Attention Felons: Evaluating Project Safe Neighborhoods in Chicago

TLDR: The authors used a quasi-experimental design to evaluate the impact of Project Safe Neighborhood (PSN) initiatives on neighborhood-level crime rates in Chicago and found that several PSN interventions are associated with greater declines of homicide in the treatment neighborhoods compared to the control neighborhoods.

Abstract: This research uses a quasi-experimental design to evaluate the impact of Project Safe Neighborhood (PSN) initiatives on neighborhood-level crime rates in Chicago. Four interventions are analyzed: (1) increased federal prosecutions for convicted felons carrying or using guns, (2) the length of sentences associated with federal prosecutions, (3) supply-side firearm policing activities, and (4) social marketing of deterrence and social norms messages through justice-style offender notification meetings. Using individual growth curve models and propensity scores to adjust for nonrandom group assignment of neighborhoods, our findings suggest that several PSN interventions are associated with greater declines of homicide in the treatment neighborhoods compared to the control neighborhoods. The largest effect is associated with the offender notification meetings that stress individual deterrence, normative change in offender behavior, and increasing views on legitimacy and procedural justice. Possible competing hypotheses and directions for individual-level analysis are also discussed.

Figures

FIGURE 4. Distribution of Index of PSN Components by Group Assignment

TABLE 6. Summary of Operation Ceasefire and PSN Effects on Log Homicide Rate (Slopes, Exp(B), Standard Errors, and p-values), 1999 to 2004

FIGURE 6. Smoothed Quarterly Homicide Rates by PSN Group Assignment, 1999 to 2004

FIGURE 5. Monthly Homicide Rate in PSN Treatment Group, 1999 to 2004

FIGURE 7. Fitted Linear Growth Curves of Homicide Rate (logged) on Predicted PSN Propensity Scores and Percent of Offenders Who attended Forum (logged) (95 Percent Confident Intervals)

TABLE 5. Multi-level Random Intercept Models of Gun Homicide Rate (logged) with and without five beats with highest intercepts. (Coefficients, (Standard Errors in Parentheses))

Full text

ISERP
Working Paper 06-06
Attention Felons
EVALUATING PROJECT SAFE NEIGHBORHOODS IN CHICAGO
ANDREW PAPACHRISTOS
DEPARTMENT OF SOCIOLOGY
CHICAGO
TRACEY MEARES
SCHOOL OF LAW
CHICAGO
JEFFREY FAGAN
S
CHOOL OF LAW
COLUMBIA UNIVERSITY
COLUMBIA UNIVERSITY IN THE CITY OF NEW YORK
SEPTEMBER 2006
PIONEERING SOCIAL SCIENCE RESEARCH AND SHAPING PUBLIC POLICY
INSTITUTE FOR SOCIAL AND ECONOMIC RESEARCH AND POLICY

ATTENTION FELONS
EVALUATING PROJECT SAFE NEIGHBORHOODS IN CHICAGO
1
Andrew Papachristos
Department of Sociology
Chicago
Tracey Meares
School of Law
Chicago
Jeffrey Fagan
School of Law
Columbia University
1
The authors would like to thank the members of Chicago’s PSN taskforce for all their assistance these past
several years. Data was provided by the Chicago Police Department and the Illinois Department of
Corrections. The opinions expressed here are those of the authors and in no way reflect those of the PSN
taskforce members, the City of Chicago, the Chicago Police Department, the Illinois Department of
Corrections, of the Bureau of Alcohol Tobacco, Firearms, and Explosives. Direct all correspondence to Andrew
Papachristos, Department of Sociology, University of Chicago, 1126 E. 59
th
St., Room 307. Chicago, IL
60637. Or via email: andrewp@uchicago.edu
.

Abstract
This research uses a quasi-experimental design to evaluate the impact of Project Safe
Neighborhood (PSN) initiatives on neighborhood level crime rates in Chicago. Four
interventions are analyzed: (1) increased federal prosecutions for convicted felons carrying or
using guns, (2) the length of sentences associated with federal prosecutions, (3) supply-side
firearm policing activities, and (4) social marketing of deterrence and social norms messages
through justice-style offender notification meetings. Using an individual growth curve
models and propensity scores to adjust for non-random group assignment, our findings
suggest that several PSN interventions are associated with greater declines of homicide in the
treatment neighborhoods as compared to the control neighborhoods. The largest effect is
associated with the offender notification meetings that stress individual deterrence,
normative change in offender behavior, and increasing views on legitimacy and procedural
justice. Possible competing hypotheses and directions for individual-level analysis are also
discussed.

2
| 1 | INTRODUCTION
riving down interstate I90, Julien passed a billboard just before exit 14B that read:
"Stop Bringing Guns to Chicago or Go Directly to Jail." Julien had seen the sign
before. In fact, it startled him enough to change his normal routine. Typically,
Julien took a Greyhound bus when transporting the illegally purchased guns he sold. This
time, however, he borrowed a car from a friend. During a phone conversation taped by
federal prosecutors, Julien remarked to a gun customer:
And there was a big ass sign when we was coming last time that said, it said, 'Do not
bring guns into Chicago.' ... I swear to God, G. It was a big ass sign. I don't know if
they did it for us or whatever, G. It is a big ass sign, G, coming from Indiana ... So
what I'm a do, is a, I'm a try to find a ride, man.
Unfortunately for Julien, his alternative plan did not work out. Julien, along with three co-
conspirators, plead guilty to conspiring to sell guns to Chicago gang members.
The billboard was posted by Chicago’s Project Safe Neighborhoods (PSN) program, a
federally-funded initiative designed to bring federal, state, and local law enforcement
together with researchers and community agencies to devise context-specific strategies for
reducing gun violence. In Chicago, this has animated a community-level mobilization of
social and legal institutions to stop the onset and spread of gun violence in targeted high-
crime neighborhoods. Chicago PSN focuses on three broad goals: (1) reduce demand among
young gun offenders, (2) reduce supply by identifying and intervening in illegal gun markets,
and (3) prevent onset of gun violence. Both the demand reduction and prevention strategies
rely on a combination of efforts to increase the perceived costs of illegal gun trafficking and
gun use, and to alter the social norms and preferences within the social networks of young
gang members and other adolescents involved in gun violence. The latter strategy includes
efforts to change the perceived legitimacy of law and legal institutions while simultaneously
changing the perceived likelihood and costs of punishment.
In this study, we use a quasi-experimental design to assess the impact of four of Chicago’s
PSN strategies—increased federal prosecutions for convicted felons carrying or using guns,
lengthy sentences associated with federal prosecutions, supply-side firearm policing that
increased the rate of gun seizures, and social marketing of the deterrence and social norms
messages through offender notification meetings. The results are promising: homicide rates
in the targeted neighborhoods decreased more than 35 percent in the two years after the
program started.
In this paper, we first provide the legislative and programmatic background of the PSN
program. A description of Chicago’s specific PSN strategies comes next. We then turn to an
explanation of the gun crime problem in Chicago to set the stage for a discussion of the
theoretical foundations of strategies developed to address Chicago’s problem. The paper
concludes with a preliminary evaluation and discussion of Chicago PSN to date along with a
discussion of next steps in the research.
D

3
| 2 | POLICY CASCADES AND ANTECEDENTS OF PROJECT SAFE
NEIGHBORHOODS
hicago’s PSN initiative is part of a nationwide PSN program that establishes a
“comprehensive and strategic approach to reducing gun crime.”
1
In 2002, Congress
allocated more than 1.1 billion dollars among the 94 federal court districts
throughout the nation specifically to develop PSN strategies to fit within local legal contexts.
In each district, an interagency taskforce overseen by the United States Attorney and
comprised of local, state and federal law enforcement agencies was directed to assess the main
factors driving gun crime in their jurisdiction and then to devise context-specific strategies to
address each area’s “gun problem.” Notably, according to national program dictates, each
district taskforce was urged to network with community partners and researchers in addition
to law enforcement agencies.
One way to understand the impetus behind the national PSN initiative is as the result of a
“policy cascade”
2
in which the public discourse around a particular problem, in this case gun
violence, intersects with a salient policy initiative against the background of a political
landscape that is receptive to the widespread promotion of the relevant policy initiative.
PSN emerged from public discourse of the “gun problem” amidst a tough-on-crime political
backdrop. This discourse was advanced by two salient policy precursors to PSN: Boston’s
Project Ceasefire and Richmond’s Project Exile, each of which was created in a political
landscape receptive to tough demand-side punishment of gun offenders.
Operation Ceasefire was a problem-oriented policing intervention focused on reducing
youth homicide and gun violence in Boston (see, Braga et al. 2001).
3
Project Exile was
started as a collaborative effort to prosecute federally all felon-in-possession, drug/gun, and
domestic/gun cases in Richmond.
4
Both programs were highly touted in the media for their
impact on the dropping crime rates. In fact, the drop in youth homicides in Boston was so
dramatic that it came to be known in the popular press as the “Boston Miracle.”
5
In
1
According to its mission statement: “The goal is to take a hard line against gun criminals through every
available means in an effort to make our streets and communities safer. Project Safe Neighborhoods seeks to
achieve heightened coordination among federal, state, and local law enforcement, with an emphasis on tactical
intelligence gathering, more aggressive prosecutions, and enhanced accountability through performance
measures” (http://www.psn.gov/).
2
Here we mean to borrow a page from Timur Kuran and Cass Sunstein (1998).
3
A multi-agency working group analyzed police intelligence and determined that approximately 1,300 gang
members (less than 1 percent of the youth population under 24) were responsible for 60 percent of all juvenile
homicides in Boston and that most of these homicides occurred in a geographically concentrated inter-gang
retaliations. To counteract the violence, the working group created a “pulling levers” strategy that concentrated
intervention and deterrence efforts law enforcement and community outreach workers directly on those gangs
and gang members responsible for gun violence. In a series of meetings with different gangs, the Boston group
told offenders of their targeted enforcement efforts and made it clear that should a violent episode occur, they
would “pull every lever” available to come down hard on the gang itself, apprehend the offenders, and prosecute
accordingly.
4
Project Exile efforts also included enhanced training for law enforcement and community organizations and a
media campaign touting the “get tough on gun crime” message.
C

Frequently asked questions

Q1. What are the contributions in this paper?

This research uses a quasi-experimental design to evaluate the impact of Project Safe Neighborhood ( PSN ) initiatives on neighborhood level crime rates in Chicago. Using an individual growth curve models and propensity scores to adjust for non-random group assignment, their findings suggest that several PSN interventions are associated with greater declines of homicide in the treatment neighborhoods as compared to the control neighborhoods. 

Q2. Why do the criminally prone pay more attention to the costs of doing crime?

Because the criminally prone potentially will be subject to legal sanctions, they pay closer attention to the costs of doing crime, assuming that they have access to information about higher potential costs with no offset from higher potential crime payoffs. 

Q3. Why do the authors want to collect primary data on such matters directly from offenders?

Because those actively involved in using, buying, or otherwise involved with guns possess the most knowledge of the problem, the authors intend on collecting primary data on such matters directly from offenders. 

Q4. How much money was funneled towards non-law enforcement issues?

More specifically, $130 million was funneled towards non-law enforcement issues, 126 million towards the hiring of federal prosecutors, and 280 million towards state, local, and community initiatives (Ludwig 2004). 

Q5. Why did the researchers design this study as a quasi-experimental panel model?

8| 5 | RESEARCH DESIGNecause political and logistic factors hindered the establishment of a true randomized experiment, the authors designed this research as a quasi-experimental panel model measuring treatment effects and using a near-equivalent control group (Shadish, Cook andCampbell 2002). 

Q6. What is the main argument for the robustness of their findings?

The fact that their findings hold under functional forms—including fixed effects OLS methods—supports the robustness of their findings. 

Q7. Why are law abiders immune to the threat of punishment?

In this account, it is the law abiders who are, in a sense, immune to the threat of sanction, but not because they are impulsive and without self-control; rather, it is because law-abiders are highly unlikely to offend in the first place due to their internalized commitment to compliance. 

Q8. What is the effect of the forum on the outcome variables?

The authors hypothesize that increasing the percentage of offenders in the target areas who have attended a forum should have a negative relationship on the outcome variables. 

Q9. What factors are considered to be important in the decline of homicide rates?

Preliminary analysis by the authors suggest that when controlling for the social, demographic, and PSN factors describe here, no statistically significant effect in the declining homicide rates during the observation period can be attributable purely to the presence of Operation Ceasefire in the PSN treatment area. 

Q10. What is the strongest PSN dimension associated with declining beat-level homicide rates?

23TABLE 4 also shows that the strongest PSN dimension associated with declining beat-level homicide rates is the percent of offenders in a beat who attend a forum (β = -0.144 , p = 0.003). 

Q11. What is the significance of the pre-intervention spike in homicides?

In particular, the pre-intervention spike in homicides suggests that the observed decline might be regression towards the mean or simply part of the nation-wide declining crime trend (Ludwig 2004).