Social Impact Bonds: Reflecting on Emerging Global Practice

A UK-based non-profit pioneered the social impact bond (SIB), a tool used by some governments to fund preventive social programs, in 2010. Like other forms of social finance, SIBs have proliferated rapidly in recent years, reflecting significant enthusiasm amongst governments and the voluntary sector. While many introduced SIBs with fanfare, they have also drawn intense scepticism. This is particularly true in Canada, which is perhaps why we have been slow to try them. When SIBs first became a topic of policy debate, the discussion centred on how the SIB would reshape the policy terrain. Proponents presented it as a panacea to society’s ills. Critics decried it as a new phase in neoliberalism’s hollowing out of the welfare state. Myriad claims were made about how these tools would function, and about their effects for vulnerable citizens (Acuña, 2014; Joy and Shields, 2013; NUPGE, 2012; Mulgan et al., 2010; Whittington, 2012; Toonkel, 2013; Rosenberg, 2012; Sheffield, 2013). Moreover, because there was – at the time – so little practice to draw upon, these arguments were inevitably based on hypotheses rather than evidence.

While few SIBs have yet concluded, there is now sufficient global practice to reflect upon. As of September 2016, 26 countries have taken action to establish SIBs, and stakeholders have proposed 151 SIB projects. Currently Canada has implemented just one: a Saskatoon project to provide assisted living to at-risk single mothers (Robinson, 2014). At least four more are in the pipeline. The Government of Ontario is poised to move forward on two SIB pilot projects (Government of Ontario, 2016), while the Public Health Agency of Canada (PHAC) has announced a SIB on heart and stroke prevention (PHAC, 2016), and the Government of Saskatchewan is planning one on education for low-income youth (Cave, 2017). It is quite possible that this will spur further interest in the use of SIBs, especially for other provinces that had previously committed to creating a SIB, including Alberta, British Columbia, and Nova Scotia.

Should we support more SIBs in Canada? As the federal government prepares to develop its Social Innovation and Social Finance strategy, this is a critical moment to assess the value of SIBs as a policy tool. In this article, I seek to reconcile early debates with data on emerging SIB practice. I begin by providing a primer on SIBs and their global use, and then present data that I have collected through a combination of primary and secondary research. Drawing on this, I argue that SIBs do not deliver on many of their purported advantages. In particular, they are often not innovative and do not transfer financial risk to the private sector. However, one advantage of SIBs is that they can accommodate collaboration.

Background: SIBs and their global use

SIBs take a variety of legal forms, but will always comprise finance from investors to pay for the cost of an intervention intended to improve social outcomes. Stakeholders measure and evaluate these outcomes based on metrics agreed to in a payment-by-results contract. Investors receive a financial return only if the intervention attains the projected outcomes (So and Jagelewski, 2013). The main players in SIBs include governments, service providers, investors, the SIB commissioner, intermediary organizations, and, in some cases, an independent auditor.

Social Finance (2013) notes that stakeholders establish SIBs through a complex process. First, they must identify and understand the problem, program, and metrics. This step can be long and expensive for governments. Next, it states that they must develop a financial model to make the case that an SIB will generate value for money. From there, the SIB moves from the feasibility stage towards commissioning: program design, procurement, and contracting.

At least 151 SIBs are in various stages of development in 26 countries on all six populated continents. There are five categories of SIB adoption. Early adopters have implemented multiple SIB projects; middle adopters are planning multiple SIB projects; new adopters are planning or implementing their first SIB; some countries are Development Impact Bond (DIB) hosts, in which the outcomes payer is not the host government but rather a foreign aid agency; and non-adopting countries have not pursued SIBs. While the UK (89 SIBs) and US (19 SIBs) clearly represent the world’s most active early SIB adopters, a few other states have also been quick to try them out. The Netherlands, Australia, and Israel each have multiple SIBs that have reached the implementation phase. Beyond these, Finland, Canada, and Ireland are middle SIB adopters. These countries are planning (or at one point planned) multiple SIBs, and may have even launched them, but they are not yet being implemented. A further nine countries represent new SIB adopters – countries in the process of designing or implementing their first SIB. This includes Germany and Sweden, as well as emerging economies such as South Korea, Mexico, and South Africa. Most other countries, including France, China, and Italy, have not pursued SIBs. A final category of SIB practice worldwide comprises DIB hosts, whether public or philanthropic. Stakeholders in Peru have completed a DIB, while one in India is currently being implemented. Others are in the planning phases for Cameroon, Palestine, and Colombia.

While SIBs are far from ubiquitous, they are being used in a variety of contexts, with deep experience in a few jurisdictions in the US and UK. Practice is sufficient to merit studying how this tool is being used, as well as how that reality compares with initial speculation. In the following sections, I explain arguments for and against SIBs and suggest how persuasive they are in light of emerging practice, looking at four themes around which the debate about SIBs has coalesced: cost, the effectiveness of service provision, innovation, and the balance of service provision.

SIBs as a cost reduction strategy

Cost has been a major theme in debates about SIBs, which makes sense since cost reductions are built into their structure. Specifically, SIBs embed a “logic of cashable savings.” The idea is that governments pay for the cost of services out of the savings produced by the positive outcomes that are achieved through preventive social programs (SACOSS, 2013; Mulgan et al. 2010; Deloitte, n.d.). For instance, a program to prevent type 2 diabetes might generate “cashable savings” for government because prevention would divert the much higher cost of treating the disease over the lifetime of an average patient. Additionally, because SIBs use performance-based contracts, they arguably transfer the risk (and cost) of project failure to the private sector, saving the government money (Mulgan et al. 2010).

In theory, these arguments both seem plausible. However, in practice SIBs have often not transferred financial risk from public actors to the for-profit sector. Instead, in large part, philanthropic organizations – or even the public sector – have borne risk. For the 62 SIB projects for which stakeholders have identified investors, 77% involve less than full risk transfer from the public to the for-profit sector. This is either because the investors are philanthropic or government actors, or because private investment is fully or partially guaranteed.

Opponents argue that other costs outweigh the cost reductions from risk transfer and cashable savings from prevention. SIBs may be more expensive than traditional financing because they require high transaction costs; because investors will demand high rates of return; and because they create the risk of costly litigation if results are disputed (author’s interview; Acuña, 2014; Azemati et al. 2013; Malcolmson, 2014; Nicholls and Tomkinson, 2013; NUPGE, 2012; Preston, 2012).

Transaction costs are important, although it is difficult to evaluate their scope because few SIBs measure them. In instances where stakeholders measure transaction costs, usually this only includes legal fees and possibly the cost of buying out-of-house outcomes metrics (NFF, 2016). It does not encapsulate the true cost of developing a SIB in time and resources. This poses a barrier to assessing whether the transaction costs are “low enough” to make SIBs worthwhile. However, the nine SIB projects that have lapsed prior to implementation suggest that transaction costs can be a serious impediment for governments. For example, the Government of Ireland had initially set out to pilot two SIB projects, one on homelessness in Dublin and the other on youth justice. Planning for these projects began in October 2013, but the project development and procurement processes have proven to be arduous (author’s interviews), so much so that policymakers discontinued the proposed project on youth justice. Ireland is still planning to pilot the SIB on homelessness in Dublin (Beattie and Ottewell, 2016). Similarly, New Zealand’s first attempt at a SIB failed prior to attaining a contract. Negotiations with the service provider collapsed in July 2016, ending nearly three years of preparatory work (New Zealand Ministry of Health, 2016; Gudsell, 2016). In the UK, it has become common practice to provide grants to local governments for SIB project development, which would seem to be tacit acknowledgement that transaction costs can be prohibitive. For example, the Big Lottery Commissioning Better Outcomes fund has granted to upwards of 50 proto-SIB projects.

The duration of project development gives some sense of the time that staff members invest in putting these projects together. On this metric, there is considerable variation. Some SIB contracts have been established in just a few months, while others have taken three or more years. For example, The Netherlands launched its sixth SIB – its fourth on youth unemployment – after less than a year. Meanwhile, planning for the South Australian homelessness SIB began in 2014 and remains ongoing. Existing practice suggests that at least two characteristics influence project development costs: whether this is the first SIB in a particular jurisdiction (including sub-nationally), and whether the project addresses a new issue area.

Like transaction costs, the rates of return for SIBs vary considerably. Maximum rates of return have in some cases been as low as 6%, as with the Brussels immigrant integration SIB (Social Finance, 2016; EVPA, 2014) or as high as 22%, as with the New York City ABLE Project for Incarcerated Youth (NFF, 2016).

Finally, thus far there is no evidence to suggest that litigious financers will pose a major cost to governments. There are no instances of funders contesting outcomes in court. However, this might change as stakeholders complete more evaluations.

Effectiveness of service provision

A second theme concerns whether SIBs will make service provision more or less effective. Proponents suggest several reasons that SIBs can enhance the efficacy of social programming.

First, they argue that SIBs align incentives for government to provide stable long-term funding to preventive services (Robinson, 2012; Loxley, 2013). Because outcomes payments are directly tied to an impact that can be demonstrably shown to “save money” for the government, SIBs arguably make it politically viable for programs to receive stable funding for longer time horizons than is usually possible. This, some posit, will enable service providers to plan and scale interventions, thus making service provision more effective. In practice, the duration of SIB projects has varied considerably. The longest SIB project features a 12-year implementation period, while others are as short as one year. The average service delivery term is just less than four years.

A second reason advocates have noted that SIBs will make service provision more effective has to do with outcomes targets and evidence-based decision-making (author’s interviews; SACOSS, 2013). Traditional contracting often identifies outputs rather than outcomes. Proponents argue that outcomes measurement is a better way to contract for service provision. Linked to outcomes contracting, some argue that SIB interventions will be more evidence-based than traditional forms of service provision, as service providers and investors have an incentive to understand how likely their intervention will meet its given outcomes target (author’s interviews; Ragin and Palandijan, 2013). Relatedly, others note that SIBs provide an opportunity for better data, since governments and service providers may need to share information in order to measure outcomes targets (SACOSS, 2013). Of course, there is disagreement about whether outcomes measurement always improves the effectiveness of service provision. Opponents of SIBs argue that too much emphasis on measuring outcomes can divert resources and time away from actually undertaking the intervention (Clann Credo, 2011; Malcolmson, 2014).

Stakeholders can achieve outcomes measurement through other arrangements; it is not unique to SIBs. However, it is worth discussing whether SIBs improve available data to evaluate programming. There is something to this idea because SIBs comprise one way to institute collaborative data collection by governments and service providers. For instance, data collected through a coalition of four service providers on the “wellbeing star” (an outcomes measurement tool for long-term health conditions) was combined with government data on the cost of secondary care for the Newcastle long-term health conditions SIB.

Finally, those that oppose SIBs have argued that, because they are a form of contracting out, SIBs are less transparent and accountable than public programs (Nicholls and Tomkinson, 2013). In addition to undermining democracy, lack of transparency may make service provision less effective because citizens are less able to criticize SIB projects. There is evidence to support this conclusion. In some instances, very little information on SIB projects is publicly available, as in the case of Austria’s domestic violence SIB. In other cases, however, SIB projects have been quite transparent. For example, the New South Wales Office of Social Impact Investment (2016) publishes extensive information on the Benevolent Society and Newpin SIBs. It is incumbent upon governments to demand transparency when they are developing SIB projects.


The third theme asks whether SIBs will encourage or stifle innovation.  There are three reasons to believe that SIBs will encourage innovation. First, many argue that SIBs enhance innovation because they are amenable to collaboration (Deloitte, n.d.; Costa and Tokasko, 2014; Gill, 2013). Second, SIBs introduce market forces and investor scrutiny into social policy, which some posit will make charity more innovative (Preston, 2012; SACOSS, 2013; Ross Philipson Consulting Ltd, 2011). Third, SIBs transfer risk to the private sector (in theory), and it is argued that this will increase innovation because private sector actors are more willing to take on risks than government (Deloitte, n.d.; Pettus, 2013). Of course, we know from above that financial risk transfer has not yet been a prominent feature of SIBs. SIB opponents have argued, moreover, that transferring risk to the private sector will stifle innovation because investors are more risk averse than governments (Acuña, 2014; Loxley, 2013; Azemati et al., 2013). They also argue that SIBs will result in less innovation because only large service providers will have the capacity necessary to participate in this complex funding mechanism, while new, and smaller, organizations are often more innovative than established ones (Rudd et al., 2013).

In practice, SIBs have not overwhelmingly funded innovative projects. We can think about innovativeness by distinguishing four types of SIB projects: demonstrate (22%), transplant (11%), new collaboration (29%), and scale (38%). Demonstrate projects are those for which the intervention itself is new (including new combinations of interventions, as was often the case). The aim of these projects is to demonstrate the effectiveness of the intervention. We might consider these projects the most innovative because they involve a relatively untested intervention. For example, the Cornwall Aging in Place SIB is a £2 million ($3.2 million) SIB that aims to reduce loneliness amongst the elderly: a new intervention that involves a variety of activities (Social Finance, 2015, 2016; Jupp et al., 2013). Next, transplant projects have been tried before but either not by the service provider or not in the jurisdiction. An example of a transplant SIB is the Israeli type 2 diabetes SIB. It brings over a set of interventions that a team in Finland had previously undertaken (Social Finance, 2016; Press, 2016). Third, new collaborations entail multiple service providers cooperating to deliver interventions. For instance, a new consortium of charitable and public agencies delivers the Massachusetts homelessness SIB (Cox, 2012; Pettus, 2013). The fourth category, scale, involve cases in which the provider had previous experience with the intervention area. In some instances, this involved genuine scaling, though not always. For instance, as part of the UK Youth Engagement Fund, participants selected Prevista Ltd. as the service provider for a second SIB, undertaking the same intervention (youth applied positive psychology) in the same jurisdiction with a target community of very similar in size – 1000, compared to 800 (Social Finance, 2016). More than a third of SIB projects are from the least innovative “scale” category.

Balance of service provision

A fourth theme is whether SIBs will result in service provision that is balanced in a substantive sense – directed to the right problems and interventions, in the right proportion.

Proponents highlight how SIBs align political incentives to fund additional preventive social programs (Ross Philipson Consulting Ltd., 2011; Keohane, 2013). The logic here is that politicians act to maximize their chances of re-election, which means that they are unlikely to fund expensive projects that are expensive up-front but whose benefits will only become clear years or decades later. SIBs, some suggest, will remedy this negative political incentive. While it may be tempting to accept this dour assessment of political incentives, the data above suggests that SIBs seldom fund entirely new interventions. This indicates that it may suffice to make the case to the public for funding preventive social programs.

Critics emphasize several ways in which SIBs might distort public services and undermine the welfare state (Joy and Shields, 2013). The first of these is the logic of cashable savings. Because cost reduction is so central to the mechanism of a SIB, they argue that SIBs will further legitimize framing the validity of social programs in terms of whether they save the government money (or not). The unit cost database created in the UK is a manifestation of that framing (New Economy Manchester, n.d.), the monetization of social problems, which some argue is to be resisted (Warner, 2015). Another less persuasive variant of this argument suggests that SIBs will distract the public’s attention from ongoing austerity and thereby make further cuts politically palatable (NUPGE, 2012).

Similarly, because SIBs are a type of contracting for public services, some have viewed them as privatization, and opposed them on that ground (NUPGE, 2012). While SIBs do involve contracting out, the vast majority of SIB projects (86%) fund a voluntary sector service provider. Out of 78 SIB projects in which stakeholders had selected the service provider, only 10 use for-profit providers.

Next, others argue that SIBs will result in less equitable service provision because investors and service providers will have incentives to “cherry pick” or “cream skim” in order to meet outcomes targets (Joy and Shields, 2013). In practice, many SIBs have been designed specifically to avoid this problem. As with transparency, governments can avoid this problem through deft negotiation.

Only a narrow cadre of interventions is appropriate for SIBs, for reasons such as the need to identify measurable outcomes targets. If political incentives drive governments to use SIBs as a norm of welfare provision, it will disadvantage programs that are not SIB appropriate (Malcolmson, 2014). SIBs may also distort voluntary sector activities if governments prefer funding charities through SIBs.  These final two arguments are premised on SIBs becoming a widespread policy practice, which does not appear to be their trajectory. Even in the UK, where they are most prevalent, SIBs make up a very small proportion of service provision arrangements.


Six years of global SIB practice suggests a mixed track record. Some fears about the effect of SIBs on the balance of service provision appear to be overblown, especially as SIBs generally fund voluntary sector organizations, not private companies. Furthermore, there is scant evidence of “cherry picking” or litigation. On the other hand, SIBs do not live up to many of their purported benefits. SIBs are amenable to collaborative governance; however, it remains unclear whether they are net cost-reducing in light of rates of return, which remain highly variable, and transaction costs, which are often uncounted. Moreover, SIBs have often not funded innovative interventions and less than one quarter of SIB projects entail full financial risk transfer.


Acuña, R. (4 March 2014). Opinion: Social Impact Bonds: An Investment in the Wrong    Direction. Parkland Institute, at [8 October 2016].

Azemati, H., M. Belinsky, R. Gillette, J. Liebman, A. Sellman, and A. Wyse. (2013).         Social Impact Bonds: Lessons Learned So Far. Community Development   Investment Review, April 2013.

Beattie, Jason and Ottewell, David. (25 February 2016). Housing Crisis Sees Sharp Rise   in Number Sleeping Rough. Irish Mirror, at   [9 October           2016].

Cave, Joanne. (9 January 2017). Signals of Transformation: What Will 2017 Bring for       Canada’s Non-profit Sector? The Philanthropist, at   bring-for-canadas-non-profit-sector/ [20 January 2017].

Clann Credo. (2011). Report of the Roundtable to Explore the Potential for Social Impact   Bonds in Ireland. Clann Credo, at             BLE_WITH%20SLIDES_Final.pdf [8 October 2016].

Costa, K., and I. Tokasko. (2014). Networking for Success – Building Networks is            Essential to Investment in Social Impact Bonds. Center for American Progress and Council of Foundations, at      rking-for-success/ [8 October 2016].

Cox, B. (2012). Financing Homelessness Prevention Programs with Social Impact Bonds.             Review of Banking and Financial Law 31, at     .pdf [8 October 2016].

Dear, Annie, Helbitz, Alisa, Khare, Rashmi, Lotan, Ruth, Newman, Jane, Crosby Sims,     Gretchen, Zaroulis, Alexandra. (July 2016). Social Impact Bonds: the Early       Years. Social Finance, at content/uploads/2016/07/SIBs-Early-Years_Social-Finance_2016_Final3.pdf.

Deloitte. (n.d.) Paying for Outcomes – Solving Complex Societal Issues through Social       Impact Bonds. Deloitte, at          issues/ca-en-insights-issues-paying-for-outcomes.pdf [8 October 2016].

European Venture Philanthropy Association (EVPA). (29 April 2014). A Belgian Première: a Social Impact Bond to Tackle Unemployment in Brussels. EVPA, at      tackle-unemployment-in-brussels/ [9 October 2016].

Gill, I. (March 2013). Social Impact Bonds: Measure for Measure. Presentation prepared   for Cause + Effect: the Social Projects Studio.

Government of Ontario. (updated 21 June 2016). Piloting Social Impact Bonds in Ontario: the Development Path and Lessons Learned. Government of Ontario, at   path-and-lessons-learned#section-3 [13 August 2016].

Gudsell, Kate. (6 July 2016). Government’s First Social Bond Collapses. Radio NZ, at   ’s-first-social-bond-    collapses [9 October 2016].

Joy, M., and J. Shields. (2013). Social Impact Bonds: the Next Phase of Third Sector        Privatization? Canadian Journal of Nonprofit and Social Economy Research 4(2),             at          third-sector-marketisation.pdf [8 October 2016].

Jupp, B., T. Symons, and L. Fulton. (2013). A Social Impact Bond to Reduce Social         Isolation and Improve Long-term Condition Self-management. Social Finance, at     /Agendas%20and%20Reports%202013/Thursday%2C%2012%20December%202        013/Item%2014%20Background%20paper.pdf [8 October 2016].

Keohane, Georgia Levenson. (5 August 2013). Can Social Impact Bonds Unlock Private Money For Public Goods? Innovation in Pay for Success and Social Finance.   Roosevelt Institute policy note, at    f [8 October 2016].

Loxley, John. (January 2013). Social Impact Bonds. CCPA Review of Economic and         Social Trends.

Malcolmson, J.D. (2014). Social Impact Bonds Cleared for Landing in B.C. Canadian        Union of Public Employees, at            _SIBs_come_to_BC.pdf [8 October 2016].

Mulgan, G, N. Reeder, M. Aylott, and L. Bo’Sher. (2010). Social Impact Investment: the   Challenge and Opportunity of Social Impact Bonds. The Young Foundation.

National Union Public General Employees (NUPGE). (April 2012). Social Impact Bonds: A New Way to Privatize Public Services, at             [8 October 2016].

New South Wales (NSW) Office of Social Impact Investment. (updated 2016). Social      Benefit Bonds. NSW Government, at               newpin-bond [13 August 2016].

New Zealand Ministry of Health. (5 July 2016). Social Bonds – Progress to Date. Government of New Zealand Ministry of Health, at  new-zealand-pilot/social-bonds-progress-date [9 October 2016].

Nicholls, A., and E. Tomkinson. (2013). The Peterborough Pilot Social Impact Bond.        University of Oxford Said Business School, at            peterborough-pilot-social-impact-bond-oct-2013.pdf [8 October 2016].

Nonprofit Finance Fund (NFF). (April 2016). Pay for Success: the First Generation. A      Comparative Analysis of the First 10 Pay for Success Projects in the United        States. NFF, at     First%20Generation_0.pdf [8 October 2016].

Pettus, Ashley. (July-August 2013). Social Impact Bonds. Harvard Magazine, at    [8 October 2016].

Press, Sarah Viva. (28 March 2016). Israel Launches Social Impact Bond to Prevent          Diabetes. Israel 21C, at        bond-to-prevent-diabetes/ [13 August 2016].

Preston, C.  (8 November 2012). Getting Back More than a Warm Feeling. New York        Times,       through-social-impact-bonds.html?pagewanted=all&_r=2 [8 October 2016].

Public Health Agency of Canada (PHAC). (28 October 2016). Minister of Health Announces Social Impact Bond for Heart and Stroke Health. Government of   Canada news release, at [7            December 2016].

Ragin, Luther and Palandijian, Tracy. (1 April 2013). Social Impact Bonds: Using Impact             Investment to Expand Effective Social Programs. Community Development      Investment Review, at   impact-bonds-impact-investment-expand-effective-social-programs.pdf [20       January 2017].

Robinson, J. (2012) Better Public Services through Social Impact Bonds – Lessons from      the US, Canada, New Zealand, and Australia. Winston Churchill Fellowship Paper.

Robinson, Britainy. (12 May 2014). Saskatoon Home Uses New Formula to Fund ‘Sweet             Dreams’. CBC News, at          home-uses-new-formula-to-fund-sweet-dreams-1.2640444 [28 August 2016].

Rosenberg, Tina. (20 June 2012). The Promise of Social Impact Bonds. The New York      Times Opinion Pages, at       promise-of-social-impact-bonds/?_r=0 [20 January 2017].

Ross Philipson Consulting Ltd. (2011). The Potential for Social Impact Bonds in New        Zealand. New Zealand Department of Internal Affairs.

Rudd, T., Nicoletti, E., Misner, K., & Bonsu, J. (December 2013). Financing Promising     Evidence-Based Programs – Early Lessons from the New York City Social   Impact Bond. MDRC, at         evidence-based-programs [8 October 2016].

Sheffield, Carrie. (15 October 2013). Private Risk, Public Benefit: Are Social Impact        Bonds the Way of the Future? Forbes, at               benefit-are-social-impact-bonds-the-way-of-the-future/#695b2654554c [20            January 2017].

So, Ivy, and Jagelewski, Adam (2013). Social Impact Bond Technical Guide for Service   Providers. MaRS Centre for Impact Investing, at           content/uploads/2013/11/MAR-SIB6939__Social-Impact-Bond-Technical-Guide-    for-Service-Providers_FINAL-ELECTRONIC1.pdf  [29 September 2016]

Social Finance. (2013). A Technical Guide to Developing Social Impact Bonds. Social      Finance UK, at content/uploads/2014/05/Technical-Guide-to-Developing-Social-Impact-            Bonds1.pdf [8 October 2016].

Social Finance. (June 2015). Investing to Tackle Loneliness: a Discussion Paper. Social      Finance UK, at content/uploads/2015/06/Investing_to_Tackle_Loneliness.pdf [13 August 2016].

Social Finance. (2016). Impact Bond Global Database. Social Finance UK, at    [13 August 2016].

South Australian Council of Social Service (SACOSS). Social Impact Bonds: New            Financing for the NFP Sector? Information Paper, SACOSS, at      20SACOSS%20Information%20Paper%20on%20Social%20Impact%20Bonds.pd           f [8 October 2016].

Toonkel, Jessica. (12 November 2013). Analysis: Wall Street Sees Social Impact Bobds     as a Way to Do Good and Well. Reuters, at         goldman-social-impact-bond-analysis-idUSBRE9AB0ZD20131112 [20 January     2017].

Warner, Mildred. (2015). Profiting from Public Value? The Case of Social Impact Bonds.             In John M. Bryson, Barbara C. Crosby, and Laura Bloom (eds.). Creating Public          Value in Practice: Advancing the Common Good in a Multi-Sector, Shared-     Power, No-One-Wholly-in-Charge World. Boca Raton, FL: Taylor & Francis         Group, 145.

Whittington, Les. (8 November 2012). Feds Introduce Controversial ‘Social Impact          Bonds’ to Fund Social Services. Toronto Star, at               ocial_impact_bonds_to_fund_social_services.html [20 January 2017].

Kristen Pue is a doctoral candidate in Political Science at the University of Toronto and a social innovation research fellow with the Innovation Policy Lab. Her dissertation research is on the nonprofitization of the welfare state.

2 thoughts on “Social Impact Bonds: Reflecting on Emerging Global Practice

    Thank you for the assessment of SIBS to date. You however fail to raise the fact that it is a flawed model. Why is the return on investment only paid to the investor and not the deliver agent? If the move is to a “business model” the nonprofit needs a return too so they can have the funds to invest in the next innovation otherwise the SIB is not in the community’s best interest. In addition you did not examine if the service system was harmed at the conclusion of the SIB. Were the programs picked up by government, wound down or the SIB extended at the premium price? If they were wound down did the participants suffer a loss of service ( become homeless again) or did they displace others waiting for service (eg. take rent subsidies from existing programs.) Lastly if the SIB was wound down did the service provider get full cost recovery for wind down costs?

    Really good overview of SIB debates. Equally good questions by Lynn – who also makes the important point that SIBs represent a higher cost unless savings are realized, especially given that the funding hasn’t really shifted from public and philanthropic funding to the private sector.

    To me this is the big question about SIBs – we don’t know if they work. Your analysis is limited to describing SIB models because there is little information on their outcomes. It’s too early to say until we have enough reporting. Neither the Peterborough evaluation by RAND ( or the Energize/Teens & Toddlers reports ( are able to point to any savings actually generated for government. They can tell us whether investors were paid back or not, but not whether society is better off as a whole because we had a long term impact on the lives of cohort participants at a cheaper cost.

    The experience of the Teens and Toddles program after completion is especially interesting – schools begun using the intervention that was funded by the SIB (which is good if, as one would expect, it is cheaper for them to be doing it rather than using a SIB model), but the investors set up a second SIB (which is bad, because why do you need to transfer risk once an intervention has shown success?).

Post a comment

Your email address will not be published. Required fields are marked *