Start Date 2015-00-00
Notes Social Impact Bonds: An Update By John Loxley and Marina Puzyreva Department of Economics, University of Manitoba CANADIAN CENTRE FOR POLICY ALTERNATIVES CCPA MANITOBA JANUARY 2015 Social Impact Bonds: An Update isbn 978-1-77125-135-8 january 2015 This report is available free of charge from the CCPA website at www.policyalternatives.ca. Printed copies may be ordered through the Manitoba Office for a $10 fee. Help us continue to offer our publications free online. We make most of our publications available free on our website. Making a donation or taking out a membership will help us continue to provide people with access to our ideas and research free of charge. You can make a donation or become a member on-line at www.policyalternatives.ca. Or you can contact the Manitoba office at 927-3200 for more information. Suggested donation for this publication: $10 or what you can afford. Unit 205 – 765 Main St., Winnipeg, MB R2W 3N5 tel 204-927-3200 fax 204-927-3201 email ccpamb@policyalternatives.ca About the Authors: John Loxley is Professor of Economics at the University of Manitoba and the Principal Investigator for the Manitoba Research Alliance. John has published extensively on a variety of economic issues including international finance and development, P3s, and community economic development. Marina Puzyreva is a graduate student in economics at the University of Manitoba. She holds an MA in Economics with a specialization in financial management and a BA in Economics from the National Research University Higher School of Economics in Moscow. Her current research focuses on the payment for successful models of financing, with a particular interest in Social Impact Bonds. Acknowledgements We are pleased to acknowledge the generous financial support of the Social Sciences and Humanities Research Council of Canada through the Manitoba Research Alliance grant: Partnering for Change — Community-based solutions for Aboriginal and inner-city poverty. social impact bonds: an update iii Table of Contents 1 Growth and Sector of SIBs, 2010–2014 3 The Development of ‘Enabling Fields’ for SIBs 4 Government Support for SIBs in the UK, USA, and Canada 8 Performance to Date 9 Growing Skepticism about SIBs? 10 Bibliography iv canadian centre for policy alternatives — manitoba social impact bonds: an update 1 In an earlier paper (Loxley, ) it was argued that Social Impact Bonds (s) are a recent innovation in the financing of social services and are becoming more popular with governments, service agencies and private and charitable financing bodies. It is difficult to gather precise information on s but a scan of the literature and web sites suggests that there are now some Growth and Sector of SIBs, 2010–20141  s either being implemented or very close to implementation around the world. Two thirds of them are located in the UK, the rest in the , Australia, the Netherlands and Canada. +e average size of s in the , however, is much larger than that in the UK so that the amount of money involved in total is greater than that of the UK. table 1 2 Social Impact Bonds 2010–2014 Number Total Canada UK USA Australia Netherlands 2010 1 1 2011 0 0 0 0 0 2012 13 12 1 2013 6 1 2 2 1 2014 3 1 1 1 23 1 15 4 2 1 1 +is paper has benefitted greatly from comments by Shauna Mackinnon and Jesse Hajer. Financial support was provided by and by the Global Political Economy Research Fund of the Faculty of Arts, University of Manitoba. We also acknowledge the generous financial support of the Social Sciences and Humanities Research Council of Canada through the Manitoba Research Alliance grant: Partnering for Change – Community-based solutions for Aboriginal and inner-city poverty. 2 +e dates when s actually commence should be treated cautiously as sometimes it refers to services actually commencing (e.g. projects in Peterborough, London, New South Wales), sometimes the contract being signed (e.g. Essex), sometimes the transaction being completed (e.g. ) and sometimes merely the project being announced (e.g. New York Rikers Island, Massachusetts etc). +e apparent lack of activity in , therefore, should also be treated with caution. 2 canadian centre for policy alternatives — manitoba rates and improve early childhood education are the main area of focus in the UK, though similar initiatives exist in the  and Australia. Half of the existing  projects focus on increasing employment, especially for youth, but these account for only  of total funding, suggesting these types of initiatives are relatively less expensive or at a smaller scale than programs aimed at offenders or children. In terms of the sector distribution of s, the bulk measured by financing are to be found in the criminal justice field, aimed at reducing recidivism among ex-inmates. Four anti-recidivism projects account for  of total  financing, with most of this expenditure in the . Programs targeting children account for  of total financing. Projects seeking to reduce the number of children in care, increase adoption table 2 Social Impact Bonds 2010–2014 Amount in Million US Dollars Canada UK USA Australia Netherlands Total 2010 7.7 7.7 2011 0.0 0.0 2012 28.5 9.6 38.1 2013 3.3 20.5 15.9 0.9 40.7 2014 0.92 15.0 27.0 43.0 0.92 54.5 57.1 15.9 0.9 129.4 table 3 Social Impact Bonds 2010–2014 Sector No. Canada UK USA Australia Netherlands Total % Total Offenders 4 7.7 50.1 57.8 44.7% Homelessness 1 7.6 7.6 5.9% Children 6 0.92 23.3 7.0 6.5 37.8 29.2% Employment 11 15.9 0.9 16.8 13.0% Home Care 1 9.4 9.4 7.2% 23 0.9 54.5 57.1 15.9 0.9 129.4 100.0% social impact bonds: an update 3 ly days. Whiteside argues the penetration of the public sector by Ps can be understood as a specific form of commodification of public services by the private sector, a form of dispossession which benefits private capital at public expense. s can also be characterized as having these properties and such an enabling field is likely required prior to any large expansion of  activity globally. +ere is, indeed, evidence of widespread activity in the development of necessary background supports for s, including significant government subsidization directly through the guarantee of returns to private investors or indirectly through project grants, as outlined below. +ere are at least   proposals in the works (Burpee, ) and, despite the increase in  activity by value since , the rate of implementation seems quite modest. Also, so far, only one has started in Canada. How do we explain this? +e answer, we believe, lies in the need for prior development of background supports for these activities in terms of legislation, government policy, budgetary practices and institutions and lobbyists promoting and enabling s. Heather Whiteside () has analyzed the evolution of an ‘enabling field’ to explain the eventual rapid takeoff of public-private partnerships (Ps) in Canada, and especially in the health field, despite strong evidence of poor performance of Ps in their earThe Development of ‘Enabling Fields’ for SIBs 4 canadian centre for policy alternatives — manitoba measured (New Economy, ). If experience with Ps is any guide, this toolkit is likely to be widely adopted in other countries. USA +ere has recently been a great flurry of activity with regard to facilitating and encouraging s in the . Legislation to permit  pilot projects has been passed recently in Hawaii, New Jersey and Minnesota, while legislation is being considered in Maryland and Utah (Eames and Terranova, ). Pilot projects or feasibility studies are under way in California, Ohio, Oregon, Rhode Island and Washington DC. Formal Requests for Information for s have been issued by the City of Denver, , with interest in ‘early childhood education, at risk youth, supportive housing and homelessness prevention’ (ibid, p. ); by Illinois in youth justice and employment; by Michigan for criminal justice and human services and by South Carolina for ‘controlling costs UK s were pioneered in the UK, where their development has been supported significantly by government. +e U.K. government has established two separate funds totalling  Million to support s (U.K. Cabinet Office, a) and Prime Minister Cameron used his presidency of G meetings in  to profile s and the broader concept of social impact investing (U.K. Cabinet Office, b). A Centre for Social Impact Bonds has been established in the Cabinet Office and this has developed a tool-kit for those wishing to consider s (Cabinet Office, ). +is consists of a Template  Service Agreement Contract which would simplify and cheapen establishing contracts for s and a cost-benefit guide and model to help proponents assess their proposals. In addition, there has recently been added a centralized database of unit costs covering crime, education and skills, employment and economy, fire, health, housing and social services against which  proposals can be Government Support for SIBs in the UK, USA, and Canada 3 +e Minnesota proposals are somewhat different from other s in that the State would guarantee private funding but pay social service agencies only if predetermined outcome targets were met. Unlike other s, the risk here lies entirely with the delivery agency and not with the private funders. It remains to be seen how delivery agencies are likely to respond to being exposed to such risks. See Perry, . social impact bonds: an update 5 +is follows the creation of Finance for Good in Alberta, a lobby group designed specifically to ‘develop the necessary tools, networks, and processes to bring s to Canada’. According to them, ‘We are on the verge of a revolution in social service delivery’ with s allowing ‘private sector champions to make profitable investments to enhance our community and save our government money’ (Finance for Good, ). In March, , the Government of Ontario’s Ministry of Economic Development, Trade and Employment made a call for  ideas that would provide ’innovative, prevention-oriented solutions in housing, youth-at-risk or ‘improving employment opportunities for persons facing barriers’ (Ontario, March , ). s had earlier been promoted by the Drummond Report () and by the Social Enterprise Strategy for Ontario (). +e Ministry will support promising ideas through workshops and the provision of expert advice. New Brunswick also announced its interest in pursuing s in its  +rone Speech (Finance for Good, ). Interest in s has been expressed in BC with regard to Aboriginal out-of-home care; in Alberta regarding Homelessness and in Nova Scotia in the areas of literacy and early childhood education (Gill, ). Recently, Saskatchewan announced its first , a million project to provide assisted living to young single mothers at risk in Saskatoon, designed to reduce the number of children taken into care (Hill, ). Half the money will be advanced by the Conexus Credit Union in Regina, matched by a contribution from two private individuals in Saskatoon. +is is a five year project which will return the original investment and a  return to the funders if  children remain and improving the health of mothers and babies enrolled in South Carolina’s Medicaid Program’ (Eames and Terranova, , p. ). Formal Requests for Proposals have been issued by Connecticut (substance abusers affiliated with the state’s child welfare program); Massachusetts (chronic homelessness and juvenile justice.) and the State of New York (health care, child welfare, early childhood development and public service) (Eames and Terranova, , pp. –). In terms of institutional encouragement and facilitation of s in the US, the Harvard Kennedy School has established a Social Impact Bond Technical Assistance Lab, which assisted the Connecticut initiative. An important factor in launching two of the earlier US s was the presence of guarantees for the funders, with the Riker’s Island  in New York obtaining a  funding guarantee by Bloomberg Philanthropies and the New York State  (youth justice) a  guarantee from Rockefeller Foundation (Finance for Good, March, ). Similar guarantees were important in the two Australian s, though in these cases it was the government which provided a  and a  guarantee (ibid). Guarantees have not been used in the UK where, it is argued, the social financing landscape is more developed (Finance for Good, ). Canada In Canada, Alberta has recently introduced legislation which provides  billion from the Alberta Heritage Trust Fund to create the Social Innovation Endowment Account, designed partly to fund the promotion and development of social impact bonds in Alberta (Acuña, ). 4  carried this news on its web site without any critical comment, raising the question of whether or not  itself is becoming part of the ‘enabling field’ for s in Canada. In response, Mike Toye, the Executive Director stresses that  does not have a formal policy position on s and that its ‘ limited capacity is directed towards making sure our members and audience are at least aware of what’s going on and have the information available to make up their own minds’ (personal communication, May, ). 6 canadian centre for policy alternatives — manitoba and philanthropic support...to grow the  Innovation Fund and to examine the potential to achieve improved  outcomes, including cost savings, through new financial structures such as social-impact bonds’ (Bonneville, ). His most recent presentation, however, is less categorical and avoids explicitly mentioning s referring, instead, to social finance and to ‘investors’ without mentioning the specific form of such finance and investment (Gill, ). +e local head of the project, a highly respected member of the Winnipeg Aboriginal and social service community, has also made it clear in personal communication that there is no prior assumption that the Boldness Project will be pursuing s. +is apparent backing off from s may perhaps reflect the lack of enthusiasm for them by the Provincial government. +e McConnell Foundation appears to be taking a lead in enabling s, both by direct finance of projects and by funding an Impact Finance Lab run by the MaRS Centre for Impact Investing which will work with ‘governments, nonprofit organizations and investors to build ecosystem’s capacity to undertake outcomes finance in Canada’ (MaRS Centre, ). At the federal level, as noted in the earlier report, Human Resources and Skills Development Canada () had expressed an interest with their mothers for six months after leaving the Sweet Dreams assisted living home. +e return will be pro-rated if  to  children stay with their mothers and nothing, neither the original investment nor the return, will be paid if fewer than  children remain with their mothers. While there has been no official expression of interest in s in Manitoba, some interest in examining their adoption has been expressed by the Boldness Project on early childhood development in Winnipeg. +e Province has provided  of the funding for the  million Boldness Project, the balance of the funding coming from the McConnell Foundation. +e Province has also created an Early Childhood Development Innovation Fund, administered by the United Way and the Boldness Project is its first initiative. One of experts funded by McConnell, Ian Gill from Vancouver, is an active proponent of s claiming that, among other attractions, they ‘Free(s) service providers from government meddling’ (Gill, ). Gill initially stated quite categorically that s would be a component of the Boldness Project (ibid). He sees s as ‘a tool for communities themselves to co-design and co-deliver interventions (and) eventually, a way for communities to invest in their own outcomes’ (ibid). More recently he has stressed that the Boldness Project is seeking ‘ Further business table 4 Concept Area Frequency Multiple domains 45 Youth 21 Health 15 Aboriginal people 13 Housing/Homelessness 13 People with disabilities 11 Public safety 6 Unemployed 5 Seniors 4 New Canadians 4 Other 17 Total 154 social impact bonds: an update 7 income families for post-secondary education, group homes for seniors, and mentorship programs for immigrant professionals (, May , pp –) +e government promises follow-up action to connect ‘great ideas with social finance partners’, building ‘on the Call’s momentum by taking several steps to bring players together, incentivize leveraging, encourage new partnerships and stimulate innovative ideas for addressing social and economic challenges’ (ibid, p. ). It is apparent, therefore, that the federal government intends to be very active in promoting s, playing a key role in their development. +e criteria it lays down for supporting  projects are that ‘they must be proven, more beneficial than the existing program and both scalable and replicable’ (Ibid, p.) in s. In November, , it put out a call for innovative ideas in Social Financing, including s. It received over  concept papers, an unspecified number of which, but ‘many’, contained proposals for s. Proposals came from all over the country, with just over a half from BC and a further  from Ontario. +e vast majority of proposals, or  were submitted by non-profit or charitable organizations and  by the private sector. Proposals were received for many sectors of the economy, but mainly for youth, health, Aboriginal People, housing and homelessness and people with disabilities. On reporting on these proposals, the Minister specifically described proposals for s in the fields of supportive housing for people with mental disabilities, employment and training for Aboriginal youth, savings by low 8 canadian centre for policy alternatives — manitoba the number of reconvictions in Cohort  when compared to the Comparison Group, the reduction was insufficient with regard to the terms set out in the contract between the Ministry of Justice and the Social Impact Bond partnership to be considered an Outcome, thus did not trigger payment (Jolliffe and Hedderman, August, , p. ). In short, targets were not met. It should be noted, however, that if the average reconviction rate falls by at least . per cent across the first and second cohorts, then investors will get their money back in . Initially, there were supposed to be three cohorts, but the third one will not proceed as the experiment was cancelled by the Ministry of Justice with effect from June . +e  is replaced by a national program for short-term offenders called Transforming Rehabilitation. +e pilot will, therefore, last only three years and not the seven years originally planned (Pay For Success Learning Hub, ). As Ainsworth has argued, ‘the pricing structure for the first  was inevitably pretty finger-in-the-air, because no one quite knew what could be achieved. In order to tempt government into putting up the money, the first  made some stretching promises. And even then the Big Lottery Fund had to promise to put up much of the cash to pay for it. As far as I can see, future s have been priced in a way that’s more generous to the investor, and indeed, several of them have already paid out’. Only six s have been completed and there is information on outcomes for only two of them, but this does not deter proponents from waxing eloquent about the success of the approach. +us, Matthew Burpee,  of Finance for Good Ltd, concludes that the results from UK s targeting youth education and prisoner recidivism ‘are incredibly promising and appear to suggest that the  model is an effective way to drive positive social outcomes’ (Burpee, ). At the same time he acknowledges that ‘Obviously it is dangerous to imply correlation between a program targeting youth education and another targeting adult recidivism; similarly, one must be cautious not to conclude too much from such a small sample size’. No mention here of possible selection bias in these two projects, or of whether or not these results could have been achieved using improved conventional delivery methods. +is rah rah approach of Finance for Good is reminiscent of the uncritical approval of Ps by the Canadian Council on Public-Private Partnerships, but it is likely to become an important part of creating an  enabling environment. What makes this positive evaluation of the  approach doubly suspect is that it appears to be premature, at least in the case of Peterborough. A recent independent assessment of the Peterborough trial, widely considered to be the model , concluded that `despite a . reduction in Performance to Date social impact bonds: an update 9 quotes its president, Lori Sigurdson: “We don’t want people to profit from the misery of others. +e motive becomes profit, not service. +e primary responsibility of government is to support vulnerable and marginalized people.” Awareness of s is not as high elsewhere in Canada. Reactionary governments are likely to promote s to undermine the public sector and promote the privatization of public services (Joy and Shields, ), while social service agencies are still likely to be attracted to s because of the guaranteed funding they offer. +e movement to promote s has generated some negative reactions, especially in Alberta. David MacDonald () has argued that ‘+ere is no free lunch with the social impact bond model. Make no mistake: +e government always pays. Even if the project misses its targets, investors will be paid off so that they’ll pony up for next year’s bond. Otherwise, the house of cards collapses’. He points out that the Alberta College of Social Workers is opposed to s, because they “allow financial institutions to turn human suffering and conditions into commodities” and Growing Skepticism about SIBs? 10 canadian centre for policy alternatives — manitoba Acuña Ricardo, , Opinion: Social impact bonds: An investment in the wrong direction Edmonton Journal, March . Ainsworth, David, , Can the Peterborough social impact bond be judged a success? Civil Society, UK. http://www.civilsociety.co.uk/finance/ blogs/content//can_the_peterborough_ social_impact_bond_be_judged_a_success Bonneville, Ruth , , ‘Early investment, longterm payoff’, Winnipeg Free Press, March . Burpee, Mathew, , Initial Results are Extremely Positive for Social Impact Bonds, Finance for Good.http://financeforgood.ca/ initial-results-are-extremely-positive-for-social-impact-bonds/, September . Drummond, Donald, , Commission on the Reform of Ontario’s Public Services, Public Services for Ontarians: A Path to Sustainability and Excellence; Department of Finance, Ontario, http://www.fin.gov.on.ca/en/reformcommission/ Eames, Sandra and Victoria Terranova, et al. , A Review of Social Impact Bonds: Financing Social Service Programs through Public-Private Partnerships, February. Bibliography Finance for Good, .Social Impact Bonds and their Role in Canada. http://financeforgood.ca/ social-impact-bonds-and-their-role-in-canada/ Finance for Good, , Compensating for Risk in Innovation: +e Myths and Realities of  Funding Guarantees (Part ) http://financeforgood.ca/compensating-risk-innovationmyths-realities-sib-funding-guarantees-part-/ March . Gill, Ian, , Social Impact Bonds: Measure for measure, Cause and Effect, +e Social Projects Studio. Gill, Ian, , A Pathway to a Canadian Indigenous Impact Partnership in Early Childhood Development: Lessons (so far) from Winnipeg. Cause and Effect, March. Hill, Andrea, , ‘New funding model supports single moms’, +e StarPhoenix, Saskatoon, May . http://www.thestarphoenix. com/life/funding+model+supports+single+ moms//story.html Human Resources and Skills Development Canada (), , Harnessing the Power of Social Finance, http://www.esdc.gc.ca/eng/consultations/social_finance/report/index.shtml social impact bonds: an update 11 forsuccess.org/resources/peterborough-sibphase-out- Perry, S. (). Minnesota explores ‘Pay for Performance’ bonds. !e Chronicle of Philanthropy. Retrieved from http://philanthropy.com/ blogs/state-watch/minnesota-explores-payfor-performance-bonds/ UK Cabinet Office. (). Social impact bonds: Introduction to social impact bonds, information on commissioning a social impact bond, sources of funding and available support. Retrieved /, , from https://www.gov.uk/ social-impact-bondssources-of-funding-forsib-projects UK Cabinet Office. (). Social impact investment forum. Retrieved /, , from https://www.gov.uk/government/news/socialimpact-investment-forum UK Cabinet Office. (). Toolkit. http://data. gov.uk/sib_knowledge_box/toolkit Whiteside, Heather, , Stabilizing Privatization: Crisis, Enabling Fields, and Public-Private Partnerships in Canada, Alternate Routes, Vol. . http://www.alternateroutes.ca/index. php/ar/article/view/ Jolliffe, Darrick and Carol Hedderman, , `Peterborough Social Impact Bond: Final Report on Cohort  Analysis`, Prepared for Ministry of Justice, QinetiQ Ltd, August. https://www. gov.uk/government/uploads/system/uploads/ attachment_data/file//peterboroughsocial-impact-bond-report.pdf Joy, Meghan and John Shields, , ‘Social Impact Bonds: +e Next Phase of +ird Sector Marketization? : !e Canadian Journal of NonProfit and Social Economy Research, Vol. , No.. Loxley, John, ‘Social Impact Bonds’,  Review: Economic and Social Trends, January . MacDonald, David, , Social Impact Bonds; +ey aren’t charity. +ey aren’t social justice. +ey’re a new way for business to profit from social services. Albertaviews, https://albertaviews.ab.ca////social-impact-bonds MaRS Centre for Impact Investing, , http:// impactinvesting.marsdd.com/strategic-initiatives/outcomes-finance-incubator/ New Economy, , Unit Cost Database. http:// neweconomymanchester.com/stories/- unit_cost_database Pay For Success Learning Hub, , Peterborough  to Phase Out in , http://pay- Unit 205 – 765 Main St., Winnipeg, MB R2W 3N5 tel 204-927-3200 fax 204-927-3201 email ccpamb@policyalternatives.ca website www.policyalternatives.ca
Updated over 5 years ago