Everybody wins? A discourse analysis of competing stakeholder expectations in Social Impact Bonds

ABSTRACT

This paper explores how diverse stakeholders frame their expectations of Social Impact Bonds (SIBs). Using discourse analysis, the authors examine competing expectations in SIB press releases, showing how they differ between stakeholders, between institutional contexts, and how they evolve over time. The paper highlights how the prioritization of social finance and collaboration discourses privileges the role of private investors, which in turn diminishes the role of service providers as innovators.

IMPACT

This paper explains the competing rationales of stakeholders from different sectors and jurisdictions engaging in SIBs, and how these rationales shift over time. Policy-makers need to reprioritize the role of service providers in SIB communications as they are often marginalized in the overall public discourse—particularly in the early stages of SIB structuring. The differences the authors found between jurisdictions reveal that institutional contexts shape the nature of SIBs, and that the SIB model should not be transferred in a standardized way to differing contexts.

KEYWORDS:

• Discourse analysis
• marketization of public services
• outcomes-based contracts
• press releases
• Social Impact Bonds (SIBs)

Introduction

Social Impact Bonds (SIBs) aim to secure additional sources of private funding for social policy interventions and repay these via outcomes-based contracts that incentivize performance (Edmiston & Nicholls, 2018). SIBs are contractual agreements between government, investors and service providers, often supported by an intermediary, linked to the achievement of measurable social outcomes (Bengo & Calderini, 2016; Dey & Gibbon, 2018). Blending the agendas of different sectors, SIBs create a number of potentially complementary objectives (Arena, Bengo, Calderini, & Chiodo, 2016) that can lead to social and policy innovations (Liebman & Sellman, 2013). However, SIBs may also lead to competing objectives (Skelcher & Smith, 2017): a topic less commonly discussed. This paper explores the interplay between complementary and competing objectives by examining how SIBs are framed by different stakeholders across institutional contexts and how this framing shifts over time.

Like all multi-actor public policy interventions, SIBs are sites of power and contestation (McHugh, Sinclair, Roy, Huckfield, & Donaldson, 2013). Discourse analysis provides an appropriate lens to understand the contested nature of SIBs by focusing on how different actors frame their expectations of SIBs, paying attention to the role of texts and language in constructing the social world and reproducing power relations (Fairclough, 2013a; Wodak & Meyer, 2009). From a conceptual and theoretical perspective, SIBs are a useful case for unpacking how language shapes power relations in cross-sectoral policy settings (Prior, Hughes, & Peckham, 2012). Discourse analysis encourages a focus on the performative role of texts and language within organizations and institutions, as it is through ‘discursive interaction that meanings are produced and transmitted, that institutional roles are contradicted and power relations developed and maintained’ (Wodak, 2000, p. 185). We draw on Weick, Sutcliffe, and Obstfeld (2005) by focusing on the ways that language and text create policy innovations, accepting that ‘situations, organizations and environments are talked into existence’ (p. 409). This performative power of discourse remains under-explored (Balogun, Jacobs, Jarzabkowski, Mantere, & Vaara, 2014).

Our focus in this paper is on the language used to describe SIBs in press releases. Within discourse analysis, press releases are considered an important element of organizational storytelling: they help explain decision-making activities and play a role in legitimation (Vaara & Tienari, 2011). Press releases are viewed as indicative of official discourse, and are used by policy-makers as a mechanism for policy implementation, particularly where policy innovations are concerned (Marston, 2000). Press releases play a performative role in constructing and communicating organizational and institutional strategies and expectations, legitimizing acceptable action, as well as reproducing and maintaining dominant discourses (Balogun et al., 2014; Marston, 2000). Press releases thereby shed light on the power effects of discourse and show how texts reproduce dominant discourses (Hardy & Thomas, 2014).

To understand how different stakeholders frame their expectations of SIBs, we analysed 183 press releases related to 29 SIBs across three English-speaking jurisdictions: Australia, the United Kingdom (UK) and the United States of America (USA). Our analysis anchors these texts in three contextual dimensions: temporal, relational, and spatial. Our findings reveal three dominant discourses: a social finance discourse, a collaboration discourse, and an impact discourse. We show the prioritization of private sector actors through the dominance of the social finance discourse, and the marginalization of service providers and the impact discourse, especially in the launch phase of SIBs. Finally, we highlight the difference between jurisdictions which shape the way in which stakeholders frame the role of the private sector in SIBs.

The discursive construction of SIBs

To understand the discursive construction of SIBs, it is necessary to consider them in the context of broader governance developments. SIBs have been variously associated with the major public management reform trends of the past three decades. The logic of contracting and performance measurement that underpins the model marks SIBs as an ‘extreme expansion of New Public Management [NPM]’ (Warner, 2013, p. 3). At the same time, the emphasis on social and public innovation, partnerships and cross-sectoral collaboration are seen as reflecting the plural character of contemporary multi-actor public management, and akin to New Public Governance (NPG) (Fox & Albertson, 2011).

SIBs have polarized since the model was first piloted at Peterborough, UK in 2010. SIBs have garnered bipartisan support from policy-makers and practitioners (Arena et al., 2016; Dey & Gibbon, 2018; Edmiston & Nicholls, 2018; Maier & Meyer, 2017), who have emphasised the opportunities presented by SIBs from increased access to funds for service innovation to increased access to investment capital (Maier, Barbetta, & Godina, 2017). SIBs have also been critiqued as an attempt by policy-makers to reshape the non-profit and voluntary sector along social investment lines (McHugh et al., 2013; Roy, McHugh, & Sinclair, 2017), or viewed as a retrograde push to reshape the welfare state (Cooper, Graham, & Himick, 2016; Dowling, 2017; Joy & Shields, 2018; Warner, 2013).

In a review of the SIB literature in developed countries, Fraser, Tan, Lagarde, and Mays (2018) identified three dominant narratives: a public sector reform narrative, a financial sector reform narrative, and a cautionary narrative. The public reform narrative presents SIBs as an opportunity to improve the efficacy of the public and non-profit sectors through bringing in new resources from the private sector to generate fiscal savings by focussing on prevention (Arena et al., 2016). Through this lens, SIBs represent a classic extension of NPM (Osborne & Gaebler, 1992) that will lead to greater accountability through applying corporatist management logics to the public sector (Fraser et al., 2018; Warner, 2013). Related to this first narrative is the market-oriented financial sector reform narrative, which focuses on bringing in financial sector expertise (Edmiston & Nicholls, 2018), growing the social finance market (Mason & Moran, 2018), and improving the financial sectors’ ‘public image’ (Fraser et al., 2018, p. 11). It highlights how SIBs appeal to NPG (Osborne, 2007) by bringing actors into governance regimes that have not historically been associated with social policy development or interventions (for example the finance sector) (McHugh et al., 2013; Warner, 2013).

A third narrative is defined as cautionary and highlights the risk associated with the marketization of public services (Fraser et al., 2018). Critiques here range from instrumental concerns around transparency and accountability (McHugh et al., 2013) to concerns about the financialization of social policy (Cooper et al., 2016; Dowling, 2017; Warner, 2013). Other challenges identified with the cautionary approach include relative weaknesses in the evidence-base, isolating outcomes from existing parallel programmes and interventions, the challenges of scaling-up, and high transaction costs of contract development (Fox & Albertson, 2011; Lowe & Wilson, 2015).

In order to build on the academic literature, we explored the contradictory and competing discourses and narratives that surround SIBs in public communications. Specifically, we examined how different stakeholders frame SIBs through conducting a discourse analysis of their press releases. In novel areas of public management, such as SIBs, practitioners from diverse organizations deploy press releases to explain, justify and promote their involvement, and legitimize policy implementation (Marston, 2000). While recognizing that press releases offer an imperfect and incomplete representation, and are by definition a curated narrative (Marston, 2000), we believe that they play a performative role (Balogun et al., 2014) and are an important artefact for exploring the perspectives of practitioners from diverse sectors. We used these ‘artefacts’ to critically explore how stakeholder perspectives shift over the lifecycle of a SIB (from launch to implementation to evaluation to payment) and how they differ across institutional contexts (within Anglo-American countries). We did this by:

• Isolating the publicly-expressed views of the key actors in SIBs: investors, governments, intermediaries and service providers.

• Analysing stakeholder objectives and expectations over time to explore how the policy discourse has evolved around the same SIB, SIBs in general, and within an individual country from a temporal perspective.

• Finally, we took a comparative perspective to compare and contrast stakeholder views in three different jurisdictions to understand how discourses diverged and converged.

Methods

To understand how SIBs are framed by diverse actors, we conducted a discourse analysis of the media releases related SIBs in Australia, the UK and the USA. Through a discourse lens, we focused on the role of language in framing SIBs, and how the language used by different stakeholders reproduces power dynamics (Fairclough, 2013a). By focusing on language and power relations we aimed to understand the relative dominance and marginalization of particular discourses and stakeholders within SIBs, and how this reflects competing expectations (Fairclough, 2013b; Phillips & Hardy, 2002). Our objective was three-fold:

• Uncover whether certain stakeholders hold more power or legitimacy by unpacking the discursive framing of actors.

• Understand how these discourses compare and contrast across jurisdictions.

• Analyse how discourses change over time.

We selected press releases as the unit of analysis for a number reasons. First, media releases offered an accessible dataset for isolating the perspectives of each of the main actors in SIBs: public agencies, investors, intermediaries and service providers. Second, we could then broadly, albeit critically, compare and contrast how these actors constructed their narrative around their participation. Third, we could discursively unpack how they framed other actors as a proxy for positioning. Finally, the staggered issuance of media releases over time—from structuring and deal development through to implementation and maturity—enabled exploration of releases temporally and spatially (across countries).

SIBs were selected on similarity rather than difference to explore convergence and divergence across different policy issues in relatively like institutional settings: developed, liberal states that have been early adopters of the model. In appreciation of the importance of language in discourse analysis, we selected countries with English as the dominant language. In addition, the countries are typified as ‘liberal regimes’ (Epsing-Anderson, 1990) in comparative public policy, with a history of cross-sectoral partnerships in public service provision and a preference for markets in public management reform. We selected two policy issue areas—housing/homelessness and child/family welfare—to ensure that differences in discourse were not purely a product of different policy domains. We collected all available media releases from the stakeholders in each SIB, including governments, investors, intermediaries, and service providers. In total, we analysed the press releases of 29 SIBs, with 183 press releases obtained from stakeholders. While this approach to case selection resulted in a dominance of cases from the UK, this bias towards the UK reflects practice where 47 SIBs have been launched in the UK compared to 30 in Australia and the USA. Table 1 shows the distribution of releases across the SIBs and different stakeholders from 2011 to 2018.

Table 1. Press releases by country, issue and actor.

Country	SIB focus		Investor	Government	Service provider	Inter-mediary	Total
	Child/family	Homeless
Australia	3	3	10	16	10	6	42
UK	6	11	31	34	19	10	94
USA	2	4	22	10	8	7	47
Totals (29 SIBs)	11	18	63	60	37	23	183

Data analysis focused on critically examining how discourses in the press releases revealed the expectations and objectives of SIBs, as well as the power dynamics between stakeholders. Building on Gioia, Corley, and Hamilton (2013), we inductively coded the press releases to uncover the underlying discourses present within each press releases. Then, through multiple rounds of analysis, we engaged in axial coding (Strauss & Corbin, 1998) to identify relationships between the underlying discourses and combine them to form three overarching discourses. Figure 1 shows how 10 underlying discourses permeate SIB discourse and how they combine to form three overarching discourses on the purpose of SIBs. The percentages show the proportion of press releases where these discourses dominated.

Figure 1. The dominant discourses.

Stage one of our data analysis involved coding the press releases linked to the eight earliest SIBs in the dataset, revealing 10 underlying discourses. Stage two involved coding press releases from all 29 bonds, to identify the underlying discourses in each release and which one of those underlying discourses was the most dominant. Stage three involved writing a narrative discourse journey for each bond which focused on: the dominant discourses for each press release; different discourses forward/emphasised by different stakeholders; how different actors framed the same discourses; shifts in the discourses used/emphasised by certain stakeholders over time; shifts in the discourses used overall over time. Stage four involved the underlying discourses and dominant discourses being coded in Microsoft Excel and transformed into pivot tables to show the relative emphasis of underlying and dominant discourses along relational, spatial and temporal dimensions (see Tables 3, 4 and 5). Stage five involved axial coding to combine the underlying discourses into three overarching discourses (see Figure 1). Table 2 provides exemplar quotations for each of the underlying discourses.

Table 2. Exemplar quotations of underlying discourses.

Discourse	Exemplar quotations
Social finance discourse
Innovative funding mechanism	‘This social benefit bond is an innovative financing solution … it is structured more like an actual bond, with capital protection measures in place to manage investor risk and rewards’ (Australia—Westpac—investor)
Build social finance sector	‘The launch of this SIB by Essex County Council is a major breakthrough and shows the potential for the development of the market in social investment’ (UK—Essex—investor)
Attracting additional/private capital	‘Now, with a new infusion of dollars and a new financing mechanism, the program will be scaled-up’ (USA—Connecticut—government)
Collaboration discourse
Cross-sector collaboration/partnering	‘ … public, private and voluntary sectors set to come together in UK first … it represents the broadest range of partners that have ever come together for a government-backed initiative of this type’ (UK—Manchester Housing—government)
Win–win–win Savings for government	‘ … the objective of providing cost savings to government … effective use of public funds’ (Australia —The Benevolent Society—government)
Flexible funds for service provider	‘Social impact bond (SIB) allows frontline staff to deliver only outcomes and gives them the freedom to tailor the approach to the precise requirements of the individual without encountering imposed and unhelpful output measures and contract requirements’ (UK—Manchester Mayday—service provider)
Financial returns for investors	‘[SIBS give investors] an opportunity to invest in their communities … with the ultimate goal of a receiving both a social and financial return’ (USA—Ohio—intermediary)
Impact discourse
Long-term/better outcomes for communities/people	‘This social impact bond is designed to provide better support to adolescents and their families on the edge of the care system’ (UK—Essex—intermediary)
Measurable benefits/ accountability /evidence/data	‘ … these service providers have an opportunity to have the impact of their services, validated by outside researchers in a randomized control trial. The study will also provide evidence of the effectiveness of these two interventions provided together’ (USA—Denver—investor)
Innovation in service provision (for example prevention, early intervention)	‘Massachusetts is on the cutting edge of efforts to end homelessness … away from overreliance on crisis health and shelters and toward more permanent solutions that provide the stability people need to end their homelessness’ (USA—Massachusetts—investor)

Table 3. Dominant discourses for different actors.

Discourses	Government	Intermediary	Investor	Service provider
Innovative funding mechanism	52%	18%	34%	24%
Build social finance sector	8%	0%	13%	3%
Attracting additional/private capital	0%	0%	5%	0%
Social finance discourse	60%	18%	52%	27%
Cross-sector collaboration/partnering	10%	27%	8%	21%
Win–win–win Savings for government	8%	9%	2%	5%
Flexible funds for service provider	0%	0%	0%	3%
Financial returns for investors	2%	18%	10%	0%
Collaboration discourse	20%	54%	20%	29%
Long-term/better outcomes for communities/people	17%	14%	16%	39%
Measurable benefits/accountability/evidence/data	2%	9%	7%	3%
Innovation in service provision (for example prevention, early intervention)	2%	5%	0%	3%
Impact discourse	21%	28%	23%	45%

Table 4. Dominant discourses during launch versus implementation.

Discourses	Launch	Implementation
Innovative funding mechanism	40%	21%
Build social finance sector	8%	5%
Attracting additional/private capital	2%	0%
Social finance discourse	50%	26%
Cross-sector collaboration/partnering	17%	0%
Win–win–win Savings for government	5%	8%
Flexible funds for service provider	1%	0%
Financial returns for investors	3%	15%
Collaboration discourse	26%	23%
Long-term/better outcomes for communities/people (solving a social problem)	18%	33%
Measurable benefits/accountability /evidence/data	3%	10%
Innovation in service provision (for example prevention, early intervention)	1%	3%
Impact discourse	22%	46%

Table 5. Dominant discourses in different jurisdictions.

Discourse	AU	UK	USA
Innovative funding mechanism	29%	37%	45%
Build social finance sector	0%	12%	6%
Attracting additional/private capital	0%	1%	4%
Social finance discourse	29%	50%	55%
Cross-sector collaboration/partnering	2%	22%	9%
Win–win–win Savings for government	5%	1%	15%
Flexible funds for service provider	0%	1%	0%
Financial returns for investors	20%	3%	0%
Collaboration discourse	27%	26%	24%
Long-term/better outcomes for communities/people	41%	17%	9%
Measurable benefits/accountability/evidence/data	0%	2%	11%
Innovation in service provision (for example prevention, early intervention)	2%	1%	2%
Impact discourse	43%	20%	22%

Findings

Our analysis suggests that the 10 underlying discourses underpin three overarching discourses that dominate the framing of SIBs: a social finance discourse, a collaboration discourse and an impact discourse. We begin by discussing these three overarching discourses through a relational lens (how discourses differ between actors), before we turn our attention to the temporal (discursive shifts over time) and the spatial (discursive framing between jurisdictions).

Relational—what are the overarching discourses? How are they framed by different actors?

While the overarching social finance, collaboration and impact discourse were forwarded by all of the main SIB stakeholders, when the releases were analysed relationally we see differences between the framing by different actors. Table 3 reveals the relative weight placed on the three overarching discourses by the different actors. Public sector actors focus their attention on selling SIBs as an innovative funding mechanism targeted at catalysing social finance markets. Investors similarly highlight the function of SIBs as instruments of social finance and building this market. Intermediaries also emphasise the importance of market building, but focus more on the collaborative dimensions. In contrast, service providers emphasise the importance of impact above other discourses. In the section below, we unpack the three overarching discourses and highlight the key differences between dominant actors in more detail.

Social finance discourse

The social finance discourse dominated the framing of SIBs for nearly half of the press releases (46%). There were three underlying discourses in the social finance discourse—innovation, sector building and additional capital.

The innovative nature of SIBs as a novel funding mechanism is the most dominant of the 10 underlying discourses (36%). The innovation framing focuses on government and investors as pioneers in addressing social problems through directing additional private capital to reshape service delivery:

This is a great example of how this government is finding new ways to do work that has never been done before for a really hard to help group of people (UK—Greenwich—government, 2014).

While not the dominant element, the underlying discourse on sector building was often one of the key foci in the language of government, investors and intermediaries. SIBs are framed as part of the broader goal of ‘catalys[ing] the development of the social finance sector’ (Australia—benevolent society—government, 2011). The UK government appears focused on market building efforts:

 … we are working with partners in the sector to grow the social investment market, developing a range of new financing options to support organizations wanting to innovate and develop their business (UK—Manchester Action for Children—government, 2015).

The social finance discourse is further supported by the underlying discourse on unlocking private capital. This celebrates the role of investors in leveraging additional capital to address social problems.

Collaboration discourse

A collaboration discourse operates as a supporting discourse to the social finance discourse by focusing on how investors, government and service providers cooperate in SIBs in ways to benefit all parties. A common trope in the collaboration discourse is the framing of SIBs as a ‘win–win–win’:

There is no doubt that this bold initiative can be a win–win–win for NSW, resulting in better solutions for NSW citizens, less risk for the NSW government and greater involvement by the private sector in working with community organizations to better solve public challenges (Australia—The Benevolent Society—intermediary, 2011).

This win–win-win rhetoric is particularly common in the early framing where advocates seem focused selling SIBs as being aligned with all parties’ interests: ‘Together I know we can-and will-succeed’ (Santa Clara, service provider, 2014). The collaborative face of SIBs is strengthened through joint announcements that tend to sing the praises of other actors and how they are coming together for the first time:

 … the ground-breaking proposals have been made possible by strong partnership working between Greater Manchester Combined Authority, the 10 councils, public services, businesses and the third sector … It is a UK first, as it represents the broadest range of partners that have ever come together for a government-backed initiative of this type (UK—Manchester Housing—government, 2017).

The collaboration discourse celebrates partnership between sectors by encouraging ‘private investors to work collaboratively with the public and community’ (Australia—benevolent society—government, 2011). The collaboration discourse privileges the role of the private sector as the novel partner in SIBs. For example, the promise of additional capital brought by private investors is framed as the point of difference between SIBs and other cross-sectoral partnerships. In this way, the collaboration discourse supports the social finance discourse by emphasising the importance of government working with private investors that are non-traditional actors in public management partnerships. By reifying the role of investors, the community and service providers remain in the background. Service providers are often presented as supplicants in need of funding, or marginalized from this collaboration discourse completely. This reification of investors could be explained by the desire to grab attention in press releases by emphasising what is the most novel element of SIBs, i.e. the involvement of private investors that have not traditionally been involved in financing social policy (as opposed to infrastructure). Nonetheless, this reification of investors was prominent, particularly in the early stages of development of individual SIBs.

Impact discourse

The impact discourse was less dominant in the press releases. The impact discourse presents SIBs as supporting new ways of addressing social problems through prevention, emphasising the importance of long-term outcomes over outputs, and measurement and accountability of outcomes rather than process. The impact discourse offers a counter-narrative that shifts rhetoric away from innovations in funding towards innovation in service provision. When government and investors are driving the impact discourse, the narrative tends to focus on acknowledging the initiative of government and investors, with little mention of the role of service providers. Service providers, on the other hand, have a clear focus on the social impact achieved by SIBs (and the people receiving services), with many of their releases including the personal stories of service users (voices largely absent from the discourses forwarded by other stakeholders).

Temporal—how do discourses change over time?

To unpack discursive shifts in the framing of SIBs over time we compared releases during the launch phase with those released throughout implementation (i.e. after launch through to completion). Table 4 reveals a palpable shift in the emphasis of releases over time, from the social finance discourse to the impact discourse. In the section below we unpack these changes.

Overall, there was a shift in emphasis from the social finance discourse to the impact discourse through the lifecycle of SIBs. Service providers were central to the discourse. Discussion gradually shifted from emphasising the innovative character of the funding mechanism and sector building efforts towards long-term impact and measurable benefits. This shift provided room for the financial returns discourse to enter, with a link between outcomes and payments. Notably, the focus on innovative finance gradually waned, with the investor and government led discourses taking a back seat as the impact discourse emerged later, with service providers wrestling the discourse back from government and investors. As SIBs progress the work of service providers comes to the fore, allowing them to emerge as more central. It appears that government participation in SIBs is generally focused on the early stages of the SIBs as they play a central role in joint-development phase, yet seem to make way for the service providers as the SIB progresses.

The majority of the SIBs we analysed went on this journey from innovation to impact. The following narrative analysis of the benevolent society SIB in Australia helps unpack this shift from innovation to impact from 2011 to 2017. For context, this SIB was one of the first launched in Australia, and has been viewed as successful in reducing children entering out-of-home-care and thereby delivering cost savings to government and a return of up to 10.5% to investors.

The early releases in 2011 and 2012 were heavily focused on how this new finance mechanism could deliver cost savings for government, with innovation a constant rhetorical device.

Harness the innovation capacity of investors and service providers … a new financial instrument … the objective of providing cost savings to government (Australia—The Benevolent Society—government, 2011).

Around 2013, we see the first shift towards the impact discourse, as service providers enter the conversation, and bring a greater focus on long-term impact:

 … intensive family support service for up to 400 families over five years … what’s most important is turning lives around (Australia—The Benevolent Society—service provider, 2013).

After the first year and second year impact reports were released in 2014 and 2015, all parties shifted towards the impact discourse, with measureable outcomes driving the conversation:

 … safely preventing children in 75 families from entering foster care, which is a win for family unity will achieve its goals over its five-year term (Australia—The Benevolent Society—service provider, 2015).

The impact discourse serves to challenge the social finance discourse—as evidenced in this reflection:

 … an investor update where attendees didn’t hear about market shares or interest margins but about the progress made by some families who went through tough times caring for their children (Australia—The Benevolent Society—service provider, 2016).

By the fourth year of impact results (2017), the conversation shifted completely to long-term, measurable social impact, with financial returns presented as ‘a side point’:

 … achieved an 89 percent preservation rate … powerfully and positively impacted hundreds of families and changed the life trajectory of the children … capital gains are only a side point (Australia—The Benevolent Society—intermediary, 2017).

Spatial—how are SIBs framed in different jurisdictions?

Analysing releases at the country-level reveals subtle, yet perceptible, differences. For example, Table 5 reveals a much stronger emphasis on long-term impact and financial returns in Australia, whereas the innovative financing and social finance discourse is most prominent in the UK and the USA.

Australian releases are distinguished by their emphasis on financial returns, accompanied by a stronger focus on long-term outcomes. Australian actors, notably governments, investors and intermediaries, appear more comfortable promoting SIBs as a fruitful investment opportunity:

 … should appeal to a range of institutional investors who not only want to secure attractive returns … we are hopeful that our transaction will be met with similarly strong investor demand (Australia—The Benevolent Society—investor, 2013).

This may be because Australian press releases from the launch phase are actively promoting SIBs as an investment opportunity, with investors yet to be secured. Further, this openness to financial returns in Australia aligns with its pragmatic, neoliberal policy discourses, where partnership with the private sector has been embraced across the political spectrum.

This contrasts to the USA and the UK where press releases are more measured, and rhetoric softened regarding returns for investors. In these contexts, the discussion of financial returns is introduced late in the releases, and either defended, downplayed (noting that they are capped) or completely absent:

Funds will be repaid if project goals are reached, and a small return on the investments will be paid if the program exceeds the target outcomes (USA—Ohio—investor, 2014).

The discussion of financial returns is more about how investors will be ‘compensated for taking risk’, rather than ‘rewarded for making an investment’:

If the goals … are met, the government compensates the investors for undertaking the investment risk. If the goals are not met, the government is not obligated to repay the investors one dime (USA—Massachusetts—service provider, 2016).

Discussion

This paper has provided a discourse analysis of the framing of SIBs during their structuring and implementation, highlighting three overarching discourses—a social finance discourse, a collaboration discourse and an impact discourse—each of which can be associated with larger waves of public management reform. The dominance of the social finance and collaboration discourses, and the privileged role of investors, aligns with Fraser et al.’s (2018) conceptualization of the financial sector reform narrative. Building on earlier studies (Edmiston & Nicholls, 2018; Warner, 2013), we see the underpinning rationales of NPM in the social finance discourse (Osborne & Gaebler, 1992), with innovation, market building and win–win the dominant tropes. Further, we argue that tropes in the collaboration discourse are in turn most closely aligned with the shift to NPG (Osborne, 2007), widely heralded as reflecting a shift to multi-nodal service provision, networks and most notably cross-sectoral collaboration. Our analysis suggests that SIBs manifest as both NPM and NPG discourses, which challenges the literature on SIBs (for example Maier et al., 2017) which treats these as competing rather than complementary governance paradigms. This finding is in line with a growing body of public management literature that observes that the shift from NPM to NPG is not decisive or linear (Brandsen, Van de Donk, & Putters, 2005).

While our relational analysis of the active participants in SIBs (who by definition are supportive of SIBs as an instrument of public management) hides explicitly critical discourses, the dominance of the social finance discourse echoes critiques of SIBs as an approach to reshaping non-profit activities along social investment lines (McHugh et al., 2013; Roy et al., 2017). Thus, while the ‘cautionary narrative’ (Fraser et al., 2018) is not explicitly present within the press releases, our findings on the marginalization of service providers and the impact discourse supports the cautionary tale. Moreover, the strong currents around social finance in the overall framing are clearly suggestive of the financialization of the public management agenda that a growing body of explicitly critical literature sees as the underpinning rationale of SIBs and social investment more generally (Cooper et al., 2016; Dowling, 2017; Warner, 2013). In opposition to these cautionary narratives we see potentially defensive posturing by both governments and investors. For example, government actors position SIBs as an innovative funding mechanism to attract additional capital, rather than as a strategy of austerity and retrenchment of public services (as is commonly charged in the critical literature—see Cooper et al., 2016; Dowling, 2017). A related discourse promoted by investors is that they are contributing to building the social finance sector and making a positive social impact (Edmiston & Nicholls, 2018; Liebman & Sellman, 2013), which could be read as defensive posturing to counter criticisms of financialization that is evident in critical narratives (Dowling, 2017).

Our overall analysis of the temporal dimensions highlighted the uneven power dynamics that early conceptual studies including McHugh et al. (2013) have speculated would play out. During the launch phases of SIBs, the dominant framing centres on the social finance discourse, and the function of SIBs as a long-term tool of social change are relatively diminished. Service providers are marginalized particularly in the early stages of SIBs and sometimes omitted from releases, while investors are reified. We did, however, observe an interesting temporal shift, with the relative dominance of investors and government gradually waning as the SIB proceeds through delivery and implementation. We attribute this to two interrelated dynamics. First, service providers’ prominence naturally increases as they play a central part in implementation and as impact data becomes available. Second, the novelty of social finance and the presence of non-traditional actors wanes over time and governments begin to centralize the critical importance of service providers. Over time, this affects the overall narrative as impact becomes a central theme. Whether this is a permanent discursive shift is an interesting line of future research that potentially challenges the social finance discourse which is consistently seen as dominant (Cooper et al., 2016; Dey & Gibbon, 2018; Dowling, 2017; McHugh et al., 2013; Roy et al., 2017).

When attention is turned to the spatial dimension, some interesting dynamics are at play. Australia seems squarely focused on financial returns for investors, while the USA and the UK take a more defensive position on this. This finding speaks to the performative nature of SIB press releases, as in the Australian context many releases were focused on selling to investors, whereas in the UK and USA many SIBs had already secured funding by the time of press release and therefore releases were aimed at a broader audience. Further, this differing focus in Australia potentially speaks to the novelty of this practice in Australia and the associated hype, as compared to the relatively mature USA and UK markets where there is greater caution in how outcomes-based funding and social investment are promoted. This finding also aligns with comparative policy research on how policy actors in Australia and UK differ in their approach to public policy in the field of social investment: Australian policy-makers adopt a notably pragmatic posture, while UK policy-makers engage in deeper dialogue (Mason & Moran, 2018). What is most notable is that the USA is measured similarly in its framing of SIBs. These findings have practical relevance for understanding how SIBs operate in different institutional contexts, and what elements of SIBs may and may not translate contexts. In terms of future research, unpacking the comparative dimensions of SIBs is also a potentially fruitful line of inquiry to explore how they differ in both structuring and implementation across institutional contexts.

Conclusion

Our paper contributes to the growing body of research regarding the evolution of SIBs as a mechanism of funding and financing service delivery. By conducting a discourse analysis of stakeholder communications in SIBs across time and jurisdictions, we have shown how different actors construct narratives regarding their participation as they discursively frame SIBs in press releases. By doing so, we have built on the extant literature in two ways. First, we have added an empirical contribution to the predominately conceptual and normative literature (Dey & Gibbon, 2018; Maier & Meyer, 2017; McHugh et al., 2013; Roy et al., 2017). Second, we have built on prior studies exploring narrative such as Fraser et al. (2018), by empirically investigating how SIBs are framed by practitioners—from policy-makers through to service providers—and how this framing constructs and justifies implementation of public policy innovations (Marston, 2000). Through showing the marginalization of service providers, our paper reveals knowledge and power dynamics among stakeholders, something that Jackson (2013) suggests may be imbalanced, despite the relative novelty of the field.

Our paper is limited by our focus on press releases as the unit of analysis. While we make an argument for the performative nature of press releases, we note that they may not necessarily be a genuine reflection of organizational expectations and intentions. Press releases often present a biased perspective, and are often carefully crafted in ways to influence and shape official and public discourses. Further, press releases are a purposeful text that are often used as a tool to gain attention from the media. Finally, larger organizations, such as government and investors, are more likely to issue press releases than smaller service providers, which may contribute to the marginalization of service providers within the overarching discourses. To overcome these limitations, future research should focus on diverse texts to better understand public and private discourses, including meeting minutes, email communications and internal policy and tender documents.

Our findings provide important practical implications for policy-makers:

• Our findings suggest that SIB press releases overly prioritize the role of investors, particularly in the early stages, with the role of service providers being marginalized. We argue that policy-makers should reprioritize the role of service providers in SIBs to overcome their marginalization in the overall discourse.

• Our analysis highlights the way SIBs shift government participation to the front end of the contract, with much of the ‘innovation’ on the part of the government done during the launch phase. This contrasts to traditional government contracting where service providers are more heavily monitored by government.

• Our findings on the differences between jurisdictions reveals that institutional contexts shape the nature of SIBs and that the SIB model should not be transferred in a standardized way to differing contexts.

Disclosure statement

No potential conflict of interest was reported by the author(s).

ORCID

Jarrod Ormiston http://orcid.org/0000-0002-5295-0126