www.policyexchange.org.uk

From the Ground Up

Policy Exchange
Clutha House
10 Storey’s Gate
London SW1P 3AY

	

£10.00
ISBN: 978-1-907689-12-3

Policy Exchange	

From the Ground Up: Promising criminal justice
projects in the US and the UK examines successful
demonstration projects in the UK and the US that
are attempting to reduce crime, drug use and
incarceration, among other challenging goals. The
report identifies a selection of innovative, groundlevel experiments in policing, probation, courts and
crime prevention that have had a real impact. In this
report the founders and lead practitioners share
their experiences of creating innovative projects
and in reviewing the projects the author identifies
key practical lessons for how to successfully plan,
implement and sustain new criminal justice
endeavours.

From the
Ground Up
Promising criminal justice projects
in the US and the UK

Aubrey Fox and Gavin Lockhart
Edited by Blair Gibbs

From the
Ground Up
Promising criminal justice projects
in the US and the UK
Aubrey Fox and Gavin Lockhart
Edited by Blair Gibbs

Policy Exchange is an independent think tank whose mission is to develop and promote new policy
ideas which will foster a free society based on strong communities, personal freedom, limited
government, national self-confidence and an enterprise culture. Registered charity no: 1096300.

Policy Exchange is committed to an evidence-based approach to policy development. We work in
partnership with academics and other experts and commission major studies involving thorough
empirical research of alternative policy outcomes. We believe that the policy experience of other
countries offers important lessons for government in the UK. We also believe that government has
much to learn from business and the voluntary sector.
Trustees
Daniel Finkelstein (Chairman of the Board), Richard Ehrman (Deputy Chair), Theodore Agnew, Richard
Briance, Simon Brocklebank-Fowler, Robin Edwards, Virginia Fraser, George Robinson, Robert Rosenkranz,
Andrew Sells, Tim Steel, Alice Thomson, Rachel Whetstone and Simon Wolfson.
The Centre for Justice Innovation is an independent agency that works to improve how the criminal
justice system functions in England and Wales. A project of the Center for Court Innovation, a New Yorkbased non-profit organisation, the Centre seeks to promote thoughtful criminal justice reform by
improving the implementation, evaluation, and dissemination of demonstration projects.

© Policy Exchange 2011
Published by
Policy Exchange, Clutha House, 10 Storey’s Gate, London SW1P 3AY
www.policyexchange.org.uk
ISBN: 978-1-907689-12-3
Printed by Heron, Dawson and Sawyer
Designed by SoapBox, www.soapbox.co.uk

Contents

1
2
3

About the Authors
Acknowledgements
About this Paper
Introduction

4
5
6
12

Getting Started: Planning a Demonstration Project
Getting it Right: Challenges of Implementation
Keeping it Going: Sustaining and Spreading
the Best Ideas

16
24
34

Conclusion

41

About the Authors

Aubrey Fox is the Director of the Centre for Justice Innovation, an
independent agency that works to improve how the criminal justice
system operates in England and Wales. The Centre is a project of the
New York-based Center for Court Innovation, a public-private
partnership that seeks to reform the justice system through
demonstration projects, original research, and technical assistance to
frontline practitioners. At the Centre, Aubrey was the founding Project
Director of Bronx Community Solutions, a one-of-its-kind initiative
launched in January 2005 that seeks to meet the ambitious goal of
changing a large and tradition-bound public agency's approach to
low-level crime. Aubrey is the co-author of Trial and Error in Criminal
Justice Reform: Learning from Failure (Urban Institute Press, 2010).
Gavin Lockhart. Having graduated from Edinburgh University with a
First Class Masters Degree and worked as a management consultant,
Gavin was recruited by Policy Exchange to help set up research units
on crime reduction and healthcare policy. He subsequently worked on
criminal justice policy for the Conservative Party and in the Downing
Street Policy Unit.
Blair Gibbs is the Research Director for Crime and Justice at Policy
Exchange. He joined the think-tank in June 2010 and most recently
co-authored the report Cost of the Cops (September 2011). He was the
editor of the report – authored by William Bratton – Fighting Crime
and Disorder: Policing Experience from America (April 2011). Blair is a regular
commentator on current policing and criminal justice issues in the
United Kingdom, and prior to joining Policy Exchange, Blair worked
as senior policy advisor to the Policing and Criminal Justice Minister,
Rt Hon Nick Herbert MP and as a director of the TaxPayers’ Alliance.

Acknowledgements

Policy Exchange and the Centre for Justice Innovation would like to
thank the Hadley Trust for their generous financial support of this
project. The authors would also like to thank Greg Berman, the
Director of the Center for Court Innovation in New York, for his help
with this document, as well as Karen Sosa at Policy Exchange and
Jessica de Grazia for suggesting the title of the paper. In addition, the
authors would like to thank the following people for their advice
and assistance with this project:
 Amy Ellenbogen, Center for Court Innovation
 Anton Shelupanov, Young Foundation
 Chris Watler, Center for Court Innovation
 Christian Steenberg, Greater London Authority
 Clive Martin, Clinks
 Diana Barran, Coordinated Action Against Domestic Abuse
 Evan Jones, St. Giles Trust
 Joe Mitton, Greater London Authority
 Mark Ferrante, New York City Department of Probation
 Merril Stevenson, The Economist
 Monica Sharma, London Criminal Justice Partnership
 Paul Pandolfo, Greater Manchester Probation Trust
 Robert Patrick, Young Foundation
 Sally Dickinson, Magistrates Association
 Sally Lewis, Avon and Somerset Probation Trust
 Vincent Schiraldi, New York City Department of Probation

About this Paper

This paper summarises the experiences of 10 innovative criminal
justice projects across the United Kingdom and the United States. The
projects, listed in order of introduction in the paper, are as follows:

Neighborhood Opportunity Network, NewYork City Department of Probation
The Neighborhood Opportunity Network (NeON) is a network of
community organisations, government agencies, local businesses,
and residents focused on connecting probation clients to
opportunities, resources and services in their home neighbourhood.
The Department of Probation is launching five NeONs in
neighbourhoods where large numbers of people on probation reside.
NeONs are a central element of Mayor Bloomberg’s Young Men’s
Initiative, which is designed to help black and Latino youth achieve
their professional, educational, and personal goals. Specially-trained
and culturally competent probation personnel will staff the NeONs.
NeON staff will share office space with community-based
organisations to provide a client-friendly environment in which a
wide range of services and supports can be accessed.1

1 neighborscampaign.word
press.com/category/wheaton
/tuesdays-together/

SOS Gang Anti-Violence Initiative, Crown Heights Community Mediation
Centre, Center for Court Innovation, New York, NY
The Save Our Streets (SOS) project was launched in Crown Heights,
Brooklyn in 2010, with support from the United States Department
of Justice. Based on a programme created by the Chicago Project for
Violence Prevention, SOS has three basic objectives: changing
community norms about violence, educating the community about
the costs of violence, and providing on-the-spot alternatives to
violence. A key part of the model is putting credible messengers to

About this paper

| 7

work as outreach workers and violence interrupters – ex-offenders
and others from the neighbourhood with direct knowledge of and
connections to gang members and other perpetrators of violence. In
2010 the project started to get traction – and between the first half
of 2010 and 2011 there was a 54% decrease in the number of
shootings in Crown Heights.
For more information, see: https://www.soscrownheights.org
Project Pegasus,Violence Reduction Unit, Glasgow, Scotland
Project Pegasus seeks to address the nexus between binge drinking and
violent crime by introducing continuous monitoring of alcohol
consumption by certain offenders via an ankle bracelet or electronic
tag. According to project plans, two groups will be outfitted with
bracelets – those sentenced to Community Payback Orders and selected
prisoners released from jail early. Participants who breach their orders
will be required to attend court for their violation of license conditions,
with a custodial sentence as a possible sanction after their third offense.2
IMPACT, Avon and Somerset Probation Trust and Avon and Somerset
Constabulary, Bristol, England
Launched in 2008, IMPACT provides enhanced support and
supervision to 800 offenders in Bristol with a history of re-offending.
The IMPACT team is made up of police officers, probation staff,
youth workers, and representatives of third-sector organisations,
who meet regularly to review cases under their supervision. In
addition to ensuring that offenders abide by the terms of their
community sentences and stay crime-free, the team offers them the
help they need – drug treatment services, accommodation, and job
training – to reduce the harm caused to the community. When it was
introduced to Bristol as a trial in 2008, serious acquisitive crime
(which includes domestic burglary, theft from motor vehicle, theft
of motor vehicle and robbery) dropped by 28% in just two years.3
For more information, see: http://www.bristol.gov/uk/node/5647

2 http://www.dailyrecord.
co.uk/news/scottishnews/2011/08/21/gangbuster
s-to-force-thugs-to-wearbooze-band-if-they-want-to-st
ay-out-of-jail-8690823360626/
3 http://www.avonand
somerset.police.uk/informati
on/documents/cache/PDF/
Document5553_442066.pdf

8 | From the Ground Up

Manchester Intensive Alternatives to Custody Project, Manchester Probation Trust,
Manchester, England
In 2009, the Manchester Probation Trust launched the Intensive
Alternatives to Custody (IAC) project. One of seven pilot projects
funded by the Ministry of Justice, IAC is aimed at offenders who
would otherwise receive a prison sentence of less than 12 months.
IAC’s core goal is to come up with a sentencing alternative for
prison-bound offenders that appeals to magistrates and judges. It
does this in two ways: first, by creating a customised community
disposal option, and second, by dedicating additional resources
(including employment-focused mentoring and family counselling)
to ensuring that individuals on IAC complete the order. The IAC order
provides additional services and enhanced supervision for young
adult offenders. Since the programme started, reoffending rates have
dropped and over a quarter of unemployed offenders on IACs have
found full-time work.4 In a sign of the success of the approach,
probation has reallocated resources and obtained commitment from
local partners and the National Offender Management Service to
continue the programme despite the withdrawal of central
government pilot funding.
For more information, see: https://www.gm-probation.org.uk/what-we-do/intensivealternative-to-custody.php

4 http://www.gmprobation.org.uk/what-wedo/intensive-alternative-tocustody.php

Project Daedalus, London Youth Reducing Re-offending Programme, Feltham
Young Offenders Institution, Greater London Authority and London Criminal
Justice Partnership
Established in September 2009, Project Daedalus seeks to help young
offenders reaching the end of their custodial sentence reintegrate into
their home communities. Individuals enrolled in Daedalus are placed
onto an enhanced resettlement regime on a separate unit at the
Feltham Young Offenders’ Institution and given help with education
and job training services. Young offenders in the unit are able to gain
work experience with local employers on day release, and so-called

About this paper

| 9

“resettlement brokers” (provided by a consortium of charities)
support each young person while incarcerated and post-release.
For more information, see: http://www.london.gov.uk/priorities/crime-communitysafety/time-action/project-daedalus
Harlem Parole Reentry Court, Harlem Community Justice Center, Center for
Court Innovation, Harlem, New York
Over 2,200 people are released from prison on parole supervision to
upper Manhattan each year, and in East Harlem, one in 20 men along
a seven-block corridor from 119th Street to 126th Street have spent
time in prison.5 The Harlem Parole Reentry Court was established in
June 2001 in the heart of this area to help parolees re-settle into
their home communities. Participants are required to return to the
Harlem Parole Reentry Court frequently to meet with case managers
and parole officers and to appear before an administrative law judge,
who closely monitors their compliance with court orders. A March
2010 study documented that the Reentry Court reduced re-arrests
and re-convictions among programme participants.6
For more information, see: http://www.courtinnovation.org/project/parole-reentry-court
Integrated Domestic Violence Advisors and Multi-Agency Risk Assessment
Conferences, Coordinated Action Against Domestic Abuse (CAADA), Bristol,
England
An average of two women are killed every week in the United
Kingdom as a result of domestic abuse.7 According to the Department
of Health, at least 750,000 children a year witness domestic violence.8
To address this problem, government officials in England and Wales
have recently launched Multi-Agency Risk Assessment Conferences
(MARACs), which bring together the police, probation, health, and
local charities to create safety plans for high-risk victims. Independent
Domestic Violence Advisors (IDVAs) support the MARACs. MultiAgency Risk Assessment Conferences are voluntary meetings where
information is shared on the highest risk domestic abuse cases between

5 The Economist, “They all
come home: Effective reentry programmes can keep
ex-prisoners out of jail”,
20 April 2011,
http://www.economist.com/n
ode/18587528
6 In summary Re-entry Court
parolees were less likely to be
re-arrested and were less likely
to be re-convicted, and the
effects were significant at one,
two, and three years (43%
versus 53% at three years)
Zachary Hamilton, “Do Reentry
Courts Reduce Recidivism?
Results from the Harlem
Parole Reentry Court”, Centre
for Court Innovation, March
2010, http://173.231.132.82/
sites/default/files/Reentry_Ev
aluation.pdf
7 http://www.caada.org.uk
/news/factsandstats.htm
8 “Alesha Dixon presents BBC
show on domestic violence”,
The Guardian, 16 November
2010, http://www.guardian.
co.uk/society/2010/nov/16/al
esha-dixon-documentarydomestic-violence

10 | From the Ground Up

local public and voluntary agencies, such as health agencies, the police
and Independent Domestic Violence Advisor services. After sharing
relevant information about a victim, the meeting then discusses
options for increasing their safety, and turns this into an action plan.
The non-profit Coordinated Action Against Domestic Abuse provides
training, education, and support for IDVAs and other domestic violence
advocates. The approach has appeared to pay off: in instances where
IDVAs and MARACs have intervened, 60% of domestic violence victims
report no further violence.9
For more information, see: http://www.caada.org.uk/
HOPE Probation, Hawaii State Judiciary, Oahu, Hawaii
Hawaii First Circuit Judge Steve Alm launched HOPE in 2004 aimed
at probationers at risk of violating the terms of their probation
mandate. Under the previous regime, breaches were not always
enforced and usually a probation officer’s recommendation was to
revoke probation and sentence offenders to long terms in prison.
With HOPE (Hawaii’s Opportunity Probation with Enforcement),
probationers are regularly drug tested, and if they fail their tests, are
given an immediate and certain, but short, two-day jail sentence as
a sanction. Breaches result in swift, certain sanctions that escalate if
transgressions persist, but punishment is minimised. Research shows
that the programme has significantly reduced crime and probation
revocations, and therefore prison costs, whilst reducing failed drug
tests and missed probation appointments. Today, nearly one in five
felony probationers in Oahu are supervised under HOPE.
For more information, see: http://www.pewcenteronthestates.org/report_detail. aspx?
id=56832
9 “Saving lives, saving money:
MARACs and high-risk
domestic abuse”, CAADA,
http://www.caada.org.uk/Res
earch/Saving_lives_saving_m
oney_FINAL_VERSION.pdf

Justice Reinvestment, Kentucky Legislature and Pew Center on the States,
Kentucky, USA
In 1990 the corrections spending in Kentucky amounted to $140
million. 20 years later that amount had grown by 214% to $414 million

About this paper

whilst recidivism rates grew beyond 1990 levels. Over the last
decade, Kentucky has seen a 45% increase in its prison population,
an unsustainable rise given significant budget shortfalls. Working
with the Pew Charitable Trust’s Public Safety Performance Project,
state officials crafted the 2011 Public Safety and Offender
Accountability Bill, which analysts believe will save the state $422
million in reduced incarceration costs over the next decade. A portion
of those savings will be reinvested in strengthened community-based
alternatives, including strengthening probation and parole as well as
programmes for substance-abusing offenders.
For more information, see: http://www.pewcenteronthestates.org/report_detail. aspx?
id=61357

| 11

Introduction
10 Public confidence in the
criminal justice system (CJS) is
measured by the proportion
of people in the British Crime
Survey (BCS) who say that
they are “fairly/very
confident” that the CJS is
effective in bringing offenders
to justice. The year ending
March 2010 reported the
proportion of people
confident in the CJS at 41%
11 The disturbances shined a
light on persistent problems
such as the need to more
positively engage and ensure
accountability among a small
group of young, mostly male,
chronic offenders. One in four
people charged over the riots
in England had committed
more than 10 past offences.
See Statistical bulletin on the
public disorder of 6- 9 August
2011 Ministry of
Justice Statistics bulletin
12 Ministry of Justice
Business Plan, Ministry of
Justice, November 2010,
http://www.number10.gov.uk
/wp-content/uploads/MOJBusiness-Plan.pdf
13 Travis, Alan, "Quango cut:
Controversy over scrapping of
Youth Justice Board". The
Guardian, 15 October 2010
14 Chambers, Max, Carter But
Smarter: Transforming
offender management,
reducing reoffending, Policy
Exchange, http://www.policy
exchange.org.uk/images/publ
ications/pdfs/Carter_But_Sm
arter_-_Nov__10.pdf

This is a challenging time for criminal justice policymakers in
England and Wales. There is significant pressure to cut expenditure
while simultaneously reducing crime and improving levels of public
confidence in justice.10 At the same time, policymakers have been
confronted with demands arising from unexpected events, such as
the civil disturbances across England in August 2011.11
In recent years, policy makers in England and Wales seeking to
improve public service delivery have tended to gravitate towards
formulating large-scale, structural changes to the criminal justice
system. In the last fifteen years, this has included the creation, in
1998of the Youth Justice Board (an agency dedicated to preventing
offending by young people under the age of 18), the establishment
in 2004 of the National Offender Management Service (which
merged prison and probation into a single national agency), and the
creation of the Ministry of Justice (responsible for the justice system
and for some areas of constitutional policy) in 2007. A more recent
example is the Coalition government’s decision to open up the
corrections marketplace and allow private companies to compete for
areas of business previously reserved for the public sector (such as
the delivery of community sentences or offender rehabilitation
schemes) and to do so with large, regional contracts.12
Yet the track record of large-scale, structural change is mixed at
best. For example, the Youth Justice Board is being abolished after
little more than a decade, with its functions reabsorbed into the
Ministry of Justice.13 As Policy Exchange explored in their 2010
report Carter But Smarter, in its short life NOMS has experienced its
own difficult growing pains, failing to deliver the integrated endto-end offender service that was promised.14 Perhaps more

Introduction

| 13

fundamentally, by its very nature, national policy change is
enormously complicated, time-consuming and fraught with all
sorts of unintended consequences.
Top-down policy changes in the crime and justice space also
reinforce a tired national debate – still prevalent in the United
Kingdom – that too often polarises into a
conflict between “soft” and “hard”
By its very nature, national
approaches to criminality. In a political and
policy change is enormously
legislative context dominated by national
complicated, time-consuming
policies, it is harder to advance the idea that
progress often means being smart on crime, and fraught with all sorts of
unintended consequences
not hard or soft.
There is another, much less visible,
approach to justice reform, however, which is to start small,
launching pilot projects (often with a minimum of publicity) in
response to specific, local, criminal justice problems, and then
See “Drug court to offer
patiently build them over time. And in fact, over the last two 15
addicts help”, BBC News
decades, England and Wales have experienced a raft of website, 9 April 2009,
http://news.bbc.co.uk/1/hi/w
demonstration projects – everything from drug courts15 to ales/7988347.stm
intensive alternative-to-custody project16 to new models of 16 Intensive Alternative to
integrated offender management.17 Furthermore, in recent weeks, Custody, Ministry of Justice
Website, http://www.gmthe Ministry of Justice has launched a number of pilot projects in probation.org.uk/what-we-do
areas that range from restorative justice to mental health to new /intensive-alternative-tocustody.php
financial incentives for local authorities to reduce the use of youth
17 Home Office, “Integrated
custody.18
Offender Management Key
Principles”, March 2011,
This paper takes a closer look at this ground-up phenomenon. http://www.homeoffice.gov.u
Rather than focus on large-scale changes to the criminal justice k/publications/crime/reducingreoffending/IOM-Key-Principles
system, it seeks to identify best practice in pilot criminal justice -Guidance?view=Binary
projects, both in England and Wales and the in United States. The 18 Ministry of Justice,
goal is to fill gaps of understanding and awareness among criminal “Breaking the Cycle: Effective
Punishment, Rehabilitation
justice policymakers and practitioners about what makes and Sentencing of Offenders”,
demonstration projects work and how to improve the chances that http://www.justice.gov.uk/co
nsultations/docs/breakingthey succeed in future.
the-cycle.pdf

“

”

14 | From the Ground Up

19 The two-year, £11 million
Diamond District Initiative was
a multi-agency approach
designed to reduce reoffending rates among
short-term, unlicensed
prisoners (e.g. released
without any additional formal
criminal justice supervision) in
six London boroughs. The
project was terminated after
an evaluation showed that it
had no impact on re-offending
rates, although some of the
boroughs involved in the work
have continued versions of
the initiative. Martin
Bentham, “£11 million
initiative to help freed
convicts fails to cut
reoffending rate,” London
Evening Standard,
18 April 2011

Criminal justice demonstration projects can take a variety of
different forms. The lead actor can vary from project to project:
experiments can be initiated by police, prosecutors, probation and
parole staff, judges, pre-trial agencies, community groups,
corrections departments and sometimes dynamic local citizens. Some
deal with hundreds of participants; others just a handful. And the
underlying problem to be addressed can be anything from minor
youth offending to serious domestic violence cases.
We have gathered ten examples of promising demonstration
projects from across the United Kingdom and the United States, and
asked them what made their projects work, what challenges they
face, and how those challenges can be overcome.
In writing this paper, our working hypothesis is that there is
much more innovation occurring at the local level than is
currently understood by national policymakers. In fact, the
examples cited herein represent only a fraction of the innovative
projects that are currently operating in both the United Kingdom
and the United States. Some of the most positive of these
programmes have been translocated between the US and UK. There
are already well known examples that have crossed the Atlantic:
for example, the recent August Riots in England highlighted the
success of the anti-gang strategy in Glasgow – a project that began
as ‘Operation Ceasefire’ in Chicago and Boston.
Almost all demonstration projects face similar conceptual and
operational challenges. What follows is an attempt to highlight a
handful of lessons from pilot projects at various stages of
development. We have chosen to focus on three distinct phases –
planning, implementation, and sustainment.
Finally, we are not saying that change of the kind described above
is always the right approach. Indeed, there is no guarantee that
demonstration projects will succeed, as the recent experience of
Diamond Districts in London shows.19 Nonetheless, innovation from
the ground up has an important contribution to make to the world

Introduction

of criminal justice. By shining a light on demonstration projects, this
paper seeks to both encourage innovation by practitioners at the
grassroots and to educate national policymakers and the media about
an approach to reform that typically receives little attention.

| 15

1. Getting Started: Planning a
Demonstration Project

20 Michael Barbaro and
Fernanda Santos, “Bloomberg
to Use Own Funds in Plan to
Aid Minority Youth,” The New
York Times, 3 August 2011

The front-page article of the 3rd August 2011 edition of the New York
Times read “Bloomberg to Use Own Funds in Plan to Aid Minority
Youth”.20 The article described a wide-ranging, $130 million
programme (which included a $30 million personal contribution
from New York City Mayor Michael Bloomberg) to try to improve
the life chances of over 300,000 young black and Hispanic men.
One component of the project was a plan, announced by the New
York City Department of Probation, to open satellite offices in five
neighbourhoods with high crime rates and high numbers of
probationers. Dubbed the Neighbourhood Opportunity Network (or
“NeON”), the offices, which will be operated in partnership with
community organisations, intend to offer enhanced mentoring,
education, and job training services to young black and Hispanic
men under probation supervision. Not every demonstration project
will start with such high-level support or a front-page article in a
major newspaper. Yet the principles of good planning remain the
same, no matter the origins of a project.
The planning stage is often the most neglected – and yet arguably
the most important – part of any demonstration project. In an
understandable rush to get projects off the ground, planners
sometimes fail to think about important building blocks such as
obtaining local support for their efforts or customising projects to
meet local conditions. They also, at times, fail sufficiently to prepare,
which involves zeroing-in on a specific set of solvable problems by
combing through data and doing extensive analysis to make sure
they are getting the facts right. Finally, the planning stage is often a

Getting Started: Planning a Demonstration Project

| 17

good moment to engage a broader set of stakeholders and potential
critics of a project to test concepts and to see if the potential
objections can be addressed – or at least partially mollified – ahead
of time.

Importance of local knowledge
In the case of the Save Our Streets (SOS) programme in Crown
Heights, Brooklyn, community engagement is embedded into the
DNA of the project. In fact, the inspiration for SOS came from a local
resident, a mother whose son had been shot and killed. “She had been
going door to door in the community, asking for help to address the community’s violence
problem” recalled Amy Ellenbogen, the Director of the Crown Heights
Community Mediation Center, which operates SOS: “From that point on,
we were committed to figuring out a way to address the community’s violence
problem.”21
The Crown Heights Community Mediation Center was created in
1998 to serve as a resource for conflict resolution in a diverse
neighbourhood notorious for three days of rioting in 1991. The
centre operates out of a small storefront office in central Brooklyn.
In addition to offering help to residents on a walk-in basis (e.g. help
finding a job, housing, or legal assistance), the centre conducts
regular workshops on conflict resolution for schools and churches
and runs a series of youth programs.
The opportunity to address gang and gun violence came in 2010,
when the Mediation Center launched SOS with funding from the
United States Department of Justice. A replication of a project
developed in Chicago called CeaseFire, SOS employs ex-offenders as
credible “violence interrupters” who work to calm conflict before it
escalates out of control. The project also organises community events
on street corners where shootings have taken place, and sends out
violence interrupters to reach out to local residents at risk of
becoming a victim of violence – or a perpetrator: “We find out what

21 Greg Berman and Emily
Gold, “From Chicago to
Brooklyn: A Case Study in
Program Replication,” Bureau
of Justice Assistance, on file
with authors

18 | From the Ground Up

their needs are – anything that’s going to shift them to a different mentality toward
gun violence,” said Lavon Walker, one of the outreach workers. “We
become like their bigger brothers, even closer than their fathers.”22
Despite being based on an already established project, project
planners immediately understood that SOS had to be customised to
the local environment. The CeaseFire model depends on detailed
knowledge of local street culture, but this landscape is different in
every city. “Many New York City gangs are changing
their organizational structure, fragmenting into smaller
Understanding the local
cliques without the traditional rules and governing
context in which a project
constitution,” said Ellenbogen.23 This theme
operates is critical – even for a
was echoed by Deputy Director Ife Charles:
project that is replicating an
“Our community violence doesn’t stem from a simple
existing initiative
red versus blue divide. We’ve got red versus red and blue
versus blue problems, too.”24 In other words, SOS
staff members have worked hard to understand where potential
conflicts might erupt on a block-by-block level.
Project planners also wrestled with the strategy employed in
Chicago of conducting regular home visits. “The home visit approach really
concerned us at first,” Ellenbogen said. “New York has more of a street scene, so
we weren’t sure it would even be effective.We also couldn’t imagine becoming comfortable
with sending programme staff into people’s apartments because of the risk of getting
swept up in police raids.”25 Over time, however, Ellenbogen and her team
grew more comfortable with the approach as SOS developed closer
relationships with specific buildings and blocks. Another factor that
22 Tim Stelloh, “Gunfire Will
No Longer Be Met With
helped
was hiring staff members from local housing projects. The
Silence,” The New York Times,
10 December 2010
outreach workers have a caseload of 15 to 20 participants, each of
23 Greg Berman and Emily
whom are attracted by what SOS is offering: help in finding a job or
Gold, “From Chicago to
getting a high school equivalency diploma, for instance.
Brooklyn: A Case Study in
Program Replication,” Bureau
More broadly, SOS benefited from the over decade-long
of Justice Assistance, on file
investment made by the Crown Heights Community Mediation
with authors
Center in Crown Heights, which helped combat the perception that
24 Ibid.
25 Ibid.
SOS was being imposed by outsiders without any understanding of

“

”

Getting Started: Planning a Demonstration Project

| 19

the local community. “Many people expressed concerns about SOS being the ‘flavour
of the month’, but when they heard the Mediation Center was involved, it helped us get
access to people and places quickly”, observed Ellenbogen. “What really helped
us was having the trust of the community and having positive relationships that spanned
a long time.”26
The key insight here is that understanding the local context in
which a project operates is critical – even for a project that is
replicating an existing initiative.

Defining the problem
The Violence Reduction Unit (VRU) sits in a nondescript police
building in Glasgow. Although the unit is formally attached to the
Strathclyde Police Force, it has a national mission to investigate, and
propose solutions to, the causes of violence in Scotland. The unit is
run by John Carnochan, a police officer for over 30 years who spent
most of his time working as a detective, and Karyn McCluskey, a
psychologist who was the Head of Strathclyde’s intelligence analysis
unit before joining VRU.
VRU is perhaps best known for a successful anti-gang violence
project called the Community Initiative to Reduce Violence. The
programme, which was launched in 2008, is an adaptation of a pilot
project developed in Boston by the American academic David
Kennedy (and subsequently replicated in a number of cities across
the United States). Initially piloted in Glasgow’s East End, it was
extended to the north of the city in 2009 and then to the whole of
Scotland. The pilot has achieved some impressive results, including
a 50% reduction in violent crime among gang members who took
part in the programme,27 and has received flattering attention from
the media and elected officials, including a mention by the Prime
Minister.28
VRU has more recently turned its attention to another thorny
problem: the connection between binge drinking and violent crime.

26 Correspondence with
authors, 14 October 2011
27 See http://www.actionon
violence.co.uk/CIRV
28 Norman Silvester, “Riots:
David Cameron’s tough
American supercop looks to
Scotland’s gangbusters for
help,” The Daily Record, 14
August 2011

20 | From the Ground Up

29 Interview with Karyn
McCluskey, 27 September
2011
30 Ibid.

A team of researchers spent four months conducting extensive data
analysis and interviewing a wide range of people, including judges,
police officers, emergency room physicians and academic experts.
VRU’s strategy was to use a blend of quantitative and qualitative
analysis to get at the issue from as many angles as possible. “I’m a
strong believer in data,” said McCluskey, “but sometimes when you’re just looking
at stats, you don’t get a complete picture.”29
From the beginning, the researchers found strong indications of
a link between excessive drinking and crime. For example,
researchers unearthed a study of prisoners processed by Strathclyde
Police in 2007 that found that approximately two-thirds of those
arrested at the scene of a crime were under the influence of alcohol
at the time of arrest. This hypothesis was confirmed when the team
went through police records of homicides committed in Strathclyde
and found that alcohol had been consumed in 80% of cases.
As they investigated further, the story got more complicated. In
informal interviews, they learned that the nature of the drinkingviolence connection had changed over the last decade. “Most of our
murders used to be outside the house” said McCluskey. “Now, with the smoking ban
[Scotland banned smoking in bars and restaurants in 2006], rising alcohol prices, and
cable television, people are drinking more in their homes.” For VRU, this raised a
challenging question: ‘How do you police inside the house?’30 This
was a particularly problematic issue with regard to domestic
violence. VRU researchers discovered that alcohol had been
consumed in close to three-quarters of all domestic violence cases.
They also knew that breaches of so-called “stay-away” orders –
which compel domestic violence offenders to stay away from their
spouses or partners – were common. Perhaps, they hypothesised,
excessive alcohol consumption was fuelling both the original acts
of violence and the worrying tendency of offenders to thumb their
noses at court orders.
Having the analysis complete, VRU proposed a potential solution:
requiring that certain groups of offenders with known drinking

Getting Started: Planning a Demonstration Project

| 21

problems, such as individuals about to be released from prison, those
given a community-based sentence as an alternative to custody, or
domestic violence offenders, be fitted with a bracelet that detects any
intake of alcohol. Having a drink would immediately trigger an
alarm and result in sanctions being imposed on the offender. The
project, known as Pegasus, is due to be trialled later this year.

A focus on outcomes
Time spent defining the problem is a crucial investment in the
planning stage of any project. When setting up Project NeON,
officials at New York City’s Department of Probation used a similar
methodology. The basis of NeON is the belief that to be successful,
probation had to get out of what New York City Probation
Commissioner Vincent Schiraldi termed “the bunker mentality” of many
probation offices. “Vinny’s analysis was that we had to get probation into
neighbourhoods [where probationers live] and away from the sterile and semi-hostile
environments they are in now,” said Mark Ferrante, Schiraldi’s Senior Policy
Advisor.31
The department’s next step was to analyse their caseload and map
the neighbourhoods where probationers live. Not surprisingly, they
found that they were clustered in certain communities throughout
the city. Officials ended up picking five sites as NeON pilots, basing
their decision on a combination of the data analysis and their sense
of which neighbourhoods had a critical mass of resources (such as
non-profit and community-based organisations with a strong track
record of success) to draw upon.
After identifying the NeON pilot neighbourhoods, the project
planning team next developed a strategy that would allow them to
test ideas about how to better engage their target population of
young black and Hispanic men. First, they are working to identify
satellite offices, or “hubs,” in each of the five neighbourhoods
identified in the planning process. Second, they plan to co-locate

31 Interview with Mark
Ferrante, 27 September 2011

22 | From the Ground Up

third sector organisations at each of the five hub offices with the goal
of expanding job training and educational opportunities for
probationers. Finally, the Probation Department has set aside grant
funds for third-sector organisations that want
to work side-by-side with probation officers
Demonstration projects
at the hub offices.
often struggle when they have
The examples of Project Pegasus in
not identified a logical model
Glasgow and Project NeON in New York City
of change
illustrate an important point: demonstration
projects often struggle when they have not identified a logical model
of change – why doing it differently will actually have an impact. As
UCLA Professor of Public Policy Mark Kleiman writes, projects run
into problems when “they don’t start from a very plausible theory of how putting
the program into practice should change outcomes.”32 The only way to avoid
that pitfall is to work carefully to define a specific problem and a
strategy designed to address that problem and deliver the desired
outcome.

“

”

Testing the concept

32 Correspondence with
Mark Kleiman, 13 October
2011
33 Interview with Karyn
McCluskey, 27 September
2011

A final step that many successful pilot projects take is to engage
outside experts and potential critics. In part this is done to keep
planners honest: as McCluskey puts it, “I don’t want my enthusiasm to get
the better of me. Because of that, I’m always actively looking for people to take apart
my road map.”33
Another, more strategic reason to engage potential critics is to head
off potential opposition before it can threaten to derail a project. In the
case of Glasgow’s Project Pegasus, VRU did so in two ways: first, by
inviting a group of prominent academics and health professionals to
review their plans, and second, by directly engaging with human rights
lawyers who they believed might object to the project’s expectation
that programme participants remain sober (given that having a drink
is not, in and of itself, an illegal act): “I feel like I should be perfectly able to say

Getting Started: Planning a Demonstration Project

to an angry violent man,‘you can’t take any more drinks,’ but I realize that not everyone
thinks like that – at least not at first,” explains McCluskey.34
McCluskey can point to several examples where outside experts
made concrete changes to the programme’s design: “We completely
changed how we were going to recruit people and what services we planned to offer,”
she said.35 If anything, having acquired this habit of actively
searching for feedback, McCluskey was disappointed that she didn’t
encounter more criticism: “After we explained what we wanted to do, the lawyers
(we spoke with) said, ‘We don’t have any problems with it.’ I was disappointed because
I thought it was a good argument to have!”36

34 Ibid.
35 Ibid.
36 Ibid.

| 23

2. Getting it Right: Challenges of
Implementation

IMPACT was launched by local Probation, Avon and Sommerset Police,
and the Bristol City Council in 2008.The IMPACT programme is charged
with supervising 800 high-risk offenders in Bristol in south west
England – individuals considered most likely to commit frequent and
serious acquisitive crime. One of IMPACT’s strongest features is that a
multi-agency team – including probation and police officers, as well as
representatives of third sector organisations – meet regularly to engage
ideas and develop new strategies for supervising offenders. At one
meeting, a probation team leader in IMPACT suggested a novel approach
to ensuring compliance with probation orders – tasking a police officer
with hand-delivering a warning letter to probationers informing them
of their appointments and the consequences for missing them.
The operational managers in IMPACT considered the suggestion
and how it could be best put into practice. It was established however
that there may be problems if police officers actually delivered the
warning letters because they may then have to appear in court if an
offender denied a breach, to prove service of the letter. After further
consideration, and consultation with the police, it was agreed that
when probation staff sent a warning letter to an offender the recipient
would receive an immediate follow-up visit from a police officer
reinforcing both the warning, the necessity to contact their probation
Offender Manager with a reason for their absence and to emphasise to
them that they must attend their next probation appointment.
While only a small example, the process described above points
to two essential truths about how to implement criminal justice
initiatives successfully. The first involves unwavering attention to the

Getting it Right: Challenges of Implementation

| 25

details: the IMPACT team spent a great deal of time working out an
approach to home visits that made sense for both probation and the
police. The second is perhaps less intuitive though no less important:
the value of going through a process of trial and error to improve
how a project operates.
The unfortunate truth, however, is that pilot projects are rarely
afforded the luxury of reflection and refinement. Sometimes that is the
result of a basic division between planners and practitioners. Initiatives
tend to be drawn up by central planners, but practitioners often find
even the most thorough plans do not conform to on-the-ground
realities. At other times, it is the result of the intense pressure that
demonstration projects are under to produce results. Many pilot projects
are funded for only two or three years, which is scarcely enough time
for initiatives to get up and running, let alone make changes in how
they are structured. Finally and relatedly, pilot projects often suffer the
burden of unrealistic expectations – the belief
that there are “magic bullet” solutions to what
Initiatives tend to be drawn
are in reality usually very complex problems.
up by central planners, but
Successful practitioners, however, are able to
practitioners often find even
navigate this tightrope by committing
the most thorough plans do
themselves to a process of patient, incremental
not conform to on-the-ground
improvement. In fact, the projects in our study
realities
have applied a handful of useful lessons that
have guided the implementation process. They
all have a focus on those people most likely to commit crime and use
available information to identify high-risk groups accurately. They all
give front-line practitioners the discretion they need to make good
decisions about individual cases and avoid burdening them with too
many rules and restrictions. They all work hard to establish
accountability among programme participants, insisting on consistent
and fair sanctions for violations of programme rules. And finally, they
are all willing to admit when they have made mistakes and to return
to the drawing board to improve what they are doing.

“

”

26 | From the Ground Up

Focusing resources on high-risk offenders

37 Magistrates’ Magazine,
Magistrates Association,
Summer 2010
38 Sarah Hansbury,
“Evaluation of the Intensive
Alternatives to Custody pilots,
Ministry of Justice”, July 2011.
http://www.justice.gov.uk/do
wnloads/publications/researc
h-and-analysis/mojresearch/intensive-alt-custod
y-research-summary.pdf

The Intensive Alternatives to Custody (IAC) Programme in
Manchester is aimed at offenders who are otherwise likely to receive
a short custodial sentence (typically between six months and a year).
Part of a seven-site, £12 million pilot project launched by the
Ministry of Justice in 2009, IAC is aimed at addressing a seemingly
intractable problem: the high number of offenders who are given
short-term sentences despite little confidence that these sentences
themselves help offenders to desist from crime.
IAC’s core goal is to come up with a sentencing alternative that
appeals to magistrates and judges. In many instances, offenders
sentenced to a short-term prison sentence have in previous instances
been given one or more community disposals that they have failed
to complete. Magistrates and judges are understandably reluctant to
give an offender another chance if they have not shown any
inclination in the past to complete a community order. As Nicola
Still, chairman of the Magistrates’ Association Sentencing Policy and
Practice Committee, wrote recently, “We [Magistrates] do not particularly
want to send these people to prison: we know that statistics show that if we do their
chances of reoffending will increase, but ... sometimes we have little, if any,
alternative.”37
IAC attempts to address this problem in two ways: first, by creating
a customised, “high-end” community disposal option, and second,
by dedicating additional resources to supervising chronic offenders.
IAC tends to be more onerous than a typical community disposal; on
average, IAC orders have twice as many requirements (3.4 versus 1.7)
and the orders last for at least 12 months.38 In Manchester, the local
Probation Trust created a dedicated IAC team, which includes a
project director and specialised staff, assigned specifically to the local
population.
One basic challenge for planners in Manchester was ensuring that
IAC orders were only given to offenders who were truly prisonbound. One way to address this concern was to have clear criteria for

Getting it Right: Challenges of Implementation

entry into the scheme. The IAC team decided to focus its attention on
young men aged 18 to 25 at risk of a short-term custodial sentence –
a population known for having a high propensity to re-offend. Today,
the IAC team works closely with probation officers stationed at
courthouses in Manchester who prepare pre-sentence reports. If a
probation officer identifies a case that might be suitable for an IAC
disposal, they contact the programme to go over the case in question.
Finally, the IAC team communicates regularly with legal advisors
(who are permanently stationed in court to assist magistrates) to
ensure that an IAC order is only given in appropriate instances.
According to the Ministry of Justice, this effort appears to be
paying off: early indications suggest that IAC orders are only going
to offenders who are truly prison-bound. For example, an audit in
Manchester found that where the court did not follow the IAC
recommendation given in a pre-sentence report, defendants almost
always received a custodial sentence.39
In another example of keeping a programme focused on a target
population, in Bristol, members of the IMPACT team came up with
a novel way of ensuring that the programme was reserved for highrisk offenders. Referrals to IMPACT can come from any of the partner
agencies if the referring officer or staff member completes a oneparagraph description of why an individual should be admitted onto
the scheme. Once this submission is received the police, probation
and drugs teams work together to gather all existing intelligence on
the individual. Each week there is a meeting during which the team
runs through all of the prospective ‘candidates’ as well as “deselecting” those people who are either doing well enough to no
longer be considered a problem or (on the other end of the scale)
those who have been given a long-term sentence to custody. This
meeting is followed by a separate session that brings together
IMPACT front-line practitioners to ensure that the team remains
focused on those most likely to commit crime in Bristol. As John
Long, the Association of Chief Police Officers (ACPO) lead for

39 Ibid.

| 27

28 | From the Ground Up

Integrated Offender Management and Assistant Chief Constable for
Avon and Somerset Police said, “IMPACT… concentrates on our top offenders,
the 20% of people that commit the 80% of crime.”40 This integrated, intensive
approach seems to be working: between 2008 and 2010, serious
acquisitive crime (including domestic burglary and motor vehicle
theft) dropped by 28%.41

Trusting front-line practitioners

40 Avon and Somerset Police
Website, “Reducing reoffending”, http://www.avon
andsomerset.police.uk/infor
mation/documents/cache/PD
F/Document5553_442066.pdf
41 Avon and Somerset Police
Website, “IMPACT – Reducing
re-offending in Avon and
Somerset”, http://www.avon
andsomerset.police.uk/operat
ions_and_initiatives/impact.
aspx
42 Interview with Bristol
IMPACT team, 21 September
2011
43 Ibid.
44 Interview with Kevin
Kehoe, 21 September 2011

Another feature of many successful demonstration projects is that
they give considerable discretion to front-line practitioners. As Sally
Lewis, the Chief Executive Officer of the Avon and Somerset
Probation Trust (and part of the IMPACT leadership group) said, “We
trust our staff to make the right decisions and try to get out of their way.”42
Staff members from different agencies assigned to IMPACT sit
together in dedicated office space. In their regular weekly meetings
and more informal day-to-day conversations, they regularly
improvise new approaches, discarding the ones that do not work
and keeping the ones that do. This has led to some innovative new
ways of working: for example, prison officers who are part of the
IMPACT team regularly conduct home visits to individuals recently
released from custody to help reinforce the message that these
offenders are being supervised closely. As one put it, “When guys open
their door and see their prison officer standing on their doorstep, it gives them a real
shock!”43
Over time, a culture of trust has developed across the disparate
agencies involved in IMPACT. “When I started [at IMPACT], I noticed that
probation managers gave their staff more freedom and flexibility to do their jobs,”
recalled Sergeant Kevin Kehoe. “I decided to follow their example and do the
same with my police officers.”44
A similar process has occurred at the Heron Unit of the Feltham
Young Offenders Institution, located in West London. A two-year
pilot project was set up in November 2009 known as Project

Getting it Right: Challenges of Implementation

Daedalus. The unit focuses on young people who are coming to the
end of a custodial sentence. The project is funded through a
payment-by-results model and co-ordinated by the London Criminal
Justice Partnership. This involves the charity Rathbone putting claims
to the contract holders, the London Development Authority (LDA).
The LDA then audits Rathbone to ensure that it is meeting the
specified aims, and Rathbone is then paid according to education,
training and employment outcomes.45
Dedicated prison staff, working with so-called “resettlement
brokers” (who work for a consortium of charities), provide help
with education, employment, and housing, linking young people to
the services they need “outside the gate” to successfully reintegrate
into their home communities.
In the beginning of the project, some understandable tensions
emerged between the resettlement brokers and members of Youth
Offending Teams (YOT), who work for local authorities and are
responsible for supervising young people after their release. After
all, post-release, the resettlement brokers and YOT workers were
providing similar services and the boundaries between the two
groups were not always clear. As a solution to this problem,
Daedalus began convening regular “champions” meetings, which
bring together all the agencies responsible for a young person to
discuss how to work together more effectively. As a result,
tensions between resettlement brokers and YOT team members
were eased.
Another factor that has helped Daedalus achieve some early
promising results is that the prison officers who work on the Heron
Unit have all been selected through a competitive internal application
process to be part of the project. With funding from the Youth Justice
Board, the Heron Unit has six additional prison officers assigned to
the project. Working in the Heron Unit is seen as an attractive
opportunity, particularly for those prison officers who want to work
more closely with young people.

| 29

45 The outcomes are: number
of young people who start the
programme; young people
involved in positive activities
who are on the Heron Unit (6
hours); number involved in
skills development (30 hours);
number that enter into
education, training or
employment (ETE) (age 15-17);
number who enter
employment (age 16+);
number in sustained
education or training for 26
weeks; and number of young
people in sustained
employment for 26 weeks

30 | From the Ground Up

With these additional resources, Feltham has been able to do
something that it cannot do in ordinary circumstances, which is to
provide more individualised assistance to inmates in the Heron Unit.
With extra resources, the prison officers “follow” the individual,
making sure to attend on days when a particular case is being
discussed during champions meetings or on the day of their release.
Under ordinary circumstances, prison officers do not control their
own schedules, meaning that they could not ensure that they could
attend particular meetings or events.
In addition, as in the case of the IMPACT project, prison officers –
working with third-sector partners – have come up with innovative
new strategies to engage young people. For example, one prison
officer has worked to increase the Unit’s use of “release on
temporary license,” or ROTL, a complicated administrative
procedure that allows young people to leave prison during the day
to complete a community payback project (groups of young people
on ROTL have cleaned parks and cleaned up graffiti), attend a job
interview or register for classes. Another staff member has worked
to form new links to community groups and businesses, including
a partnership with Arsenal football club (Arsenal players have visited
the Heron Unit, and one former participant has gone on to work
for the club).

Demanding accountability
The projects profiled in this report also work hard to ensure
accountability with programme conditions. For example, the Harlem
Parole Reentry Court works with adult offenders released from
prison under parole supervision who are returning to the Harlem
neighbourhood of Manhattan. The project is part of the Harlem
Community Justice Center, a neighbourhood-based court launched
as a partnership by the non-profit Center for Court Innovation and
the New York State Unified Court System. Participants are required

Getting it Right: Challenges of Implementation

to return to the Reentry Court frequently to meet with case managers
and parole officers, and to appear before an administrative law judge,
who closely monitors their compliance with their parole conditions.
The goal is to prevent parolees from reoffending by helping them
find jobs and assume familial and personal responsibility. The
Reentry Court uses graduated sanctions and rewards to encourage
compliance. Sanctions may include curfews, increased court
appearances and, in the most serious cases, return to prison.
Rewards, which provide positive reinforcement for positive
behaviour, include reduced court reporting, cinema tickets and
relaxation of travel restrictions. As Grace Bernstein, a Harlem Reentry
Court judge said, her role was “to help parolees understand their conditions of
parole, helping them see the consequences of what would happen if they violated parole,
and to help celebrate their accomplishments.You need to help people find a reason to
break the cycle.”46
To help meet this goal, the Harlem Reentry Court engages parolees
prior to release and convenes regular case conferences with case
managers, parole officers and the administrative law judge. The team
discusses both “macro” issues such as new service providers
interested in working with the Court, as well as “micro” issues such
as individual problematic cases.
Those on the IMPACT scheme are also held to account very
closely. Rapidly available intelligence ensures that when those
probationers on IMPACT miss an appointment they don’t simply
receive a letter, but the police officer assigned to that offender
will know that appointment has been missed and will visit that
individual. As Assistant Chief Constable John Long explains:
“There’s lots of rehabilitative work that goes on that is very much supported by
enforcement.We make sure that offenders abide by the terms of their sentencing and
that they are keeping their appointments to help deal with their drugs problems.The
offenders know that if they don’t attend their appointments they will be back inside
the police station or before a court, or answering to someone, as to why they
haven’t.”47

| 31

46 http://rethinkingreen
try.blogspot.com/search?upd
ated-max=2011-0629T09%3A09%3A00-07%3A00
&maxresults=10
47 “Reducing re-offending”,
Avon and Somerset Police
Website, http://www.avon
andsomerset.police.uk/infor
mation/documents/cache/PD
F/Document5553_442066.pdf

32 | From the Ground Up

Admitting mistakes and changing course
The projects profiled in this study have made a commitment to
continuous incremental improvement; they have all shown a
willingness to admit failure and make mid-course corrections to
operations.
The Harlem Parole Re-Entry Court was designed to meet two
goals which often come into conflict with one another: reducing reoffending among participants while at the same time cutting returns
to prison, which can come from new offenses or from technical
violations of parole (such as missing a parole appointment or drug
test). The reason these two goals often clash is due to the
“supervision” effect observed in multiple research studies of similar
projects: the reentry court team is more likely to detect violations of
programme rules among participants because they are watching
them more closely than a typical parolee. And indeed, an evaluation
of the Harlem Parole Reentry Court found that while re-offending
dropped significantly among participants, returns to prison increased
due to increased detection of technical parole
violations.
They have all shown a
Rather than attempt to bury these negative
willingness to admit failure and
findings, the Parole Reentry Court used them
make mid-course corrections to
as
an opportunity to introduce basic changes
operations
to the program. The Reentry Court created a
new approach to technical violations, including the introduction of
an actuarial risk and needs assessment tool, which gave parole
officers a more structured way of distinguishing technical violations
of parole that might lead to more serious offending from ones that
were less threatening, and the introduction of new cognitive
behavioral group sessions for high risk participants.
In Manchester, the IAC team made changes to programme
operations in response to a Ministry of Justice-commissioned process
evaluation of the scheme. Project Director Paul Pandolfo believes that
the feedback from the evaluation focused the team on several

“

”

Getting it Right: Challenges of Implementation

important issues. One example was the evaluator’s finding that
compliance rates were lower among members of the Black and
Ethnic Minority Ethnic (BME) community. In response, the IAC team
created a workshop specifically for this BME group; the training was
so successful that it has been filmed and made available to other IAC
projects. The team also learned that substance misuse – most often
binge drinking or heavy cannabis use – was preventing probationers
from making progress on their education and training goals. By
working with the National Health Service, IAC was able to recruit
on-site drugs workers for the programme.

| 33

3. Keeping it Going: Sustaining
and Spreading the Best Ideas

48 Correspondence with
Diana Barren, 28 September
2011
49 Ibid.
50 Ibid.

When Diana Barran decided to leave her job as a hedge fund
manager, she began talking to children’s charities across England and
Wales about the problems they were having trouble addressing. “I
asked them for the most hidden social problem in this country which was the hardest
to raise money for and they all independently said domestic abuse,” she wrote.48 She
then began visiting charities working in the field and was struck by
a key gap: while most agencies were focused on removing women
and children from their homes, there was little attention being paid
to those majority of cases where it made more sense for victims and
their families to stay in their homes. That approach would require
wholesale cultural change, knitting together the various disparate
public (police, probation, the courts and local authorities) and third
sector organizations involved in domestic violence cases and refocusing their attention on the needs of victims. As Barran wrote, “at
the time, there was relatively little practical multi-agency work, little focus on risk and
no real trained support for victims.”49
This educational process led Barran to start a new non-profit
organization that would be devoted to helping victims of domestic
violence. She dubbed it Co-Ordinated Action Against Domestic Abuse
(or CAADA). As Barran wrote, “CAADA was born on my kitchen table!”50 Six
years later, CAADA has become an established part of the national
domestic violence landscape, helping to launch (and providing
support to) Multi-Agency Risk Assessment Conferences (or
MARACs), which bring together all the relevant public and thirdsector stakeholders to share information and create a safety plan for
the highest risk victims of domestic abuse., CAADA also defined and

Keeping it Going: Sustaining and Spreading the Best Ideas

developed the role of the Independent Domestic Violence Advisor
(IDVA), who offer practical support to victims and represent them at
the MARAC meeting. All told, there are 240 MARACs operating in
England and Wales, and CAADA has provided accredited training to
over 1,100 IDVAs. And in the year to June 2011, MARACs analysed,
and created safety plans for, over 45,000 high-risk adult victims of
domestic violence and over 63,000 associated cases involving
children.
Barran’s approach to this issue stands out from all the other projects
described in this paper. Although she has been deeply involved in
supporting the roll out of both MARACs and IDVAs, her organization
does not operate either program directly. Instead, CAADA acts as an
advisor and technical assistance provider, supporting these projects on
a national basis. CAADA’s success brings into focus a third, and final
challenge faced by all practitioners and policymakers who work on
demonstration projects: how do you not only ensure their survival,
but spread the best ideas and practices throughout the system?

Getting beyond the pilot stage
Many of the projects described above are facing challenges in trying to
get beyond the pilot stage. For example, central government funding for
Intensive Alternative to Custody projects was withdrawn after two years,
and the Manchester project has had to scramble for local funding to
keep going, a difficult challenge in a tough fiscal climate. In a sign that
the project is seen as effective by local stakeholders, the project has been
successful in obtaining funding for a third year of operations. In
addition, IAC has recently extended its approach to 16 and 17 year olds.
Bristol had another approach to sustainability: despite being part
of a multi-site national demonstration project, from the onset they
have not taken any additional central government funding to support
operations, meaning that the work of the IMPACT team is paid for
with existing resources. Their logic in being part of the national

| 35

36 | From the Ground Up

network of demonstration sites was that they would benefit from the
opportunity to learn from other projects, as well as from being
included in any national evaluations. “We purposely never sought additional
public funding [to support IMPACT]” recalled Sally Lewis, the Chief Executive
Officer of the Avon and Somerset Probation Trust and the national lead
on offender management with the Probation Chiefs Associations.51
The Violence Reduction Unit in Glasgow has worked to sustain
initiatives by mainstreaming them within the police service. For
example, its gangs project has been adopted by the Strathclyde police
force, although members of VRU still sit on its executive board. In the
case of Project Daedalus, pilot funding was initially scheduled to run
out in November 2011, but planners were able to negotiate a sixmonth extension and are working with the Greater London Authority
to secure the future of the project.

Attracting champions

51 Correspondence with
author, 18 October 2011
52 Angela Hawken, School of
Public Policy “HOPE for
Probation: How Hawaii
Improved Behavior with HighProbability, Low-Severity
Sanctions”, Pepperdine
University

As the example of Diana Barran and CAADA shows, one powerful
way to sustain criminal justice projects is to attract powerful
individuals or institutions to serve as champions.
This was the case in Hawaii where an enterprising judge, Steve Alm,
launched a project called “HOPE Probation” in 2004. The project was
borne out of Alm’s frustrations on the bench: appointed to Hawaii’s First
Circuit in 2001, Alm regularly saw individual offenders with up to 30
different probation violations come before him. Alm believed that by
that time it was too late to change offenders’ behaviour: “If my son
misbehaved, I would talk to him about what he had done wrong and warn him that he
shouldn’t do it again,” Alm said. “Then, if he did it again, I would give him a swift and
sure, but proportionate punishment for breaking the rules.That way, he would learn from his
mistake. I thought that it made sense to apply that thinking to the probation system.”52
HOPE was born from this analysis. With HOPE, probationers are
regularly drug tested, and if they fail their tests, they are given an
immediate and certain short jail sentence as a sanction (typically

Keeping it Going: Sustaining and Spreading the Best Ideas

several days; servable at the weekend if employed). During their first
two months in HOPE, probationers are randomly tested for drugs at
least six times per month. They are assigned a colour code at the
warning hearing and are required to call the HOPE hotline each
weekday morning. Those probationers whose colour is selected must
appear in court before 2pm that day for a drug test. Good behaviour
through compliance and negative drug tests is rewarded with an
assignment of a new colour associated with less-regular testing. But
if a probationer fails to appear for testing, an arrest warrant is issued
immediately and is served by the Honolulu Police Department.
Probationers who test positive for drug use or fail to appear for
probation appointments are arrested and held in custody.
Over time, individuals who repeatedly fail drug tests are assigned to
drug treatment programmes. The logic is that their inability to change
their behaviour despite repeated negative consequences is in itself a
sign of drug addiction. Another benefit of the HOPE approach is that
referrals for drug treatment are reserved for this group of offenders,
which not only economises on the use of drug treatment but removes
the need for expensive upfront screening and assessment.
A probationer found to have violated the terms of probation is
immediately sentenced to a short jail stay, with credit given for time
served. The probationer resumes participation in HOPE and reports
to his or her probation officer on the day of release. While those on
probation may request a treatment referral at any time, probationers
with multiple violations are mandated to intensive substance abuse
treatment services (typically residential care). The court continues to
supervise the probationer throughout the treatment experience and
consistently sanctions noncompliance.
Research shows that the programme has reduced crime and
probation revocations, and therefore saved on prison costs. By 2009,
positive drug tests were reduced by 86% and missed appointments
by 80%. Revocations were reduced by 50% and arrests for new
crimes were reduced by more than 50%.

| 37

38 | From the Ground Up

53 See, for example,
“Managing Drug Involved
Probationers with Swift and
Certain Sanctions: Evaluating
Hawaii’s Project HOPE,”
Angela Hawken and Mark
Kleiman, United States
Department of Justice,
December 2009. Also see
Mark Kleiman, When Brute
Force Fails: How to Have Less
Crime and Less Punishment,
Princeton University Press,
2010. Also see Mark Kleiman,
“Jail Break: How smarter
parole and probation can cut
the nation’s incarceration
rate,” Washington Monthly,
July/August 2009.
54 See, for example, “The
Impact of Hawaii’s HOPE
Program on Drug Use, Crime,
and Recidivism,” Pew Center
on the States, January 2010.
Available at http://www.pew
centeronthestates.org/upload
edFiles/PSPP_HOPE_Brief_we
b.pdf.
55 Correspondence with
Diana Barren, 11 September
2011
56 Ibid.
57 Ibid.

Alm did more than create the programme, however: he worked
tirelessly to promote it both locally and nationally, including appearing
at conferences and collaborating with academics to run a randomised
controlled trial. He attracted the attention of Pepperdine University
Professor of Public Policy Angela Hawken and Mark Kleiman, a
Professor of Public Policy at the University of California-Los Angeles,
who have made HOPE a centrepiece of their written work, including
a number of articles and books.53 Working together Alm, Hawken and
Kleiman have recently persuaded the United States Department of
Justice to launch a multi-site replication of the HOPE model, and
Kleiman has travelled to California and other states to encourage
policymakers to create their own versions of HOPE. Finally, the Pew
Charitable Trusts have worked to promote the programme nationally,
summarising the findings of Hawken and Kleiman in a number of
publications that have been widely distributed.54
As Barran points out, these types of tasks – raising money,
organising research and evaluation, and advocating at a national
level – are often (though not always, as in the case of Judge Alm)
performed more effectively by an outside agency: “You need an
organisation whose primary role is to do the roll out”, she wrote. “You can’t
expect the manager of a local project to do this. It’s a different skill set and both jobs
are more than full time.”55 These advocates, however, need to work
closely with the people actually operating the projects. “Our approach
has been to try and get people to do their day job well rather than convert them to
the importance of our cause,” she said. “We reckon that if the approach works,
and they see the benefits for themselves, then they will be more committed.”56
Again, the key is a tight relationship between policy and practice.
“All too often ideas fail because the expectations of partner agencies are unrealistic
in practice, albeit reasonable in theory.”57
The work done by Alm, Kleiman, and others provided clarity to
“what works” about a particular programme model. As numerous
evaluations have shown, even small tweaks to a programme can result
in big differences in outcomes. Gathering relevant data, defining the

Keeping it Going: Sustaining and Spreading the Best Ideas

| 39

delivery model, and keeping an intense focus on outcomes is
essential if a local programme is to make a successful transition from
local to national. Without this, there is a real risk that the approach
itself will be diluted and the original project will lose momentum.

Getting the politics right
A final role that outside advocacy groups can play is helping to get
the politics of criminal justice innovation right. This has been the
role played by the Pew Center on the States’ Public Performance
Project, which has helped states across the US implement “justice
reinvestment” strategies – shifting spending away from new prison
construction and towards investments in community-based
alternatives. Part of Pew’s approach has been to remain strictly
nonpartisan about the policies it advocates.
“We are a neutral, objective third party without
Outside advocacy groups
ideological baggage or scars from prior state turf battles”
can
help to get the politics of
wrote Adam Gelb, Director of Pew’s efforts.
criminal justice innovation right
“That gives our analysis extra credibility.”58
Another advantage of Pew’s approach is
that they avoid the simplistic ‘soft versus hard’ trap that many
criminal justice initiatives fall into. According to Gelb, “We remain
strictly agnostic about policy reforms in a state until we’ve analysed the drivers of the
prison population and costs.There’s no cookie-cutter approach. It’s tailored to the unique
circumstances of each state as revealed by their data. The data analysis itself is guided
by a bipartisan, interbranch group of policy makers, and that buy-in to the process
helps focus the discussions on sound science rather than sound bites.”59
In Kentucky, Pew worked tirelessly to advocate for a new approach
to corrections, including conducting analysis and appearing before
various state committees. They worked closely with Tom Jensen, the
Republican Senate Judiciary Chairman, and John Tilley, the 58 Correspondence with
Adam Gelb, 29 September
Democratic House Judiciary Chairman. Their efforts paid off: earlier 2011
this year, Kentucky Governor Steve Beshear passed the Public Safety 59 Ibid.

“

”

40 | From the Ground Up

60 “Governor Beshear signs
bill aimed at lowering
Kentucky prison population”,
3 March 2011:
http://www.courierpress.com
/news/2011/mar/03/govbeshear-signs-bill-aimed-lowe
ring-kentucky-pri/

and Accountability Act by a resounding 96-1 vote. The bill shifts state
spending away from incarceration and towards community-based
alternatives like probation. In signing the bill, Governor Beshear
announced that the changes “enabled the State to continue to be tough on crime
while at the same time being smarter about it.”60
The experience of the Pew Public Safety Performance Project in
Kentucky reinforces Barran’s point about the importance of outside
champions. Politics can be a tough challenge for criminal justice
practitioners, particularly those who are focused on operating a
project at a local level. Even the best and most effective
demonstration projects cannot succeed without political support.
As credible outsiders, groups like CAADA and Pew can be
particularly effective in helping demonstration projects navigate
these tricky waters.

Conclusion

Making the criminal justice system fairer and more effective is the
ultimate goal of many, if not all, reformers. The question is how best
to achieve this goal. This paper has sought to highlight one approach
to change: creating demonstration projects at the local level, finding
out what works and what doesn’t, and then disseminating new ideas
and new practices more widely.
There are many drawbacks to this approach. It can be timeconsuming and pain-staking. It favours incremental reform as
opposed to wholesale change. And since execution is crucial, it relies
on the active engagement of frontline practitioners such as police
officers, magistrates, probation officers, and social workers.
Nonetheless, there is ample evidence that demonstration projects are
in fact capable of both solving local public safety problems and
generating knowledge that is capable of transforming the wider field of
criminal justice. Indeed, the demonstration projects described in this
paper have achieved some encouraging results, addressing problems
like repeat offending, prison overcrowding, and domestic violence.
The ink isn’t dry on many of these experiments; it is too soon to tell
whether they will be able to sustain their early success – and translate
that success into broader systemic impact. It is, however, possible to
mine the demonstration projects profiled above for a number of
valuable lessons for the policymakers and practitioners of tomorrow.

Key Lessons for Successful Innovation
 Start Small. One factor that has helped the projects described

above is that many started small, beginning with limited tests of
the concept before expanding their operations. In the case of

42 | From the Ground Up

Bristol IMPACT, the project team began with a caseload of only
125 offenders before eventually growing to 800, which had
multiple advantages. Early funding demands are therefore less
but perhaps most important, starting small provided them with
the opportunity to work out the kinks of the model. The same
process was at work in or on the Hawaii HOPE project, which
started with a caseload of only 35 practitioners in a single
courtroom before expanding throughout Oahu. Starting small
and engaging in a rigorous trial-and-error process is crucial to
the ultimate success of demonstration projects.
 Real Change Takes Time. In the era of the perpetual, 24/7 news

cycle, one of the hardest things to ask for – from the media, from
politicians and from the public – is patience. But the reality is
that change within a public institution as large and as complex
as the criminal justice system doesn’t happen overnight. Many
of the pilot projects described above were funded for a two-year
period, which is barely enough time for a project to establish
itself, let alone start to achieve (and capture) measurable results.
Unfortunately, there are numerous examples of promising
initiatives and interesting ideas being jettisoned not because they
were proven to be ineffective, but simply because policymakers
did not give them enough time to achieve their goals.
 Don’t Go It Alone. External champions can play a key role in

demonstration projects, helping local practitioners meet
challenges like obtaining political support or finding funding.
For example, CAADA provided critical support to practitioners
throughout England and Wales working with victims of domestic
violence, and Pew’s Public Safety Performance Project helped
policymakers in Kentucky analyse their prison population and
come up with new solutions designed to cut crime and spending
simultaneously. Interestingly, these champions came from outside

Conclusion

government, which points to the role that credible intermediary
organisations can play in aiding local practitioners and
promoting innovation more broadly.
 Politics Matters. Local political interest in pilot projects is to be

welcomed, and over time, political attention can lead to more
media exposure and financial support. However, the most
sustainable projects recognise the need to remain apolitical in
their leadership and to reach out to all parties, not just those in
office when a project is favoured. Political support is often
temporary and can be reliant on relationships that can end
unexpectedly and rarely exceed the duration of a project. The
politics of a project can go wrong when that scheme becomes
too closely tied to the fortunes of incumbent politicians, and
therefore vulnerable to termination or funding cuts if the
political landscape changes. Projects can also be discredited if
political rhetoric runs ahead of the on-the-ground reality, or if
inflated claims are made about programme impact that do not
match the tentative or modest successes that a scheme has
achieved so far, thus raising expectations unfairly.
 Research Matters. Thoughtful champions and political support

make a difference, but nothing is more important than being able
to document impact. The ideas that have successfully moved from
a single programme to broader implementation can usually point
to hard data, and robust independent analysis, not anecdotes, as
one of the key forces behind replication. At the most successful
demonstration projects, research isn’t an afterthought or a
necessary evil. Nor is it something that is out-sourced and then
forgotten about. Even in cases where there is an independent
researcher, it is crucial for those engaged in implementation to
play a key role in evaluation, helping to define realistic, robust
and measurable outcomes for their project.

| 43

44 | From the Ground Up

 Keep focused. Successful schemes target resources effectively and

keep focused on where they can make the most impact. In
criminal justice, this often means working with finite resources
and addressing the most prolific offenders or those that present
the greatest risk of harm. The best projects in our study, from the
HOPE probation scheme in Hawaii to reduce drug-related
offending to the Save Our Streets programme designed to reduce
gang-related violence in New York, have a relentless focus on
those people most likely to commit crime and use all the
information available to identify that high-risk group accurately.
Allowing mission creep and attempting to solve too many
problems across too many areas can dilute the impact of a
programme and undermine the distinctive nature of a project
which can be the key to gaining funding and support.
 Don’t Take Funding for Granted. Once a project has secured

enough seed capital to begin, it is natural to divert attention to
the immediate operational issues around implementation and
evaluation. But even when projects receive local acclaim or
national recognition, ongoing funding can still be problematic.
In the criminal justice sector, although some schemes benefit
from secure funding, most support – from the private, voluntary
or public sector – is not usually long-term, and can run out just
when a project is close to proving its impact. Diversity of funding
provision can help. Larger grants may come with strings. Start-up
investment and bridging income may be needed if contracts are
outcomes-based. Even after initial success, funding can never be
taken for granted. Today’s funding is no guarantee of tomorrow’s
likely income if a project needs more money to expand.
Demonstration projects are not the only path to change in criminal
justice. There will always be a place for legislation, litigation, protest,
and other forms of advocacy. But as the case studies in this report

Conclusion

make clear, demonstration projects are an important tool for both
policymakers and practitioners. New approaches to preventing crime
depend upon imagination, experimentation, and robust evaluation.
This favours a localised approach that frees professionals to innovate,
creates new understanding, and thereby provides policy makers with
more and better evidence to effect change more widely. If we want
to be smart on crime we need to rely less on national policies devised
and imposed by central government and more on fostering
innovation and demonstration projects at a local level that over time
will show how best to improve public safety.
Despite the political emphasis in the UK on top-down
government solutions and large, structural reforms, England and
Wales have a good record in recent years of stimulating change from
the ground up. The challenge in future in both the UK and the US is
to make sure that demonstration projects are supported, that they
are implemented and evaluated thoughtfully, and that the lessons
arising from them are spread as broadly as possible.

| 45

www.policyexchange.org.uk

From the Ground Up

Policy Exchange
Clutha House
10 Storey’s Gate
London SW1P 3AY

	

£10.00
ISBN: 978-1-907689-12-3

Policy Exchange	

From the Ground Up: Promising criminal justice
projects in the US and the UK examines successful
demonstration projects in the UK and the US that
are attempting to reduce crime, drug use and
incarceration, among other challenging goals. The
report identifies a selection of innovative, groundlevel experiments in policing, probation, courts and
crime prevention that have had a real impact. In this
report the founders and lead practitioners share
their experiences of creating innovative projects
and in reviewing the projects the author identifies
key practical lessons for how to successfully plan,
implement and sustain new criminal justice
endeavours.

From the
Ground Up
Promising criminal justice projects
in the US and the UK

Aubrey Fox and Gavin Lockhart
Edited by Blair Gibbs