Housing First
Premier Ministre
Ministère de l’Écologie,
du Développement durable,
des Transports et du Logement
Housing
First
Nicholas Pleace European Observatory oN Homelessness
Acknowledgements
The author would like to thank Marie Gaffet at DIHAL for
coordinating the drafting and production of this report, as
well as Ruth Owen, Policy Officer at Feantsa for her help with
this review and her comments. Responsibility for any errors
lies with the author.
1
Housing First
Table of Contents
Summary..................................................................................................................................................3
What is Housing First? ............................................................................................................................................................................................................3
Pathways Housing First (PHF)...........................................................................................................................................................................................................3
Communal Housing First (CHF)..........................................................................................................................................................................................................4
Housing First ‘Light’ (HFL).......................................................................................................................................................................................................................4
Evidence in favour of Housing First................................................................................................................................................................................5
The Global Influence of Housing First............................................................................................................................................................................6
The Case for Housing First...................................................................................................................................................................................................6
1.	Introduction........................................................................................................................................7
About this Report...........................................................................................................................................................................................................................7
The Structure of the Report.................................................................................................................................................................................................8
2.	 Pathways Housing First (PHF)..........................................................................................................9
Introduction..........................................................................................................................................................................................................................................9
Origins......................................................................................................................................................................................................................................................9
Problems with ‘staircase’ services..................................................................................................................................................................................................9
What is Pathways Housing First? ................................................................................................................................................................................10
The PHF Philosophy..................................................................................................................................................................................................................................11
Delivering a Pathways Housing First Service .....................................................................................................................................................14
Which groups of homeless people is PHF designed to help?.................................................................................................................................14
What are the aims of a PHF service?........................................................................................................................................................................................14
How is the PHF service accessed? ...........................................................................................................................................................................................15
How is the PHF service delivered?..............................................................................................................................................................................................15
What does PHF cost?.............................................................................................................................................................................................................................17
Risk management........................................................................................................................................................................................................................................17
Summary of PHF service delivery...............................................................................................................................................................................................18
Evidence of effectiveness.......................................................................................................................................................................................................18
Housing Stability...........................................................................................................................................................................................................................................18
Changes in Drug and Alcohol Use...............................................................................................................................................................................................18
Changes in Mental Health....................................................................................................................................................................................................................19
Social Inclusion.............................................................................................................................................................................................................................................20
Economic Inclusion...................................................................................................................................................................................................................................20
Cost Effectiveness......................................................................................................................................................................................................................................21
3.	 Communal Housing First Services...............................................................................................................23
Introduction......................................................................................................................................................................................................................................23
Origins..................................................................................................................................................................................................................................................23
What is Communal Housing First? .............................................................................................................................................................................23
CHF Philosophy ...........................................................................................................................................................................................................................24
2
The Housing First philosophy and CHF services in Finland...................................................................................................................................25
Delivering a Communal Housing First Service...................................................................................................................................................26
Which groups of homeless people is a CHF service designed to help?......................................................................................................26
What are the aims of a CHF service? ....................................................................................................................................................................................27
How is a CHF service accessed?.................................................................................................................................................................................................27
How is a CHF service delivered? ................................................................................................................................................................................................27
What does a CHF service cost?...................................................................................................................................................................................................28
Risk management.......................................................................................................................................................................................................................................29
Summary of CHF service delivery..............................................................................................................................................................................................29
Evidence on effectiveness...................................................................................................................................................................................................29
Housing stability...........................................................................................................................................................................................................................................29
Changes in Drug and Alcohol Use.............................................................................................................................................................................................30
Changes in mental health......................................................................................................................................................................................................................31
Social and economic inclusion..........................................................................................................................................................................................................31
Cost effectiveness.......................................................................................................................................................................................................................................31
4.	 Housing First ‘Light’ Services........................................................................................................................32
Introduction......................................................................................................................................................................................................................................32
Origins..................................................................................................................................................................................................................................................32
What are HFL services?......................................................................................................................................................................................................32
HFL philosophy.............................................................................................................................................................................................................................34
Delivering a Housing First Light Service.................................................................................................................................................................35
Which groups of homeless people is a CHF service designed to help? .....................................................................................................35
What are the aims of a HFL service? ....................................................................................................................................................................................35
How is an HFL service accessed?..............................................................................................................................................................................................36
How is an HFL service delivered? .............................................................................................................................................................................................36
What does an HFL service cost? ...............................................................................................................................................................................................37
Risk Management......................................................................................................................................................................................................................................38
Summary of the HFL service model..........................................................................................................................................................................38
Evidence on effectiveness...................................................................................................................................................................................................39
Housing stability...........................................................................................................................................................................................................................................39
Changes in drug and alcohol use.................................................................................................................................................................................................39
Changes in mental health.....................................................................................................................................................................................................................39
Social and economic inclusion.........................................................................................................................................................................................................40
Cost effectiveness......................................................................................................................................................................................................................................40
5.	 Housing First in France...................................................................................................................................41
Introduction........................................................................................................................................................................................................................................41
The Global Influence of Housing First..........................................................................................................................................................................41
Questions about Using a Housing First Approach in France.................................................................................................................42
The Limits of Housing First................................................................................................................................................................................................................42
Housing First is “not designed” for France............................................................................................................................................................................43
Housing First cannot address all forms of homelessness in France..............................................................................................................44
Why use Housing First in France?..............................................................................................................................................................................45
3
Housing First
Summary
What is Housing First?
Housing First differs from other homelessness
services because it immediately provides homeless
people with either secure independent or communal
housing. This means that Housing First gives
homeless people ‘housing first’ before it does
anything else. Housing First ‘separates’ housing
and support. This ‘separation’ means that homeless
people are immediately given secure housing without
being required to enter psychiatric treatment or
becoming abstinent from drugs and alcohol. In
addition, homeless people can remain in the secure
housing provided by a Housing First service without
there being any expectation that they comply with
psychiatric treatment or stop drinking alcohol or
using drugs. Housing First services use a ‘harm
reduction’ approach that attempts to stabilise and
reduce mental health problems, problematic drug
and alcohol use. Housing First services also have
‘recovery orientation’ that aims to encourage
homeless people away from behaviour that is
causing them harm.
Housing First was originally developed by the
organisation Pathways in New York. Since
the Pathways Housing First service was first
established, several different forms of Housing First
service have appeared in the USA and in a number
of EU member states including Denmark, Finland,
Ireland, France, Hungary, the Netherlands, Portugal,
Austria and the UK. These services all share a
similar philosophy but can work in quite different
ways. The three main groups of Housing First
service are:
❱❱ Pathways Housing First (PHF),
❱❱ Communal Housing First (CHF)
❱❱ Housing First ‘Light’ services (HFL)1
Pathways Housing First (PHF)
❱❱ PHF works with ‘chronically homeless people’2
who are characterised by severe mental illness,
problematic drug and alcohol use, nuisance
behaviour, low-level criminality, sustained
worklessness and long periods living in
homelessness shelters and on the street.
❱❱ PHF provides independent housing with security
of tenure immediately or as soon as possible to a
chronically homeless person.
❱❱ PHF provides low-level support designed to
promote housing stability, service brokerage (to
connect chronically homeless people to essential
services) and also directly provides psychiatric,
drug and alcohol, social work, medical and other
services. The team providing these services is
mobile and visits people using the service in their
homes or at other agreed locations.
❱❱ PHF services give considerable choice and
control to chronically homeless people by
following a harm reduction model which allows
them to continue drinking alcohol and using
drugs. Chronically homeless people can chose
not to use the psychiatric and drug and alcohol
services that are made available by PHF whilst
remaining in the housing provided by PHF. This
is called a ‘separation’ of housing and support.
1	 Please note that the acronyms PHF, CHF and HFL are used consistently within this report hereafter
2	 The US Federal government defines a chronically homeless person as either an unaccompanied homeless individual
with a disabling condition who has been continuously homeless for a year or more; or an unaccompanied individual with
a disabling condition who has had at least four episodesof homelessness in the past three years. The term is used more
and more frequently in European contexts to describe long-term homeless people with more complex needs.
4
Communal Housing First (CHF)
❱❱ CHF services are focused on chronically
homeless people. CHF offers communal housing
(single rooms or apartments) with security of
tenure provided immediately in a building only
lived in by homeless people using the CHF
service. The communal housing has often been
specifically modified, or designed, to provide a
service for chronically homeless people. Support
and medical services are situated in the same
building or are very nearby.
❱❱ CHF provides communal housing with security
of tenure immediately to a chronically homeless
person.
❱❱ CHF directly provides psychiatric, drug and
alcohol services and medical services and may
also use service brokerage to arrange access to
other services.
❱❱ CHF services give considerable choice and
control to chronically homeless people as part of
following a harm reduction model. If chronically
homeless people opt not to use the services that
are on offer, or choose not to stop drinking and
using drugs, this does not place their housing in
the CHF service under threat.
Housing First ‘Light’ (HFL)
❱❱ HFL is delivered by using ordinary private rented
or social housing and a team of mobile support
workers.
❱❱ HFL provides independent housing with security
of tenure immediately or as soon as possible to
a homeless person.
❱❱ HFL may be used to help prevent homelessness
where an individual or household who has never
been homeless is assessed as being at risk of
homelessness.
❱❱ HFL provides low-level mobile support services
designed to help promote housing stability.
❱❱ HFL uses service brokerage to arrange access
to psychiatric, drug and alcohol services and
medical services where these are required and
may also use service brokerage to arrange
access to education, training and other services
where these are needed.
❱❱ HFL services do not directly provide medical,
psychiatric or drug and alcohol services.
❱❱ HFL can support chronically homeless people but
may also be used for other groups of homeless
people, including homeless people with lower
support needs. The other groups that might
be targeted by HFL services can include young
homeless people and homeless families.
❱❱ HFL services give considerable choice and
control to homeless people as part of following
a harm reduction model. If homeless people do
not use the medical and support services which
can be arranged, or choose to continue drinking
alcohol and using drugs, this does not place their
housing under threat.
The differences between PHF, CHF and HFL
services are summarised in Table S.1
5
Housing First
Table S.1 : Broad Types of Housing First Services
Service offered
Pathways
Housing First
Communal
Housing
First
Housing
First Light
Housing with security of tenure in private rented sector or in social
housing provided immediately or as soon as possible
Yes No Yes
Offers communal housing (single rooms or apartments) with security
of tenure provided immediately in a building only lived in by homeless
people using the service
No Yes No
Homeless people have to stop using drugs No No No
Homeless people have to stop drinking alcohol No No No
Homeless people have to use mental health services No No No
Harm reduction approach Yes Yes Yes
Uses mobile teams to provide services Yes No Yes
Directly provides drug and alcohol services Yes Yes No
Directly provides psychiatric and medical services Yes Yes No
Uses service brokerage Yes Yes Yes
Provides support to promote housing stability Yes No Yes
Evidence in favour of Housing First
The bulk of evidence about Housing First is
American. There is also some evidence from Europe
and in particular from Finland. Large scale testing
of the Housing First model is occurring across the
European Union and within France. For example,
Housing First Europe3
is a Social Experimentation
project funded in the framework of the PROGRESS
programme of the European Commission. It will
evaluate and provide mutual learning on 10 Housing
First projects in European cities and deliver
conclusions in 2013. France is implementing a major
social experimentation called “Un Chez-soi d’abord”4
to test the effectiveness of Housing First services
for homeless people with mental illness.
There is very strong evidence from the USA that
PHF services provide stable housing for chronically
homeless people with very high support needs and a
long history of homelessness. There is more limited
evidence on CHF and HFL services that also show
success in providing stable housing for homeless
people. There is some evidence that by promoting
housing stability and following a harm reduction
approach, Housing First services can stabilise and
in some cases reduce mental health problems and
problematic use of drugs and alcohol.
PHF has shown far more success in providing
housing stability for ‘chronically homeless’ people
than ‘staircase services’. Staircase services require
homeless people to move through a series of ‘steps’
before they are given access to housing. They often
require abstinence from alcohol and drugs. There is
strong evidence that the ‘staircase’ approach fails to
provide stable housing for the majority of chronically
homeless people who use staircase services.
3	 http://www.servicestyrelsen.dk/housingfirsteurope
4	 http://www.developpement-durable.gouv.fr/Programme-Chez-Soi-d-abord.html
6
The Global Influence of Housing
First
Housing First has become globally important because
of the unparalleled success that it has shown in
providing stable housing and accommodation for
chronically homeless people. Housing First has
been central to the Federal homelessness strategy
in the United States under both the Bush and
Obama administrations. Housing First services have
been integrated into the homelessness strategies of
France, Denmark, Finland, Ireland, the Netherlands
and Sweden and are being widely tested in pilot
programmes across many EU Member States. The
Jury for the European Consensus Conference on
Homelessness, held in Brussels in December 2010,
recommended that ‘housing-led’ approaches were
the most effective solution to homelessness and that
the different forms of Housing First service were
good examples of these ‘housing-led’ services.
The Case for Housing First
There are three main reasons to consider using
Housing First services in France.
❱❱ The core philosophy of Housing First is the
reason for the success that has been achieved
in delivering stable housing and accommodation
for formerly chronically homeless people. The
Housing First philosophy can be drawn upon
without copying the detailed operation of Housing
First services working in the USA. The ideas of
Housing First can be used to design services that
can be used to reduce long-term homelessness
in different countries with different welfare
systems. An example of this has been the use
of the Housing First philosophy to effectively
redesign homelessness services in Finland.
❱❱ Housing First is highly flexible. The core
philosophy has been adapted from the PHF model
and employed in CHF and HFL services that can
be focused on both chronic homelessness and
on other forms of homelessness.
❱❱ Housing First services can support other
types of homelessness services, for example,
by stopping very long stays in emergency
accommodation beds by chronically homeless
people. It is important to note that Housing First is
not designed to replace all existing homelessness
services and has been used as one part of a
mixture of services to respond to homelessness.
By supporting the work of other homelessness
services, Housing First services can potentially
enhance their capacity to respond to issues such
as migrant homelessness and homelessness
among families with low support needs in France.
7
Housing First
1.	 Introduction
About this Report
The report is intended to act as an introduction
to Housing First and is designed to help stimulate
and inform debate. It is not a blueprint of how to
run Housing First services in France, because the
existing evidence on Housing First is concentrated
on other countries and the Housing First model
will need to be adapted to the specific conditions in
France to work well. A Housing First service in Paris
would not face the same situation or challenges as
a Housing First service in another country, or a
Housing First service working in another city or in
a rural Department within France.
The report describes what Housing First is, how
it works, explores the evidence base and then
discusses the case for the wider use of Housing
First in France. The report is intended for all the
organisations involved in tackling homelessness
in France, including NGOs, social workers, policy
makers, regional councils, social and private housing
providers, health professionals and elected officials.
Housing First has become increasingly important
within strategic responses to homelessness across
the economically developed World. However, there
can be some confusion about what is meant by
‘Housing First’. This is because new ‘Housing First’
services do not always follow the same model.
Various forms of ‘Housing First’ services exist in
Denmark, Finland, France, Hungary, Ireland, the
Netherlands, Portugal, the USA and the UK5
. This
report describes the three main types of ‘Housing
First’ services and reviews the evidence base for
each one.
The report is a summary that uses broad definitions
to try to encompass both the philosophy and
operational reality of Housing First services. The
descriptions are broad because they encompass
sometimes very different services that share
several elements of a core philosophy. It is very
important to note that the sections on Communal
Housing First (CHF) and Housing First Light (HFL)
describe a range of services rather than providing
a description of a single, specific type of service.
The report argues that Housing First services would
be effective in France. There is very strong evidence
that Housing First can produce housing stability – a
permanent exit from homelessness – for a majority
of those chronically homeless people with very high
needs who have been homeless a long time. Achieving
very high rates of housing stability for this group of
chronically homeless people is not something that
other types of homelessness services, including
the ‘staircase’ model, have been able to achieve.
This report does not suggest that Housing First is
a ‘perfect’ homelessness service or that Housing
First services are able to deal effectively with all
the forms of homelessness, such as homelessness
amongst migrants who face administrative or legal
barriers to welfare and housing, as well as family
homelessness, which exist in France. However, it is
argued that that the small group of very vulnerable
people spending years on the streets and in beds
in emergency accommodation can be significantly
reduced by Housing First. Furthermore, it seems
that elements of the Housing First philosophy can
provide policy makers and service providers with
useful insights into how to address other forms
of homelessness e.g. amongst people with lighter
support needs.
5	 http://www.servicestyrelsen.dk/housingfirsteurope
8
The Structure of the Report
The rest of the report is structured as follows:
❱❱ Chapter 2 focuses on Pathways Housing First.
The chapter begins by describing the PHF
model and then reviews the evidence on the
effectiveness of PHF.
❱❱ Chapter 3 looks at ‘Communal Housing First’
and follows the same structure as Chapter
2, describing how CHF services work and
reviewing the evidence on the effectiveness of
the CHF approach.
❱❱ Chapter 4 looks at Housing First ‘Light’ services
and again follows the same structure as chapters
2 and 3, describing HFL services and reviewing
the evidence.
❱❱ Chapter 5 discusses the use of Housing First
services in France, looking at why Housing
First has become globally influential, exploring
some of the questions that have been raised
about whether or not Housing First is suitable
for France and finally summarises the case for
considering using Housing First in France.
9
Housing First
2.	Pathways Housing First (PHF)
Introduction
This chapter looks at the original Housing First
service model which was developed by Pathways
in New York City. The chapter begins by looking at
the origins of the Pathways service and then moves
on to discuss its philosophy and operation. Finally,
the evidence base for Pathways Housing First is
discussed.
Origins
Problems with ‘staircase’ services
The ‘staircase’ model is widely used in the USA6
and in several European countries7
to try to end
enduring ‘chronic’ homelessness among people
who have high rates of severe mental illness and
problematic use of drugs and alcohol. Each staircase
service has a series of steps that are designed to
make a chronically homeless person increasingly
‘ready’ for housing. Chronically homeless people
are expected to achieve goals that a staircase
service sets for them in order to progress from
one step to the next. These ‘steps’ involve moving
between accommodation, with each step allowing a
chronically homeless person more independence
until they finish the process with access to their
own home. At the end of the ‘staircase’ a chronically
homeless person is supposed to be able to live
independently in their own home and to no longer
have problems related to severe mental illness and
drug and alcohol use8
.Staircase services have strict
rules about behaviour, compliance with psychiatric
treatment and also the use of drugs and alcohol.
During the 1990s a series of evaluations in the USA
reported that staircase services were expensive to
operate and failed to end the homelessness of most
of the chronically homeless people they worked
with9
. Chronically homeless people quite often
became ‘stuck’ in staircase services, unable to ‘climb’
to the next ‘step’. In addition, chronically homeless
people were also frequently abandoning staircase
services10
because of harsh rules requiring total
abstinence from drugs and alcohol11
. Conditions
within some staircase services in the US were
also criticised as being inhuman because of their
extensive and strictly enforced rules and because
staff had a hostile and judgemental attitude towards
homeless people12
. Only a minority of chronically
6	 Collins, S.E.; Clifasefi, S.L.; Dana, E.A. et al (2011) ‘Where harm reduction meets Housing First: Exploring alcohol’s role in
a Communal Housing First setting’ International Journal of Drug Policy doi: 10.1016/j.drugpo.2011.07.010
7	 Sahlin, I. (2005) ‘The staircase of transition: Survival through failure’ Innovation 18, 2, pp.115-136; Busch-Geertsema, V.
and Sahlin, I. (2007) ‘The Role of Hostels and Temporary Accommodation’ European Journal of Homelessness 1 pp.67-
93; Johnsen, S. and Teixeira, L. (2010) Staircases, Elevators and Cycles of Change: Housing First and Other Housing
Models for People with Complex Support Needs London: Crisis.
8	 Ridgway, P. and A. M. Zipple (1990) ‘The paradigm shift in residential services: From the linear continuum to supported
housing approaches’ Psychosocial Rehabilitation Journal 13, pp. 11-31.
9	 Gulcur, L., A. Stefancic, et al. (2003). ‘Housing, hospitalization and cost outcomes for homeless individuals with psychiatric
disabilities participating in Continuum of Care and housing first programmes’Journal of Community and Applied Social
Psychology 13, 2, pp. 171-186.
10	 Bebout, R. R.; Drake, R.E. et al.(1997) ‘Housing status among formerly homeless dually diagnosed adults’ Psychiatric
Services 48, 7, pp. 936-41;
11	 Dordick, G. A. (2002) ‘Recovering from Homelessness: Determining the «Quality of Sobriety» in a Transitional Housing
Program’ Qualitative Sociology 25, 1, pp. 7-32; Carr, E. S. (2006).
«’Secrets keep you sick’: Metalinguistic labor in a drug treatment program for homeless women.» Language in Society
35, 5, pp. 631-653.
12	 Dordick, G. A. (2002) op cit; Lyon-Callo, V. (2000) ‘Medicalizing Homelessness: The Production of Self-Blame and SelfGoverning
within Homeless Shelters’ Medical Anthropology Quarterly 14, 3, pp. 328-345.
10
homeless people were being helped by expensive
staircase services that took quite a long time to
achieve results13
. Research in Europe has reported
very similar problems with staircase services in
countries that include Sweden14
. The experience
of homeless people using staircase services can
be one of a series of disruptions or ‘ruptures’ as
they move between ‘steps’ in a programme and
have to readjust to a new living environment with
new rules several times before they are eventually
able to have a settled home. This can mean that
homeless people never feel secure while they are
in the staircase, because each step except the last
step (if they progress that far) is not a permanent
home, i.e. they are unable to settle because they
know each step is temporary.
There is a view that homeless people are maintained
in an insecurity regarding their accommodation as
well as their social network by staircase services.
The whole system of support bases progression up
the staircase on the individual effort that the person
makes. The person must “earn” housing through
behaviour, conformity, and engagement with care. In
systems of care in collective shelter/accommodation,
people develop skills for group living. But once they
eventually reach independent housing, it will be
individual and personal skills that they require15
. If
unsuccessful in the progression up the staircase,
the person feels individually responsible for failing
to maintain their efforts. In this view of the staircase
system, it is not the nature of the support that is
being questioned but the failure of the person.
All observations show that one of the constants
amongst homeless people is a severing of all social
ties. The staircase system causes ruptures with the
neighbourhood, social workers etc at each step,
contributing to instability.
What is Pathways Housing First?
The Pathways Housing First service, founded by
Dr Sam Tsemberis, first appeared in New York
in 199216
Pathways Housing First (PHF) has the
following key characteristics17
:
❱❱ provides independent housing with security of
tenure immediately or as soon as possible to a
chronically homeless person.
❱❱ provides support designed to promote housing
stability and service brokerage to connect
chronically homeless people to essential services
❱❱ directly provides psychiatric, drug and alcohol,
social work and medical services. The team
providing this support is mobile and visits people
using the service in their homes or at other
agreed locations.
❱❱ gives considerable choice and control to
chronically homeless people by following a harm
reduction model. Chronically homeless people
can chose not to use the psychiatric and drug
and alcohol services that are made available by
PHF and can still remain in the housing provided
by PHF. This is called a ‘separation’ of housing
and support.
13	 Devine, J. A., Brody, C.J. et al (1997) ‘Evaluating an alcohol and drug treatment program for the homeless: An
econometric approach’ Evaluation and Program Planning 20, 2, pp. 205-215; Pleace, N. (2008) Effective Services
for Substance Misuse and Homelessness in Scotland: Evidence from an international review Edinburgh: Scottish
Government.
14	 Sahlin, I. (2005) op cit.; Busch-Geertsema, V. and Sahlin, I. (2007) op cit.
15	 Tsemberis, S. and Stefancic, A. (2007) ‘Housing First for Long-Term Shelter Dwellers with Psychiatric Disabilities in a
Suburban County: A Four- Year Study of Housing Access and Retention’ American Journal of Public Health 28, pp.
265-279.
16	 Ridgway, P. and Zipple, A.M (1990) Ibid; Tsemberis, S. (2010a) Housing First: The Pathways Model to End Homelessness
for People with Mental Illness and Addiction Center City, Minnesota: Hazelden.
17	 Tsemberis, S. (2010b) ‘Housing First: Ending Homelessness, Promoting Recovery and Reducing Costs’ in I. Gould Ellen and
B. O’Flaherty (eds) How to House the Homeless Russell Sage Foundation: New York.
11
Housing First
The PHF Philosophy
The Pathways organisation takes the view that
PHF is not simply a different way of delivering
services to chronically homeless people. Instead
Pathways believes that PHF represents a shift in the
philosophy of service delivery to homeless people
which is in part a reaction against the approach that
has been taken by staircase services. The founder
of Housing First has often been critical of what
he sees as the failure of staircase services in the
USA, criticising both their operational effectiveness
and how the staff in staircase services regard
chronically homeless people18
. The PHF philosophy
is described as following these principles::
❱❱ Housing as a basic human right.
❱❱ Respect, warmth and compassion for all clients
(a ‘client’ is a chronically homeless person using
the PHF service).
❱❱ A commitment to working with clients for as long
as they need.
❱❱ Scattered site housing, independent apartments
(that clients should live in the community in
ordinary apartments, not in a single apartment
block).
❱❱ Separation of housing and services.
❱❱ Consumer choice and self-determination.
❱❱ A recovery orientation.
❱❱ Harm reduction.
Housing First: Housing as a ‘Basic Human Right’
PHF is ‘housing first’ because the service works by
immediately providing chronically homeless people
with an apartment, or providing an apartment as
quickly as possible19
. Unlike staircase services, there
is no requirement that the chronically homeless
person completes a training programme to be made
‘housing ready’ before they move into an apartment.
In New York, apartments are provided through
an arrangement with a private landlord. In many
cases, the PHF signs a lease or tenancy agreement
with the landlord and asks the service user to sign a
sub-tenancy or sub-lease with PHF. This is intended
to guarantee the housing rights of the service user
while providing reassurance to the private landlord.
PHF avoids concentrating the people using its service
within single apartment blocks20
. The reason for
this policy is that PHF seeks to reintegrate formerly
homeless people into mainstream community life,
something that would be difficult if all PHF service
users were concentrated in one or two apartment
buildings.
It is made clear by PHF that, unlike in a staircase
service, housing is not provided as an incentive to
stop using alcohol and drugs or as an incentive to
comply with psychiatric or detoxification treatment.
Instead housing is provided as a ‘basic human right’.
This means housing is provided to someone using a
PHF service on the following basis::
❱❱ There is no requirement to stop or reduce
alcohol or drug consumption in order to receive
and to remain in housing provided via PHF.
❱❱ There is no requirement to comply with treatment
for mental health problems or with detoxification
treatments in order to receive and remain within
housing provided via PHF.
Respect, Warmth and Compassion for Clients
PHF places particular emphasis on showing
respect, warmth and compassion to the chronically
homeless people using its services. The terms
‘respect’, ‘warmth’ and ‘compassion’ are used in the
general sense. As noted, Pathways argue that this
makes their service philosophy different from that of
staircase services because of evidence that staff in
some US staircase services can have judgemental
attitudes towards homeless people21
.
18	 Tsemberis, S. (2010a) Housing First: The Pathways Model to End Homelessness for People with Mental Illness and
Addiction Hazelden: Minnesota, p. 18.
19	 Immediately where possible; sometimes there may be a short wait in temporary accommodation but this is not supposed
to exceed four weeks.
20	Ibid.
21	 Dordick, G. (2002) op cit.
12
A Commitment to Working with Clients as Long
as They Need
The PHF model is designed to provide support for
as long as is needed. There may be a point at which
services are reduced because someone is coping
well on their own and services might eventually
be withdrawn because an individual becomes fully
independent. PHF does not stop supporting a
formerly chronically homeless person if they lose
the housing they have been provided with by the
PHF service (for example, for breaking the terms
of the tenancy agreement). While the PHF service
will not keep rehousing a service user again and
again, the PHF service will give an individual who
has lost the housing they were provided with two,
three or more chances depending on their needs.
PHF services will also stay in contact with someone
if they return to homelessness, are imprisoned for
short periods or have to enter hospital (including
psychiatric care)22
.
Scattered Site Housing, Independent Apartments
The use of ordinary housing is one of the key
features of PHF. In the context of New York, there is
very little social rented housing available and there
are not the resources available to buy or develop
housing because of the costs that would be involved,
so private renting is the only option. There are
some areas of New York where PHF cannot afford
the rent for housing, which means the PHF project
is not active throughout the city. Within the financial
limitations of the PHF service, homeless people are
allowed some choice about which area they live
in and what sort of housing they have. The PHF
approach is summarised in the guidance:
This scattered site feature of the housing
model helps ensure that people with psychiatric
disabilities are not all housed together in one
building but are integrated into the buildings
and into their communities. In this model, clients
[people using the PHF service] don’t move into
a ready-made unit of a housing program –
they move into their own apartments in their
neighbourhood of choice23
.
A key goal of this approach is to promote social
inclusion. PHF operates on the assumption that in
living alongside other citizens and being within an
ordinary neighbourhood in ordinary housing, the
formerly chronically homeless people using PHF are
brought back into contact with ‘normal’ life. Support
is also provided to enable service users to establish
and maintain new friendships and relationships and
to get back in contact with their family if they have
one.
In addition, PHF seeks to reconnect formerly
chronically homeless people with paid work,
through the provision of support services that are
designed to enable access to training and education
and help with getting into employment. Through this
rehousing in the community, the fostering of positive
social and family relationships and help to become
economically active in paid work, PHF seeks to
‘resettle’ formerly chronically homeless people back
into society.
Separation of Housing and Services
A person using PHF does not have to receive
support related to severe mental illness and
problematic use of drugs and alcohol if they chose
not to. Their housing will not be affected if they
refuse to use drugs and alcohol services or to
comply with psychiatric treatment. This is described
as the ‘separation’ of housing from support, because
being given and being allowed to stay in housing is
not dependent on accepting treatment and support
services.
This ‘separation’ of housing and services is not total.
PHF service users have to agree to a weekly visit
from PHF staff to check on their well-being and to
ensure there are no problems with the apartment24
.
Consumer Choice and Self-Determination
Consumer choice and self determination refer to
the emphasis within the PHF on enabling chronically
homeless people to make their own decisions
about where and how they live their lives and, to
a considerable extent, what services and support
22	 Tsemberis, S. (2010a) op cit.
23	 Tsemberis, S. (2010a) op citp. 22.
24	 Tsemberis (2010a) op citp. 48.
13
Housing First
they use. This means that the people receiving
PHF services can act as (to use the American
terminology) as ‘consumers’ and exercise ‘self
determination’.
In practice, this means that the people receiving a
PHF service can decide what their own goals are
(this is what is meant by self-determination), for
example to drink less alcohol, to cease drinking
alcohol and, because this flexibility exists within the
PHF model, not to set themselves any goals about
how much alcohol they drink. The PHF model is built
around an assumption that chronically homeless
people will generally not, when given access to
housing and support, opt to refuse all support and
choose courses of action that are harmful to their
own well-being.
As noted, the consumer choice and selfdetermination
offered by PHF are not limitless.
Alongside the compulsory weekly support visit,
someone using a PHF service must comply with
the terms of their lease, just the same as any other
tenant. However, the level of control given to people
using PHF services is sufficient to mean that they
can and do get into difficulty:
Honouring client [people using PHF services] selfdetermination
is especially important in times of
difficulty, such as when clients deplete their financial
resources, when a landlord threatens eviction, or
when a client has relapsed into addiction. In these
situations, staff must resist the impulses to control
or resolve a chaotic situation. Instead, staff must
make every effort to help clients explore their
options during a crisis25
.
A Recovery Orientation
The ‘recovery orientation’ of PHF refers to a focus
on encouraging people using PHF services in the
right direction. This is not viewed as contradicting
the emphasis on ‘consumer choice’ in the PHF model.
Staff are expected to encourage people using PHF
services to believe that they can permanently cease
to be homeless, can stop using drugs and alcohol
and get treatment that will help with their mental
health problems.
The recovery orientation in PHF services is
designed to also encourage a greater degree of
self-reliance among the people using PHF services.
Over time, staff are expected to maximise the extent
to which people using PHF services do things for
themselves, encouraging and reinforcing their ability
to live independently.
Harm Reduction
Harm reduction is an approach to stopping people
making heavy or dangerous use of drugs and
alcohol that does not require abstinence from drugs
and alcohol. Harm Reduction is also a means of
reducing the harmful consequences of mental health
problems. The emphasis is on minimising the risks
associated with problematic behaviour and/or drug
and alcohol use and trying to reduce, with the hope
of eventually ending, those behaviours that harm an
individual.
The PHF harm reduction approach works by
emphasising to an individual the aspects of their life
that are being harmed by their behaviour. Extensive
support is made available - if a chronically homeless
person decides to use it -to help them end behaviour
that causes them harm.
From the Pathways perspective, harm reduction
should not be viewed as an ‘alternative’ approach to
treatment; the ultimate goal of harm reduction is the
same as for services trying to enforce abstinence
and compliance with treatment, to engage people
with severe mental illness with psychiatric services
and to end problematic drug and alcohol use26
.
However, harm reduction is centred on respecting
an individual’s current wishes and behaviour, with
the aim of encouraging an individual to use services
rather than ‘requiring’ an individual to use services
or stop using drugs and alcohol.
25	 Tsemberis (2010a) op citp. 27.
26	 Tsemberis (2010a) op citp. 30.
14
Delivering a Pathways Housing
First Service
There is extensive guidance on the delivery of a
PHF service in the ‘manual’ written by the founder
of PHF, Sam Tsemberis, which is entitled Housing
First: The Pathways Model to End Homelessness
for People with Mental Illness and Addiction27
. In
addition, guidance is available on the Pathways
website28
.
This section provides a summary overview of
the operation of PHF, including information on the
following areas: 
❱❱ Which groups of homeless people is a PHF
service designed to help?
❱❱ What are the aims of a PHF service?
❱❱ How is the service accessed?
❱❱ How is a PHF service delivered?
❱❱ What does a PHF service cost?
❱❱ Risk management
Which groups of homeless people is
PHF designed to help?
PHF is targeted only at chronically homeless people.
The needs, characteristics and experiences of
chronically homeless people can include:
❱❱ A disrupted childhood and/or experience of the
child care system.
❱❱ Problematic/addictive use of drugs, including
illegal drugs
❱❱ Problematic drinking of alcohol (i.e. unhealthy or
dangerous levels of consumption).
❱❱ Severe mental illness, i.e. psychiatric conditions
that are severe enough to impede capacity to
live independently, secure and sustain work and
which in some instances might represent a risk
to an individual or to those around them.
❱❱ Low level criminality, including ‘survival’ crime to
feed and clothe themselves and also crime to
sustain problematic drinking and/or drug use.
❱❱ Nuisance behaviour, sometimes linked to low
self esteem, poor mental health and to drug and
alcohol use.
❱❱ Poor physical health, linked to poor diet, alcohol
and drug consumption, poor physical environment
(i.e. living on the street and in emergency
accommodation for sustained periods).
❱❱ Sustained worklessness, linked to low educational
attainment, health and support needs.
❱❱ Alienation from mainstream society linked to poor
self-image and low self-esteem and sustained
exclusion from ordinary social and economic life.
❱❱ Repeated and sustained homelessness.
What are the aims of a PHF service?
PHF is ‘housing first’ but it is not ‘housing only’.
Through a harm reduction led philosophy that
emphasises choice and control for the people using
the PHF service, a series of goals are pursued:
❱❱ Housing stability, including developing the capacity
of formerly chronically homeless people to live
largely or wholly independently over time.
❱❱ Reductions and where possible cessation of
problematic drug and alcohol use.
❱❱ Reductions in criminal behaviour (if present).
❱❱ Reductions in severe mental illness and other
mental health problems among service users.
❱❱ Improvements in physical health by emphasizing
well-being and ensuring contact with medical
services where necessary.
❱❱ Reengagement with normal social and community
life, developing friendships, re-establishing family
ties where possible and developing and sustaining
successful personal relationships.
27	 Tsemberis, S. (2010a) op cit.
28	 www.pathwaystohousing.org
15
Housing First
❱❱ Engagement with work related activity, including
productive activity that is similar to work,
education and training as well as securing paid
work where this is possible and practical.
PHF is a service intended to deal with the
consequences of repeated and sustained
homelessness among people with high support
needs in a number of ways. Alongside tackling
homelessness through the provision of stable
housing, PHF is also intended to promote positive
outcomes in terms of social and economic inclusion
and in terms of health and well-being, using a harm
reduction framework that separates housing and
support and which emphasizes choice for people
using the PHF service.
How is the PHF service accessed?
The PHF service in New York takes referrals from
homelessness services. There can also be referrals
from psychiatric services and prisons, either when
someone has been referred to those services from a
situation of homelessness or seems likely to become
homeless on leaving. There is also the capacity to
take self-referrals from chronically homeless people
who decide they need help29
. Each individual who is
referred to PHF is assessed individually.
The process of ‘engagement’ with service users is
not something that PHF expects to be completed
very quickly. According to the PHF manual, trust
has to be established with service users who have
sometimes become used to being promised suitable
housing by staircase services, only to find a great
many conditions and requirements being placed
upon them once they started using a staircase
service. PHF therefore aims to emphasize service
users’ choice and control, seeking to give the
service user confidence in the service and adopting
an approach that is focused on asking service users
how PHF can help, rather than behaving like a
staircase service which ‘tells’ people using services
what to do30
. For this reason, the initial meeting
with a potential service user is at a venue they
choose (within safety constraints) and there is a
requirement to allow the service user to control the
pace at which their engagement with PHF moves.
Locating an apartment centres on taking into
account a service user’s preferences before making
an offer of an apartment. Service users are shown
the apartment, to check that it is acceptable to them,
before they are asked to sign a lease. In the case of
the New York PHF service, the process of securing
suitable private rented housing usually takes place
within two to four weeks.
PHF offers a full property management service
to private rented landlords, meaning all aspects of
housing management are handled by PHF, including
rent collection31
. As noted, this enables access to
private rented apartments because the lease or
tenancy is often signed by PHF, not by the person
using the service and any issues with rent collection
or other housing management issues are handled
by PHF. This means a private landlord need do
nothing but collect the rent.
How is the PHF service delivered?
PHF is not a very complex service. Essentially PHF
immediately places chronically homeless people in
ordinary rented apartments and provides them
with services that are delivered by a mobile team
of workers and professionals who visit them in
their apartments or another venue which they have
chosen.
PHF service users must accept a weekly visit,
devote 30% of their income to paying rent, and sign
a lease or sub-lease agreement for their apartment.
The three requirements of the PHF service are as
follows:
❱❱ A weekly home visit from PHF staff
❱❱ Signing a lease or sub-lease, which gives the
service user some housing rights alongside
responsibilities for the apartment they live in.
❱❱ Signing rental agreements guaranteeing 30% of
available income be devoted to payment of rent.
The core components of a PHF service are:
❱❱ Ordinary apartments. In the USA these are
usually found in the private rented sector, but
29	 Tsemberis, (2010a) op cit.
30	 Tsemberis, (2010a) op cit, p. 41.
31	 Tsemberis, (2010a) op cit.
16
PHF could also use social housing. Apartments
are furnished by the PHF project. A housing
specialist within the PHF team arranges access
to suitable housing. Chronically homeless people
using PHF must usually be in receipt of welfare
benefit payments linked to severe mental illness
in order to make a contribution to the rental
costs of their apartment.
❱❱ Support with maintaining housing stability and
living an independent life. This might include
assistance with claiming welfare benefit payments
to which a service user is entitled, help in getting
used to living independently and learning about
their neighbourhood, help managing relationships
with the private landlord, maintaining their home,
budgeting and shopping. The main mechanism
for this is the weekly home visit to each service
user in the apartment by a PHF staff member.
❱❱ An Assertive Community Treatment (ACT)
team of mobile support workers. This part of
the service is closely modelled on the ACT teams
developed in mental health services in the US.
A PHF ACT team includes a Team Leader who
coordinates the services provided, along with
a part-time psychiatrist, a part time provider of
primary medical care (either a doctor or nurse-
practitioner32
) and a full time nurse. In addition,
the ACT team will include a qualified social worker,
usually with specialist knowledge of mental health,
and specialists in supported employment, a drug
and alcohol specialist and an administrative
assistant. The ACT team must also include a
‘peer specialist’. This is an individual qualified
to provide support who has been through the
experience of chronic homelessness themselves.
Alongside providing practical support, the ‘peer
specialist’ is also seen as a ‘living illustration’
that ‘recovery’33
from chronic homelessness is
possible.34
An ACT team may also include what
is termed a ‘family specialist’; this is essentially
a worker whose role centres on positive
reconnection between a formerly chronically
homeless person and their family. In addition,
an ACT team may also include a ‘wellness
management and recovery specialist’, a role
that centres on helping a formerly chronically
homeless person develop and manage positive
personal relationships and which encourages
a generally healthy lifestyle.35
The ACT team
provides intensive support. A ten person ACT
team would be responsible for around 70
formerly chronically homeless people, a ratio of
one staff member for each seven service users.
The ACT team works with service users with the
following characteristics:
〉〉 Severe mental illness (e.g. manic depression,
bi-polar disorder) without problematic drug or
alcohol use.
〉〉 Severe mental illness and low use of drugs
and/or alcohol.
〉〉 Severe mental illness and heavy use of drugs
and/or alcohol.
❱❱ An Intensive Case Management (ICM) team.
The ICM is also based on a service model used
for people with mental health problems in the
US. The ICM team has a service brokerage
role. This means that the ICM connects a
service user with services they need that are
not directly provided by PHF, this can include
support services, medical services and welfare
benefit payments that can be accessed by any
US citizen but can also include specialist services,
such as support with problematic drug use. The
ICM refers service users to external services
and supports them in accessing those services.
The ICM team also provides some direct support
itself. ICM team staff are each assigned up to 20
service users. The ICM team works with service
users with the following characteristics:
32	 A nurse practitioner has some of the training used for doctors and is qualified to a higher level than an ordinary nurse,
though their training is less extensive than for a doctor. A nurse practitioner can prescribe some drugs.
33	 Recovery’ is a term used by PHF to describe the process by which someone ceases to be a chronically homeless person.
34	 Tsemberis, S. (2010a) op cit.
35	Ibid.
17
Housing First
〉〉 Mental health problems (e.g. depression)
without problematic drug or alcohol use.
〉〉 Mental health problems with low use of drugs
and/or alcohol.
〉〉 Mental health problems with heavy use of
drugs and alcohol.
What does PHF cost?
While it is more economical than the staircase
approach (see discussion under Evidence of
Effectiveness), the PHF service is not a low cost
service. PHF provides an extensive range of
support services. A sample budget provided in the
manual on PHF produced by Pathways suggests
that a PHF scheme can be developed and run for
one year at a cost of $1.5 million (approx. €1.23
million)36
. It is difficult to exactly cost the service
for France because the salaries, housing costs and
other operating costs will not be the same as is the
case for New York. There are important differences
in context. These centre on the amount of welfare
payments available to chronically homeless people,
which will be significantly higher in France and
also the fact that social housing is available to use
in France. Using US costs, a PHF team handling
70-80 chronically homeless people would cost
approximately €482,500 in direct salary costs for
one year of operation.
To these costs must be added: benefits for
employees, transport for the mobile workers,
furniture purchase, medical supplies, moving
expenses, apartment maintenance (on behalf of
private landlords) and costs for transport. The PHF
sample budget adds another €270,000 to cover
these costs, another €57,000 for administrative and
supervisory costs and a further €287,000 for an
office in which the team are based, taking the total
budget to about €1.23 million.
A French PHF service would probably have lower
direct costs. There are several reasons for this.
Direct salary costs may be less because France
has extensive healthcare provision that would
be accessible to service users, i.e. it might not be
necessary to directly employ as many medical
professionals in a French PHF service. If a French
PHF service were able to mobilise the social
housing stock to implement PHF, there might be
less reliance on housing specialists. In addition, if
a PHF service was operated by a medium or large
size French NGO, many administrative tasks could
be handled by existing bureaucracy, rather than
requiring dedicated staffing. Some of the additional
costs of the PHF service might be less in France,
for example public transport is much more widely
available and cheaper than in the USA.
Risk management
Delivering a service by using mobile support that is
provided to people in their own homes presents a
number of risks that are not present in a communal
service. The evidence base on how PHF manages
risk is not extensive, although there is some
descriptive information in the PHF manual37
. The
key risks that could potentially arise centre on
someone using the PHF service becoming seriously
ill, or overdosing, and the possibility that a service
user might harm themselves or (less likely) harm
someone else, including one or more staff from the
mobile support services. The techniques used to
manage risk are:
❱❱ Assessment of needs on referral to the
service. PHF would not refuse to engage with a
chronically homeless person because they were
seriously ill, very likely to overdose or presented
a potential risk to themselves or others. However,
judgements are made about how practical it is
for PHF to support each individual.
❱❱ Twenty-four availability of one member of the
ACT or ICM team by mobile telephone.
36	 Tsemberis, S. (2010a) op cit p. 219-222.
37	 Tsemberis, S. (2010a) op cit.
18
❱❱ Keys to apartments are held by the ACT/ICM
team and by the housing staff as well as by the
service user.
❱❱ Checks on service users by the PHF staff,
including the weekly visit.
Summary of PHF service delivery
❱❱ PHF works with ‘chronically homeless people’
who are characterised by severe mental illness,
problematic drug and alcohol use, nuisance
behaviour, low-level criminality, sustained
worklessness and long periods living in
homelessness shelters and on the street.
❱❱ PHF provides independent housing with security
of tenure immediately or as soon as possible to a
chronically homeless person.
❱❱ PHF provides low-level support designed to
promote housing stability, service brokerage (to
connect chronically homeless people to essential
services) and also directly provides psychiatric,
drug and alcohol, social work, medical and other
services. The team providing these services is
mobile and visits people using the service in their
homes or at other agreed locations.
❱❱ PHF services give considerable choice and
control to chronically homeless people by
following a harm reduction model which allows
them to continue drinking alcohol and using
drugs. Chronically homeless people can chose
not to use the psychiatric and drug and alcohol
services that are made available by PHF but can
remain in the housing provided by PHF. This is
called a ‘separation’ of housing and support.
Evidence of effectiveness
Housing Stability
There is very strong evidence that PHF produces
a high level of housing stability among chronically
homeless people in the US. Beyond the evidence
from the studies of the PHF in New York, there
is also strong evidence from a number of other
American cities where PHF models have been
implemented.38
PHF has delivered unprecedented levels of housing
stability among chronically homeless people, often
more than doubling the success rates reported by
staircase services. In 2004, a study on PHF in
New York reported that 80% of PHF service users
were stably housed after two years. Over time,
the success rates reported by PHF have increased,
with housing stability rates of 88% being reported39
.
By contrast, research on staircase services
reported that, at best, housing stability was being
produced for 30-40% of the chronically homeless
people using staircase services40
.The founder of the
PHF service in New York has described the main
success of PHF in very simple terms:
Housing First ends homelessness. It’s that
simple.41
Housing First has become globally important in
homelessness policy because of what PHF has
achieved in delivering housing stability for chronically
homeless people (see final chapter).
Changes in Drug and Alcohol Use
There is evidence that PHF achieves the goal of
harm reduction in drug and alcohol use among
formerly chronically homeless people. Alcohol use
38	 Pleace, N. (2008) Effective Services for Substance Misuse and Homelessness in Scotland: Evidence from an
international review Edinburgh: Scottish Government; Johnsen, S. and Teixeira, L. (2010) op cit.
39	 Tsemberis, S. (2010b) op cit.
40	 Bebout, R.R.; Drake, R.R. et al (1997) op cit; Pleace, N. (2008) op cit; Tsemberis, S. (2010b).
41	 Tsemberis, S. (2010a) op cit p. 4.
19
Housing First
and drug use tend to stabilise and also fall among
people using PHF services and there is no evidence
that PHF services produce any increase in drug or
alcohol use42
.
Some US academics have raised concerns that
PHF does not produce large enough improvements
in problematic drug and alcohol use43
. There are
arguments about whether it is better to deliver total
abstinence for a minority of chronically homeless
people, which is what ‘staircase’ services try to
deliver, or harm reduction for a greater number
(with only relatively few becoming abstinent), which
is what PHF delivers44
. There have also been
criticisms that PHF does not work with the very
heaviest users of drugs and alcohol at the same
rate as staircase services do45
.
The counter argument is that PHF tends to reduce
alcohol and drug use among many of the people
using it, most of whom remain with the PHF service.
By contrast, while a US staircase service might
eventually produce an end to drug and alcohol
use for some of the chronically homeless people
using it, such staircase services tend to eject, or
be abandoned by, between 60-70% of chronically
homeless people before an end to drug or alcohol
use has been achieved46
. Pathways has accepted
that it is not always possible for PHF to engage
with people making extremely heavy use of drugs
and alcohol, but also reports that it is working with
a group of chronically homeless people who very
often have problematic use of drugs and alcohol47
.
To set the findings of the research on PHF (and
staircase) services in context, it is important to note
that problematic drug and alcohol use are generally
quite difficult to treat. Service outcomes for people
with problematic drug use who are not homeless
may be little or no better than for service outcomes
for homeless people with drug problems48
.
Changes in Mental Health
There is evidence that housing conditions and
housing stability have direct impacts on mental
health and that the wrong housing in the wrong
neighbourhood can be ‘toxic’ to mental health. This
means there may be particular benefits to mental
health when someone is given some choice about
where they live49
. While these are difficult concepts
to measure, it seems a sense of safety, security
and privacy at home promotes good mental health.
Homelessness disrupts or removes all that is
associated with feeling ‘at home’50
.
Some research in the US by Padgett has argued
that PHF gives service users the dignity and control
of having one’s own home and home life. Padgett
has argued that in providing chronically homeless
people with their own ‘home’ in which they can make
their own choices, PHF creates a stable platform
from which a recovery from mental health problems
can begin51
.
More generally there is some evidence from
evaluative research showing that mental health tends
42	 Kertsez, S.G.; Crouch, K.; Milby. J.B.; Cusimano, R.E. and Schumacher, J.E. (2009) ‘Housing First for Homeless Persons
with Active Addiction: Are we overreaching?’ The Milbank Quarterly 87, 2, pp. 495-534; Tsai, J.; Mares, A.S. and
Rosenheck, R.A. (2010) ‘A Multisite Comparison of Supported Housing for Chronically Homeless Adults: “Housing First”
Versus “Residential Treatment First”’ Psychological Services 7, 4, pp. 219-232.; Padgett, D.K.; Stanhope, V.; Henwood,
B.F. and Stefanic, A. (2011) ‘Substance Use Outcomes Among Homeless Clients with Serious Mental Illness: Comparing
Housing First with Treatment First Programmes’ Community Mental Health 47 pp. 227-232.
43	 Lipton, F. R., Siegel, C. et al (2000) ‘Tenure in supportive housing for homeless persons with severe mental illness’
Psychiatric Services 51,4, pp. 479-86; Tsai et al (2010) op cit
44	 Collins, S.E. (2011) op cit.
45	Tsai et al (2010) op cit; Padgett et al (2011) op cit.
46	 Pearson, C. et al (2007) op cit.
47	 Tsemberis, S. (2010b) op cit.
48	 McKeganey, N., Bloor, M., Robertson, M., Neale, J. and MacDougall, J. (2006) ‘Abstinence and drug abuse treatment:
Results from the Drug Outcome Research is Scotland study’, Drugs: education, prevention and policy 13, 6, pp. 537-550.
49	 Srebnik, D.; Livingston, J.; Gordon, L.; King, D (1995) ‘Housing Choice and Community Success for Individuals with Serious
and Persistent Mental Illness’ Community Mental Health Journal, 31, 2, pp. 139-152.
50	 Padgett, D. (2007) ‘There’s no place like (a) home: Ontological security among persons with a serious mental illness in the
United States’ Social Science and Medicine 64, pp. 1925-1936
51	Ibid.
20
to improve among users of PHF services. Scores
on short questionnaires designed to assess mental
health52
also tended to improve53
. However, there is no
expectation that formerly chronically homeless people
using PHF services should comply with psychiatric
treatment, which means that access to treatment for
severe mental illness or mental health problems may
not occur for people using PHF services54
.
Criticism of the extent to which PHF can deliver
improvements to mental health takes two main
forms. First, there are the arguments that because
there is no requirement that chronically homeless
people using PHF services comply with psychiatric
treatment, mental health problems can go untreated.
The second criticism centres on what PHF services
are able to deliver in terms of housing choices.
Some researchers have the view that necessary
compromises resulting from limited funds for
PHF can mean that neither the housing nor the
neighbourhoods where the housing is located are
beneficial to mental health55
.
Social Inclusion
PHF is intended to improve the social inclusion
of chronically homeless people by housing
them as independently as possible in ordinary
neighbourhoods and communities. PHF is based
on an assumption that social inclusion is being
generated because chronically homeless people are
living in the community alongside other citizens and
are not physically separated from other people in a
separate service or apartment block.
Evidence on whether or not social inclusion is being
promoted by PHF is not extensive. There has been
some work that suggests that users of PHF ‘feel’
more included in society than was the case when
they were homeless,56
but a sense of social inclusion
can vary according to individual perception. There
is research suggesting that intolerance and hostility
towards people with support needs, or even simply
towards people on lower incomes, can be an issue
in socially and economically mixed neighbourhoods57
and this may sometimes be an issue for people
using PHF services.
Economic Inclusion
PHF seeks to promote economic inclusion, which in
the context of the US means access to education,
training and paid employment. There is evidence
that securing work can improve self-esteem,
promote social inclusion and generate benefits in
mental and physical health for formerly homeless
people. However, these benefits are associated with
work that offers fair pay and is rewarding58
.
There is no real evidence that PHF has been able
consistently to secure access to education, training or
paid employment. In part, this is because this aspect
of Housing First services has not been systematically
investigated. There is evidence that it is relatively
difficult to secure access to paid employment for
homeless people, particularly those with a criminal
record or history of drug use59
, because employers
are reluctant to employ this group of people, even
during times of relative economic prosperity.
52	 A short series of questions that are used to test mental well-being which are ‘validated’ measures that are repeatedly
tested to ensure that results are consistent and they are a good indicator of mental health. The example used in some
evaluations of PHF was the Colorado Symptom Index (CSI).
53	 Greenwood, R.M.; Schaefer-McDaniel, N.J.; Winkel, G. and Tsemberis, S. (2005) ‘Decreasing Psychiatric Symptoms by
Increasing Choice in Services for Adults with Histories of Homelessness’ American Journal of Community Psychology
36, 3/4, pp. 223-238.
54	 Tsemberis, S. (2010a) op cit.; Gilmer, T.P.; Stefanic, M.A.; Ettner, S.L.; Manning, W.G. and Tsemberis, S. (2010) ‘Effect of
Full-Service Partnership on Homelessness, Use and Costs of Mental Health Services and Quality of Life among Adults
with Serious Mental Illness’ Archive of General Psychiatry 67,6, pp. 645-652.
55	 Tabol, C.; Drebing, C. and Rosenheck, R. (2009) ‘Studies of “supported” and “supportive” housing: A comprehensive review
of model descriptions and measurement’ Evaluation and Program Planning 33 pp. 446-456
56	 Yanos, P.T.; Barrow, S.M. and Tsemberis, S. (2004) ‘Community Integration in the Early Phase of Housing Among
Homeless Persons Diagnosed with Severe Mental Illness: Successes and Challenges’ Community Mental Health Journal
40, 2, p.133- 150
57	 Bretherton, J. and Pleace, N. (2011) ‘A Difficult Mix: Issues in achieving socioeconomic diversity in deprived UK
neighbourhoods’ Urban Studies, Volume 48 Issue 16, pp. 3429 - 3443.
58	 Rio, J., Ware, L., Tucker, P., Martinez, J. (2008) Ending Chronic Homelessness through Employment and Housing: A
Program and Policy Handbook for Successfully Linking Supportive Housing and Employment Services for Chronically
Homeless Adults New York: Corporation for Supportive Housing and Advocates for Human Potential.
21
Housing First
Cost Effectiveness
In the US, PHF is seen as cost effective because
it significantly reduces the amount of time that
chronically homeless people spend outside a settled
home. Producing housing stability among formerly
chronically homeless people has four main potential
cost-saving effects60
:
❱❱ Reductions in the use of emergency homeless
shelters by chronically homeless people who
often make repeated and sustained use of
emergency homeless shelters (this has been
demonstrated). Some estimates are that the
10% of the US homeless population who are
chronically homeless may use up to 50% of
the bed spaces that emergency shelters have
available during the course of one year61
, because
they occupy beds for such long periods of time.
❱❱ Reductions in the use of emergency medical
and psychiatric services as chronically
homeless people using PHF have direct access
to psychiatric and medical services and may
experience improvements in health and wellbeing
(this has been demonstrated). Studies in
the US have suggested that there is reduced
use of emergency psychiatric treatment and that
net savings have been produced ranging from
$4,000 to $8,880 a year for each formerly
chronically homeless person using a PHF service
(€2,900 to €6,400)62
.
❱❱ Reductions in arrests and imprisonment for
chronically homeless people using PHF services,
producing savings in expenditure on criminal
justice services (this has been demonstrated).
❱❱ If PHF were to demonstrate effectiveness in
getting formerly chronically homeless people into
paid work, meaning they were no longer reliant
on welfare benefit payments, there would be a
cost saving (this has yet to be demonstrated).
Pathways has estimated the relative costs of its
services on a per-night basis and drawn a broad
comparison with other services. The comparison is
broad, because chronically homeless people would
not tend to stay in the most expensive places for
very long periods (Table 2.1).
Tableau 2.1 : Relative costs of PHF compared to alternatives in the USA according to Pathways
Service Cost per night ($) Cost per night (€)
PHF $57 $42
Emergency homeless shelter $73 $54
Jail $164 $120
Hospital ER $519 $381
Psychiatric hospital $1 185 $869
Source: Pathways63
59	 Kemp, P. and Neale, J. (2005) ‘Employability and problem drug users’, Critical Social Policy 25, 1, pp. 28-46.
60	 Culhane, D.P. (2008) ‘The Cost of Homelessness: A Perspective from the United States’ European Journal of
Homelessness 2, pp. 97-114; Culhane, D.P.; Gross, K.S.; Parker, W.D. et al (2008) ‘Accountability, Cost-Effectiveness, and
Program Performance: Progress Since 1998’ National Symposium on Homelessness Research http://works.bepress.
com/dennis_culhane/22
61	 Kuhn, R. and Culhane, D.P. (1998) ‘Applying cluster analysis to test a typology of homelessness by pattern of public shelter
utilization’ American Journal of Community Psychology 26, 2, pp. 207-232.
62	 Tsemberis, S. (2010a) op cit, p. 187-188.
63	 http://www.pathwaystohousing.org/content/our_model
22
A large scale study in New York found that 95%
of the costs of providing PHF were covered by the
savings to health, criminal justice and other services
that PHF generated. The study looked at 4,679
homeless people placed in PHF type housing in New
York and compared utilization of public shelters,
public and private hospitals, and correctional
facilities with a matched control group. Each unit of
permanent supportive housing saved $16,282 per
year in public costs for shelter, health care, mental
health, and criminal justice. These savings offset
nearly all of the $17,277 cost of implementing the
PHF package of housing and support. Almost all the
costs of providing a PHF service were effectively
met by saving money that would otherwise have
been spent elsewhere64
.
A recent review of PHF and other Housing First
costs in the US identified a potential limitation to the
cost effectiveness of PHF and other Housing First
services65
. This limitation is that the cost offsets
are greatest when PHF services are used with
chronically homeless people with the highest levels
of need. When people using PHF and other Housing
First services have slightly lower levels of need, the
financial benefits of PHF and other Housing First
services can be less clear66
. Some commentators in
the US have responded to this criticism by arguing
that purely economic evaluation of PHF and other
Housing First services is not the basis on which
policy decisions should be made67
.
64	 Culhane, D.P., Metraux, S. & Hadley, T. (2002). ‘Public service reductions associated with placement of homeless persons
with severe mental illness in supportive housing’ Housing Policy Debate, 13, 107-163.
65	 Culhane, D.P. (2010) Ending Chronic Homelessness: Cost-Effective Opportunities for Interagency Collaboration White
Paper Commissioned for New York State Office of Mental Health and NYC Department of Homelessness Services http://
works.bepress.com/dennis_culhane/94
66	 Kertesez, S.G. and Weiner, S.J. (2009) ‘Housing the Chronically Homeless: High Hopes, Complex Realities’ Journal of the
American Medical Association 301, 17, p. 1823.
67	 Culhane, D.P. (2008) op citp. 109.
23
Housing First
3.	Communal Housing First Services
Introduction
The PHF philosophy has become very influential.
However, different NGOs and governments have
taken some of the ideas on which PHF is based and
have altered those ideas to create their own versions
of ‘Housing First’. There are now many different
examples of Housing First service and one group
of services can be broadly defined as ‘Communal’
Housing First services. This chapter describes
the origin of this group of services, explores the
differences in how they operate compared to PHF,
looks at the evidence on the effectiveness of CHF
services and finally provides a description of how
CHF services are delivered.
It is important to note that this chapter describes
a range of services that operate in broadly similar
ways, rather than providing a description of a single,
specific type of service.
Origins
As the influence of PHF spread across the US,
some NGOs took some of the elements of the PHF
philosophy and applied them to staircase services.
These services retained the physical structure of a
staircase service, but the way in which the service
was delivered was very different68
. Some new
Communal Housing First services have also been
developed.
Communal Housing First (CHF) was the result of
applying Housing First principles to communal
homelessness services. CHF services give
considerable choice and control to service users,
including choices over the range of support they
receive and whether or not to stop using drugs
and alcohol. CHF offers communal housing (single
rooms or apartments) with security of tenure
provided immediately in a building only lived in by
homeless people using the CHF service. Housing
is also provided on a permanent or long-term
basis, there is no expectation that service users
will move out, which means there is no programme
to create ‘housing readiness’ or any ‘steps’ to
climb.69
These forms of ‘Housing First’ service have
started to sometimes be described as ‘ProjectBased
Housing First’ (PBHF) services in the US70
,
though it is perhaps clearer to describe them as
Communal Housing First services, in which groups
of chronically homeless people and support staff
are all within one building.
What is Communal Housing First?
CHF services are not all the same. Some services
have been developed as new services, others are
modifications of staircase services. All draw on the
ideas developed for PHF services, but different CHF
services do not all reflect the philosophy of PHF to
the same extent. CHF services do however share
some basic characteristics. The basic features of
CHF services are:71
:
❱❱ Housing in a building that is only for chronically
homeless people using the service (larger
services might employ several buildings). The
type of housing on offer varies between CHF
services, some have small self-contained
apartments, some have single rooms and some
have only semi-private spaces in communal living
areas72
. All offer some security of tenure. The
variety of living arrangements offered by CHF
services is considerable, and some would not
be considered by many commentators to be
housing in the true sense.
68	 A studio apartment contains a living area, kitchen and sleeping space, sometimes in one room and a bathroom.
69	 Pearson, C.; Locke, G. et al (2007) The Applicability of Housing First Models to Homeless Persons with Serious Mental
Illness Washington DC: HUD
70	 Collins, S.Eet al (2011) op cit;Pearson, C. et al (2007) op cit; Larimer, M.E.; Malone, D.K.; Garner, M.D. et al (2009) ‘Health
Care and Public Service Use and Costs Before and After Provision of Housing for Chronically Homeless Persons with
Severe Alcohol Problems’ Journal of the American Medical Association 301, 13, pp. 1349-1357.
71	 Pearson, C. et al (2007) op cit.
72	 Collins, S.E. et al (2011) op cit.
24
❱❱ Support is provided on site, with the staff
delivering the support being based within the
building (or nearby if more than one building
is used). The range of support on offer varies
between services. Some mirror the extensive
service provision of PHF services, including
mental health and drug and alcohol services,
other CHF services might have less extensive
support services and make more use of service
brokerage (arranging access to externally
provided services).
❱❱ Housing is either long-term or permanent; there
is no pressure or expectation on service users
to move out of the CHF service, though those
wishing to move on may be supported to do so.
❱❱ Service users have choices over whether or not
to use support services and have a role in setting
their own goals.
❱❱ CHF service users have the option to continue
drinking and using drugs, they can also refuse
to work with psychiatric services, and this does
not impact on their right to stay, i.e. people can
stay in the housing provided by the service and
continue to drink and use drugs, or chose not to
engage with mental health services.
❱❱ CHF services follow a harm reduction and
recovery orientation model.
CHF Philosophy
CHF services do not always operate in the same way
as one another. This makes a detailed discussion of
CHF philosophy difficult, because CHF services are
not consistent. However, it is possible to compare
the basic CHF philosophy with the PHF philosophy:
❱❱ Housing as a basic human right. Those CHF
services that offer a single room or a semiprivate
space in a shared environment could
be said to provide accommodation but are not
providing housing. In the case of CHF services
that offer self-contained studio apartments with
secure tenures it could be argued that ‘housing’ is
being provided. Not everyone would necessarily
share that view because concentrating formerly
homeless people in one place may segregate
them and limit their social integration73
.
❱❱ Respect, warmth and compassion for all clients
(service users).A CHF service does adopt a nonjudgemental
attitude, with staff treating chronically
homeless people as individuals worthy of respect,
warmth and compassion.
❱❱ A commitment to working with clients as long
as they need. A CHF service does make an open
ended commitment to working with homeless
people for as long as they might need the service.
Some CHF services will encourage and support
people to move on, but they are designed to offer
permanent accommodation.
❱❱ Scattered site housing, independent
apartments. A studio apartment in a CHF
accommodation block does give someone their
own front door, but it is a front door within a
block in which the staff support team and other
service users are the only neighbours. The
single rooms and the ‘semi-private’ living spaces
offered in some CHF schemes cannot offer
the privacy or independence of an apartment.
Chronically homeless people using CHF services
are also physically isolated from other citizens,
rather than living among them in the way that
PHF service users do. CHF therefore does not
offer scattered site housing and independent
apartments.
❱❱ Separation of housing and services. PHF
separates housing and services because it
gives chronically homeless people ‘housing first’
without expecting them to accept drug, alcohol
and mental health services. A CHF service does
not expect homeless people to be ‘housing ready’.
CHF services will not force service-users to
73	 Tsemberis, S. (2011) op cit.
25
Housing First
move out if they do not use the drug, alcohol or
mental health services which are offered to them.
Whether this is comparable to PHF is debatable,
because it is sometimes only accommodation –
or just one type of housing in a block with other
service users – that is being offered by the CHF
service.
❱❱ Consumer choice and self-determination. The
choices offered by PHF are not limitless: service
users have to accept a weekly visit from a
support worker, have to follow the terms of their
lease or tenancy agreement and have only some
choices about where to live set by the available
budget. However, the PHF model offers more
choice and control than a CHF model because
it is designed to provide service users with a
range of options about where and how they will
live. In addition, service users in a PHF model
are physically separated from the mobile support
services that are on offer to them, whereas the
users of CHF are generally on the same site
as the support services. CHF services offer far
more ‘consumer choice’ than a staircase service,
because they allow chronically homeless people
choices about which services to use and whether
or not to drink and use drugs, but they do not
allow chronically homeless people any choice
about where to live.
❱❱ A recovery orientation and harm reduction74
.
CHF services do follow the recovery orientation
and the harm reduction approach which together
form the seventh and eighth elements of the PHF
philosophy.
In summary, a CHF service shares many aspects
of the PHF philosophy, but it differs in one
important respect, in that some CHF services offer
‘accommodation’ or a single housing option first,
rather than genuine community-based ‘housing’ first.
Such services may offer some security of tenure
and may even be a self-contained apartment, but it
will be an apartment in an apartment block that is
designed only for chronically homeless people. This
is not the same as the ordinary apartments that are
scattered across a community which are used by
PHF services.
The Housing First philosophy and CHF
services in Finland
In February 2008, the Finnish government
introduced a strategy that was intended to halve
long term (i.e. chronic) homelessness. Finland
decided on an approach that involved extensive
use of a CHF service model75
in the context of
their strategy to reduce long-term homelessness.
A key part of the strategy was to redesign and
modify what were defined as ‘residential homes’ (i.e.
institutional communal accommodation with on site
services) to make them into ‘residential units’ that
would be supported using services that followed a
Housing First philosophy. The main goals were76
:
❱❱ to provide secure permanent housing with a
tenancy agreement to long-term (chronically)
homeless people;
❱❱ to reduce the use of conventional homelessness
shelters/emergency accommodation by changing
these services into blocks of supported, rented
apartments;
❱❱ the prevention of eviction by means of housing
advice services and financial support;
❱❱ drafting plans for individual rehabilitation and
services (i.e. involving chronically homeless
people in planning and choosing their own
support services);
❱❱ provide guidance in the use of normal social
welfare and health services (i.e. provide service
brokerage services), and
❱❱ promote social inclusion for formerly homeless
people.
74	 Tsemberis, S. (2010a) op cit, p. 18.
75	 Busch-Geertsema, V. (2010) The Finnish National Programme to Reduce Long-Term Homelessness: Discussion Paper
Peer Review: Finnish national programme to reduce long term homelessness http://www.peer-review-social-inclusion.eu/
peer-reviews/2010/the-finnish-national-programme-to-reduce-long-term-homelessness
76	 Luomanen, R. (2010) Long term homelessness reduction programme 2008-2011 Peer Review: Finnish national
programme to reduce long term homelessness http://www.peer-review-social-inclusion.eu/peer-reviews/2010/the-finnish-
national-programme-to-reduce-long-term-homelessness
26
The second goal of the Finnish strategy was very
close to the goals of CHF services in the USA78
.
Finland sought to alter the way in which existing
communal homelessness services worked by using
the PHF philosophy as a reference point. However,
while some US service providers sought only to
modify the operation of existing staircase services,
Finland embarked on an extensive programme of
refurbishment and new building that was designed
to deliver one person apartments for chronically
homeless people. There has been some policy
within debate in Finland as to how far an approach
largely based on a CHF model can fully reflect the
philosophy of PHF77
.
Delivering a Communal Housing
First Service
Unlike PHF, there is no standardised or agreed set
of guidance on how to deliver a CHF service. This
is because CHF services vary in the detail of their
operation.
This section looks at the operation of CHF in six
respects:
❱❱ Which groups of homeless people is a CHF
service designed to help?
❱❱ What are the aims of a CHF service?
❱❱ How is the service accessed?
❱❱ How is a CHF service delivered?
❱❱ What does a CHF service cost?
❱❱ Risk management
Which groups of homeless people is
a CHF service designed to help?
Like PHF services, CHF services in the USA are
designed primarily to help chronically homeless
people (see Chapter 2). In Finland, the term ‘longterm
homeless’ is used to describe the group of
homeless people that CHF services are designed
to help, but this group is defined in a very similar
way to ‘chronically homeless people’ i.e.people with
mental health problems, problematic alcohol/drug
consumption and other complex support needs
which mean they could not access housing without
support79
.
Sometimes, as in Finland, the main target group for
CHF services are chronically homeless people who
have been in the homelessness shelter system for
a long time. CHF services may also be targeted on
individuals who are judged to be at risk of chronic
homelessness, such as people with a history of
homelessness who have high support needs and
who are about to leave psychiatric hospital or the
prison system without any home of their own to go
to. Some CHF services in the USA are focused on
chronically homeless people with the very highest
support needs only. In practice this means that CHF
services are designed for:
❱❱ Chronically homeless people who have been
using homelessness services for a long time.
❱❱ Chronically homeless people with high levels of
need associated with mental health problems and
problematic alcohol and drug consumption.
❱❱ People thought to be at risk of chronic
homelessness because of their history and/or
their characteristics.
What are the aims of a CHF service?
The key aim of a CHF service is to provide stable
housing that will reduce the harm that chronic
homelessness has caused to an individual. Alongside
providing a stable and secure place in which to live,
CHFservicesareintendedtohelpminimiseandwhere
possible reduce mental health problems and the use
of drugs and alcohol that are strongly associated
with chronic homelessness. In the USA and Finland,
services are being used as a replacement for
staircase and emergency accommodation services
that were found not to be effective in ending the
homelessness of most of the people who used them.
In both countries, CHF services are primarily used
77	 Larimer, M.E. et al (2009) op cit.
78	 Social Innovation Europe interview with JuhaKaakinen 2011 http://www.socialinnovationeurope.eu/magazine/local-
development-and-communities/interviews/ending-homelessness-finland
79	 Busch-Geertsema (2010) op cit.
27
Housing First
to reduce sustained street homelessness among
very vulnerable chronically homeless people. The
key objectives are:
❱❱ To provide a stable, secure living situation that
ends a life of living on the street and in emergency
homelessness services.
❱❱ To reduce the harm caused by sustained
chronic homelessness on the people who have
experienced it, such as poor diet, exposure to
the weather and dangers of living on the street
and in emergency shelters and health problems
associated with being homeless for a long time.
❱❱ To reduce the harm of severe mental illness,
mental health problems and problematic alcohol
and drug use, which are often associated with
chronic homelessness.
❱❱ In the USA, to increase the cost effectiveness
of homelessness services by redirecting
expenditure on chronic homelessness from
homelessness shelters, emergency health
services, psychiatric services and the criminal
justice system into more effective services.
There may also be a goal to reduce expenditure
on chronic homelessness by health and criminal
justice services.
❱❱ In some situations, the reason for using CHF
service approaches as opposed to PHF is to
provide a lot of affordable housing over a relatively
short time frame using existing infrastructure.
One difference in the aims of the Finnish approach
is that the CHF services are intended for a
chronically homeless population that is viewed as
likely to decline over time. This is because a range
of preventative measures are being put in place to
reduce the rate at which people become chronically
homeless. Eventually, this could mean that provision
of CHF services will be reduced and might, at least
theoretically, cease altogether.
How is a CHF service accessed?
Referrals for CHF services come from existing
homelessness services and health, mental health
and drug and alcohol services as well as from
different parts of the criminal justice system. Some
services may be focused on a particular population,
such as the long-term homeless population in
Finland or chronically homeless people who are the
heaviest users of medical services in the USA.
How is a CHF service delivered?
The delivery of CHF services can involve the design
and building of an entirely new service, such as an
apartment block or communal housing with single
rooms, which is designed to accommodate and
support chronically homeless people with on-site
support, care and medical services. The delivery of
CHF can involve the modification of existing staircase
and other shared or communal accommodation for
homeless people. In some cases, a change is made
to the design of an existing building, for example a
shared sleeping area or single rooms in a former
staircase service might be replaced with small
apartments. In Finland, the modification of existing
emergency and short-term shelters for homeless
people into self contained apartments has created
a form of CHF.80
The most important aspect of a CHF service is the
use of a ‘Housing First’ philosophy. A CHF service
is different from a staircase or emergency shelter
service in five ways:
❱❱ Providing housing immediately without any
requirement that drug and alcohol use stop.
❱❱ The use of a harm reduction approach in
response to problematic drug and alcohol use
and mental health problems.
❱❱ Choice and control being given to chronically
homeless people using the service over whether
or not they use the drug and alcohol and mental
health services that are on offer.
80	 http://www.asuntoensin.fi/en/housing_first/ (website available in Finnish and English)
28
❱❱ The separation of tenancy and support, meaning
that chronically homeless people can stay even
if they choose not to use the mental health, drug
and alcohol services that are made available.
❱❱ Security of tenure.
❱❱ An emphasis on non-judgemental approaches
to homelessness and homeless people, which is
part of providing support within a framework of
harm reduction.
Alongside providing accommodation, a CHF service
will also offer:
❱❱ Support workers who are available on site,
sometimes on a 24-hour basis. These workers
may help access outside services that are not
provided on site, they may also provide low level
support, answering questions and providing
information as well as listening and talking to
service users.
❱❱ Drug and alcohol support services that follow
a harm reduction philosophy and which are
available on site.
❱❱ Mental health services that follow a harm
reduction philosophy and which are available on
site.
❱❱ There may be provision of services that
are designed to enable social and economic
integration; however, the available evidence
on CHF services does not suggest these are
provided by any service that has been evaluated81
.
What does a CHF service cost?
The costs of a CHF service will vary considerably.
Some services take an existing staircase service
and extensively modify its philosophy and approach,
changing how the service operates and sometimes
the range of supports it offers. What this means in
financial terms is difficult to quantify because it will
vary between services. Some staircase services
have extensive drug, alcohol and mental health
services which require an alteration in philosophy,
but which may not require any more expenditure,
in order to create a CHF service. In other cases,
where existing buildings are modified to create a CHF
service, or an entirely new building is constructed,
the development costs to provide a CHF service can
be considerable.
As noted above, while precise or typical costs
are not available because CHF services vary
considerably, the Finnish programme has created
1,250 units of CHF accommodation and some
other housing at a cost of some €201.1 million82
.
In the USA, a CHF service for which a purpose
built building was developed cost some $12 million
in building and setting up costs, with an estimated
additional typical daily cost of $37 (€28) per day for
providing services to each service user83
.
As is the case with PHF services, CHF is not
necessarily cheaper than existing homelessness
services (see further discussion under Evidence
of Effectiveness). CHF services can be relatively
expensive to develop and to deliver. While there is
evidence that CHF either cost no more than the
existing expenditure on chronic homelessness by
health and criminal justice services and may in some
cases (particularly when working with people with
particularly high support needs) cost less overall,
CHF services are not a ‘low cost’ homelessness
service.
Risk management
CHF services differ from PHF services in how they
are able to manage risk and in the nature of the risks
they face. One difference is that unlike PHF services,
a CHF service can continually monitor the well-being
and behaviour of a formerly chronically homeless
person because the support services are either in
the same building or are very close by. This may
be more of a challenge for PHF services, which can
be some distance away from a formerly chronically
homeless person who is living in an apartment in
an ordinary neighbourhood. This might mean that
the management of risks associated with formerly
chronically homeless people with very high support
81	 Social Innovation Europe interview with JuhaKaakinen 2011 http://www.socialinnovationeurope.eu/magazine/local-
development-and-communities/interviews/ending-homelessness-finland
82	 Luomanen, R. (2010) op cit
83	 Larimer, M.E. et al (2009) op cit.
29
Housing First
needs may be less challenging for a CHF service
than is the case for a PHF service. However, risk
management is not a part of CHF service delivery
which has been evaluated systematically or which
has been directly compared with PHF services.
Although it may be easier to monitor chronically
homeless people in a CHF service and for
services to reach them very quickly if a problem
or emergency occurs, the communal nature of
CHF services can also present challenges for risk
management. The challenges centre on several or
many formerly chronically homeless people sharing
communal living areas or living next to each other
in self-contained apartments. In Finland, there
are some concerns that the disruptive behaviour
of some individuals, or the disruptive behaviour of
groups of chronically homeless people associating
with one another, may be having a detrimental
effect on the effectiveness of CHF services84
. The
perceived risk in Finland has two aspects. First,
there is a concern that exposure to other drug and
alcohol users may make the process of reducing
harm associated with drug and alcohol consumption
more difficult in CHF services and second, some
chronically homeless people might find some CHF
services rather distressing environments in which
to live.
Summary of CHF service delivery
The CHF service model can be summarised as
follows:
❱❱ CHF services are focused on chronically
homeless people. These services are delivered
in communal buildings that are only lived in by
people using the CHF service. This can be single
rooms with shared living areas or individual selfcontained
apartments in an apartment block.
The accommodation has often been modified,
or designed, to provide a service for chronically
homeless people. Support and medical services
are situated in the same building or are very
nearby.
❱❱ CHF provides communal housing in a shared
building with security of tenure immediately to a
chronically homeless person.
❱❱ CHF directly provides psychiatric, drug and
alcohol services and medical services and may
also use service brokerage to arrange access to
other services.
❱❱ CHF services give considerable choice and
control to chronically homeless people as part of
following a harm reduction model. If chronically
homeless people opt not to use the services that
are on offer, or choose not to stop drinking and
using drugs, this does not place their housing in
the CHF service under threat.
Evidence on effectiveness
Housing stability
Some CHF services do not deliver true ‘housing’
stability because they do not offer self-contained
housing. CHF services may provide a single room
or semi-private area in a shared block that is
accommodation rather than “ordinary” housing. The
evidence on CHF services in the USA shows that
these services do create a stable living situation, in
that most of the service users do not often return
to homelessness85
. CHF services can end sustained
street homelessness and also sustained use of
emergency beds in homeless shelters, but while the
chronically homeless people using these services
are accommodated, it is sometimes difficult to see
them as being ‘housed’ in the sense that an ordinary
citizen in their own apartment is ‘housed’.
The extent to which CHF services are providing
‘housing’ when they offer self contained apartments
that are available only to chronically homeless
people using the CHF service is debatable. By some
criteria, CHF services like some of those in Finland
are offering ‘housing’ because people using the
service are provided with their own apartment and
have some security of tenure. Yet these services
84	 Kettunen,M. and Granfelt, R. (2011) op cit.
85	 Larimer, M.E. et al (2009) op cit; Hanratty, M. (2011) ‘Impact of Heading Home Hennepin’s Housing First programs for
long-term homeless adults’ Housing Policy Debate 21,3, pp. 405-419.
30
do not offer any choice about where to live, the
‘housing’ is only accessible to people using the
service and is physically separated from the rest of
the community because it is in one or more purpose
built or converted buildings. Staff providing services
will also often be on site. From some perspectives,
these services are delivering ‘housing’ stability, from
others, such as that of Sam Tsemberis, the founder
of PHF, these services are not actually offering what
most ordinary citizens would recognise as ‘housing’
and therefore only delivering ‘accommodation’
stability86
.
In Finland, success has been reported in creating
greater stability in housing or accommodation for
long term homeless people. Again, it is important
to note that Finnish services that follow a CHF
model may offer long term, secure, tenure that is
not always intended to be permanent. There can be
encouragement for some service users to move on
to fully independent apartments in the community.
Since the introduction of a ‘Housing First’ strategy
in Finland, overall levels of long-term homelessness
have either been reduced, or been halved, in the
participating municipalities87
. However, it has also
been noted that some of the reports of higher rates
of stability are anecdotal and that there are not as
yet extensive data showing sustained stability in
accommodation or housing88
.
Changes in Drug and Alcohol Use
The research available on CHF services in the
US suggests a pattern of stabilisation and at least
some reduction in drug and alcohol use, which is
similar to the results reported for PHF. Two studies
found that CHF services that allowed chronically
homeless people to drink alcohol in their rooms saw
this pattern, but again, did not report that alcohol
consumption had actually stopped among most
service users.89
Evidence on the Finnish experience is mixed. Some
reports suggest that there has been a reduction
in alcohol consumption in some CHF services
in Finland90
. Other research has indicated that
management of high levels of drug and alcohol
use among some residents of CHF services has
sometimes been problematic, in that tolerance of high
levels of drug and alcohol use in what is communal
housing has sometimes been difficult to manage.
There are some concerns about high rates of drug
and alcohol use in blocks of CHF apartments and it is
being suggested that more provision should be made
to enable people to move away from CHF services
and into their own independent apartments91
. These
arguments are linked to a possible limitation of CHF
services in respect of drug and alcohol use, the
evidence that exposure to high levels of use (and
therefore to various sources of supply) can make
reduction and recovery from problematic alcohol
and drug use more difficult for some people92
.
Changes in mental health
US research does not suggest that CHF services
deliver significant improvements in mental health,
although there is no evidence to suggest that mental
86	 Tsemberis, S. (2011) Observations and Recommendations on Finland’s “Name on the DoorProject” From a Housing
First Perspective Housing First Finland http://www.asuntoensin.fi/files/1242/Tsemberis_2011_-_Observations_and_
Recommendations.pdf
87	 Kettunen,M. and Granfelt, R. (2011) Observations from the first year of the Finnish Name on the door project –
recommendations for the long-term homelessness reduction programme for years 2012-2015 http://www.housingfirst.fi/
en/housing_first/reading_room/general_reading/observations_and_conclusions/ Note: While the programme has had a
specific effect in reducing long-term homelessness, overall homelessness levels in Finland have not yet fallen.
88	 Busch-Geertsema, V. (2010) op cit.
89	 Larimer, M.E. et al (2009) op cit; Collins, S.E. et al (2011) op cit.
90	 Luomanen, R. (2010) op cit, p. 31.
91	 Kettunen,M. and Granfelt, R. (2011) op cit.
92	 Pleace, N. (2008) op cit.
31
Housing First
health deteriorates as a consequence of using CHF
services93
. There is also some evidence showing
that use of emergency mental health services is
significantly reduced among chronically homeless
people using some CHF services94
.
There has not been a detailed evaluation of the
impacts of the Finnish Housing First programme
using CHF services on mental health. As already
noted, there are some emerging concerns about
whether CHF services containing a number of long
term homeless people with high support needs are
always suitable places in which to recover from
mental health problems, as well as from problematic
drug and alcohol use95
.
Social and economic inclusion
There is no clear evidence that CHF models
promote either social or economic inclusion. The
chronically homeless people using CHF services are
accommodated in physically separate blocks, which
in at least some cases are architecturally distinct
from surrounding buildings and houses. Promoting
economic inclusion, in the sense of enabling people
into work related activity, education, training or
paid employment can present a challenge for any
service providing support to chronically homeless
people, though dedicated employment programmes
have achieved some positive outcomes96
. In Finland,
there has been some discussion of how to socially
and economically include people living within CHF
services97
.
Cost effectiveness
There is evidence from the US and from Finland that
CHF services can generate significant ‘cost offsets’.
These ‘cost offsets’ are reductions in expenditure
on emergency medical services, including drug
and alcohol and mental health services as well
as hospital emergency facilities, and the result
of less contact with criminal justice systems by
chronically homeless people. When in stable and
secure accommodation, chronically homeless
people generally have less contact with emergency
services, with the police and also make much
less use of emergency homelessness shelters98
.
Development and running costs for some US and
Finnish CHF services tend to be quite high, with
a US study indicating savings of some $12 million
(€9.1 million) from a new CHF service that had cost
$11 million to develop (€8.3 million)99
and the Finnish
programme providing 1,250 units of CHF and other
housing at a cost of some €201.1 million100
.
93	 Sadowski, L.S.; Kee, R.A.; Vanderweele, T.J. et al.(2009) ‘Effect of a Housing and Service brokerage Program on
Emergency Department Visits and Hospitalizations Among Chronically Ill Homeless Adults: A Randomised Trial’ Journal
of the American Medical Association 301, 17, pp. 1771-1778.
94	 Larimer, M.E. et al (2009) op cit.
95	 Kettunen,M. and Granfelt, R. (2011) op cit.
96	 Burt, M.R. (2007) Evaluation of LA’s HOPE: Ending Chronic Homelessness through Employment and Housing Final
Report Los Angeles: City of Los Angeles Community Development Department
97	 Social Innovation Europe interview with JuhaKaakinen 2011 http://www.socialinnovationeurope.eu/magazine/local-
development-and-communities/interviews/ending-homelessness-finland
98	 Culhane, D.P. et al (2008) op cit.
99	 Larimer, M.E. et al (2009) op cit.
100	Luomanen, R. (2010) op cit.
32
4.	Housing First ‘Light’ Services
Introduction
In both Europe and in the USA, there are low-intensity
homelessness services that use mobile workers to
support formerly and potentially homeless people
living in independent apartments. These services
were sometimes developed at the same time as the
PHF service but there are also some services that
were operating before PHF existed. This group
of services has either adopted parts of the PHF
philosophy or have independently developed a
philosophy that is similar to the ideas of Housing
First. This chapter describes the origins of what it
terms ‘Housing First Light’ (HFL) services, explores
the differences between HFL and PHF services,
provides a description of how HFL services are
delivered and finally looks at the evidence on the
effectiveness of HFL services.
It is important to note that this chapter describes
a range of services that operate in broadly similar
ways, rather than providing a description of a single,
specific type of service.
Origins
HFL services immediately provide housing with
security of tenure to homeless people and support
them in their own housing by using low-intensity
mobile support services. In the UK, low intensity
mobile services were first developed to resettle
people who had been long-term residents of large
dormitory homelessness services101
.These services
were then used by social housing providers to
stop housing management problems including rent
arrears, nuisance behaviour and abandonment
of housing by ‘vulnerable’ homeless people102
.
HFL services then began to replace the use of
emergency accommodation for homeless people
with high needs, using immediate access to housing
with low intensity mobile support. Eventually these
services extended beyond the social housing sector
and began using both social and private rented
housing, becoming the single most common form of
homelessness service in the UK103
. In the USA, HFL
services have been used in experiments looking at
alternatives to the staircase services that failed to
end the majority of chronic homelessness104
.
What are HFL services?
HFL services are not a single type of service.
These services have sometimes been developed
with reference to the PHF philosophy and have
sometimes been developed wholly independently
without reference to the ideas of Housing First.
The term ‘Housing First Light’ is used to refer to
this group of services in this document, but in the
countries in which these services operate they
are known as ‘supported housing’ services, as
‘resettlement services’ as ‘tenancy sustainment’ and
as ‘homelessness support services’ and sometimes
referred to using other terminology105
. All services
of this type share a number of broad characteristics:
❱❱ Like PHF and CHF services, these services provide
‘housing first’, meaning that homeless people are
placed immediately (or as quickly as possible) into
independent housing without any requirement that
they are ‘housing ready’. Sometimes this housing
101	 Dant, T and Deacon, A (1989) Hostels to Homes? The Rehousing of Single Homeless People Aldershot: Avebury
102	‘Vulnerable’ homeless people had priority access to social housing under the terms of the 1977/1978 homelessness
laws (this group included many chronically homeless people) see Pleace, N. (1995) Housing Single Vulnerable Homeless
People York: University of York http://www.york.ac.uk/inst/chp/publications/PDF/pleace.pdf
103	Pleace, N. and Quilgars, D. (2003) Supporting People: Guide to Accommodation and Support Options for Homeless
Households London: Homelessness Directorate; Johnsen. S. and Teixeira, L. (2010) op cit.
104	Tabol, C.; Drebing, C. and Rosenheck, R. (2009) ‘Studies of “supported” and “supportive” housing: A comprehensive review
of model descriptions and measurement’ Evaluation and Program Planning 33 pp. 446-456;Caton, C.L.M; Wilkins, C. and
Anderson, J (2007) ‘People who Experience Long Term Homelessness: Characteristics and Interventions’ Paper given at
the National Symposium on Homelessness Research and accessed on 6th September 2010 at http://aspe.hhs.gov/hsp/
homelessness/symposium07/caton/index.htm
105	Tabol, C. et al (2009) op cit.
33
Housing First
is scattered throughout a municipality, city or
region, sometimes it is concentrated in specific
apartment blocks. HFL services provide access to
housing in several ways
〉〉 An HFL service might be provided directly
by a social housing provider and use its own
social housing stock.
〉〉 An HFL service may function independently
and work in cooperation with social housing
providers and private rented housing providers
to provide homeless people with housing.
〉〉 An HFL service may not have direct access to
any housing, instead it will provide support to
a homeless person or household so that they
are able to access private rented or social
housing.
❱❱ Housing is separated from support, meaning
that access to housing and security of tenure is
not linked to whether or not a homeless person
accepts the support services that are offered,
including mental health and drug and alcohol
services.
❱❱ There is no requirement for abstinence from
drug and alcohol use.
❱❱ HFL services use a service brokerage model, i.e.
support workers enable access to mainstream
health and welfare services for homeless people,
ensuring they receive all the support they are
entitled to. Support workers might also arrange
access to specialist services, such as drug and
alcohol detoxification or psychiatric services106
.
HFL services might also facilitate access to
education, work-related and employment related
services, as well as services designed to promote
social inclusion.
❱❱ HFL services use low intensity support, centring
on service brokerage and on limited support
with housing-related needs. Alongside arranging
access to welfare and health services, an HFL
worker might also provide advice and assistance
with dealing with electricity bills, ensuring the
rent is paid and ensuring a homeless person
is eating healthily. A worker might only see a
homeless person using an HFL service once
a week, or less, and HFL services are often
designed to gradually withdraw support as a
homeless person becomes better equipped to
live independently.
❱❱ An HFL service can be used for homeless people
with a range of support needs. This means
that an HFL service may support chronically
homeless people but that it may also support
homeless people with lower levels of support
need.
❱❱ An HFL service can be focused partially or
largely on homelessness prevention, such
as some UK ‘tenancy sustainment services’,
targeting people who have been identified as at
risk of homelessness who have not actually ever
been homeless.
Services following this basic model operate in a
number of EU member states107
. However, the
extent of these services is difficult to map because,
with the exception of France,108
, and a few other
countries, databases of homelessness services are
often not available at the national level. As noted,
services of the HFL type are the most common
form of homelessness service provided in the UK109
,
but these services are less common than staircase
services in the USA and may be less common than
CHF services110
.
106	Goldfinger, S. M., R. K. Schutt, et al (1999) ‘Housing placement and subsequent days homeless among formerly homeless
adults with mental illness’ Psychiatric Services 50, 5 pp. 674-9; Hickert, A.O. and Taylor, M.J. (2011) ‘Supportive Housing
for Addicted, Incarcerated Homeless Adults’ Journal of Social Service Research 37, pp. 136-151.
107	Busch-Geertsema, V. (2005). «Does re-housing lead to reintegration? Follow-up studies of re-housed homeless people.»
Innovation 18, 2, pp. 202-226.
108	Marpsat, M. (2007) 149 Documents de travail Services for the Homeless in France: Description, official statistics, client
recording of information INED
109	Centre for Housing Research (2010) Supporting People Client Records and Outcomes: Annual Report 2009-2010 (St
Andrews: University of St Andrews).
110	 Tabol, C. et al (2009) op cit.
34
HFL philosophy
HFL services vary in how they operate and in
the range of support that they provide111
. As noted
above, it is important to remember that although the
term ‘Housing First Light’ is used in this document,
there are services of this type that were developed
independently of the ideas of Housing First and,
while they reflect the philosophy of Housing First,
were operating before ‘Housing First’ services
appeared. This makes a detailed discussion of HFL
philosophy difficult, because HFL services are not
consistent, but it is possible to compare a broad
‘HFL’ philosophy with the PHF philosophy:
❱❱ Housing as a basic human right. HFL services
work by immediately providing or arranging
housing for homeless people. These services may
operate by providing mobile support services to
a single apartment block, but HFL services can
use apartments or houses scattered throughout
a municipality or city.
❱❱ Respect, warmth and compassion for all clients
(service users). An HFL service does adopt the
same non-judgemental attitude as PHF, with staff
treating homeless people as individuals worthy of
respect, warmth and compassion.
❱❱ A commitment to working with clients as
long as they need. An HFL service can make
an open ended commitment to working with
homeless people for as long as they might need
the service. HFL services will often encourage
and support people to live independently and
may be designed with the intention that support
levels are reduced over time. Some HFL services
operate with time limits, providing support for a
set period of six or nine months. This use of
set time periods of support has sometimes
been discouraged, as the needs of homeless
people are not predictable, with more flexible
arrangements being encouraged112
. For example,
arrangements by which support can be quickly
redeployed in the case of a crisis.
❱❱ Scattered site housing, independent
apartments. Many HFL services do provide
scattered, independent apartments. The extent
to which they offer choice about where to live
may vary. Some services primarily make use
of private rented housing, while others may
be more reliant on social housing, which can
limit the range of housing options available to
homeless people. Some HFL services may group
homeless people together in neighbourhoods or
in apartment blocks, which does not reflect the
PHF philosophy.
❱❱ Separation of housing and services. An HFL
service provides access to housing with security
of tenure without a requirement that service users
use drug and alcohol services or mental health
services. Unlike PHF services, HFL services
rely primarily on service brokerage, rather than
using a mixture of directly provided health and
support services and service brokerage.
❱❱ Consumer choice and self-determination. As
is the case for PHF services, HFL services do
not offer limitless housing choices. There will
be a requirement to agree to the terms of a
lease for their housing and there may also be a
requirement to agree to a minimum number of
visits by a support worker.
❱❱ A recovery orientation and harm reduction.
HFL services do follow the recovery orientation
and the harm reduction approach, which
together form the seventh and eighth elements
of the PHF philosophy.
In summary, an HFL service shares many aspects
of the PHF philosophy. However, HFL services
may not fully adopt all eight key aspects of the
PHF philosophy, compromising on issues such as
the duration for which support is provided or the
extent to which apartments are scattered. As is
the case with CHF services, a less than complete
adoption of the PHF philosophy would not been seen
by Pathways as meaning that a service could truly
111	 Tabol, C. et al (2009) op cit; Pleace, N. and Quilgars, D. (2003) op cit; Jones and Pleace (2010) op cit; Busch-Geertsema
et al (2010) op cit.
112	 Pleace, N. and Quilgars, D. (2003) op cit.
35
Housing First
be a form of ‘Housing First’113
. The point at which
a service becomes, or ceases to be, Housing First
is debatable. However, HFL services are ‘housing
first’ in the sense that they separate housing and
support, do provide ‘housing first’ and adopt a nonjudgemental
harm reduction led approach. For
these reasons, HFL services can be viewed as
following much of the PHF philosophy.
Delivering a Housing First
Light Service
This section looks at the operation of HFL services
in six respects:
❱❱ Which groups of homeless people is a HFL
service designed to help?
❱❱ What are the aims of a HFL service?
❱❱ How is the service accessed?
❱❱ How is a HFL service delivered?
❱❱ What does a HFL service cost?
❱❱ Risk management
Which groups of homeless people is
a CHF service designed to help?
HFL services can be used to support chronically
homeless people. HFL services are also employed
to support other groups of homeless people with
varying levels of need. These other groups of
homeless people can include:
❱❱ Individuals and households assessed as being
at risk of homelessness but who have not
actually become homeless. This might include
people whose support needs have placed their
housing stability at risk, for example because they
have not paid rent or created a nuisance due to
mental health problems or problematic drinking
of alcohol. Some HFL services can be largely or
entirely focused on this form of homelessness
prevention.
❱❱ Groups of homeless people with particular
needs. For example HFL services might be
focused on young homeless people, homeless and
potentially homeless people with mental health
problems or homeless families and use workers
who are specifically trained for that group of
homeless people. An HFL service for homeless
families might, for example, include workers who
were familiar with providing parental support,
education and social services for children.
In the USA, the Homelessness Prevention and Rapid
Re-housing Program (HPRP) uses what can be
broadly defined as an HFL model alongside direct
financial support of families and other households
who are at risk of becoming homeless. Workers
are provided for a period of up to 18 months for
households at risk of becoming homeless and
provide service brokerage that is focused on
ensuring housing stability for the household at
risk of becoming homeless114
. In the UK, there are
examples of HFL services that are focused on
specific groups, such as HFL services for homeless
families containing someone with high support
needs and HFL homelessness prevention services
for people who are at risk of homelessness because
nuisance behaviour means they are threatened with
eviction115
.
What are the aims of a HFL service?
The primary aim of a HFL service is to ensure
housing stability. HFL services are designed to
either prevent a return to homelessness for people
who have already been homeless, or to prevent
a potentially homeless person or household from
becoming homeless.
A HFL service may have a number of secondary
aims, which may be specific to the particular groups
of homeless people and potentially homeless people
it is designed to assist. For example, an HFL service
working with chronically homeless people will have
the goal to minimise the risks to housing stability
from severe mental illness and problematic drug and
113	 Tsemberis, S. (2011) op cit.
114	 http://www.hudhre.info/index.cfm?do=viewHprpProgram
115	 Jones, A. et al (2002) Firm Foundations: An evaluation of the Shelter Homeless to Home service, London: Shelter;
Jones, A. et al (2006) Addressing Antisocial Behaviour: An independent evaluation of the Shelter Inclusion Project,
London. Shelter.
36
alcohol use through the use of service brokerage
to arrange the necessary mental health and drug
and alcohol services. By contrast, an HFL service
for young homeless people might concern itself
particularly with the social and economic exclusion
of the young people it is working with, focusing on
access to education, training and employment as
well as promoting housing stability.
The aims of an HFL service can include:
❱❱ Promoting housing stability.
❱❱ Reductions and where possible cessation of
problematic drug and alcohol use (if present).
❱❱ Reductions in criminal behaviour (if present).
❱❱ Reduction and where possible an end to severe
mental illness and mental health problems among
service users (if present).
❱❱ Improve physical health by emphasizing wellbeing
and ensuring contact with medical services
where necessary.
❱❱ Reengagement with normal social and community
life, developing friendships, re-establishing family
ties where possible and developing and sustaining
successful personal relationships.
❱❱ Engagement with work related activity, including
productive activity that is similar to work,
education and training as well as securing paid
work where this is possible and practical.
How is an HFL service accessed?
As the operation, focus and nature of HFL services
is subject to variation, there is no set mechanism
by which referrals are made to HFL services.
Some HFL services may accept referrals from
homelessness services including emergency
shelters; daycentres or soup kitchens and HFL
services may also work in cooperation with mental
health and social work services. Some HFL services
are operated by social housing providers and may
be accessed primarily by new tenants who have
been homeless and existing tenants who may be at
risk of homelessness. It is possible that individuals
may be able to self refer to some services, but
this will depend on the specific arrangements and
which groups of homeless people an HFL service is
working with.
How is an HFL service delivered?
HFL services use a combination of mobile support
workers and ordinary housing. An HFL service
moves homeless people into housing immediately
or as soon as possible, and/or seeks to stabilise
the situation of a potentially homeless person who
is at risk of losing their existing home when it has
a homelessness prevention role. As noted, HFL
services are primarily low intensity support services
that rely heavily on service brokerage. The mobile
support workers employed by an HFL service will
provide:
❱❱ Low intensity support with the domestic
management of a home, this might include:
〉〉 Advice and support in claiming welfare benefits
〉〉 Advice and support on cooking healthy meals
〉〉 Advice and support on managing household
bills
〉〉 Advice and support on managing any existing
debt
〉〉 Support in securing furniture, kitchen
appliances such as cookers and fridges for
unfurnished housing (some services may
provide furnished apartments).
❱❱ Service brokerage, which can include helping
service users to access health care, psychiatric
services, drug and alcohol services, social work
support, legal advice, education, training, debt
and money management advice and education,
training and employment related services.
❱❱ Low intensity support with issues such as isolation
and boredom, though this will not necessarily be
offered by all HFL services.
An HFL service will not directly provide any of the
following services, though, as noted, an HFL service
will seek to arrange access to these services via
service brokerage, when these forms of support
are required:
37
Housing First
❱❱ Psychiatric services
❱❱ Health care
❱❱ Drug and alcohol services
❱❱ Social work support
❱❱ Education, training and employment related
services
❱❱ More intensive support with social isolation
One large American research review suggests
some core components that are a necessary part
of any HFL service116
:
❱❱ Settled housing should be provided immediately.
Homeless people have a right to housing and
without housing in place successful and sustained
exits from homelessness will not be possible.
❱❱ Housing provided should be affordable, adequate
and located in a suitable neighbourhood.
❱❱ No requirements or conditions about staying in
housing.
❱❱ Housing and mobile support services are
separate.
❱❱ Support staff are a mobile team, they are
physically separate from where service users
live.
❱❱ Choice in housing options for service users.
❱❱ Housing provided should be integrated with
housing for non-service users (i.e. it is scattered
across a community and not within one
apartment block, this is not an approach followed
by all HFL services).
❱❱ Housing is long term or permanent.
❱❱ Normal tenancy agreements are used.
❱❱ Housing does not look any different from housing
around it (appearance fits with neighbourhood).
❱❱ Service users have privacy (own front door).
❱❱ Individualised and flexible social support.
❱❱ 24/7 crisis cover.
❱❱ Services are nearby.
❱❱ Shared decision making, i.e. person-centred
planning and person-centred delivery of services,
which involves homeless people in the choices
made about the services they are offered and
provided with.
This list of minimum requirements does not list the
duration of the service, which in the case of PHF
and CHF models is not fixed. Alongside the other
elements necessary to deliver an HFL service,
setting no limit on the time for which formerly
chronically homeless people can have support might
also be included as an essential part of an HFL
service.
What does an HFL service cost?
HFL services do not exist in one form and have
not been subject to extensive systematic evaluation,
which means that it is difficult to be precise about
costs117
. The direct cost of providing an HFL service
can be quite low, as the service provision centres
on a mobile team of support workers, which means
that an office base, salary costs, administrative
support and a travel budget are the main forms
of expenditure. Unlike a PHF service, HFL does
not directly employ highly qualified staff such as
psychiatrists or nurse practitioners and unlike a
CHF service, there is no expenditure on developing
or renovating a specially designed building in which
to house homeless people and support services.
An HLF service will however also generate additional
costs for other services. Service brokerage by
HFL services enhances access to health, social
work, drug and alcohol services and a wide
range of other service provision for homeless and
potentially homeless people, raising the expenditure
for those services. There is the potential that an
HFL service may still reduce total expenditure by
other services, because service brokerage is being
used, rather than expensive emergency provision.
For example, it is cheaper for health services for
an HFL service to enable access to a family doctor
for a homeless person through service brokering
than to have a homeless person seek treatment in
116	 Tabol, C. et al (2010) p.448 op cit. This work reported a concern that HFL services in the USA were not consistent and
sought to describe what were thought to be the key components of a successful service.
117	 Tabol, C. et al (2010) op cit
38
the emergency facilities in a hospital. The promotion
of housing stability may also improve mental health,
help stabilize drug and alcohol consumption and
reduce contact with the criminal justice system in
much the same way as for PHF services.
Risk Management
There are a number of risk management issues
that are present when supporting vulnerable people
with complex needs in the community. These issues
centre on the potential that someone might overdose,
commit harm to themselves or to others or become
seriously ill and not be able to summon assistance.
Some HFL services will seek to manage these
risks through the assessment process, not allowing
homeless people who are judged to represent a risk
that cannot be properly monitored or managed by
an HFL service.
In some cases, HFL services supporting people with
different levels of need will provide relatively more
support to people judged to represent a higher risk
and relatively less support to those who are judged
to be low risk. The mobile support workers will also
maintain contact and check on the well-being of
people using the service on a regular basis. A well
run HFL service should use service brokerage to
ensure that were risks are present the homeless or
potentially homeless person concerned has specific
support in place. For example, if someone has risks
associated with severe mental illness, HFL workers
should ensure that person is being monitored and
cared for by appropriate psychiatric services. The
techniques used to manage risk in HFL services can
be summarised as:
❱❱ Using assessment processes to ensure that
individuals or households who represent a risk
that is difficult to manage are referred to more
appropriate services.
❱❱ Using service brokerage to ensure that, were
risks linked to health or support needs are
present, appropriate external services are in
place to monitor and manage those risks.
❱❱ Monitoring of the well-being of service users
through contacts and visits by mobile support
workers.
Summary of the HFL service
model
The HFL service model can be summarised as
follows:
❱❱ HFL is delivered by using ordinary private rented
or social housing and a team of mobile support
workers.
❱❱ HFL provides housing with security of tenure
immediately or as soon as possible to a homeless
person.
❱❱ HFL may be used to help prevent homelessness
where an individual or household who has never
been homeless is assessed as being at risk of
homelessness.
❱❱ HFL provides low-level mobile support services
designed to help promote housing stability.
❱❱ HFL uses service brokerage to arrange access
to psychiatric, drug and alcohol services and
medical services where these are required and
may also use service brokerage to arrange
access to education, training and other services
where these are needed.
❱❱ HFL services do not directly provide medical,
psychiatric or drug and alcohol services.
❱❱ HFL can support chronically homeless people but
may also be used for other groups of homeless
people, including homeless people with lower
support needs. The other groups that might
be targeted by HFL services can include young
homeless people and homeless families.
❱❱ HFL services give considerable choice and
control to homeless people as part of following
a harm reduction model. If homeless people do
not use the medical and support services which
can be arranged, or choose to continue drinking
alcohol and using drugs, this does not place their
housing under threat.
39
Housing First
Evidence on effectiveness
Housing stability
There is some evidence that HFL services can
deliver housing stability by using low-intensity
support and service brokerage. While the UK
has achieved sustained reductions in chronic
homelessness, direct evidence on the effectiveness
of the HFL services used in the UK is either out
of date, or limited to statistical data collected at the
point at which homeless people ceased to use HFL
services118
. More comprehensive evidence on HFL
services is available in the US which suggests high
rates of housing stability for formerly chronically
homeless people can be delivered by these
services119
.
It is important to remember that HFL services are
much more reliant on external services, particularly
when supporting chronically homeless people,
because they do not provide as many services as
PHF or CHF. The HFL service brokerage model
requires external psychiatric services, drug and
alcohol services and other services to be available
so that the HFL support workers can connect
homeless people to those services. An HFL service
therefore requires a ‘service rich’ environment, as it
cannot function well without access to the services
homeless people and chronically homeless people
need. This is because the mobile support workers
cannot provide all the support homeless people need
themselves and use service brokerage to arrange
it. Research within the UK has linked failures within
HFL services to required external services being
underfunded (and thus difficult to access) or because
external services, such as health and social work,
failed to cooperate with an HFL service120
. This
means that success in delivering housing stability
for an HFL service is far more dependent on what
other services are available to homeless people than
is the case for a PHF or CHF service.
Changes in drug and alcohol use
The evidence on HFL services is less extensive
than for PHF and CHF services. However, there
is some research that suggests a similar pattern
exists, in that alcohol and drug use stabilise and may
fall, but tend not to stop altogether among homeless
people using HFL services. Some research has
emphasised that HFL services can achieve high
rates of housing stability without there being any
expectation of abstinence from drugs or alcohol121
.
There is no systematic evidence that HFL services
increase drug or alcohol use.
Changes in mental health
There is evidence that housing with security of
tenure is of benefit to mental health and well-being,
but there is no systematic research on the impact
of HFL services on mental health. Some statistical
evidence from the UK indicates improvement in
mental health among homeless and potentially
homeless people using HFL services, but these
data do not employ recognised and validated clinical
measures122
. Having housing with some security of
tenure will often benefit mental health, as feeling
that there is nowhere secure to live can undermine
mental health, but this is only if that housing is
adequate and feels safe. Housing with some security
of tenure that is in poor condition or in a location
that feels unsafe, for example because of high levels
of crime, will not benefit mental health123
. In other
words, much may depend on what sort of housing
an HFL service is able to arrange for someone with
mental health problems.
Social and economic inclusion
Some statistical evidence from the UK indicates that
improvements in social and economic engagement
do occur when homeless and potentially homeless
people are using HFL services. However, these
118	 Centre for Housing Research (2010) op cit.
119	 Edens, E.L; Mares, A.S.; Tsai, J. and Rosenheck, R.A. (2011) ‘Does Active Substance Use at Housing Entry Impair
Outcomes in Supported Housing for Chronically Homeless Persons?’ Psychiatric Services 62, 2, pp. 171-178; Tabol, C.
et al (2009) op cit; Hickert, A.O. and Taylor, M.J. (2011) op cit.
120	Pleace, N. and Quilgars, D. (2003) op cit.
121	 Tabol, C. et al (2009) op cit; Hickert, A.O. and Taylor, M.J. (2011) op cit.
122	Centre for Housing Research (2010) op cit.
123	Dupuis, A. and Thorns, D. (1998) Home, home ownership and the search for ontological security, Sociological Review, 48,
pp. 24-47.
40
statistics are not based on tested measures of
social or economic inclusion and cannot be assumed
to be robust; the indicators used are also very
restricted. What evidence there is suggests limited
gains in social inclusion but no effect on economic
inclusion124
. There is no systematic evidence on the
extent to which HFL services may influence social
and economic inclusion.
Cost effectiveness
HFL services have not often been subject to detailed
cost analysis125
. The total costs are difficult to
calculate because of the reliance of HFL services
on service brokerage. While the costs of delivering
an HFL service can be calculated, the HFL will
create costs elsewhere as it connects homeless and
potentially homeless people to health, social care
and welfare services, as well as to other support
services. In addition, HFL services are also used
as a homelessness prevention service for people
who are at risk of homelessness, but who have not
actually been homeless. Cost effectiveness can be
difficult to assess because it involves projection of
the costs of homelessness that was thought to be at
risk of occurring, but which did not actually occur.
There is the potential for HFL services to reduce
overall costs on the same basis as for PHF and
CHF services, simply creating housing stability
appears to reduce the use of emergency medical
services and reduce the rate of contact that some
homeless people have with the criminal justice
system. However, PHF and CHF have their cost
effectiveness assessed on the basis of working with
chronically homeless people, individuals who tend to
have a high financial cost for emergency services.
HFL services would seem likely to generate similar
savings if working with chronically homeless people,
but, as is the case with PHF services, the savings
may be less obvious if the homeless people with
whom a HFL service is working have lower support
needs. This is because homeless people with
lower support needs may make much less use of
emergency health services and have less or little
contact with the criminal justice system, so delivering
housing stability for this group may generate less
of a financial saving. As noted above, there are
strong humanitarian reasons for not assessing
the effectiveness of HFL and other Housing First
services in simply financial terms126
.
124	Centre for Housing Research (2010) op cit.
125	Tabol, C. et al (2009) op cit; Hickert, A.O. and Taylor, M.J. (2011) op cit.
126	Culhane, D.P. (2008) op cit
41
Housing First
5.	Housing First in France
Introduction
This final chapter considers the applicability of the
Housing First service model to France. The chapter
begins by looking at the global influence of Housing
First. The second part of the chapter considers
some of the questions about the applicability of the
Housing First approach to France. The chapter
concludes by listing three reasons why it is worth
considering using Housing First services in France.
The Global Influence of Housing
First
During the 1980s, enduring street homelessness and
sustained use of shelter beds among a small group
of people with very high support needs appeared
in many societies with advanced welfare systems127
.
This ‘chronic homelessness’ was a difficult problem
to solve. Policy failure occurred at four levels:
❱❱ Chronically homeless people sometimes became
semi-permanently resident in staircase services
or other supported housing which failed to rehouse
them.
❱❱ Chronically homeless people often became
caught in a ‘revolving door’, repeatedly returning
to homelessness after making use of services,
such as staircase services, that, while they
temporarily stopped street homelessness, did not
provide a lasting solution.
❱❱ A small number of chronically homeless people
were not reached by any services or abandoned
staircase services because of their strict regimes
and requirement for abstinence from drugs and
alcohol. This group remained in the emergency
beds in homelessness shelters for very long
periods, restricting the capacity of homelessness
shelters.
❱❱ Public expenditure on attempts to counteract
chronic homelessness was often considerable
but was – at best - only partially effective.
In New York in the 1990s, Pathways Housing First
achieved something that had not been achieved
before. The PHF service targeted chronically
homeless people and produced lasting housing
stability for the majority of this group of very
vulnerable people, using a mix of ordinary private
rented apartments and mobile support services.
As variations of the Housing First model were
introduced elsewhere in the USA, including
Communal Housing First and Housing First Light
models, similar successes in achieving housing
stability were reported.
The level of success that Housing First services
have shown in delivering housing stability for
chronically homeless people has never been seen
before. It is this success that has led to Housing
First being incorporated in homelessness strategies
in Denmark, Finland, Portugal, the Netherlands and
Ireland, as well as within France.
Housing First has become very influential because
it is the first homelessness service that can clearly
demonstrate that it provides an enduring solution
that provides housing stability for the majority of
chronically homeless people. There is strong
evidence that Housing First does what no other
homelessness service has managed to achieve,
providing lasting solutions to the most extreme form
of homelessness.
The Jury for the European Consensus Conference
on Homelessness held in Brussels in December
2010 recommended that ‘housing-led’ approaches
were the most effective solution to homelessness
and that the different forms of Housing First service
were examples of these services128
. The Jury
identified as ‘housing-led’ all those approaches to
homelessness which focus on:
❱❱ Access to permanent housing as the primary
response to all forms of homelessness.
❱❱ Prevention of loss of housing.
❱❱ Provision of adequate mobile support services on
the basis of need.
127	Daly, G.P. (1996) Homeless: policies, strategies, and lives on the street London: Routledge.
128	European Consensus Conference on Homelessness: Policy Recommendations of the Jury http://ec.europa.eu/social/
main.jsp?catId=88&langId=en&eventsId=315&furtherEvents=yes
42
Questions about Using a Housing
First Approach in France
Housing First services can deliver housing or
accommodation stability for most chronically
homeless people, something that has not been
achieved by other homelessness services. This
achievement makes the Housing First model worthy
of consideration in France.
The other arguments in favour of Housing First
centre on two sets of benefits:
❱❱ Some evidence that housing stability has a positive
effect on the well-being of chronically homeless
people, including improvements in mental health,
stabilisation and some reductions in problematic
drug and alcohol use.
❱❱ Evidence that, whilst improving outcomes for
homeless people, Housing First services can
generate significant cost savings for other
services, in particular by reducing the use made
of emergency medical services, homelessness
shelters and the level of contact between
chronically homeless people and criminal justice
systems.
The idea of Housing First has met some resistance
in France129
. There are several reasons to be
cautious about the introduction of the Housing First
model in France and these include130
 :
❱❱ The limits of Housing First: While the evidence
that Housing First can deliver housing stability for
most chronically homeless people is very strong,
some critics have suggested that some models of
Housing First service cannot always work with
people who present a high risk. There are also
uncertainties about how far some Housing First
services are able to bring an end to problematic
use of drugs and alcohol and the extent to which
Housing First services are able to improve social
inclusion for homeless people.
❱❱ Housing First “is not designed” for France:
Housing First is not a French innovation.
Housing First was developed in the USA, a
country with a radically different and far more
restricted welfare system than exists in France.
The concern here is that foreign service design
cannot simply be ‘imported’ into France and
expected to work well131
.
❱❱ Housing First cannot address all forms of
homelessness: Housing First has shown the
greatest success in working with chronically
homeless people. This group may not exist in
quite the same form or to the same extent in
France, not least because its welfare system is
very different from that in the USA. In addition,
France has forms of homelessness, such as
homelessness amongst migrants with precarious
legal or administrative status, which Housing First
services are not explicitly designed for.
The remainder of this chapter will discuss each of
these reasons for caution before concluding that
Housing First has much to offer in the French
context, albeit taking account of a certain number of
risks and preconditions.
The Limits of Housing First
Housing First may not be able to deliver an end
to problematic drinking and alcohol use for all
chronically homeless people or meet all other needs
of the chronically homeless people that use these
services. The founder of PHF, Sam Tsemberis, has
himself noted there are limitations and accepted
that PHF cannot always work with individuals who
represent a serious risk or have extremely high
needs132
.
The counterargument is simple. No homelessness
service can deal with all aspects of homelessness or
meet all the needs of all homeless people. To claim
that Housing First, that any homelessness service,
can ‘solve’ all the problems of all homeless people
would not be realistic133
. The advocates of Housing
129	Houard, N. (2011) ‘The French Homelessness Strategy: Reforming Temporary Accommodation, and Access to Housing to
deliver ‘Housing First’: Continuum or Clean Break?’ European Journal of Homelessness 5, 2, pp. 83-98.
130	Pleace, N. (2011) ‘The Ambiguities, Limits and Risks of Housing First from a European Perspective’ European Journal of
Homelessness 5, 2, pp. 113-127.
131	 Johnsen, S. and Teixeira, L. (2010) op cit.
132	Tsemberis, S. (2010a) op cit.
133	Busch-Geertsema, V. (2005) op cit.
43
Housing First
First do not argue it is the solution to every aspect
of homelessness, but there is very strong evidence
that it is an answer to the most distressing and
damaging part of chronic homelessness. Housing
First stops the physical homelessness of chronically
homeless people and it stops most of them from
becoming homeless again. Having a stable home
immediately improves the situation and well-being
of chronically homeless people. Housing First may
not answer all their needs, but the majority of
chronically homeless people are in a much better
situation than they were before they had contact
with a Housing First service.
Housing First is “not designed” for
France
The second reason for caution is that Housing First
is not a French innovation; it is an ‘import’ from the
USA. It is certainly the case that a model like the New
York PHF model could not simply be transferred to
French cities without adaptation. There are aspects
of the operation of PHF in New York that make
little sense in the French context. The focus on
only private rented housing, for example, makes less
sense in France which has both social and ‘very’
social housing available. France also possesses a
far more extensive welfare and healthcare system,
meaning that the direct provision of services such as
psychiatrists and nurse practitioners in PHF might
also make less sense, particularly when service
brokerage could enable chronically homeless people
to access the extensive health and welfare services
that exist.
However, the core aspects of the Housing First
philosophy are not closely linked to any one society
or welfare system and can be adopted in a wide
range of situations. Importantly, the evidence shows
that the success of Housing First stems not from
those parts of the design that are specifically linked
to a particular welfare regime but from the general
philosophy of:
❱❱ Providing housing immediately or as soon as
possible.
❱❱ Promoting choice and control for homeless
people and treating them with compassion and
respect.
❱❱ Following a harm reduction approach and
separating housing and support services. This
means allowing homeless people to continue
drinking alcohol and using drugs - and to choose
not to use the psychiatric and drug and alcohol
services that are made available – while at the
same time allowing homeless people to remain in
the housing or accommodation provided.
The core ideas of Housing First have already been
adapted and used in flexible ways. The combination
of secure housing and mobile support services
can be used to respond to the needs of a wide
range of homeless people, including those groups
who need only limited short term assistance, those
with high needs and specific groups, such as young
homeless people or homeless families. The Finnish
implementation of Housing First has involved the
development of a Finnish service for a Finnish
context; it has not involved a simple ‘copying’ of
American services.
Much of the evidence drawn upon in this report
has been American. This is because it is the USA
where Housing First services are most developed
and where the most research has been done.
Evidence on Housing First is however developing at
European level as well as at national and local level
within Member States. As mentioned previously, the
European Consensus Conference on Homelessness
devoted much attention to housing-led and Housing
First approaches. Housing First Europe134
is a
Social Experimentation project funded in the
framework of the PROGRESS programme of the
European Commission. It will evaluate and provide
mutual learning on Housing First in 10 European
cities. The project was launched in August 2011and
will report in 2013. Its outcomes will give a more
detailed insight into the operation of Housing First
services in Europe than is possible at present. This
will inform greater clarity on the potentials and limits
of the approach and its impact on homeless people’s
lives.
In France, a major social experimentation called “Un
Chez-soi d’abord” is being carried out to test the
effectiveness of Housing First services for homeless
people with mental illness135
. A randomized control
trial over four sites (Paris, Lille, Marseille, Toulouse)
will evaluate a Pathways-type service for this target
134	http://www.servicestyrelsen.dk/housingfirsteurope
135	http://www.developpement-durable.gouv.fr/Programme-Chez-Soi-d-abord.html
44
population. The total study population will be 800
people. Half of this population will be randomly
allocated to the experimental programme and the
other half will continue to receive the conventional
service offer available to them. The evaluation will
compare both the outcomes for beneficiaries and
cost effectiveness of these services. The results will
provide a more detailed insight into the effectiveness
of the approach and the key elements for success in
the French context.
Housing First cannot address all
forms of homelessness in France
The third reason for caution centres on the
differences that may exist between the nature of
homelessness in France and homelessness in the
USA. There is almost certainly a significantly higher
rate of general homelessness and a higher rate
of ‘chronic’ homelessness in the USA. However,
survey evidence shows that the French homeless
population does appear to include a small group
of chronically homeless people, including homeless
individuals who have difficulty accessing welfare
services136
. This means it is likely that there is a
group of chronically homeless people who could
benefit from Housing First services in France.
However, there are other groups of homeless
people in France, including families and migrants
with precarious legal or administrative status.
In considering this question, it is vitally important
to note that Housing First is not the only part of
the homelessness strategy in the USA. Federal
policy in the USA assumes that there are three
groups of homeless people, which are chronically
homeless people, episodically homeless people and
transitionally (temporarily) homeless people. The
first two groups have high support needs, with
chronically homeless people having the highest
needs, and people who experience episodic
homelessness (i.e. they keep becoming homeless
again after a few weeks or months away from the
street or homelessness shelters) also having high
needs. The third group, of temporarily homeless
people, is by far the largest group and also tends
to have low needs compared to other homeless
people. This group includes individuals, couples and
families with children, all of whom are similar in
characteristics to poor people who are housed. This
third group is targeted by extensive preventative
services including the ‘Homelessness Prevention
and Rapid Re-Housing Program’ (HPRP), which
provides assistance with rent and some low level
support services that can be classified as Housing
First Light (HFL) services.
The USA recognises multiple forms of homelessness
and uses a mixture of preventative, low level and
Housing First services to counteract these different
forms of homelessness. Beyond this, the USA
also employs not just one form of Housing First
service but what are in reality multiple forms of
Housing First following the Pathways Housing
First (PHF), Communal Housing First (CHF) or
Housing First Light (HFL) approaches outlined in
this report. Housing First is therefore not seen as
the ‘only’ solution to homelessness in the USA. The
chronically homeless group which is addressed
by most Housing First services is clearly not the
only target of homelessness policy. Other forms
of homelessness exist and other services, besides
Housing First, are in place to try to deal with
those forms of homelessness. Finland, too, does
not confine its homelessness strategy or services
simply to Housing First.
There are particular issues with migrant
homelessness in France. A high proportion of the
places in homelessness services in Paris have been
reported as being taken up by Eastern European
and African migrants, both with documents and
without documents137
. An argument that has been
presented against Housing First is that it would
undermine the capacity of homelessness services
to support those service users who due to their
administrative or legal status face specific barriers
to accessing permanent housing. Housing First
136	Brousse, C. (2009) ‘The Homeless Support Network: A segmented world’ in F2009/06 Economie et Statistique: Special
Issue on the Homeless (English Version) INSEE: Paris, pp. 17-42
137	Horréard, J. P. (2007) Migration and Homelessness in Paris, Homeless in Europe Autumn 2007: Social Demographic
Change and Homelessness(Brussels: FEANTSA).
45
Housing First
in this sense has been presented as a threat to
unconditional shelter in France that could lead to
increases in homelessness in Paris and other cities.
However, this argument assumes that Housing First
would ‘replace’ all other forms of homelessness
service. This would not be the case if something like
the American or Finnish approach to Housing First
were followed in France, where Housing First is
just one type of the several forms of homelessness
services, including homelessness prevention, which
are used.
It is also worth noting that Housing First could have
a beneficial effect in increasing the capacity of
homelessness shelters and daycentres in France.
Adopting a Housing First approach does not mean
abandoning existing services where there is a need
for them. In the USA, a key argument for Housing
First has been that it reduces the use of homelessness
shelters significantly. This was because it was found
that chronically homeless people, while representing
perhaps 10% of all homeless people, stayed in
shelters for so long that they used 50% of the beds
that were available during the course of a year (i.e.
each chronically homeless person used a bed in
homeless shelter many times, meaning it could not
be used by other homeless people). As Housing First
services took chronically homeless people out of the
shelter system, those beds became available again.
Over time, it is true that it was hoped that some
reductions on spending on shelters could occur in
the USA, as fewer beds would be needed overall
due to Housing First. In France, Housing First might
be used to allow shelters to better meet the needs of
those who are homeless and are unable to access
housing immediately, such as migrants that face
barriers due to their legal or administrative status.
Why use Housing First
in France?
There are three main reasons to consider using
Housing First services in France.
❱❱ The core philosophy of Housing First is the
reason for the success that has been achieved
in delivering stable housing and accommodation
for formerly chronically homeless people. The
Housing First philosophy can be drawn upon
without copying the detailed operation of Housing
First services working in the USA. The ideas of
Housing First can be used to design services that
can be used to reduce chronic and other forms
homelessness in different countries with different
welfare systems. An example of this has been the
use of the Housing First philosophy to effectively
redesign homelessness services in Finland.
❱❱ Housing First is highly flexible. The core
philosophy has been adapted from the Pathways
Housing First (PHF) model and employed in
Communal Housing First (CHF) and Housing
First Light (HFL) services that can be focused on
both chronic homelessness and on other forms
of homelessness.
❱❱ Housing First services can support other types of
homelessness services, for example by stopping
very long stays in emergency accommodation
beds by homeless people that could access
permanent housing. It is important to note that
Housing First is not designed to replace all
existing homelessness services and has been
used as one part of a mixture of services to
respond to homelessness in the USA and Finland.
By supporting the work of other homelessness
services, Housing First services can potentially
enhance the capacity of the overall homeless
service provision and ensure that the needs of
diverse groups within the homeless population
are met.
It is important to end this discussion with a note
of caution. Housing First services are effective if
they follow the harm reduction approach and other
core elements of the Housing First philosophy. A
service is not ‘Housing First’ if it adopts some of
the core elements of this philosophy and not others.
For example a service that gives immediate access
to housing and uses mobile support services, but
which requires abstinence from drugs and alcohol
is not a Housing First service. Similarly, a service
that requires chronically homeless people or other
homeless people to complete one or more steps in a
programme to make them ‘ready for housing’, before
granting access to secure housing is still following
a staircase model, it is not Housing First. Adoption
of the core philosophy of Housing First cannot be
partial. Although there is some scope for flexibility,
immediate access to housing, the separation
of housing and support and a harm reduction
approach are crucial elements of the Housing First
philosophy and of the effectiveness of the approach.
This core philosophy can be adapted to a range of
operational contexts and service delivery models but
the key principals must be addressed in order to
end situations of homelessness.
46
notes
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
47
Housing First
notes
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
48
notes
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
............................................................................................................................................................................................................................................................
Premier Ministre
Ministère de l’Écologie,
du Développement durable,
des Transports et du Logement
Printedon100%recycledpaperstockusingbiovegetablebasedinks.Design:www.beelzepub.com
www.developpement-durable.gouv.fr www.feantsa.org