Abstract
Social impact bonds (SIBs) are attracting an increasing amount of critical scholarly attention. As an outcomes-based mechanism for financing social services, SIBs financialize social policy through the logic of impact investing. Responding to calls for attention to the politics of SIBs’ development, and breaking with the literature’s focus on cases from the UK and USA, this article explores the emergence of SIBs in Australia. It employs the concept of “fast policy,” which theorizes why and how policies move across borders, and describes the contemporary conditions that enable them to do so. Using document analysis, the article explores the discursive devices and practices used to justify the “pulling in” of SIBs to states in Australia. It finds that key actors in the Australian social impact world justified SIBs’ adoption using their synergy with powerful, popular policy discourses and practices, rather than engaging in political debates about their desirability. The Australian experience illuminates the power of intermediaries and the investors they represent over the design and proliferation of SIBs, as well as the roles played by austerity politics, policy experimentalism, and fast policy infrastructures in producing a context in which SIBs could be made real.
Social impact bonds (SIBs) are attracting an increasing amount of scholarly attention. As is typical of a project dedicated to marketizing social policy, important debates over the contradictions of market forces and neoliberalism, the spread of inequality, and the financialization of public policy are being applied to and refined through discussions of SIBs (see, e.g. Cooper et al., 2016; Neyland, 2018; Sinclair et al., 2019; Warner, 2013). However, as Williams (2018) contends, scholarship must delve past debates on the features of SIBs into interrogating the political economy of their development, identifying points of contradiction and tension.
SIBs are partnerships between public, private, and third-sector actors organized to privately fund a social program. The outcomes produced by that program, meaning changes in the behavior or conditions of a specified group of service recipients, are measured, and the investors who funded it are paid returns determined by those outcomes. Unsurprisingly, this structure has provoked significant criticism from critical scholars, but that critique is largely based on examinations of SIBs in the countries with the largest SIB markets: most prominently the UK, but also the USA. This focus has produced valuable work situating SIBs in those national political economies and policy-making cultures, particularly in relation to austerity politics after the Global Financial Crisis of 2007–2008 (GFC) (see, e.g. Berndt and Wirth, 2018; Dowling and Harvie, 2014; Harvie and Ogman, 2019). However, existing literature does not extensively engage with SIBs’ geography, meaning that countries outside of these “market leaders” are understudied and that there is little attention to how and why SIBs move across borders.
This article seeks to attend to those gaps. In the interest of diversifying the base from which theory on SIBs is built, I examine the Australian experience rather than the US or UK experiences. The Australian SIB experience is largely absent from the critical literature. This is surprising, as the state of New South Wales (NSW) expressed interest in SIBs in 2010, making it the first jurisdiction outside of the UK to do so. Australia currently boasts more than 10 programs, which is one of the highest totals outside of the USA and UK. Moreover, all Australian SIBs thus far have been undertaken at the state level, as state governments in Australia are responsible for the vast majority of publicly provided social reproduction. SIBs’ adoption in NSW and subsequent proliferation across state borders means that the Australian SIB world is multi-sited. However, compared with US SIBs (Tse and Warner, 2018), Australian sites are closely entwined, offering a novel case study of their mobility.
To problematize and analyze SIBs’ adoption in Australia and proliferation through its states, I engage with the concept of “fast policy” advanced by Jamie Peck and Nik Theodore (2015). 1 “Fast policy” is concerned with the movement of policies across borders. It contextualizes moving policies within conditions that precipitate the reproduction of neoliberal politics through networks of grounded, yet connected, actors and institutions in social policy worlds (Peck and Theodore, 2015). I also draw on Robinson (2015) to conceptualize policy movement as actors reinventing ideas and models to suit contingent ideologies and interests, rather than policies “arriving” through free-floating travel. I operationalize these concepts by exploring the synergies between the politics of policy networks and the devices used by proponents of SIBs to justify their adoption. This article therefore seeks to illustrate the utility of “fast policy” as an explanatory concept.
The article proceeds by explaining how SIBs function, before reviewing the existing literature analyzing them. Next, I discuss debates around “fast policy.” I emphasize the tendency of the conditions it describes to make neoliberal policy models mobile, while ensuring that alternatives are left behind or reshaped. After describing the methodology of the study, I outline the Australian SIB experience and identify some of its key actors. I then present the article’s main contribution by analyzing the discursive devices and technologies deployed by actors to justify the adoption of SIBs in Australia. I organize this analysis around five themes: invoking elsewhere, austerity, experimentalism, mutation, and speed-up.
To conclude, I reflect upon the contributions of the article for understanding the emergence of SIBs in Australia and their broader mobility. I find that SIBs were “pulled in” to Australia because of their synergy with powerful policy discourses and harmonization with local political economic conditions. They were depoliticized: positioned more as convenient savings mechanisms, evidence machines, and a global trend than as a policy response to a crisis or problem of social reproduction. Moreover, the article’s analysis highlights the influence held by actors at the boundaries of the state over SIBs’ design and trajectory, particularly on the production of “lessons” and best practices, further contributing to their depoliticization. The article demonstrates that responding to Williams’ (2018) call for better analysis of how and why SIBs emerged can be approached through engaging with “fast policy” and the broader policy mobilities literature.
Social impact bonds?
SIBs, often called “social benefit bonds” in Australia, are an experimental model for funding social services associated with impact investing. 2 They are partnerships between public, private, and third-sector actors, in which investors are paid returns indexed upon the success of a social program (Warner, 2013). Typically, a government agency enters a contract with a financial intermediary, who coordinates the development of the program and attracts investment. The intermediary also engages a service provider, usually a charity or not-for-profit organization, who carries out the agreed program (Warner, 2013). As the program proceeds, its “outcomes,” meaning the amount of positive change to a specific metric made to the target recipient population, are measured by an independent evaluator. The results of the evaluation are compared with a previously agreed-upon benchmark, typically determined by a counterfactual, random control trial or historical data. How close the outcomes of the program come to reaching this benchmark, or the degree to which they exceed it, dictates the return paid to investors by the government agency who commissioned the program over the period of the bond, which is usually 5–10 years. If the outcomes do not meet the benchmark, investors are not paid, generating the “pay-for-success” effect: Only successful programs earn investors a return.
The value of the return paid by SIBs originates in the money saved by government organizations. Proponents argue that the positive outcomes of the SIB intervention mean that the state will not have to pay for social services that it would have had to fund otherwise (Dear et al., 2016). Interventions target specific populations in the areas of un/underemployment, poverty reduction, homelessness, child and family welfare, health, education, and criminal justice (Brookings Institute, 2019). These are considered areas of high cost to government containing “wicked” problems that, if solved, will benefit society broadly and strengthen government budget positions in the long term.
SIBs were first launched in the UK in 2010 as part of Conservative Prime Minister David Cameron’s “Big Society” agenda. Big Society sought to remove power and funding from politicians and centralized government bodies and build it in community organizations at the local level (Williams et al., 2014). The Peterborough SIB, which targeted recidivism, was the first program implemented. Sir Ronald Cohen, a wealthy venture capitalist, played a major role in this push, as the head of Big Society Capital since its establishment in 2011. Big Society Capital used the money in “dormant” bank accounts to fund social impact projects. In 2013, building enthusiasm for social investment in the UK led to the launch of the G8 Social Impact Investment Taskforce, which in 2015 morphed into the Global Steering Group for Impact Investing. The Group was launched and is still led by Cohen.
SIBs have proliferated across the world throughout the past decade. As of 2019, there were 170 programs in 31 countries (Brookings Institute, 2019). 3 The reach of SIBs is therefore wide, not deep, in that they exist in many territories, but there are usually few of them. This is reflected in the diversity of SIBs’ structures and designs. Features considered core to the model in the abstract are often not present when programs are negotiated and implemented (Arena et al., 2016). The Peterborough SIB has the hallowed status of original SIB partly because it was one of the only programs to contain all of the model’s key elements.
Situating and understanding SIBs
The GFC and ensuing recession is significant in critical scholarship researching SIBs (Ogman, 2019). In a landmark early work, Dowling and Harvie (2014) argue that Cameron’s Big Society agenda emerged as a policy response to a three-pronged crisis of capitalism in the early 2010s, and that SIBs should be understood within that context. Conceptualizing SIBs as policy is important, as it foregrounds the role of the state in their development, against some impact investing proponents who downplay it (see, e.g. Dear et al., 2016; see Berndt and Wirth, 2018 for a critique). Dowling and Harvie (2014) argue that the GFC precipitated crises of capital accumulation, social reproduction, and public debt. Harvie and Ogman (2019) add a fourth: a crisis of legitimacy for liberal democracy, manifest in the global wave of anti-establishment politics in the latter half of the 2010s. By situating SIBs in this political economic context, their emergence is theorized as a policy response to their structural conditions by a neoliberal national government. SIBs provide an avenue for private investment, an impetus for the development of social service programs, funding for those programs without increasing expenditure, and a public image that makes governments appear as if they are solving problems. Austerity—reverence for balanced budgets and saving government money in order to cut taxes for middle- and upper-class taxpayers—is given a makeover, but is essentially unchallenged in what Ogman calls a “progressive neoliberal compromise” (2019: 1). These structural accounts therefore analyze SIBs as a product of dynamics in their national political economies, situated within global trends in capitalist development.
Other scholarship theorizes the relationship of SIBs with the production of knowledge, markets, and market subjects (see, e.g. Berndt and Wirth, 2018, 2019; Langley, 2020; Neyland, 2018). Scholars in this vein argue that SIBs represent an example of “market making”: positioning the actors involved to act in a manner consistent with idealized market principles (Rosenman, 2019: 151; see Neyland, 2018 for a critique). This occurs by sculpting the practices of service providers toward businesslike efficiency (Cooper et al., 2016); “nudging” beneficiaries toward rational consumer behavior (Berndt and Wirth, 2019); and prompting investors to identify financial value in social programs (Barman, 2015). For Williams (2018: 10), SIBs are “the product of, and embedded within, a particular network of actors and tools with close ties to the world of finance and are underlain by a belief in the merits of investment as a way to bring focus, rigor, and results to the public and charitable sectors.” SIBs are thus analyzed in relation to knowledge regimes and discourses, and the actors that are embedded in them.
Both strands of critical scholarship on SIBs are valuable, but neither extensively engages with their geography. While there are exceptions (see, e.g. Bryan and Rafferty 2014; Mitropoulos and Bryan, 2015; Williams, 2018), the influential strains of theory on SIBs discussed previously are overwhelmingly informed by case studies from the UK and USA, with the UK featuring particularly prominently. This is understandable. SIBs originated in the UK, and the US and UK have the largest SIB markets. However, analyzing SIBs as part of the UK government’s policy response to the GFC does not explain why SIBs have proliferated in their wide, not deep, pattern, or what territorializing claims are made on their behalf. This is not just a problem for approaches focused on political economy. Discursive approaches must also confront the differences in imaginaries and rationalities between geographies more and less “plugged in” to institutional, cultural, and ideological understandings of the crisis and austerity. SIBs have not appeared independently in 31 countries, but relationally, with each informing the development of the others. Ignoring this leads to what McCann (2011: 124) calls a “dangerous tendency toward diffusionism,” in which new policy ideas are thought to originate in “creative cities” and spread outwards, carried by their innovative features. Avoiding this tendency means interrogating how actors invoke, elevate, twist, and evaluate programs made elsewhere to advance their interests. Theory attending to the transnational and transscalar mobility of policy ideas is a productive frame for confronting these concerns.
Fast policy
The most influential elaboration of “fast policy” appears in the work of Jamie Peck and Nik Theodore (2012, 2015). They view “fast policy” as summarizing a set of conditions of contemporary policy-making, including:
A partly shared consciousness between actors in policy-making networks across different territories, leading to the transnationalization of policy discourses and debates and a flow of ideas, norms, practices, fixes, and stories between cosmopolitan actors. Deference to technocratic “best practices” and shortening of research and development phases, alongside an ethos of experimentality and innovation, meaning that ideas are rapidly adopted and iterated upon, or “sped up.” An expanded, normalized global infrastructure of policy development, including frequent conferences and field trips, social media, and various genres of policy document, as well as the constructed fora of international institutions (Peck and Theodore, 2015: 224–225).
“Fast policy” emerged from and propelled conversation in policy mobilities scholarship (see, e.g. McCann and Ward, 2013). These scholars conceptualize policies as powerful technologies, undergirded by and productive of ideas and knowledge, whose movement is embedded in power relations, ideologies, and political conflict, and which are always incomplete and in contention (McCann and Ward, 2013). This literature is typically informed by empirical work using interview-based, multi-sited methodologies, named “follow the policy” by Peck and Theodore for their emphasis on tracing the movement of policies across territories (2012: 21).
At its core, “fast policy” is about the construction of normalized transnational and transscalar network infrastructures within which policy ideas and models flow and are consolidated. Tangibly, this means sites like workshops, conferences, communication via social media, field trips, and performance awards. Peck and Theodore (2015: 12) highlight that policy learning at these sites is “normatively prefiltered,” as networked actors deliberately use policy examples from elsewhere to develop their own capacities and advance their agendas. Importantly, this means that actors who “pull in” policies are, to varying degrees, accustomed to doing so. Policy models are produced with a networked audience in mind, so the ideas and discourses that are accepted by this audience become powerful and mobile, or “fast” (Peck and Theodore, 2015). Networks therefore possess sedimented political interests and ideological content as actors develop relationships, consensus, and shared interests as they iteratively “pull in” policies that suit their interests. 4
This means that speed-up is uneven, not total. Those ideas that synergize with sedimented politics in policy networks are quickly “pulled in” to new jurisdictions, but those that challenge or collide with them are subject to lag, slowdown, or stoppage (Clarke et al., 2015: 28; see also Robinson, 2015). Ideas that cut against the grain—that challenge the wrong actors, deploy the wrong technology, or originate in the wrong locale—must be fought for. At the same time, policies that synergize with sedimented politics but are slowed by, for example, long research phases or difficult implementation, must be sped up to fit with the iterative rounds of reform that characterize fast policy worlds. Fast policy therefore tends to facilitate the movement of some ideas and not others.
For social policy, “fast” ideas are overwhelmingly neoliberal: they seek to restructure social relations as market relations through variegated political projects (Peck and Theodore, 2019). Neoliberalization is composed from adaptive policy interventions that empower market actors through diverse institutional configurations (Peck and Theodore, 2019). Fast policy worlds propel that adaptive tendency. The dynamic of “failing forward”—attempting to fix problems of neoliberal social reproduction with further market-oriented reform (Peck and Theodore, 2019: 258)—is locked into policy networks within conditions of fast policy. The application of this to SIBs is obvious: They are an attempt to create financial markets to fund policy “fixes” for the social problems of neoliberalism.
Marketization, best-practice technocracy, financialization, experimentalism, austerity, and other hallmarks of the neoliberal project are therefore embedded in fast policy conditions. Peck and Theodore’s (2015) empirical work examining participatory budgeting and conditional cash transfers supports this assertion. Participatory budgeting, originally a social democratic reform, had its radical aspirations “leeched out,” becoming a best-practice model of budget decentralization (Peck and Theodore, 2015: 221), and while the authors are enthusiastic about the opportunity for counterhegemonic practice that conditional cash transfers brought in Brazil, they acknowledge that it is much weaker than the dominant neoliberal version.
This is not to say that mobile policies are fundamentally neoliberal. The interactions between policies and the political, institutional, cultural, and material specificities that influence reinvention and enactment at the local and regional scales are too diverse for a policy monoculture to eventuate (Peck and Theodore, 2015). Indeed, rather than arriving in new places fully formed, actors invent, or pull in, elements of policy models from elsewhere to create local instantiations, meaning that policies “mutate” as they move (Robinson, 2015: 832). The unevenness of fast policy worlds means that neoliberal elements are easier to “pull in” than alternatives, not that neoliberal policies made at the transnational level are replicated in the national. By examining what moves, what does not, and what justifications are presented for those decisions, we can theorize how policies’ “harmonization” or “synergy” with powerful discourses in neoliberal policy networks is used to “pull in” policies to new jurisdictions. Though policy mobilities scholarship argues that mutation is intrinsic to policy movement (McCann and Ward, 2013), mutation is particularly visible in the case of SIBs, which rarely adhere to the precepts of their typical model (Arena et al., 2016).
Methodology
This study diverges from “follow the policy” by undertaking document analysis rather than interviews or ethnography. I analyzed approximately 80 publicly available documents relating to SIBs in Australia, created by actors operating at both federal and state levels. Documents from NSW, Victoria (VIC), South Australia (SA), Queensland (QLD), and Western Australia (WA) written between 2010 and 2019 were included. The genre of documents analyzed was diverse, consisting of policy documents, media releases and fact sheets, presentations, government discussion papers and their submissions, reports, evaluations, web pages, information memoranda, and media articles. The authors included government agencies, journalists, research organizations, individual advocates, financial intermediaries, and service providers. The target audience of the documents was therefore also diverse, encapsulating advocates highly embedded in SIB discourse as well as the public and first-time investors. I coded the documents thematically in NVivo, using themes highlighted in the literature on SIBs and “fast policy” as well as those that emerged inductively.
The purpose of employing document analysis is to illuminate the politics and sedimented logics that serve to justify SIBs for the actors deploying them and the audiences they speak to. As has been argued in various discourse analytic approaches to critical policy research (see, e.g. Bacchi, 2012; Jessop, 2010), examining discourse through policy documents provides a view into the process of problematization, or how actors are able to justify a model or idea as a “fix” to a (socially constructed) problem. The “Australian SIB experience” is conceptualized as a multifarious entity, in which the actions and ideas promulgated by government organizations and actors at the boundaries of the state are influenced by and mutually characterize one another. Identifying the discursive devices used to justify the adoption of SIBs reveals which problems they were positioned to fix, who by, and how that discursive labor was conducted, suggesting which interests, actors, and ideas were powerful and effective, which were contestable, and which were sedimented.
Contextualizing Australian SIBs
A visit to Sydney by Ronald Cohen in 2010 is credited with inspiring SIB enthusiasm in NSW (KPMG, 2014). Premier Kristina Keneally announced NSW’s interest in that year. Two programs were launched in 2013: Newpin Social Benefit Bond (Newpin) and The Benevolent Society Social Benefit Bond. Both interventions targeted children in or at risk of entering out-of-home care, with the goal of reducing money spent on out-of-home care by training parents in creating positive home environments. Both programs were labeled as successes early on. In 2013, other state governments began to advertise their interest in SIBs, with discussion papers produced for SA, WA, QLD, and VIC. All except WA launched pilot programs after 2015, targeting a range of issues including youth unemployment, homelessness, and out-of-home care. As of 2019, there had been 10–13 SIBs in Australia, depending on what is counted as a SIB: seven in NSW, three in QLD, two in VIC, and one in SA. 5 They have been launched by both Liberal and Labor governments and typically enjoy cross-party support.
Post-GFC conflict around austerity contextualized the emergence of Australian SIBs. In the immediate aftermath of the crisis, Prime Minister Kevin Rudd spoke against neoliberalism and enacted a massive stimulus package. However, by 2012, austerity was becoming entrenched, with Prime Minister Julia Gillard’s Labor government passing an austerity budget more as a strategic choice than a fiscal imperative (Whiteside, 2015), though mechanisms for raising revenue were prominent in Labor’s austerity compared to that of the Liberal budget enacted two years later. Also in 2012, in the hope of protecting its AAA credit rating, the newly elected Liberal–National coalition government of NSW passed the Fiscal Responsibility Act, enshrining a commitment to “responsible and sustainable spending” (S.7.2). This was partly a reflection of the institutional structure of Australian federalism, which is characterized by a high degree of “vertical fiscal imbalance,” meaning that the federal government is responsible for a vast majority of revenue collection, but the minority of service provision. This makes it more difficult for state governments to operate in periods of crisis. 6 An “austerity rationality” proceeded to dominate Australian politics at all scales throughout the 2010s.
Key actors
NSW government and the Office of Social Impact Investment
The enthusiasm of successive NSW state governments for SIBs propelled their adoption in NSW and the other states. In 2015, the Office of Social Impact Investment (OSII) was formed as a joint group between the Department of Premier and Cabinet and the Treasury. OSII was tasked with formulating and providing updates on NSW’s actions in the impact investing space, including procuring two new impact investments each year, and with “developing the capacity” of government agencies and service providers to produce impact investments, especially SIBs (NSW OSII, 2015). NSW also released the first policy statement on impact investing in Australia.
The Centre for Social Impact
The Centre for Social Impact (CSI) is a research and teaching institute spanning three business schools in Australian universities. It was founded under the leadership of influential public servant Professor Peter Shergold in 2008, and its establishment was supported monetarily by the federal government, several banks, and individual donors. The CSI’s purpose is described on its website as “to catalyze positive social change, to help enable others to achieve social impact … through transformational research and education” (CSI, 2019).
The CSI’s research and advocacy, especially through Shergold, was pivotal in the adoption of the flagship NSW SIBs (Mitropoulos and Bryan, 2015). Shergold is quoted substantially in the Premier’s initial announcement that NSW was interested in SIBs (NSW Government, 2010), and can be seen interviewing Ronald Cohen for television in 2010 (CSI, 2011a). Also in 2010, the CSI was commissioned to write the initial report analyzing the possibilities for SIBs in NSW, which made positive recommendations and suggested mutations to the typical model (CSI, 2011b). The CSI also organized the 2012 Social Finance Forum in Sydney, bringing together figures in the Australian social impact world with some influential UK figures (CSI, 2012).
The CSI’s early reports on SIBs were informed by interviews with investors, advocates, lawyers, not-for-profit organizations, and public servants. They were positive about the potential of SIBs and the private sector generally, but they did not harshly criticize government agencies. Notably, CSI’s enthusiasm for the model mellowed in the mid-2010s. A leading CSI academic publicly criticized SIBs in 2015, arguing that they lack structural protection against program failure for the communities they are designed to benefit (Barraket in Easton, 2015).
Social Ventures Australia
Social Ventures Australia (SVA) is a not-for-profit organization that “influences systems to deliver better social outcomes for people,” through consulting, impact investing, advocacy, and supporting social ventures (SVA, 2020). SVA acted as a coordinating intermediary on five SIBs across NSW, QLD, and SA, which is more than any other organization. It has also made submissions to government inquiries, run SIB workshops for staff in the public and not-for-profit sectors, and produced guides on designing SIBs. SVA consistently acted and advocated across state borders, affecting the mobility of SIBs through Australia. Because of SVA’s extensive engagement with SIBs and the resources it has devoted to them, it maintains an interest in their continued use.
“Pulling in” SIBs
This section brings the previous discussions together by analyzing the discourse around Australian SIBs and exploring its synergy with the neoliberal politics of fast policy worlds. I focus on the devices used to justify SIBs’ adoption and the practices, ideas, structures, and actors those devices foreground. It is organized around five themes: invoking elsewhere, austerity, experimentalism, mutation, and speed-up. The first three are aspects of the SIB rationale, or discursive devices, while the final two are aspects of the SIB technology, or practices. Due to the limited ability of document analysis to delve into the material realities of making, delivering, and receiving policy, my analysis is exploratory and suggestive.
Invoking elsewhere
Actors strategically used SIBs elsewhere to justify their enthusiasm for exploring the model in Australia. This was the case with all genres of text, and ranged from short lists of exemplary programs to multiple paragraphs analyzing singular examples.
Invoking elsewhere took multiple forms. One was “learning” from best practice. In the early stages, “best practice” was substituted for “only practice” through frequent references to the UK’s Peterborough SIB. In the 2011 CSI report on the viability of SIBs in NSW, the CSI was described as having “privileged access to the groundbreaking work” of actors involved in Peterborough (CSI, 2011b: 30). The CSI argued that “the UK SIB has identified and addressed a number of challenges and in doing so provided valuable learning” (CSI, 2011b: 8). Three learning areas were highlighted: the appropriate structure of SIBs; the process of creating economic and financial models; and developing audit and performance criteria. Rather than encouraging replication, the CSI sought to adapt the lessons of Peterborough to the NSW context. When looking at the appropriate structure for SIBs, it argued that the role adopted by financial intermediaries in the UK—issuing the bonds—may not be appropriate for NSW, as investors would prefer to contract directly with service providers (CSI, 2011b: 33). Peterborough became “ground zero”: a reference point embedded in transnational consciousness through which actors could establish their expertise and authority to influence SIBs’ construction.
The specific historical relationship between the UK and Australia in progressive neoliberal social policy restructuring is important context for Peterborough’s effectiveness as a “ground zero,” and constitutes a distinct aspect of Australian SIBs’ fast policy world. The roots of Cameron’s Big Society agenda were partly embedded in the previous Labour government’s “Social Investment Taskforce,” a Giddensian third-way project for devising methods for recasting government spending as investment in its population (Harvie and Ogman, 2019). Some scholars writing in the policy transfer tradition argue that ideas that became known as the third way in the UK and USA were first enacted in Australia in the 1980s and 1990s, and that close relations between the UK and Australian Labour/Labor parties were formed during this time (see, e.g. Pierson and Castles, 2002). This line of mobility was visible in the movement of the public–private partnership model—similar to SIBs in many ways—during the 1990s and 2000s, for which the UK and Australia are considered “first movers” (Whiteside, 2019). This historical relationship undergirds Australian actors’ construction of the Peterborough program as “good policy,” even in its infancy.
Programs elsewhere were also invoked to depict Australian SIBs as part of an emerging global phenomenon. For example, the 2013 WA report opens with the statement “SIBs are attracting worldwide attention as an innovative form of social service sector investment” (Charlton et al., 2013: 2). Similarly, in 2010, Premier Keneally’s first statement on her government’s interest in SIBs identified NSW as the “second jurisdiction to examine SIBs—after the UK,” positioning NSW on the forefront of a new policy innovation (NSW Government, 2010: 1). Potential investors were also targets of this legitimation strategy. The information memoranda for SIBs coordinated by SVA across three states each contained a section outlining their geographical reach, including case studies of prolific programs (see, e.g. SVA, 2013, 2017a, 2017b). Newpin’s memorandum mentioned the interest in SIBs in Canada, among other countries, at a time when all the Canadian government had done was call for impact investing concept proposals (SVA, 2013: 10). Whether programs were successful and who they benefited was irrelevant. They only need exist to be used to prove the existence of a shift at the global scale. Invoking elsewhere thus reframed the adoption of SIBs in a particular territory as part of a global discovery of “good” policy. Observers’ attention was removed from debates on the desirability and politics of SIBs and directed toward the place of Australia in a globalized world.
Austerity
Government savings are the basis from which investors in SIBs are paid returns. The economic case for SIBs in Australia and elsewhere is founded upon this idea, and it was repeatedly used to justify their adoption in discussion papers, principles statements, and information memoranda (see, e.g. CSI, 2011b; SA Government, 2013; SVA, 2013). SIBs thus harmonize with the neoliberal ideal of austerity, speeding their adoption into any jurisdiction whose government wishes to proclaim its expertise at balancing budgets. In post-GFC Australia, this was a major concern for state governments. The rationality of austerity is visible in KPMG’s evaluation of NSW’s first SIB development process, in which choosing policy areas seen as having a high cost to government was argued to be key to successfully developing programs (KPMG, 2014). Here, the logic of SIBs is more important than their reality. While their logic relies on their capacity to produce government savings, whether they tend to do this is questionable (Harvie and Ogman, 2019). Even if they did, at present their scale means that they do not constitute a large enough portion of social services budgets to have a real impact (Sinclair et al., 2019). For that to happen, they would need to be massively expanded in unexplored policy areas, which even some of their proponents acknowledge is undesirable (Easton, 2015). Using savings to justify SIBs is therefore a device that positions them favorably in relation to the austerity rationality with little purchase on their reality.
Despite Rudd’s stimulus package briefly disrupting the influence of anti-government intervention thought in Australia, early texts stressed the inability of government to respond to the social impacts of the crisis. This was a version of the austerity rationality that reduced the sphere of government intervention rather than simply demanding efficiency. For example, Ronald Cohen’s submission to a federal Senate inquiry argued “everywhere, government resources are failing to keep up with increasing social need” (Cohen, 2011: 15). NSW’s Social Impact Investment Policy (2015: 1) argued that “when it comes to tackling our most pressing social issues, we recognize that Government cannot solve them alone.” Similarly, a CSI report on the 2012 Social Finance Forum quoted the NSW Minister for Family and Community Services as saying “governments need non-government organizations to help them make difficult decisions in a world of limited resources” (CSI, 2012: 3). As in the UK, NSW needed to fund the surplus it committed to in the Fiscal Responsibility Act, or be perceived as doing so. SIBs were a vehicle that could contribute to this without creating the impression that services would suffer. However, these lines also refer to a reconfiguration of the institutions of government, in which the state’s incapability necessitates the transfer of power and responsibility to organizations outside it. This is reminiscent of classic neoliberal reason, and its invocation in support of SIBs in these examples demonstrates SIBs’ synergy with neoliberal politics.
Experimentalism
Experimentalism idealizes the construction of policy on the basis of “objective” evidence produced by quasi-scientific evaluation of interventions. Despite their own lack of evidence, SIBs’ synergy with experimentalism and evidence-based policy is well established (Berndt and Wirth, 2019).
As is the case elsewhere, Australian SIBs were described as enabling the development of new preventive interventions that tackle intractable “social challenges” in innovative ways. NSW’s Social Impact Investment Policy (2015: 1) explained that “better services and results” can be obtained using impact investment, as it provides “an opportunity to identify and test new and innovative ways to address social challenges, with a focus on measurement and delivery of outcomes.” Interventions are “tested” and “measured” rather than debated, and produce “results” rather than effects. The contradiction discussed in critical literature (see, e.g. Warner, 2013) between evidence-based interventions and innovative interventions was known to actors, especially in later discourse. It was viewed as a healthy way to establish high- and low-risk programs to appeal to a wider variety of investors and government actors, with riskier experiments bringing greater rewards, both for governments, through new evidence, and for investors, through outcomes payments (see Sainty, 2019; VIC Department of Treasury and Finance, 2018). Through this device, experimentalism is linked to finance, illuminating SIBs’ derivative logic (see Bryan and Rafferty, 2014).
In 2015, after it had gained some experience developing SIBs, SVA identified three preferable experimental mechanisms: comparison with a control group, preferably through randomized controlled trials; comparison with historical data for the selected population; and comparison with outcomes in a correlated population (Sainty, 2015). Such methods are thought to produce clear outcomes measurements and, therefore, clear conclusions on the degree of success enjoyed by programs. SIBs were thus represented by their proponents as “evidence production machines” for new policy areas. More nuanced than simply solving problems, SIBs created evidence that could define what a better intervention would look like. Positioning them as such legitimized SIBs because, even if they lacked evidence for their own success, they harmonized with the privileged experimental methods sedimented in fast policy networks. They offered the alluring prospect of expanding the domain of evidence-based policy into new areas.
Mutation
“Mutation” describes the alterations and twists made to the typical SIB idea in Australia. Examining the significant deviations in Australian programs is important; not because they represent a betrayal of or innovation upon the typical SIB model—which is not usually the version present in reality anyway (Arena et al., 2016)—but because they suggest how relations of power and jurisdictional contingencies are implicated in the “pulling in” of some SIB elements and the exclusion of others.
Australian SIBs, including the 2013 flagship programs in NSW, exhibit different structures of risk transfer and accountability to the typical model. Some of these differences can be traced to the CSI’s 2011 assessment of the viability of pilot programs for NSW. The CSI argued that the Peterborough risk structure would be unattractive to potential investors in Australia, as it exposed the full amount of investors’ principal to outcomes measurements (CSI, 2011b). Instead, it suggested a “shared risk” structure in which the program is partly paid for by a “standing charge” from the government agency, paid at the program’s outset and not subject to change by outcomes measurements (CSI, 2011b: 10). This structure protects an amount of investors’ principal proportional to the standing charge, as investors do not have to fund the entire program, limiting losses and gains. The CSI’s suggestion of standing charges was incorporated into most Australian SIBs. Also with the goal of protecting investors, the Benevolent Society Bond featured a two-tranche structure, with one tranche protecting investors’ entire principal from performance-based metrics at the cost of earning a lower interest rate. Australian SIB proponent Tomkinson (2017) links these changes to the contracting process, in which governments, intermediaries, and service providers must attempt to attract investment after bond structures are negotiated, meaning that a failure to attract investment will waste the time and money spent on the negotiation process. Further reflecting this trend away from exposing private capital, in 2018 VIC’s Labor government stopped describing SIBs as bonds, preferring the label “Partnerships Addressing Disadvantage” (PADS). Rather than risk transfer, PADS emphasize “risk sharing” between investors and government, on the rationale that investors should not have to bear all the risk of program failure (VIC Department of Treasury and Finance, 2018: 4).
These mutations destroy the argument that SIBs detach government from the risk of program failure. However, the mutability of the model has contributed to its Australian adoption, as its mutations have enabled the structural empowerment of investors to the point where it is profitable for them to take part. The negotiation of standing charges provides investors—typically represented by intermediaries—a powerful avenue through which they can alter the distribution of risk between actors in a SIB. It gives them more control over how profit is made and distributed than they would otherwise have.
This suggests that the CSI and SVA had significant influence over the development of Australian SIBs, and that they have been successful in communicating the demands of investors considering contributing capital to SIB programs. However, comparing the Australian experience with Williams’ (2018) analysis of Canada, the USA, and UK suggests that mutations may be better described as SIB proponents’ coping strategy than their strident agenda. The lack of a large, dedicated funding organization like Big Society Capital has meant that SIBs are underfunded at the national level in Canada and the USA compared with the UK (Williams, 2018). Despite a campaign by advocates, Australia also lacks such an organization. The concessions made by governments and service providers to attract private investment when negotiating SIBs reflect the necessity of private capital to these programs and the leverage that necessity grants. These perverse power dynamics—in which government needs private capital to make contracts worthwhile, but remains the initiator of SIB programs that offer little real savings—have led some scholars to identify a trend toward decommercialization in Australia, with public and third-sector actors becoming skeptical toward the SIB model and developing alternatives that do not privilege the demands of investors (Mollinger-Sahba et al., 2020).
Speed-up
A problem with framing SIBs as fast policy is that they are notoriously slow to negotiate and have high “transaction costs” (Dear et al., 2016; Easton, 2015). However, the Australian experience illustrates how they can be “sped up” by proponents through the use of normal practices of policy development in the political struggle over policy speed. After the flagship NSW SIBs took over 12 months to develop, their protracted development process became a major challenge for proponents (KPMG, 2014). This difficulty had been identified elsewhere (Warner, 2013), but was unavoidable, as staff at service providers and government agencies had only recently been introduced to the SIB model (KPMG, 2014). Slowness is resource-intensive for agencies involved and paints SIBs as complicated and difficult to implement, challenging the argument that they introduce businesslike efficiency to service providers and overcome sluggish government processes. This represented a struggle over SIBs’ timescale. Their proponents had to either make SIBs develop faster or make it seem that they did.
Partly, proponents approached this problem by stating that SIBs would get “faster” as more of them were (inevitably) designed (Easton, 2015). Simultaneously, SVA and OSII stressed the importance of “building the capacity” for SIB development in service providers and government agencies. This meant training staff through workshops and conferences in the specific skills, rationalities, and modes of valuation required to develop impact investments. In SA, SVA ran “Laying the Foundations” workshops prior to the government requesting proposals for SIB programs to prepare potential applicants (SVA, 2014). In NSW, one of the first steps after developing SIBs was to advertise the idea through conferences and OSII’s “Expert Advice Exchange” (see, e.g. Haigh, 2015). Both of these “business as usual” practices sped up the process of developing SIBs by resituating the time required to introduce them to staff from within the official SIB development process to outside of it. If the skills and knowledges necessary for developing SIBs are introduced before an official development process starts, that development process becomes shorter, creating the impression that SIBs are being developed faster. However, it is unclear whether these efforts were successful. In QLD, the Treasury team responsible for negotiating its three pilot SIBs was nominated for procurement industry awards for completing the process in 9 months, which is an example of both SIBs “speeding up” and the rewards that incentivize the fast creation of policy programs. But, at the time of writing, SIBs still exist only at the level of trials and pilots in most Australian states, and even in NSW, their rate of development does not approach that of the UK. The temporal effects of fast policy on SIBs are thus visible in the Australian experience, but in complex, nonlinear ways.
Conclusions
This article has sought to contribute to the literature on SIBs by explaining how and why they emerge in territories outside of the typical cases: the UK and USA. It has pursued this aim by exploring the connections between Australian experience and “fast policy,” utilizing the notion that conditions of fast policy tend to facilitate the spread of neoliberal social policies and slow down alternatives. I explored the synergy between powerful discourses and neoliberal politics embedded in policy networks and the discursive devices and practices deployed to justify the adoption of SIBs and “pull them in” to Australia. The analysis was organized around five themes: invoking elsewhere; austerity; experimentalism; mutation; and speed-up. I argued that SIBs were “pulled in” to Australia as a fast policy fix, because key actors were able to harmonize SIBs’ perceived benefits with powerful, sedimented discourses from transnational policy networks and local political economic and institutional conditions.
Bringing the ideas in this article together illuminates some key lessons from the Australian experience for SIB scholarship. First, the perceived benefits of SIBs synergize with policy discourses, ideologies, and ideas that are sedimented in transnational policy networks, especially experimentalism and austerity. The way they are justified reflects that. Rather than participate in debates on the merits and politics of SIBs, actors invoked programs elsewhere to demonstrate their relevance and desirability. SIBs were “pulled in” to Australia as convenient savings mechanisms, evidence machines, and a global trend rather than as a crisis response. Analyses that privilege their structural role in national political economies must contend with this.
Second, this analysis of the Australian experience suggests that intermediaries, research organizations, and the investors that they represent have exerted significant influence over the design and proliferation of SIBs. The influence of the CSI and SVA on the rationale and structure of SIBs was enormous, and acted across jurisdictions. Each tended to represent the interests of investors: the CSI through its early research reports, which underlined the importance of the needs of investors and privileged market mechanisms; and SVA through negotiating and coordinating SIB programs as an intermediary while laying the groundwork for their proliferation through multiple states. Both organizations used existing fast policy infrastructures like workshops and conferences to mobilize enthusiasm for SIBs, twisting them into new shapes deemed suitable for potential investors while still attractive to governments and service providers. This was also apparent in which actors and ideas were represented in SIB discourse. SIBs are clearly a project of reform in social service policy propelled by advocates within the state and on its boundaries. There is no suggestion that SIBs have the potential to fund “bottom-up” initiatives with redistributive aims.
Finally, responding to Williams’ (2018) call to interrogate the political process and conditions through which SIBs emerge involves engaging with the mobility of policy ideas, and therefore with fast policy. SIBs are a transnationally mobile policy model. Each instantiation exists relative to its others, and SIB proponents use that fact to justify SIBs’ adoption. Critical attention must be paid to this device, lest we risk accepting the mythos surrounding SIBs as their reality. Moreover, this analysis of the Australian experience demonstrates that conditions of fast policy play a role in producing the contexts that SIBs are pulled into, demonstrating the utility of “fast policy” as a conceptual tool. SIBs emerged in Australia, and perhaps elsewhere, partly because of their synergy with powerful transnational policy discourses and ideologies, which could be utilized by actors who had an interest in exploiting those synergies. In a broader sense, this point also connotes the importance of documentary analysis for the policy mobilities and “fast policy” literature. Stories about policy articulated in interviews exist for a different purpose than those in publicly available texts, and accessing those in official documents grants insight into the discourses, material conditions, and institutional configurations that permit the “pulling in” of mobile policies.
Focusing on how SIBs were represented in official discourse means that little attention was paid to the everyday life of policy-making in this study. This is not because it was not worth attending to, but because it is necessary to engage directly with the individuals who were involved in making SIBs to engage with the prosaic elements of their construction. I also focused on the adoption of SIBs over their results and effects. As more programs complete and publish their results and the profits made by investors, more outcomes-focused research will be paramount in articulating a critical response to SIBs. This article hopes to lay some groundwork for future research into Australian SIBs attending to both of those areas, as well as how SIBs have been “pulled in” to jurisdictions elsewhere. Extra attention is necessary on “development impact bonds” in the Global South, where the power dynamics of policy mobility can be starkly different.
Footnotes
Acknowledgments
The author is grateful to Dr. Kelly Gerard and Associate Professor Glenn Savage for their supervision, to Dr. Jordan Soukias Tchilingirian and two anonymous reviewers for their comments, and to the participants at the December 2019 Bringing Life’s Work to Market symposium at the University of Auckland for their feedback. All mistakes remain my own.
Declaration of conflicting interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research is supported by an Australian Government Research Training Program Scholarship.
