‘A Real Gap’: Consequences of Removing Reintegration Support in Tasmania, Australia

As Australian imprisonment and recidivism rates continue to rise, it is increasingly important to better understand how to support “returning citizens” to adequately prepare them for post-release life. Using a case study approach focused on service provider perspectives, this article examines the consequences of removing a transitional accommodation support service on service providers and returning citizens. Participants highlighted significant consequences, such as: gaps in services; lack of support and housing; persons remaining in the system unnecessarily; persons being released without transitional support or accommodation; health implications; loss of relationships and trust; and suicide of returned citizens. Drawing on critical criminological theory, we argue that the most significant consequence was the loss of “floating support”—which involves a case worker who brokers support between agencies working with a single client—that was integrated into the program. Study findings suggest serious consequences of government decisions to defund programs in this sector.


Introduction
The Australian incarceration rate is at an all-time high and continues to rise. The national incarceration rate increased from 205 to 214 persons per 100,000 adult population in 2020-2021 (Australian Bureau of Statistics 2021b), with recidivism rates as high as 45.2% nationally. This figure is 50.4% in Tasmania, with 59.3% returning to corrective services within two years (Productivity Commission 2022). While Tasmania's incarceration rate has declined the last two years, overall, it has increased sharply since 2014-its lowest point in the last decade-from 112 per 100,000 to 149 per 100,000 adult population (Australian Bureau of Statistics 2021b). This has resulted in prison population growth from 448 to 642 (Australian Bureau of Statistics 2021b), causing overcrowding that has seen incarcerated persons double-or triple-bunking in cells designed for one person (Department of Justice 1 The labels we place on individuals are powerful and difficult to shed. As such, we have chosen to remove any term with a negative association to it, such as "inmate," "offender" or "prisoner," and instead refer to "returning citizens" in line with current research. On the importance of language and word choice on the stigmatization and exclusion for "returning citizens," see Jackl (2021); see also Cox (2020). 2 Given the disadvantages and social exclusion experienced by returning citizens, some researchers question the term, "reintegration," which assumes a previous level of integration most incarcerated persons have not experienced, and use "(re)integration" to highlight its inappropriateness (see Borzycki and Baldry 2003;Burke et al. 2019;Ward 2001). outlook, decrease self-harm and the risk of suicide, decrease alcohol and drug dependency, and reduce anxiety and negative effects of incarceration (The Salvation Army n.d. ) An independent evaluation in 2013 deemed the program successful in achieving its aims of housing and assisting persons released from prison, with service providers reporting a 6.5% recidivism rate (Lloyd et al. 2013) compared to the general recidivism rates of 39.3 percent in 2013 (Productivity Commission 2013). This follows previous support programs with successful recidivism rates (McCrae 2011;van Aaken 2010;White and O'Halloran 2011). In addition to decreased recidivism, a Department of Health and Human Services evaluation found REO had a cost-efficiency factor of one to four, and that the AUS$257,761 invested in REO saved the Tasmanian Government more than AUS$1 million in incarceration costs-not including police or court costs (Parliament of Australia 2015; Tenants' Union of Tasmania 2016). The program was cheaper compared to the Tasmanian recurrent expenditure of AUS$330.43 per incarcerated person per day in 2013 (Productivity Commission 2022)-an average of approximately AUS$120,000 per annum per person. Clients reported high levels of satisfaction and advocated for continuation and expansion throughout Tasmania (Lloyd et al. 2013).
Despite the positive evaluation of REO and its precursor programs, in 2015, the Tasmanian Liberal government decided the program would no longer receive state funding. The rationale for this defunding was that Housing Connect 3 would provide a more "equitable and fair approach" prioritizing "those most in need, not who they are" (Smiley 2015). This was fiercely opposed by practitioners and politicians, who argued that REO was not merely a housing program and that Housing Connect would not be able to meet incarcerated persons' needs (Parliament of Tasmania 2015).
Nevertheless, defunding proceeded, and Housing Connect acquired the accommodation support service for incarcerated persons (Billings 2015;Smiley 2015), which saw currently incarcerated persons incorporated into general accommodation waiting lists (Parliament of Tasmania 2015; Tenants' Union of Tasmania 2016). While incarcerated persons could submit an Expression of Interest six weeks before release, they were not offered a home until after release and assessment by Housing Connect-meaning individuals were released into homelessness (Housing Tasmania 2018). Because an application was not active until postrelease, The Parole Board (2016) was unable to consider an application for a prospective releasee, as accommodation is a requirement for parole. While Housing Connect changed its policy to afford priority status to prospective releasees, this had no impact in providing housing to parole applicants (Parole Board 2016). This differs greatly from how REO workers assessed and set up participants in accommodation prior to release and provided additional floating support upon release. As of October 2016, 109 incarcerated persons had applied for housing through Housing Connect, but none had been allocated a house (Tenants' Union of Tasmania 2016), and no returning citizens were provided with any form of support work. The Parole Board (2016Board ( , 2018 reported an increase in parole denial since 2015, and noted the absence of suitable, stable accommodation since defunding "previous providers." In mid-2018, almost three years after REO's defunding, a new Salvation Army program-Beyond the Wire-was implemented.

Support Needed for Returning Citizens
Returning citizens' needs are multifaceted (Halsey 2010;Halsey and Harris 2011;Parsell et al. 2015;Willis 2018)-especially when crime stems from homelessness, low socioeconomic and educational status, poverty, and unemployment (Bradley et al. 2001;Payne et al. 2015). They are among the most socially excluded members of society (Jones-Finer and Nellis 1998) as the skills and knowledge required in prison are vastly different from those required in society (Halsey 2010). These issues need amelioration pre-release to enhance community (re)integration (Maruna and LeBel 2003), or such individuals are less likely to feel as if they are valued community members (Halsey 2010). While not discussed in this article, it is important to acknowledge that we have a collective responsibility-shared among the returning citizen, the government, and the community-to achieve crime desistance and facilitate social bonds that support (re)integration (Burke et al. 2019).
Incarcerated persons are overrepresented in low socioeconomic status populations and highly disadvantaged in many aspects of education, employment, family and social relationships, and health (Baldry et al. 2003b;Willis 2004). For example, they experience chronic social disadvantage, poor physical and mental health, and high rates of substance misuse prior to imprisonment-and at higher rates post-release-with a high probability of recidivism, suicide, and fatal drug overdoses in immediate months post-release (Kinner and Williams 2006). To lessen these risks, it is vital to provide transitional and accommodation support for returning citizens because accommodation stability decreases offending and drug use (Willis 2018), and each time a person is incarcerated, that person is likely to lose resources like housing.
Incarcerated persons rely more heavily on public housing and rentals from landlords and housing authorities than the general public, with fewer incarcerated persons owning their own homes (Baldry et al. 2003b). Baldry et al. (2003b) found that eighteen percent of the prison sample was homeless prior to incarceration, compared to one percent of the general population. Most returning citizens lack suitable accommodation and accessible pre-release support to secure accommodation (Baldry et al. 2003b;Willis 2004), which can contribute to reoffending. As noted above, appropriate accommodation is also a requirement for parole, with research finding that nineteen percent of incarcerated males and fifteen percent of females had no accommodation arrangements four weeks prior to release (Kinner, 2006). Those homeless prior to incarceration, and transient post-release, are also significantly more likely to be reconvicted as they engage in "survival offending" (Baldry et al. 2006;Payne et al. 2015;Williams et al. 2012). Importantly, studies have found a clear relationship between homelessness and recidivism rates (Baldry et al. 2003a(Baldry et al. , 2003b2006;Williams et al. 2012) because the stress of unstable housing may induce or exacerbate current mental health issues, causing people to self-medicate with alcohol and drugs, and to offend to support such use (Kirkwood & Richley 2008;Payne et al. 2015). This should serve as evidence for the need for a dedicated holistic service focused on transitional support alongside accommodation-not just accommodation by itself-to address difficulties emerging upon release (McCrae 2011), thus reducing criminal processing costs (Willis 2018).
This article discusses what happens when this support is removed. We argue that defunding the REO program represents the dissolution not just of an accommodation service, but also of floating support. Floating care or support involves a case worker who brokers support between agencies working with a single client (Borzycki & Baldry 2003). Allen (2003: 1.3) argues that floating support workers represent "welfare intermediaries" that manage "the interstices between (rather than within)" housing and supervisory services (like probation) "to overcome problems of inter-professional fissure." They work with-and refer to-a range of different support services for returning citizens to ensure they are provided with the best possible chance of (re)integration and avoid recidivism, and to "help them to sustain independent living" (Allen 2003: 1.3). In the case of REO, the service provider case workers constituted floating support for returning citizens that assisted with their housing and other needs. We suggest that tasking Housing Connect with the work of REO was unsuccessful because this form of floating support for returning citizens ended when REO was defunded-and, as we have suggested thus far, accommodation is not enough to ensure (re)integration success (Halsey 2010;Halsey and Harris 2011). This is corroborated by previously incarcerated persons and service providers who, in Hardcastle et al. (2018) study, argue that returning citizens are impacted by numerous factors of disadvantage that need to be addressed collectively through holistic support systems. The criminal processing system 4 which-as evident by globally growing imprisonment and recidivism rates-can quite arguably be considered ineffective, is built on, and driven by, crime control and punitive ideologies. Researchers, such as Elliott Currie (e.g., 2010Currie (e.g., , 2011, David Garland (e.g., 1998Garland (e.g., , 2017Garland (e.g., , 2020a, and Russell Hogg (e.g., 2002), have long argued that the dysfunctionality of prisons serve as an extension of punitive government responses that do not recognize the developmental nature of (re)integration and desistance which occurs over time by a process of lapse and relapse (Buck 2018;Graham and McNeill 2017). In addition, the association between crime and welfare-and, indeed, the role of persistent poverty-as Currie (2010Currie ( , 2011 and Garland (2017Garland ( , 2020a, in particular, explain-is crucial when assessing the inadequacies of our penal systems. The prevalence of crime and violence is intertwined with the functioning of communities, families, and schools, as well as the mainstream processes of socialization, social integration, and social control. Penal practices are connected with welfare practices, with both serving as remedial control mechanisms that target a "suspect" population, such as incarcerated persons (Garland 2020a). Institutions, such as prison, are merely a means to control and govern these "threats" and have expanded their focus on risk management and community protection through a lens of punishment, control, and monitoring-all to the detriment of rehabilitation (Hogg 2002). The introduction of populist law and order measures, such as mandatory sentencing regimes, registers, and targeted policing, categorize individuals as risks to be managed and segregated, further perpetuating the "us vs them" dynamic (Hogg 2002). This has produced a system where the focus is on managing the problem through punishment, rather than addressing the underlying factors and social institutions of crime: criminogenic opportunities, neighborhoods, schools, and the labor market, or lack thereof (Garland 2017). Garland (1998) describes, the criminal processing system is reactive and blame-allocating, primarily concerned with delivering sanctions to convicted offenders.
From a public point of view, this approach appears immediate, cost-effective, and morally apt-"giving offenders the punishment they deserve" (Garland 2020a: 336)-which fits with the "tough on crime" narrative that continues to dominate public discussion about criminality and punishment. Australia's criminal processing system, for one, is driven by an ideology of incapacitation, and our punitive society and policy decisions are politically and culturally motivated by the attitudes of the general public (Brookman and Wiener 2017;Jones and Weatherburn 2010;Roberts and Indermaur 2007;Spiranovic et al. 2011). What this approach fails to recognize is that, as Garland (2017) argues, penal sanctions are after-the-fact remediations to control social problems and thus neither rehabilitative nor reintegrative-further illustrating the need for long-term commitment in this area.

Methods
This research examined the defunding and removal of a successful program in Tasmania that provided transitional support and accommodation to returning citizens. Semi-structured interviews were conducted to document the consequences of defunding the program through the voices of the people closely associated with the REO program, resulting in rich qualitative participant data (Berg and Lune 2017). Interviews sought to address two research questions: (1) What were the consequences of defunding this program on current and formerly incarcerated persons and on service providers? and (2) What factors (such as external funding and community perceptions) affect the sustainability of (re)integration programs?
We employed qualitative research methods, including data triangulation (Guion et al. 2011). Participants (P#) were categorized into two cohorts: (1) service providers/case workers involved directly with REO; and (2) public servants with contact with incarcerated persons through their employment (such as those from Community Corrections, the Tasmania Prison Service, legal services, and other NGOs). The final sample included eleven participants: four service providers and seven public servants. This enabled us to triangulate issues raised by both cohorts, thereby enhancing the validity of our findings (Schwandt et al. 2007) and eliciting nuanced data about the impact of precarious (re)integration environments like Tasmania.
We employed a consultative approach that depended on the generous cooperation of the key service provider: The Salvation Army. Collaborating with the research team, Salvation Army workers gathered background information and facilitated recruitment by serving as the initial contact for potential participants. Participants were recruited through snowball sampling, supplemented through purposively targeting persons with specialized knowledge (Berg and Lune 2017). Participants had extensive knowledge about and involvement in Tasmanian (re)integration programs and were connected closely with incarcerated persons.
The Social Science Human Research Ethics Committee (Tasmania) granted ethics approval for the study, and interviews were conducted from July 2018 to October 2018, ranging from twenty-five minutes to an hour and thirty minutes in length. Interviews were audio-recorded, transcribed, and analyzed using inductive thematic analysis (Braun and Clarke 2013), and data were approached without preconceived ideas about the consequences of defunding a supported accommodation program. Analysis was conducted in three phases by the authors to identify key patterns in the data: first, we printed the transcripts and color-coded core patterns based on our research questions; second, we categorized data by identifying groups impacted by REO's defunding and distinguished consequences relevant to each group; and third, we conducted more fine-grained analysis of data sub-themes and tested categories and themes using coders unfamiliar with the data. This process enabled themes to emerge from narratives of participants as guided by the research questions (Braun & Clarke 2013).

Loss: Consequences of Defunding a Tasmanian-Supported Accommodation Program
The consequences of defunding the REO program were far reaching and significant, summarized well by one participant: "It is a real shame, and it is still a loss, and I still think we feel it today" (P8). The participants identified a range of consequences, including gaps in service provision, the loss of a specialized service, increased prison populations, and loss of organizational reputation: "It looks bad on them that they've lost a program, even though they had nothing to do with it" (P11). Participants also noted individual consequences, including irrevocably devastating outcomes: "Just taking away a program at the end of the day was enough to actually push at least one of our people over the line. And he hung himself from the doorhandle in his unit. Jail style" (P2).

Loss of Specialized Service and Relationships
As noted above, Housing Connect took over the accommodation component, even though prior to REO's dissolution, numerous people emphasized that REO also provided transitional support (Billings 2015;Parliament of Tasmania 2015;Smiley 2015)-what we have referred to as "floating support" and which Housing Connect did not have the resources to continue. There was no doubt participants considered this loss of transitional support a significant service gap resulting from REO's defunding: "nothing provided, no routine, nothing. All of a sudden, they just have to do it on their own" (P9) because "nothing was happening" (P10) in the rehabilitation space.
The loss of floating support like this was evidenced most prominently in how, even though the program had ceased to operate, many of the workers continued to assist returning citizens on a voluntary basis so that people would not be left completely without support: [W]hen the life or the activity period of the project was coming to an end, people were left high and dry. And it was sometimes only through the goodwill of people supporting them, following the conclusion of the funding, on a voluntary basis-out of the goodness of their hearts (P4).
The complexity of the gap left by the defunding of the REO program was captured by one participant who highlighted loss of relationships as a major concern: The difficulties when a program like REO gets defunded, and even if it is, I guess, reconstituted in a slightly different format later, there has not only been a gap in terms of providing that support for people who would have been able to access it, but also the relationships lapse a little bit, the continuity lapses, the capacity to strategically plan a program is lost (P1).
Participants experienced a significant loss as a result of defunding REO, which was further emphasized with comments in relation to loss of specialized services.
REO assisted returning citizens with their accommodation needs pre-release, but also provided holistic transitional support in the form of a floating support case worker who, as noted above, assisted with integral life skills, such as budgeting, education, and employment through an individual case plan (McCrae 2011). Participants noted that "REO wasn't just bricks and mortar" (P10), but a "specialized service" (P8): "Prisoners have specialized needs, because prisoners-generally speaking-have mental health issues, often have poor physical health, they have sometimes chronic behavioral issues" (P7). Participants also referred to the "lack of expertise in the area" (P2) since its dissolution. This lack of specialized services meant that "prisoners went into a pool where they had to compete for time" (P3), which ultimately meant "that if you are a prisoner, you are down the bottom of the list" (P7). The decision to task Housing Connect with housing returning citizens was criticized by participants because this service was "for everybody, and to expect this population to fit into that somehow, it is unrealistic" (P8). Participants suggested Housing Connect did not have "specialist knowledge to work with this particular group [returning citizens]": "that real classical transitional support was lacking" (P9). Participants were highly dissatisfied with this situation: "It is just crazy. There is no support there. Yes, they got the house-well done, they have a place to live, terrific, but what support? Nothing" (P11). It is evident that the loss of a specialized service with floating support, tailored to meet the needs of returning citizens, resulted in a forgotten cohort lacking access to housing or transitional support, which led to further unmet needs.

Unmet Needs
With REO's defunding leaving such a substantive service gap, "most inmates that needed that service, they were just lost basically" (P9). The interviewees emphasized that "people aren't getting their needs met" (P5) and this meant "they are still looking for that support" (P11). Assigning returning citizens to an accommodation provider without re-integrative transitional support knowledge about releasees and their needs meant "people [who] would ordinarily need the intensive, or really specifically tailored, support on release would just go in with everybody else through the same, and they just fell through the cracks" (P8). Participants suggested "REO was able to make those connections, but without them,... [t] here was a whole, huge group of people with really, really serious issues who were missing out completely" (P9).
Unmet needs were a core concern for people involved with REO. This, however, was not just a time-specific moment after the program's dissolution, but an ongoing issue for service providers and returned citizens. Participants talked at length about how "quite a few inmates after-even years after REO left-would ask about it" (P8), and that service providers said they "still get calls from prisoners just coming out and they just want to know if I can help them" (P11). REO's effectiveness in meeting returning citizens' needs is evidenced in how they return to service providers "just to debrief, to download some things that are going on in [their lives] and will reflect on some of the things that are happening for [them]" (P2), and that service providers "still talk to these people... [So-and-so] just called in the other day" (P5). Defunding REO "left a huge hole" (P9) comprised of specialized floating support case workers, and subsequently resulting in multiple consequences for returning citizens.

Consequences for Returned Citizens
In the absence of floating support aligned with REO as a transitional supported accommodation service, participants in this study reported a range of social consequences, from a lack of support generally, a loss of trust, and social exclusion, to physical consequences, such as a lack of housing options and suicide of returning citizens.

Lack of Support
One of the core issues noted by participants was the lack of support since REO's dissolution, and how that impacted returning citizens that "would get out without that sort of support" (P9) as now "there's simply not the rehabilitation and the relevant support" (P5). Sometimes, people "had somewhere to go, but they needed that extra sort of support, a person walking with them... And that's kind of missing" (P9). This lack of support was highlighted by participants, who argued that "current inmates don't get the help that they require" (P5) referring to a forgotten cohort with 'the clients that could have been helped but weren't" (P11).
Reoffending was a core concern, with participants emphasizing this as highly probable without support: "the lack of support when somebody gets out-like a REO type of thing-that certainly doesn't help [with someone reoffending]" (P8). Release and parole were recognized as "already hard enough" (P8); as one participant put it, being released after incarceration "is kind of a big shock" (P10). Support was vital because "the stress of actually applying for parole and the impending release is sometimes more, causes more anxiety than perhaps getting sent to jail in the first place does" (P2). (Re)integration success was contingent on support and opportunities: "prisoners are vulnerable, returning to the community requires support if they are to become successful, returning citizens" (P7). In the absence of these, "common sense tells you that you are going to get worse outcomes" (P7) because returning citizens were lacking transitional support after REO was defunded. There was no evidence in participants' narratives to suggest that assigning Housing Connect to provide accommodations for returning citizens would make it a "fair and equitable" system prioritizing "those most in need, not who they are" (Smiley 2015). In fact, there is significant evidence that housing was not provided to previously incarcerated persons.

Lack of Housing
Despite promises that Housing Connect would ensure a fair approach prioritizing people in need, a Right to Information request showed that in the year after REO's dissolution, 109 incarcerated persons had applied for housing, but none were allocated a house (Tenants' Union of Tasmania 2016). Statements from participants in this study provided further evidence of this: "I have not known any prisoners to have been housed by Housing Connect" (P7); and "no one had been housed since REO was defunded" (P6). At the time of our interviews, the program had been defunded for three years. REO's success in providing housing and the gap since its dissolution meant that "quite a number of inmates had no access to any accommodation services, or they were scratching around looking for them" (P10).
The handover to Housing Connect did not necessarily mean that incarcerated persons were not offered houses. In fact, "through that process, about three or four have been offered properties" (P9). The problem, however, was that the houses that were offered were deemed unsuitable, due to location and lack of proximity to supervision: "some of the properties that were being offered were on the east coast, and Parole doesn't have any offices over there, and most of them would be knocked back" (P9). Participants highlighted unique complexities making housing suitable for releasees, and Housing Connect was unable to accommodate these subtleties. As such, despite best intentions to provide properties, at the time of interviews, no releasees had been successfully housed since 2015.
As noted earlier, complexities of housing releasees have repercussions for parole eligibility, as the Parole Board "can't parole someone who has nowhere to go" (P10). This means incarcerated persons serve their full sentence, which contributes to prisons reaching capacity, as well as persons being released without any support. It also increases the risk of incarcerated persons being released into potential homelessness, creating circumstances that could lead them to "just go back to what they did before" (P11). Upon release, returning citizens have fourteen days to contact Housing Connect, upon which their application will be backdated to the original Expression of Interest date. Considering the housing climate in Tasmania at the time of the study, however, where rental properties were expensive and difficult to access (and remain so), and that the waiting list for those requiring priority housing reached seventy-two weeks in 2018 (Humphries & Whitson 2018), one could argue that incarcerated persons were not receiving required assistance and could not access housing. Participants discussed how this, in turn, led to a loss of trust in the systems established to support returning citizens.

Loss of Trust
A core reason for REO program success was support workers and their ability to build relationships and trust with returning citizens-a core form of floating support. Trust was integral to support work as service providers "can't move forward without it" (P5). Developing this trust "takes time" (P3) with people who have "been let down time and time again" (P8) because they do not "have a great deal of trust in people" and "a complete lack of trust in the system" (P5). Participants highlighted the central importance of relationships and trust, stating that "once they trust you, then it can be all talk, and you can help people more" (P11). Relationship-building happened over a long period of time to "find out about their families" (P2) and trust was built through reliability and dependability: "what we said we would do, we did" (P5), including providing returning citizens with necessities, such as food and "all the furniture they needed" (P11). Defunding REO resulted in a "breakdown of trust in the actual community that [service providers] were trying to service" (P5), a lack of "faith in organizations", and "significant cynicism" (P9): The things that I thought about after that were what message are we sending to people that are finally trying to change their attitudes towards trusting us? Those particular people. They aren't interested when I say to them I haven't got the money to help you anymore. They hear I don't want to help you anymore. There's a very big difference in those two things… That's the biggest implication I saw, is that we broke down trust that took years to build (P5). This is an important consideration, as effective transitional support relies on trust, thereby further impacting a person's propensity to apply positive change in their life, which would ultimately enhance community safety. This lack of trust has flow-on effects when combined with the apparent lack of options for returning citizens, and a lack of hope.

Without Options and Hope
All participants highlighted a lack of options for currently incarcerated persons and returning citizens as well as emergent feelings of hopelessness after defunding REO: "It's like giving someone hope and then taking it away" (P11). This contributes to "doubt that [incarcerated persons] can do something about it, or that they have some sort of control over their lives" (P8), and this hopelessness was very challenging when returning citizens struggled "at the best of times to think that the world is a fair place" (P8). Participants highlighted a pre-existing lack of trust due to having "been in contact with NGOs and State services all their lives and basically been failed" (P5). REO's dissolution resulted in a lack of "alternatives for those in the position of having nothing" (P11), and "the helplessness and hopelessness" (P8) that emerged when people realized services were removed. This is a core function of floating support: case workers would broker relationships with other services for returning citizens so that they had options upon release. Dissolving REO meant a lack of pre-and post-release options in a prison system where "you're told what to do" (P11) and was seen as exacerbating an already pressured environment: "behavior can sometimes go a bit off because if people feel that they are optionless, they get scared and angry" (P8). " [B]ecause there [are] no other options for you" (P11), returning citizens must focus on "how they can survive" (P5)-something compounded when released without floating support to help broker connections with support services. Worse still, the lack of options sometimes means remaining in the system for longer than required.

Remaining in the System
Having "a release address to go to before you can release" (P11) is a requirement for parole, which inevitably meant that, after REO's defunding, people who were otherwise eligible for parole were "needlessly languishing in prison" (P2) due to a lack of accommodation: "they would spend a lot more time in jail, when all things being equal they were ready for parole, apart from the fact they had nowhere to go" (P10). According to participants, in the absence of floating support "to line up properties for them, they just stay incarcerated" (P3). This meant people remained incarcerated even though they could "do really well on parole with that extra support-the counseling, the drug testing, and the learning to manage in the community, which is hard" (P8). It also meant that "if somebody goes straight to release, with no post-sentence order like probation or parole, they get nothing" (P8). Most importantly, participants stressed "the continued deprivation of liberty for people who are otherwise eligible for parole" (P7) as a grave consequence of defunding REO: "to keep people in prison simply because they have nowhere to live is unconscionable" (P7). Housing is imperative for (re)integration and "critical to a successful application for parole" (Parole Board 2016). Participants commented that not having accommodation "is not a good enough reason to keep some poor soul in [prison] that could actually do really well on parole with that extra support" (P8), with particular concern expressed around how this could motivate survival offending.

Reoffending
All participants stressed the importance of stable housing, emphasizing that "the more times somebody moves from exiting prison, the more chances you've got of increasing the recidivism rate" (P2). For example, Baldry et al. (2006) found that those who move twice or more are more likely to be re-incarcerated nine months post-release than those who do not (59% compared to 22%). Participants, themselves, pointed to the link between homelessness, reoffending, and survival offending, arguing that the lack of support "generally inspires more crime, and then you go back, and you have a roof over your head, three meals a day" (P10; see also P2) because "you are desperate, you have no hope, you gravitate back to the antisocial groups... committing offences to get some cash to get a room for the night" (P7).
These comments are paradoxical in light of government officials claiming incarceration to be a rehabilitative process, that will halt, or at the very least curtail, offending. Yet the comments of participants in this study highlight clearly that without a supported transition process-including explicit floating support-release, itself, serves as a criminogenic factor that leads to reoffending. These complex circumstances are only further exacerbated by the significant social exclusion experienced by previously incarcerated people.

Social Exclusion
All consequences outlined above contribute to social exclusion, with returning citizens being among the most socially excluded members of society (Jones-Finer and Nellis 1998), lacking housing and therefore a sense of community and belonging (Baldry et al. 2006). The interviewees made it clear that the REO program offered more than accommodation and floating transitional support. It also provided prosocial modeling and mentoring, which "is invaluable and probably the most valuable thing for any kind of support program like that" (P8).
Participants wholeheartedly agreed that returning citizens "need to have a roof over their head, have a friend-they need to be touched, they need to be looked at, and they need to be listened to" (P5); they also need a job. These comments highlight the central importance of release from prison as a social experience. The REO program provided opportunities for housing, employment, and prosocial relationships-and presented an opportunity for extra floating support alongside their probation officer. The loss of these social bonds was felt significantly by people involved. The defunding of REO "remove[d] that extra contact... that friendship or support" with individuals who, because of their offending, "have no real contacts anywhere" (P10). Participants recognized that institutionalization causes many difficulties because "if people stay in jail long enough, that becomes their culture... their support network" (P3). Prison is "quite a society... like a little town" (P8) with "a lot of your associates, your mates" (P2). In stark contrast, the experience of release is palpably different, "with nothing provided, no routine, nothing, all of a sudden they just have to do it on their own" (P9), which "can actually mean that they are very much on their own in a little unit somewhere in an area where they haven't lived before" (P8). This social isolation can become particularly pronounced for returning citizens with parole conditions that state they are not to associate with certain persons-an onerous undertaking to fathom, let alone uphold, as Halsey (2010) suggests.

Consequences for Service Providers
The consequences for service providers ranged from frustration and hopelessness, to loss of human resources, to adverse health implications. It was abundantly clear that the consequences for incarcerated persons were also profoundly felt by service providers.

Confusion, lack of Understanding, Frustration, Powerless, and Optionless
There was an apparent confusion and lack of understanding about why REO was defunded when it was "clear as day that persons who do receive treatment, who are assisted with (re)integration into community, are more likely to be a good outcome" (P6). This confusion was coupled with an inability to understand the reasoning for its defunding: "It was really confusing, we couldn't understand why they were taking the money and not replacing it with anything specialized at all" (P9) when "we've proven the money we saved by doing this, and the people's lives got back together" (P11). The REO program successfully reduced the rate of recidivism to 6.5 percent (Lloyd et al. 2013), "despite the fact that [service providers] took on some clients who everybody in the world had no expectations of-they said they would never rehabilitate, they would never reintegrate" (P2). Participants reported disbelief and stated that "if anything, you should do the opposite" (P11) and "expand it" (P7). This was the original recommendation from the independent evaluation of the program (Lloyd et al. 2013) and precursor evaluations (van Aaken 2010; White and O'Halloran 2011). This lack of understanding stemmed from a belief that "governments got a responsibility to fund these programs because they are in the public's interest and deliver public benefit" (P1).
Participants expressed significant frustration about the lack of a reason to defund a successful program: "it's frustrating" (P5), "very frustrating" (P6), and "you can't understand the reasoning behind it all, why this is happening" (P11). This frustration was coupled with feeling "powerless" (P11) because "all that work, for all those years, meant nothing" (P11). This raised feelings of impertinence with participant five noting "[i]t was a waste of my time. It's kind of an insult, really," with this comment immediately followed by "well, it wasn't a waste because we got a number of people that are doing the right thing" (P5). This discouragement became more apparent as participants spoke of how floating support case workers "lose a bit of hope" (P8). The frustration and feelings of powerless and hopelessness for their clients left support case workers feeling like they did not have any options to provide to returning citizens.
Participants spoke of REO's defunding as producing a lack of options for them to provide a service to returning citizens, because "you no longer have that option" (P7), leaving them to ask: "who do we contact about this?" (P8). This notion of being optionless can be further appreciated from the following statements: We hear this a lot. "Have you got somewhere to live?", "No." And I used to be able to put people in touch with REO. And I did. But I can't do that anymore, and I think that is an indictment of our society that that happens (P7).
And very hard for us as workers too, to be doing, having conversations like that. I remember seeing a young guy who was about early 20s, and he was leaving in two or three days, and he just kept saying, "I need somewhere to go, I have nothing-what am I going to do?", and the only thing I could offer him was a Housing Connect phone number (P9).
Being unable to provide options went hand in hand with the hopelessness discussed above, with participants saying, "it is disheartening" (P8) and "it alarms us that we can't place people. We can't say to people: 'Have you contacted REO?'" (P7). The unmet needs and lack of floating case work support since REO's dissolution meant that returning citizens were "still reaching out for that bit of help" (P11), but providers were limited in support they could provide given they were unpaid and they no longer had relationships with services they brokered in the case work role. REO's defunding caused disbelief, frustration, hopelessness, and feelings of not being able to provide returning citizens with any options-issues exacerbated by the loss of human resources and relationships.

Loss of Human Resources and Relationships
REO's defunding meant losing vital human resources: "the service to clients was reduced and the human resource was lost" (P3). This had a significant impact on human capital: "I had a really good team... a brilliant team of people, who went and never returned to doing this kind of work" (P2). Participants said that "one of the first things we lost were the people who were working in the program" (P3), which "was a great loss, because they ended up having years of experience" (P2). The loss of human resources was keenly felt as a successful program "needs the right people" (P5): "experience is 70% of the job" (P3). Furthermore, participants reported that "people get very nervous at the end of a financial year... because we need to know who's coming and who's going and where the gaps are" (P9), which had a knock-on effect on "staff retention and staff morale" (P9).
Losing organizational human resources also meant flow-on effects for inter-agency relationships because "you lose those sorts of connections [between agencies]" (P2): they lost their status as welfare intermediaries. It was acknowledged it "takes a long time to actually develop good, trusting professional working relationships" (P2), and when a program is defunded, "you have to go through a process of seeking people to develop the networks" (P3), as the "relationships and trust are gone during those years" (P5). Losing connections and networks presented considerable barriers to good service provision for returning citizens. Most importantly, participants shared how the substantial stress caused by these processes created extensive implications for their long-term personal health and well-being.

Pressure and Health Implications
Numerous participants highlighted pressures associated with starting and facilitating a program, as the complexities of applying for funding in a competitive tender environment with short-term funding "stifles so much good work" (P4), with a potential risk "that [incarcerated persons] would be forgotten" (P10). Applying for funding requires "a great deal of energy, and thought, and time and money" (P5), but it was also about how defunding REO meant adding this service onto an already saturated system-including Housing Connect. This was "not fair on untrained workers... and those organizations" (P9), resulting in negative consequences for service providers: "caseloads had doubled, so people's stress levels had doubled" (P3). It is clear short-term programs, and intermittent program defunding, brought about concerning work circumstances for service providers.
With increased caseloads and stress, it was unsurprising participants noted these consequences impacted their long-term health. This added pressure on service providers, coupled with workers investing "a lot of themselves in their jobs... in their clients" (P4), was highlighted as a potential for "stress leave, family relationship breakdown-people [who] were providing the service... are now requiring [that] service because of their breakdowns, the level of drug abuse and self-medicating to get over things" (P5). Participants spoke of the difficult nature of their jobs hearing the tragic life stories told by returning citizens: "just heartbreaking... it is part of our jobs to hear those stories" (P9). More concerning, though, was taking on some of the returning citizens' hopelessness and "feel[ing] so disheartened about this business that you just give up or become cynical" (P8). Vicarious trauma was a reality for people working in the sector and participants had "seen evidence of that" (P4) throughout their journeys. The added pressure, and the uncertain nature of the work, was exacerbated by not being able to get anywhere, and being "willing to give [funding bodies] more than what you're going to pay me for, because I am passionate about this, but you constantly take that away" (P5), resulting in burnout.

Conclusion
Defunding the REO program had substantial consequences felt by all parties. Removing the specialized service led to unmet needs of returning citizens, including a lack of housing, which contributed to recidivism, as well as parole-eligible incarcerated persons remaining in prison. Most importantly, even if Housing Connect could provide housing for returning citizens, they were ultimately unable to provide a key component of REO which enabled successful (re)integration of returning citizens: floating support. The lack of floating support meant a lack of brokered connections to welfare services which, in turn, meant REO was still talked about, and returning citizens still asked to be referred and sought this form of floating support with former program workers. Program dissolution and support discontinuation destroyed relationships and trust developed through significant time and labor, and all this in the very community service providers were tasked to assist. Interviews evidenced sustained disbelief, frustration, and hopelessness about the defunding of a successful program. Service providers experienced increased pressure because they lacked options for support referral due to lost networks with welfare services and they became disheartened and cynical-something that produced long-term health outcomes, such as burnout and vicarious trauma.
Certain limitations must be considered when interpreting the findings of this research. First, this is a small-scale study conducted in the Australian state of Tasmania. Second, the number and type of participants in this study were restricted to service providers. Third, this study was completed as part of a Master's degree and due to time and ethics restrictions, we were unable to interview returning citizens with lived experiences of being released, meaning our findings may not reflect their experiences-an important focus needed in future research. Fourth, we were not able to interview Housing Connect workers and, as such, they could not comment on these issues. This could be another important focus of future research. Furthermore, our interviewees may not be representative of the wider (re)integration sector. Considering that four participants were service providers directly involved with REO, we must recognize that there is a potential for bias in favor of the service. It should be noted, however, that narratives from all participants-seven of which had no direct involvement with REO, and thus no vested interest in its ongoing operation-have been uniform.
These limitations aside, there is no doubt Tasmania has seen increased recidivism and incarceration rates (Australian Bureau of Statistics 2021b), and an increase in parole applications being denied, due to the lack of this specialized accommodation service provided by REO (Parole Board 2018). All participants recognized, which has been confirmed by previous research, that floating support, accommodation, and employment are imperative to change offending behavior, which in turn improves community safety. As such, REO's defunding was, and still is, considered a great loss to Tasmania's (re) integration sector. Further research is needed that unpacks the impact of defunding successful (re)integration programs on returning citizens and service providers. There are clear benefits of third sector organizations, such as NGOs, in the absence of government commitment, but this can be complex. For instance, service providers working with NGOs and the returning citizens being supported by them are subject to what we call "program churn"-starting, defunding, and reinstating programs rather than continuous funding and support. This, too, requires further examination.
The second part of this study focused on implications such as lack of support, relationship breakdown and loss of trust, perpetuated failure, marginalization, and stigma for returning citizens. We also considered the loss of human capital and relationships, cynicism, disengagement, well-being, and burnout for service providers, as well as the underlying factors to program churn: competitive tendering, populist tough on crime politics, and fear of crime (Herrlander Birgerson 2018).
Overall, the data from this project demonstrate the volatile environment created by government whim for some of the most vulnerable people in our community-and how short-term political economies perpetuate crime and poverty (Garland 2017(Garland , 2020a, meaning returning citizens are unable to "live securely and with dignity" (Currie 2010: 120). This environment is fueled by the substantial public anger directed at attempts to reintegrate returning citizens and leaves politicians little room to support successful re-integrative initiatives when they are concerned about reelection (Garland 1998). Returning citizens are "a suspect population" (Garland 2020b: 313), "a breed apart" (Hogg 2002: 239) almost universally devalued as human beings, but most importantly, seen by politicians as "a cost to be minimized" (Garland 2020b: 313). This situation is compounded further when the (re)integration programs are aligned with third sector organizations where labor is voluntary, charitable, and not funded by government or is focused on profit (Hinde and White 2019). Re-integrative programs operating in this space are directly shaped by politics. All these ideas collide in the data analyzed in this article: returning citizens are devalued and subsequently denied the opportunity to live a safe, dignifying life supported by (re)integration through REO. This program was a cost minimized by a short-term government with no regard for the needs of returning citizens.
We might argue that government decision-making that saw the discontinuation of REO was driven by what Currie (2011) calls "spurious prudence"-the inclination to ignore the sizeable evidence showing that (re)integration works and that punitive approaches do not. (Re)integration initiatives, such as REO, must be grounded in "antipoverty strategies, job creation, educational transformation, and public health" (Currie 2010: 122) to ensure they are able to reintegrate returning citizens in line with indicators of success expected in these programs. Yet, even when success is well evidenced in evaluation data, political agendas can result in program termination and can engender even deeper uncertainty in the lives of returning citizens. Data presented in this article demonstrate unequivocally that government decisions made in a political vacuum can have profound effects on the lives of returning citizens, perpetuate the cycle of poverty and crime, and greatly reduce their life chances and their capacity to shape a life without reoffending. There is a very clear need for programs like this to be funded in an ongoing way by governments, for legislation to ensure that spurious prudence is not a possible option, and to remove decision-making about these programs from political influence.
Funding Open Access funding enabled and organized by CAUL and its Member Institutions.

Conflict of interest
The authors declare that they have no conflict of interest.