Content uploaded by Matthew Jones
Author content
All content in this area was uploaded by Matthew Jones on Feb 18, 2025
Content may be subject to copyright.
Dr. Matthew Jones and Mrs. Ella Rabaiotti
SCHOOL OF SOCIAL SCIENCES, SWANSEA UNIVERSITY
Exploring the potential of non-cash physical tokens
(‘BillyChips’) as a public health intervention
1
Title: Exploring the potential of non-cash physical tokens (‘BillyChips’) as a public health
intervention
Authors: Matthew Jones and Ella Rabaiotti, School of Social Sciences, Swansea University
Acknowledgements: This study received funding from Public Health Wales. The authors would
like to thank Public Health Wales, BillyChip and all the participants for their support and
contributions.
Date: November 2024
Final Version
List of Figures and Tables
Figure 1: Image of a BillyChip which also serves as the BillyChip logo Pg.2
Figure 2: Proposed theory of change model for BillyChip Pg.5
Table 1: Respondent characteristics Pg.14
Table 2: Interview participant sample Pg.15
Table 3: Likert scale response correlations Pg.18
Table 4: Additional purchasable product suggestions Pg.26
2
Contents Page No.
1. Executive Summary Pg.3
2. Introduction Pg.6
2.1. Overview Pg.6
2.2. Exploring poverty and homelessness Pg.6
2.3. Food insecurity, homelessness and inclusion health Pg.6
2.4. The introduction of the BillyChip scheme Pg.7
2.5. Research rationale Pg.8
2.6. Research objectives Pg.8
3. Methodology Pg.9
3.1. Research design Pg.9
3.2. Scoping review of the literature Pg.9
3.3. Survey approach Pg.10
3.4. Interviews and focus group approach Pg.10
4. Findings Pg.11
4.1. Scoping Review Findings and narrative synthesis Pg.11
4.2. Survey and Interview Findings Pg.14
4.3. Theme 1 - Responding to population needs & the case for BillyChip Pg.16
4.4. Theme 2 - Communication and Engagement of BillyChip Pg.20
4.5. Theme 3 – Place-based matters Pg.22
4.6. Theme 4 – Implementation challenges and opportunities Pg.24
4.7. Theme 5 – Developing the BillyChip scheme Pg.26
5. Discussion and Conclusion Pg.28
5.1. Discussion Pg.28
5.2. Limitations Pg.31
5.3. Concluding remarks and recommendations Pg.32
References Pg.33
Appendix 1 – Scoping review table of ndings Pg.37
Appendix 2 - Proposed Theory of Change model for BillyChip Pg.46
Appendix 3 – Copy of poster presented at WISERD conference (separate le)
Figure 1: Image of a BillyChip which also serves as the BillyChip logo
3
1. Executive Summary
Introduction
Research suggests links between homelessness, food insecurity and poor health. This study
explores the development of the BillyChip scheme amongst the homeless population, where
BillyChips can be exchanged for food and drink to help alleviate food insecurity in parts of
England and Wales. A mixed methods study, which included a scoping review of the relevant
literature, a survey (40 respondents), interviews (20 participants) and a focus group (3
participants), aimed to understand the perceived benets and pitfalls of BillyChip and wider
issues which may inuence its development and success.
Findings
A scoping review of the available literature demonstrates the paucity of peer reviewed research
evidence concerning community-based interventions for the alleviation of poverty. The most
promising avenues of research and development appear to be interventions utilising subsidies,
cash transfers, vouchers, case management and educational components. Five key themes
were identied through the qualitative ndings. Survey ndings have been integrated into these
themes – notably themes 1, 2 and 5.
Theme 1 – Responding to population needs and the case for BillyChip
All interview/focus group participants (100%, n=23) spoke about BillyChip as an intervention for
the homeless, and most (82%, n=19) saw its role for people in poverty. Other inclusion health
population groups mentioned included people with addictions (61%, n = 14) or experience of
the Criminal Justice system (43%, n = 10). After housing, the most common need identied was
food (91%, n= 21). Participants acknowledged places giving out food can be restricted to certain
days, times and locations, which didn’t help if you were ‘hungry at midnight’, according to one
health worker. Other essential needs described by over half of participants included clothes,
toiletries, and transport.
Survey respondents’ agreement with Likert scale items concerning BillyChip’s potential to ‘help
people in food poverty’, ‘helping people feel cared for’, and ‘safety and ease of exchange’ were
signicantly associated with experience of the scheme. Whilst interview and focus group
participants identied key benets of gifting and receiving a BillyChip as:
• Overcomes the barrier around giving cash to the homeless (78%, n=18).
• Addresses concerns around gifted money being used for drugs or not being needed at all
(professional begging) (52%, n=12).
• Oers choice (52%, n=12) and provides dignity (48%, n=11) to the person in need.
Theme 2 - Communication and engagement of BillyChip
Less than half of survey respondents had heard of BillyChip prior to completing the survey,
(40%, n=16), whilst a third had rst-hand experience of the scheme (33%, n=13). Most interview
participants and all focus group members (78%, n=18) articulated low awareness of BillyChip
4
and most (91%, n=21) felt more public engagement was needed. Over half (52%, n=12) saw a
promotion role for professionals working within the sector – such as signposting to BillyChip’s
‘blue symbol’.
Interview and focus group participants (39%, n=9) expressed how BillyChip can help individuals
in need beyond the initial transaction - through educating and engaging people, promoting
community safety and safeguarding. Over half (57%, n=13) felt BillyChip could play a greater
role in the broader sector. For example, six participants who worked with homeless people
ended up buying food from their own pockets and would welcome giving BillyChip.
Theme 3 – Place-based matters
When talking about places, issues raised within the qualitative research included - people being
problematic (30%, n=7); the sustainability and limited availability of facilities for the population
group (n=7) and limits around transport and access to services (n=7). Participants highlighted
‘exclusion’ of the population group (78%, n=18), including stigma, NIMBYism (‘Not In My Back
Yard’; a term to describe community opposition), stereotyping, insecurity, ineligibility,
invisibility, and being banned from places.
Most interview/focus group participants (82%, n=19) described BillyChip use in urban areas. As
well as cafes (40%, n=16) or grocery shops (39%, n=9) being seen as traditional BillyChip
outlets, participants saw community hubs or similar (65%, n=15) and schools (22%, n=5) as
potential BillyChip outlets. Many participants (61%, n=14) saw a well-known fast food bakery
chain as a key BillyChip outlet, despite its limited involvement. Participants saw this, other fast
food chains and supermarkets as important, describing them as ‘somewhere that’s
everywhere’.
Theme 4 - Implementation challenges and opportunities
Despite positive regard for the concept of BillyChip, there were negative comments from
interview participants about the implementation of the scheme (43%, n=10). The focus group
had some concerns about the impact the scheme would make in light of increasing food costs,
saying, ‘you gonna have to use them at like cheap places because you can't get a lot for two
quid these days’.
Nine interview participants (six of which had direct BillyChip experience) described negative
issues at the micro-business level ranging from provider engagement, to till operating system
issues to lost chips, errors, and inventory issues. All bar one participant highlighted some low-
level risks and issues. The most frequently aired concern was BillyChip becoming an alternative
currency longer term (48%, n=11). However participants did not think these issues should be a
barrier to developing the scheme.
Theme 5 - Developing the BillyChip scheme
Most interview participants saw the potential of increasing what BillyChip could be exchanged
for, with one observing, ‘I think expanding…will make it more useful’. Others suggested this ‘with
caution as the beauty of BillyChip is its simplicity’. Additional items suggested included clothes
5
(57%, n=13); toiletries (43%, n=10); dog food (26%, n=6); bus fare (39%, n=9). Within the survey,
an average of 3.03 (SD 3.42) purchasable items were suggested by each participant.
Interview and focus group participants saw opportunities for BillyChip in terms of expanding
retailers (52%, n=12) and its application (39%, n=9); having a role within health and well-being
and community safety sector (26%, n=6); and supporting families (17%, n=4).
Conclusions
Our exploratory study has found that participants welcome the BillyChip scheme as an
alternative to cash giving which alleviates concerns and gives recipients choice and dignity in
exchanging the chip for food or drink. This is particularly important for a population group – the
street homeless – who experience food insecurity, stigma, social exclusion and related health
issues. However, we found a low level of awareness of BillyChip and some issues with the
model working in practice (such as availability of outlets). The acceptability of BillyChip is
evident in that where people are more experienced with BillyChip, they are more likely to see the
benets it can oer. There was an emphasis on national food retailers being key to BillyChip
success and recipients being able to access food in universally welcome places. The concept
has broader applicability for the development and formalisation of other local alternative giving
schemes. Further research, including an understanding of experiences of food insecurity, as
well as the testing of the proposed Theory of Change Model (see Figure 2) could assist with the
development of BillyChip as an evidenced-led intervention.
Figure 2: Proposed theory of change model for BillyChip
6
2. Introduction
2.1 Overview
The current research explores a developing public health intervention which aims to help
alleviate the harms associated with poverty and homelessness within England and Wales. As
the rst study into this scheme, it centres on understanding the use of non-cash physical
tokens called ‘BillyChips’ (see gure 1), which can be exchanged for food and drink, and
explores alternative ways of giving, and supporting people experiencing food insecurity.
2.2 Exploring poverty and homelessness
To understand the BillyChip scheme, and the concept of alternative giving schemes more
broadly, it is necessary to understand poverty and homelessness as the context in which the
scheme has originated, and within which it operates.
We know that people experiencing poverty are at greater risk of physical and mental illness, and
of injuries including drug overdose (Mitchell et al., 2023). Complex health needs among
homeless people are compounded by diiculty in accessing healthcare (Riley et al., 2003), and
homelessness is a robust predictor of premature mortality (Demakakos et al., 2020).
Additionally, poverty and homelessness are related to risk of engagement in crime (Wu & Wu,
2012), and of being a victim of crime (especially violent crime) (Nilsson et al., 2020).
Researchers wishing to explore poverty and homelessness face numerous methodological
challenges. Firstly, accurately gauging the prevalence of poverty and homelessness is diicult.
Various prevalence gures are provided by UK governmental and third-sector charitable
organisations, each of which dier in methodology and so vary in their reported estimates.
Prevalence data do share a consensus of general increase over previous years however (Crisis,
2024; Shelter England, 2023; UK Government, 2024).
Further diiculties lie in the lack of a universally accepted denition of either poverty or
homelessness. This is perhaps typologically unavoidable due to the multiple dimensions of
poverty experienced by people, ranging from food poverty, lack of clothing or items essential for
hygiene, to digital or communications poverty. Similarly, the term homeless can apply to people
living in unstable or unsuitable conditions, or to people experiencing street homelessness e.g.
‘rough sleeping’, wherein they have no access to shelter at all. The latter group are usually those
considered non-statutory homeless, which means that these people are not oicially registered
as homeless with their local authority. Statutory homeless people whose situations are
registered with their local authority often include those who are vulnerable due to chronic
health conditions, older age, or who have dependent children and who experience unstable or
unsuitable housing conditions (UK Government, 2018).
We will present BillyChip in context, and consider the methodological challenges related to the
study of poverty and homelessness in this report.
2.3 Food insecurity, homelessness and inclusion health
People living with poverty and homelessness experience food insecurity. Food insecurity arises
from a lack of money, aects the quality and variability of diets and results in disruptive eating
7
patterns or eating less than desirable foods (Francis-Devine, 2024). Whilst there is a rise in UK
food insecurity (Francis-Devine, 2024), a recent review of research on food insecurity and
homelessness indicates a lack of national studies (Lippert & Lee, 2021). Available research
commonly indicates concerns regarding the nutritional quality of food available to homeless
people (Ravikumar et al 2021) and in one European study, those experiencing homelessness
‘rarely payed attention to food hygiene or the nutritional value of food’ (Kolanowski & Trafialek,
2021). Ravikumar et al (2022) highlighted that dietary inequalities arise from obstacles such as
lack of food supply, lack of cooking facilities as well as poverty.
Particular population groups are known to be vulnerable to UK food insecurity, including
children, disabled people and ethnically diverse populations (Francis-Devine, 2024). Inclusion
health populations are also aected, often experiencing food insecurity in combination with
other health challenges. For example, Pautz & Dempsey (2022) studied the impact of Covid-19
on those susceptible to food insecurity in the UK, including asylum seekers, people with a
disability, as well as the homeless. They found the pandemic drove a need linked to the cost-of-
living crisis, which has ‘created new and intensied food access challenges’ as well as having ‘a
signicant impact on the operation of food banks and their important 'wrap-around' services’.
Another inclusion health group, people in conict with the criminal justice system, were found
to have positive associations with reported food insecurity in one US study, as well as positive
associations between their physical health symptoms and food insecurity (Fitzpatrick & Willis,
2021).
Potential links have been made between some mental health conditions and food insecurity
among homeless adults (Loftus et al 2021). Ravikumar et al (2022) found food insecurity
resulted in perceived depressive symptoms and stress. Whilst Lachaud (2020) suggested ‘a
persistently high level of food insecurity is common among individuals with mental illness who
have experienced homelessness, and the presence of certain mental health disorders
increases this risk.’ More research is needed to understand the links between food insecurity,
homelessness and mental health (Loftus et al, 2021).
2.4 The introduction of the BillyChip scheme
Foodbanks and other related schemes such as community fridges have become commonplace
in the UK (Sosenko et al., 2022). However, as the prevalence of poverty increases and the scope
of poverty widens over time, more novel methods may be needed to manage the problem of
poverty and food insecurity. One such example is an alternative to cash scheme known as
‘BillyChip’, which originated in Bristol, UK. BillyChip aims to put coee shops and food outlets
at the centre of community support for the homeless
1
. Members of the public can buy a
BillyChip at participating outlets and pass them to a homeless person where they can be
exchanged at the same or any other participating outlet for food and drink. The social enterprise
was created in memory of 20-year-old ambulance driver Billy Abernethy-Hope, who died in
2018. It enables members of the public to give to people experiencing homelessness even if
they do not carry cash or have reservations about giving cash (Barradale, 2023). BillyChip states
that it is centred on ‘compassion, connection, choice’; the aim is to help those in need by giving
a BillyChip, creating a connection between the givers and the recipients, and giving them more
1
See ‘How it works’ on the BillyChip website for more details: https://www.billychip.com/how-it-works
8
choice than being given a specic item such as a coee directly (Barradale, 2023; BillyChip,
2023).
BillyChip is reminiscent of the Italian philanthropic practice of ‘il caè sospeso’ or ‘suspended
coee’. Customers buying a coee order an extra coee for a someone in need; Homeless people
in Naples, for example, are known to come and ask for a caè sospeso (Buscemi, 2015). Whilst
there has been some scepticism about how this practice has been used by modern coee chains
for commercial interests (Buscemi, 2015), there is recent evidence that enabling people to
access free hot drinks and a warm space during the cost-of-living crisis has wider public health
benets such as developing social connections and reducing social isolation (Rabaiotti, 2024;
Butler, 2023). In addition, other forms of non-cash credits have been known to be successful
within community work. For example, Tempo time credits, a reward and recognition scheme
where a pro-social volunteering activity such as befriending is rewarded with an activity based
reward such as cinema tickets, has been found to improve physical and mental health of
participants (Apteligen, 2020).
2.5 Research rationale
Whilst there has not been any formal research into BillyChip to date, their own business
monitoring indicates uptake and engagement has been positive, with the scheme expanding
across Bristol and the South-West, London and more recently, South Wales. Over 350 outlets are
listed on the BillyChip website
2
. The type of outlets and recipients has grown, with public and
voluntary sector community type organisations (charities, churches, and council hubs)
increasingly becoming involved. One such example of diversication has included sales of
BillyChips at a pet retail store which during its rst six weeks in operation led to £25,000 worth of
donations for homeless people and their pets
3
. Given the growing use of BillyChip, there is a need
to understand its potential as an intervention for accessing support, promoting engagement, and
improving health and well-being amongst a population known to suer poor health outcomes and
limited access to services (Folsom et al., 2005; Honer et al., 2017). Furthermore, we know that
homeless people and those in unstable housing are at an increased risk of drug and alcohol
related harms (Beiser et al., 2019; Collins et al., 2014; Doran et al., 2022), and also violent crimes
including muggings (Nilsson et al., 2020; Pain & Francis, 2004). Therefore, there is a need to
understand risks and benets to personal and community safety in respect of the use of
BillyChip. Given that BillyChip is a novel scheme with plans to expand, we expect that the lessons
we learn by carrying out exploratory research can inform the rollout of BillyChip or similar
schemes in other locales, and hopefully help rene the model to improve reach and acceptability.
2.6 Research objectives
We sought to explore the experiences of people using BillyChip, both outlets who sell and
redeem the chips and the people who benet from them. We accepted that accessing this latter
group would be diicult and therefore sought to recruit providers who help and support people
in poverty or experiencing homelessness to aid recruitment. Our primary aim and secondary
2
See ‘Find an outlet’ on the BillyChip website for outlet locations https://www.billychip.com/outlets
3
As reported within the media. For example, https://www.charity-links.co.uk/single-post/jollyes-new-
charity-partnership-generates-25k-donations-in-six-weeks
9
aims can be understood according to the following research queries: What are the perceived
benets and pitfalls of Billy Chip? What are the characteristics of BillyChip providers and
recipients? What are the wider contextual issues which may inuence the development and
success of BillyChip?
3. Methodology
3.1 Research design
Following a scoping review to provide context, we sought to explore the experiences of
professionals employed at BillyChip outlets as well as providers supporting people in poverty or
experiencing homelessness. Our target population included café and shop sta, plus health,
social welfare and charity workers. Using a mixed methods approach, the research centred on a
survey and semi-structured interviews designed to capture data related to experience and
beliefs and attitudes toward BillyChip. Participants were recruited via emails to participating
outlets and key agencies working with homeless populations. A focus group was subsequently
held with adults experiencing poverty and homelessness. The ndings were analysed separately
and then together thematically to develop a proposed Theory of Change (ToC) Model
4
and
recommendations for development of the scheme and a broader research and evaluation
project. The study was given ethical approval by Swansea University’s Faculty of Humanities
and Social Sciences Research Ethics Committee.
3.2 Scoping review of the literature
The rst step in our evaluation of the ‘BillyChip’ scheme was to conduct a scoping review of the
relevant academic literature concerning community-based interventions for harms associated
with poverty and homelessness. The ndings of this review will help us contextualise the
BillyChip scheme and assess how BillyChip ts in the existing landscape of comparable
interventions. In carrying our review, we posed the following research questions:
• How are poverty and homelessness dened in the available literature?
• What kinds of community-based interventions are described in the available literature?
• What kinds of designs, settings and outcomes are used by researchers when
investigating these interventions?
• Who are the typical beneciaries of these interventions?
• Is there any consensus amongst the literature regarding these interventions?
• What are the current knowledge gaps?
Our population of interest were people living in extreme poverty including those in unstable
housing or rough sleeping who live near each other, in a shared geographic area e.g. neighbourhood,
village or town. . The intervention of interest was broadly dened as any community-based
intervention to alleviate poverty including lack of shelter, food, heating, lighting, clothing and
shoes, medicine or healthcare, and basic toiletries. We did not place any restrictions on study
design, the inclusion of comparators or specic outcomes of interest. We sought only to include
4
Dened as ‘a comprehensive description and illustration of how and why a desired change is expected
to happen in a particular context’ (Center for Theory of Change, 2021).
10
peer-reviewed articles reporting qualitative or quantitative studies of community-based
interventions designed to alleviate poverty in our population of interest in both urban and rural
settings.
We searched ve large databases of peer-reviewed research articles pertinent to the elds of
social care and public and community-based social and interventions. Articles included in our
search were independently reviewed by the researchers for inclusion and relevant data
extracted for summary and reporting. A technical protocol outlining how we conducted this
review is available at OSF | Review protocol v0.1 24.7.2024.pdf.
3.3 Survey approach
We designed a survey to capture demographic data, data related to experience and data related
to beliefs and attitudes toward BillyChip. The survey was split into three parts. In the rst part of
the survey participants were asked to provide their age, sex, what kind of organisation they
worked for and their role within that organisation. In the second part participants were asked
whether they had heard of BillyChip, and if they answered positively they were then asked to
provide a brief description of the scheme. If they answered negatively then they were directed to
a video on the BillyChip website explaining the scheme, and then asked to summarise their
understanding of the scheme. Participants were also asked if they had any direct experience of
the scheme, and if they answered positively then they were prompted to provide a summary. To
conclude this part of the survey participants were tasked with listing any products they would
like to see available in exchange for BillyChip which were not currently available. Prompts for
items of toiletries, female hygiene products, contraception, over the counter medicines, reading
and writing materials, bus fare, phone top-up vouchers, pet food and clothing were included as
well as a free-text box for other items.
In the nal part of the survey participants were asked to answer several Likert scale questions
along a 5-point scale of ‘strongly disagree’, moderately disagree’, ‘neither agree nor disagree’,
‘moderately agree’ and ‘strongly agree’. The items were: ‘BillyChips can help people struggling to
provide food or drink’, ‘BillyChips can help people feel cared for’, Cash oer more choice than
BillyChips’, ‘Carrying BillyChips is safer than carrying cash’, ‘It is easy to exchange a BillyChip for
food or drink’, ‘BillyChips can be sold for cash’, ‘It’s easier to just give someone food or drink
than to give them a BillyChip’.
The survey was hosted on the Qualtrics online survey platform from 20/1/2024 to 27/6/2024.
Survey respondents were recruited via emails to participating outlets. Participants were
presented with a participant information sheet and a consent form upon clicking the survey link
contained in the emails. A minimum proportion of survey items was set at 60%. Responses
which returned lower than this threshold were not included at analysis. Participants were given
the opportunity to attend for follow up interviews, and to enter a rale for a £20 online shopping
voucher as a thank you for their time and eort in participating.
3.3 Interviews and focus group approach
Survey participants were invited to optionally leave their email addresses to engage in a follow-
up semi-structured interview. A small number of participants were recruited through this
11
approach. Other interview participants were recruited through a purposive sample via BillyChip,
who provided a list of potential ‘key informants’ (people engaged with the scheme), some of
whom agreed to take part in the interviews, as well as an element of snowball sampling, where
interviewees provided additional names of potential participants. Interviews primarily took
place online and lasted an average of 30 minutes each.
In addition, an in-person, hour long, focus group took place at a supported housing premises
with three adult males with experience of homelessness. Both interview and focus group
participants were asked about their understanding of BillyChip, the perceived benets,
potential pitfalls and the scheme’s potential for development.
4. Findings
4.1 Scoping review ndings and narrative synthesis
A total of 11 articles were included with dates of publication ranging from 2011 to 2022. Most
studies (8/11) were carried out in urban settlements in higher income category countries
(Australia, Spain, USA, Canada and Japan) with the remaining 3 reported studies taking place in
rural areas of middle and lower-income category countries (Panama, Uganda and Malawi).
One of the studies was cross-sectional, with the remaining 10 being carried out over a
prolonged observation period. Three of the studies were experimental (1 RCT clustered by area
and 2 individually randomized RCT), with the remaining 8 being quasi-experimental or
observational. Most of the studies utilized interviews and survey methods. Median observation
period (where data were available) was 15 months [IQR=17] with a range of 1.6 to 49 months.
Average attrition of participants between intake and follow up was 7% (SD=0.13). Median
sample size was n=65 [IQR=137] with a range between n=16 to n=1800. The average age (where
data were available) was 38.32% (SD=12.71) and an average of 41.96% (SD=0.26) of
participants were male.
Study characteristics, including design, methodology and approach to data analysis are
summarized in Appendix 1. Included articles reported outcomes related to accommodation
(e.g. access to accommodation, measures of housing security), food security, and general
income. Some of the included articles also reported outcomes outside of the scope of this
study, and these were omitted from the narrative synthesis.
Accommodation
Jacups et al (2018) reported that of all 37 study participants who engaged with the ‘Return to
Country’ or R2C scheme reported access to stable accommodation in their respective
communities. This, alongside qualitative and survey data regarding the acceptability of the
scheme to recipients led the authors to conclude that "a partial payment can be a long-term
strategy, and this is acceptable to homeless populations as a valuable service that is not
considered a ‘hand-out’.” Though these results are positive, the use of a convenience sample of
people motivated to return to their home community makes generalizing ndings diicult. In
addition, self-report bias limits validity of the survey results exacerbated by low response rates
to certain items.
Fowler & Chavira (2014) compared a voucher programme designed to help recipients purchase
decent housing with a control group. They found that families who received the vouchers were
12
signicantly less likely to report living in precarious housing or literal homelessness situations at
follow up. The authors concluded that the positive impact of the voucher program on housing
was signicant but conceded that the follow-up period of 10 months does not allow for
understanding of the longer-term outcomes of the programme, or to consider changes in family
relationships or housing needs. In addition, measurements lacked specicity ignoring
signicant variation in participant’s individual circumstances.
Marr (2012) made qualitative comparisons between participants in transitional housing
programmes (THPs) in Los Angeles and Tokyo. The author found that four of the 17 American
participants exited homelessness and secured rented accommodation, whilst seven of the 17
Japanese participants exited homelessness. In both LA and Tokyo sta allies, social ties,
minimum wage employment, and less time in homelessness were the main facilitators to
exiting homelessness. The main dierences between LA and Tokyo based participants appeared
to be in that subsidy provided by the government and nonprot organisations, along with family
support were seen as less important in Japan compared with America. The author’s qualitative
study was limited by a dissimilar sex ratio and ethnic makeup between groups, variable
observation periods for participants, and an analysis method that claims to identify causal
factors in exiting homelessness but is unable to control for confounding variables.
Kidd et al (2019) compared a complex intervention called the Housing Outreach Program—
Collaboration (HOP-C) programme which included weekly contacts with a transitional case
manager, peer support and group or if necessary individual psychotherapy wit case
management alone. Using a self-report scale of perceived housing security (Frederick et al.,
2014), the authors found no dierence between groups and concluded that the intervention
was feasible good engagement from recipients. The study was underpowered due to low
sample size, and participants with low service needs may have been overrepresented due to
youth with higher needs being less able to engage. Similarly low trust among participants meant
that blinding was diicult to apply, with intensive engagement to address concerns necessary
on behalf of study sta.
Wallace et al (2019) carried out a survey study with a cohort of homeless adults engaged in a
THP and resident at an emergency shelter. Participants received case management, rent
supplements, monthly grocery vouchers, bus passes and prolonged support where possible. It
was found that at follow up after exiting the THP 12 of 111 participants (23.5%) were in
subsidised rental housing, 17 (33.4%) were in social housing, and 12 (23.5%) were living
independently. The authors concluded that though a signicant minority of participants
achieved stable accommodation, these successes did not constitute entry into the housing
market or economic self-suiciency. In addition, individuals often remained in poverty and
reliant on subsidies. As is common when utilising survey methodologies, self-report bias, and
high attrition rates must be considered when making sense of these results.
Food security
Zamora-Sarabia et al (2019) describe an ethnographic investigation of a community-maintained
supermarket foodbank in Spain. By observation and by carrying out interviews with parent
foodbank users the authors found that the foodbank was only able to ensure basic and
nonperishable food items, and a person relying solely on the foodbank could not maintain a
high‐quality diet. They therefore concluded that such community-organised schemes should
not be relied upon in the same way as public services might be. However, the study’s ndings
may not be generalisable outside of the study population, which were mostly from immigrant
13
backgrounds living in Spain. Saturation was reached in some themes, but other themes
emerged which warrant further exploration. Finally, the study focused solely on parental
perceptions and did not capture data regarding children's perceptions of the foodbank.
Schmelzer & Leto (2018) reported on the delivery of a Participatory Action Research (PAR) study.
PAR involves the practical participation and leadership in the research study by a population
group experiencing issues relevant to the interests of the collaborating researchers. They found
that adults experiencing food poverty who engaged with an educational programme regarding
food preparation beneted from statistically signicant increases in the number of meals they
were able to make using limited ingredients according to a specialized measure developed by
the authors called the Making Meals Performance Measure (MMPM) (Schmelzer et al., 2022).
The study concluded that their ndings supported the feasibility and positive impact of
occupation-based programme using a PAR approach, but that further research was needed to
determine whether benets were maintained over time long enough to be considered sustained
food security. Limited demographic data, limited sample size, and a shorter than ideal
observation period limit the generalizability of ndings. Additionally, further work needs to be
done in validating the MMPM measure.
Miller et al (2011) carried out a longitudinal evaluation of a social cash transfer scheme (SCTS)
in Malawi which allocates cash to families in poverty to purchase food and ensure food security.
They found a 63% dierence in participants reporting less than adequate food consumption
over the past month in households allocated to the SCTS. Total food expenditures increased by
around 8% in households not allocated SCTS and by 86% in households in receipt of the SCTS.
SCTS recipient households were also signicantly more likely to eat a varied diet and have food
stores able to last a month if necessary. The authors concluded that the SCTS provides income
necessary for households to increase food expenditures, however ndings may be biased as
food buckets were given to households not receiving SCTS. It is also possible that participants
may have borrowed against future cash transfer and changed their consumption behaviour in
anticipation of the cash, or that SCTS households overstate their food expenditure motivated by
social desirability.
Income
Green et al (2016) randomised villages in rural Uganda to receive an entrepreneurship training
and grant scheme for women versus villages on the waiting list for the scheme. Sixteen months
after randomisation, monthly cash earnings doubled from UGX 15529 to 31842 (UGX 800 is
roughly US$1) for those participants assigned the intervention; signicantly higher than any
change in the control group. The authors also estimated that entrepreneurial businesses activity
doubled, and that recipient’s cash savings roughly quadrupled in the experimental group. These
ndings led the authors to conclude that though their data is encouraging, further evidence is of
needed to inform policy makers and donors in a position to roll out or fund poverty alleviation
programs in low-income countries, especially in post-conict settings. It remains unknown to
what extent these results could be generalised to other low income or post-conict settings, or
dierent socioeconomically structured societies.
Alvarez et al (2018) compared a cash transfer scheme to a scheme which provided recipients in
rural Panama with vouchers for food or domestic products with the aim of helping participants
become nancially independent. Cash recipients reported a signicantly higher income than
voucher recipients, whereas cash recipients reported improvement of household living
conditions and general health of household members. The authors concluded that their results
14
showed that autonomy-oriented nancial interventions can be empowering for recipients. The
voucher scheme was perceived as being highly acceptable to participants but did not
demonstrate success in facilitating nancial independence. Lack of baseline measurements, a
sample over representative of women, and a cross-sectional design precluding inferences
regarding causality all limit the generalisability of the ndings.
Matheson et al (2022) evaluated the experiences of recipients of a case management and
psychotherapy-based intervention for harmful gambling. The authors interviewed participants
resident at an emergency shelter for people experiencing poverty and homelessness.
Participants reported reduced debts, increased expendable incomes and better access to food
following engagement with the intervention leading the authors to conclude that the
intervention supported recovery from harmful gambling, which in turn was helpful in facilitating
nancial independence. The study was limited by a small sample, and the intervention delivery
and content were relatively unstructured, and so may be hard to reproduce in other settings.
4.2 Survey and Interview ndings
A total of 40 respondents completed the minimum proportion of survey items. Most
respondents were in the 35-44 age bracket and were female. Most respondents worked in
charitable organisations, followed by local government. A variety of dierent roles were
described by 36 respondents. Most described supporting people in poverty (35%, n=14).
Participant demographics are summarised in Table 1.
Table 1. Survey respondent characteristics
Age
18-24
6 (15%)
25-34
17 (42.5%)
35-44
8 (20%)
45-54
7 (17.5%)
54+
2 (5%)
Sex
Female
32 (80%)
Male
8 (20%)
Organisation
Charitable organisation
12 (30%)
Local
government/council
8 (20%)
Private business
6 (15%)
Housing association
4 (10%)
Health service
(including third sector
addictions services)
4 (10%)
15
Community hub/centre
(including Library)
4 (10%)
Religious organisation
2 (5%)
Role
Supportive
14 (35%)
Managerial
12 (30%)
Voluntary
5 (12.5%)
No data
4 (10%)
Administrative
2 (5%)
Healthcare
2 (5%)
Sales orientated
1 (2.5%)
A total of 20 people were interviewed and 3 participated in a focus group.
The interview sample breakdown can be seen in Table 2. Six participants were part of a BillyChip
outlet/s. The other participants were involved in supporting BillyChip or working with the client
group. Five worked in the private sector, seven in the public sector and eight in the voluntary
sector. The majority (70%, n=14) were based in South Wales.
Table 2. Interview participant sample
Participant
number
BC
Outlet
Sector
Role
P1
No
Private
Managerial
P2
No
Public
Managerial
P3
Yes
Private
Managerial
P4
Yes
Voluntary
Managerial
P5
Yes
Voluntary
Managerial
P6
No
Public
Managerial
P7
No
Private
Managerial
P8
No
Voluntary
Managerial
P9
No
Private
Managerial
P10
Yes
Voluntary
Team
member
P11
No
Public
Team
member
P12
Yes
Voluntary
Team
member
P13
No
Public
Team
member
P14
No
Voluntary
Team
member
P15
No
Public
Team
member
P16
No
Public
Team
member
P17
No
Public
Managerial
P18
No
Voluntary
Managerial
P19
Yes
Private
Managerial
P20
Yes
Voluntary
Managerial
16
Five key themes were identied through the qualitative ndings. These include:
1. Responding to population needs and the case for BillyChip
2. Communication and Engagement of BillyChip
3. Place-based matters
4. Implementation challenges and opportunities
5. Developing the BillyChip scheme
Survey ndings have been integrated into these themes – notably themes 1, 2 and 5.
Theme 1
‘Hungry at midnight’: Responding to population needs and the case for BillyChip
Participants described the population group that BillyChip responds to (or potentially could)
and their needs. There was a clear sense of how BillyChip can play a role in responding to food
insecurity within homelessness, as well as a potential wider role in supporting inclusion health.
Population group
All participants (100%, n=23) spoke about BillyChip as an intervention for the homeless, with
some focusing on street homelessness and others talking more broadly about unseen
homelessness. Participants also described people experiencing extreme poverty in all its forms
(including specic mention of those with no recourse to public funds or having issues with
welfare benets) (82%, n=19). However, when discussing the potential reach of BillyChip, there
were also a variety of other groups mentioned, including inclusion health populations:
• People with substance misuse/addiction (61%, n=14)
• People with experience of the Criminal Justice System (43%, n=10)
• Children, young people and students, and families, students (26%, n=6)
• Asylum seekers & refugees (17%, n=4)
Population needs
After housing, the most common need identied was food and having to access free food via
foodbanks or community providers (91%, n= 21). In addition, essential consumables needed
included: clothes (including underwear, socks, gloves, shoes or sandals) (52%, n=12); pet food
(26%, n=6); toiletries, hygiene products and towels (52%, n=12); transport or bus fare (52%,
n=12); and phone data/sim cards (22%, n=5). Those working most closely with the population
group had a strong understanding of the particular range of needs:
The longer you're embedded in an area, you’re meeting people in the community, you
come more aware of range of needs. So yes, its food, hygiene products, clothes,
showers, laundry, advice, haircuts. (P4)
We have a lot of people coming straight out of prison…we provide as much as we can
but …we don't provide towels or a cleaning kit (P8)
We get loads of donations come Christmas time, so loads of food, loads of
clothes…people forget they need suntan lotion in the summer and summer clothes?
Maybe ip ops? Also self care [toiletries, haircuts, towels] as well as food. (P8)
17
Members of the focus group, emphasised that they would sometimes like new clothes:
For people on the streets, when winter comes and the cold weather, there is places
where you can get, like, donated clothes but it would also be nice for people to get brand
new. And make them feel maybe a bit more normal, actually going in and buying, picking
what they want really (Focus Group member)
Food insecurity
Participants acknowledged there are places where free food is available to the homeless or
those living in extreme poverty. However, places giving out food can be restricted to certain
days, times and locations:
We've been very fortunate in [city] that we've managed to put together a package
where… seven days a week, there is food available, but that food is only available at
lunchtime pretty much, apart from on a Sunday (P13)
I was working in [city] and there was a lot more services oering free meals, hot meals in
particular, and seemed to be a lot more of a community, I suppose, for homeless
individuals. But there's not so much at all in [town]. I think we've got the [support agency
name], but I think that's only open sort of two or three days a week. So people are limited
[in where they can get food] (P16)
A few participants recognised that the population group needs more support during the night
and BillyChip could help ll the gap if available in outlets that open late
5
:
You can get food in some places, but the food is at certain times, whereas the BillyChip
allows them to get food when they're hungry and when they're needed. That's the
point…If you want something you're hungry at midnight, you can go [Fast food outlet] at
midnight. You know, if you've got a place that serves food, if it's only serving food
between 12 and two. You might not even be, you know, conscious between then, if
you're, you know, you never know, do you? (P17)
Furthermore, a lack of transport can aect opportunities to travel to access free food:
Some of our clients…will not be able to travel necessarily or have the means to travel to
some of the food banks for example outside of [city] so therefore they would miss out
(P5)
Whilst P4 explained that ‘people will travel a distance to get to us’.
BillyChip meeting population needs
When talking about participant understanding of BillyChip and how it may assist population
needs, all participants understood and supported the premise of people exchanging a BillyChip
for food and drink i.e. all participants thought it was a good concept. When talking about the
benets of gifting and receiving a BillyChip, there were two key areas highlighted by the
participants:
Firstly - It’s not cash; The main advantage for people who want to give to the homeless, is that
BillyChip can overcome the barrier around giving cash (78%, n=18), especially for members of
the public who may be wary of how the individual will use the cash. There was a particular
5
The opening hours of outlets are currently unclear on the BillyChip website directory.
18
concern from half of the participants (52%, n=12) around gifted money being used for drugs or
even not being needed at all (in reference to professional begging).
Secondly - It oers dignity and choice; For around half of the participants, they see the benet of
BillyChip for the recipient is that it oers dignity (48%, n=11) and over half of participants talked
about BillyChip giving the recipient choice over what they ate (as opposed to food donations)
(52%, n=12), with participants consistently using those particular words. This can be summed
up by P4 and P20:
[BillyChip] brings empowerment and dignity to an individual. Some people don’t want to
give money as they think they maybe facilitating a problem. BillyChip is giving someone a
choice. (P4)
The one thing I like about it is because it gives somebody on the streets a little bit of
independence so they can go and choose things to purchase rather than being given
them. And I think that… gives a level of dignity back that I think is often excluded you
when you are on the streets. (P20)
There were some examples given of people benetting from BillyChip:
There's a woman that sits in [area] and...she's always so grateful [for BillyChip] (P12)
[One outlet] get a homeless person in the area and he uses [BillyChip] every single day.
Someone goes in every day buys a chip gives it to him…it's made him a regular
customer…so it has made him sort of like the known in the community is not just a face
that's homeless, it's a person as well. (P19)
We sold 40 in the last 3 months - that’s 40 people that will receive the chip and help that
goes with it. (P5)
Survey ndings also highlighted the benets of BillyChip. In particular, agreement with Likert
scale items concerning was BillyChip’s potential to help people in food poverty, helping people
feel cared for, safety and ease of exchange were signicantly associated with experience of the
scheme. Correlations are summarised in Table 3.
Table 3. Likert scale response correlations
Statement
Correlations
Kendall's
tau B
p
Lower
95% CI
Upper
95% CI
BillyChips can help people struggling
to afford enough food or drink
Agreement
-
Experience
0.529
< .001
0.388
0.670
Agreement
-
Familiarity
0.116
0.440
-0.076
0.308
BillyChips can help people feel cared
for
Agreement
-
Experience
0.577
< .001
0.436
0.719
Agreement
-
Familiarity
0.175
0.232
-0.023
0.372
Cash oers more choice than
BillyChips
Agreement
-
Experience
-0.091
0.532
-0.245
0.062
19
Statement
Correlations
Kendall's
tau B
p
Lower
95% CI
Upper
95% CI
Agreement
-
Familiarity
-0.172
0.238
-0.363
0.018
Carrying BillyChips is safer than
carrying cash
Agreement
-
Experience
0.351
0.018
0.202
0.501
Agreement
-
Familiarity
0.035
0.815
-0.159
0.228
It is easy to exchange a BillyChip for
food or drink
Agreement
-
Experience
0.299
0.039
0.135
0.464
Agreement
-
Familiarity
-0.095
0.510
-0.294
0.103
BillyChips can be sold for cash
Agreement
-
Experience
0.021
0.881
-0.155
0.198
Agreement
-
Familiarity
-0.166
0.246
-0.366
0.033
It's easier to just give someone food or
drink than to give them a BillyChip
Agreement
-
Experience
-0.071
0.618
-0.209
0.066
Agreement
-
Familiarity
-0.081
0.571
-0.267
0.104
BillyChip resonates with humanistic values
Interview and focus group participants portrayed a sense of humanistic values about
responding to homelessness and poverty, sometimes driven by charitable (70%, n=16) including
Christian values (26%, n=6), as well as a broader community response. Members of the focus
group talked about ‘kindness’ and ‘help’ oered by individual members of the public. Whilst one
participant (P7) directly referenced the human message from BillyChip: ‘There's a message on
every [BillyChip] and it's Billy's words. You're amazing. Don't ever forget it’. And P20 said:
‘[BillyChip] is raising that awareness that actually these are people, it could be you. You know,
we're all two mortgage payments away from that situation’.
Alternatives to BillyChip
Interview and focus group participants did mention alternative initiatives to BillyChip within the
discussions (56%, n=13). This included community places which provide free food, warm hubs,
diverted giving schemes, community fridge, gift cards or vouchers, free bus tickets, donated
bags of shopping, and paid forward coees. For example, one public sector worker said:
Warm hubs already provide refreshments for free, no exchange or token required. During
funded school meal provision in summer, some organisations chose to buy vouchers
which worked well for the asylum seeker cohort. (P2)
20
Theme 2:
‘Look for the blue symbol’: Communication and Engagement of BillyChip
Less than half of respondents had heard of BillyChip prior to completing the survey (40%, n=16),
and close to a third had rst-hand experience of the scheme (32.5%, n=13). However, as
previously indicated, the survey also found that the more people had experience of the scheme,
the greater they viewed it as a benecial intervention.
Most interview participants (78%, n=18) articulated a low or limited level of awareness of
BillyChip. P9 said ‘I hadn't actually heard of Billy Chip, and that's what surprised me because we
do work in the [homeless] community’. The focus group had not heard of BillyChip at all and
thought a homeless person being given the chip on its own was insuicient:
People would be like, what’s this? Its money? People may think they really come from a
gambling place. I thought it was a poker chip. When you look properly you know then. It
should have more information to say how much its worth really. It could be clearer.
Indeed, one participant involved in outreach and distributing BillyChips explained how raising
awareness and understanding with the recipients can take time:
When we rst started handing them out, a few people knew [about BillyChip], but they
were very few…People were a bit suspicious. But we said give it a try and we told them of
various places where they might be able to exchange them and explained about looking
for the blue symbol on the door of the café. (P12)
Most participants (91%, n=21) felt there was much to do around engaging the public with
BillyChip. One participant (P4) who works directly with homeless people said they ‘did not see a
lot of signage’ and ‘do not think the general public know enough about how it works’. They
emphasised the importance of word of mouth with the client group, saying ‘they tell each other.
That’s the best form of advertising’. P4 added:
I think it needs a lot of advertising. People need to feel ready condent and clear in how
do it and why it a good thing, why it's better than giving someone couple pounds or giving
them nothing at all.
Getting this right was clearly particularly important for those working in the homelessness and
poverty sector. As P2 said, this includes dealing with ‘sensitivities around the messaging around
the scheme, what it means and what it implies and who the message is coming from’. Over half
of participants (52%, n=12) saw a role for professionals working within the sector to help
promote the scheme. For example, participants explained:
[I] talked to the volunteers about [BillyChip] so everyone was on board, this can make or
break or really stunt somethings eectiveness (P4)
We introduced it to all our workforce, support workers, ...we have a leaet that they can
give out to somebody in need which actually tells them the places where they can
redeem the chip and what they need to do and where they can go (P5)
21
Around half of participants (48%, n=11) had direct engagement with the BillyChip team and
described them in positive terms. For example, one participant said:
They're lovely people…they're expanding…extremely busy. I have no problems with them
at all. They’re very approachable and very professional. (P6)
However, for a few participants that engagement had levelled o. They ‘hadn’t heard from them
for months’ and noted that the team member who was engaging regularly had moved on.
BillyChip supporting work of public and voluntary sector
The potential mutual benets of BillyChip (‘its not cash’ and ‘provides dignity/choice’, and the
Likert scale positive associations for those experienced with the scheme), also extends to the
context in which the scheme operates. A range of professionals have an interest in BillyChip –
health, homelessness, poverty, community safety – and there was evidence to suggest that
BillyChip can play a role in supporting their work. Over a third of the participants (39%, n=9)
expressed how BillyChip can help individuals in need in a broader way. For example, through
educating and engaging people – both those in need and the wider public by ‘raising
understanding of homelessness’ (P2). This was particularly important for those involved in
outreach work who can use BillyChip as a tool for engagement; ‘if they’ve got BillyChips in their
pocket, they're able to talk to people' (P10) and ‘it encourages people on the street who are very
isolated to come and have a chat’ (P20). This appears of interest to police forces in three
locations where BillyChip is used. For example:
Its been brought up in the Community safety partnership meetings, the sargeant… is
very, very keen. He’s happy for his oicers and PCSOs to have BillyChips…It's a good
way then for the PCSOs and street wardens interacting with the youth in the area (P6)
The people who were handing it out were [Redacted] Police. They were kind of…support
teams who were actively out…talking to the people who were rough sleeping (P7)
The focus group took this a step further, seeing BillyChip as having a safeguarding role for the
homeless; being 'safer for addicts, so people would be less likely to be robbed if they had
BillyChips as opposed to if they had been given a lot of money'. Whereas P6 saw BillyChip as
having a role of safeguarding the community:
One of my primary concerns was the amount of money that was being handed over by
elderly people to the people begging and I seen elderly people go to the bank, take £20
out and give them to the people of the begging and they were quite aggressive on times.
Whilst there is limited evidence that BillyChip deters professional begging
6
, one participant
described an interaction where his 'mate gave this guy who begs a billychip but he refused it, he
doesn’t give him money anymore as he knows he's a professional begger' (P1). However, the
focus group was less certain about the potential impact of BillyChip on substance misusers:
‘These people begging for money, they're still going to get the money for the drugs either
way. It doesn't matter through crime, shoplifting or begging.’
6
The BillyChip team have received feedback that professional begging has reduced in areas where
BillyChip is used in concentration.
22
Despite this, over half of participants (57%, n=13) felt BillyChip could play a role in supporting
the broader system (homelessness, health and community safety). For example, six
participants who were professionals working with homeless people talked about not being able
to give service users cash when they were in need. Some ended up buying food from their own
pockets and saw BillyChip acting as a solution to this.
Theme 3:
'Somewhere that’s everywhere': Place-based matters
Participants from urban versus rural settings had dierent views about how BillyChip can assist
where they live and work. Various place-based issues exist for people experiencing
homelessness, not least social exclusion, which may also aect the implementation of
BillyChip.
Dierences in urban versus rural settings
The majority of participants (83%, n=19) described BillyChip with reference to urban areas –
cities and towns. There was a recognition that BillyChip is most appropriate in places with more
homeless people, which participants tended to view as urban areas. However, one participant
recognised that police eorts in one city centre has had a displacement eect, moving the
street homeless to the outskirts, and would like to see more BillyChip outlets to respond to such
displacement.
Two participants recognised that homelessness and food insecurity is also an issue for more
rural areas, particularly because they do not have the support services readily available in more
urban areas. This includes access to food banks or free food provision. Therefore, BillyChip
outlets could ll a gap:
In [urban area] there was a lot more services oering free meals, hot meals in particular,
and seemed to be a lot more of a community, I suppose, for homeless individuals. But
there's not so much at all in [more rural area] I think we've got [venue], but I think that's
only open sort of two or three days a week. So people are limited then on other
occasions (P16)
If you've got people in the middle of nowhere, in the middle of [rural area] …they're not
be able to get to some, you know, coee house (P17)
BillyChip places
As well as cafes (70%, n=16) or grocery stores (39%, n=9) which are traditionally seen as
BillyChip Outlets, participants referenced the role (or potential role) of community hubs or
similar (including foodbanks) (65%, n=15) and schools (22%, n=5) as BillyChip places (where
BillyChips can be distributed and/or exchanged).
Notably, almost two thirds of participants (61%, n=14) specically saw a well-known national
fast food bakery outlet as a BillyChip place (although it is known only to use BillyChip within a
23
limited locality). Such a place was, similar to fast food and major supermarkets, ‘somewhere
that’s everywhere’ (P17), i.e. a well-known chain that’s available and accessible for all.
7
Place issues for people experiencing homelessness and food insecurity
When talking about places, issues were raised which can be grouped into three sub-themes:
• People who inhabit places as being seen as problematic (due to being banned from
places, or because of their begging or seen as substance misusers) (30%, n=7). As P1
remarked about potential BillyChip outlets…they're quite happy to support but didn’t
want them sitting in their café...[due to the] stereotype of what the beggar is’.
• Sustainability and limited availability of facilities for the population group (30%, n=7). For
example, P16 explained that homeless people could be ’waiting in A&E for hours without
food and drink’ and also that they ‘have to leave emergency [hostel] beds by 7am’.
• Limits around transport and access to services (30%, n=7). As P5 remarked - ‘some of
our clients particularly [those using street homeless service] will not be able to travel
necessarily or have the means to travel to some of the food banks or for example outside
of [city] so therefore they would miss out’.
Inclusion versus exclusion
A strong sub-theme within all participants responses can be brought together as social
inclusion and exclusion. Discussions grouped as ‘exclusion’ including comments around
‘stigma’, ‘NIMBY-ism’, ‘stereotyping’, the transactional nature of society, not being eligible, being
‘invisible’, or ‘banned’ and the ‘big divide’ between rich and poor (78%, n=18). The issue of
dignity, as previously mentioned in Theme 1, was a reoccurring theme. P4 said ‘sometimes you
can see in people’s eyes that they just don't want to come and ask a complete stranger for a pair
of underwear’.
Participants recognised that stigma exist for homeless people:
People walk past homeless people and they look away. Or they're embarrassed or
they’re uncomfortable… And its not always that they're addicted to drugs…a really
common perception, but these people are there because they’re ill or unable to do
anything about it or they've had a run of bad luck (P20)
Although the focus group were generally positive about BillyChip, one member suggested the
scheme may not necessarily overcome the barrier of stigma:
Some people, they might not want to go in and they might feel ashamed, right? Using
them [BillyChips]. So I would think a little bit, I would feel like I'm trying to hide from
people and I'm standing there like a f**king k**b, like Oliver twist! Denitely [stigma may
exist] for some people. When everybody knows what they [BillyChips] are.
In contrast, responses also highlighted ‘inclusion’ (57%, n=13) - including warm hubs,
‘comfortable’ and ‘accessible’ places, support provisions, community fridge, foodbanks,
schools, and cafes that give away free items. P2 thought that BillyChip could help with inclusion
7
The BillyChip team has also reported that this ‘BillyChip place’ is more desirable by the homeless as one
outlet has redeemed more chips than they sold.
24
by raising ‘the prole of what homelessness looks like in cities’ and P19 explained that
‘homeless can go into our stores feeling welcome’.
Theme 4:
‘Game changing?’: implementation challenges and opportunities
Whilst there was a genuine sense of positivity associated with BillyChip (as described in Theme
1), there were issues described in relation to the business model at an outlet level and at the
organisational level. Some of these were directly in relation to BillyChip, but other comments
were about the context it operates. Risks and issues exist alongside this, but also opportunities
which could be ‘game changing’ for the success of the scheme (which is developed further in
Theme 5).
Micro-level business issues
Nine interview participants (39%) described some negative issues at the micro-business /outlet
level ranging from provider engagement, to till operating system issues to lost chips, errors, and
inventory issues. Two thirds of these participants were from outlets or had worked closely with
outlets so were drawing from direct experiences, whilst the remaining third were commenting
on perceived challenges. For example, P20 reported low sales of BillyChips but felt the scheme
was 'not promoted' within their own organisation in the way it should be. They felt there ‘needs
somebody driving it at the top of the organisation’. Whereas P1 suggested employee
engagement in the scheme rests on ‘if you’re a manager and its part of your job description or
KPIs’. In addition, there were a few positive (13%, n=3) and neutral responses (9%, n=2) at the
outlet level, suggesting BillyChip can t within business processes:
‘BillyChip can be done anywhere with anyone with any systems, complements what
people do, monthly checkups, inputting monthly gures is simple’ (P3)
Macro model
At the more strategic level in relation to BillyChip’s model, there were a range of participant
comments which could be described as neutral (43%, n=10). One participant wanted BillyChip
to be ‘available online to make it easier to purchase and give out’ (P1). Whilst one outreach
worker will hand out ‘two at a time, because you don’t know how generous the vendors are going
to be’ (P12). A few participants observed that it doesn’t quite work for the voluntary sector to
become outlets; with one suggesting BillyChip works best in 'typical nancial model of
business' (P4) and another saying that ‘as a charity, we would want to sell chips and keep the
money’ (P9). There were also a range of negative comments (43%, n=10). For example, one
outlet worker noted that people would say 'I'll buy one but I'm never going to [location] or
[location] to give one out to the homeless…98% of people that buy BillyChips don’t want one'
(P10). Furthermore, although they thought BillyChip was a good idea, the focus group still had
some concerns about how the scheme works in practice and the impact it would make:
This is going to be just like one or two people giving them out…they're still going to be
begging for money… But a two quid chip, you gonna have to use them at like cheap
places because you can't get a lot for two quid these days. So you're gonna have to like,
you know, maybe cheap shops to take them. Find out what shops where they could
actually get something for it. Where you can use it is important (Focus group participant)
25
One participant felt that although BillyChip was a ‘great idea’, potentially helping alleviate the
food insecurity issues in relation to poverty for some ‘as an interim’ approach but it did not help
to address the structural issues. They stated, ‘underlying all of this there will still be children
who can't access food or people, families who are hungry, who can't get access to those
BillyChips and the driver of poverty is still there’ (P14). They felt that BillyChip could use data
and their platform to seek policy change. Whilst another agged up ‘accessibility [of getting to
outlets] for disabled’ (P15).
In addition, 11 (n=48%) of the negative or neutral participant comments related to accessing
outlets, including the greater involvement of a well-known bakery/fast food chain or
supermarkets, highlighting that ‘a problem is not many outlets accept BillyChip’ (P20) but ‘if
[retailer] and [retailer] came on board then its game changing!’ (P1). Getting the right outlets on
board was seen as a signicant opportunity to develop the scheme and engaging the homeless
population group. As one third sector manager explained:
The hope is that one of like the bakeries or a fast-food outlet would get involved in this
because obviously some of the clients that we look after they don't you know have
access to kind of regular clean dry clothes etcetera so sometimes they may not want to
go into like you know a more formal looking establishment. (P5)
The focus group were sceptical about big retailers getting involved, asking ‘do you think it will
ever happen?’ Whilst some participants had taken the step of speaking to retailers; P17 had
been told its unlikely the big named retailer would have BillyChip in their location and another
said, ‘they're very keen but they have to change their complete till system’ (P6). Having such
retailers involved in BillyChip was important for ease and accessibility so those in need ‘havent
got to go looking’ (P8) and P13 agreed that they ‘need to be accepted in lots of places, those
open 7 days a week and late’ (P13).
Risks and Issues
All bar one participants highlighted some low-level risks and issues – the main concern
expressed was related to BillyChip becoming an alternative currency longer term (48%, n=11).
As P1 explained, in relation to expansion, ‘what you've got to be careful with is that the more
things you add to it, the more it becomes an alternative currency’. One focus group member, in
reference to potentially exchanging BillyChip for drugs, said - ‘if you had enough of them, say
you had 30-40 quids worth, somebody may give you like a 15-20 rock, even a 10 rock. People
would do it’. Whereas another focus group member said, 'most dealers wouldn’t want stu like
that, they'd want money'.
Whilst the professionals also suggested that potentially people ‘could swap them for drugs or
steal o each other but this would less of an issue than the positives’ of the scheme (P13) and it
should not act as a barrier, saying, ‘you’re gonna get that anyway with anything’ (P16). In
addition, there was little concern raised around misuse of BillyChip within the outlets. Only P19
mentioned some initial worries - ‘once [outlet sta] were aware that people can buy it for two
pounds and redeem it for something higher…but we have noticed it's only a handful of times’.
Other potential issues suggested were around whether people want to take a BillyChip or may
be unsure about participating in the scheme (this included participating as a provider/outlet,
purchaser or recipient) (43%, n=10). And there was a bit of, what could be described as, outlet
26
fatigue (17%, n=4). With one outlet saying, ‘I'm not really sure [BillyChip] is having the eect that
it is meant to have and that's kind of helping the most in need’ (P5) and another has 'taken as far
as we could but don’t want to close the door on it' (P4). A small number of participants (17%,
n=4) thought the chip looks like a poker chip and that it could have a negative association with
gambling.
We've had a couple of comments about it's a poker chip and that kind of connection, if
you will, through to gambling and that for some people that might be why they've ended
up as a rough sleeper (P7)
Theme 5 – ‘Expanding cautiously’: Developing the BillyChip scheme
Most interview participants saw the potential of increasing what BillyChip can be exchanged for,
with one observing, ‘I think expanding…will make it more useful’. Others suggested doing this
‘with caution as the beauty of BillyChip is its simplicity’. One outlying participant suggested that
BillyChip should ‘stick to food and drink only’, as other items could increase the misuse of the
scheme. Additional items suggested in interviews and the focus group included clothes (57%,
n=13); toiletries (43%, n=10); dog food (26%, n=6); bus fare (39%, n=9). Notably, the focus group
members reeled a list o, including - ‘clothes, milk, bread, newspapers, toiletries, dog food, dog
leads, bus fare, phone top ups, anything that’s not a lot of money, eggs’.
Within the survey, an average of 3.03 (SD 3.42) purchasable items were suggested using by
participants. The frequency of suggestions is displayed in the Table 4. No participants opted to
use the free text option to describe unlisted items. Bus fare and clothing were the most popular
additional products for survey participants, closely followed by pet food, toiletries and phone
top-up voucher.
Table 4. Additional purchasable product suggestions
Bus fare
18
45%
Pet Food
16
40%
Feminine hygiene/period
products
15
37.5%
Clothing (including shoes)
17
42.5%
Toiletries (e.g. soap, toothpaste)
16
40%
Phone top-up voucher
16
40%
Reading & writing materials (e.g.
books, magazines, stationary)
9
22.5%
Contraception
8
20%
Over the counter
8
20%
27
Expanding the BillyChip scheme
Interview and focus group participants saw opportunities for BillyChip in terms of expanding
retailers (52%, n=12) and its application (such as, exchanging the chip for more items) (39%,
n=9); having a role within health and well-being and community safety sector (26%, n=6); and
supporting families (17%, n=4). For example, participants saw the opportunities for BillyChip to
be distributed to support ‘health engagement’ (P13) or ‘pastoral support or safeguarding teams’
for children (P7) as well as supporting ‘summer meals for kids’ (P2). Whilst P1 said ‘every city
needs to have a venue where people can claim it and there’s the additional facilities there for
the homeless’ and P13 emphasised the outlet needs to be ‘open 7 days a week and late’.
Testing of billy chip
Eight participants (35%) discussed the testing of BillyChip or other similar schemes. BillyChip is
being used creatively and in a variety of outlet types – as well as cafes, it is used in charity shops
(to exchange for clothes or books), pet stores, and within community venues already engaged
with population group. As two participants explained:
If someone comes to us with something and we think it will be good, we will always try it.
(P4)
Where somebody turns up to the [community] fridge without donation, it's usually quite
apparent why they haven't got a donation, so we actually have had some chips donated
to us from the people that have purchased it. So we will then put in a donation for them
so there's no awkwardness (P5)
Testing BillyChip and similar schemes was also linked to outreach activity. P16 described that
they had seen vouchers used as an incentive within inclusion health outreach:
We had a consultant…come out with us and he had [bakery chain] vouchers and he was
sort of using those as an incentive to get people to come over to talk about health and
things. And it was really eective.
Whilst ‘PCSOs have distributed BillyChips to homeless people to get drinks… in the local cafes
that are on board and that's been successful’ (P6). However, there was a recognition by P12 that
with BillyChip it takes time, but now they are ‘giving out lots’ of chips:
We get people that don't know about them, who refuse them because they don't know
and they're worried about, who are these people? What are they giving me? But most
homeless…now know about us, and they know about BillyChips and are always so
incredibly grateful (P12)
28
5. Discussion and conclusion
5.1 Discussion
BillyChip and existing literature on comparable interventions
The scoping review of the available literature demonstrated the paucity of peer reviewed
research evidence concerning community-based interventions for the alleviation of poverty, and
the distinctiveness of the BillyChip scheme.
The included articles include evaluations of interventions with the aim of alleviating
accommodation, food and general poverty, but include little experimental research. The quality
of the data, though not scrutinized here on the outcome level, suers from the lack of
experimental methodologies. The lack of experimental data is understandable however
considering the peripatetic and unstable characteristics of the target population making
engagement in clinical trials diicult (Strehlau et al., 2017). Notably however, the average
attrition rate was comparatively low compared to longitudinal research generally (Gustavson et
al., 2012). The limitations of the included studies mostly related to self-report and interview
methods and were well reported in the included literature. The combined sample of participants
in the selected studies represent a broad range of ages, but the sex ratio was slightly skewed
towards women, which given the greater prevalence of homelessness and poverty amongst
men renders the sex ratio in the current literature unrepresentative of the population of interest
(Oice for National Statistics, 2021).
In terms of generalizability, studies carried out in low and middle-income countries are of
limited use in informing practices in higher income countries, and vice-versa. Regarding
research carried out in higher income countries, travel subsidies, vouchers to facilitate house
purchases, and approaches that included educational and case management approaches
seemed to garner positive results (Fowler & Chavira, 2014; Jacups et al., 2018; Marr, 2012;
Matheson et al., 2022; Schmelzer et al., 2022). In lower income countries cash transfer and
educational interventions also provided positive data (Green et al., 2016; Miller et al., 2011).
BillyChip is a token which is purchased from participating outlets by members of the public, and
which is imbued with purchasing value by the participating outlet rather than being a voucher or
alternative to the usual currency. BillyChip can therefore be considered unique amongst the
community-based interventions described in the literature. As BillyChip expands and grows in
popularity, the necessity of evaluation, initially using observational methods and then ideally
using experimental methodology, increases.
The role of BillyChip in helping to alleviate food insecurity
BillyChip exists in a landscape of emerging novel approaches to food insecurity, ranging from
retailer led initiatives, such as prepacked food parcels
8
to community led initiatives such as a
salad garden in Cardi
9
or the growth of warm hubs as mentioned by some of our participants.
The ndings from this study have shown that BillyChip can respond to gaps in population needs
in terms of providing a route to access free food. Homeless populations face food insecurity
challenges, both in terms of ‘the quantity and quality of food’ but also the ‘acquisition of meals
on weekends’ (De la Cerda et al 2024). Indeed, despite the increase in foodbanks and, more
8
https://www.morrisons-corporate.com/media-centre/corporate-news/morrisons-introduces-food-
parcels-for-customers-to-purchase-for-those-in-need2/
9
https://cardisaladgarden.co.uk/
29
recently, warm hubs (see Rabaiotti, 2024), participants recognised the gap BillyChip can play in
food security, including the importance of exchanging BillyChip in fast food or convenience
stores to access some type of food during the night.
Participants saw a key benet of BillyChip as providing choice in what it can be exchanged for
and that recipients should be able to choose what they wanted to eat or drink using the
BillyChip, without any judgement. Moreover, given quantity of food is of more concern than the
quality, it follows that standard interventions, such as typical nutritional advice are not realistic
for people experiencing homelessness which has implications for health policy (Graham et al,
2022). To improve the health and well-being of homeless populations, health interventions
(including mental health) need to be combined with access to basic needs (Lachaud, 2020).
Indeed, in addition to food and drink, participants considered that BillyChip has the potential to
be exchanged for a variety of essential consumables, including clothing, toiletries and period
products and bus fare. This growing need of wider essentials can also be seen within the growth
of ‘multi-banks’
10
. Our study has also highlighted that pet food is seen as a key need, in
recognition of the prevalence of homeless people who have dogs (Scanlon et al, 2021).
Notably, participants suggested that people prefer to use BillyChip rather than giving cash,
which may be reective of the decreasing use of cash in society but is also due to concerns that
the recipient will spend the money funding addictions. Indeed, associations have been made
between alcohol abuse and increased food insecurity within homeless populations which need
to be addressed to reduce chronic disease (Reitzel et al, 2020). Therefore, BillyChip may enable
those with addictions to receive more public support through giving, as well as access to food.
Other sources of support, such as alcohol interventions may also reduce food insecurity levels,
but more research is needed to understand whether alcohol consumption takes precedence
over food (Reitzel et al, 2020) as well as how BillyChip can assist.
Furthermore, food insecurity and the challenges homeless people face to feed themselves (as
well as their pets too) can contribute to feelings of stigma and shame (Johnson et al, 2020).
Issues exist around homeless accessing food, in terms of not just the amount and quality of
food, but the type of food wanted (Kloubec and Harris, 2021). As well as choice, interview
participants recognised BillyChip provides an element of dignity to the recipient to ‘buy’ what
they want with the chip. This is signicant given that ‘the hardship, insecurity, and stigma of
homelessness reect the extent to which one’s dignity is tied to the need to nd and eat food.’
(Graham et al, 2022, p.128). Whilst there was some limited concern around BillyChip becoming
an alternative currency, only one participant highlighted the potential for the chip to contribute
to, rather than reduce stigma. There is some evidence that cash alternatives undermine those in
poverty to live with dignity (for example, the ‘Azure payment card’, see Carnet et al, 2014).
However, it has also been suggested that there has been a shift in stigma since the growth of
food insecurity and increased use of foodbanks within a wider population group (Glass et al,
2021).
Increasing opportunities for social and health inclusion
Issues around place and social exclusion for homeless populations, as well the availability of
BillyChip outlets were key ndings within this study. Links have been made between lack of
access to health, social capital and food insecurity among homeless populations (Fitzpatrick &
10
https://faithinfamilies.wales/2024/03/06/faith-in-families-launches-waless-rst-multibank-cwtch-
mawr-to-donate-surplus-essentials-to-more-than-40000-families-in-poverty-across-south-wales/
30
Willis, 2021). Research into warm hubs found benets in communities coming together for a hot
drink (Rabaiotti, 2024). However, for the most socially excluded, the everyday social practice of
food, such as making a cup of tea, is diicult to achieve and can be both psychologically and
socially challenging, as sourcing and eating food in public ‘disrupts the public-private divide’
(Graham et al, 2022, p.127). BillyChip may contribute to help overcome this and other barriers
to access food; these include a lack of healthy and aordable food in proximity to supported
housing or restrictive business policies around donating food to the homeless (Johnson et al,
2020).
There was a consistent suggestion by participants that universal spaces, such as a well-known
fast food bakery chain, can help provide a suitable and informal place for excluded populations
to access food. Within key urban eating spaces, a ‘cosmopolitan canopy’ can exist where
population groups usually segregated can co-exist to ease tension and increase civility
(Anderson, 2004). Recent research has found that accessible and aordable spaces can help to
bring together socially divided groups (DuPont et al, 2024). Indeed, the warm hubs introduced
during the cost-of-living crisis found an increase in social inclusion (Rabaiotti, 2024). Feelings of
safety are also important in such community spaces (DuPont et al, 2024, Rabaiotti, 2024) and
those with direct experience of BillyChip were more likely to see the aspects of help, care and
safety it can oer. Whilst overall, participants saw the benets of the scheme, there were
opportunities to improve public engagement and implementation. Moreover, it is unclear how
successful the current outlets are because there is a low awareness of the scheme within the
sector that is working directly with the population group. The participants within this study who
ran outlets did describe sales of chips but generally a low level of exchange of chips. Resultingly,
it was noted that some of the outlets had adapted their model by passing on their chips to
outreach workers for example.
Indeed, participants saw the opportunity to use BillyChip as an engagement tool and its value of
expansion the scheme within health, community safety and education settings. Inclusion
health nurses described using their own cash to feed the homeless and it has been reported
that teachers are also spending their own money on pupils arriving at school hungry (Weale,
2024). Participants saw the opportunity to develop the scheme beyond the basic model of
street homeless exchanging a BillyChip for a hot drink or food. This was primarily informed by
the professionals working with the population group. More work (and research) is needed to
engage BillyChip outlets (predominantly the private sector) and the beneciaries of the chip to
develop their interest and gather their views on how the application of BillyChip can be applied
further.
A proposed Theory of Change model
This exploratory study has found evidence of some benets (and perceived benets) of
BillyChip. There appears potential for long term impact through reducing food insecurity,
increasing well-being, harm reduction, social inclusion and community safety. This depends on
the direction of the BillyChip scheme, and the engagement of the various stakeholders involved
in it. There are also key gaps in understanding the amount and pattern of BillyChips in
circulation.
Further research is required to understand what those benecial outcomes from the scheme
are and whether they are being achieved. To guide BillyChip and further research, a proposed
Theory of Change (ToC) model has been drafted in discussion with BillyChip team to help steer
key activities and data collection (see Appendix 2). ToC models have been found to help support
31
the development of evaluations of services, by ‘identifying gaps in data required to evidence
outcomes’ (Cauleld et al, 2023).
As BillyChip circulation increases, there are opportunities to use linked data sets to provide an
estimation of the impact of BillyChip on outcomes related to service use by applying a coding
framework developed in collaboration with colleagues in the SAIL databank (Jones et al., 2014).
This framework is designed to capture routine primary care, hospital and drug and alcohol
service data related to homelessness or extreme poverty as a means of identifying members of
this marginalised community, their patterns of service utilisation, and to gauge the feasibility of
identifying prevalence and incidence of injuries and illnesses associated with homelessness,
unstable housing and poverty.
In addition, it has been argued that support services need to better understand the health and
well-being of the population group to maximise their access to food (Fitzpatrick & Willis, 2021).
Further qualitative research would be benecial directly with the service user group, including
developing a greater understanding of experiences of food insecurity beyond BillyChip,
including what food sources are available, what food sources are preferred and the potential of
community interventions.
This breadth of additional activity will be relevant for policy development across a range of areas
– health, homelessness, poverty and community safety.
Application – the potential for alternative giving schemes
The BillyChip scheme shows the potential for the development and use of alternative giving
schemes as an intervention for supporting inclusion health populations or people in poverty.
The scheme developed with a strong compassionate message and purpose, working with key
collaborators, to respond to concerns for street homelessness accessing a hot drink or food.
This concept has broader applicability for the development and formalisation of other
alternative giving schemes, such as, for example, the use of gift vouchers for those experiencing
food insecurity.
This study has shown a strong level of acceptability for such an alternative giving approach that
overcomes the barrier of giving cash. The challenges faced by such as a scheme as it expands,
such as low awareness, could be combatted by communities co-creating their own alternative
giving schemes to respond to local challenges. This would potentially help to create greater
ownership and reduce the business issues associated with scaling up of such schemes. This
would also give the potential for the local area to decide what the vouchers, for example, could
be exchanged for, based on local need.
5.2 Limitations
This evaluation of BillyChip suers from several limitations linked to the methodological
complexities around poverty and homelessness research, and practical limitations in terms of
time and resources.
Firstly, in setting the scene for our evaluation we carried out a scoping review. Due to the
multidimensional nature of poverty, it could be argued that our search terms were not expansive
32
enough to capture the full range of ways in which poverty can be experienced and so this may in
turn have led to the omission of useful articles. We also focused on articles reporting the
alleviation of poverty e.g. schemes or initiatives which aim to provide shelter, food or clothing
where it is lacking. We did not include studies where in the aim of the stated intervention was to
improve aspects of health such as mood, wellbeing, or physical problems apart from
malnutrition such as infectious diseases. As poverty is such a pervasive and pernicious issue, it
could be argued that the alleviation of poverty extends out beyond the focus of our review.
Secondly, our survey and interview sample were over representative of people working in
charitable and local government in supportive and managerial roles. A third survey respondents
had direct experience of the scheme and only one was able to answer from the perspective of a
frontline sales representative, which reduces the applicability of our ndings. The focus group
had experience of homelessness but not using BillyChip. The limited sample size is not a
concern in that our study was not intended to test hypotheses, but a broader range of
perspectives would be advantageous.
Thirdly, survey respondents and those of whom who wished to be interviewees were recruited
from the southwest of England and the South of Wales. The focus group also took place in South
Wales. Therefore, it is recognised that one limitation is that most participants contributing to the
qualitative element of the study were drawn from an area where BillyChip is less developed.
5.3 Concluding remarks and recommendations
Our exploratory study has found that participants welcome the BillyChip scheme; an alternative
to cash which alleviates concerns to giving cash yet gives recipients choice and dignity in
exchanging the chip for food or drink. This is particularly important for a population group – the
street homeless – who experience food insecurity, stigma, social exclusion and related health
issues. However, we found a low level of awareness of BillyChip and some issues with the
model working in practice, such as limited availability of outlets. Where people are more
experienced with BillyChip, they are more likely to express a positive attitude towards the
scheme. There was an emphasis on national food retailers being key to BillyChip success and
recipients being able to access food in universally welcome places.
5.3.1 Future research
Additional survey, interview and ethnographic observational work should be carried out to
further explore the opportunities and challenges related to the trajectory of BillyChip. In
particular this could include research into direct experiences of recipients of the scheme (and
alternative giving schemes more generally), preferred outlets and items they would like to
exchange. The testing of the proposed Theory of Change Model could assist with the
development of alternative giving schemes like BillyChip as evidenced-led interventions.
5.3.2 Implications for policy
The concept of BillyChip has broader applicability for health promotion policy for inclusion
health populations and those known to suer food insecurity and social exclusion. This may
help to inform the development and formalisation of other alternative giving schemes, such as,
for example, the use of gift vouchers. The development and adoption of BillyChip or other
alternative giving schemes should be concentrated at a community level with an emphasis on
building awareness and responding to local needs. In areas where food insecurity poses a
33
signicant problem, alternative giving schemes should be considered as highly acceptable
interventions from the perspectives of those in need and those in a position to help. Retail
outlets belonging to larger retailers should be approached for inclusion prior to smaller
businesses, as according to our ndings, these businesses are more likely to engage, facilitating
early growth.
5.3.3 Implications for practice
We have discussed our ndings and recommendations with the BillyChip team who are open to
the potential the scheme has and testing a theory of change. Indeed, there has been some
adaptation to the model already (such as outlets collecting purchased chips, passing them on
for outreach activity) and whilst some participants were cautious, most saw the potential for the
greater application of such a scheme in their community. As well as businesses, public and
third-sector services who work with those in poverty may benet from the appointment of a
local ‘champion’ who works directly for the alternative giving scheme as a charitable enterprise
and liaises with other participants on a regular basis to maintain engagement.
References
Anderson, E. (2004). The Cosmopolitan Canopy. The ANNALS of the American Academy of
Political and Social Science, 595(1), 14-31. https://doi.org/10.1177/0002716204266833
Apteligen. (2020). Evaluation of impact tempo time credits. Apteligen/Tempo. https://f7305a46-
1323-499e-b34f-
90054fe06f85.lesusr.com/ugd/165ea9_939f0d2f58534046a3a343e2d02c007c.pdf
Barradale, G. (2023, 6 January). The inspiring story behind the Greggs poker chips used to help
rough sleepers. The Big Issue. https://www.bigissue.com/news/activism/billy-chip-greggs-
homeless-poker-chip/
Beiser, M. E., Smith, K., Ingemi, M., Mulligan, E., & Baggett, T. P. (2019). Hepatitis C treatment
outcomes among homeless-experienced individuals at a community health centre in Boston.
International Journal of Drug Policy, 72, 129–137. https://doi.org/10.1016/j.drugpo.2019.03.017
BillyChip. (2023). How does BillyChip work? https://www.billychip.com/how-it-works
Buscemi, F. (2015). How “il caé sospeso” became “suspended coee”: The neo-liberal re-
’invention of tradition’ from Bourdieu to Bourdieu. European Journal of American Culture, 34(2),
123–136. https://doi.org/10.1386/ejac.34.2.123_1
Butler, P. (26 April 2023). More than 500,000 people in UK visited ‘warm rooms’ during the winter.
The Guardian. https://www.theguardian.com/society/2023/apr/26/warm-rooms-winter-
loneliness
Cauleld, L., Quigg, Z., Adams-Quackenbush, N., Timpson, H., & Wilson, S. (2023). Reections
on good practice in evaluating Violence Reduction Units: Experiences from across England and
Wales. Evaluation, 29(3), 276–295. https://doi.org/10.1177/13563890231183993
Collins, S. E., Saxon, A. J., Duncan, M. H., Smart, B. F., Merrill, J. O., Malone, D. K., Jackson, T. R.,
Clifase, S. L., Joesch, J., & Ries, R. K. (2014). Harm reduction with pharmacotherapy for
homeless people with alcohol dependence: Protocol for a randomized controlled trial.
Contemporary Clinical Trials, 38(2), 221–234. https://doi.org/10.1016/j.cct.2014.05.008
34
Doran, K. M., Fockele, C. E., & Maguire, M. (2022). Overdose and Homelessness—Why We Need
to Talk About Housing. JAMA Network Open, 5(1), e2142685.
https://doi.org/10.1001/jamanetworkopen.2021.42685
Folsom, D. P., Hawthorne, W., Lindamer, L., Gilmer, T., Bailey, A., Golshan, S., Garcia, P.,
Unützer, J., Hough, R., & Jeste, D. V. (2005). Prevalence and Risk Factors for Homelessness and
Utilization of Mental Health Services Among 10,340 Patients With Serious Mental Illness in a
Large Public Mental Health System. Https://Doi.Org/10.1176/Appi.Ajp.162.2.370, 162(2), 370–
376. https://doi.org/10.1176/APPI.AJP.162.2.370
Glass, J., Shucksmith, M. D., Chapman, P., & Atterton, J. (2021). Covid19, lockdowns and
nancial hardship in rural areas: Insights from the Rural Lives project. Rural Lives.
https://www.rurallives.co.uk/ rural-lives-nal-report.htm
Honer, W. G., Cervantes-Larios, A., Jones, A. A., Vila-Rodriguez, F., Montaner, J. S., Tran, H.,
Nham, J., Panenka, W. J., Lang, D. J., Thornton, A. E., Rauscher, A., & Schultz, K. (2017). The
Hotel Study—Clinical and Health Service Eectiveness in a Cohort of Homeless or Marginally
Housed Persons. Canadian Journal of Psychiatry, 62(7), 482–492.
https://doi.org/10.1177/0706743717693781
De La Cerda, C. M. P., Soares, G. M. E., Pinheiro, A. K. B., Lacthim, S. A. F., Dias, A. L. F.,
Arcencio, R. A., & De Freitas, G. L. (2024). Access to food and food quality: perception of the
homeless population. Acta Paulista de Enfermagem, 37. https://doi.org/10.37689/acta-
ape/2024AO0023611
Dupont Picard, P., Ali, S., Manley, D., Schaefer, C., Taylor, L., & McKeown, S. (2024).
Understanding the spatial dimension of youth intergroup contact in a postaccord society..
Peace and Conict: Journal of Peace Psychology, 0 https://doi.org/10.1037/pac0000719
Fitzpatrick, K. M., & Willis, D. E. (2021). Homeless and hungry: food insecurity in the land of
plenty. Food Security, 13(1), 3–12. https://doi.org/10.1007/s12571-020-01115-x
Francis-Devine, B. (2024). Who is experiencing food insecurity in the UK? House of Commons
Library. https://commonslibrary.parliament.uk/who-is-experiencing-food-insecurity-in-the-uk/
Graham, R., Stolte, O., & Atatoa-Carr, P. (2022). Cup of Tea to Keep Warm: Food-Related
Practices in Aotearoa New Zealand for Survival while Homeless. International Journal on
Homelessness, 1–17. https://doi.org/10.5206/ijoh.2022.1.14242
Johnson, K., Drew, C., & Auerswald, C. (2020). Structural violence and food insecurity in the
lives of formerly homeless young adults living in permanent supportive housing. Journal of
Youth Studies, 23(10), 1249–1272. https://doi.org/10.1080/13676261.2019.1667492
Jones, K. H., Ford, D. V., Jones, C., Dsilva, R., Thompson, S., Brooks, C. J., Heaven, M. L.,
Thayer, D. S., McNerney, C. L., & Lyons, R. A. (2014). A case study of the secure anonymous
information linkage (SAIL) gateway: A privacy-protecting remote access system for health-
related research and evaluation. Journal of Biomedical Informatics, 50, 196–204.
https://doi.org/10.1016/j.jbi.2014.01.003
35
Kloubec, J., & Harris, C. (2021). Food Acquisition Strategies of Homeless Youth in the Greater
Seattle Area: A Cross- Sectional Study. Journal of the Academy of Nutrition and Dietetics,
121(1), 47-58.e1. https://doi.org/10.1016/j.jand.2020.05.013
Kolanowski, W., & Trafialek, J. (2021). Eating behaviour of homeless people aged 35–40 in
Warsaw. Food Control, 119, 107477-. https://doi.org/10.1016/j.foodcont.2020.107477
Lachaud, J., Mejia-Lancheros, C., Wang, R., Wiens, K., Nisenbaum, R., Stergiopoulos, V.,
Hwang, S. W., & O’Campo, P. (2020). Mental and substance use disorders and food insecurity
among homeless adults participating in the at Home/Chez Soi study. PloS One, 15(4),
e0232001–e0232001. https://doi.org/10.1371/journal.pone.0232001
Lippert, A. M., & Lee, B. A. (2021/04//). Adult and Child Food Insecurity Among Homeless and
Precariously-Housed Families at the Close of the Twentieth Century. Population Research and
Policy Review, 40(2), 231-253. https://doi.org/10.1007/s11113-020-09577-9
Loftus, E. I., Lachaud, J., Hwang, S. W., & Mejia-Lancheros, C. (2021). Food insecurity and
mental health outcomes among homeless adults: a scoping review. Public Health Nutrition,
24(7), 1766–1777. https://doi.org/10.1017/S1368980020001998
Nilsson, S. F., Nordentoft, M., Fazel, S., & Laursen, T. M. (2020). Homelessness and police-
recorded crime victimisation: A nationwide, register-based cohort study. The Lancet Public
Health, 5(6), e333–e341. https://doi.org/10.1016/S2468-2667(20)30075-X
Pain, R., & Francis, P. (2004). Living with crime: Spaces of risk for homeless young people.
Children’s Geographies, 2(1), 95–110. https://doi.org/10.1080/1473328032000168796
Pautz, H., & Dempsey, D. (2022). Covid-19 and the crisis of food insecurity in the UK.
Contemporary Social Science, 17(5), 434–449.
https://doi.org/10.1080/21582041.2022.2044069
Rabaiotti, E. C. (2024). Safe and welcoming “warm hubs”: Building social connections and
inclusion in Welsh communities. Journal of Community Safety and Well-Being, 9(1), 46–51.
https://doi.org/10.35502/jcswb.370
Ravikumar, D., Nic Gabhann, S., & Kelly, C. (2021). A systematic review of the determinants of
the nutritional quality of food for homeless populations. European Journal of Public Health,
31(Supplement_3). https://doi.org/10.1093/eurpub/ckab165.403
Ravikumar, D., Vaughan, E., & Kelly, C. (2022). Diet Quality, Health, and Wellbeing within the
Irish Homeless Sector: A Qualitative Exploration. International Journal of Environmental
Research and Public Health, 19(23), 15976-. https://doi.org/10.3390/ijerph192315976
Reitzel, L. R., Chinamuthevi, S., Daundasekara, S. S., Hernandez, D. C., Chen, T. A., Harkara, Y.,
Obasi, E. M., Kendzor, D. E., & Businelle, M. S. (2020). Association of problematic alcohol use
and food insecurity among homeless men and women. International Journal of Environmental
Research and Public Health, 17(10), 3631-. https://doi.org/10.3390/ijerph17103631
36
Scanlon, L., Hobson-West, P., Cobb, K., McBride, A., & Stavisky, J. (2021). Homeless People and
Their Dogs: Exploring the Nature and Impact of the Human–Companion Animal Bond.
Anthrozoös, 34(1), 77–92. https://doi.org/10.1080/08927936.2021.1878683
Weale, S. (2024, June 5) Eight in 10 primary teachers in England spending own money to help
pupils | Education | The Guardian.
https://www.theguardian.com/education/article/2024/jun/05/eight-in-10-primary-teachers-in-
england-spending-own-money-to-help-pupils?CMP=oth_b-aplnews_d-5
APPENDIX 1
Authors
Title
Year
Country
Seng
Duraon/
Obs period
Sample
descripo
n
Sample at
intake
Sample
male %
Sample
age
Arion
Design &
methods
Interven
on
Compariso
n
Analysis
Jacups,
Rogerson
and
Kinchin
An
innovave
approach
to address
homelessn
ess in
regional
Australia:
Parcipant
evaluaon
of a co-
payment
model
2018
Australia
Urban
(City of
Cairns)
n/d
Urban
homeless
aboriginal
Australian
s who
were able
to access
adequate
housing in
their
home
communi
es
n=37
67.60%
Mean
37.2,
median
34, range
19-63
0%
Observao
nal: Survey
and
interview
study
Return 2
Country or
R2C
iniave.
Subsidised
travel for
homeless
people in
an urban
seng to
facilitate
return to
rural
home
communi
es
None
Quanta
ve
descripve
s,
themac
analysis of
interview
data
1
Zamora-
Sarabia et
al.
Child
health and
the
possibilie
s for
childcare
in a
context of
poverty
and food
insecurity:
The
narraves
of parents
aending
a self-
managed
foodbank
in Spain
2017
Spain
Urban
(Tetuan,
Madrid)
7 months
Parents
and
children
experienci
ng food
poverty
n=37 (15
adult
parents,
22
children
under 18)
33.3% of
parents
Children
age range
1.5-17
years. No
other data
0%
Observao
nal:
Ethnograp
hic
observao
n and
interview
study
Supermark
et food
bank
None
Qualitave
content
analysis
Green et
al.
Does
poverty
alleviaon
decrease
depression
symptoms
in post-
conict
sengs? A
cluster-
randomize
d trial of
microente
rprise
assistance
in
Northern
Uganda
2016
Uganda
Rural
Uganda
18 months
People
displaced
by military
conict
n=1800
(896
experimen
tal group,
n=896,
904
control
group)
14%
27.02 (7.2)
experimen
tal group,
27.63
(7.29)
control
group
3.7% total
sample, no
signicant
dierence
between
groups
Experimen
tal: Cluster
RCT
Women's
Income
Generan
g Support
(WINGS)
program.
The
program
consists of
5 days of
business
skills
training
designed
for
illiterate
populaon
s, an
individual
start-up
grant of
roughly
US$150,
and 3 to 5
visits over
approxima
Experimen
tal versus
waing list
Intent-to-
treat
regression
analysis
2
tely 18
months by
trained
communit
y workers
to provide
business
advice and
encourage
ment
Alvarez et
al.
Empoweri
ng the
poor: A
eld study
of the
social
psychologi
cal
consequen
ces of
receiving
autonomy
or
dependen
cy aid in
Panama
2018
Panama
Rural,
Ngäbe-
Buglé
district
Panama
n/d
People in
extreme
poverty
equal to
individual
income of
less than
US $2.15
per day.
n=154 (77
cash
recipients,
77
voucher
recipients)
5.84%
no data
no data
Quasi-
experimen
tal: Cross-
seconal
evaluaon
interview
study
Cash
recipients
received
transfer of
US$100
cash every
2 months.
Voucher
recipients
received
equivalent
of $100
every 2
months in
the form
of
vouchers
that can
be used at
local shops
to acquire
food or
domesc
products
Cash
transfer
versus
shopping
vouchers
Regression
analysis
using
Linear
mixed
eects
models
Fowler
and
Chavira
Family
Unicaon
Program:
Housing
Services
for
Homeless
Child
Welfare–
2014
USA
Urban
(City of
Chicago)
26 months
Families at
risk of
separaon
solely
because of
living
circumstan
ces and
condions
of poverty
n=65 (31
experimen
tal group
houeholds
, 34
control
group
household
s). No
other data
93.5%
experimen
tal group,
100%
control
group
32.81
(9.1)
experimen
tal group
parental
age, 31.35
(8.4)
control
group
8% total
sample, no
signicant
dierence
between
groups
Experimen
tal: RCT
Experimen
tal group
recieved
Family
Unicaon
Program
(FUP)
vouchers
to lease or
purchase
Experimen
tal versus
control
group
Intent-to-
treat
regression
analysis
3
Involved
Families
parental
age
“decent,
safe, and
sanitary
housing”
plus usual
care of
referrals to
domesc
and
homeless
shelters,
educaon
on
housing
searches,
talk with
landlords,
budget,
and nd
employme
nt .
Control
parcipant
s recieved
usual care
only
Matheson
et al.
Filling the
GAP:
Integrang
a gambling
addicon
program
into a
shelter
seng for
people
experienci
ng poverty
and
homelessn
ess
2022
Canada
Urban
(City of
Toronto)
18 months
People
experienci
ng
homelessn
ess, or at
risk of
homelessn
ess
n=17 (11
homelessn
ess, 6 at
risk of
homelessn
ess)
64.71%
Mean 54,
range 37 -
80
0%
Observao
nal:
Interview
study
Individual
case
managem
ent and
facilitated
group
sessions
ulising
Cognive
Behaviour
al Therapy
(CBT) and
Life Skills
(LS) for the
treatment
of harmful
gambling.
Intraparc
ipant at
intake and
follow up
Qualitave
content
analysis
4
M. Marr.
Pathways
out of
Homeless
ness in Los
Angeles
and Tokyo:
Mullevel
Contexts
of Limited
Mobility
amid
Advanced
Urban
Marginalit
y
2012
USA &
Japan
Urban
(City of
Los
Angeles &
Tokyo)
Range 3 -
64 months
(LA
median
24,
months),(
Tokyo
median 12
months)
People
experienci
ng
homelessn
ess in LA
and Tokyo
n=34 (17
LA, 17
Tokyo)
LA 53%,
Tokyo
100%
Most LA
parcipant
s over 50,
Most
Tokyo
parcipant
s under
50. No
other
data.
0%
Quasi-
experimen
tal:
Interview
study
Publicly
funded
transiona
l housing
programs
(THP)
which aim
to provide
temporary
shelter
and help
people
experienci
ng
homelessn
ess secure
stable
housing.
LA and
Tokyo THP
parcipant
s
Fuzzy-set
qualitave
compara
ve analysis
(fsQCA)
Schmelzer
& Leto
Promong
Health
Through
Engageme
nt in
Occupao
ns
That
Maximize
Food
Resources
2018
USA
Urban,
City of
Greeneld
, Indiana
1.6
months
Adults
experienci
ng food
poverty
n=16
25%
Mean
46.42
(10.22)
0%
Observao
nal:
Parcipa
on Acon
Research
(PAR)
study
Educaon
al group of
seven 2.5
to 3 hour
sessions
including
didacc
teaching
componen
t on
managing
food
resources,
graded
cooking
task that
either
directly
applies the
informao
n just
learned
(e.g.,
measuring
porons)
or
contribute
Intraparc
ipant at
intake and
follow up
Means
compariso
n
5
s to the
larger
focus of
managing
food
resources
(e.g.
deboning
a chicken
to store
the
leovers,
creang
meals that
use
ingredient
s that spoil
quickly).
Sessions
begin with
feedback
from
previous
week's
tasks
Miller et
al.
The
impact of
the social
cash
transfer
scheme on
food
security in
Malawi
2011
Malawi
Rural,
Mchinji
District
7 months
Household
s in lowest
naonal
economic
quinle, or
having no
assets, or
consuming
only one
meal per
day
n=819
household
s (408
experimen
tal group
household
s and 411
control
group
household
s). No
other data
37%
experimen
tal group,
34%
control
group
Mean 61
Interven
on, 63
comparato
r. No other
data
10%
experimen
tal group,
6% control
group
Quasi-
experimen
tal:
Longitudo
nal
evaluaon
survey and
interview
study
Malawi
Social
Cash
Transfer
Scheme
(SCTS).
Recipients
receive
cash to
purchase a
variety of
healthy
foods,
ferlizer,
and farm
tools, and
to direct a
poron of
their
harvest to
Interven
on versus
control
group
Dierence
in
dierence
(DiD)
regression
analysis
6
food
stores.
Control
recipients
did not
receive
cash
transfer
but did
receive a
food
bucket.
Kidd et al.
Trialing
the
feasibility
of a crical
me
interven
on for
youth
transioni
ng out of
homelessn
ess
2020
Canada
Urban
6 months
People
aged 16 or
over with
rcent
experience
of
homelessn
ess and at
risk of
future
homelessn
ess
n=65 (34
interven
on group,
31 control
group)
50.80%
Mean
21.75
(2.07),
range 17–
26
6.15%
experimen
tal group,
4.62%
control
group
Experimen
tal: RCT
Housing
Outreach
Program—
Collabora
on (HOP-
C)
program
consists of
weekly
contacts
with a
transiona
l case
manager
to assist in
nding
housing,
educaon,
employme
nt and also
jusce and
health
support.
Also
includes
peer
support
including
monthly
organised
acvies
and
weekly 90
Experimen
tal versus
control
group
Regression
analysis
using
generalize
d linear
models
(GLM)
7
minute
therapy
group
based on
psychoedu
caon and
mindfulne
ss
principles.
One on
one DBT
(Dialeccal
Behaviour
Therapy)
was also
available if
needed.
Control
group
received
only
weekly
contacts
with case
manager.
Wallace et
al.
Where's
the
Housing?
Housing
and
Income
Outcomes
of a
Transion
al Program
to End
Homeless
ness
2018
Canada
Urban
(Victoria,
Brish
Columbia)
49 months
Homeless
adults
resident at
an
emergenc
y shelter
n=111
65.80%
Mean 44.
No other
data
45.95%
Observao
nal: Survey
study
Transion
al housing
program
(THP)
providing
accommo
daon in a
15 bed
house and
case
managem
ent, rent
suppleme
nts (up to
a year) on
exit,
monthly
grocery
vouchers,
and bus
Intraparc
ipant at
intake and
follow up
Descripv
e
compariso
n of
proporon
s
8
passes.
Where
possible,
sta
provide
support to
former
residents
through
outreach
to
maintain
housing
and social
supports
9
APPENDIX 2
