U.S.

Department of Justice
Office of Justice Programs
National Institute of Justice
NOV. 04

Research R e p o r t

Reducing
Gun Violence
The St. Louis
Consent-to-Search Program
U.S. Department of Justice
Office of Justice Programs
810 Seventh Street N.W.

Washington, DC 20531

John Ashcroft
Attorney General

Deborah J. Daniels
Assistant Attorney General

Sarah V. Hart
Director, National Institute of Justice

This and other publications and products

of the National Institute of Justice can be
found at:

National Institute of Justice

www.ojp.usdoj.gov/nij

Office of Justice Programs

Partnerships for Safer Communities
www.ojp.usdoj.gov
 NOV. 04

Reducing Gun Violence:

The St. Louis Consent-to-Search Program

Acknowledgments
The authors gratefully
acknowledge the assis­
tance of many individuals
who made this research
possible and helped to
improve it, especially:
Sgt. Simon Risk (ret.),
Capt. Joe Richardson,
Sgt. Terry Sloan,
Sgt. Bob Heimberger,
Sgt. Tom Malacek, and
Maj. Lawrence O’Toole
from the St. Louis
Metropolitan Police
Department; and George
Burruss of Southern
Illinois University–
Carbondale, Bruce Jacobs
of the University of Texas–
Dallas, Steve Mastrofski
of George Mason
University, and Phil Cook
of Duke University.
Lois Felson Mock was the
NIJ Program Manager for
this grant. Her oversight
Findings and conclusions of the research reported here are those of the
was integral to the
authors and do not reflect the official position or policies of the U.S.
successful completion of Department of Justice.
this project.
This research was sponsored by the National Institute of Justice, U.S.
Department of Justice, under grant number 95–IJ–CX–0067.

NCJ 191332
S T. L O U I S C O N S E N T- T O - S E A R C H P R O G R A M

Foreword

This Research Report is part obstacles encountered. Both

of the National Institute of successes and failures are dis­
Justice’s (NIJ’s) Reducing Gun cussed, and recommendations
Violence publication series. are made for future programs.
Each report in the series de­
scribes the implementation While the series includes im­
and effects of an individual, pact evaluation components,
NIJ-funded, local-level pro­ it primarily highlights imple­
gram designed to reduce mentation problems and
firearm-related violence in a issues that arose in design­
particular U.S. city. Some stud­ ing, conducting, and assess­
ies received cofunding from ing the respective programs.
the U.S. Department of Jus-
tice’s Office of Community The Research Reports should
Oriented Policing Services; be of particular value to any­
one also received funding one interested in adopting a
from the Centers for Disease strategic, data-driven, problem-
Control and Prevention. solving approach to reducing
gun violence and other crime
Each report in the series de­ and disorder problems in
scribes in detail the problem communities.
targeted; the program de­
signed to address it; the prob­ The series reports on firearm
lems confronted in designing, violence reduction programs
implementing, and evaluating in Boston, Indianapolis, St.
the effort; and the strategies Louis, Los Angeles, Atlanta,
adopted in responding to any and Detroit.

iii
S T. L O U I S C O N S E N T- T O - S E A R C H P R O G R A M

Contents

An Innovative Gun Recovery Program .....................................1

Program Phases and Results...................................................7

Evaluating a Changing Program .............................................13

Implementation Problems .....................................................15

Implications for Community Policing .....................................19

The Promise of Consent Searches ........................................23

Notes .....................................................................................25

v
 S T. L O U I S C O N S E N T- T O - S E A R C H P R O G R A M

Scott H. Decker and Richard Rosenfeld

An Innovative Gun Recovery

Program

In 1994, facing an epidemic of children might have hidden.

violent crime, St. Louis insti­ Any guns found were confis­
tuted an innovative program cated, with no followup pros­
to reduce its alarmingly high ecution. Parents and young
gun violence rates among people who requested help
youths. The idea for the pro­ were referred to agencies or
gram came from a communi­ community-based groups that
ty meeting of residents and offered appropriate services.
police. During the meeting a
woman complained about a
house where teenagers were Program success and
This report is based on known to possess guns. When shifts
“From Problem Solving to police arrived, the children ran
Crime Suppression to into the house. Officers at the The Consent-to-Search pro­
Community Mobilization: meeting told the woman there gram emerged during the
An Evaluation of the was nothing they could do be­ national epidemic of youth
St. Louis Consent-to- cause there was no legal way violence in the late 1980s
Search Program,” final
to get in and not enough infor­ and early 1990s.2 (See
report by the authors to “Homicides and Other Vio­
mation to obtain a search war­
the National Institute of lent Crime in St. Louis.”) In its
Justice, 2001, available
rant. She then asked a single
question that changed every­ first year, police confiscated
at http://www.ncjrs.org/
thing: “Why don’t you just 402 guns from juveniles. They
pdffiles1/nij/grants/
knock on the door and ask encountered remarkable
188291.pdf.
that mother if you can search cooperation within the com­
the house?” munities that were most
affected by gun violence.
About the Authors
The officers realized that the
Scott H. Decker, Ph.D., The program won national
woman was right. Thus, the
is Curator’s Professor of recognition. It was covered
Criminology and Criminal
Consent-to-Search program
was born.1 widely in the media, nominat­
Justice at the University ed for a prestigious award,
of Missouri–St. Louis.
The program involved police and reviewed by a congres­
Richard Rosenfeld, Ph.D.,
knocking on doors in high- sional committee. In October
is professor and chair of
the Criminology and crime areas and asking par­ 1995, the National Institute of
Criminal Justice Depart­ ents of high-risk youths for Justice funded an evaluation
ment at the University of permission to search their of the program to assess its
Missouri–St. Louis. homes for guns that their effectiveness and determine
its applicability to other cities.

1
 R E D U C I N G G U N V I O L E N C E / N O V. 0 4

As a strategic problem-solving several years, local govern­

approach to reducing gun ment decisions resulted in
violence by getting guns out the program starting and
of the hands of juveniles, stopping twice—each time
Consent-to-Search appeared with a different format and
destined to become a model set of objectives. The pro­
community policing initiative. gram was terminated in
August 1999.
This situation changed abrupt­
ly in December 1995, when
the chief of police who had Lessons learned
established Consent-to-Search
stepped down. Over the next The Consent-to-Search story
offers valuable insights about

HOMICIDES AND OTHER VIOLENT CRIME IN ST. LOUIS

After falling for several years, national homicide rates of persons between 14 and 24 years old
escalated rapidly after 1985, peaking in 1993. By the early 1990s, guns were readily available to
children in many cities through street corner markets.a
The homicide increase in St. Louis during the late 1980s and early 1990sb was concentrated
among African-American adolescents and young adults. By the early 1990s, the city’s homicide
rate reached 380 per 100,000 for black males ages 15–19 and an astonishing 600 per 100,000 for
black males ages 20–24. More than 97 percent of these deaths involved firearms.c
The levels of criminal violence have been sharply higher in St. Louis than in most other cities,
although the demographic patterns of risk are quite similar to other cities.d Changes in St. Louis
homicide rates over the past 30 years correspond closely with national rates,e which suggests
that interventions shown to be effective in St. Louis might be promising candidates for broader
adoption.

Notes
a. American Psychological Association, Violence and Youth: Psychology’s Response: Volume I: Summary Report of the
American Psychological Association Commission on Violence and Youth, Washington, DC: 1993.

b. Cook, P.J., and J.H. Laub, “The Unprecedented Epidemic in Youth Violence,” in Youth Violence, vol. 24 of Crime and Jus­
tice: A Review of Research, ed. M. Tonry and M.H. Moore, Chicago: University of Chicago Press, 1998.

c. Rosenfeld, R., and S. Decker, “Consent to Search and Seize: Evaluating an Innovative Youth Firearm Suppression Program,”
Law and Contemporary Problems 59 (1996): 197–220 (hereinafter “Consent to Search and Seize”).

d. See Jones, M., and B. Krisberg, Image and Reality: Juvenile Crime, Youth Violence, and Public Policy, Washington, DC:
National Council on Crime and Delinquency, 1994. Also see Snyder, H., and M. Sickmund, Juvenile Offenders and Victims:
A Focus on Violence, A Statistics Summary, Washington, DC: U.S. Department of Justice, 1995, NCJ 153570.

e. Rosenfeld and Decker, “Consent to Search and Seize.”

2
 S T. L O U I S C O N S E N T- T O - S E A R C H P R O G R A M

reducing youth gun violence. great risks, threatens public …a central

These insights are especially order, and will not be tolerat­
significant for community ed by the police or the com­ theme of the
policing as it tries to gain a munity. The success of such story is the
toehold in the traditional cul­ an effort depends heavily on
ture of policing and institu­ the quality of the interaction
difficulty of
tional responses to crime. between the community and sustaining law
law enforcement. enforcement
The original Consent-to-Search
program was based on the But a central theme of the innovations
standard community policing story is the difficulty of sus­ in the face
approach of responding to a taining law enforcement
problem identified by citizens. innovations in the face of of strong
By drawing citizens into the strong organizational resist­ organizational
process of identifying and ance and weak external
confiscating illegal firearms, support. It is not enough to
resistance and
officers relied on community have a good idea for an inter­ weak external
expertise—a central tenet of vention, or even to have a support.
problem-solving policing.3 For good working intervention.
example, officers in the first Problem-solving initiatives
phase of the program believed must build support at all lev­
that its success depended on els within the police depart­
scrupulous adherence to the ment. They must also be
promise of no prosecution. anchored within the commu­
They were willing to ignore nity, by design and through
evidence of all but the most partnerships and routine con­
serious crimes in return for tact. Programs that do not
access to the homes of juve­ establish these elements are
niles with firearms. This view— vulnerable to changes in lead­
that arrest opportunities are ership, policy, or other influ­
worth trading for the chance to ences that may counteract or
get guns out of kids’ hands— shut down the intervention.
is essential to this type of
intervention’s success. Communities seeking inter­
ventions to reduce gun vio­
Consent-to-Search can be lence, especially among
viewed as a variant of both the youths, might consider using
aggressive order maintenance consent searches as part of a
and targeted deterrence comprehensive gun recovery
strategies.4 The intervention strategy. (See “Gun Recovery
sent a signal that juvenile Strategies—Consent-to-Search
firearm possession poses in Context.”)

3
 R E D U C I N G G U N V I O L E N C E / N O V. 0 4

GUN RECOVERY STRATEGIES—CONSENT-TO-SEARCH IN CONTEXT

Police use seven tactics to recover firearms and potentially reduce gun violence:
■ Search warrants ■ Pedestrian stops ■ Arrests
■ Gun buybacks ■ Traffic stops ■ Gun turn-in campaigns
■ Consent searches

As each tactic has something to contribute to illegal firearm recovery, the police should employ a
mix of complementary tactics to be effective in reducing illegal firearm use. Communities consider­
ing gun violence reduction strategies may want to carefully balance factors of risk, outcome, and
costs (see exhibit 1).
Risk. A gun recovery intervention should first consider the level of risk for each strategy. For exam­
ple, search warrants and arrests generally identify high-risk offenders who pose a danger to offi­
cers. Because consent searches target individuals considered to be at risk for involvement in crime,
either as victims or offenders, they are likely to have a greater crime-reduction payoff than pedes­
trian or traffic stops—while posing less risk for police.
Probability of gun seizure. A second criterion for choosing a gun recovery intervention is the prob­
ability that a gun will be found. Those efforts most likely to yield guns—search warrants and gun
buybacks—are the most dangerous and least dangerous tactics, respectively. Although traffic and
pedestrian stops are deemed the least likely to get guns, they account for the majority of gun seizures
because of the sheer volume of these stops—literally thousands per year.a
Search warrants, gun buybacks, and gun turn-in campaigns have a high yield in firearms but account
for only a fraction of the guns recovered by the police. Guns recovered through buybacks and turn-
in campaigns are the least likely to have been involved in crime.
Crime reduction and social costs. Another consideration is whether the removal of guns through a
given tactic or set of tactics results in a net reduction in crime. Search warrants and arrests usually
are executed because an offense has occurred or is imminent; therefore, they are most likely to
reduce criminal activity in the near term. All of the other tactics, including consent searches, are
less likely to identify an individual involved in crime during gun seizure or in the near future.
Tactics that have an immediate effect on crime have an intuitive appeal. Search warrants, arrests,
and some traffic and pedestrian stops have this potential if officers are trained to look for firearms.b
But traffic and pedestrian stops can have substantial social implications. A major complaint about
U.S. law enforcement is the alleged use of racial profiling to stop minorities in proportions far greater
than their representation in the population. Thus, traffic and pedestrian stops generate distrust of
the police for many Americans.

4
 S T. L O U I S C O N S E N T- T O - S E A R C H P R O G R A M

Gun Recovery Strategies (continued)

Search warrants, arrests, and consent searches exact more moderate social costs because, al­
though these tactics are invasive, the perceived crime-control benefits offset their intrusiveness.
When done in close partnership with the community, consent searches may actually increase
police-citizen cooperation and therefore have a low social cost.
Collaboration. A final dimension for comparing these strategies is whether a collaborative partner
is needed. Activities that rely on collaboration between police and the community—such as con­
sent searches—are more difficult to execute than those that the police can perform themselves.
Nonetheless, consent searches based on police-citizen collaboration have proved to be an effec­
tive, relatively low-risk tactic to recover illegal firearms from juveniles.

Exhibit 1. Impact of gun recovery tactics

Level Probability Difficulty Crime
of risk of getting of getting reduction Social Collaboration
from subject a gun a gun impact Cost ($) cost required

Search warrants High High High High High Medium No

Arrests High Medium Medium High Medium Medium No

Traffic stops Medium/Low Low Low Low Low Medium/High No

Pedestrian stops High Low Low Low Low High No

Consent searches Medium/Low Medium Medium Medium Medium Low Yes

Gun buybacks Low High Low Low Medium Low Yes

Gun turn-in
campaigns Low High Low Low Low Low No

Notes
a. See Burruss, G.W., and S.H. Decker, “Gun Violence and Police Problem Solving: A Research Note Examining Alternative Data
Sources,” Journal of Criminal Justice 30(6) (Nov./Dec. 2002): 567–574.

b. See Sherman, L., J. Shaw, and D. Rogan, The Kansas City Gun Experiment, Research in Brief, Washington, DC: U.S. Department
of Justice, National Institute of Justice, 1995, NCJ 150855.

5
 S T. L O U I S C O N S E N T- T O - S E A R C H P R O G R A M

Program Phases and Results

In its earliest stage, Consent- by police (see exhibit 2.) This

to-Search was a police prob- made it easier to identify the
lem-solving tool directed at effects of program changes.
one of St. Louis’s most seri­ (See exhibit 3 for a timeline
ous crime problems. In its depicting pivotal events and
later stages, however, the pro­ program changes.)
gram fundamentally changed.
To evaluate the program in the
face of the major changes that Phase I—Problem
occurred in 1996 and 1998, Solving/Aggressive
researchers divided it into Order Maintenance
three phases that correspond
to the changes in operational The St. Louis Consent-to-
philosophy and approach used Search program in Phase I

Exhibit 2. St. Louis Consent-to-Search program phases

Phase I Phase II Phase III
Problem Solving/ Crime Targeted Intervention/
Aggressive Order Control/ Attempted Community
Maintenance Suppression Mobilization

Program goals and methods

Orientation ■ Problem solving ■ Crime control ■ Community policing

■ Aggressive order maintenance ■ Suppression ■ Community mobilization
Tools ■ Consent searches ■ Search warrants ■ Consent searches
■ Arrests ■ Referrals
Sources of ■ Community meetings ■ Police ■ Police
search target ■ Intelligence ■ Police Information Record
information System
■ Gang unit
■ Intelligence

Objectives ■ Respond to citizens’ concerns ■ Make arrests ■ Seize weapons

■ Seize weapons from juveniles ■ Seize weapons ■ Make referrals
■ Notify parents ■ Gain intelligence ■ Involve a community partner
■ Refer parents to assistance ■ Serve warrants ■ Notify parents

Program process ■ Home visits ■ Warrants ■ Home visits

■ Consent requests ■ Consent requests

7
 R E D U C I N G G U N V I O L E N C E / N O V. 0 4

combined two proven aspects The program was operated

of community policing: prob­ by the Mobile Reserve unit, a
lem solving and aggressive squad without a specific geo­
order maintenance. In its first graphic assignment that re­
full year of operation, the pro­ sponds to pockets of crime
gram enjoyed spectacular suc­ and violence throughout the
cess, especially in three areas: city. Training was conducted
citizen cooperation, the tar­ within the unit. Officers in­
geting of locations that were volved in this phase attributed
likely to yield guns, and the its early success to its low-key
confiscation of firearms. Of approach.6 Two officers and a
citizens who were approached, Mobile Reserve sergeant vis­
98 percent consented to a ited the residence in question,
search. This degree of cooper­ spoke with an adult resident,
ation is quite remarkable, and requested permission to
given the historic pattern of search the home for illegal
distrust between the police weapons.
and the black community in
St. Louis.5 In 1994, the Mobile Reserve
unit conducted between 5 and
Cases in this phase were ini­ 30 searches each night the
tiated mostly by citizens who program was in operation.
identified homes to search. Guns were found in half of

Exhibit 3. St. Louis Consent-to-Search program timeline

su es ns
m f r gi
en s

d
d

inv d
d

d
sp ign

te
ra ie be

te
te

re te
de

Pr m nde
d
te

en

ina
ta

m ta
og Ch n
tia

og uspe
Pr ice atio

ins

ra ins

rm
ini

re

og re

te
Po valu

s
m

m
m
ra

ra

ra

ra
Je

ra
og

og

og
og
l
NI
Pr

Pr

Pr

Pr
Pr

1993 1994 1995 1996 1997 1998 1999 2000

Phase I Phase II Phase III
Problem Solving/ Crime Control/Suppression Targeted Intervention/
Aggressive Order Maintenance Attempted Community
(June 1996–March 1998)
(Sept. 1993–Dec. 1995) Mobilization
(Nov. 1998–Aug. 1999)

8
 S T. L O U I S C O N S E N T- T O - S E A R C H P R O G R A M

the homes searched; an aver­ commander was unaware of

age of three guns were the program’s accomplish­
seized per household. Anec­ ments; researchers showed
dotal evidence indicated him a newspaper article to
strong support for the pro­ convince him. He agreed to
gram among adults in neigh­ resume the program, but only
borhoods that experienced in a new form. The Phase II
high levels of gun violence. program had a different set of
One parent even offered to objectives and procedures.
sign several predated forms The primary goal of consent
so that the police could return searches changed from seizing
at any time; another wanted guns to arresting offenders.
to give police a key to her
house so they could search The changes were drastic: the
while she was at work. department abandoned the
problem-solving approach of
An innovative feature of the removing weapons from juve­
program was the use of a niles through consent in favor
“Consent to Search and Seize” of traditional crime control and
form to secure legal access to suppression. Arrests, search
the residence (see exhibit 4). warrants, and intelligence
Residents were assured that gathering replaced the original
the sole purpose was to con­ goals of consent, no arrests,
fiscate illegal firearms pos­ and referrals for services.
sessed by juveniles and that
by providing written consent Consent searches were rele­
to search, they would not be gated to a marginal role as
charged with illegal posses­ more aggressive tactics—
sion of a firearm. especially search warrants—
received top priority. The new
lieutenant in charge of the
Phase II—Crime Mobile Reserve unit explained
Control/Suppression the shift in philosophy: “Why
only get a gun with a consent
One month after the police search, when you can get a
chief who founded the gun and a criminal with an
Consent-to-Search program arrest or search warrant?”
resigned, his successor sus­
pended the program. In early The program received little
1996, the lead evaluator met support from members of the
with the new commander of newly constituted Mobile
the division housing the Reserve unit. No training was
Mobile Reserve unit. The provided; many in the unit

9
 R E D U C I N G G U N V I O L E N C E / N O V. 0 4

Exhibit 4. The Consent to Search and Seize form

were unaware of the pro­ reestablished as a result of

gram. The pledge of no pros­ support from the U.S. Attor-
ecution was removed from ney’s Office,7 the availability
the consent-to-search form. of Federal funding,8 and con­
After 9 months of warranted tinued national and local atten­
searches that yielded relative­ tion to youth firearm violence,
ly few guns, the program was which kept pressure on the
discontinued, although it was police department to do some­
not officially terminated until thing about guns in the hands
12 months later. of juveniles.

Phase III operated differently

Phase III—Targeted from the earlier phases. Moved
Intervention/Attempted from Mobile Reserve to the
Intelligence unit, the program
Community Mobilization was conducted on an over­
In January 1999, the Consent- time basis. Officers were
to-Search program was
10
 S T. L O U I S C O N S E N T- T O - S E A R C H P R O G R A M

trained and program activities Program results

were documented. The origi­
nal consent-to-search form Gaps in data resulting from
with promise of no prosecu­ programmatic changes and
tion was restored. Participating operational lapses constrained
officers were chosen because the evaluators’ ability to meas­
they had served in Phase I ure overall program impact.
and were committed to the However, program changes
policy, elements not present and some outcomes were
in Phase II. This link to Phase I documented.
lent credibility to the new
Phase I. The comparative suc­
initiative.
cess of the Phase I approach
A hallmark of Phase I was its as a means of gaining parental
reliance on community input consent and seizing illegal
as a major source of target ad­ guns from juveniles is evident
dresses for consent searches. from the nearly total rate of
In contrast, Phase III relied pri­ parental compliance and the
marily on internal police data high number of guns seized
sources to select juveniles (see exhibit 5).
who had been arrested or
Phase II. Guns seized in this
mentioned in a field incident
phase were largely from
report.
search warrants and arrests,
Another new element in as consent searches were not
Phase III was a formal part­ used. During the approximate­
nership with a coalition of ly 9 operational months of
African-American churches Phase II, only 6 percent as
called the African-American many guns were recovered
Churches in Dialogue. This as during the 12 operational
group was formed to present months of Phase I.
a unified voice and leadership
Phase III. Despite some favor­
regarding social issues affect­
able features, Phase III did not
ing the black community in
match the achievements of
St. Louis. The police request­
Phase I. Granting of consent
ed the group’s help with re­
fell by nearly 50 percent, and
ferrals from consent searches,
the number of guns seized
and the group also agreed to
remained nearly 90 percent
counsel parents and youths.
below Phase I levels. This
The officers were relieved to
phase was heavily police driv­
have somewhere to refer dis­
en, which may account for the
traught or desperate parents.

11
 R E D U C I N G G U N V I O L E N C E / N O V. 0 4

much lower level of compli­ Another factor that also could

ance by parents or guardians have affected the number
than in Phase I. As the police of illegal firearms seized in
assumed greater responsibili­ Phase III is that overall levels
ty for determining where to of youth firearm violence had
search, relying on neighbor­ fallen in St. Louis and other
hood sources less, they met large cities during that period
with more refusals and recov­ (1998–99). There may have
ered significantly fewer guns been fewer illegal firearms in
from youths.9 the community.

Exhibit 5. St. Louis Consent-to-Search program results

Phase I Phase II Phase III

Consent given 98% Not applicable 42%

Hit rate (percent of
searches netting a gun) 50% Not applicable <25%
Program duration 18 months 9 months 9 months
Guns seized 510 31 29
Community partner Parents and neighbors None Clergy

12
S T. L O U I S C O N S E N T- T O - S E A R C H P R O G R A M

Evaluating a Changing Program

The Consent-to-Search pro­ Process evaluation

gram evaluation suffered from
implementation lapses and
obstacles
changes in program design Ideally, a process evaluation
and execution. Phase I rec­ would identify the attributes
ords were lost;10 to evaluate of individuals, the program
that phase, researchers inter­ components, and the commu­
viewed police about their rec­ nity characteristics associat­
ollections and relied on the ed with a high level of citizen
public record. Phase II data compliance with police search
sources were activity logs, requests and with a high ratio
ride-alongs, observations, and of firearm confiscations to
interviews. Phase III sources searches. But researchers
were police records, reports, faced major challenges in col­
interviews, and ride-alongs. lecting such process data be­
cause of the changing nature
Researchers tried unsuccess­ of the program. For example,
fully to convince the police to
concentrate on a few high- ■ Targets of searches were
crime neighborhoods, employ­ determined by different
ing similar neighborhoods as sources in each phase and
controls. Otherwise, the thus were not comparable
team argued, the impact of throughout the program
the program on firearm vio­ phases.
lence would be diluted and
difficult to measure. How­ ■ Degree of compliance could
ever, the police preferred a not be a process measure
problem-solving approach since searches in Phase II
that focused on high-risk were warrant searches, not
youths rather than high-risk consent searches.
areas.11 This made an out­
come evaluation infeasible.
■ Phase III introduced a com­
munity partnership element
not present in the other two
phases.

13
 R E D U C I N G G U N V I O L E N C E / N O V. 0 4

Impact evaluation ■ Measure the program’s

effect on the level of com­
obstacles munity safety, including
To measure program impact, firearms crime and victim­
a model outcome evaluation ization in neighborhoods.
of the Consent-to-Search
program would have pursued Ideally, juveniles whose
three objectives: homes were searched would
have been contacted and
■ Determine whether the pro­ interviewed regarding their
gram reduced youth firearm rearming and sense of per­
possession, i.e., did youths sonal security. Because
rearm after their guns were evaluators could not deter­
confiscated? mine the effectiveness of
Consent-to-Search in meeting
■ Determine whether confis­ the first two objectives, they
cation of guns threatened could not evaluate the pro-
the personal security of gram’s impact on community
young people, i.e., did they safety.
believe that without a gun
they were more at risk to
become victims of violence?

14
S T. L O U I S C O N S E N T- T O - S E A R C H P R O G R A M

Implementation Problems

Despite the intuitive sense problem solving or community

Consent-to-Search made for policing. The prevailing depart­
a city with very high rates mental philosophy from early
of youth violence, the pro­ 1996 into 1998 was that con­
gram could not be sustained sent searches were ineffec­
because of fundamental defi­ tive and too soft on offenders.
ciencies in how it was imple­ Departmental leadership
mented and maintained. placed a higher priority on
Attempts in Phase III to cor­ taking offenders rather than
rect some of these problems guns off the streets.
were insufficient, although
Phase III did raise awareness Officer resistance. The tough­
of the program within the est impediment to implement­
department and among other ing and sustaining problem-
law enforcement and city solving initiatives is officer
agencies. resistance. Even in police
departments that encourage
problem-solving strategies,
Why the program organizational support for an
ended intervention such as Consent-
to-Search can be precarious.
The Consent-to-Search pro- For example, an ongoing
gram’s demise is rooted in issue throughout the program
six broad areas. was whether it constituted
“real police work.” Some offi­
Uncommitted leadership. A cers viewed the program as a
distinctive organizational cul­ community relations exercise
ture is required to encourage with little impact on crime.
and sustain favorable attitudes
about problem-solving initia­ Program isolation. Problem-
tives among police officers. solving or community policing
Strong leadership is especial­ initiatives are often handled
ly needed. After the first chief by a single police unit or divi­
left, the St. Louis Police De­ sion, as opposed to across
partment had no organiza­ the department. Subunit
tional commitment to either autonomy helps to insulate

15
 R E D U C I N G G U N V I O L E N C E / N O V. 0 4

officers from the traditional partner (the African-American

norms and procedures of Churches in Dialogue) to fol­
“real police work,” i.e., mak­ low up on referrals received
ing arrests, investigating from the police. The evaluation
crimes, pursuing offenders.12 team was unable to document
The drawback is that subunit a single case of followup. This
autonomy can lead to unit breakdown created a dilemma
isolation, which works against for some officers and fueled
integrating innovations into others’ cynicism about police
the department’s mainstream. involvement in this type of
In this case, the Mobile Re­ intervention.
serve unit’s isolation pre­
vented other captains and Lack of institutional memory.
lieutenants from accepting The St. Louis Police Depart­
program goals and methods. ment kept few records of
Consent-to-Search until the
Lack of training. Without a for­ program’s last year. No entries
mal training manual, program were made in the confiscated
description, and documenta­ firearms logs regarding meth­
tion of past achievements, an od of confiscation. Such
innovation cannot diffuse records would have document­
throughout the department. ed the effectiveness of con­
What training existed was sent searches compared with
limited to the unit, which insu­ other confiscation methods.
lated the program from rou­ In the absence of documen­
tine inservice training and tation, officers could legiti­
adoption by the department mately question the veracity
as a whole. of the program’s early reputa­
tion for success. As noted
Absence of followup. The previously, failure to establish
final blow to Consent-to-Search outcome measures made
was the apparent inability of program evaluation difficult,
the Phase III community if not impossible.

16
 S T. L O U I S C O N S E N T- T O - S E A R C H P R O G R A M

A unique legacy Operation Ceasefire charac­

teristics were absent from
Where in the inventory of Consent-to-Search.
innovations is the Consent-
to-Search program best However, the program’s ex­
situated? Perhaps the best traordinary achievements in
known police problem-solving Phase I give it a unique status
partnership—and a good among gun violence interven­
benchmark—is the Boston tions. Coupling this success
Gun Project’s Operation with the program’s failures
Ceasefire program.13 The suc­ makes Consent-to-Search a
cess of that intervention potent example of the diffi­
hinged on the broad array of culties in program implemen­
constituents who embraced tation that policymakers and
the program, the continuous practitioners must bear in
review of data to assess pro­ mind. Even the most suc­
gram progress, and the close cessful program will face
linking of the program to challenges.
review findings. These

17
S T. L O U I S C O N S E N T- T O - S E A R C H P R O G R A M

Implications for Community

Policing

Problem-solving policing works Absence of a

best if police take their cues
from the community. When
constituency
Consent-to-Search lost that Probably the greatest reason
connection, gun seizure lev­ for Consent-to-Search’s over­
els fell substantially below all failure was the program’s
the levels achieved in Phase I lack of a constituency, either
when the community was internally within the police
involved. It is not unusual for department or externally
a police department to return among city or community
to the philosophy and tactics leadership or the citizenry.
that were the backbone of
law enforcement activity for Internal. Problem-solving
most of the 20th century. innovations cannot survive
Such a reversion is more like­ without internal proponents
ly if an innovation was never who have sufficient prestige
routinely implemented or and influence to protect them
institutionalized. from internal and external
challenges. For the Consent-
Past research shows that fail­ to-Search program to have
ure to develop a constituency, survived the first police chief’s
lack of support from commu­ resignation, supporters within
nity residents, and officer re­ the department’s leadership
sistance have been the main and rank-and-file officers
impediments to community would have been necessary.
policing efforts.14 The St. None came forward until
Louis Consent-to-Search pro- Phase III.
gram’s experience, it seems,
was not unique in this regard. Community and parental.
In Phase I, these impediments As critical as internal con­
were absent and the program stituencies are, external con­
enjoyed considerable success. stituencies are even more
However, each of these three important. Participation by
impediments played a role in community groups willing to
the program’s difficulties in work with law enforcement
Phases II and III. is difficult to achieve in high-
crime communities where

19
 R E D U C I N G G U N V I O L E N C E / N O V. 0 4

tension between residents developing a community con­

and police can run high. stituency. Such constituencies
are best developed through
Phase I’s success in gaining sustained contact between
consent to enter a home may officers and citizens.
have been due to a program
emphasis on officer civility, Phase II made little to no effort
The presence but it also may reflect parental to involve the community.
concern for their children’s
of the police safety and well-being—in During Phase III, the commu­
on their effect a parental constituency. nity partner did not follow up
The presence of the police on on police referrals for services.
doorsteps This unanticipated develop­
their doorsteps seemed to
seemed to indicate or reinforce to many ment reveals the difficulties in
indicate or parents that a problem exist­ forging effective partnerships
ed and that they should coop­ between law enforcement
reinforce to erate. Phase I was the only and community organizations
many parents phase to exploit this intangi­ during problem-solving experi­
ble factor. Phase II ignored ments. What went wrong is
that a this element, and Phase III unclear. One implication is
problem restored it only partially. the need to review how com­
munity partnerships function
existed and in problem-solving initiatives
that they Lack of community and whether the partners
should support have the requisite skills and
commitment.
cooperate. Phase I achieved significant
community cooperation in
providing information about Overcoming officer
homes to target for searches; resistance
however, these sources
were not recruited as active Even though only officers with
participants—as an advisory a demonstrated commitment
board or support group, for to the program participated in
example. Had neighborhood Phase III, few of them believed
groups been strongly commit­ that the program would lead
ted to the program, it might to lower levels of violence
not have been suspended and among juveniles. Most felt
redesigned. Isolation of the that adolescents could obtain
program in a small police unit illegal firearms easily. Some
that was not neighborhood acknowledged the possibility
based also worked against that the program could reduce

20
 S T. L O U I S C O N S E N T- T O - S E A R C H P R O G R A M

youth violence if they got the greater support from police

“right” guns at the “right” officers. Most officers saw
time, but that prospect was value in such elements of the
considered unlikely, given the program as—
large pool of available firearms
in a city with high levels of Problem solving. Officers
firearm violence. valued using problem solv­
ing to address youth vio­
What broke down the resis­ lence and other issues,
tance of these Phase III offi­ especially assisting with
Community-
cers was the return to working non-law enforcement prob­ based
with the community, aug­ lems such as housing code selection
mented by training and inter­ violations, school problems,
nal support. This led to an and unemployment. of search
increased understanding of targets
consent searches’ deterrent Referral. Officers believed
effect. that the relatively modest seems to
crime-reduction effects of have secured
weapon seizure should have
What works been complemented by
greater
other interventions. Frus­ cooperation
In Phase I, Consent-to-Search trated by their inability to and,
depended on city residents address underlying condi­
and organizations (block tions, they appreciated ultimately,
groups, Neighborhood Watch,
parents) to determine which
having local clergy as com­ more guns.
munity partners to whom
homes to search. Community- they could refer parents
based selection of search tar­ and their children.
gets seems to have secured
greater cooperation and, ulti­ Deterrence. Many Phase III
mately, more guns. Perhaps officers simply sought a
residents have a more intu­ safer community for youths.
itive sense or better knowl­ They were not as con­
edge of which juveniles are cerned with the target of
likely to be harboring illegal the consent search—the
guns. guns—as they were with
the message sent to par­
However, the researchers
ents and guardians in the
believe that Phase III’s more
community.
coordinated approach is the
best model for this type of Intelligence gathering. Most
program because it enjoyed officers felt that consent

21
 R E D U C I N G G U N V I O L E N C E / N O V. 0 4

searches provided an op­ light on what works in this

portunity to gather informa­ type of community policing
tion that otherwise would response to gun violence if
have been unavailable. They approaches from Phases I
interacted more supportively and III are combined:
with residents and observed
the family circumstances of ■ Use community sources to
many juveniles. identify where to search.

Improved police image. ■ Place a higher priority on

Some officers hoped that seizing guns than on prose­
residents would see them cuting some young people.
in a different light, more as
partners than opponents.
■ Emphasize deterrence and
Although evaluators had no referral to social services.
objective measure of this ■ Train officers and ensure
perception, they noted that
departmental support.
some parents who were
skeptical at the doorstep ■ Secure and maintain
were cooperative by the community participation
end of the search. and support.
For those seeking to imple­ ■ Establish procedures
ment gun recovery programs, for recordkeeping and
Consent-to-Search sheds evaluation.

22
S T. L O U I S C O N S E N T- T O - S E A R C H P R O G R A M

The Promise of Consent Searches

The consent search interven­ Consent searches

tion has the potential to be
part of a broader repertoire
today
of tools the police can use St. Louis still grapples with
to reduce firearms violence. high rates of violence, espe­
Hundreds of guns were cially homicide, aggravated
seized under the auspices of assault, and robbery. In 2000,
Consent-to-Search, and the the city became a test site
police made many referrals for a National Institute of Jus­
to youth-serving agencies. tice program called Strategic
Approaches to Community
An irony of the Consent-to- Safety Initiative (SACSI). At
Search program is the evalua­ each SACSI site, the U.S.
tors’ finding that Phase III was Attorney, local decisionmak­
more successful in the eyes ers, and a research partner
of police officers than Phase I, collaborate to reduce crime.
despite lower gun recovery Researcher participation en­
levels. Poorer results in Phase sures that strategies will be
III may be mitigated by other based on real data, such as
factors. From the officers’ the finding that homicides in
viewpoint, more is at stake St. Louis are concentrated in
with this type of program than 10 neighborhoods.
how many guns are taken.
Convinced that seizing fire­ In 1998, St. Louis established
arms alone does not ade­ a program based on the Bos­
quately address youth gun ton model to address violent
violence, police preferred the crime by focusing on the
community partnership and city’s “hot spots.” The program
neighborhood targeting char­ includes consent searches
acteristics of Phase III. They conducted by a trained team
also recognized the aggres­ using the original tactics de­
sive order maintenance and veloped in 1993. The pres­
deterrent effects of searches, ence of a powerful external
which sent a message to the constituency—in this case,
community that illegal guns the U.S. Attorney—combined
in the hands of juveniles will with police department sup­
not be tolerated or ignored.15 port has helped integrate

23
 R E D U C I N G G U N V I O L E N C E / N O V. 0 4

consent searches into routine Thus, within the context of

policing. Teaming with com­ coordinated programs rooted
munity groups that have a in the community, consent
stake in program success has searches have become a valu­
completed the advocacy net­ able element of a comprehen­
work needed to sustain the sive gun violence reduction
program. program.

In 2002, St. Louis became part

of Project Safe Neighborhoods A final observation
(PSN), a national initiative to
reduce gun violence in Ameri- The Consent-to-Search expe­
ca.16 Like SACSI, Project Safe rience shows that although
problem-solving initiatives
Neighborhoods fosters part­
can be quite effective, they
nerships of Federal, State, and
are relatively fragile. An inno­
local law enforcement and
vative problem-solving inter­
community-based groups.
vention must rest on an
One aspect is based on the
established network of sup­
premise that many youths are
port among local justice pro­
“career” offenders or victims
fessionals, elected officials,
(or both). Police follow up on
police leadership and field
emergency room gunshot and
officers, and the community.
stab wound victims to prevent
The challenge is to sustain
retaliatory violence and repeat
internal and external support
victimization. Followup may
for consent searches as part
include consent searches.
of a broader community
Consent searches may also
mobilization to reduce
be integrated into another
firearms violence.
PSN program that will tackle
gun violence in schools and
among juveniles.

24
 S T. L O U I S C O N S E N T- T O - S E A R C H P R O G R A M

Notes

1. See Bryan, B., “Soft Sell of Searches 6. Officers deliberately avoided con­
Nets Police 3,900 Guns,” St. Louis frontational or intimidating behavior.
Post-Dispatch, April 10, 1995: 13a See Bryan, B., “Soft Sell of Searches.”
(hereinafter “Soft Sell of Searches”). Also see Decker, S., and R. Rosenfeld,
“From Problem Solving to Crime
2. Cook, P.J., and J.H. Laub, “The Suppression to Community Mobiliza­
Unprecedented Epidemic in Youth tion: An Evaluation of the St. Louis
Violence,” in Youth Violence, vol. 24 Consent-to-Search Program,” final
of Crime and Justice: A Review of report, Washington, DC: U.S. Depart­
Research, ed. M. Tonry and M.H. ment of Justice, National Institute of
Moore, Chicago: University of Justice, 2001, NCJ 188291: 6. Avail­
Chicago Press, 1998. able at http://www.ncjrs.org/pdffiles1/
nij/grants/188291.pdf (hereinafter
3. See Goldstein, H., “Improving Po­ “From Problem Solving to Crime Sup­
licing: A Problem-Oriented Approach,” pression to Community Mobilization”).
in Thinking About Police: Contempo­
rary Readings, ed. C. Klockars and S. 7. The U.S. Attorney heard about the
Mastrofski, New York: McGraw Hill, Consent-to-Search program at a
1991; and Goldstein, H., New Polic­ national conference and returned to
ing: Confronting Complexity, Re­ St. Louis only to find that the idea
search in Brief, U.S. Department of had developed there years before.
Justice, National Institute of Justice, He used his leverage to ensure its
Washington, DC: 1993, NCJ 145157. reinstatement.

4. See Kennedy, D., “Pulling Levers: 8. Funds were allocated from the
Getting Deterrence Right,” National Local Law Enforcement Block Grant,
Institute of Justice Journal (July 1998): which also elevated the need for
2–8; and Kelling, G., and C. Coles, Fix­ accountability—use of Federal funds
ing Broken Windows: Restoring Order to pay officer overtime to conduct
and Reducing Crime in Our Commu­ consent searches required develop­
nities, New York: Free Press, 1996. ment of a system for reporting activity.

5. The Consent-to-Search program’s 9. Because in many cases parents

constitutionality was an issue for were absent or no one was home,
some local observers, but there are results may underestimate the num­
some legal precedents supporting ber of parents who might have per­
this approach. Federal courts have mitted searches. See Decker, S., and
ruled that the property of children R. Rosenfeld, “From Problem Solving
can be searched with consent from to Crime Suppression to Community
their parents or guardians. For exam­ Mobilization”: 15.
ple, see U.S. v. Rith, 164 F.3d 1323
(10th Cir. 1999); and U.S. v. Ladell,
127 F.3d 622 (7th Cir. 1997).

25
 R E D U C I N G G U N V I O L E N C E / N O V. 0 4

10. Ibid.: 8. 14. See Sadd, S., and R.M. Grinc,

Implementation Challenges in
11. Ibid.: 12. Community Policing: Innovative
Neighborhood-Oriented Policing in
12. See Van Maanen, J., “Kinsmen in Eight Cities, Research in Brief, Wash­
Repose: Occupational Perspectives ington, DC: U.S. Department of
of Patrolmen,” in The Police and Soci­ Justice, National Institute of Justice,
ety: Touchstone Readings, 2d ed., 1996, NCJ 157932.
ed. V. Kappeler, Prospect Heights, IL:
Waveland Press, 1999. 15. Deterrence was also a critical
factor in the Indianapolis Directed
13. See Kennedy, D., A. Piehl, and A. Patrol Project. See McGarrell, E.F., S.
Braga, “Youth Violence in Boston: Gun Chermak, and A. Weiss, Reducing
Markets, Serious Youth Offenders, Gun Violence: Evaluation of the
and a Use-Reduction Strategy,” Law Indianapolis Police Department’s
and Contemporary Problems 59 Directed Patrol Project, Research
(1996): 147–196; and Kennedy, D., A. Report, Washington, DC: U.S. Depart­
Piehl, A. Braga, and E. Waring, Reduc­ ment of Justice, National Institute of
ing Gun Violence: The Boston Gun Justice, 2002, NCJ 188740.
Project’s Operation Ceasefire, Re­
search Report, Washington, DC: U.S. 16. For more information on
Department of Justice, National Insti­ Project Safe Neighborhoods, see
tute of Justice, 2001, NCJ 188741. http://www.psn.gov.

26
 The National Institute of Justice is the
research, development, and evaluation
agency of the U.S. Department of Justice.
NIJ provides objective, independent,
evidence-based knowledge and tools
to enhance the administration
of justice and public safety.

NIJ is a component of the Office of Justice

Programs, which also includes the Bureau
of Justice Assistance, the Bureau of Justice
Statistics, the Office of Juvenile Justice
and Delinquency Prevention, and the
Office for Victims of Crime.
U.S. Department of Justice
PRESORTED STANDARD
Office of Justice Programs
POSTAGE & FEES PAID
National Institute of Justice
DOJ/NIJ
Washington, DC 20531 PERMIT NO. G–91
*NCJ~191332*
Official Business
Penalty for Private Use $300

MAILING LABEL AREA (5” x 2”)

DO NOT PRINT THIS AREA
(INK NOR VARNISH)

NOV. 04