
incarceration, data on arrest trends points to an urgent need to focus more
deliberately on one of the problem’s primary points of origin: policing
practices. (For more on arrest data, see “Arrest Trends: Unlocking police
data” on page 18.) Police officers, as gatekeepers of the criminal justice
system, hold almost exclusive authority—by way of citations, arrests,
and even physical force—to enforce and regulate the law. And they have
increasingly been asked to do this in situations that involve societal
problems that would be better resolved in the community—problems like
homelessness, mental illness, and substance use. Although arrest volume
is down across almost all offense categories since its high-water mark of
15 million in 1997, nationally there are still roughly 28,000 arrests every
day, which equates to one arrest every three seconds or approximately 10.5
million every year.4 By virtue of their arrest, all these people face probable
jail incarceration. This volume does not reflect an increase in arrests for
serious crimes. In fact, the proportion of serious violent crimes among
all arrests—less than 5 percent—has not changed in decades.5 Rather,
arrests most often occur in response to minor offenses—including drug
use violations and disorderly conduct—which account for more than 80
percent of total arrests.6 This mass enforcement of relatively minor law
2

Vera Institute of Justice

violations suggests that policing practices currently tend toward punitive
approaches—that is, those that prioritize arrest and frequently lead to time
behind bars—in ways that are often not necessary to achieve public safety.7
But jail incarceration isn’t the only negative consequence of police
overreliance on arrest. Just a few days in jail increase a person’s likelihood
of being found guilty, receiving a harsher sentence, and committing a
future crime.8 As a result of a bewildering number of legal and regulatory
penalties, disabilities, or disqualifications that flow from involvement
with the criminal justice system, people with arrest records—even if the
charges against them are later dropped—have a harder time maintaining or
finding employment, credit, or housing.9 In one state alone, Maryland, the
American Bar Association’s National Inventory of Collateral Consequences
of Conviction (NICCC) catalogued 348 occupational licensing or
certification restrictions limiting employment opportunities for people
with a criminal record, including those merely accused of a crime.10

Someone arrested today is much more likely to spend time in jail

THEN

100

arrests

1994

2016

In 1994, for every

In 2016, for every

there
were

70

jail admissions

100

arrests

there
were

NOW

99

jail admissions

Source: For 1994: Craig A. Perkins, James J. Stephan, and Allen J. Beck, Jails and Jail Inmates 1993-94 (Washington, DC: Bureau of Justice Statistics, 1995), 13, https://
bjs.gov/content/pub/pdf/jaji93.pdf. For arrests in 2016, see Vera Institute of Justice, “Arrest Trends," (arrest rates from 1980–2016), https://perma.cc/QCY2-TWJD; for jail
admissions in 2016, see Zhen Zeng, Jail Inmates in 2016 (Washington, DC: Bureau of Justice Statistics, 2018), https://perma.cc/HN73-Q8FY.

Gatekeepers: The Role of Police in Ending Mass Incarceration

3

Young people who are arrested but not convicted are more likely to be
impoverished compared with those who have a similar background but no
history of arrest.11 Overreliance on enforcement also produces profound
emotional impacts. Anxiety and other symptoms of trauma are more
common among those stopped by police.12 The more intrusive the stops, the
more pronounced the negative effects—an especially concerning finding
given that a history of trauma is itself strongly correlated with subsequent
justice system involvement and incarceration.13 Moreover, some people are
arrested repeatedly without ever having their underlying needs identified
or treated, precipitating a cycle of enforcement that for some ends only on
death and results in significant human and financial costs.14
American policing’s overreliance on punitive enforcement—
particularly for low-level offenses—has also had real life consequences for
communities of color that extend far beyond the arrest of any one person.15
Communities of color have endured practices such as stop, question,
and frisk; excessive enforcement of minor offenses; and concentrated
policing in targeted areas far disproportionate to their percentage of the
population and likelihood of violating the law. All of this has resulted in
increasingly fractured community-police relationships—and far worse.16 In
2016, for example, black people were arrested at more than twice the rate
of whites, approximately 5,313 and 2,444 per 100,000, respectively.17 This
rate has remained consistent for the past 15 years.18 Police violence against
black Americans is regularly documented in the media.19 Numerous highprofile deaths of unarmed black people during seemingly routine police
encounters——from Michael Brown in Ferguson, Missouri, to Eric Garner
in Staten Island, New York, to Sandra Bland in Waller County, Texas,
and many others—have further galvanized community mistrust in law
enforcement, leading to national movements such as #BlackLivesMatter
that aim to build local power to end police violence against black and other
marginalized communities.20 According to the Washington Post, which
tracks police shootings of civilians through reviews of media reports in
its “Fatal Force” database, an estimated 996 people were shot and killed by
police during 2018, following 987 such deaths in 2017—disproportionately
people of color.21
Policing in the United States has historically been used by powerful
elites as a mechanism of social control over marginalized communities,
from its roots in the subjugation of black people from slavery to the Jim
4

Vera Institute of Justice

Crow era to the Civil Rights movement; to the ethnic violence committed
against Mexican-Americans by the Texas Rangers; to the targeting of
LGBTQ people at Stonewall; to recent collaborations between some local
law enforcement agencies and Immigration and Customs Enforcement
(ICE) to apprehend immigrants.22 Deep cleavages between police and
communities persist, as evidenced by findings of the U.S. Department of
Justice following investigations of local police departments in Baltimore,
Chicago, Cleveland, and Ferguson.23 Such rifts continue to raise questions
of basic fairness, undermining the perceived legitimacy of police by the
public they serve and heightening the risk that communities may be
unwilling to rely on and cooperate with police in order to help prevent and
solve crimes.24

This mass enforcement of relatively minor
law violations suggests that policing
practices currently tend toward punitive
approaches in ways that are often not
necessary to achieve public safety.

Overrelying on punitive enforcement—particularly for low-level
offenses—can also have negative effects on police departments. It can
place officers themselves in jeopardy, either directly—in 2017, according
to the FBI, more than 60,000 police officers were assaulted on the job
and nearly 100 died—or indirectly by harming their physical and mental
well-being.25 Officer suicide rates, for example, far exceed that of the
general population.26 In part because of these risks, policing today faces a
crisis in recruitment and attrition, resulting in many police departments
being understaffed; officers excessively working overtime, contributing
to exhaustion and other personal and intrapersonal challenges; and, in

Gatekeepers: The Role of Police in Ending Mass Incarceration

5

some instances, agencies lowering the standards to become a police officer
simply to fill the ranks, resulting in a less qualified workforce.27
Yet despite the harm it has brought to individuals, communities,
and police, enforcement remains an officer’s principal tool. To chart a
new course in American policing, police should use arrest sparingly,
intentionally, and transparently. To foster dialogue and set out a roadmap
to achieve this goal, this report explores national data to understand
the scope of mass enforcement today—most specifically examining the
wide reach of arrests through publicly available data—and discusses the
determinants of its growth. It explores some of the reasons why police
officers currently default to punitive action and identifies some of the
dynamics that are likely at play, including agency-level messages and
policies set by police leadership, as well as an officer’s level of experience,
temperament, and personal biases. It also analyzes larger structural
incentives that drive the use—and overuse—of enforcement, such as widely
adopted police performance measures that reward detection, capture, and
sanction while ignoring or even chastising what are often considered
softer responses. And it highlights the scarcity of effective alternatives to
punitive enforcement, reviewing research into the reach and efficacy of
such programs.
What is clear is that far from being synonymous with rote enforcement
of laws, policing actually operates in a gray area: it depends on an officer’s
judgment, in which the law is just one variable and enforcement only one
of many possible responses. Police have wide discretion to make choices
other than arrest: they can, for example, choose to cite and release, issue a
warning, or do nothing. There is always a risk that officers can misuse or
abuse such discretion. But police departments can encourage officers to use
their discretion to rein in punitive enforcement and to employ what are
often safer and more effective responses.
As this report discusses, a number of mostly small-scale experiments
in alternatives-to-arrest programs provide some evidence of what police
can choose to do, along with some cautionary lessons for their future
replication. They are just the beginning of reimagining the gatekeeper role
for more effective and equitable public safety outcomes.
This change is not the responsibility of police alone. The problems that
have led to mass enforcement are, to some extent, the result of societal
issues that have been laid at the feet of police but are not theirs alone to
6

Vera Institute of Justice

solve. Addressing issues like mental illness, homelessness, and substance
use will require a coordinated effort by local elected officials, public
health providers, and social services agencies working with and investing
in community-based services to prevent crime at its roots. An open
exchange and dialogue between local officials, police, and the community is
needed to come up with alternative solutions. And all of this will require
commitment by elected officials and agency-level leadership to make the
necessary investments and stay the course to effect organizational change
in entrenched policing culture.

To chart a new course in American
policing, police should use arrest
sparingly, intentionally, and
transparently.

This paper concludes by proposing a set of actions that, if applied
collectively nationwide, have the potential to stem the tide that is pushing
people toward the jailhouse door and into deeper involvement in the
criminal justice system. They include
›› identifying alternative responses to societal problems outside the

criminal justice system, such as reengineering the nation’s 911
systems and investing in community-based resources, that if taken
to scale could establish other entities beyond police agencies as the
default responders to noncriminal but critical circumstances;
›› homing in on categories of offenses that do not require police
enforcement, recognizing that in many instances it is not the best
response;

Gatekeepers: The Role of Police in Ending Mass Incarceration

7

›› expanding the reach and scope of current alternative-to-arrest

programs, policies, or procedures, in particular by rethinking how
police should respond to violence;
›› investing in institutional changes that support and reward officers
when they properly use nonpunitive problem-solving tools in the
service of justice and public safety (such as referring people to
needed services); and
›› researching and evaluating the impact of implemented policing
reforms to ensure they are achieving their intended outcome by, for
example, studying whether increasing the use of citations in lieu of
arrest is a genuine pathway away from—rather than a back door to—
incarceration.

This work will likely require systemic
shifts in how everyone from local elected
officials to the public to law enforcement
agencies views policing itself.

These steps may require leadership to provide clear guidance on how
officers should exercise their discretion—or even limiting that discretion
outright—in order to reduce an overreliance on enforcement. And such
guidance should be based on arrest data, particularly with the aim of
reducing unwarranted disparities across race, ethnicity, and gender. But,
more broadly, this work will likely require systemic shifts in how everyone
from local elected officials to the public to law enforcement agencies views
policing itself.

8

Vera Institute of Justice

Mass enforcement by
the numbers

B

y and large, policing in America is a local endeavor. Most of the
roughly 18,000 police agencies nationwide are city or county entities.28
State police, by comparison, generally have less contact with the
public, except in isolated rural areas and through highway patrol.29 Because
policing happens and is overseen locally, there is very little national-level
data quantifying and describing it. Most existing national data focuses
on arrests, although there have been some attempts—mainly at the local
level—to collect and understand data on other enforcement activities, such
as citations in lieu of arrest. The International Association of Chiefs of
Police, for example, surveyed U.S. police departments in 2016 and found that
nearly 87 percent of responding agencies offer officers the option of writing

Defining arrest
The Uniform Crime Reporting Program (UCR)—launched
more than 80 years ago by the FBI—is a nationwide,
cooperative, statistical effort to collect and collate data
on crimes brought to the attention of law enforcement
agencies.a Today, many local law enforcement agencies
voluntarily report data through this program.
The UCR is widely known for generating the most reliable
and comprehensive set of statistics on crimes reported to
the police. This program is also responsible for the most
expansive data available nationally on enforcement through
the “persons arrested” tables. These tables include each
separate instance—categorized by month, age, sex, race,
and ethnicity—in which a person is arrested, cited, or
summoned for all violations, except traffic offenses. But
because reporting data to the UCR program is voluntary,
it does not capture information on every law enforcement
agency, arrest, or demographic characteristic. To account

for missing data, the FBI adjusts the reported data with
estimated counts. For the purposes of the UCR program, the
FBI defines an arrest as “when a law enforcement officer
detains an adult with the intention of seeking charges against
the individual for a specific offense(s) and makes a record of
the detention.”b (For more information on persons arrested,
see Vera’s report, Every Three Seconds: Unlocking Police
Data on Arrests, the associated data visualization tool Arrest
Trends, and “Arrest Trends: Unlocking police data” on page 18
of this report.)

a	

FBI, “Uniform Crime Reporting (UCR) Program,” https://perma.
cc/6DXX-26HL.
b	

FBI, Uniform Crime Report Program: Summary Reporting System
User Manual (Washington, DC: FBI, 2013), 140, https://perma.
cc/53TX-Y499.

Gatekeepers: The Role of Police in Ending Mass Incarceration

9

citations in lieu of arrest. However, data on and research demonstrating the
application and impact of this practice is currently limited.30

Volume of arrests nationally
In the most recent published data—for 2017—the FBI reported an adjusted
figure of more than 10.5 million arrests.31 (For an explanation of the FBI
methodology in arriving at this figure, see “Defining arrest” on page
9.) Even with the adjustment, the FBI count is generally regarded as
conservative, meaning the actual number of arrests—combined with other
forms of punitive enforcement—is likely much higher.
Although arrests have been declining for more than a quarter-century,
10.5 million is still a lot of arrests, even in a country with more than 300
million people.32 Arrests are not spread evenly across the population; they
are concentrated among both younger people and marginalized people and
disproportionately affect people of color. In 2016, the most recent year for
which data on racial disparities is available, black Americans made up 27

Black Americans are disproportionately arrested

Black Americans made up 27%
of all arrests in 2016...
...while constituting just 13% of the U.S. population.

% of U.S. population
% of arrests

Sources: FBI, “2016 Crime in the United States,” table 21, https://perma.cc/CCK5-P8A9; and U.S. Census Bureau, “QuickFacts,” https://www.census.gov/quickfacts/fact/
table/US/RHI225218.

10

Vera Institute of Justice

percent of all arrests nationally even though they constitute just 13 percent
of the U.S. population, and Native Americans were arrested at nearly
double the rate of white people.33 According to one study, fully one out of
three Americans will have experienced arrest by age 23.34 Among young
black men, that number is almost one in two.35 (The FBI has not collected
arrest data by ethnicity since 1991, so current Hispanic/non-Hispanic
arrest rates are not known.)36
Nor are arrests spread evenly geographically. Police enforcement is
largely concentrated in metropolitan and suburban areas. Suburbs within
metropolitan areas have the highest average arrest rates (at 4,604 per

Arrest numbers in the United States are staggering

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
1 out of 3
Americans
will have
experienced
arrest by
age 23.

The rate
among young
black American
men is almost...

1

in

2.

Source: Robert Brame, Michael G. Turner, Raymond Paternoster, and Shawn D. Bushway, “Cumulative Prevalence of Arrest from Ages 8 to 23 in a National Sample,”
Pediatrics 129, no. 1 (2012), 21-27; and Robert Brame, Michael G. Turner, Raymond Paternoster, and Shawn D. Bushway, “Demographic Patterns of Cumulative Arrest
Prevalence by Ages 18 and 23,” Crime & Delinquency 60, no. 3 (2014), 471-486, https://perma.cc/E49B-SGHY.

Gatekeepers: The Role of Police in Ending Mass Incarceration

11

100,000), followed by cities outside of metropolitan areas (4,090 per
100,000).37 Principal cities—cities in a metropolitan area with populations
of 50,000 people or more—and nonmetropolitan counties have the lowest
arrest rates at 3,332 per 100,000 and 2,322 per 100,000 respectively.38 In
2017, fewer than 20 percent of the nation’s total arrests were made outside
of cities and suburban areas.39

Serious or not?
The FBI includes type of offense in the arrest data but doesn’t always
distinguish between more and less serious forms of the same type of
crime. It is impossible, for example, to separate shoplifting from grand
larceny—both are grouped under the category “larceny/theft.”40 This way
of organizing the data precludes drawing a broad distinction between
arrests for serious versus nonserious offenses (such as between felony
and misdemeanor arrests.) Nevertheless, the FBI data provides compelling
evidence that very few arrests annually are for the most harmful crimes
and, conversely, a large number of arrests are for behaviors commonly
considered nuisance offenses.41

Serious violent offenses have constituted
less than 5 percent of all arrests annually
for decades, despite fluctuations in the
total number of arrests year to year.

For example, in 2017, just under 5 percent (or 518,617) of all recorded
arrests pertained to the four crimes that involve the most significant
violence or threat of violence against a person: murder and nonnegligent
manslaughter, rape, robbery, and aggravated assault.42 This followed a
12

Vera Institute of Justice

Figure 2

Total arrests and arrests for serious violent offenses, 1980–2017

18M

Total arrests

10M

Serious violent
arrests (Part I
violent)

17
20

04
20

92
19

19

80

0

Source: Derived from the FBI’s Uniform Crime Reporting program, available at FBI, “Uniform Crime Reporting (UCR) Program,” https://fbi.gov/services/cjis/ucr. Part I
violent crimes include murder, nonnegligent manslaughter, rape, robbery, and aggravated assault.

lengthy trend: serious violent offenses have constituted less than 5 percent
of all arrests annually for decades, despite fluctuations in the total number
of arrests year to year.43 (See Figure 2 on page 13.)
At the other end of the spectrum, there were 1.2 million arrests in 2017
for disorderly conduct, public drunkenness and other liquor law violations
(not including driving while under the influence), vagrancy, loitering and
curfew violations, and vandalism.44 And, even though drug arrests are
down in recent years after an increase of 171 percent between 1980 and
2016, they continue to account for more than 1.5 million arrests annually—
the vast majority of which are made for drug possession generally, and
marijuana possession most often.45 (See Figure 3 on page 14.)
Independent studies of local arrest practices corroborate the national
data that suggests that significantly more arrests are made for minor
offenses than for serious ones. For example, in 2016, the New York
City Police Department made twice as many misdemeanor arrests
(approximately 179,000) than felony arrests (approximately 88,000) and
issued more than triple the number of criminal summonses (271,000),
which include a “notice-to-appear” or “desk appearance” ticket (a form of
noncustodial arrest that secures a promise from the person to appear in
court when required).46 This data parallels the findings of the Criminal
Gatekeepers: The Role of Police in Ending Mass Incarceration

13

Figure 3

Low-level and serious arrests by offense type, 2017
drug abuse
violations

15.
5%

*

rs

uct
ond
c
y
l
rder
diso

al l

ot

he

ss
ne
en
k
n
dru

11.1%
4.9%

27% of arrests
are for drug and
trivial offenses; 5%
are for serious
offenses.

ws
liquor la
vandalism
curfew and loitering
law violations
vagrancy
aggrava
ted assa
ult
robb
ery
rap
e
mu
ma rder a
nsla nd
ugh non
ter neg
lige
nt

Source: Federal Bureau of Investigation, “UCR: 2017 Crime in the United States,” table 29, https://perma.cc/7P43-SYT4.
* Burglary | Larceny-theft | Motor vehicle theft | Arson | Other assaults | Forgery and counterfeiting | Fraud | Embezzlement | Stolen property; buying, receiving, possessing
| Weapons; carrying, possessing, etc. | Prostitution and commercialized vice | Sex offenses (except forcible rape and prostitution) | Gambling | Offenses against the family
and children | Driving under the influence | All other offenses | Suspicion

Justice Agency (CJA)—a pretrial services organization in New York City
that interviews defendants after arrest, just prior to arraignment.47 Of the
272,294 defendants that CJA interviewed in 2016, 73 percent had a most
serious charge of a misdemeanor, and 9 percent a violation or infraction.48
Similarly, in Oklahoma City, municipal charges—a violation of Oklahoma
City’s municipal code, which includes offenses such as public drunkenness,
driving without a license, trespassing, or petit larceny—constituted the
most common arresting charges in 2015 that landed people in jail.49
Many offenses are deemed to be so trivial that the people arrested aren’t
even fingerprinted. In New York State, for example, the list of offenses for
which suspects need not be fingerprinted runs to more than 160 pages,
including common arresting charges such as multiple forms of disorderly
conduct, vagrancy, loitering, criminal trespass, and public intoxication.50 If
a charge is not significant enough to trigger fingerprinting, it is reasonable
to question whether the arrest itself is necessary.

14

Vera Institute of Justice

Lesser forms of enforcement
Full custody arrest is the most dramatic form of police enforcement, but
it is not the only type. In response to nonserious offenses, in most states
officers can employ “least harm” or less intrusive forms of enforcement.
These include issuing a written order requiring a person to appear in a
designated court at a specified time and date (known variously as “tickets,”
“citations,” “summonses,” “desk appearance tickets,” or “cite and release”).51
Police officers can issue such tickets in the field—at (or near) the scene of a
disturbance—or at the stationhouse.52 Citations are commonly associated
with misdemeanors, traffic offenses, and local ordinance violations or
infractions and can come in many forms.53 They are often issued in lieu
of arrest, but can also be issued after an arrest, with officers releasing a
person with instructions to appear in court, pay a fine, and/or adhere to
other outlined conditions. Such citations are common, yet there has been
no routine and systematic collection of national-level data describing the
volume issued annually, for what types of offenses, or their impact.54
However, the data that does exist—some of which dates back 40
years—confirms that many law enforcement agencies use citations in
lieu of custody arrest. As noted above, a 2016 survey conducted by the
International Association of Chiefs of Police found that citation in lieu
of arrest is a widely used law enforcement tool, with nearly 87 percent of
1,300 responding agencies engaged in the practice—more than 80 percent
of those for 10 years or more.55 Drilling down to the local level, data also
suggests that the use of citations may be on the rise in certain places. The
New Orleans Police Department, for example, is just one among many law
enforcement agencies that is relying more on citation and release than
on custodial arrest in certain situations. After the passage of a new city
ordinance mandating the use of a summons rather than arrest for most
municipal offenses, the use of summonses in cases other than domestic
violence and public intoxication increased from 41 percent to more than 70
percent between 2009 through 2011.56 Arrests correspondingly dropped
from 59 percent to 30 percent.57
The use of citations is less invasive of individual liberty than is arrest
and avoids many of the hardships associated with custodial arrest and jail
detention—including damage to reputation or financial burdens such as
the need to raise bail. Nevertheless, an increasing emphasis on citations
Gatekeepers: The Role of Police in Ending Mass Incarceration

15

may inadvertently fuel their use in cases where officers would otherwise
not have made an arrest, potentially widening the net of enforcement and
thus the criminal justice system writ large. No research has yet tracked
and analyzed whether this phenomenon has occurred, either locally or
nationally.
It is also true that, in some cases, citation in lieu of arrest may only
delay rather than replace more punitive action. For example, without
support to address a person’s underlying needs, it may be unrealistic to
expect someone with untreated mental illness and drug addiction to appear
in court at a specific day and time. And, if people fail to show up in court
as required, judges often issue arrest warrants that police officers are
compelled to enforce.58

Disparate racial impact
America’s overreliance on punitive enforcement has not affected everyone
equally. According to the most recent national data, the estimated volume
of arrests of black people across the country rose by 23 percent between
1980 and 2014 and now accounts for an estimated 28 percent of all
arrests—even though black people make up only 13 percent of the U.S.
population. The greatest racial disparities occur for low-level offenses of
all types.
›› Black Americans are nearly three times more likely than whites to

be arrested for disorderly conduct.59 (See Figure 4 on page 17.)
›› In 2014, black people were an estimated 2.39 times more likely to be
arrested for “drug abuse violations” than white people—even though
research suggests that black people and white people use drugs at
similar rates.60
›› Racial disparities are especially acute for low-level marijuana
possession. A 2010 study found that black people were 3.7 times
more likely to be arrested for marijuana possession than white
people, even though both groups use the marijuana at similar rates.61
These enforcement disparities are likely a result of widely used policies
that encourage more aggressive enforcement of minor drug and quality-of-

16

Vera Institute of Justice

life violations through vehicle and pedestrian stops and other practices that
often result in increased police presence in neighborhoods where large
numbers of people of color and economically disadvantaged people reside.62
(For more information about such policies, see “The consequences of
‘broken windows’” on page 30.) Pursuit of such seemingly trivial offenses
can become a pretext for investigatory actions that lead to enforcement—
and heightened risk of future arrests—pulling disproportionate numbers of
black and poor people into the criminal justice system. Such disparities in
enforcement strategies then create disparities in other parts of the criminal
justice system, including incarceration rates.63
At the same time, the inverse also seems to be accurate: white people
are more likely to benefit from lesser forms of enforcement—and perhaps
leniency—than people of color. It is difficult to track or quantify what many
people believe to be true: that police are more likely to ignore or informally
confront white people committing low-level offenses than they are people
of color in the same circumstances. But a 2015 study in New York City

Figure 4

Arrest trends for certain low-level offenses, 1980–2017

1.4M

ss

e
nn
ke
un

800K

Dis

ord

600K

Vandalism

200K

04
20

92
19

80
19

con
d

uct

Curf
ew
& loitering

Vagrancy

0

y

nearly 3 times
more likely

Whites

Blacks

17

400K

erl

The greatest
racial
disparities
occur for
low-level
offenses of
all types.

Disproportionate arrest
rate for disorderly conduct

20

1M

Dr

1.2M

Source: Information on arrests by type from 1980–2017 is derived from the FBI’s Uniform Crime Reporting (UCR) Program, https://perma.cc/8NB2-X6YP.

Gatekeepers: The Role of Police in Ending Mass Incarceration

17

Arrest Trends: Unlocking police data
Researchers, practitioners, and advocates alike need
information on the dynamics of policing to better understand
how the overreliance on arrests drives mass incarceration
and deepens other social inequities. Until recently, finding
out how many people are arrested every year where you live,
whether the number is rising or falling, and how your city,
county, or state ranks nationally would have been difficult for
anyone without the skills of a data analyst and willingness to
devote significant time and effort.
Arrest Trends—a new online tool at arresttrends.vera.org—
makes it easy for anyone to look at arrest volumes and rates,
types of offenses, and the people who are targets of those
arrests broken out by race, sex, age, and location—and to
examine trends over time, in most cases as far back as 1980
and in some cases back to 1964. Users can choose to display
the results of their queries as charts and graphs or as tables
and can also compare what is happening where they live
with trends elsewhere.a There are gaps in this data based on
what local departments report to the FBI—data on arrests by

ethnicity, notably, is not currently provided—but it represents
a broad starting point to begin to understand the far reach
of arrests.
In addition to arrest data, results from victimization surveys
provide another measure of crime rates and also include
statistics on how often victims report these crimes to the
police. Separately, clearance rates indicate how successful
police are at solving reported crimes.

a	

In the future, Arrest Trends will incorporate more census
data—drilling down to the municipal level to augment higherlevel population data that is already available—and incorporate
socioeconomic data, enabling a rich intersectional analysis of how
arrest practices affect different groups of people. There are also
plans to deepen the ability to compare law enforcement agencies—
not only by type, which is already possible—but also by factoring
in Law Enforcement Management Statistics such as agency size and
particular policies and practices.

confirmed that a white person arrested for possession of marijuana is still
more likely to be offered a desk appearance ticket (DAT)—in lieu of full
custodial arrest—than a black or Latino person, even after a 2013 policy
change that directed the police to issue a DAT in all arrests with a top
charge of marijuana possession.64 This trend continued despite a 2014
policy change that directed police to stop making arrests for possession
of small amounts of marijuana in public view.65 (In these instances, police
were instructed to issue a criminal court summons for a noncriminal
violation.66) Conviction rates for black and Latino people, in both cases,
were still nearly twice those of white people.67

The revolving door
National data captures the number of arrests annually, but it does not
reveal how many people are arrested repeatedly over the course of a year. If
the goal is to break this cycle, it is important to know how many people are
being arrested and incarcerated repeatedly and why.
Research published in 2018, for example, shows that a very small
number of people in Camden, New Jersey, make up a disproportionate
share of all arrests in the city. Specifically, just 5 percent of people arrested

18

Vera Institute of Justice

The revolving door of arrests and emergency room visits in Camden, New Jersey
226 “super-util
the
ize
f
rs.
ho
c
.. ”
a
E

had
7+
arrests

(

and
16+
hospital
ED
visits.

= 1 arrest )

(

= 1 hospital visit )

Of the 226 people with high
dual-system use:

Nearly all (96%)
had at least one arrest
for disorderly conduct

A majority (65%)
had at least one
drug-related arrest

A minority (35%)
had been arrested at any
point for a violent crime
(includes weapons-related offenses
and property crimes)

Source: Anne Milgram, Jeffrey Brenner, Dawn Wiest et al., Integrated Health Care and Criminal Justice Data—Viewing the Intersection of Public Safety, Public Health and
Public Policy Through a New Lens: Lessons from Camden, New Jersey (Cambridge, MA: Harvard Kennedy School, 2018), 7-8, https://perma.cc/9HQF-AE62.

in Camden during the five-year study period were involved in one out of
every four arrests.68 Furthermore, a small subset of 226 people, described as
“super-utilizers,” were arrested at least seven times and also made at least
16 visits to hospital emergency departments—suggesting that health issues
may be driving their criminal behavior.69
Not surprisingly, the same dynamic is at play in jail incarceration: a
relatively small number of people are responsible for a large share of jail
admissions. In Chicago, for example, 21 percent of people admitted to
jail between 2007 and 2011 accounted for 50 percent of all admissions.70
In New York City, from 2008 through midyear 2013, approximately 400

Gatekeepers: The Role of Police in Ending Mass Incarceration

19

people were admitted to jail 18 times or more, accounting for more than
10,000 jail admissions and 300,000 days in jail.71 Vera analyses of data from
Oklahoma City and Oklahoma County found similar results. For example,
104 people were admitted to the Oklahoma County jail six or more times
in 2015.72 The research further found that there was a significant likelihood
that these people suffered from mental illness, substance use disorders, or
both.73 In addition to jail and police resources, frequent users of the criminal
justice system consume many other services, at significant cost to the
community. A 2005 study suggests that one cause of frequent admissions is
the exhaustion of community resources for housing and health care among
this high-need population, leaving jails as an inadequate social safety net—
and police on the front lines responding to social service needs.74

What’s driving the overreliance
on punitive enforcement?

H

ow do so many interactions between the public and police—many
of which either do not involve crimes or revolve around minor
low-level offenses—escalate toward policing activities that end
in arrest?75 After all, a punitive approach to enforcement is only one of
many possible responses that the police can take in a case. There are many
tools in an officer’s toolbox, including the decision to cite and release,
issue a warning, make a referral to community-based treatment, redirect
to another responding agency or, indeed, do nothing. (See “An officer’s
toolbox,” on page 21.) There is good reason for having a variety of possible
responses: the same offense can occur under very different circumstances.76
Consider, for example, the difference between someone with a suspended
license driving home alone after meeting friends at a local bar versus
driving an injured child to the emergency room for stitches.

20

Vera Institute of Justice

An officer’s toolbox
Person goes free, no
criminal record.
Person’s
underlying needs
are assessed and
addressed through
services and
programming.

Person goes
free, no criminal
record.

Refer to
communitybased
treatment.

Issue a
warning.

Do
nothing.

Make a
custodial arrest.

Redirect
to another
responding
agency.

Cite and
release.

A more
appropriate first
responder, such as
fire or emergency
medical personnel,
addresses issues.

Person goes free
with directions to
appear in court at
a later date.

Person is detained by police and, if bail is
set and they cannot afford it, they may be
incarcerated in the local jail pending trial.

A number of factors other than the law—singly or in combination—
influence an officer’s decision to invoke the criminal process and enforce
the law in a given circumstance. These include everything from the in-themoment, person-to-person dynamics of a particular interaction; to training
and job performance measures; to individual biases; to enforcement
priorities set by a department or municipality—all of
which can combine to induce officers to believe that there is no choice
but to arrest someone in order to prevent crime, maintain order, and keep
streets safe.
There is a subtle but important distinction between the terms “law
enforcement” and “policing,” which are often used interchangeably.
Enforcement is simply an action taken in response to the letter of the
law. Policing, by comparison, describes a much wider set of actions
beyond enforcement and often includes more pragmatic, problem-solving
approaches that can help address long-term needs and problems.

Gatekeepers: The Role of Police in Ending Mass Incarceration

21

Person-to-person factors
Routine policing usually requires officers to make split-second judgments
when responding to a 911 call or when on patrol, not only relying on
what they have observed, been told by witnesses, or deduced from other
available evidence, but also on the basis of a mix of training and street
smarts gained through on-the-job experience. Consider an officer on
foot patrol who smells marijuana in a jurisdiction where public use is
illegal and sees a young man smoking what appears to be a joint. If she
approaches and confirms her suspicion, she has a few options. She could
respectfully inform him that smoking pot in public is against the law and
ask him to stop. She could make the same statement and then hand him a
summons, which might result in the imposition of a fine—or worse if he
fails to pay or appear in court as may be required. Or she could arrest him.
Some studies suggest the young man’s demeanor and, in particular,
how he responds to the officer will have a strong influence on what action
she decides to take. In one such study, researchers observed numerous
police encounters in which officers had sufficient evidence to make an
arrest, mostly for low-level offenses, and found that they were less likely
to resort to arrest when the person they encountered was respectful,
compliant, and sober.77 Similarly, a study on traffic enforcement in a large
police department showed that leniency by an officer corresponded to the
person’s attitude and demeanor during the stop.78
Of course, police commonly arrest and cite people who are both sober
and essentially compliant. An officer’s temperament and level of experience
come into play. Race and ethnicity in some instances also matter. As noted
above, black people are nearly three times more likely than whites to be
arrested for disorderly conduct in circumstances where the behavior is
comparable.79 Some researchers and other experts have suggested that
analysis of data from police body cameras should be the first step to better
understand officer decision making and minimize the influence of officer
variations and biases.80 In one such study, Stanford University researchers
analyzed Oakland Police Department body camera data and found that
officers consistently spoke less respectfully to black community members
as compared to their white counterparts.81 This finding was consistent
despite the race of the officer and the type, severity, location, and outcome
of the encounter.82
22

Vera Institute of Justice

Messages from above
Law enforcement agencies operate as quasi-military organizations.83
Officers wear uniforms, use military ranks, carry weapons and are
authorized to use force, engage in military-style tactics, such as patrolling,
and follow a body of rigid rules.84 Most important, they function according
to a strict authoritarian organizational style with a clear hierarchy and
chain of command.85 At the agency level, official policies and informal
messages from the top of this hierarchy thus have a significant effect on
how frequently officers resort to enforcement in the course of policing.

Many police departments—whether
deliberately or not—incentivize the use
of enforcement, measuring officers’
performance in large part by the
number of stops, arrests, summonses,
and tickets they generate—all of which
are much easier to quantify than
enforcement alternatives.

Many police departments—whether deliberately or not—incentivize
the use of enforcement, measuring officers’ performance in large part
by the number of stops, arrests, summonses, and tickets they generate—
all of which are much easier to quantify than enforcement alternatives
(especially when an officer decides to do nothing).86 And this may be driven
by an executive policy decision in a department or precinct to prioritize
certain arrest-based crime enforcement and response strategies, such as
those that focus on the enforcement of low-level misdemeanor offenses
Gatekeepers: The Role of Police in Ending Mass Incarceration

23

and civil infractions (including such things as disorderly conduct, public
drinking, public marijuana smoking and other minor drug offenses, noise
pollution, public indecency, verbal harassment, aggressive panhandling,
vagrancy, and obstruction). (See “The breaking of ‘broken windows’” on
page 30.) A supervisor in such a department is likely to praise one officer
for making several arrests during the week, particularly if it is in service of
a department’s adopted enforcement priorities, while calling someone else
who has made none a “zero.”87 At least anecdotally, officers who log more
arrests have been reported to receive preferential treatment in schedules,
assignments, and promotions.88
One of the most widely studied examples of how departmental
policies can greatly incentivize—or de-incentivize—police behavior comes
from New York City. In response to surging crime rates in the 1990s —
particularly violent crime—the New York City Police Department (NYPD)
developed a management framework called Computer Statistics (CompStat)
that the city still uses.89 It allows police executives to track geographic
trends in crime, requiring the commanders responsible for those areas to
account for fluctuations through regular CompStat meetings.90 As part
of this data-driven approach, middle managers are held responsible and
accountable for reducing crime in their regions.91 A majority of large—and
many small- and medium-sized—police departments nationwide now use
an adapted version of the CompStat model.92
Through this performance-management approach, NYPD commanders
quickly established a systemwide expectation: they measured officers’
success in combating crime in part through high levels of enforcement
activities including stop, question, and frisks; citations; summonses; and
arrests. Immediately following CompStat’s implementation, enforcement
activities grew dramatically, especially in poor communities and black
communities.93 The NYPD began tracking the use of stops in 2003 and, in
that year, almost 161,000 stops were reported.94 In the years that followed,
police executives encouraged the use of stops as a crime-fighting tactic
at CompStat meetings and continued tracking this metric.95 The number
grew precipitously, up to a peak of almost 686,000 stops in 2011, a 326
percent increase.96 Felony and misdemeanor arrests, as well as summonses,
increased on a similar trajectory—though not as pronounced—even as
crime was decreasing significantly.97 After several lawsuits and resulting
reforms designed to address policing culture, including updated policies
24

Vera Institute of Justice

and training, reported stops plunged to about 46,000 in 2014, a 93 percent
decrease from 2011.98
In contrast to the incentives that the original CompStat method
spawned, a police department may choose to discourage the use of
enforcement in response to low-level offenses and support officers in
making different choices where appropriate. By prioritizing non-arrest
options in some cases, such as issuing a warning to that young man caught
smoking marijuana or not approaching him in the first place, leadership
can communicate that sometimes doing nothing is the best response.
Even in departments that give clear agency-level guidance to rein
in enforcement responses to specific offenses, officers may still default
to it. They do so in part because of longstanding incentive structures
and the fact that the vast majority of police training focuses on tactical
enforcement operations such as pursuing suspects, operating firearms
and other weapons, executing arrests, and understanding the concept of
probable cause (the legal prerequisite for police enforcement activities).99
Most police training, in fact, follows the military model and emphasizes
four areas: driving, firearms, apprehension, and arrest.100
Alternatives to enforcement and other problem-solving tactics are not
yet prominent features of formal police training. Where such training does
exist, communication about priorities from agency executives will heavily
influence its quality and content, as well as how seriously officers approach
it.101 Some departments have made the use of alternatives to enforcement a
priority in training.102 (See page 45 for a discussion of one such department,
in Durham, North Carolina.)

Lack of alternatives
Reliance on enforcement may simply be the result of the absence of policy
or lack of appropriate training in nonenforcement options—from donothing policies to use of summons or tickets— but it is also likely to flow
from a scarcity of or lack of working relationship with vital community
referral services, such as those that focus on mental illness, substance use,
employment, or homelessness.
Take, for example, an officer who encounters a person in a park who
is clearly inebriated, possibly mentally ill, and scaring local residents.

Gatekeepers: The Role of Police in Ending Mass Incarceration

25

Complicating the situation, the person is homeless. Without a mental
health clinic, sobering center, or fully equipped homeless shelter ready
to take in the person, an officer might default to arrest, not having other
available options or thinking it is the easiest way to ensure both the person
and community are safe. Even where these services exist, officers might
still default to arrest in this situation because they are either unaware of
such services or resources, or because the incentive structure or ideological
climate of their agency may not yet support channeling the exercise of
police discretion toward alternative action.

A growing number of police
departments are beginning to explore
alternatives to enforcement, focusing on
creating or fine-tuning community crime
prevention strategies.

Yet the fact is that police are frequently the first and sometimes only
responders in situations like these, where the tools and stock responses
at their disposal are a mismatch with what is needed and may make the
situation worse.103 This issue is especially acute in nonmetropolitan and
rural communities, where appropriate community-based interventions
or treatment modalities are less available than in urban areas.104 Even if
partnerships or programs exist in these areas, distance and scarcity—as
well as stigmatization—can make their practical application impossible.105
However, a growing number of police departments are beginning to
explore alternatives to enforcement. These jurisdictions are focused on
creating or fine-tuning community crime prevention strategies—often
partnering with local health departments and behavioral health or human
service providers to develop alternatives to making an arrest or booking a
person into jail.106 Such partnerships increase police capacity to respond
26

Vera Institute of Justice

more constructively to people who regularly come into contact with
police, such as those who have mental health or substance abuse issues or
who are homeless. The community-based services that these departments
are tapping into typically provide everything from 24-hour respite care;
crisis beds for short-term stays; case management; detox services; health
education; and referrals to ongoing treatment, housing, and other social
services.107 These policies and programs hold the promise of reducing
the volume of unnecessary arrests and jail bookings and, for the people
involved, avoid some of the negative consequences that follow arrest,
prosecution, and incarceration—including deteriorating mental or physical
health, imposition of court fines and fees, and disruption of employment or
education. Nevertheless, these programs remain small in scope and reach.
(See “Working at the margins: Alternatives-to-arrest programs” on page 32.)

Perverse incentives
Perverse incentive structures sometimes exist in policing, turning arrests
into a tool to achieve some other purpose. For example, officers may use
their power to arrest to increase the pool of fingerprints and DNA samples
available to solve past or future crimes.108 Officers may also default to arrest
to ensure the alleged suspect’s appearance in court via pretrial detention.109
Or they might use arrest in response to financial incentives.110
Indeed, money is often the root of unnecessary arrests.111 In the NYPD,
for example, an arrest triggers overtime pay—a phenomenon known
as “collars for dollars”—because processing an arrest can be very time
consuming.112 Officers may have a financial incentive to wait until the
ends of their shifts to make arrests so they are eligible for overtime pay.
It is unclear how common this practice is, but there have been several
high-profile cases where officers have arrested innocent people in order to
receive overtime.113 Money at the agency level may also drive arrests. One
study revealed the negative impact of the Edward Byrne Memorial State
and Local Law Enforcement Assistance Program, which makes federal
grants that police agencies can use to fund enforcement of drug-related
and violent crimes and is designed to incentivize agencies to develop and
expand their drug interdiction activities.114 When researchers looked at
cities that received Byrne grants between 1987 and 2004, they found a

Gatekeepers: The Role of Police in Ending Mass Incarceration

27

corresponding increase in drug-trafficking arrests that disproportionately
affected black community members.115
Civil asset forfeiture—a legal tool that gives law enforcement agencies
the authority to seize cash, property, and other possessions believed to
be associated with alleged illegal activity—can also serve as a perverse
incentive motivating enforcement.116 Through this process, police can
seize property, including cars, houses, jewelry, and cash, even when the
owner has not been charged with a crime.117 According to a 2015 study,
seven states and Washington, DC, prohibit law enforcement agencies from
accessing forfeiture proceeds for department use; while 42 states and the
federal government allow law enforcement to access at least 45 percent—
and in many cases, including the federal government’s, up to 100 percent—
of the value of forfeited property.118
Asset forfeiture has been a feature of law enforcement for centuries,
but its use expanded significantly during the 1980s with the advent of the
War on Drugs and the 1984 Comprehensive Crime Control Act.119 Civil
asset forfeiture was originally proposed to halt large criminal enterprises
by seizing their resources, but it has been used to seize everything from
vehicles in suspected drunk driving cases to homes in which illegal drug
sales have occurred—whether or not the homeowner was aware of these
sales.120 This practice has disproportionately affected poor communities and
communities of color and focuses too often on people suspected of lowlevel offenses—such as a $60 marijuana sale or a minor traffic infraction.121
Police departments and prosecutors (including at the federal level)
who employ this practice typically use seized property as a direct revenue
source to pay for equipment, vehicles, building improvements, and even
officer salaries and overtime pay.122 Annual reports submitted to the
Department of Justice show that civil asset forfeiture funds are also used to
purchase surplus military-grade equipment and vehicles, such as automatic
weapons, night vision gear, armored cars, and Humvees.123 While some law
enforcement executives argue that this spending is necessary, it contributes
to an increasingly militarized local law enforcement presence in many
communities around the country. As a 2014 study by the ACLU notes, “[t]he
change in equipment is too often paralleled by a corresponding change in
attitude whereby police conceive of themselves as ‘at war’ with communities
rather than as public servants concerned with keeping their communities
safe.”124 In contrast, seized assets are seldom spent on nonenforcement
28

Vera Institute of Justice

activities such as community engagement. For example, data from 43,000
state and local police agency annual reports on forfeiture (obtained through
a series of Freedom of Information Act requests by the Washington Post)
revealed that less than 1 percent of money the agencies gained through
civil asset forfeiture since 2008 had gone to community outreach.125
Criminal justice fines and fees are another example of how money
can influence policing activities in a potentially damaging way. “Fines” are
financial obligations imposed on someone as punishment or restitution;
“fees”—sometimes referred to as “user fees”—are financial obligations
imposed on people ostensibly as payment for their passage through the
criminal justice system and whose explicit purpose is to raise revenue to
fund courts or other governmental services and functions.126
Most municipalities generate just a small amount of their total revenue
from fines and fees. But some have come to rely on them to support core
government operations, with a direct and distorting impact on police
officers’ work.127 The example of Ferguson, Missouri, stands out. After
the August 2014 police shooting of Michael Brown, an investigation by
the U.S. Department of Justice found that the city was funding much of
its government operations through fining poor black residents for lowlevel offenses, along with other patterns of racially biased applications of
enforcement such as stops, arrests, and use of force.128 A 2016 review of the
characteristics of cities that rely most heavily on fines for revenue found
that they have significant black populations—five times greater than the
national median.129
Even a small fine can place a significant burden on poor and lowincome people.130 When money is tight, people face tough choices. They
weigh the importance of paying a fine versus, for example, paying the
utility bill, buying school supplies for their children, or fixing the broken
taillight that generated the ticket in the first place. When the amount
owed is compounded by interest on unpaid fines, the choices become
even harder, which is why the people who police target for fines and fees
sometimes end up in jail.131 This trickle of revenue may not even offset the
direct cost of punishing those who cannot pay: a study conducted by Vera
found that the City of New Orleans spent approximately $2 million more
to jail people for failure to pay bail, fines, and fees than the total amount
collected from these revenue sources.132 It is a cautionary tale for any
jurisdiction that seeks to reduce arrests by increasing the use of fines.
Gatekeepers: The Role of Police in Ending Mass Incarceration

29

The consequences of “broken windows”
Order-maintenance policing—sometimes referred to as
“broken windows” policing—is a crime and disorder control
strategy based on the idea that police can stop or mitigate
the growth of more serious crime patterns and problems from
taking hold in any one neighborhood by systematically and
aggressively using their enforcement authority against minor
crimes and other public order incivilities—from graffiti to
loitering to trespassing to public intoxication.a
Research has shown that policing strategies focused
on disorder are associated with a small but statistically
significant reduction in crime overall, with community and
problem-solving interventions showing more beneficial
effects than aggressive strategies targeting minor individual
disorderly behavior that often has little if any impact on
major crime.b These aggressive strategies—relying primarily
on arrests—have, however, been found to have direct
negative effects on community-police relations, making
community members feel unsafe and reducing trust in and
cooperation with the police.c
Broken windows evolved in an era of rising crime as a way
for police departments to attempt to increase order in
challenged communities, with the goal of reducing crime.
The strategy’s logic may seem intuitively attractive. But
in practice, its widespread application has had many
negative unintended consequences for the neighborhoods
most affected—usually places with concentrated poverty
and often communities of color. This has been particularly
apparent when implemented in conjunction with other
proactive policing activities such as hot-spots policing—a
strategy that focuses resources on the locations where crime
or 911 calls for police service are highly concentrated—as well
as extensive use of pedestrian stops (the use of temporary
detention and interrogation in the field) to detect and disrupt
potential criminal activity.d
Indeed, because of increases in the probability of police
contact for people in targeted communities—again, often
disproportionately those with large concentrations of
poor people and people of color—such impacts include
heightened risk of further downstream criminal justice
involvement, along with the full breadth of collateral
consequences that flow from such contact, including barriers
to housing, employment and credit.e Beyond the personal
toll, the large-scale cycle of enforcement, arrest, removal,
and return of people damages familial and community
relationships and disrupts neighborhood life, including the
number of social networks that can exert informal social
control and encourage social cohesion among community
members.f

30

Vera Institute of Justice

At the extreme, this kind of overpolicing can turn fatal. One
example is the violent confrontation that led to the death
of Eric Garner in 2014—the pretext for which was Garner’s
participation in the tobacco gray market through the sale
of single cigarettes on a Staten Island street corner.g Deaths
at the hands of police have also occurred during traffic
stops, such as the 2015 death of Walter Scott, who was shot
in the back as he fled the police after being stopped for a
nonfunctioning brake light; and the 2016 death of Philando
Castile, who police shot as Castile was reaching for his
license and registration during a stop purportedly for a
broken taillight.h The aggressive use of such intrusive policing
tactics can have wider adverse impacts for communities
and police: from an increase in the number of unproductive
police contacts that can sow the seeds of mistrust in the
communities police are meant to serve, to the noxious effect
of encouraging racially problematic behaviors to grow
and proliferate under the guise of police discretion and/or
officers’ perceived fears for their personal safety.i

a	

George L. Kelling and James Q. Wilson, “Broken Windows: The
Police and Neighborhood Safety,” Atlantic, March 1982, 29-38,
https://perma.cc/3MJ3-ZY6J.
b	

For a synthesis of research on policing strategies focused on
disorder, see Anthony A. Braga, Brandon C. Welsh, and Cory Schnell,
“Can Policing Disorder Reduce Crime? A Systematic Review and
Meta-analysis,” Journal of Research in Crime and Delinquency 52,
no. 4 (2015), 567-588, https://perma.cc/T39B-X7TA. The researchers
identified 30 studies that met the Campbell Collaboration criteria.
In combining the results of these studies, the researchers found
that general policing disorder strategies are associated with a
small reduction in crime. However, when the researchers compared
“community and problem-solving interventions” to “aggressive order
maintenance strategies” they found differing results. “Community
and problem-solving interventions,” defined as community
problem-solving programs that attempted to engage residents,
local merchants, and others in the identification of local crime and
disorder problems and the development and implementation of
appropriate responses, produced a large and significant effect in
reducing crime. On the other hand, “aggressive order maintenance
interventions” that focus on individual disorderly behaviors, primarily
relying on arrests, ordinance violation summonses, and other law
enforcement strategies, and not typically involving the community,
resulted in much smaller crime reductions.
c	

Jacinta M. Gau and Rod K. Brunson, “Procedural Justice and Order
Maintenance Policing: A Study of Inner-City Young Men’s Perceptions
of Police Legitimacy,” Justice Quarterly 27, no. 2 (2010), 255-279.
d	

Bernard E. Harcourt, Illusion of Order: The False Promise of Broken

Windows Policing (Cambridge, Massachusetts: Harvard University
Press, 2004).
e	

Elizabeth Hinton, LeShae Henderson, and Cindy Reed, An Unjust
Burden: The Disparate Treatment of Black Americans in the Criminal
Justice System (New York: Vera Institute of Justice, 2018), 10 & n.103,
https://perma.cc/UQ8D-5KUB.
f	

Research that focuses on the community-level impact of
incarceration has demonstrated that these harms extend far beyond
direct collateral consequences for formerly incarcerated people.
Incarceration strains ties not only between incarcerated people and
their community, but also between their family and the community—
and diminishes social capital for themselves, their children, and
their extended family. When an incarcerated parent is a mother,
these harms can be compounded in significant ways as childcare
needs go unmet and children enter the foster system. Communities
disorganized by financial and personal strain in this way are unable
to form and sustain informal social control mechanisms that can
prevent antisocial behavior by community members. Normative
behaviors in communities affected by incarceration tend to be less
social, less cooperative, and less likely to be trusting of law or law
enforcement. And the disenfranchisement and exclusion of formerly
incarcerated people in matters of civic participation or labor
opportunities make them less likely to be able to reintegrate or regain
social capital in the community. Dorothy Roberts, “The Social and
Moral Cost of Mass Incarceration in African American Communities,”

Stanford Law Review 56, no. 5 (2004), 1271-1305, 1281-96, https://
perma.cc/6YM5-XV4B. Besides undermining the development of
informal social controls and trust in formal enforcement, community
exposure to incarceration also undermines trust in informal social
controls as a positive influence. Dina Rose and Todd Clear, “Who
Doesn’t Know Someone in Jail? The Impact of Exposure to Prison on
Attitudes toward Formal and Informal Controls,” Prison Journal 84,
no. 2 (2004), 228-247. See also Todd Clear, Imprisoning Communities:
How Mass Incarceration Makes Disadvantaged Neighborhoods Worse
(New York: Oxford Press, 2007).
g	

Joseph Goldstein and Nate Schweber, “Man’s Death after
Chokehold Raises Old Issue for the Police,” New York Times, July 18,
2014, https://perma.cc/3JH9-G4HC.
h	

For the death of Walter Scott, see Matthew Vann and Erik Ortiz,
“Walter Scott Shooting: Michael Slager, Ex-officer, Sentenced to 20
Years in Prison,” NBC, December 7, 2017, https://perma.cc/Q55Q5K37. For the death of Philando Castile, see Bill Kirkos and Holly Yan,
“Philando Castile Death: Fate of Police Officer Now in Jury’s Hands,”
CNN, June 12, 2017, https://perma.cc/U2FQ-6F5L.
i	

For negative community impacts, see Gau and Branson,
“Perceptions of Police Legitimacy,” 2010. For problematic patterns
of enforcement under the guise of discretion, see U.S. Department
of Justice (DOJ), Civil Rights Division, Investigation of the Baltimore
City Police Department (Washington, DC: DOJ, 2016), 26.

Gatekeepers: The Role of Police in Ending Mass Incarceration

31

Working at the margins: Alternatives-to–arrest programs*
Despite an array of factors incentivizing reliance on
punitive enforcement, police have long understood that “full
enforcement”—the expectation that they enforce the letter of
the law in every situation—is not practicably possible; nor is
it in many circumstances an effective response to widespread
social problems that may underlie lawbreaking—from
poverty, to homelessness, to substance use, to mental illness.1
More and more, society is coming to understand that there
are some problems that, in the frequently used phrase, “We
cannot arrest our way out of.”2
Some police departments, individual officers, and
communities are recognizing that they must moderate the
use of punitive enforcement as a matter of policy and
practicality as well as public safety. Thus, police routinely use
a process called “diversion,” a variety of informal and formal
practices that filter certain cases out of traditional criminal
justice processes.3 For police, this can mean employing
informal responses such as issuing warnings or requesting
people to discontinue certain behaviors.4 Sometimes police
also refer people to third parties—friends, family members,
community-based services, or other resources—when they
perceive that cases may be better handled outside the
criminal justice system.5
Some law enforcement agencies have collaborated with
community organizations and other criminal justice
stakeholders to develop formal diversion programs and
practices that channel certain categories of people—
usually those considered low risk or in need of specialized
treatment—away from the criminal justice system.6 A number
of these collaborations have been in existence for decades.7
They represent a notable example of law enforcement
agencies making deliberate decisions to elevate and
validate nonpunitive approaches—rather than depending
on ad hoc decisions made by street-level officers or the
informal policing culture of a particular department—as an
organizational crime response and prevention strategy that
prioritizes the values of restraint and minimization of harm.
Formal diversion programs remain limited in both scale
and scope. Yet they have demonstrated that alternatives to
enforcement not only can produce public safety gains, they
can also align with the broader goals and culture of law
enforcement agencies. But the challenges these programs
face in terms of acceptance, impact, and effectiveness also
demonstrate the obstacles that stand in the way of bringing
them to scale.

Prevalence, scale, and scope of police
diversion

32

In a large survey of police agencies nationwide conducted
by the Center for Court Innovation (CCI) and published in
2016, roughly a third of police agencies surveyed reported
Vera Institute of Justice

engaging in diversionary practices of some kind, but only one
in five were operating what could be described as a formal
and intentional program.8 Not surprisingly, the survey found
that formal programs are more common in larger police
departments.9 Thirty-nine percent of sampled agencies with
500 or more officers reported operating a formal initiative of
some type, but that rate dropped to 25 percent for agencies
with fewer than 50 officers and 12 percent for agencies with
10 or fewer officers.10 The impact of these programs also
ranged widely, from diverting a few thousand people to none
at all during the period studied.11

Percent of police agencies operating a
formal diversion initiatiative, CCI survey
39% of sampled
agencies with

500+ officers

25% of sampled
agencies with

< 50 officers

10% of sampled
agencies with

< 10 officers

The CCI survey, along with research compiled by the
International Association of Chiefs of Police in partnership
with other organizations, also suggests that when police
control diversion they tend to focus narrowly on specific
groups of people: typically juveniles, people with mental
illness, sex workers, and people who use drugs, all of whom
are usually charged with relatively minor, nonviolent offenses
and/or are being charged with a crime for the first time.12
There is growing interest in diverting people who come into
contact with the criminal justice system by repeatedly
committing the same or similar low-level, nonviolent offenses
because of underlying needs and circumstances, but rules
governing eligibility for such programs still shut out too many
people—often excluding those who have certain offenses in
their criminal histories.13

There is also evidence—most robust in the realm of juvenile
diversion—that racism and other biases unfairly influence
who reaps the benefits of these alternatives, as they do in
other realms of police enforcement.14 In 2016, researchers
examined how demographic factors affect juvenile justice
system outcomes and the likelihood of diversion by analyzing
a year’s worth of data from eight different police agencies
and prosecutors’ offices in a large Midwestern metropolitan
county.15 The results suggest that many more juveniles were
eligible for diversion than police and prosecutors diverted
and, moreover, that they were likely to overlook young
people of color.16 Specifically, in five of the eight agencies
(63 percent), police enrolled eligible nonwhite juveniles in
alternative programs less often than their white peers, even
when all other factors were the same or controlled for through
statistical means.17 

Police enrolled eligible nonwhite juveniles
in alternative programs less often than
their white peers

ci
en
ag
lic
e

po

white juvenile
white juvenile
white juvenile
white juvenile
white juvenile
white juven

es

ENROLLED

in alternative programs

Formal alternative-to-arrest programs
Most formal alternative-to-arrest programs keep people under
the umbrella of the criminal justice system. Diversion can
occur before someone is arrested and formally charged with
a crime (known as pre-arrest diversion); or it can happen
after normal arrest procedures, but prior to arraignment
and/or a potential stay in local custody (known as pre- or
post-booking diversion).18 Regardless of when diversion
occurs, legal coercion is involved. The criminal case against
the person is not dismissed, but held in abeyance and if
participants do not successfully fulfill obligations imposed by
a diversion program—which can be anything from remaining
arrest-free to completing community service to participating
in a drug education program—the criminal case is reinstated,
and the person will face prosecution and trial.19

Law Enforcement Assisted Diversion (LEAD) is perhaps the
most widely replicated police diversion program, although
the point at which diversion occurs varies by city.20 In some
jurisdictions, police divert people instead of arresting them,
while in others, diversion happens after an arrest but before
the person is booked into jail.21 But everywhere the goal
is the same: to connect people, often through referrals to
community-based providers, with services and supports that
can help them change their lives for the better.22 It might be
medical care, mental health or drug treatment, safe housing,
counseling, or something else.23 In this way, LEAD addresses
the root causes of crime.24
Pre-arrest diversion. The Pre-Arrest Diversion Initiative (PAD)
in Fulton County (Atlanta), Georgia—similar to LEAD, but
limited, as its name indicates, to pre-arrest action—takes a
harm reduction approach, which addresses people’s needs
while trying to minimize the adverse impacts of criminal
justice involvement. The program also takes a housingfirst approach and is both trauma-informed and culturally
competent.25 PAD was spurred by grassroots organizers and
the very people it seeks to help: mainly poor people, many of
them sex workers, whose health care and other needs have
gone unmet while they have had multiple negative encounters
with the criminal justice system.26 Today, PAD partners
with the Atlanta Police Department to connect people who
otherwise would be arrested with the care they need.27
Post-booking diversion. A prime example of a post-booking
diversion program is Richmond County, New York’s Heroin
Overdose Prevention and Education (HOPE) program.28
Officers have the option to process people with little to no
criminal history who are arrested on low-level drug possession
charges and qualify for a desk appearance ticket (DAT) for
a seven-day return date, rather than the 30-day return date
typical of DATs.29 At the same time, officers contact a peer
coach who meets participants on their release, provides
them with a Naloxone kit and training on how to use it, and
encourages them to visit a local resource and recovery
center that can provide assessment and treatment option
recommendations.30 If they do so before their return date,
their cases are adjourned for 30 days; if they engage with
treatment within the adjournment period, they will not be
prosecuted.31
Outcomes of formal diversion programs. Because formal
diversion is still relatively uncommon and small in scale, the
research on its effects is limited. There is, however, evidence
that diversion reduces recidivism. For example, two metaanalyses examining the effects of youth diversion programs
showed they are more effective at reducing recidivism than
traditional criminal justice interventions such as arrest,
probation, and incarceration.32 Another systematic literature
review, looking at the effects of diversion for people with
mental illnesses, also found reductions in recidivism as
measured by the number of subsequent arrests—although
the authors note the need for more research.33

Gatekeepers: The Role of Police in Ending Mass Incarceration

33

Despite a general lack of literature, there are evaluations of
programs from single jurisdictions with encouraging findings.
Leon County, Florida, for example, operates a PAD program
that allows responding officers to cite instead of arrest
people suspected of committing select misdemeanors—
including trespassing, disorderly conduct, petty theft,
possession of small amounts of marijuana, non-domestic
simple battery/simple assault, and others—for whom this
would be the first offense.34 Those in PAD have to complete
a program tailored to the nature of their alleged offenses—
for example, drug education, behavior management,
counseling sessions, and/or community service—and, if they
successfully complete the program, the case is closed.35 A
study of 854 PAD participants published in 2017 shows that
91 percent successfully completed the program.36 There was
no comparison group, but 87 percent of participants avoided
re-arrest through the end of the study, a period ranging
from five to 44 months post-program completion.37 The study
showed that PAD was especially effective for women and
for men and women without substance use or behavioral
health problems, as they were the least likely to drop out or
recidivate.38
Several evaluations show that in jurisdictions that have
adopted LEAD programs, the programs are associated not
only with lower recidivism but also with greater stability

One study of LEAD outcomes found that
participants were...

Twice as likely to
find shelter

89% more likely to
obtain permanent
housing

46% more likely to be

employed or training for
employment

33% more likely

to connect with
supplemental income
and benefits

34

Vera Institute of Justice

and quality of life for those who participate. In one study,
researchers tracked LEAD participants’ housing, employment,
income, and benefits before and after program referral
and found significant improvements in all of these areas.39
Participants were twice as likely to find shelter; 89 percent
more likely to obtain permanent housing; 46 percent more
likely to be employed or training for employment; and 33
percent more likely to connect with income and benefits such
as veteran pensions, unemployment, supplemental security
income, and Temporary Assistance for Needy Families.40
The value of diversion programs is not limited to participants:
a cost-benefit study of Seattle’s LEAD program in 2015 also
indicates the program’s economic value to the jurisdiction.
Participation in LEAD was associated with a net savings
of $2,100 per person, compared with a cost increase of
$5,961 for a matched comparison group.41 The difference
is a function of the reduction of costs associated with jail
bookings, days spent in jail, and prison incarceration.42 Thus,
while the programmatic costs of diversion are relatively high,
long-term participation in LEAD is associated with statistically
significant reductions in criminal justice system use and the
higher costs associated with it.
However, other cost-benefit studies in the area of pre-arrest
diversion have produced widely differing results: several
meta-analyses reveal that there is a widely varying but
positive chance of producing benefits greater than the overall
costs.43 Every evaluation of pre-arrest diversion programs
that the researchers studied in these comprehensive metaanalyses found either slight or significant improvements in
outcomes, suggesting that even small-scale programs can
generate positive results and that greater investments and
attention can expand improved outcomes.44

Bypassing the criminal justice system entirely
Some law enforcement agencies have developed problemsolving programs aimed at keeping people out of the criminal
justice system altogether. For instance, police officers
have increasingly become the primary first responders to
situations involving people living with a mental or behavioral
health disorder—and, in many cases, their de facto health
providers.45 Rather than arrest people who are struggling,
these departments have focused on making police contact a
pathway to care.46
The Crisis Intervention Team (CIT) model is a collaborative
effort between law enforcement and mental health
professionals that is designed to allow officers to become
conduits to treatment in situations where enforcement
is neither necessary nor the best option.47 The heart of
the model is a 40-hour training course for officers in the
dynamics of mental illness and how to respond effectively
to people in crisis.48 Most agencies that have adopted
the model aim to train roughly 25 percent of uniformed
staff.49 Several Texas counties have taken a more expansive

approach: Bexar County (San Antonio), had 95 percent of its
officers CIT-trained by 2016.50 The Houston Police Department
expanded its CIT initiative department-wide and had more
than 3,100 CIT-certified officers by 2018.51 Houston police also
added an escalated response option for especially difficult
CIT calls: Crisis Intervention Response Teams (CIRTs) in which
a licensed masters-level clinician is paired with a CIT-certified
officer.52 As of 2016, there were 12 CIRTs serving Houston’s Harris
County, with plans to add more, making it one of the largest
co-responder programs in the nation.53
While full-scale evaluations of CIT programs are limited,
research has shown that they improve officers’ knowledge,
attitudes, self-efficacy, and inclination to use de-escalation
strategies.54 Evidence also suggests that CIT programs increase
referrals and linkages to mental health services.55
CITs are not the only such co-response model being
investigated by police departments. Disability Response
Teams (DRTs) bring together law enforcement, victim service
providers, legal professionals, and advocates to uncover and
problem-solve community-specific challenges around the
intersection of criminal justice and disability; co-responder
teams embed mental health clinicians with responding police
officers; and case management teams provide coordinated
follow-up from police and mental health professionals

for people who have been identified as frequent users of
emergency response systems.56
Meanwhile, the current epidemic of drug overdose deaths—
particularly those related to opioids—has prompted some
law enforcement agencies to take other proactive measures,
such as joining wider community and government overdose
prevention efforts by creating non-arrest, stigma-free pathways
to treatment and recovery. For example, in 2015, the thenpolice chief of Gloucester, Massachusetts, established its Angel
Program, which encourages drug users to walk into the police
station with their drugs and/or paraphernalia and ask for
treatment.57 Rather than processing these people for arrest,
police officers contact treatment facilities to find a space for a
presenting person and contact a local volunteer recovery and
addiction expert who can provide emotional support.58 With
the Angel Program now incorporated into the Police Assisted
Addiction and Recovery Initiative (PAARI)—a movement of police
departments across 35 states and the federal government—
more than 20,000 people had received treatment through
nearly 500 participating law enforcement agencies as of
March 2019.59

* Box notes at end of report.

Gatekeepers: The Role of Police in Ending Mass Incarceration

35

Shifting the paradigm

T

here is a real appetite for reducing America’s reliance on incarceration.
Yet there have been only scattered efforts to approach the goal by
starting at the point of first contact for the vast majority of people
who end up in jail or prison—through changing policing practices and
priorities—and scant study of these initiatives. Expanding alternatives
to police enforcement from a marginal practice to a defining aspect of
American policing requires a clear path forward. Such a monumental shift

A monumental shift in policing culture
and practice will require a nationally
concerted effort that engages local
elected officials and police leadership,
collaboration with and investment in
community-based service providers, and
the support of the general public.

in policing culture and practice will require a nationally concerted effort
that engages local elected officials and police leadership, collaboration with
and investment in community-based service providers, and the support
of the general public. Patience, as police leaders update policies, innovate
training, and work to alter an entrenched organizational culture that
rewards and encourages arrest, is necessary. America will not reduce its
reliance on enforcement overnight.
36

Vera Institute of Justice

Specifically, for police to figure out how they can move away from the
mass enforcement status quo, they need a comprehensive set of strategies
on how to better channel their wide reservoir of discretion when deciding
what laws to enforce, against whom, and in what circumstances. To protect
against risks of arbitrary, haphazard, or discriminatory enforcement that
have led to so many unnecessary arrests—whether through practices such
as overintensive and misplaced police surveillance or unnecessary use
of force—these strategies also need to make police exercise of discretion
much more focused, deliberate, reflective, and transparent. In many
circumstances, this may mean giving greater discretion to police not to use
their authority to arrest, and providing the appropriate infrastructure and
set of incentives to support less intrusive and nonenforcement options.133
To help spur discussion and inspire change, Vera offers a number of
recommendations about how to move along this path based on lessons
drawn from the small body of research on alternatives to date and an
assessment of what has driven the current overreliance on enforcement
(including the messages received from superiors, departmental priorities or
directives, institutional incentives, and the current scarcity of alternatives).
Based on what is known, it will then be possible to build from existing
informal and formal alternative-to-arrest programs, practices, and policies
and lay the foundation for a different paradigm in American policing.
ALTERNATIVES
Recommendation 1: Identify,
promote, and invest in
alternatives to enforcement that
don’t involve the criminal justice system

The earliest stage at which alternatives to punitive enforcement can come
into play is prior to any call for service or observation of criminal activity.
And those alternatives depend in large part on police partnering with—
and localities investing in and enhancing—community-based services and
supports that focus on dealing with the root causes of crime and social
disorder, such as mental illness, homelessness, substance use, and poverty
outside the criminal process. There are many circumstances in which an
officer can provide potentially helpful information and walk away—or not
Gatekeepers: The Role of Police in Ending Mass Incarceration

37

engage at all—but there are other situations in which doing nothing is
neither safe nor responsible. It is precisely in these difficult circumstances
that officers need a validated alternative to arrest as well as an established
protocol for accessing these services.
Thus, a necessary step in creating robust alternatives to enforcement
is expanding the service-provision capacity of health care and ancillary
systems in the community. This will require local political will beyond
policing, though the support of police can help catalyze such efforts. The
broader success of such services rests on their capacity to evaluate people
and respond accordingly without delay. Such rapid response requires that
police agencies and officers develop sustained connections with an array of
local human service providers and other community-based organizations
and resources to connect people with whom they often come into
contact to such things as behavioral health care, housing, and other social
services as an alternative to arrest and detention. Doing so can deepen the
engagement of police with the communities they serve.

Community drop-off centers
One option is for police departments to expand their use of communitybased drop-off centers, which can provide shelter, clinical, and other social
services for people in crisis—particularly those with a mental illness
and/or substance abuse problem, who are more likely to be arrested and
are overrepresented in jail and prison populations.134 Police in multiple
localities—for example, Johnson County, Kansas; Miami-Dade County,
Florida; Salt Lake County, Utah; and the city of San Antonio, Texas—have
increasingly used such centers to connect people with appropriate clinical
assessments, crisis services, and referrals to other resources and services.135
In Houston, Texas, the Houston Recovery Center, which is open 24/7
year-round, provides short-term crisis care as well as a voluntary sobering
center that offers a safe, supportive environment for intoxicated people,
including those referred by law enforcement.136 The authors of a 2001
review of three substance abuse and mental health diversion programs (in
Memphis, Tennessee; Montgomery County, Pennsylvania; and Multnomah
County, Oregon) suggest, as best practices, that such centers must adopt a
no-refusal policy—accepting people whether or not they meet criteria for

38

Vera Institute of Justice

involuntary treatment—and that the process must be less difficult and less
time consuming for officers than making an arrest.137 In Tucson, according
to the city crisis center’s former CEO, police can drop someone at the
center and be back on the street within 15 minutes, compared to the several
hours required to process an arrest.138
However, the mere existence of such centers may not be enough to
reduce criminal justice contact in the long term, particularly among people
with behavioral health problems. A 2010 meta-analysis of research into
programs both inside and outside the criminal justice system aimed at
people with mental illness found that even evidence-based mental health
interventions that have proven successful in improving clinical outcomes
are largely ineffective in lowering police contact and arrest or reducing risk
of incarceration or recidivism.139 The analysis indicates that such programs
have not yet yielded positive criminal justice outcomes primarily because
they begin with the assumption that reoffending is caused solely by
untreated mental illness, rather than by a variety of factors.140 This suggests
that if community-based interventions such as community drop-off centers
are to be successful in lowering criminal justice involvement, they need
to offer, or have access to, comprehensive services addressing the holistic
needs of high-risk populations (including unemployment, homelessness,
and involvement with peers who are engaged in criminal behavior) and not
simply focus on behavioral health symptoms and treatment.
As officials in Camden, New Jersey, discovered, addressing people’s
needs more comprehensively requires agencies to share crucial health
data across systems to help both community-health providers and
law enforcement responders better identify people’s needs and tailor
appropriate interventions.141 (See “Camden’s policing transformation” on
page 41.) Vera researchers found a similar phenomenon and need through
an effort to share behavioral health information across justice and health
systems in the District of Columbia. Vera’s research revealed that although
clinicians, officers, and other decision makers infrequently have full access
to behavioral health information about the justice-involved people with
whom they come into contact, together health and justice agencies had
access to behavioral health data for six out of every 10 arrestees.142 If
shared across agencies, this information could help clinicians and police
officers alike fashion better-tailored interventions or treatments for people

Gatekeepers: The Role of Police in Ending Mass Incarceration

39

who need it. The findings suggest that greater access to behavioral health
data can better identify people for treatment and improve continuity of
care, with the goal of reducing their contacts with both the criminal justice
and health care systems.143

Responses that tackle underlying causes that
trigger police enforcement
Because enforcement is the response to far too many violations rooted
in poverty, police need alternatives to arrest for such encounters as
well. In response to the 2016 death of Philando Castile, who was shot
and killed by a Minnesota police officer after a traffic stop purportedly
involving a broken taillight, Minnesota-based nonprofit MicroGrants
created Lights On!, a program that authorizes officers who stop motorists
for minor violations such as broken taillights to issue a $50 voucher—in
lieu of a traffic ticket—that the motorist can redeem at participating auto
repair shops.144 The program is a more humane and effective response
that provides law enforcement with a creative intervention to decrease
unnecessary arrests and avoid a vicious cycle of mounting fines when
people who cannot afford to repair their vehicles get ticketed repeatedly.145
The program has been replicated in Iowa City, Iowa, and other jurisdictions
in the Midwest hope to adopt the idea.146

Crime prevention programs
Seeding community-based violence prevention programs and
neighborhood watch organizations that deal directly with crime and related
community issues can also enhance safety while reducing arrests and
incarceration. The Audre Lorde Project in Brooklyn, New York, provides
two such programs. The Safe Neighborhood Campaign encourages
community members to promote nonviolence and to intervene in the
wake of a violent event.147 The Safe OUTside the System (SOS) Collective is
another antiviolence program with the goal of creating community-based
strategies to address crime instead of relying on the police.148

40

Vera Institute of Justice

Camden’s policing transformation
Over the past five years, Camden, New Jersey, has remade
its law enforcement culture. In December 2013, Rolling Stone
featured Camden in an article titled “Apocalypse New Jersey:
A Dispatch from America’s Most Desperate Town.”a At the
time, Camden, across the Delaware River from Philadelphia,
was one of the nation’s most dangerous cities, with a murder
rate of about 85 per 100,000 residents.b Today, while this city
of roughly 75,000 residents still faces endemic poverty as well
as myriad social and economic challenges, crime is falling.c
Camden County Police Chief J. Scott Thomson jumpstarted
the changes by going back to the drawing board. In 2013,
he disbanded the city’s police department, rethinking its
approach to policing and reconstituting it as the Camden
County Police Department, with fewer officers and an
emphasis on community policing.d Crime overall is at a 50-year
low, and 2017 marked the lowest number of murders since 1969
(the number of aggravated assaults with a firearm, however,
rose over the past three years).e The turnaround happened, in
large part, through community policing initiatives.
Thomson, a Camden native, began the transformation by
instructing his officers to get out of their cars and walk the
streets so they could get to know people.f The department
also began hosting free communitywide events such as
“Movies with Metro” film events that include popcorn, snacks,
and drinks, and pop-up block parties featuring burgers,
hotdogs, and hoops.g Because violent crime remains a
concern in Camden, the department leads an Office of

a	

Matt Taibbi, “Apocalypse, New Jersey: A Dispatch from America’s
Most Desperate Town,” Rolling Stone, December 11, 2013, https://
perma.cc/D5EY-GW57.
b	

For crime rates in Camden, see Vera Institute of Justice, “Arrest

Trends,” (Camden), https://perma.cc/87XJ-YK55. See also Andy
McNeil, “2013 Marks Deadly Year in Camden,” Courier-Post, January
5, 2014, https://perma.cc/AHS3-663K.
c	

Camden County, “Camden Officials Celebrate Topping-Off of New
Office Headquarters,” press release (Camden, NJ: Camden County,
June 12, 2018), https://perma.cc/2DQK-B2R4.
d	

Sara Holder, “What Happened to Crime in Camden?,” CityLab,
January 10, 2018, https://www.citylab.com/equity/2018/01/whathappened-to-crime-in-camden/549542/.
e	

“Camden Had Lowest Homicide Rate in 33 Years in 2017,” SNJ
Today, July 24, 2018, https://perma.cc/8RCH-3QRQ; and Sara
Holder, “What Happened to Crime in Camden?,” 2018.
f	

J. Scott Thomson, “Five Voices on Reforming the Front End of Justice:
Where Police are Told to Be ‘More Peace Corps than Special Forces,’”
The Marshall Project, July 17, 2016, https://perma.cc/254S-3MCS.

Juvenile Justice and Delinquency Prevention-funded program
called G.R.E.A.T., for Gang Resistance and Education
Training; as well as Project Guardian, which taps local
churches and community centers to spread messages of
nonviolence to young people.h
Today, the department is working closely with the Camden
Coalition of Healthcare Providers to understand and
address the considerable overlap between people who make
repeated visits to hospital emergency rooms and are also
arrested multiple times.i This data-sharing initiative, known as
Camden ARISE (Administrative Records Integrated for Service
Excellence), has identified a group of people known as superutilizers, who each experienced 16 or more emergency room
visits and seven or more arrests over a five-year period—a
sign that neither system is meeting their needs.j Because
an analysis of this data revealed that most of these arrests
were for nonviolent, low-level offenses—disorderly conduct in
particular—the police department is now working with health
care and other local service providers to create a screening
tool to divert such high-need people away from the criminal
justice system.k
Camden faces steeper challenges than most cities—37.4
percent of the population lives below the poverty line,
including nearly half of residents under 18—but the city
continues to benefit from a police department that views its
officers as guardians of the community as opposed to just
enforcers of the law.l

h	

For more on G.R.E.A.T., see G.R.E.A.T. Gang Resistance Education
and Training, “History,” https://perma.cc/9HEJ-LX5T; and Mike
Dougherty, “Dozens of Camden Students Complete Gang Resistance
Program with Police,” 3 CBS Philly, June 14, 2017, https://perma.cc/
L7PR-Y5SE. For more on Project Guardian, see George Woolston,
“Camden Police Program Helps Get At-Risk Teens on the Right Path,”
Tap into Camden, June 16, 2018, https://perma.cc/6ZRM-R845.
i	

Camden Coalition of Healthcare Providers (CCHP), About the
Camden Coalition (Camden, NJ: CCHP, 2018), https://perma.cc/
SC4G-TYD4.
j	

Anne Milgram, Jeffrey Brenner, Dawn Wiest et al., Integrated Health
Care and Criminal Justice Data—Viewing the Intersection of Public
Safety, Public Health and Public Policy Through a New Lens: Lessons
from Camden, New Jersey (Cambridge, MA: Harvard Kennedy
School, 2018), 4-6, https://perma.cc/9HQF-AE62.
k	

CCHP, “Creating a Police Screening and Diversion Tool with Camden
ARISE,” CCHP News, July 10, 2017, https://perma.cc/UWW7-4T3F.
l	

For poverty statistics in Camden, see U.S. Census Bureau,
“American Fact Finder: Community Facts—Camden, NJ,” Selected
Economic Characteristics 2013-2017 American Community Survey
g	 ESI Econsult Solutions Inc., The Positive Impacts of Investments
5-Year Estimates, https://perma.cc/6NF2-26PD. For the police
in Camden, NJ on Social Determinants of Health (Camden, NJ: ESI
department’s role, see Thomson, “‘More Peace Corps than Special
The Role of Police in Ending Mass Incarceration
Econsult Solutions Inc.,Gatekeepers:
2019), 12, https://perma.cc/77ZL-FQ4P.
Forces,’” 2016.

41

Recommendation 2: Reengineer
the 911 system

Icon

911

Nationally there are an estimated 240 million calls to
911 over the course of a year.149 The number of such calls that are for
police—rather than emergency medical or fire—service is unknown, but
a study in San Francisco found that police calls for service constituted
just under half of all calls.150 Many calls for service involve people in crisis
or other situations that police are not well-equipped to handle.151 Others
involve minor inconveniences, such as a parked car blocking a driveway.
The nation’s police officers have become the first responders to society’s

Mapping 911 calls in two counties
Tracking call trends is a crucial first step in overhauling the
911 system. Research in Surrey, British Columbia, for example,
found that mental health-related calls occurred most often
on weekdays (especially Mondays and Wednesdays), and
domestic violence calls peaked on weekends (especially
Sundays).a Other differences emerged as well, including that
mental health-related calls peaked midafternoon through
10:00 p.m., whereas domestic violence-related calls peaked
between 9:00 p.m. and midnight.b The nature and timing of
calls may vary greatly among jurisdictions, but the study
highlights the value of examining 911 call data to inform
resource allocation and strategic planning.
Ultimately, much more research is needed to fully understand
and sharpen the role of 911 call-takers and police dispatchers
with the goal of reducing unnecessary enforcement. Vera is
currently engaged in one of the most in-depth studies to
date of 911 data, looking at calls placed in Tucson, Arizona,
and Camden, New Jersey, over a two-year period. The goal
is to develop a comprehensive picture of who is seeking help
and why as the first step toward developing responsive,
practical, and scalable alternatives to police enforcement.
The analysis looks at type, frequency, and time of calls,
along with caller demographics and location, to identify
salient trends in help requests. The study will also examine

42

Vera Institute of Justice

outcomes—the time it takes police to arrive on the scene,
what (if any) enforcement actions officers take, and any
social service referrals they make.
More can be done to better analyze existing 911 data and
to standardize data collection across jurisdictions to aid
research and policy development.c Using technology that
recognizes and sorts text to discover patterns would allow
systems to, for example, create more meaningful menu
options in lieu of “fill in the blank” fields for requests,
responses, and outcomes. Having these menu options would,
in turn, help future researchers identify trends more quickly.

a	

Adam Vaughan, Kathryn Wuschke, Ashley Hewitt et al., “Variations
in Mental Health Act Calls to Police: An Analysis of Hourly and IntraWeek Patterns,” Policing: An International Journal 41, no. 1 (2018),
58-69, 64.
b	
c	

Ibid., 62-63.

See S. Rebecca Neusteter, Maris Mapolski, Mawia Khogali, and
Megan O’Toole, The 911 Call Processing System: A Review of the
Literature as it Relates to Policing (New York: Vera Institute of Justice,
2019), www.vera.org/911-call-processing-review.

ailments, major and minor, and too often respond with the tool that is most
familiar, expedient, and rewarded: enforcement.
This trend is at odds with the widely shared belief that police should
be developing better community relations and preventing and solving
serious crime.152 The search for better options in responding to 911 calls
is not just about conserving scarce resources, it is also about building
bridges between law enforcement and the communities it serves. 911
call centers and police dispatchers can and likely will play a critical role
in advancing the use of alternative responses to enforcement.153 For
example, by directing the right resources to the right callers, dispatchers
can save scarce police resources and bolster community relationships.
But before they can do this, they will need to be able to identify calls that
require highly specialized responses—such as those involving mentally
ill people and others in crisis—and differentiate them from calls meriting
a nonemergency, nonclinical response.154 (See “Mapping 911 calls in two
counties” on page 42.)

Divert calls to crisis or nonemergency
help hotlines
Calls that do not require an emergency law enforcement response should
be diverted to more appropriate resources. Review of data from two very
different places—New York City and Tucson, Arizona—suggests that
three out of four calls in which police respond don’t involve a crime in
progress.155 Data from Seattle offers an even more nuanced picture. In 2017,
of the approximately 450,000 911 calls for police service citywide, two
of the top three concerns were related to traffic and other disturbances.156
Seattle police were just as likely to respond to a traffic-related 911 call as
they were to a call about a suspicious circumstance.157 Many of these trafficrelated calls likely do not involve serious damage or injury: nationally
about 70 percent of traffic accidents involve property damage only.158 The
default to police—especially in places with already frayed communitypolice relations—can lead to worse outcomes than if no one had responded,
including unnecessary arrests.159
Ideally, dispatchers can direct nonemergency 911 calls away from
immediate police response, preserving law enforcement resources and

Gatekeepers: The Role of Police in Ending Mass Incarceration

43

deploying responses tailored to the individual circumstance. Training
call-takers and dispatchers on the manifestations of mental illness can
promote more accurate routing of calls. A survey by the Police Executive
Research Forum found that while some police agencies believe it is easy
to determine if a call is mental-health related, others believe this judgment
requires officers on the scene.160 Some police agencies have already
directly linked 911 call-takers and police dispatchers with mental health
professionals to more accurately assess the nature of calls and trigger the
best response.161 There is also evidence that in communities known to have
mental health crisis intervention teams, callers may specifically ask for the
team.162 Crisis hotlines, too, can reduce the use of expensive—and often less
effective—hospital emergency services by connecting people with muchneeded behavioral health services.163
Many cities already have nonemergency help lines that ease the
burden on 911.164 In the United States and Canada, 211 is reserved as a
nonemergency number for information and referrals to nonemergency
health services, human service providers, and social service organizations.
In 2017, more than 13 million 211 calls were placed, the majority of which
were referrals for physical and mental health services, employment
opportunities, homelessness prevention services, and housing assistance.165
Many cities, including New York City, also have 311 lines, where callers
can get city-specific help. Callers to 311 can receive information about local
services, make complaints, or report issues such as graffiti or potholes.166

Recommendation 3: Increase the
number and types of offenses
that don’t require punitive
enforcement

other
response

Efforts to pursue targeted reductions in punitive enforcement should be
rooted in data analysis. These analyses should look at both the volume of
punitive enforcement and where the greatest disparities in enforcement
occur vis-à-vis race, ethnicity, age, gender, and neighborhood. (See
“Understanding and mitigating racial and ethnic disparities in policing

44

Vera Institute of Justice

practices” on page 47.) Such reforms must incorporate policies or protocols
that track enforcement activities demographically and outline the
discretion of individual officers, such as establishing do-nothing policies
related to some low-level offenses and requiring a supervisor’s approval
before arresting or citing someone for a particular category of offense
under certain circumstances.
Work in Durham, North Carolina, provides a good example of targeted
reductions in law enforcement use of arrests. For decades, going back
to 1980, police in Durham overrelied on arrests, logging roughly 10,000
annually.167 In 2008, this trend began to shift dramatically and, over the
next nine years, the number of arrests dropped 64 percent, down to
about 3,500 in 2016.168 Several factors have contributed to these declines.
In 2016, the Durham Police Department changed its policy to explicitly
require officers to use their “personal judgment, knowledge, skill, and
insight” to evaluate all possible responses.169 The department also permitted
noncustodial alternatives unless an arrest is required by law or there
is a clear safety concern.170 In 2016, the department directed officers to
cite, rather than detain, people charged with misdemeanor marijuana
offenses when “there is no danger to person or property, the suspect has
valid identification and address, isn’t wanted on other charges, and has no
previous charges other than traffic citations.”171 That policy appears to be
working: arrests for misdemeanor marijuana possession or possession of
drug paraphernalia fell 65 percent between 2013 and 2017.172 (See Figure 5
on page 46.)
For jurisdictions looking to significantly lower the incidence of
unnecessary arrests, law enforcement agencies can consider do-nothing
policies in relation to offenses that drive arrest volume but have little
impact on public safety. Liquor law violations, curfew violations, and
loitering are good examples of such offenses. Between 2010 and 2016
in the United States, there were more than 430,000 arrests for curfew
violations and loitering, and more than 2.6 million arrests for liquor
violations.173 These types of offenses clearly constitute a significant portion
of arrests nationwide, but jurisdictions instituting such a policy should
confirm whether these offenses drive overall arrests in their localities.
Reform-oriented policy changes of this kind hold much promise. But
law enforcement agencies should be mindful that there may be unintended

Gatekeepers: The Role of Police in Ending Mass Incarceration

45

detrimental effects in their implementation. For example, after Colorado
legalized marijuana in 2012 for personal use, arrests were cut in half, but
the racial disparity in arrests among people less than 21 years of age—who
are not allowed to use marijuana under the new law—widened.174 A survey
by the Colorado Department of Public Health and Environment found little
difference in teenage marijuana use by race or ethnicity in the state, but
the arrest rate for white 10- to 17-year-olds fell by 8 percent from 2012 to
2014, while arrest rates for Latino and black youths rose 29 percent and
58 percent respectively.175 Arrest disparities also exist for adults. Again,
although there is little difference in marijuana use by race and ethnicity
among adults in Colorado, after legalization, the arrest rate fell by 51
percent for white people, but only by 33 percent for Latino people and 25
percent for black people.176
Similarly, Durham, North Carolina, has seen significant overall
reductions in marijuana arrests, but racial disparities in those arrests

Figure 5

Total arrests, Durham, North Carolina, 2008–2016
12K
64%
decrease

9,905
8K

3,546

4K

16
20

14
20

12
20

20
10

20
0

8

0

Source: Vera Institute of Justice, “Arrest Trends” (Durham Police Department, 1980-2016), https://arresttrends.vera.org/arrests?location=1570
1&estimated=0#bar-chart.

46

Vera Institute of Justice

have persisted, and there has been a rise in marijuana-related citations as
a proportion of overall marijuana-related charges.177 And while Durham
has made strides in reducing its reliance on custodial arrests for lowlevel marijuana offenses, an increase in the use of citations can still
inadvertently lead to arrest down the line: if people are unable to appear
in court, pay required fines, or fulfill any other condition imposed by the
court, a judge may issue a bench warrant that can exponentially increase
the risk of future jail time. More work is thus required to ensure that
people’s heightened risk of jail time because of conditions imposed by
citations—even if a delayed risk—can be mitigated.

Understanding and mitigating racial and ethnic disparities in police practices
In order to understand where racial and ethnic disparities
exist in police enforcement, law enforcement agencies need
to develop research and management protocols designed to
find and address them. Data is needed to

›› track enforcement patterns and practices by
demographic group;

›› identify institutional factors that may lead to disparate
racial or ethnic outcomes;

›› track and assess how officers are applying their

police action.c Existing implicit bias training modules are
limited in nature: at present, there are few specific guidelines
for police departments on what courses should include,
how to teach the material, whether it is effective, or how to
measure any effects it may have. A 2016 meta-analysis of
implicit bias studies showed that unconscious biases can
be altered, but it remains to be seen what type of training—
and how much—will be most effective in addressing this
unconscious process.d

discretion; and

›› implement policies and strategies focusing on racial and
ethnic fairness to correct discriminatory practices where
discovered.

Some law enforcement agencies are beginning to address
implicit bias in policing through officer training about the
phenomenon.a Implicit biases are unconscious attitudes
that people hold about groups of people—often linked to
race, ethnicity, economic class, and gender—that influence
perception and behavior despite conscious intention to
view people as individuals.b Because people have a strong
propensity to lean on stereotypes to fill in the blanks about
people they don’t know, such reflexive responses may affect

a	

Bailey Maryfield, Implicit Racial Bias (Washington, DC: Justice
Research and Statistics Association, 2018), https://perma.cc/Q3J857FL.
b	

Ibid., 1.

c	

Cheryl Staats, Kelly Capatosto, Robin A. Wright, and Danya
Contractor, State of the Science: Implicit Bias Review 2015
(Columbus, OH: Ohio State University Kirwan Institute, 2015), 5, 9-12,
https://perma.cc/UAZ3-UTTE.
d	

Patrick Forscher, Calvin K. Lai, Jordan R. Axt et al., “A MetaAnalysis of Change in Implicit Bias,” Journal of Personality & Social
Psychology (2016), https://perma.cc/XA3U-33L2.

Gatekeepers: The Role of Police in Ending Mass Incarceration

47

rnat

Alte

Icon

ives

Recommendation 4: Expand the
reach of alternatives to arrest

Existing police diversion programs remain limited in scope
and in application. Narrow eligibility criteria restrict most programs to
people who are alleged to have committed low-level, nonviolent offenses—
often excluding anyone with a lengthy criminal history, particularly
those charged with typically ineligible violent and sex offenses. Indeed,
research discussed earlier highlighted the small scale and reach of existing
formal diversion programs—and suggested that police officers may not be
diverting people even in circumstances where they have met all objective
prerequisites or eligibility requirements, resulting in smaller programs
than if all eligible people were participating.178 (See “Working at the
margins: Alternatives to arrest programs” on page 32.)
Given that the majority of arrests nationwide are for minor violations—
drug possession, disorderly conduct, public drunkenness, vagrancy,
loitering, curfew violations, and vandalism—current alternatives to
punitive enforcement could cast a wider net and mitigate the number of
traditional arrests.
To reduce the number of people needlessly burdened by the collateral
impacts of criminal justice contact, police and policymakers should link
both criteria for diversion and decisions on when to use such programs to
the drivers of arrest volume in their localities. By focusing on the groups
or activities responsible for the most arrests, police can optimize efforts
to divert more people more quickly. More importantly, the exercise of
police discretion may need to be more clearly operationalized and limited.
For example, police authorities should define the circumstances in which
officers should reroute cases to an alternatives-to-arrest track in order to
minimize ambiguity or confusion. Existing alternative-to-arrest programs
have only begun to scratch the surface in terms of who might be eligible
and benefit from such approaches.

48

Vera Institute of Justice

Recommendation 5: Create
structural incentives for police to
use alternatives

Icon

Rank and file officers will only begin to treat arrests as an outcome with
broad negative collateral consequences for the people they arrest—rather
than a simple output of policing activity—when police executives and
middle managers convey that message clearly. Leadership must back up
those messages with policies—and performance metrics—that encourage
and reward officer use of alternatives to punitive enforcement.179 This could
include everything from measuring levels of community satisfaction, trust,
and perceptions of safety to the number of times officers refer someone to a
community service provider instead of arresting them—and attaching these
metrics to professional incentives and rewards.180 What law enforcement
agencies choose to measure can send a strong signal to both officers and
the public about what is valued and important in policing. Such a change
would amount to a major shift in police culture, but one for which there
are no published examples from which agencies can learn.
As the CompStat discussion on page 24 demonstrated, data can be a
powerful tool for constructing law enforcement interventions. Indeed,
CompStat’s implementation and replication across the country proved how
counting something—in this case, punitive enforcement—can inherently
“confer a status on, or suggest the importance of, the thing counted.”181
After CompStat’s widespread adoption, high numbers of arrests, citations,
and stops became synonymous with police success.182
Police agencies that want to change how they define success
need to create quantifiable measurements for interventions beyond
punitive enforcement and then measure and reward officers’ use of
those alternatives. With this goal in mind, Vera and the National Police
Foundation have created a new framework to facilitate this challenging
work: CompStat360.183 Developed through a public-private partnership
between Vera and the National Police Foundation, and with support
from the MacArthur and Ford foundations and the U.S. Department of

Gatekeepers: The Role of Police in Ending Mass Incarceration

49

Justice’s Community Oriented Policing Services (COPS) Office and Bureau
of Justice Assistance, this management and accountability framework
offers a new performance management model that is rooted in problemsolving and community satisfaction as a crucial balance to CompStat’s
traditional enforcement metrics. In contrast to CompStat’s original primary
(and often exclusive) focus on violent crime and responses that typically
encourage and incentivize enforcement, CompStat360 uses a proactive,
problem-solving approach that factors in community needs and feedback
to its interpretation of quantitative data. By encouraging creativity and
collaboration in setting police policy, it aims to avoid defaulting to and
rewarding enforcement-based responses.

Recommendation 6: Find
out what works best through
experimentation, research, and
analysis
Research on alternatives to enforcement is very limited. The reasons
why and under what circumstances officers default to enforcement are
still unclear. There is more to learn about what is needed in terms of
alternative responses—and, most importantly, what types of alternativesto-arrest work, for whom, and why. Studies should also examine
common challenges to implementation. If interventions aim to reduce
enforcement—as measured by arrests, citations, and incarceration—then
research is needed to assess whether they achieve these goals. One
necessity to further research, experimentation, and analysis is complete
data. Local, state, and federal governments should prioritize and invest
in infrastructure and processes to collect and analyze arrest data in
uniform ways so that it can be compared across agencies and jurisdictions,
including closing one huge data gap: the number and nature of arrests by
ethnicity. Moreover, additional data is needed to further understand the
application, spread, and impact of enforcement and diversion activities
more broadly.
One pitfall when expanding alternatives to enforcement such as
citations in lieu of arrest or diversion programs is that they still require
50

Vera Institute of Justice

justice system oversight. Thus, police initiatives intended to apply a lighter
touch can end up pulling people into the criminal justice net who wouldn’t
be there otherwise. As these initiatives are implemented, it is essential for
police departments, policymakers, and the public to track whether they
inadvertently lead to justice system net-widening and if certain groups are
less likely to reap the benefits of diversion from arrest.
Some funders are taking note of this gap in both practice and empirical
evidence. The MacArthur Foundation’s Safety and Justice Challenge funds
work in 52 jurisdictions nationwide to reduce the use of local jails and
to support needed research that seeks to advance policy and practice in
this space, in part by reining in punitive enforcement.184 This initiative
involves deep collaborations between local justice stakeholders—including

Local, state, and federal governments
should prioritize and invest in
infrastructure and processes to collect
and analyze arrest data in uniform
ways so that it can be compared across
agencies and jurisdictions.

law enforcement leaders—to examine and implement policies, programs,
cultural transformations, and incentives all toward the outcome of
reducing the local criminal justice footprint.185 And, in September 2018,
the Laura and John Arnold Foundation (now Arnold Ventures) released
a request for proposals from organizations and individuals seeking to
evaluate programs that aim to increase the capacity of police and other
first responders to recognize people in crisis and channel them toward
treatment and other services in lieu of enforcement.186

Gatekeepers: The Role of Police in Ending Mass Incarceration

51

Conclusion

T

here is little doubt that when people and communities call on the
police for help—with maintaining order, responding to crime, or
resolving community problems like homelessness, substance use,
mental illness, and intrafamilial disputes—police too often employ a set of
punishing responses they should use more sparingly. Indeed, police have
become the default first responders to a number of issues that would be
better handled by the public health system or community-based social
services. And, in carrying out their duties, police officers have increasingly
relied on and defaulted to more and more intrusive measures: aggressively
patrolling communities, stopping and frisking residents, issuing tickets and
summonses, and making far too many custodial arrests in situations where
other responses—including nonlegal strategies—may be more appropriate.
These policing practices leave residents feeling both overpoliced and
underprotected, undermining police legitimacy and overall community
health and safety.
Communities of color, which often also suffer from concentrations
of poverty, have suffered most under this paradigm. Mass enforcement
ends up criminalizing people—disproportionately black Americans—who
pose no or very little threat to public safety, and it too often results in
incarceration. But after landing in jail, sometimes repeatedly, people who
have been arrested fall more deeply into the criminal justice system, often
saddled with debt from unpaid fines and fees. Moreover, the criminalizing
process rarely addresses the underlying life issues that led them to criminal
justice contact in the first place. Rank and file police officers suffer too,
from the risks inherent in such interventions to the strain that the job of
policing places on their mental and physical well-being—and that of their
families.
This leaves police and communities in an untenable position. Police
officers remain the go-to source of help for many people, but they lack the
training, tools, and resources to adequately respond to or solve many of
the issues they encounter. Instead, by defaulting to enforcement and often
arrest, police allow fear, distrust, and rancor to germinate between them

52

Vera Institute of Justice

and the communities they serve. A new model is needed: one that rebuilds
trust, maintains public safety, and reserves arrest as a last resort. There is
a moral and strategic imperative to reshape police culture and incentives
to ensure that the exercise of police discretion is more deliberate,
transparent, and accountable and that policing practices are focused more
tightly on dramatically reining in arrests, while promoting lesser forms of
enforcement and holding officers and departments accountable.187

A new model is needed: one that rebuilds
trust, maintains public safety, and
reserves arrest as a last resort.

This will require hard work, patience, political will, and an enormous
shift in worldview. Police leaders must give officers training, support,
and incentives to forego punitive responses and problem-solve instead,
as well as to collaborate with and, in many circumstances, defer to nonlaw enforcement agencies and others—such as community organizations,
mental health service providers, social workers, detoxification centers,
mediators, and even families—that are far better positioned to respond
to many of the crises that police now routinely find themselves forced
to reflexively react and respond to first. And elected officials and other
leaders who are responsible for appointing or otherwise supporting police
executives must clearly focus in on policing culture, data, and outcomes the
communities they serve wish to see and provide those communities and
police executives with the resources and time required to achieve those
desired results.
Police officers alone cannot repair the breach in trust or uphold
the peace. It takes a village to fix a village. The work requires a holistic
approach to investing in long-neglected communities that demands the

Gatekeepers: The Role of Police in Ending Mass Incarceration

53

engagement of civil society and community groups as well as professionals
in policing, public health, and social services. Approaching these issues
on multiple fronts is essential if America wants to end the crisis of
criminalization—and, by extension, mass incarceration—and to refashion
policing into a pursuit that truly operates in the public interest, worthy of
public trust.

54

Vera Institute of Justice

Endnotes
1	

The Federal Bureau of Investigation’s Uniform Crime Reporting
(UCR) program gathers voluntarily produced data from more than
18,000 city, university and college, county, state, tribal, and federal
law enforcement agencies. In 2016, the UCR program reported that
there had been 10,662,252 arrests. In the same year, the Bureau
of Justice Statistics reported 10.6 million jail admissions. Federal
Bureau of Investigation (FBI), “UCR: 2016 Crime in the United
States,” table 18: Estimated Number of Arrests, United States, 2016,
https://perma.cc/9RPD-G7MZ; and Zhen Zeng, Jail Inmates in 2016
(Washington, DC: U.S. Department of Justice (DOJ), Bureau of
Justice Statistics (BJS), 2018), 1, https://perma.cc/R57H-9XQD.

2	

In 1994, there were 14 million arrests and approximately 9.8 million
jail admissions. Craig A. Perkins, James J. Stephan, and Allen J.
Beck, Jails and Jail Inmates 1993-94 (Washington, DC: BJS, 1995),
13, https://bjs.gov/content/pub/pdf/jaji93.pdf.

3	

In 1970, the violent crime rate was 363.5 per 100,000 and the
property crime rate was 3,621 per 100,000; in 2014, the most recent
year available in the database, the rates were 375.7 and 2,596.1
per 100,000, respectively. FBI, UCR, “Uniform Crime Reporting
Statistics, UCR Data Online, Violent Crime and Property Crime
rates from 1960–2014,” database (Washington, DC: BJS), https://
www.bjs.gov/ucrdata/Search/Crime/State/StatebyState.cfm.
In 1980, the arrest rate was 4579.1 per 100,000 people; by 2016
it had fallen to 3,239.3 per 100,000. Vera Institute of Justice,
“Arrest Trends,” (arrest rates from 1980-2016), database (New
York: Vera Institute of Justice), https://arresttrends.vera.org/
arrests?unit=rate#bar-chart.

4	

Vera Institute of Justice, “Arrest Trends,” https://arresttrends.vera.
org/arrests.

5	

For current arrest data and proportion of offenses, see FBI, “UCR:
2016 Crime in the United States,” table 18; and FBI, “UCR: 2017
Crime in the United States,” table 29: Estimated Number of Arrests,
United States, 2017, https://perma.cc/6M35-NFP4. For historical
comparison data see FBI, “UCR: 2010 Crime in the United States,”
table 29: Estimated Number of Arrests, United States, 2010, https://
perma.cc/LQ9B-7V89; Howard N. Snyder, Arrest in the United States,
1990-2010 (Washington, DC: BJS, 2012), 1-15, https://perma.cc/
U2VN-UXTA; DOJ, Office of Justice Programs, Crime and Justice
Atlas 2000 (Washington, DC: DOJ, Office of Justice Programs,
2000), 34-41, https://perma.cc/HHK5-CDZY; and FBI, “Uniform
Crime Reporting Program Data [United States]: County Level Arrest
and Offenses Data, 1977-1983,” database (Ann Arbor, MI: Interuniversity Consortium for Political and Social Research, 2006),
https://www.icpsr.umich.edu/icpsrweb/NACJD/studies/8703.

6	

For the frequency and nature of arrests, see Vera Institute of
Justice, “Arrest Trends”; Vera Institute of Justice, “Incarceration
Trends,” database (New York: Vera Institute of Justice, 2018),
https://www.icpsr.umich.edu/icpsrweb/NACJD/studies/8703; and

S. Rebecca Neusteter and Megan O’Toole, Every Three Seconds:
Unlocking Police Data on Arrests (New York: Vera Institute of Justice,
2019), 6, https://perma.cc/YBL9-GNXT.
7	

Whether increased enforcement of low-level offenses improves
public safety is a subject of ongoing debate; research has not
definitively demonstrated that it does. See George L. Kelling and
James Q. Wilson, “Broken Windows: The Police and Neighborhood
Safety,” Atlantic, March 1982, 29-38, https://perma.cc/5FCD-3JXE;
Bernard E. Harcourt, Illusion of Order: the False Promise of Broken
Windows Policing (Cambridge, MA: Harvard University Press, 2004);
and Jacinta M. Gau and Rod K. Brunson, “Procedural Justice and
Order Maintenance Policing: A Study of Inner-City Young Men’s
Perceptions of Police Legitimacy,” Justice Quarterly 27, no. 2 (2010),
255-279, https://perma.cc/Z3EB-TJQS.

8	

Mary T. Phillips, Pretrial Detention and Case Outcomes, Part 1:
Nonfelony Cases (New York: New York City Criminal Justice
Agency, 2007), 4-5 & 55-60, https://perma.cc/TLK7-EARH;
Christopher T. Lowenkamp, Marie VanNostrand, and Alexander
Holsinger, Investigating the Impact of Pretrial Detention on
Sentencing Outcomes (Houston, TX: Laura and John Arnold
Foundation, 2013), 10-11, https://perma.cc/YE7L-RDEF; and
Christopher T. Lowenkamp, Marie VanNostrand, and Alexander
Holsinger, The Hidden Costs of Pretrial Detention (Houston, TX:
Laura and John Arnold Foundation, 2013), 10-11, https://perma.cc/
YM25-4MZL.

9	

See for example Gary Fields and John R. Emshwiller, “As Arrest
Records Rise, Americans Find Consequences Can Last a Lifetime,”
Wall Street Journal, August 18, 2014, https://perma.cc/CH3X-32LH;
Benjamin Geffen, “The Collateral Consequences of Acquittal:
Employment Discrimination on the Basis of Arrests without
Convictions,” University of Pennsylvania Journal of Law and Social
Change 20, no. 2 (2017), 81-103, https://perma.cc/662T-NJD8;
Corinne Carey, No Second Chance: People with Criminal Records
Denied Access to Public Housing (New York: Human Rights Watch,
2004), https://perma.cc/WQK8-EPQX; and Rebecca Oyama, “Do
Not (Re)Enter: The Rise of Criminal Background Tenant Screening as
a Violation of the Fair Housing Act,” Michigan Journal of Race and
Law 15, no. 1 (2009), 182-222, https://perma.cc/8LYM-J9JM.

10	 Alexander Williams Jr., Final Report of the Collateral Consequences
Workgroup (College Park, MD: University of Maryland College
of Behavioral and Social Sciences, 2016), 7, https://perma.
cc/9MVF-LQSC; and NICCC, “Collateral Consequences Inventory,”
database (Washington, DC: CSG Justice Center), https://niccc.
csgjusticecenter.org/database.
11	 Mike Males and Elizabeth Brown, “Teenagers’ High Arrest Rates:
Features of Youth or Poverty?” Journal of Adolescent Research 29,
no. 1 (2014), 3-24.

Gatekeepers: The Role of Police in Ending Mass Incarceration

55

12	 Amanda Geller, Jeffrey Fagan, Tom Tyler, and Bruce G. Link,
“Aggressive Policing and the Mental Health of Young Urban Men,”
American Journal of Public Health 104, no. 12 (2014), 2321-2327,
https://perma.cc/JFF4-UPU6.
13	 For links between trauma and justice system involvement of
youth, see Geller, Fagan, Tyler et al., “Aggressive Policing,” 2014;
Carly B. Dierkhising, Susan J. Ko, Briana Woods-Jaeger et al.,
“Trauma Histories among Justice-Involved Youth: Findings from the
National Child Traumatic Stress Network,” European Journal of
Psychotraumatology 4, no. 1 (2013), 1-12, https://perma.cc/RH4XQUFW; Julian D. Ford, John Chapman, Daniel F. Connor, and Keith
R. Cruise, “Complex Trauma and Aggression in Secure Juvenile
Justice Settings,” Criminal Justice and Behavior 39, no. 6 (2012),
694-724, https://perma.cc/75B4-XP98; and Julian D. Ford, J. Kirk
Hartman, Josephine Hawke, and John F. Chapman, “Traumatic
Victimization, Posttraumatic Stress Disorder, Suicidal Ideation, and
Substance Abuse Risk Among Juvenile Justice-Involved Youth,”
Journal of Child & Adolescent Trauma 1, no. 1 (2008), 75-92.
14	 See for example Malcolm Gladwell, “Million Dollar Murray,” New
Yorker, February 13, 2006, https://perma.cc/S5ZL-ZJTU.
15	 Overpoliced—and, by extension, overincarcerated—communities
suffer significant damage to social networks and long-term
opportunities. The damage affects children, parenting relationships,
family functioning, mental and physical health, employment
opportunities, and economic and political infrastructures.
For examples of ways in which black communities experience
disproportionate policing and enforcement, see Elizabeth Hinton,
LeShae Henderson, and Cindy Reed, An Unjust Burden: The
Disparate Treatment of Black Americans in the Criminal Justice
System (New York: Vera Institute of Justice, 2018), https://perma.cc/
KRN7-4SFD. For the impacts of these policies, see Todd Clear, “The
Effects of High Imprisonment Rates on Communities,” Crime and
Justice 37, no. 1 (2008), 97-132.
16	 For example, a 2018 study undertaken by researchers from Harvard
and Boston universities showed a correlation between police
shootings of unarmed black people and the mental health of black
Americans living in the states where the shootings happened.
Using individual-level data from the 2013–2015 U.S. Behavioral Risk
Factor Surveillance System, researchers mapped responses from
black Americans on their current mental health to police killings of
unarmed black people occurring within the previous three months
in the same state. They found that each killing directly correlated
to as much as two weeks’ worth of poor mental health and that
the same effects did not occur for police killings with different
characteristics, such as the killings of armed black people or
unarmed white people. Jacob Bor, Atheendar S. Venkataramani,
David R. Williams, and Alexander Tsai, “Police Killings and Their
Spillover Effects on the Mental Health of Black Americans: A
Population-Based, Quasi-Experimental Study,” The Lancet 392, no.
10144 (2018), P302-P310, https://perma.cc/6SH7-L8LC.
17	 Vera Institute of Justice, “Arrest Trends: Demographics,” https://
arresttrends.vera.org/demographics.

56

Vera Institute of Justice

18	Ibid.
19	 See for example James Poniewozik, “A Killing. A Pointed Gun. And
Two Black Lives, Witnessing,” New York Times, July 7, 2016, https://
perma.cc/9E6R-3MUJ; “Officer Convicted of Murder in Slaying of
Laquan McDonald,” Associated Press, October 5, 2018, https://
perma.cc/V9WX-EP2H; Mihir Zaveri, “Black Man Killed by Officer
in Alabama Mall Shooting Was Not the Gunman, Police Now Say,”
New York Times, November 24, 2018, https://perma.cc/69S8-J5ML;
Maxine Bernstein, “Portland Officer Said He Fired AR-15 Rifle at
Quanice Hayes to Defend Himself, Other Officers,” Oregonian,
March 27, 2017 (updated March 29, 2017), https://perma.cc/42LCZZ8K; Lynn Thompson, “Seattle Police Fatally Shoot Black Seattle
Mother; Family Demands Answers,” Seattle Times, June 18, 2017
(updated June 20, 2017), https://perma.cc/7X76-CLTV; and Melissa
Etehad, “Cleveland Policeman Who Shot Tamir Rice Is Fired, But
Not Because of the 12-Year-Old’s Death,” Los Angeles Times, May 30,
2017, https://www.latimes.com/nation/nationnow/la-na-tamir-riceofficer-fired-20170530-story.html.
20	 See Teddy Grant, “A Look Back on the Death of Mike Brown in
Ferguson,” Ebony, August 9, 2018, https://perma.cc/NW2MPJPC; PR Lockhart, “Eric Garner Died During a 2014 Police
Encounter. An Officer Involved Might Lose His Job,” Vox, June 7,
2019, https://perma.cc/96G2-NS7C; and Tracy Swartz, “Sandra
Bland Documentary: What We Learned about the Chicago-Area
Native and Her Mysterious Death,” Chicago Tribune, December
4, 2018, https://perma.cc/L8G8-GQA6. For the frequency of
police shootings by race, see Roland Fryer, “Reconciling Results
on Racial Differences in Police Shootings,” American Economic
Review (Papers and Proceedings) 2 (2018), 1-6, 6, https://perma.
cc/78YB-73S8. For an analysis of the community cost of deaths
from police use of force, see Anthony Bui, Matthew Coates, and
Ellicott Matthay, “Years of Life Lost Due to Encounters with Law
Enforcement in the USA, 2015–2016,” Journal of Epidemiology and
Community Health 72 (2018), 715-718, https://perma.cc/RSB3-4FJ2.
21	 “Fatal Force,” Washington Post, accessed January 11, 2019,
https://www.washingtonpost.com/graphics/2018/national/policeshootings-2018/; and https://www.washingtonpost.com/graphics/
national/police-shootings-2017/.
22	 For a brief history of laws and enforcement policies deliberately
targeting black communities after the Civil War, see Ruth
Delaney, Ram Subramanian, Alison Shames, and Nicholas Turner,
Reimagining Prison (New York: Vera Institute of Justice, 2018),
37-39, https://perma.cc/D392-2FW2. For the Texas Rangers, see
Maria Muñoz Martinez, The Injustice Never Leaves You: Anti-Mexican
Violence in Texas (Cambridge, MA: Harvard University Press, 2018).
For Stonewall, see Michael Gold and Derek M. Norman, “Stonewall
Riot Apology: Police Actions Were ‘Wrong,’ Commissioner Admits,”
New York Times, June 6, 2019, https://perma.cc/93X7-9AQ4. For ICE,
see Caitlin Dickson, “ICE Expands Controversial Partnerships with
Local Law Enforcement,” Yahoo News, May 25, 2018, https://perma.
cc/7VF3-37WA.

23	 Although much has changed since the civil rights movement of the
1950s and 1960s when many explicitly racist laws were dismantled,
the interconnection between racism and policing clearly persists in
troubling ways, and law enforcement agencies still grapple with this
history. For example, the findings of the 1968 Kerner Commission
report, which examined the causes of 1967’s deadly riots in Detroit,
Michigan, and Newark, New Jersey, echo more recent findings
of DOJ following investigations of local police departments in
Baltimore, Chicago, Cleveland, and Ferguson, Missouri. While
separated by 50 years, the investigations found similar racially
disparate policing patterns and practices—from searches to arrests
to uses of force—as well as outright racism among some officers.
For the findings of the Kerner Commission, see National Advisory
Commission on Civil Disorders, Report of the National Advisory
Commission on Civil Disorders (Washington, DC: U.S. Government
Printing Office, 1968), https://perma.cc/284S-48FN. For analysis
of the report in recent context with a focus on policing, see Nicole
Lewis, “The Kerner Omission,” The Marshall Project, March 21, 2018,
https://perma.cc/Y7EB-8EHN. For DOJ’s investigations, see DOJ
Civil Rights Division, Investigation of the Baltimore City Police
Department (Washington, DC: DOJ, 2016), https://perma.cc/AB4U4QJY; DOJ Civil Rights Division, Investigation of the Chicago Police
Department (Washington, DC: DOJ, 2016), https://perma.cc/9D4B2M82; DOJ Civil Rights Division, Investigation of the Cleveland
Division of Police (Washington, DC: DOJ, 2014), https://perma.
cc/QQE4-TE67; and DOJ Civil Rights Division, Investigation of the
Ferguson Police Department (Washington, DC: DOJ, 2015), https://
perma.cc/UPH2-6MX6
24	 Tom Tyler and Jeffrey Fagan, “Legitimacy and Cooperation: Why
Do People Help the Police Fight Crime in Their Communities?”
Ohio State Journal of Criminal Law 6, no. 231 (2008), 231-275,
262, https://digitalcommons.law.yale.edu/cgi/viewcontent.
cgi?article=4027&context=fss_papers.
25	 For 2017 assault data, see FBI, “2017 Crime in the United States,”
table 83: Law Enforcement Officers Killed and Assaulted, 2017,
https://perma.cc/84MD-YJFK. For data on officers who died in
the line of duty in 2017, see FBI, “2017 Law Enforcement Officers
Killed & Assaulted” (officers feloniously killed), https://perma.cc/
P6EF-4CME; and FBI, “2017 Law Enforcement Officers Killed and
Assaulted,” (officers accidentally killed), https://perma.cc/6QDARPPJ.
26	 John M. Violanti, Cynthia F. Robinson, and Rui Shen, “Law
Enforcement Suicide: A National Analysis,” International Journal of
Emergency Mental Health and Human Resilience 15, no. 4 (2013),
289-297, 293 (finding that detectives/police/investigators as a
group have a suicide risk 82 percent higher than the U.S. working
population comparison group), https://perma.cc/6QYP-G2XX .
27	 For a discussion of the shortage of law enforcement personnel
and its roots, see Martin Kaste, “Shortage of Officers Fuels Police
Recruiting Crisis,” NPR, December 11, 2018, https://perma.cc/3QDSD2GR; and Nicole Cain, “A Profession in Crisis: Addressing
Recruitment and Hiring Practices in Law Enforcement,” PoliceOne,
December 8, 2018, https://perma.cc/M8A5-NUUT. For discussions

of its effects, see for example Daniel Connolly, “Memphis Police
Department Loosens College Requirement for Recruits,” Memphis
Commercial Appeal, June 25, 2018, https://perma.cc/QP7V-ASDA;
Mike Maciag, “With Fewer Police Applicants, Departments Engage
in Bidding Wars,” Governing, April 2018, https://perma.cc/7LYPULUQ; and Rebecca Muller, “Are Police Officers the New Face of
Burnout?” Thrive Global, August 21, 2018, https://thriveglobal.com/
stories/police-burnout.
28	 DOJ, Bureau of Justice Assistance (BJA), National Sources of Law
Enforcement Employment Data (Washington, DC: BJA, 2016), 5 &
table 4, https://perma.cc/G3FJ-WL34.
29	 For the limited interactions between state and rural police with
communities, see generally Gary Cordner, State Police and
Community Policing (Washington, DC: DOJ, Community Oriented
Policing Services and Kutztown University, 2014), https://perma.cc/
YU3D-EPC4.
30	 International Association of Chiefs of Police (IACP), Citation in Lieu
of Arrest: Examining Law Enforcement’s Use of Citation across the
United States (Alexandria, VA: IACP, 2016), 3, https://perma.cc/
M3KD-A9PP.
31	 FBI, “2017 Crime in the United States” (persons arrested), https://
perma.cc/9YJW-9ZUS.
32	 U.S. Census Bureau, “National Population Totals and Components
of Change: 2010-2018,” https://www.census.gov/data/tables/timeseries/demo/popest/2010s-national-total.html.
33	 For percentage of arrests by race, see FBI, “2016 Crime in the
United States,” table 21: Arrests by Race and Ethnicity, 2016,
https://perma.cc/CCK5-P8A9. For population distribution by race,
see U.S. Census Bureau, “QuickFacts,” https://perma.cc/B9Q8TQSB. For Native American arrests, see Vera Institute of Justice,
“Arrest Trends: Demographics,” https://arresttrends.vera.org/demo
graphics?race=N&unit=rate#bar-chart.
34	 Using data from the National Longitudinal Survey of Youth 1997
for the period 1997–2008, researchers found that, by age 23, the
arrest prevalence among American youth overall for those years
was approximately 30 percent, with upper and lower bounds of
25.3 percent and 41.4 percent if no assumptions were made about
missing cases, and a lower bound of 30.2 percent if missing cases
were equally likely to have been arrested as the observed cases.
Robert Brame, Michael G. Turner, Raymond Paternoster, and Shawn
D. Bushway, “Cumulative Prevalence of Arrest from Ages 8 to 23 in
a National Sample,” Pediatrics 129, no. 1 (2012), 21-27, 21, https://
perma.cc/ZN23-RJ2B.
35	 Data from the National Longitudinal Survey of Youth 1997 for
the period 1997–2008 showed that, by age 23, approximately
49 percent of black men and 38 percent of white men had been
arrested at least once. Robert Brame, Michael G. Turner, Raymond
Paternoster, and Shawn D. Bushway, “Demographic Patterns
of Cumulative Arrest Prevalence by Ages 18 and 23,” Crime &
Delinquency 60, no. 3 (2014), 471-486, https://perma.cc/E49B-SGHY.

Gatekeepers: The Role of Police in Ending Mass Incarceration

57

36	 See Vera Institute of Justice, “Arrest Trends: Demographics,” https://
arresttrends.vera.org/demographics.
37	 “Metropolitan areas” are counties that contain a city with at
least 50,000 inhabitants as well as adjacent counties that have
“a high degree of economic and social integration” with that city
as measured through commuting. Cities outside of metropolitan
areas are incorporated cities that do not meet the population or
integration requirements for inclusion in a metropolitan area. FBI,
“2017 Crime in the United States” (area definitions), https://perma.
cc/BBH9-EYTP.
38	 Neusteter and O’Toole, Every Three Seconds, 2019, 8.
39	 FBI, “2017 Crime in the United States,” table 31: Number and Rate of
Arrests, by Population Group, 2017, https://perma.cc/8Y3L-L32F.
40	 FBI, “2017 Crime in the United States” (larceny-theft), https://
perma.cc/E763-LCGH.
41	 FBI, “2017 Crime in the United States,” table 29: Estimated Number
of Arrests, 2017, https://perma.cc/Y6JB-K34L.
42	 The UCR program reports 12,208 arrests for murder and nonnegligent manslaughter, 23,436 for rape, 94,046 for robbery, and
388,927 for aggravated assault in 2017, out of a total of 10,554,985
arrests. FBI, “2017 Crime in the United States,” table 29.
43	 FBI, “2016 Crime in the United States,” table 18; FBI, “2017 Crime in
the United States,” table 29; FBI, “2010 Crime in the United States,”
table 29; and Snyder, Arrest 1990-2010, 2012, 2.
44	 FBI, “2017 Crime in the United States,” table 29.
45	 An estimated 580,900 drug arrests occurred in 1980, and 1,572,572
occurred in 2016. Estimated drug arrest rates also showed an
increase of 88 percent over the same period, jumping from 254 per
100,000 in 1980 to 478 per 100,000 in 2016. Vera’s Arrest Trends tool
features data on drug arrests generally, but this category is not
broken out by specific drug offense type because of inconsistencies
in that data’s availability; however, the FBI provides publicly
available, national-level aggregate data on drug arrest breakouts.
Although 2016 FBI drug arrest breakdown tables do not exist, 2017
data indicates that 85 percent of drug arrests were made for
possession (not sales), with the leading possession drug type being
marijuana, making up 37 percent of all drug arrests. FBI, “2017
Crime in the United States” (arrests for drug abuse violations),
https://perma.cc/R5YP-YMVD.
46	 Preeti Chauhan, Shannon Tomascak, Celina Cuevas et al., Trends
in Arrests for Misdemeanor Charges in New York City, 1993-2016
(New York: John Jay College of Criminal Justice Misdemeanor
Project, 2018), 16, https://perma.cc/5XL3-6EAA. For a description of
summonses and their function in New York, see New York County
Lawyers’ Association (NYCLA), New York City Criminal Courts
Manual (New York: NYCLA, 2011), 96.
47	 New York City Criminal Justice Agency, “Our Mission,” https://
perma.cc/P4FZ-CZWQ.

58

Vera Institute of Justice

48	 Russell Ferri, Stephen Koppel, Lauren Wilson, and Raymond
Caligiure, New York City Criminal Justice Agency Annual Report
2016 (New York: New York Criminal Justice Agency, 2018), 11,
https://perma.cc/KVZ8-ZVUK.
49	 Nancy Fishman, Kaitlin Kall, Rebecca Silber et al., Report to the
Greater Oklahoma City Chamber Criminal Justice Task Force (New
York: Vera Institute of Justice, 2016), 14, https://perma.cc/CJ6UMLGD.
50	 New York State Division of Criminal Justice Services (NYDCJS),
DCJS Coded Law File as of 01/30/2019 (New York: NYDCJS, 2019),
https://perma.cc/AR8W-EY7E. In Philadelphia, fingerprinting is
required on arrest for all felonies and misdemeanors, but only
for some “summary offenses”: minor traffic violations such as
speeding or going through a red light as well as nontraffic offenses
such as disorderly conduct, underage drinking, harassment, and
shoplifting. 18 Pa. Cons. Stat. § 9112, https://perma.cc/8E9S-BBTP;
and Pennsylvania Bar Association, Traffic Violations and Summary
Offenses (Harrisburg, PA: Pennsylvania Bar Association, 2013),
https://perma.cc/N37Y-Y43D.
51	 National Conference of State Legislators (NCSL), “Citation in Lieu of
Arrest,” https://perma.cc/SJ3C-3YG3.
52	IACP, Citation in Lieu of Arrest, 2016, 7.
53	 Citations are most commonly applied to misdemeanors and other
low-level violations, but some states permit citations to be issued
for low-level felonies. According to the NCSL, four states—Alaska,
Louisiana, Minnesota, and Oregon—permit citations to be issued
for some felonies. NCSL, “Citation in Lieu of Arrest.”
54	IACP, Citation in Lieu of Arrest, 2016, 17.
55	 Ibid., 3.
56	 Criminal Justice Leadership Alliance, “Use of Summonses versus
Custodial Arrest for Municipal Offenses,” December 8, 2010; and
Criminal Justice Leadership Alliance, “Use of Summonses versus
Custodial Arrest for Municipal Offenses,” July 14, 2011, unpublished
reports provided to Vera in its role as a member of the alliance.
57	Ibid.
58	 For an explanation of how tickets and fines can lead to jail
incarceration, see Jessica Brand, “How Fines and Fees Criminalize
Poverty, Explained,” The Appeal, July 16, 2018, https://perma.
cc/7PFA-7DRN. For a discussion of the specific problems and
consequences associated with issuing citations to people who are
homeless, see Editorial Board, “Everyone Knows We Can’t Arrest
Our Way out of Homelessness. So Why Is L.A. Still Trying?” Los
Angeles Times, February 16, 2018, https://perma.cc/ULD3-NWJY.
59	 FBI, “2016 Crime in the United States,” table 18.
60	 Rebecca Neusteter and Megan O'Toole, Every Three Seconds:
Unlocking Police Data on Arrests (New York: Vera Institute of
Justice, 2019), 9, https://perma.cc/AQQ6-7GFY. Vera researchers

calculated disparities using 2014 (the most recent year available)
UCR estimated arrest volumes and United States Census Bureau
population data. For statistics on prevalence of drug use among
black and white people, see Center for Behavioral Health Statistics
and Quality, Results from the 2017 National Survey on Drug Use and
Health: Detailed Tables (Rockville, MD: Substance Abuse and Mental
Health Services Administration, 2018), Table 1.29A.
61	 Ezekiel Edwards, Will Bunting, and Lynda Garcia, The War on
Marijuana in Black and White: Billions of Dollars Wasted on Racially
Biased Arrests (Washington, DC: American Civil Liberties Union,
2013), 9, 21-22 & 47, https://perma.cc/G4QP-9K9P.
62	 Harry Levine, “Marijuana Arrests: Gateway into the Criminal
Justice System,” Brooklyn Rail, October 3, 2011, https://perma.
cc/C3LN-U7NZ. Research has demonstrated that race can affect
arrest decisions (see for example Tammy Rinehart Kochel, David
B. Wilson, and Stephen D. Mastrofski, “Effect of Suspect Race on
Officers’ Arrest Decisions,” Criminology 49, no. 2 (2011), 473-512);
vehicle and person searches (see for example Robin S. Engel and
Richard Johnson, “Toward a Better Understanding of Racial and
Ethnic Disparities in Search and Seizure Rates,” Journal of Criminal
Justice 34, no. 6 (2006), 605-617); traffic stops (see for example
Patricia Warren, Donald Tomaskovic-Devey, William Smith et al.,
“Driving While Black: Bias Processes and Racial Disparity in Police
Stops,” Criminology, 44, no. 3 (2006), 709-738); pedestrian stops
(see for example Andrew Gelman, Jeffrey Fagan, and Alex Kiss,
“An Analysis of the New York City Police Department’s ‘Stop-andFrisk’ Policy in the Context of Claims of Racial Bias,” Journal of the
American Statistical Association 102, no. 479 (2007), 813-823); and
drug arrests (see for example Katherine Beckett, Kris Nyrop, and Lori
Pfingst, “Race, Drugs, and Policing: Understanding Disparities in
Drug Delivery Arrests,” Criminology 44, no. 1 (2006), 105-137).
63	 Jessica Eaglin and Danyelle Solomon, Reducing Racial and Ethnic
Disparities in Jails: Recommendations for Local Practice (New York:
Brennan Center for Justice, 2015), 7-8, https://perma.cc/8NEVQNNF.
64	 Mary T. Phillips, Misdemeanor Marijuana Arrests New York City
2012–2014 (New York: New York City Criminal Justice Agency, 2015),
33 & 38, https://perma.cc/R2SR-NFF2.
65	 Ibid., 35.
66	Ibid.
67	 Ibid., 34 & 39.
68	 Anne Milgram, Jeffrey Brenner, Dawn Wiest et al., Integrated
Health Care and Criminal Justice Data—Viewing the Intersection
of Public Safety, Public Health and Public Policy Through a
New Lens: Lessons from Camden, New Jersey (Cambridge, MA:
Harvard Kennedy School, Program in Criminal Justice Policy and
Management, 2018), 5, https://perma.cc/GXD6-UHPB
69	 Ibid., 6.

70	 David E. Olson and Koert Huddle, “An Examination of Admissions,
Discharges & the Population of the Cook County Jail, 2012,” Social
Justice (March 2013), 6, https://perma.cc/D6K3-RADN.
71	 Mayor’s Task Force on Behavioral Health and the Criminal Justice
System, Action Plan 2014 (New York: Office of Mayor Bill de Blasio,
2014), 1-2, http://criminaljustice.cityofnewyork.us/wp-content/
uploads/2018/04/BHTF-First-Status-Report-8.15.pdf.
72	 Fishman, Kall, Silber et al., Greater Oklahoma City Chamber
Criminal Justice Task Force, 2016, 19.
73	 Ibid., 33.
74	 Marilyn Chandler Ford, “Frequent Fliers: The High Demand User in
Local Corrections,” Californian Journal of Health Promotion 3, no. 2
(2005), 61-71, 62, https://perma.cc/NZL9-YAUN.
75	 For example, in a year’s worth of dispatched police calls for
service for the Eastern District of Baltimore, Maryland, only about
26 percent were related to a crime requiring a written report; a
further 35 percent were for police services not involving a crime,
and the remainder were coded as not requiring a police response
(for example, when officers could not locate the caller or identify an
emergency). Peter C. Moskos, “911 and the Failure of Police Rapid
Response,” Law Enforcement Executive Forum 7, no. 4 (2007), 137149, 140, https://perma.cc/CDU5-TZEG.
76	 Christopher Hawk, “Do Police Officers Have Too Much or Too Little
Discretion?” PoliceOne, February 19, 2014, https://perma.cc/MM3K3Q7Y.
77	 William Terrill and Eugene A. Paoline III, “Nonarrest Decision Making
in Police–Citizen Encounters.” Police Quarterly 10, no. 3 (2007), 308331, 319, https://perma.cc/M6SM-JK9Y.
78	 Joseph A. Schafer and Stephen D. Mastrofski, “Police Leniency
in Traffic Enforcement Encounters: Exploratory Findings from
Observations and Interviews,” Journal of Criminal Justice 33, no. 3
(2005), 225-238, 235.
79	 FBI, “2016 Crime in the United States,” table 18.
80	 Rob Voigt, Nicholas P. Camp, Vinodkumar Prabhakaran et al.,
“Language from Police Body Camera Footage Shows Racial
Disparities in Officer Respect,” Proceedings of the National
Academy of Sciences 114, no. 25 (2017), 6521-6526, https://perma.
cc/GUP3-7BF4. However, the mere act of deploying body-worn
cameras may affect officer interactions with the community,
decreasing the likelihood of negative interaction. See for example
Michael D. White, Janne E. Gaub, and Natalie Todak, “Exploring
the Potential for Body-Worn Cameras to Reduce Violence in
Police-Citizen Encounters,” Policing 12, no. 1 (2017), 66-76, 66
(“Following BWC deployment, the percentage of officers with a
complaint in each group declined by 50% and 78% (Control and
Treatment, respectively); the percentage of officers with a use of
force declined notably (39%) for one group only”), https://perma.
cc/253H-NU82.

Gatekeepers: The Role of Police in Ending Mass Incarceration

59

81	 Voigt, Camp, Prabhakaran et al., “Racial Disparities in Officer
Respect,” 2017, 6521.
82	Ibid.
83	 Jon M. Shane, “Key Administrative and Operational Differences in
the Police Quasi Military Model,” Law Enforcement Executive Forum
10, no. 2 (2010), 75-106, https://perma.cc/G5VM-6DBN.
84	 Ibid.; and Thomas J. Cowper, “The Myth of the “Military Model” of
Leadership in Law Enforcement,” Police Quarterly 3, no. 3 (2000),
228-246, https://perma.cc/6X22-JJQK.
85	 Shane, “The Police Quasi Military Model,” 2010, 77-79; and Cowper,
“Myth of the Military Model,” 2000, 238.
86	 Malcolm K. Sparrow, Measuring Performance in a Modern Police
Organization (Washington, DC: DOJ, National Institute of Justice,
2015), 18-20, https://perma.cc/5BJV-Q2U3.
87	 Floyd v. City of New York, 959 F. Supp. 2d 540, 599 (S.D.N.Y. 2013);
and Al Baker, “Police Evaluations Should Focus on Lawfulness
of Stops, Monitor Says,” New York Times, October 20, 2017,
https://www.nytimes.com/2017/10/20/nyregion/new-york-policeevaluations-stops-monitor.html. Supervisors also face pressure to
meet specific metrics. For example, failure to lower crime rates in
a precinct can lead to public shaming, reassignment, or demotion
for these supervisors. John F. Timoney, Beat Cop to Top Cop: A Tale
of Three Cities (Philadelphia, PA: University of Pennsylvania Press,
2010), 168-69.
88	 Saki Knafo, “How Aggressive Policing Affects Police Officers
Themselves,” Atlantic, July 13, 2015 (officers who fail to fulfill
their arrest quotas may be denied vacation requests, assigned to
undesirable shifts, or limited in organizational growth), https://
perma.cc/5BUR-4YAP ; Aegis Staff, “Two Bel Air Police Officers Are
Promoted, Honored by Town Officials,” Baltimore Sun, August 21,
2018, (attributing two officers’ promotions to the number of arrests
and citations they issued throughout their career), https://perma.
cc/SX6Y-648Y; and Shaun Ossei-Owusu, “Race and the Tragedy
of Quota-Based Policing,” American Prospect, November 3, 2016
(noting that even in the absence of a formal quota policy officers
may still be informally required to reach a minimum number of
arrests to move forward in their careers), https://perma.cc/UHN27DMG .
89	 David Weisburd, Stephen D. Mastrofski, Rosann Greenspan, and
James J. Willis, The Growth of Compstat in American Policing
(Washington, DC: Police Foundation, 2004), 1-3, https://perma.cc/
F48E-FD38.
90	Ibid.
91	 Paul E. O’Connell and Frank Straub, Performance-Based
Management for Police Organizations (Long Grove, IL: Waveland
Press, Inc., 2007), 81-82.
92	 See Weisburd, Mastrofski, Greenspan, and Willis, The Growth of
CompStat, 2004, 12 & figure 6.

60

Vera Institute of Justice

93	 Examining the rise of the New York City Police Department's “stop,
question, and frisk” policy concurrently with the implementation
of CompStat reveals a pattern of de facto racial profiling. Peter
Hanink, “Don’t Trust the Police: Stop Question Frisk, CompStat, and
the High Cost of Statistical Over-reliance in the NYPD,” Journal of
the Institute of Justice & International Studies 13 (2013), 99-114, 100.
In fact, in just the first two years after CompStat was implemented,
communities of color faced additional overpolicing issues.
John A. Eterno and Eli B. Silverman, The Crime Numbers Game:
Management by Manipulation (Boca Raton, FL: CRC Press, 2012),
223-24.
94	 Preeti Chauhan, Todd Warner, Adam Fera et al., Tracking
Enforcement Rates in New York City, 2003–2014 (New York: John
Jay College of Criminal Justice, 2015), 11, https://perma.cc/X8CRUR74.
95	 New York State Office of the Attorney General, New York City
Police Department’s “Stop & Frisk” Practices: A Report to the People
of the State of New York from the Office of the Attorney General
(Albany, NY: NY Office of the Attorney General, 1999), 81, https://
perma.cc/MSN3-BA25. See also John A. Eterno, “Policing by the
Numbers,” New York Times, June 17, 2012, https://www.nytimes.
com/2012/06/18/opinion/the-nypds-obsession-with-numbers.html.
96	 Chauhan, Warner, Fera et al., Enforcement Rates 2003-2014, 2015,
11.
97	 Ibid., 10.
98	 For the number of stops, see ibid., 11. Perhaps the best known
lawsuit concerning NYPD’s “stop question frisk” policies is Floyd v.
City of New York, 959 F. Supp. 2d 540, 599 (S.D.N.Y. 2013), which
held that the NYPD had violated plaintiffs’ Fourth Amendment rights
as part of a pervasive pattern of racially biased unconstitutional
searches. For a discussion of Floyd’s effect, see John MacDonald
and Anthony A. Braga, “Did Post-Floyd et al. Reforms Reduce
Racial Disparities in NYPD Stop, Question, and Frisk Practices?
An Exploratory Analysis Using External and Internal Benchmarks,”
Justice Quarterly (2018), 1-30, 5. However, even with the reforms
implemented after Floyd, concerns remain that reliance on
CompStat as a departmental metric leads officers to effectively
manipulate data by misclassifying incidents, according to recently
filed suits. For one such suit, see Lopez-Lopez v. City of New York,
No. 1:2019cv02188 (E.D.N.Y., filed April 15, 2019), https://perma.cc/
HFK6-9J6X.
99	 Seth Stoughton, who teaches law and studies police issues at the
University of South Carolina, described police training for Michel
Martin on “All Things Considered” in 2016. See “Looking at How
Police Are Trained,” NPR, July 9, 2016, https://perma.cc/ZDP8-4XL8.
100	 LawEnforcementEDU.net, “Police Officer Academy Training,”
https://perma.cc/E9EP-SP6P.
101	 For example, the Commission on Accreditation for Law Enforcement
Agencies, Inc. (CALEA), which provides accreditation programs for
public safety agencies, lists 459 standards for law enforcement;

fewer than a dozen are focused on discretion, diversion, and
alternatives to traditional enforcement. CALEA, “Law Enforcement Standards Titles,” https://perma.cc/GS7S-EZCW.
102	 For example, officers in Miami receive training in how to work
with people suffering from mental health issues, which has led
to a significant reduction in arrests and the ultimate closure of
one of Miami-Dade County’s five jails. Law Enforcement Leaders
to Reduce Crime & Incarceration, “Increasing Alternatives to
Arrest and Prosecution, Especially Mental Health and Drug
Treatment,” https://perma.cc/B3VP-GPP6. Washington State has
also adopted a new training style that emphasizes the role of
officers as guardians and de-emphasizes their role as warriors.
Ashley Balcerzak and Julie Tate, “Creating Guardians, Calming
Warriors: A New Style of Training for Police Recruits Emphasizes
Techniques to Better De-Escalate Conflict Situations,” Washington
Post, December 10, 2015, https://www.washingtonpost.com/sf/
investigative/2015/12/10/new-style-of-police-training-aims-toproduce-guardians-not-warriors/?noredirect=on&utm_term=.
a9498a913d5a.
103	 The Houston Police Department (HPD), for example, saw calls for
service for mental health issues more than double between 2007
and 2014, from 15,000 to 37,000. In response, HPD implemented
a crisis call diversion program to direct these calls to more
appropriate responders. HPD Mental Health Division, “Crisis Call
Diversion Program,” October 18, 2017, https://perma.cc/BJJ6-FM2E.
For reviews of police department capacity to handle mental health
issues and emerging alternatives, see Martha Deane, Henry J.
Steadman, Randy Borum et al., “Emerging Partnerships Between
Mental Health and Law Enforcement,” Psychiatric Services 50,
no. 1 (1999), 99-101, https://perma.cc/V5QP-WFFP; and Henry
J. Steadman, Martha W. Deane, Randy Borum, and Joseph P.
Morrissey, “Comparing Outcomes of Major Models of Police
Responses to Mental Health Emergencies,” Psychiatric Services 51,
no. 5 (2000), 645-649, https://perma.cc/86QW-8GXB.
104	 Rural Health Information Hub, “Rural Mental Health,” https://
perma.cc/X4UU-MRMC.
105	Ibid.
106	 For example, both Memphis and San Diego County use specially
trained officers and teams to establish links to community-based
resources and crisis teams; in Baltimore County, social workers ride
with officers on mental health responses. Melissa Reuland, A Guide
to Implementing Police-based Diversion Programs for People with
Mental Illness (Rockville, MD: U.S. Department of Health and Human
Services, Substance Abuse and Mental Health Administration,
2004), 12-14, https://perma.cc/P7SD-WGZY.
107	Reuland, Police-based Diversion Programs, 2004.
108	 Elina Treyger, “Collateral Incentives to Arrest,” University of Kansas
Law Review 63 (2015), 557-631, 558-59, https://perma.cc/C2HG26VY.

109	 Rachel A. Harmon, “Why Arrest,” Michigan Law Review 115, no. 3
(2016), 307-364, 336-37, https://perma.cc/L9GT-YML3.
110	 In 2015, retired Major Neil Franklin spoke about the “significant
impact” that financial incentives have on policing styles. See
German Lopez, “The Twisted Financial Incentives behind The War
on Drugs,” Vox, April 14, 2015, https://perma.cc/AG3H-6FXH. In 2014,
Jeff Smith, a former Missouri senator, described how reliance on
revenue from traffic stops helped shape the financial incentives
that drove racial disparity in law enforcement in Ferguson,
Missouri—and eventually led to Michael Brown’s death. Jeff Smith,
“In Ferguson, Black Town, White Power,” New York Times, August 17,
2014, https://www.nytimes.com/2014/08/18/opinion/in-fergusonblack-town-white-power.html.
111	 Michael D. Makowsky, Thomas Stratmann, and Alex Tabarrok,
“To Serve and Collect: The Fiscal and Racial Determinants of
Law Enforcement,” George Mason University Working Paper in
Economics No. 16-17 (2018), https://perma.cc/BX2D-TXWD.
112	 Alan Feuer and Joseph Goldstein, “The Arrest Was a Bust. The
Officers Got Overtime Anyway,” New York Times, February 18, 2018,
https://www.nytimes.com/2018/02/19/nyregion/new-york-policeovertime-pay-trial.html.
113	 Ibid.; Rocco Parascandola, “Exclusive: NYPD Internal Affairs Chief
Sends Scathing Memo Warning Cops Against Making Arrests Just to
Earn Overtime,” New York Daily News, April 15, 2014, https://perma.
cc/USQ8-5YYG; and Kevin Sheehan, “NYPD Cop Accused of False
Arrest Admits He Faked Overtime Reports,” New York Post, February
20, 2018, https://perma.cc/SV3B-XT2Y.
114	 Robynn Cox and Jamein P. Cunningham, “Financing the War on
Drugs: The Impact of Law Enforcement Grants on Racial Disparities
in Drug Arrests,” CESR-Schaeffer Working Paper Series, Paper no.
2017-005 (2017), https://perma.cc/8U8X-B54Z.
115	 Ibid., 3.
116	 Alexandra Rogin, “Dollars for Collars: Civil Asset Forfeiture and the
Breakdown of Constitutional Rights,” Drexel Law Review 7 (2014),
45-82, https://perma.cc/J9MX-XXS5; and Dick Carpenter, Lisa
Knepper, Angela Erickson, and Jennifer McDonald, Policing for
Profit: The Abuse of Civil Asset Forfeiture (Arlington, VA: Institute
for Justice, 2015), https://perma.cc/XB22-CG5A. The future of civil
forfeiture law is uncertain following the Supreme Court’s decision
in Timbs v. Indiana. The Court explicitly applied the Excessive
Fines Clause of the Eighth Amendment to the states. What is an
“excessive” fine will continue to be a question of fact, but the
Court held that seizure of an asset worth four times more than the
maximum fine assessable for the underlying crime could fall within
the description. Timbs v. Indiana, ___ U.S. ___ (2018) (slip opinion),
https://perma.cc/A3RS-A2VC.
117	 Chip Mellor, “Civil Forfeiture Laws and the Continued Assault on
Private Property,” Forbes, June 8, 2011, https://perma.cc/NLA3C57U.

Gatekeepers: The Role of Police in Ending Mass Incarceration

61

118	 Ibid.; and Carpenter, Knepper, Erickson, and McDonald, Policing for
Profit, 2015.
119	 Mellor, “Assault on Private Property,” 2011; and Comprehensive
Crime Control Act of 1984, Pub. Law No. 98-473 (1984).
120	 Note, “How Crime Pays: The Unconstitutionality of Modern Civil
Asset Forfeiture as a Tool of Criminal Law Enforcement,” Harvard
Law Review 131 (2019), 2387-2408, 2387, https://perma.cc/26WUH7VD. For direct examples of forfeiture, see Edgar Walters and Jolie
McCullough, “Texas Police Made More Than $50 Million in 2017 from
Seizing People’s Property. Not Everyone Was Guilty of a Crime,”
Texas Tribune, December 7, 2018, https://perma.cc/AQ9B-4W6G.
121	 “How Crime Pays,” 2019. For a discussion of how civil asset
forfeiture unevenly affects black people, see Vanita Saleema Snow,
“From the Dark Tower: Unbridled Civil Asset Forfeiture,” Drexel Law
Review 10, no. 1 (2017), 69-125, https://perma.cc/X687-ZSP5. For an
investigation of how black people who committed low-level offenses
were unevenly targeted in one city, see DOJ Civil Rights Division,
Investigation of the Ferguson Police Department, 2015, 9-15 (fines
and fees), 62-78 (racial bias in enforcement.)
122	 Carpenter, Knepper, Erickson, and McDonald, Policing for Profit,
2015.
123	 Robert O’Harrow Jr., Steven Rich, and Shelly Tan, “Asset Seizures
Fuel Police Spending,” Washington Post, October 11, 2014,
https://www.washingtonpost.com/sf/investigative/2014/10/11/
asset-seizures-fuel-police-spending/?noredirect=on&utm_
term=.623f6fdf6698.
124	 American Civil Liberties Union (ACLU), “Police Militarization,”
ACLU.org, https://perma.cc/UU4R-LE8Y. For a deeper look into the
trend of militarization see ACLU, War Comes Home: The Excessive
Militarization of American Policing (New York: ACLU, 2014), https://
perma.cc/BEW3-XP65.
125	 O’Harrow, Rich, and Tan, “Asset Seizures,” 2014. In addition to police
departments, other justice system entities can be beneficiaries of
asset forfeiture and may have different spending priorities. For
example, a 2018 report on New York City district attorneys found
that of the $730 million in forfeited assets, 0.03 percent had been
spent on community-based programs (such as drug-treatment
facilities, job skills training, or programs designed to deter youth
from drugs and crime) and 5.8 percent on “drug, gang, and
other education awareness programs” via 11 community-based
organizations through a competitive bidding process. New York
City Independent Budget Office (NYCIBO), Nest Egg: City’s
District Attorneys Holding Millions in Off-Budget Funds From Asset
Forfeitures (New York: NYCIBO, 2018), https://perma.cc/D8W9P8KH; appendices at https://perma.cc/SR4Y-QF2V.
126	 Harvard Law School Criminal Justice Policy Program, Confronting
Criminal Justice Debt: A Guide for Policy Reform (Cambridge, MA:
Harvard Law School, 2014), 1, https://perma.cc/7DZ8-LP5Z.

62

Vera Institute of Justice

127	 Dan Kopf, “The Fining of Black America,” Priceonomics, June 24,
2016, https://perma.cc/E6NG-YR7M. See also Torie Atkinson, “A
Fine Scheme: How Municipal Fines Become Crushing Debt in the
Shadow of the New Debtors’ Prison,” Harvard Civil Rights-Civil
Liberties Law Review 51, no. 1 (2016), 189-238, https://perma.cc/
BQ66-2MLK; and Mike Maciag, “Skyrocketing Court Fines Are Major
Revenue Generator for Ferguson,” Governing, August 22, 2014,
https://perma.cc/R93P-E6LL.
128	 Joseph Shapiro, “In Ferguson, Court Fines and Fees Fuel Anger,”
NPR, August 25, 2014, https://perma.cc/DV67-7YTB; and DOJ Civil
Rights Division, Investigation of Ferguson, 2015, 9-15.
129	 Kopf, “The Fining of Black America,” 2016.
130	 Brand, “Fees Criminalize Poverty,” 2018.
131	Ibid.
132	 Mathilde Laisne, Jon Wool, and Christian Henrichson, Past Due:
Examining the Costs and Consequences of Charging for Justice
in New Orleans (New York: Vera Institute of Justice, 2017), https://
perma.cc/XTZ9-HYCB.
133	 Harold Pepinsky, “Better Living through Police Discretion,” Law and
Contemporary Problems, 47, no. 4 (1984), 249-267, 266, https://
perma.cc/37PA-JCBC.
134	 On the increased risk of criminal justice involvement for people
with serious mental illnesses, see Jacques Baillargeon, Ingrid A.
Binswanger, Joseph V. Penn et al., “Psychiatric Disorders and
Repeat Incarcerations: The Revolving Prison Door,” American
Journal of Psychiatry 166, no. 1 (2009), 103-109, https://perma.
cc/2GSU-NTLA; Anne G. Crocker, Kathleen Hartford, and Lisa
Heslop, “Gender Differences in Police Encounters among Persons
with and without Serious Mental Illness,” Psychiatric Services 60,
no. 1 (2009), 86-93, https://perma.cc/Z2NR-3KCW; James D.
Livingston, “Contact Between Police and People with Mental
Disorders: A Review of Rates,” Psychiatric Services 67, no. 8 (2016),
850-857, https://perma.cc/F6C8-6HU6; and Seth J. Prins, “Does
Transinstitutionalization Explain the Overrepresentation of People
With Serious Mental Illnesses in the Criminal Justice System?”
Community Mental Health Journal 47, no. 6 (2011), 716-722, https://
perma.cc/QE2Z-WQYP.
135	 For Johnson County, Kansas, see National Association of Counties
(NACo), Mental Health and Criminal Justice—Case Study: Johnson
County, Kan. (Washington, DC: NACo, 2014), https://perma.
cc/96NN-QR7E. For Salt Lake County, Utah, see Salt Lake City
Police Department, “Community Connection Center,” https://
perma.cc/6WPY-NHWX. For Miami-Dade County, Florida, see
Daniel Chang, “Criminal Mental Health Program in Miami-Dade
Seen as a Model for Nation,” Miami Herald, May 21, 2016, https://
www.miamiherald.com/news/health-care/article79004057.html.
For San Antonio, Texas, see Center for Health Care Services,
“Restoration Center: San Antonio’s Answer to Mental Health,”
Center for Health Care Services, January 30, 2017, https://perma.
cc/54D5-PKVX; Scott Helman, “The San Antonio Way: How One

Texas City Took On Mental Health as a Community—and Became a
National Model,” Boston Globe, December 10, 2016, https://perma.
cc/X34J-9MWT; Steven R. Pliska, Bruce Adams, Sally E. Taylor et
al., “Mental Health Crisis in Bexar County a New Normal?,” My San
Antonio, August 13, 2016, https://perma.cc/U9C7-E2QL; and Haven
for Hope, “What is Haven for Hope?,” https://perma.cc/N73GQVYX.

149	 National Emergency Number Association, “9-1-1 Statistics,” https://
perma.cc/6786-WBPF.

136	 Houston Recovery Center, “Home,” https://perma.cc/P37T-5DKM.

150	 A team of analysts found that 56 percent to 63 percent of 911 calls
generated a computer-aided dispatch response, and 83 percent of
those required police response rather than fire or medical. Diara
Dankert, James Driscoll, and Nancy Torres, San Francisco’s 911
Call Volume Increase (Mountainview, CA: Google, 2015), 9, https://
perma.cc/QHU3-C4AH.

137	 Henry J. Steadman, Kristin A. Stainbrook, Patricia Griffin et al., “A
Specialized Crisis Response Site as a Core Element of Police-Based
Diversion Programs,” Psychiatric Services 52, no. 2 (2001), 219-222,
https://perma.cc/8792-CJG7.

151	 For example, the Houston Police Department responded to 37,032
calls for mental health issues in 2014, more than twice as many
as in 2007. HPD, Mental Health Division, “Crisis Call Diversion
Program,” 2017.

138	 Sheryl Kornman, “Pima County’s New Crisis Center: 11,000 Calls for
Help Each Month,” TusconSentinel.com, October 11, 2012, https://
perma.cc/N2V2-Q4DF.

152	 See for example DOJ Community Relations Services, Importance of
Police-Community Relationships and Resources for Further Reading
(Washington, DC: DOJ, 2015), https://perma.cc/F7FZ-SYK2.

139	 Jennifer L. Skeem, Sarah Manchak, and Jillian K. Peterson,
“Correctional Policy for Offenders with Mental Illness: Creating a
New Paradigm for Recidivism Reduction,” Law and Human Behavior
35, no. 2 (2011), 110-126, 113 & table 1, https://perma.cc/V8VC-GX77.

153	 Policy Research Associates, The Sequential Intercept Model:
Advancing Community Based Solutions for Justice-Involved People
with Mental And Substance Use Disorders (Delmar, NY: Policy
Research Associates), https://perma.cc/KHZ2-WM69; and Reuland,
Police-Based Diversion Programs, 2004, 13-15.

140	Ibid.
141	 Milgram, Brenner, Wiest et al., Lessons from Camden, New Jersey,
2018.
142	 Marilyn Sinkewicz, Yu-Fen Chiu, and Leah Pope, Sharing
Behavioral Health Information Across Justice and Health Systems:
Opportunities in the District of Columbia (New York: Vera Institute of
Justice, 2018), 16, https://perma.cc/9GEZ-66ZP.
143	 Ibid., 38.
144	 See Brandon Stahl, “Officer’s Lawyer: Castile Pulled Over Because
He Matched Robbery Suspect,” StarTribune, July 11, 2016, https://
perma.cc/8LVM-RCBS; Andy Mannix, “Police Audio: Officer Stopped
Philando Castile On Robbery Suspicion,” StarTribune, July 12,
2016, https://perma.cc/YWG6-FT7C; “Lights On! A Program of
MicroGrants,” https://perma.cc/KN8R-FH6Z; and Dan Simmons,
“Two Years After Philando Castile’s Death, Program Aims to
Transform Relations Between Police, Residents,” Washington Post,
July 7, 2018, https://www.washingtonpost.com/national/two-yearsafter-philando-castiles-death-programs-aim-to-transform-relationsbetween-police-residents/2018/07/06/c65a3a82-7e28-11e8-b0effffcabeff946_story.html?noredirect=on&utm_term=.8de5503ed815.
145	 See Simmons, “Two Years After Philando Castile’s Death,” 2018.
146	Ibid.
147	 The Audre Lorde Project, “Safe Neighborhood Campaign,” https://
perma.cc/4M2J-ESPX .
148	 The Audre Lorde Project, “Safe OUTside the System (SOS),” https://
perma.cc/22TF-39YU.

154	 For more on the challenges and promises of mining 911 police calls
for service data, see S. Rebecca Neusteter, Maris Mapolski, Mawia
Khogali, and Megan O’Toole, The 911 Call Processing System: A
Review of the Literature as it Relates to Policing (New York: Vera
Institute of Justice, 2019), www.vera.org/911-call-processing-review.
155	Authors’ unpublished analyses of Tucson and New York calls for
service data for FY2015 and calendar year 2015, respectively, on
file with Vera.
156	 Seattle Police Department, “Calls for Service Dashboard,”
database (Seattle, WA: Seattle Police Department) (search: 2017
Calls by Call Group), https://www.seattle.gov/police/informationand-data/calls-for-service-dashboard.
157	Ibid.
158	U.S. Department of Transportation, National Highway Traffic
Safety Administration (NHTSA), Traffic Safety Facts—2016 Data
(Washington, DC: NHTSA, 2018), 2 & table 1, https://perma.cc/
C5DV-NHLE.
159	 See for example Elizabeth Dias, John Eligon, and Richard A. Oppel,
“Philadelphia Starbucks Arrests, Outrageous to Some, Are Everyday
Life for Others,” New York Times, April 17, 2018, https://www.nytimes.
com/2018/04/17/us/starbucks-arrest-philadelphia.html; Associated
Press, “Black Men Arrested at Philadelphia Starbucks Feared for
Their Lives,” Guardian, April 19, 2018, https://perma.cc/D6S8-XQ5F.
160	Reuland, Police-Based Diversion Programs, 2004, 13.
161	 One example of this principle being deployed in a medical context
comes from Washington, DC, where triage nurses sit with 911
dispatchers to set up routine medical appointments for minor issues

Gatekeepers: The Role of Police in Ending Mass Incarceration

63

like sprains or coughs. Selena Simmons-Duffin, “Can Triage Nurses
Help Prevent 911 Overload?” NPR, April 19, 2018, https://perma.cc/
A3YB-K9YW.
162	 Personal communications between the authors and Tucson Police
Department, September 7, 2018, on file with Vera.
163	 For a discussion of the nature and need for comprehensive crisis
services beyond police and hospitals, see Technical Assistance
Collaborative, A Community-Based Comprehensive Psychiatric
Crisis Response Service: An Informational and Instructional
Monograph (Boston, MA: Technical Assistance Collaborative,
2005). For evaluations of the effectiveness of crisis hotlines
and crisis intervention resources, see John Kalafat, Madelyn
S. Gould, Jimmie L. Harris Munfakh, and Marjorie Kleinman,
“An Evaluation of Crisis Hotline Outcomes, Part 1: Non-Suicidal
Crisis Callers,” Suicide and Life-Threatening Behavior 37, no. 3
(2010), 322-337; Madelyn S. Gould, John Kalafat, Jimmie L. Harris
Munfakh, and Marjorie Kleinman, “An Evaluation of Crisis Hotline
Outcomes, Part 2: Suicidal Callers,” Suicide and Life-Threatening
Behavior 37, no. 3 (2010), 338-352; and Shenyang Guo, David E.
Biegel, Jeffrey A. Johnsen, and Hayne Dyches, “Assessing the
Impact of Community-based Mobile Crisis Services on Preventing
Hospitalization,” Psychiatric Services 52, no. 2 (2001) 223-228,
https://perma.cc/UMP5-7TRJ.
164	 Matthew L. Saxton, Charles M. Naumer, and Karen E. Fisher, “2-1-1
Information Services: Outcomes Assessment, Benefit-Cost Analysis,
and Policy Issues,” Government Information Quarterly 24, no. 1
(2007), 186-215; and Nancy Shank, “A Review of the Role of CostBenefit Analyses in 2-1-1 Diffusion,” American Journal of Preventive
Medicine 43, no. 6 (2012), S497-S505.
165	 211.org, "About 211," https://perma.cc/U4D4-4PZU.
166	 Colin Wood, “What is 311?” Government Technology, August 4, 2016,
https://perma.cc/7UZR-Q7UP.
167	 Vera Institute of Justice, “Arrest Trends,” (Durham Police
Department, 1980-2016), https://arresttrends.vera.org/arrests?locat
ion=15701&estimated=0#bar-chart.
168	Ibid.
169	 Durham Police Department, General Orders Manual (Durham, NC:
Durham Police Department, 2018), 60-61, https://perma.cc/TVP2HJS6.

of marijuana paraphernalia. See N.C. Gen. Stat. § 90-113.22(a)
(possession of drug paraphernalia) and § 90-113.22A (possession of
marijuana drug paraphernalia).
173	 For curfew and loitering statistics, see Vera Institute of
Justice, “Arrest Trends,” https://arresttrends.vera.org/
arrests?offense=28#arrest-volume. For liquor violations, see Vera
Institute of Justice, “Arrest Trends,” https://arresttrends.vera.org/arr
ests?offense=22&start=2010&year=2016#bar-chart.
174	 For marijuana use rates by race and ethnicity, see Colorado
Department of Public Health & Environment, Data Brief: Colorado
Marijuana Use 2017 (Denver, CO: Colorado Department of Public
Health and Environment, 2018), 2, https://perma.cc/GWJ2-95CX.
For disparities in arrests, see Colorado Department of Public
Safety, Marijuana Legalization in Colorado: Early Findings (Denver,
CO: Colorado Department of Public Safety, 2016), 8, https://
perma.cc/U4NC-F6AK.
175	 For rates of marijuana use among teens in Colorado, see Colorado
Department of Public Health and Environment, “Colorado Youth
Marijuana Use 2018,” https://perma.cc/R53T-PTDY. For arrest rates,
see Colorado Department of Public Safety, Marijuana Legalization
in Colorado: Early Findings, 2016, 8.
176	 For rates of marijuana use among adults in Colorado, see Colorado
Department of Public Health and Environment, “Marijuana Use
in Colorado,” https://perma.cc/9BBM-UNEP. For arrest rates, see
Colorado Department of Public Safety, Marijuana Legalization in
Colorado: Early Findings, 2016, 5.
177	 Sarah Willets, “In Durham, Fewer Stops and Searches and Pot
Arrests, But Still Racial Disparities,” Indy Week, April 11, 2018,
https://perma.cc/E8XV-PKRX. In 2015, before the policy change, 25
percent of people charged with these marijuana-related offenses
received citations; in 2017, that number rose to 39 percent. Ibid.
178	 Jennifer A. Tallon, Melissa Labriola, and Joseph Spadafore,
Creating Off-Ramps: A National Review of Police-led Diversion
Programs (New York: Center for Court Innovation, 2016), https://
perma.cc/8WZH-L8TM.
179	 See IACP/IACP/University of Cincinnati (UC) Center for Police
Research and Policy, Deconstructing the Power to Arrest: Lessons
from Research (Cincinnati, OH: IACP/UC Center for Police Research
and Policy, 2018), xvii, https://perma.cc/5MEB-6WFK.

170	Ibid.

180	 Marc Krupanski, Police and Harm Reduction (New York: Open
Society Foundations, 2018), 6, 21-23, https://perma.cc/2DKB-AS8K.

171	 Virginia Bridges, “Durham Police to Issue Citations for Misdemeanor
Marijuana Offenses,” News Observer, November 10, 2016, https://
www.newsobserver.com/news/local/community/durham-news/
article114052823.html.

181	 James Gilsinan, “The Numbers Dilemma: The Chimera of Modern
Police Accountability Systems,” Saint Louis University Public Law
Review 32, no. 1 (2013), 93-110, 97.

172	 Durham Police Department, 2017 Misdemeanor Marijuana
Review (Durham, NC: Durham Police Department, 2018), https://
perma.cc/6A83-T5A6. This decline could also be attributed to a
2014 change to the law that lowered penalties for possession

64

Vera Institute of Justice

182	 William J. Bratton and Sean W. Malinowski, “Police Performance
Management in Practice: Taking COMPSTAT to the Next Level,”
Policing 2, no. 3 (2008), 259-265, 263, https://perma.cc/74DYW2CV; and Sparrow, Measuring Performance, 12-14.

183	 CompStat360, “Welcome to CompStat 360,” https://perma.cc/
TH3W-25HC.

were put in place. Jennifer A. Tallon, Melissa Labriola, and Joseph
Spadafore, Creating Off-Ramps: A National Review of Police-led
Diversion Programs (New York: Center for Court Innovation, 2016),
30, https://perma.cc/8RB6-D7V3.

184	 Safety and Justice Challenge, “The Challenge Network,” https://
perma.cc/L5E4-DP9T.
185	Safety and Justice Challenge, “About the Challenge,” https://
perma.cc/798C-YUZR.

8	

Tallon, Labriola, and Spadafore, Creating Off-Ramps, 2016, 15.

9	

Ibid., viii.

186	 Laura and John Arnold Foundation (LJAF), Request for Proposals:
Reimagining America’s Crisis Response Systems (New York: LJAF,
2018), https://perma.cc/EW9Y-P39S.

10	 Ibid.

187	 See Ram Subramanian and Leah Skrzypiec, To Protect and Serve:
New Trends in State-Level Policing Reform, 2015-2016 (New York:
Vera Institute of Justice, 2017), 10, https://perma.cc/9539-XDEV.

12	 Ibid.; and Engel, Worden, Corsaro et al., Deconstructing the Power
of Arrest, 2018.

Spread notes
“Working at the margins: Alternatives-to–arrest
programs” p. 32
1	

2	

3	

Robin Engel, Robert E. Worden, Nicholas Corsaro et al.,
Deconstructing the Power of Arrest: Lessons from Research
(Alexandria, VA: IACP/UC Center for Police Research and
Policy, 2018) 1-2, https://perma.cc/BUY7-D4BK; and Robin S.
Engel, Nicholas Corsaro, and M. Murat Ozer, “The Impact of
Police on Criminal Justice Reform: Evidence from Cincinnati,
Ohio,” Criminology & Public Policy 16, no. 2 (2017), 375-402.
See for example, Police Executive Research Forum, The
Unprecedented Opioid Epidemic: As Overdoses Become a Leading
Cause of Death, Police, Sheriffs, And Health Agencies Must Step Up
Their Response (Washington, DC: Police Executive Research Forum,
2017), 14, 24, & 56, https://perma.cc/XLA5-FD9P.
Micah Kubic and Taylor Pendergrass, “Diversion Programs Are
Cheaper and More Effective Than Incarceration. Prosecutors
Should Embrace Them,” ACLU, December 6, 2017, https://perma.cc/
ANC4-Q2EG.

4	

Ibid.

5	

Ibid.

6	

Young adults, who are in need of specialized services and who
face additional disruption through contact with the justice system,
constitute one population that is a growing focus of diversion
efforts. National League of Cities, Introduction to City Strategies to
Reduce the Use of Jails for Young Adults (Washington, DC: National
League of Cities, 2017), https://perma.cc/J29R-49JJ.

7	

The Center for Court Innovation reviewed police-led diversion
programs and found that some had been in existence since the
1980s and 1990s. Others were more recent, depending on models
that had not been developed when the first diversion programs

11	 Ibid., 16.

13	 For example, the first LEAD program in Seattle/King County,
Washington, restricts eligibility to people who are charged with
low-level drug offenses and engaging in prostitution and restricts
people who have a history of having been charged with a number
of violent offenses. LEAD, “About,” https://perma.cc/57LS-E9C4.
14	 See for example Rebecca D. Ericson and Deborah A. Eckberg,
“Racial Disparity in Juvenile Diversion: The Impact of Focal
Concerns and Organizational Coupling,” Race and Justice 6, no. 1
(2016), 35-56, 50-53.
15	 Ibid., 42.
16	 Ibid., 45.
17	 Ibid.
18	 For an overview of pre-arrest diversion programs and resources, see
SAFE Project, Law Enforcement Pre-Arrest Diversion Resource Guide
(Arlington, VA: SAFE Project, 2019), https://perma.cc/BCQ6-995Z.
19	 Elena Schwartz, “Do Jail Diversion Programs Really Work?” Crime
Report, July 3, 2018, https://perma.cc/S3SF-SWRS.
20	 LEAD National Support Bureau, “About LEAD,” https://perma.cc/
RW9S-82GD.
21	 For example, in Albany, New York, diversion occurs pre-booking,
and people who would otherwise be booked into jail are diverted to
other social services. See Robert E. Worden and Sarah J. McLean,
“Discretion and Diversion in Albany’s LEAD Program,” Criminal
Justice Policy Review, 29, no. 6-7 (2018), 584-610. See generally
LEAD Albany, Report to Albany on the LEAD Program (Albany, NY:
Katal Center for Health, Equity, and Justice, 2017), 2, https://perma.
cc/MJ9A-YDQJ. Seattle, Washington, where the LEAD program
was developed, incorporates both pre-arrest and pre-booking
diversion, where people can also be diverted after being arrested.
Seattle Police Department, King County, “Law Enforcement Assisted
Diversion (LEAD),” https://www.kingcounty.gov/depts/communityhuman-services/mental-health-substance-abuse/diversion-reentryservices/lead.aspx. Other programs, such as the Stop, Triage,
Engage, Educate and Rehabilitate (STEER) program in Montgomery
County, Maryland, primarily divert people who have been arrested
but not yet booked away from the criminal justice system and

Gatekeepers: The Role of Police in Ending Mass Incarceration

65

toward treatment services. Center for Health & Justice at TASC,
“STEER Deflection Model: Pre-booking Diversion Option for Law
Enforcement,” https://perma.cc/66YC-936W.
22	 LEAD National Support Bureau, “Essential Principles for Successful
LEAD Implementation,” https://perma.cc/WEW8-8Z9D.

40	 Ibid.

23	 Ibid.
24	 Seema L. Clifasefi, Heather S. Lonczak, and Susan E. Collins,
“Seattle’s Law Enforcement Assisted Diversion (LEAD) Program:
Within-Subjects Changes on Housing, Employment, and Income/
Benefits Outcomes and Associations with Recidivism,” Crime and
Delinquency 63, no. 4 (2017), 429-445, 442.
25	 Atlanta/Fulton County Pre-Arrest Diversion Initiative (PAD), “What
We Do,” https://perma.cc/7LHJ-WDSV.
26	 PAD, “History,” https://perma.cc/Q3WG-KQ2B.
27	 Ibid.
28	 Criminal Justice Knowledge Bank, “Heroin Overdose Prevention and
Education (HOPE) Program,” https://perma.cc/MLJ3-3VKG.
29	 Ibid.
30	 Ibid.
31	 Ibid.
32	 Holly A. Wilson and Robert D. Hodge, “The Effect of Youth Diversion
Programs on Recidivism: A Meta-Analytic Review,” Criminal Justice
and Behavior 40, no. 5 (2013), 497-518, 509-10, https://perma.cc/
LV8N-WPFU; and Jennifer S. Wong, Jessica Bouchard, Jason Gravel
et al., “Can At-Risk Youth Be Diverted from Crime? A Meta-Analysis
of Restorative Diversion Programs,” Criminal Justice and Behavior
43, no. 10 (2016), 1310-1329, 1311.
33	 Carolyn S. Dewa, Desmond Loong, Austin Trujillo, and Sarah
Bonato, “Evidence for the Effectiveness of Police-Based Pre-Booking
Diversion Programs in Decriminalizing Mental Illness: A Systematic
Literature Review,” PLOS|ONE 13, no. 6 (2018), https://perma.
cc/9EWL-MKVT.
34	 Civil Citation Network, Leon County/Tallahassee Pre-Arrest
Diversion - Adult Civil Citation Program: A Model Program with
National Implications (Tallahassee, FL: Civil Citation Network, 2017),
1-3, https://perma.cc/ZA85-JHDK.
35	 Ibid.
36	 Albert M. Kopak and Gregory A. Frost, “Correlates of Program
Success and Recidivism among Participants in an Adult Pre-Arrest
Diversion Program,” American Journal of Criminal Justice 42, no. 4
(2017), 727-745, 736-37.
37	 Ibid.
38	 Ibid., 741-43.

66

39	 Seema L. Clifasefi, Heather S. Lonczak, and Susan E. Collins, LEAD
Program Evaluation: The Impact of LEAD on Housing, Employment
and Income/Benefits (Seattle, WA: Harm Reduction Research and
Treatment Lab University of Washington – Harborview Medical
Center, 2016), 1, https://perma.cc/AXG3-PX96.

Vera Institute of Justice

41	 Susan E. Collins, Heather S. Lonczak, and Seema L. Clifasefi, LEAD
Program Evaluation: Criminal Justice and Legal System Utilization
and Associated Costs (Seattle, WA: Harm Reduction Research and
Treatment Lab University of Washington – Harborview Medical
Center, 2015), 2, https://perma.cc/V8AJ-T7SB.
42	 Ibid.
43	 Estimates of the benefits of diversion range from a 25 percent
chance that a police diversion for people with mental illness will
produce benefits greater than costs, to an 86 percent chance
of this kind of success for police pre-arrest diversion for lowseverity offenses. Washington State Institute of Public Policy,
“Police Diversion for Individuals with Mental Illness,” 2017-2018;
and Washington State Institute of Public Policy, “Police Diversion
for Low-Severity Offenses (Pre-Arrest),” 2017-2018, https://perma.
cc/8QPT-Y7LG. One variable that may have a significant effect
on the cost-benefit value of diversion programs is the population
diverted. For example, people with mental illness may be at higher
risk for recidivism than people who are diverted for a misdemeanor
that is their first criminal offense.
44	 Kathleen Hartford, Robert Carey, and James Mendonca, “PreArrest Diversion of People with Mental Illness: Literature Review
and International Survey,” Behavioral Sciences & the Law 24, no. 6
(2006), 845-856, https://perma.cc/UJ5B-A7P8.
45	 William H. Fisher, Lorna Simon, Kristen Roy-Bujnowski et al., “Risk
of Arrest among Public Mental Health Services Recipients and the
General Public,” Psychiatric Services 62, no. 1 (2011), 67-72, 67,
https://perma.cc/9BFW-JTLB. A review of studies primarily from the
United States, Canada, United Kingdom, and Australia concluded
that 25 percent of people living with a mental illness have been
arrested at some point in their lives; 12 percent have had the
police involved in their pathway to mental health services. James
D. Livingston, “Contact Between Police and People with Mental
Disorders: A Review of Rates,” Psychiatric Services 67, no. 8 (2016),
850-857, 6 (arrest statistics) and 7 (pathways to services), https://
perma.cc/M2V9-8FC3.
46	 In general, these responses to people in mental health crisis
are likely to fall into three categories: specially trained police
may directly provide services and act as liaisons; police
departments may hire mental health consultants to accompany
or provide telephone support to officers dealing with a crisis;
or a mobile mental health crisis response team may have a
special relationship with the police department but operate
independently. Martha Williams Deane, Henry J. Steadman, Randy
Borum et al., “Emerging Partnerships Between Mental Health
and Law Enforcement,” Psychiatric Services 50, no. 1 (1999),

99-101, https://perma.cc/33LT-3Q4P; and Henry J. Steadman,
Martha W. Deane, Randy Borum, and Joseph P. Morrissey,
“Comparing Outcomes of Major Models of Police Responses
to Mental Health Emergencies,” Psychiatric Services 51, no. 5
(2000), 645-649, https://perma.cc/P8A5-2QQA. For a discussion
of police perceptions of different response options, see Randy
Borum, Martha Deane, Henry Steadman, and Joseph Morrissey,
“Police Perspectives on Responding to Mentally Ill People in Crisis:
Perceptions of Program Effectiveness,” Behavioral Sciences & the
Law 16, no. 4 (1998) 393-405, https://perma.cc/X9RZ-76KT.
47	 Amy C. Watson and Anjali J. Fulambarker, “The Crisis Intervention
Team Model of Police Response to Mental Health Crises: A Primer for
Mental Health Practitioners,” Best Practices in Mental Health 8, no.
2 (2012), 71, https://perma.cc/V2KF-SWRE.
48	 Ibid.
49	 Nick Margiotta, “The Five-Legged Stool: A Model for CIT Program
Success,” Police Chief Magazine, October 2015, https://www.
policechiefmagazine.org/the-five-legged-stool-a-model-for-citprogram-success.

Christian Ritter, and Mark R. Munetz, “Police Officer Perceptions
of the Impact of Crisis Intervention Team (CIT) Programs,”
International Journal of Law and Psychiatry 37 no. 4 (2014), 341350; and Michael T. Compton, Michelle L. Esterberg, Robin McGee,
et al., “Crisis Intervention Team Training: Changes in Knowledge,
Attitudes, and Stigma Related to Schizophrenia,” Psychiatric
Services 57, no. 8 (2006), 1199-1202, https://perma.cc/W7T4-9LAQ.
For a discussion of the systemic effects of CIT training in the
department, its effectiveness, and retention, see Michael Compton
and Victoria Chien, “Factors Related to Knowledge Retention After
Crisis Intervention Team Training for Police Officers,” Psychiatric
Services 59, no. 9 (2008), 1049-1051, https://perma.cc/5P9Y-3CFK;
Megan Davidson, “A Criminal Justice System-Wide Response to
Mental Illness: Evaluating the Effectiveness of the Memphis Crisis
Intervention Team Training Curriculum Among Law Enforcement
and Correctional Officers,” Criminal Justice Policy Review 27, no.
1 (2016), 46-75; and Michael T. Compton, Roger Bakeman, Beth
Broussard et al., “Police Officers’ Volunteering for (Rather than
Being Assigned to) Crisis Intervention Team (CIT) Training: Evidence
for a Beneficial Self-Selection Effect,” Behavioral Sciences & the
Law, 35, no. 5-6 (2017), 470-479, https://perma.cc/5FCC-9JPT.

51	 Houston now provides CIT (which the department modified to
stand for “Crisis Intervention Training”) training to all cadets; older
officers may engage in training on a voluntary basis. As older
officers retire, the percentage of CIT-trained officers is growing.
Houston Police Department, “Crisis Intervention Training (CIT)
Program,” https://perma.cc/2XCW-Z7E8.

55	 For discussions of how CIT affects the community, including
likelihood of use of force and resolution options, see Compton,
Bakeman, Broussard et al., "Crisis Intervention Team (CIT) Model
II," 2014, 523-529; Steadman, Deane, Borum, and Morrissey,
“Comparing Outcomes,” 2000; Sema Taheri, “Do Crisis Intervention
Teams Reduce Arrests and Improve Officer Safety? A Systematic
Review and Meta-Analysis,” Criminal Justice Policy Review 27,
no. 1 (2016), 76-96; and Sheryl Kubiak, Erin Comartin, and Edita
Milanovic, “Countywide Implementation of Crisis Intervention
Teams: Multiple Methods, Measures and Sustained Outcomes,”
Behavioral Sciences & the Law 35, no. 5-6 (2017), 456-469.

52	 Houston Police Department, “Crisis Intervention Response Team,”
https://perma.cc/ARM7-44LE.

56	 Jackson Beck, “Collaboration Helps Police Serve Safely,” Vera
Institute of Justice, May 23, 2019, https://perma.cc/JE8D-M5QD.

53	 Tallon, Labriola, and Spadafore, Creating Off-Ramps, 2018, 37.

57	 Brian MacQuarrie, “‘Angel’ Opioid Initiative Thrives Despite Exit of
Gloucester Police Chief,” Boston Globe, February 21, 2017, https://
perma.cc/D3TK-TWQL.

50	 Rachel Gandy and Erin Smith, Prioritizing Treatment over
Punishment: An Overview of Mental Health Diversion from Jail in
Texas (Austin, TX: LBJ School of Public Affairs, 2016), 4, https://
perma.cc/S6TF-JU4M.

54	 For discussions of the impact of CIT training on police officers, see
Horace Ellis, “Effects of a Crisis Intervention Team (CIT) Training
Program upon Police Officers Before and After Crisis Intervention
Team Training,” Archives of Psychiatric Nursing 28, no. 1, (2014)
10-16; Michael T. Compton, Roger Bakeman, Beth Broussard et al.,
“The Police-Based Crisis Intervention Team (CIT) Model: I. Effects on
Officers’ Knowledge, Attitudes, and Skills,” Psychiatric Services 65,
no. 4 (2014), 517-522, https://perma.cc/4AUG-SY67; Natalie Bonfine,

58	 Gloucester Police Department, “For Addicts and Their Friends,
Families, and Caregivers,” https://perma.cc/EC8X-ZZSV.
59	 Police Assisted Addiction and Recovery Initiative (PAARI), “About
Us,” https://perma.cc/D2HH-PAGV; and PAARI, “PAARI to Host 4-Year
Anniversary Celebration & Awards,” https://perma.cc/FB8M-YBJA.

Gatekeepers: The Role of Police in Ending Mass Incarceration

67

Acknowledgments
The authors wish to thank the following people for serving as thought
partners in developing this report: Susan Shah, Nicholas Turner, and Daniel
Wilhelm. The authors also thank Jeremy Travis for serving as a mentor
in considering the role of policing in incarceration and criminal justice
reform; Chris Magnus and J. Scott Thomson for their visionary roles in
transforming policing practices in Tucson, Arizona; Camden County, New
Jersey, and nationwide; Sylvia Moir for her commitment to understanding
the role of policing in the larger criminal justice system; and Jim
Bueermann, Ronal Serpas, and Brandon del Pozo for sharing a vision to
reduce American incarceration through changes in police enforcement.
We are also grateful to Mary Crowley, Nancy Fishman, and Jim Parsons at
Vera, and Aisha Edwards, Laurie Garduque, and Bria Gillum at MacArthur,
for their thoughtful review of early drafts; Patrick Griffin for his support
of the project; Megan O’Toole for assistance with the development of this
concept; and Abdul Rad and Henessy Pineda for research assistance. We
thank Alice Chasan for editing the report; Paragini Amin for designing it;
Maris Mapolski for cite checking; Khusbu Bhakta and Karina Schroeder for
editorial support; and Tim Merrill for proofreading.
This report was created with support from the John D. and Catherine T.
MacArthur Foundation as part of the Safety and Justice Challenge
initiative, which seeks to address over-incarceration by changing the way
America thinks about and uses jails. Core to the challenge is a grants
competition designed to support efforts to improve local criminal justice
systems in jurisdictions across the country. The Foundation is supporting
a nationwide network of selected local jurisdictions committed to finding
ways to safely reduce jail incarceration—particularly the disproportionate
incarceration of racial and ethnic minorities. More information is available
at www.SafetyandJusticeChallenge.org.

About citations
As researchers and readers alike rely more and more on public knowledge
made available through the Internet, “link rot” has become a widely
acknowledged problem with creating useful and sustainable citations. To
address this issue, the Vera Institute of Justice is experimenting with the
use of Perma.cc (https://perma.cc/), a service that helps scholars, journals,
and courts create permanent links to the online sources cited in their work.
68

Vera Institute of Justice

Credits
© Vera Institute of Justice 2019. All rights reserved. An electronic version of this report is posted on
Vera’s website at www.vera.org/gatekeepers.
Cover image: © Christopher Anderson/Magnum Photos, 2017
Graphics: Paragini Amin
The Vera Institute of Justice is a justice reform change agent. Vera produces ideas, analysis, and
research that inspire change in the systems people rely upon for safety and justice, and works in
close partnership with government and civic leaders to implement it. Vera is currently pursuing
core priorities of ending the misuse of jails, transforming conditions of confinement, and ensuring
that justice systems more effectively serve America’s increasingly diverse communities. For more
information, visit www.vera.org.
For more information about this report, contact Rebecca Neusteter, policing program director, at
RNeusteter@vera.org. For more information about Vera’s work to reduce the use of jails, contact
Nancy Fishman, project director at Vera’s Center on Sentencing and Corrections, at nfishman@
vera.org.

Suggested citation
S. Rebecca Neusteter, Ram Subramanian, Jennifer Trone, Mawia Khogali, and Cindy Reed. Gatekeepers: The Role of Police in Ending Mass Incarceration. New York: Vera Institute of Justice, 2019.

Gatekeepers: The Role of Police in Ending Mass Incarceration

69

This report was created with support from the John D.
and Catherine T. MacArthur Foundation as part of the
Safety and Justice Challenge, which seeks to reduce overincarceration by changing the way America thinks about
and uses jails.

Vera Institute of Justice
233 Broadway, 12th Floor
New York, NY 10279
T	 212 334 1300
F	 212 941 9407

Washington DC Office
1111 14th Street, Suite 920
Washington, DC 20005
T 202 465 8900
F 202 408 1972

New Orleans Office
1307 Oretha Castle
Haley Blvd., Suite 203
New Orleans, LA 70113
T 504 312 4510
F 504 312 4517

Los Angeles Office
634 S Spring Street, #300A
Los Angeles, CA 90014
T 213 416 0058
F 213 416 0075