Introduction
Government-funded, nonprofit-provided services are the norm in many social policy areas (Salamon, Reference Salamon2015). Given this pivotal role, understanding the consequences of nonprofit social welfare is important for the future of social policy (Cammett and MacLean, Reference Cammett and MacLean2014). When governments acquire nonprofit social welfare services (SWS), they create institutions of acquisition – institutions which can be, but are not necessarily, markets. The rules and practices that governments adopt in acquiring SWS define who is able to participate, on what bases, and how prices will be determined. Through a comparison of Canadian and English homelessness, this paper demonstrates that the effects of contracting vary depending on the characteristics of government regimes for acquiring nonprofit SWS.
Building from insights in the public service contracting literature (e.g. Greer et al., Reference Greer, Breidahl, Knuth and Larsen2017; Considine et al., Reference Considine, Lewis and O’Sullivan2011), this paper argues that SWS acquisition regimes can be marketised to a greater or lesser extent, and that this is influential through its effect on nonprofit competition. Depending on the institutions that they select, governments embed different values in their SWS acquisition regimes. Those regimes that emphasize free market competition create incentives for nonprofits to compete on market values, especially through price-based competition. In contrast, regimes that embed non-market values such as collaboration and service stability are more likely to produce collaborative, trusting relationships – albeit at the expense of accountability. Overall, this paper finds that marketised SWS acquisition regimes are more likely to erode service expansiveness in the long-run, because they create incentives for participants to reduce prices by loss-leading or ratcheting down service quality.
This paper uses a case comparison approach, based on document analysis and semi-structured interviews, to illustrate the argument. Two case studies present distinct systems with differing levels of marketisation. England’s competitive tendering process for acquiring homelessness services is highly marketised. Nonprofits participating in this regime were most likely to compete on the basis of price and least likely to collaborate, suggesting that incentives reward market-like behaviour. In contrast, Canada’s regime of homelessness service agreements and grants emphasized provider stability, perhaps at the expense of accountability. Nonprofits in this regime preferred to differentiate their services and were positively disposed to collaboration.
Context
Neoliberal Privatisation and Welfare (Quasi-) Markets
Since the 1980s, governments have increasingly sought to provide SWS by contracting with businesses and nonprofits (le Grand, Reference le Grand1991). The rise of public services contracting has been framed as part of a broader neoliberal transformation of welfare, so much so that all contracting tends to be described using this privatisation narrative (e.g. Cordelli, Reference Cordelli2020). The neoliberal privatisation narrative envisions the state as outsourcing SWS to private actors in the pursuit of efficiency gains through market competition – even if public contracting systems never truly resemble conventional markets, making them quasi-markets (le Grand, Reference le Grand1991; Flynn et al., Reference Flynn, Pickard and Williams2009; Knapp et al., Reference Knapp, Hardy and Forder2001; Lowery, Reference Lowery1998).
Welfare markets or quasi-markets have received considerable research attention. Scholars have pointed to various ways in which welfare markets potentially promote efficiency (or not) (e.g. Bradley et al., Reference Bradley, Crouchley, Millington and Taylor2000). Yet welfare markets also transform the nature of citizenship, weaken social rights and accountability, and under certain circumstances undermine service access, equity, sustainability, and quality (Bradley et al., Reference Bradley, Crouchley, Millington and Taylor2000; Cordelli, Reference Cordelli2020; Dean et al., Reference Dean, Goodlad and Rosengard2000; Knapp et al., Reference Knapp, Hardy and Forder2001; Pue et al., Reference Pue, Westlake and Jansen2021; Buckingham, Reference Buckingham2012; Greer et al., Reference Greer, Breidahl, Knuth and Larsen2017). These debates highlight the importance of understanding the characteristics of SWS contracting and their implications for providers, governments, and citizens.
The neoliberal privatisation narrative has become so ubiquitous that all contracted SWS provision tends to be viewed within this lens. But, as Greer and colleagues (Reference Greer, Breidahl, Knuth and Larsen2017) have identified, contracting regimes can be marketised to a greater or lesser extent. Competition is fundamental to the creation of welfare quasi-markets and is a defining feature of a marketised contracting regime (Greer et al., Reference Greer, Breidahl, Knuth and Larsen2017). In practice, however, welfare markets are not always very market-like. Even the most marketised systems – those in which nonprofit and for-profit organizations compete in open processes following public procurement rules (Greer et al., Reference Greer, Breidahl, Knuth and Larsen2017) – often only have a few bidders. Market forces are tempered through political commissioning processes and ‘soft’ contracting cultures (Rees et al., Reference Rees, Miller and Buckingham2016; Mackintosh, Reference Mackintosh2000). Previous studies have found that European contracting regimes exhibit a mix between logics of hierarchy, competition, and co-operation, even though they are ostensibly market systems (Ascoli and Ranci, 2002; Bifulco and Vitale, Reference Bifulco and Vitale2006). So, even marketised welfare contexts are more nuanced than the privatisation narrative suggests.
The neoliberal privatisation narrative is still appropriate for welfare (quasi-) markets, even if these systems do not perfectly resemble markets. But it gives short shrift to the nonprofit welfare state, which is not merely a story of neoliberalism. There are many other reasons that governments turn to nonprofits to provide welfare services and formalise the arrangement in a funded contract: for example, a nonprofit’s service expertise, ability to reach populations that distrust government, and knowledge of the local community context, as well as a government’s desire to maintain service stability or to support the voluntary sector (Pue, Reference Pue2021; Taylor-Gooby, Reference Taylor-Gooby1999). And the institutional arrangements established to mediate nonprofit welfare do not always resemble markets, or even quasi-markets. As this paper will show, government homelessness services in Canada are contracted to nonprofits, yet there is virtually no competition for these contracts. Instead, the acquisition regime embeds service stability and collaboration as primary values.
The existence of non-marketised SWS contracting points to a need for market-neutral terminology. It also poses an important question: how do marketised, quasi-market, and non-marketised SWS acquisition regimes differ, and what are the implications of these differences for participating organisations and service users? This study contributes to that question through an analysis of institutional variation and its impact on nonprofit behaviour.
Public Services Contracting and Nonprofit Commercialisation
In nonprofit studies research, competition is viewed as a driver of commercialisation. It is also seen as a mechanism for decreased government spending in contracted social welfare. As such, competition is at the heart of the debate on nonprofit welfare. Competition is defined as “the pursuit of the same objective by two or more firms” (Tuckman Reference Tuckman1998, 176). Nonprofits are in competition with one another for revenue, personnel (board members, volunteers, and staff), beneficiaries, access, and authority (Tuckman, Reference Tuckman1998; Stroup and Wong, Reference Stroup and Wong2017). To the extent that contracted SWS drives nonprofit competition, this raises concerns about how contracting could change the way that nonprofits operate: whether they might become more business-like or lose their unique character.
Some of these questions have been explored by the literature on nonprofit competition. For the most part, this corpus focuses on attitudes toward competition (e.g. Sharp, Reference Sharp2018), strategies of competition (e.g. Chetkovich and Frumkin, Reference Chetkovich and Frumkin2003), and the confluence of competition and collaboration amongst nonprofits (Bunger, Reference Bunger2012; Lammers, Reference Lammers1990). But the body of research which specifically examines the effects of nonprofit competition, whether in nonprofit-only or mixed-mode contexts, is more limited (e.g. Lammers, Reference Lammers1990; Wolff and Schlesinger, Reference Wolff and Schlesinger1998). These studies broadly point to a concern that nonprofit competition will erode the unique characteristics of the nonprofit sector. Eikenberry and Kluver (Reference Eikenberry and Kluver2004), for instance, argue that the marketisation of nonprofits undermines the sector’s role in fostering civil society. Grogan and Gusmano (Reference Grogan and Gusmano2009) find that competition for government revenue can increase political activity to secure a favourable role under a new contracting regime; however, these efforts can undermine a nonprofit’s ability to represent poor beneficiaries.
Competition is conceived as one of the drivers behind a broader trend toward the marketisation or commercialisation of the sector (Eikenberry and Kluver, Reference Eikenberry and Kluver2004; Weisbrod, Reference Weisbrod and Weisbrod1998; Tuckman, Reference Tuckman1998). However, Weisbrod (Reference Weisbrod and Weisbrod1998) notes that competition may in some circumstances strengthen the distinction between nonprofit and for-profit sectors, as well as nonprofit collaboration. Insofar as competition is expected to drive the adoption of market-based strategies, the effect is thought to be especially acute in mixed-form markets where nonprofits compete directly with for-profit organizations (Marwell and McInerney, Reference Marwell and McInerney2005). Competition with for-profits is also argued to change the way in which nonprofits provide services – for instance, influencing the types of beneficiaries they serve (Wolff and Schlesinger, Reference Wolff and Schlesinger1998). The literature has identified competition for government contracts in particular as a source of nonprofit marketisation (Eikenberry and Kluver, Reference Eikenberry and Kluver2004; Alexander, Reference Alexander1999; Adams and Perlmutter, Reference Adams and Perlmutter1991; Smith and Lipsky, Reference Smith and Lipsky1993). The supposition is that government contracting leads to a loss of the public character of nonprofits, as these organizations adapt to the business-oriented approach necessary to succeed (Alexander, Reference Alexander1999).
Although the shift from grants to contracts has been identified as a sort of bogeyman for the sector, the literature has devoted relatively little detailed attention to the characteristics of government service markets themselves. Tuckman (Reference Tuckman1998) acknowledges that public policy can influence whether for-profits and nonprofits compete directly, but this is a passing reference in an article about how nonprofits compete. As such, this paper seeks to contribute to better understanding of the link between the contracting revolution and nonprofit commercialisation through an analysis of government regimes for acquiring nonprofit SWS. It asks: do governments take different approaches to acquiring nonprofit SWS and, if so, how does this affect competition (and nonprofit behaviour more broadly)?
Theory: Characterizing Nonprofit Service Markets
When governments acquire SWS, they create institutions of acquisition. These institutions of acquisition have been referred to in other work as markets (e.g. Gingrich, Reference Gingrich2011), but this paper uses the more neutral term of regimes to emphasize the differing extent to which free-market principles are reflected. Government SWS acquisition regimes establish rules and practices which can emphasize different values, creating varying levels of competitiveness and different bases of competition. Depending on government’s choices, these institutions can reflect free market principles to a greater or lesser degree. This paper begins from the logical supposition that to the extent that an acquisition regime looks like a free market (reflects capitalist values), nonprofits are more likely to face pressures to become more business-like.
Governments can choose to emphasize other values in structuring their SWS acquisition regimes: they can protect nonprofits from direct competition with for-profits, create a presumption of service continuity except in extraordinary circumstances, encourage collaboration, and promote environmental, social, and governance values. Rules and practices such as these seek to place boundaries on competition or to shape the bases of competition away from a focus on price and service quality. In creating institutions such as these, governments shape the nature of the market, as well as the bases upon which nonprofits compete. This section discusses key institutions of nonprofit service markets and introduces their connection to different bases of competition.
There are at least three institutions of SWS acquisition that are potentially salient for nonprofit competition: rules of entry, participation, and assessment. First are the rules of entry: governments determine eligibility rules for potential service-providers. In doing so, they can ‘ring-fence’ the process such that only nonprofits can compete. Or they can allow for direct competition with for-profits. Rules of entry also include whether government invites a small pool of applicants, versus an open process any organization can join. Next, the rules of participation set out how a nonprofit applies for funding. Procedurally, participation can entail written submissions or in-person interviews. The length of written submissions, as well as substantive requirements, can make applications more or less onerous for prospective participants. Finally, governments can establish assessment rules that prioritize different values. For instance, assessment criteria may turn on the lowest available cost or they could privilege values such as service stability, provider experience, partnership and collaboration, or ties to the local community. The inclusion of values outside of the cost-efficiency frame in assessment criteria suggests a less marketised acquisition regime. Assessment rules also determine who is empowered to decide. The inclusion of certain actors, such as people with lived experience, may shape the way in which funding decisions are made. Additionally, modes of evaluation after funds are awarded, as well as the length of the agreement, rules of payment, and the procedures of defining the service potentially shape nonprofit competition in less direct ways.
The content of these institutions of government SWS acquisition shape the manner by which nonprofits compete. Rules of entry can constrain the extent to which there is competition – if, say, a government chooses to value stability by continuing to fund the same organization year-after-year. Rules of procedure, when they are onerous or complex, can limit competition by making it difficult for nonprofits to participate. But perhaps most significantly, different institutions of acquisition shape the bases for competition. As defined by Tuckman, basis for competition refers to “the way that competitors choose to compete to meet their customer/constituent needs” (1998: 176). Rules of entry can create direct competition with for-profit organizations, which might lead nonprofits to compete differently than if the pool was restricted to nonprofits alone (Marwell and McInerney, Reference Marwell and McInerney2005). Bureaucratically onerous procedures for participation can privilege larger, professionalized nonprofits. And, of course, assessment criteria will dictate the characteristics of nonprofits that are likely to be successful, as well as the manner by which nonprofits tailor the service that they are offering.
Research Design
This paper uses case comparison to analyse variation in regimes of acquiring nonprofit SWS. It compares the institutions of acquiring nonprofit homelessness services in Canada and England. The study focuses England and covers two Canadian provinces. Both Canada and England are liberal welfare states with similar approaches to defining and subsidizing the nonprofit sector (Esping-Andersen, Reference Esping-Andersen1990; Salamon and Anheier, Reference Salamon and Anheier1998; Phillips and Smith, Reference Phillips and Smith2014), thus limiting external variation to allow for focused study of variation in the institutions of interest. Homelessness was selected as a policy area because it is a SWS with significant nonprofit provision, while government simultaneously accepts a duty to support affected citizens. The cases were developed based on data from 56 semi-structured interviews with public servants and nonprofit staff, undertaken as part of a broader study on the nonprofitization of the welfare state between April 2017 and April 2019. 38 interviews were conducted in Canada, while 14 interviews were conducted in England. For a full list of participating organizations, see Appendix 1. The larger number of Canadian interviews reflects the decision to analyse two provincial systems with distinct histories and institutions, as well as the asymmetry of secondary data on SWS acquisition: whereas the British service commissioning system has been studied extensively (e.g. Knapp et al., Reference Knapp, Hardy and Forder2001; Macmillan and Paine, Reference Macmillan and Paine2020; Buckingham, Reference Buckingham2011, Reference Buckingham2012), no existing studies characterise the regimes for acquiring social welfare in Canada. The author identified possible interview participants through background research as well as a snowball method, aiming to include perspectives from both government and homelessness nonprofits, as well as nonprofit sector infrastructure organizations (e.g. the National Council of Voluntary Organisations). The interview sample in Canada is weighted more heavily toward nonprofit organizations, while in England participants were largely government officials. This could have impacted the findings, and the author was mindful of this when triangulating interview data with document analysis.Footnote 1
Recognizing that devolution has resulted in separate homelessness policies in the regions of the UK, this paper focuses on homelessness service acquisition in England. Similarly, this study focuses on just two Canadian provinces. Alberta and Ontario were selected for their size – Ontario is the most populous province, while Alberta is the fourth most populous – as well as the anticipated range of variation, since Alberta is typically viewed as the country’s most neoliberal province, while Ontario approximates the country’s middle (Haddow, Reference Haddow2015). This selection leaves out Quebec, a province with high welfare generosity and a distinct history (Haddow, Reference Haddow2015). Thus, the full range of Canadian welfare variation is not covered here.Footnote 2
Interview data provided information on the experiences, perceptions, and expectations of actors participating in the acquisition or provision of social welfare. The interview research received ethics clearance from the University of Toronto Research Ethics Board. For each interview, the author created an interview report, which served as the primary reference for future interview analysis. Interview reports were paragraph-form documents that identified key messages, including some notable direct quotes and anecdotes. Where recordings were possible, these were revisited as needed for clarification or whenever the interview report suggested that a quote might be worthwhile. Where an interview recording was not available for review – largely because the participant did not grant permission to record, but in a small number of instances because interview circumstances were not conducive to recording – direct quotes were taken as recorded in the interview report. The author utilised the interview reports to identify key themes and to compare perceptions. Utilising interview reports is a helpful method for inductively identifying trends and themes. However, it does have weaknesses, including the potential for recall and confirmation bias. To minimise the potential for recall bias, interview reports were completed shortly after the interview was conducted based on notes taken during the interview. To minimise confirmation bias, interview data was triangulated wherever possible through document analysis drawing on legislation, policy guidance, annual reports, and secondary literature. Policy documents were analysed using a deductive approach (Natow, Reference Natow2019), with an eye to verifying (or not) claims made by participants in the interviews.
Case Studies
Canada and England are societies in which homelessness policy is delivered through contracts to private actors, primarily nonprofits. But the two countries use different approaches to acquire homelessness services. In England, local authorities use a competitive service contracting regime that emphasizes free market values. This case closely resembles the classic welfare quasi-market that is familiar in the literature. Contrasted is the Canadian case, in which nonprofit homelessness services are funded through a combination of stable service agreements and grants for project and capital spending. The Canadian case bears little resemblance to a market, and instead emphasizes values of service stability and collaboration.
Competitive Contracting: Nonprofit Homelessness Service Acquisition in England
Policy Context
The United Kingdom has a legally enforceable housing entitlement, which is implemented through statutory obligations held by local authorities. Local authorities thus have a duty to rehouse the statutorily homeless – people who are deemed to meet criteria such as not being ‘intentionally homeless’, having a ‘local connection’ to the housing authority, and being in a situation of ‘priority need’ (Burrows et al., Reference Burrows, Pleace and Quilgars1997). In addition to the main duty to rehouse, local authorities also have a duty to prevent homelessness, provide housing advice, and provide temporary relief from homelessness. As a result, homelessness policy in England is standardized and places more emphasis on permanent rehousing, relative to the Canadian case.
Contracting for homelessness services in England was made possible through legislative changes in the mid-1990s (Pue, Reference Pue2021). Local authorities can – and do – contract out most elements of their statutory obligations to private providers. Contracted tasks include the provision of core services like supported accommodation, as well as ancillary interventions like security or concierge services, service infrastructure, and research, planning, and evaluation. Homelessness policy in England and Canada share an important feature: the contracting of service provision to private actors, primarily nonprofits. However, the two countries have set up starkly different rules and processes to acquire nonprofit homelessness services.
In England, homelessness services are acquired through a competitive social care commissioning process, which is rooted in public procurement procedures and implemented by local authorities. Public procurement rules in England are geared toward providing open, fair, and transparent competition. However, four considerations temper this general policy aim: the local care market, small providers, social value, and service disruption.
The SWS Procurement Process
Public procurement is a process by which government purchases goods or services. In England at the time of writing, public procurement was subject to European Union rules covering public contracts, as well as domestic legislation which implements these rules. The use of procurement – rather than allocating funding to a provider – is an indicator of marketisation, since procurement rules are designed to institutionalize markets (Greer et al., Reference Greer, Breidahl, Knuth and Larsen2017). Legally, local authorities must undertake public procurement processes where the value of a social care contract exceeds £615 278 (CCS, 2017). Below this threshold, local authorities can establish their own procedures. It is usual for local authorities to use procurement procedures for contracts of a larger size but below the threshold, while smaller contracts might be achieved by inviting one or a few providers to submit a quote (author’s interviews).
Procuring for social care entails commissioning, where policy-specific commissioners work with more generalist procurement officers to define and procure a service (Geldards LLP, Reference Geldards2013). The commissioner typically develops the specification, a document defining the service to be provided, and works with procurement officers to develop a suite of documents that set out the procedure for applying, criteria of evaluation, and proposed contract. Once this package is prepared it is posted for a specified period during which prospective providers can tender their submissions. Once the deadline has passed, submissions are evaluated on “quantitative and qualitative” criteria, referring to pricing and service quality, respectively (author’s interviews). The overall aim of including both elements is to find the “most economically advantageous tenderer” (MEAT) – which can mean “Best Value for Money” or the lowest price (CCS, 2016; authors interviews). In general, local authorities achieve this by weighting quantitative and qualitative elements of the bid. Weightings are sometimes set by the Council, while in other cases commissioners have discretion. A common ratio is 70:30, where price is 70% of the weighting (author’s interviews). “Commissioners want high quality and procurement wants a low price […] We try to get them [commissioners] to construct the ITT so you get innovation – so there’s something newer and better” (author’s interview).
Homelessness Contracts 2015-2017
Local authorities in England are required to publish contract notices on the Contracts Finder database (Cabinet Office Efficiency and Reform Group and CCS, 2015). According to a search of this database, 168 homelessness contracts were published or awarded from 2015 through to 9 October 2017. The Contracts Finder data show a high level of nonprofit participation in homelessness contracting, with some for-profit participation. Nonprofits were selected as suppliers for the majority (75 percent) of homelessness contracts, receiving the largest share of the total value of contracts (£242 million out of a total £500 million). For-profit organizations represented 17 percent of the total number of contracts and 6 percent of the total value (£28 million). The second largest category of contracts by total value is a small number of hybrid contracts, encompassing nonprofit and for-profit providers (£216 million).
The Contracts Finder data also suggest that most homelessness contracts are awarded through a competitive tendering process. As shown in Figure 1, the majority of the contracts in this dataset, 81 percent, were awarded through an open or restricted procedure (the two procedures that allow for the widest range of competition). In an open procedure, “any interested economic operator may submit a tender in response to a contract notice” (PCR, 2015: 27). Restricted procedures are similar, but entail pre-qualifying bidders (Mills & Reeve, 2015).
UK Homelessness Contracts 2015-2017, Competition Type.
Source: Contracts Finder database 2015-Oct 2017.
A Market for Homelessness Services?
Public procurement procedures institute a competitive market for contracting which adheres to three values: transparency, fairness, and equal access to opportunity. Thus, this acquisition regime is based on free market principles to a considerable degree. However, there are four considerations which temper the marketisation of homelessness contracting in England: promoting the local care market, small providers, social value, and incumbents and service disruption.
First, the Care Act 2014 requires local authorities to promote a local care market (Department of Health and Social Care, 2018). The government staff that participated in this study all expressed an awareness of the market in the service areas in which they operated and their obligation to facilitate market development (author’s interview). Facilitating markets can entail encouraging new entrants as well as preserving existing participants. For the former, procurement officers and commissioners seek to ensure that providers have heard about opportunities and listen to agencies about barriers to applying for contracts (author’s interviews). To preserve participants, the market shaping imperative tempers the drive for low-cost bids, where this is unsustainable. Procurement officers expressed an awareness that bidders sometimes “loss-lead” and explained that they look for financially sustainable bids. “We don’t want to be squeezing businesses so much that they go out of business,” one interview participant said (author’s interview).
Second, local authorities can take certain limited actions to encourage the participation of small providers. While procurement officers and commissioners stressed that for-profit and nonprofit providers are “all treated equally”, statements of this nature were often followed by a discussion of rules to facilitate the participation of small and medium-sized enterprises (SMEs). The Public Contracts Regulations 2015 incorporated a number of recommendations to encourage SMEs. To make it easier for SMEs to bid for services, procurement officers are encouraged to break larger services into ‘lots’ (smaller, parcelled out services) for which organizations can bid (CCS, 2016). Service commissioners are, furthermore, moving toward consortia-based service delivery (author’s interview). For small charities, joining a consortium can sometimes be the only way to competitively vie for a service contract (author’s interview).
Third, although local authorities cannot restrict procurement processes to nonprofit providers, the Public Services Act (Social Value) 2012 (SVA) requires that social value – “enriched lives and social justice” – be considered in public service contracting (Cabinet Office, 2018: 19). In some instances, SVA requires public sector agencies to consult on social value. For example, evidence of social value might include stakeholder engagement or ethical certifications. But there have been challenges in implementing SVA since its passage. A Cabinet Office report found that “too often the [Social Value] Act is treated as an afterthought, and a ‘tick box exercise’ to be performed for compliance only” (2018: 115).
A final consideration concerns incumbent providers. Formally, procurement processes cannot favour incumbents. However, interview participants expressed that commissioners do consider the effects of de-funding incumbents in their decision-making. For instance, one local government official described a situation in which an incumbent provider has two contracts with the local authority (author’s interview). In that case, commissioners would consider whether losing one contract might affect the provider’s ability to carry out the other. Although the procurement officers and commissioners that participated in this study expressed that this would not occur in a way that breached procurement regulations, the inherent tension between these two ideals was palpable as they described the issue of incumbents (author’s interviews).
Stable Service Agreements and Grants: Nonprofit Homelessness Service Acquisition in Canada
Policy Context
Canada does not have legally enforceable entitlement to housing that sets the policy framework across the country. Instead, provincial governments establish their own homelessness policies and, in some cases, download responsibility for homelessness to local governments. The federal government is also involved in homelessness policy as a funder, through its matched funding program called Reaching Home. In most provinces, governments primarily provide temporary supports for acute cases of homelessness, such as emergency sheltering – although there has been a recent emphasis on permanent rehousing (Smith, Reference Smith2016). Because homelessness is decentralized, homelessness policies differ depending on the province. As such, the analysis here, which focuses on Alberta and Ontario, may not capture the situation elsewhere in the country.
In Alberta, homelessness policy is set at the provincial level, with funding flowing from the province. Alberta’s Ministry of Community and Social Services primarily delivers homelessness services through ‘community-based organizations’, which are nonprofits. Nonprofits in Alberta are funded to provide rehousing programs, adult homeless shelters, women’s shelters, and supportive housing (AMCSS, 2018). Ontario has devolved homelessness policy implementation to local governments, though the province sets the framework for municipalities (York Region, 2017). While municipalities have primary responsibility for homelessness in Ontario, the province provides funding for homelessness through the Community Homelessness Prevention Initiative, as well as a handful of other grant programs (Farthing-Nichol and Pries, Reference Farthing-Nichol and Pries2018). Municipal homelessness programs tend to be made up in large part by funding to nonprofit-operated homeless shelters, although support services and outreach are also typical.
Like England, governments in Ontario and Alberta rely on private providers to deliver publicly funded homelessness services. In Canada, however, the field of providers is typically limited to nonprofits. For example, 94% of the country’s homeless shelters are operated by nonprofits and 5% are government-run (Pue, Reference Pue2021). Both Canadian provinces acquire nonprofit homelessness services in a similar manner, allowing the identification of a Canadian regime of stable service agreements and grants. The Canadian regime bears little resemblance to a market, in contrast to the English system of competitive contracting. Homelessness services in Canada are funded through stable service agreements for core homelessness services, as well as supplemental short-term grants for capital projects and programs outside of the core areas of government responsibility. Under this system there is functionally no competition for service agreements, though nonprofits compete with one another for grant funding.
Stable Service Agreements
Governments in Ontario and Alberta fund nonprofits to deliver core homelessness services through service-specific agreements. Emergency sheltering in both provinces is organized through this model, as well as some supportive housing, homelessness prevention, outreach, and rehousing programs. Canadian homelessness policy is relatively new – arising only in the early 1980s – and the initial homelessness service agreements were arrangements to fund charities already providing homeless sheltering on a philanthropic basis (Smith, Reference Smith2016; Bacher, Reference Bacher1993; Côté, Reference Côté2012). A similar pattern has continued as government homelessness services have expanded over time: when a new program or site is needed, public servants will canvas trusted partners for interest in providing the service, then establish a service agreement (author’s interviews).
In a typical service agreement, government pays a nonprofit operator to provide a service based on usage: e.g. nightly occupancy, for a shelter (author’s interviews). Service agreement funding rates are usually uniform and established through policy, but sometimes bilateral negotiation occurs (author’s interviews). Funding can typically only be used for direct costs of service provision, as a result of which homeless shelters often use philanthropy to subsidize administration. Most governments have shelter operating standards, which establish expectations as to the quality of the service. While shelter operators felt that funding levels were inadequate, and often eroding due to inflation, they expressed that funding was very secure: operators perceived little risk that funding would be withdrawn (author’s interviews). “I can’t think of an example where we have de-funded” a service provider, one public servant said. They continued, saying that “in theory” agencies know funding could be discontinued, “but the reality is we wouldn’t de-fund an agency,” (author’s interview).
Competitive, But Supplemental, Grants
In addition to service agreements for core services, governments in Ontario and Alberta provide project and capital funding through short-term, competitive grants. Alberta organises its homelessness grant funding through the federal Reaching Home program, while Ontario has a handful of grant programs with different rules and focus areas (Pue, Reference Pue2021). Grant funding, whether for capital projects or short-term program funding, typically encompasses a request for proposals that is competitively assessed based on established criteria such as feasibility and program impact (TSSHA, 2018; OTF, April 2017). To encourage collaboration, it is common for government funders to require that applicants obtain letters of support from other community organisations (author’s interviews).
Competition and Bases for Competition
The cases above show that SWS acquisition regimes can differ, even amongst two liberal welfare states. In England, homelessness services are procured through a competitive tendering process that emphasizes transparency, fairness, and equal access. However, it is worth emphasizing that the most open procedures resulted in just a handful of bidders – with five or seven being the high end of participants’ estimates (author’s interviews). These bidders were “usually all” local nonprofits (author’s interview). The Canadian homelessness system is carried out through a combination of grants and service agreements. Generally speaking, grant processes feature more competition than service agreements, since in the latter case existing service providers tend to receive renewals on a more or less automatic basis. Table 1 summarizes the two regimes.
Government Social Welfare Service Acquisition Regimes
With regard to rules of entry, the Canadian homelessness system tends to constrain participation to nonprofits and government – explicitly or tacitly, depending on the policy. In contrast, ‘ring-fencing’ the nonprofit sector was viewed as an illegitimate practice in England. Nonprofits in that case competed directly with for-profits, while the use of procurement rules situated competition in a free-market frame of selecting the best ‘value for money’ amongst businesses. While Canadian grant programs often invited open competition through requests for proposals – similar to the English system of posting contract notices online – there tended not to be a new competition to determine providers of core services like homeless shelters, once the provider was established. As to rules of participation, both homelessness processes included procedures which were onerous, especially reporting requirements. Thus, both cases create incentives for nonprofits to have a professionalized bureaucratic structure. However, in the English system adapting to contracting processes requires knowledgeable procurement specialists, whereas in the Canadian context specialized knowledge is geared toward reporting homelessness data. Thus, only the English context produces market-oriented professionalization. In both cases government officials expressed an aim of reducing procedural barriers to participation.
Finally, assessment criteria in England favoured cost-efficiency, although steps had been taken to incorporate other characteristics. As such, competition rewarded price-reducing behaviour. For example, participants described the tendency for bidders to loss-lead – to deliberately propose a service price below the actual cost of providing the service – in order to gain or retain market share (author’s interviews). Providers would then fund the government deficit through cross-subsidizing with revenue from more profitable contracts or philanthropic donations. Participants estimated that cross-subsidization could cover between ten percent and thirty percent of contract values (author’s interviews). In Canada, grant programs tended to value project impact, local connection, and partnership. Participants described an expectation that grant applicants obtain letters of support from other local providers (author’s interview). Service stability was clearly a core value in Canadian shelter service agreements. Without an ongoing competitive process, however, price was in most cases determined by government on the basis of a uniform price per user formula. This led to the erosion of payments through inertia. As such, in both systems one could identify driving factors that might create downward pressure on the funding provided by government. However, the English homelessness system’s emphasis on competing for cost-efficiency introduced an additional mechanism for decreased service expansiveness that was not present for Canadian homelessness.
The acquisition regimes influenced the bases on which providers competed for funding. The competitive contracting process adopted in England was more likely to result in providers competing on the basis of price. In some cases, nonprofits took on loss-leading contracts in order to gain or retain market share (author’s interviews). Canadian service providers were more likely to differentiate themselves from other providers by emphasizing their own unique contributions (author’s interviews). Table 2 summarizes the effects of the SWS acquisition regimes for competition, collaboration, accountability, and long-term investment.
Government Social Welfare Service Acquisition Regime Effects
The researcher observed a reticence to collaborate amongst nonprofits in the English case, connected to the competitive tendering process – which discourages providers from sharing information and connecting with one another. The transition from grants to competitive contracting in England resulted in organizations taking an increasingly competitive attitude toward one another. For instance: “People started to ask one another, ‘Are you going to bid for my service?’ People that had been working in close partnership became competitors,” (author’s interview). Participants also expressed a perception that the potential for a service to transition from one provider to another resulted in lower staff loyalty and decreased the incentive for long-term investments (author’s interviews). In contrast, Canadian nonprofits were more likely to view themselves as competitors in the context of philanthropic funding sources. Certainly, government grant funding was sometimes seen as a source of competition amongst nonprofits, especially under scarcity. But in general, nonprofits in Canadian homelessness saw advantages in collaborating, and saw collaboration as something government funders wanted. On the other hand, the competitive tendering process allows an ostensibly objective process by which to de-fund agencies. In this sense, it is potentially more accountable. In the Canadian context, failing homeless serving nonprofits were likely to “limp along” (author’s interview).
Conclusions
Many social services that we associate with the welfare state are provided through government-funded nonprofits. While much of the literature has focused on welfare markets, the arrangements that governments use to acquire nonprofit SWS can vary substantially, presenting different institutional incentives for the actors that participate. This paper has examined variation in the institutions for acquiring nonprofit-provided homelessness services in Canada and England, with attention to how these regimes create incentives for nonprofits to behave and compete differently. Case analysis revealed that England’s competitive tendering process is more marketised than Canada’s system of homelessness service agreements and grants. The English regime produced incentives that rewarded market-like behaviour, as a result of which more price-based competition and less collaboration was observed. The Canadian regime, in contrast, produced competition through service differentiation and emphasis on collaboration. Overall, the case studies demonstrate that the institutions of SWS acquisition can vary in ways that are meaningful and impact nonprofit behaviour, even within two liberal welfare regimes. Marketised acquisition regimes appear to have the potential to undermine the welfare state in the long run, through incentives to lower the price of a service.
This study compares just two SWS areas in two very similar welfare states. As such, this study cannot capture the full range of variation: many other arrangements for acquiring nonprofit SWS exist. Emergency management in Canada and England, for instance, utilizes a system of reciprocity and cost-reimbursement that is distinct from both homelessness cases. Nevertheless, this paper has contributed to the literature on SWS contracting by theorizing the constituent elements of acquisition regimes – rules of entry, participation, and assessment – and showing how different values pose long-run consequences for the welfare state, through their influence on nonprofit behaviour. For instance, systems modelled on market competition seem to be especially linked to eroding service quality over time, as a result of incentives for providers to bid down prices.
This study has important implications for debates on privatization and mixed welfare provision. First, insofar as the theory and findings link competition to service erosion, this paper implies that welfare quasi-markets may have long-run drawbacks. On the other hand, this paper presents a case against viewing all contracted social services with the same lens. The institutions of SWS acquisition can vary in important ways. One trade-off identified in this paper is between accountability and long-run service expansiveness. However, there are many more such trade-offs yet to be discovered. Gaining a more fulsome understanding of the different institutional features of SWS acquisition regimes would enable researchers to more reliably identify the causes of variation in outcomes, especially for service quality, accessibility, and equity.
Acknowledgements
I would like to extend my sincerest thanks to Dan Breznitz, Amos Zehavi, Rodney Haddow, Joe Wong, Jane Gingrich, Caitlin McMullin, Susan Phillips, and the two anonymous reviewers for their helpful feedback on drafts of this paper.
Competing interests
The author declares none.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S0047279421000751
