Social Impact Bonds Pay for Success Pros Cons

Social Impact Bonds / Pay for Success: Pros, Cons, Promises and Risks Kelly Walsh, Ph. D Justice Policy Center – Urban Institute October 7, 2014 1: 00 PM ET John K. Roman Justice Policy Center – Urban Institute Kyle Mc. Kay Texas Legislative Budget Board J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.

• Leads a national movement • State-based juvenile justice coalitions and organizations • Laws, policies and practices that are fair, equitable and developmentally appropriate for all children, youth and families J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. Photo: Moriza

Pay for Success: Funding the Infrastructure for Evidence-Based Change Kelly Walsh, Ph. D J U S T I C E P O L I C Y C E N T E R Washington, DC The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.

What is Pay for Success? • New financial instrument to support discretionary social programming o Brings evidence-based programs to the necessary scale • Can support programs that o Involve significant start-up costs, but provide long-term savings o Serve large populations o Involve political or programmatic risks • PFS addresses these programs with private capital o Offers an investment return if the program meets performance goals o Government only pays for successful outcomes • Sometimes called “social impact bonds” o PFS does not operate like a bond J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 4

How Does Pay for Success Work? • An intermediary selects an evidence-based program • The program is supported with private capital solicited by the intermediary • The intermediary o oversees program implementation and o ensures model fidelity • An independent evaluator determines if the program has met performance targets • Government pays the intermediary for successful outcome J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 5

The Structure of Pay for Success Intermediary J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 6

Advantages and Disadvantages Advantages Drawbacks • Transfers risk away from the government • Requires significant expertise from multiple fields • Develops or scales local service networks • Complicates execution • Could limit nonprofit innovation by focusing on programs with a proven track record • May reallocate existing impact and philanthropic capital rather than drawing new capital • Addresses the “wrong pockets” problem • Allows agencies to pool resources and knowledge • Provides significant flexibility to deploy programs across jurisdictions or regions J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 7

Building PFS in 5 Steps • PFS projects should target justice system “inefficiencies” (e. g. , prison population drivers) • Once drivers are identified, PFS is developed in 5 steps Price the product Strategic Planning 1 Develop infrastructure 2 3 Make the deal J U S T I C E P O L I C Y Evaluate the program 4 5 Deliver service and TTA C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 8

PFS in the US The US PFS market: almost $100 million in total investment Youth and the CJS: • $ 9. 6 million invested in New York City Rikers Island (2012) • Adolescent Behavioral Learning Experience (ABLE) Program • $27 million invested in Massachusetts’s Roca, Inc. (2014) J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 9

Pay for Success: The Promise of Scale Finance John K. Roman, Ph. D J U S T I C E P O L I C Y C E N T E R Washington, DC The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.

Strategic Planning Identify target population 1 2 Assess PFS Suitability Price the Product 4 1 3 Find evidencebased solutions J U S T I C E 2 Make the Deal P O L I C Y 3 4 5 Product Development Strategic Planning Identify population and cost drivers Evaluate Develop the Infrastructure Program Deliver Service and TTA C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 11

Strategic Planning Identify target population 1 2 Assess PFS Suitability Price the Product 4 1 3 Find evidencebased solutions J U S T I C E 2 Make the Deal P O L I C Y 3 4 5 Product Development Strategic Planning Identify population and cost drivers Evaluate Develop the Infrastructure Program Deliver Service and TTA C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 12

Strategic Planning • PFS relies on good strategic planning to maximize cost effectiveness • Before starting a program, governments should do the following: o o Identify population and cost drivers Identify the problems that generate these drivers Find evidence-based solutions to the problems Determine if the evidence-based solutions are PFS-compatible • Governments should identify a research partner to assist with strategic planning • Justice program databases can inform implementation: Office of Justice Programs’ Crime Solutions Washington State Institute for Public Policy (WSIPP) Vanderbilt University Peabody Research Institute Blueprints for Violence Prevention The Urban Institute’s Meta cost-benefit analyses for District of Columbia Crime Policy Institute (DCPI) o National Reentry Resource Center’s What Works in Reentry Clearinghouse o o o J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 13

PFS Opportunities in Juvenile Justice • Juvenile justice o Family-based intervention and prevention programs can prevent delinquent conduct and reduce recidivism § Implementing these programs requires start-up investment to train local staff • Adolescent diversion o Adolescent diversion from secure confinement has been found to have a large cost-benefit ratio § Training and project implementation impose significant upfront costs J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 14

Scale of the Problem J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 15

Strategic Planning – Likely Results From Steven H. Goldberg, Caffeinated Capital • Drivers of costs and populations: states and counties spend – $5. 7 billion every year taking legal and physical custody of more than 60, 000 juveniles adjudicated delinquent • Early-intervention programs are under-utilized: – Could reduce juvenile incarcerations by 70% – At a cost that is 80% less than confinement – Fewer than 5% of eligible families benefit from these programs. • Potential for enormous cost savings: – Family-Based therapies cost less than $15, 000 per family – Compared to an average out-of-home placement cost of $88, 000 per youth per year. J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 16

Florida Redirection as a National Model for Scaled Implementation Florida Redirection was managed by Evidence-Based Associates (EBA). From 2004 -2013: § Florida paid EBA a total of $65. 5 million to manage the delivery of three intensive family therapy programs: § Multisystemic Systemic Therapy® (MST), § Functional Family Therapy (FFT), and § Brief Strategic Family Therapy (BSFT) § More than 10, 000 families in 18 of the state’s 20 judicial districts we served. Source: J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 17

How Cost-Effective are MST Family. Based Therapies? Source: Washington State Institute for Public Policy J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 18

PFS Opportunities in Juvenile Justice Source: National Center for Justice Planning J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 19

PFS Opportunities in Juvenile Justice: from Caffeinated Capital Source: Social Finance, UK J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 20

PFS Opportunities in Juvenile Justice Source: National Center for Justice Planning J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 21

Future of PFS • PFS efficacy depends on maintaining a research emphasis • Broad adoption of PFS will be a learning process o Governments, for-profits, and nonprofits learn how to collaborate o PFS offers a new instrument for sharing resources and benefits J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. 22

Comments? Questions? Email: Jroman@Urban. org Twitter: @John. KRoman J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.

Social Impact Bonds: The cost of complexity Kyle Mc. Kay Presentation to National Juvenile Justice Network, October 7, 2014 All opinions expressed below are solely the author’s and should not be attributed to any of the individuals or organizations with which Mc. Kay is associated.

How complex are social impact bonds? The idea of paying for success sounds simple. However, even basic performance-based contracts without investors can become unwieldy. Social impact bonds (SIBs) are particularly complex. • There is “uncertainty arising from defining outcomes, estimating the potential impact of interventions, measuring and attributing change, valuing benefits, demonstrating a fiscal return and getting interventions to scale. …The weakness of the chain arises not from its weakest link but from the sum total of the weakness of every link. ” Fox and Albertson, Journal of Criminology and Criminal Justice

How complex are social impact bonds? “Complexity in some instances meant that the actual transfer of risk is not clear” – RAND Europe

How complex are social impact bonds? � In Peterborough, “key staff who deliver services to cohort members include: � 6 full-time caseworkers provided by St. Giles Trust � Fifty lay volunteers recruited, trained and managed by Sova. � Up to six St Giles Trust volunteers (individuals who may be currently on probation or have personal experience of the criminal justice system). � Two, part-time family specialist practitioners from Ormiston Children and Families Trust. � One part-time recovery worker from Mind. � One trainer from John Laing Training who delivers a construction skills course. ” Source: RAND Europe

Can SIB borrowing grow the pie? Commercial investors likely to seek secured source of income State law and fiscal policy can create non-appropriation risks Appropriations for payments were made in advance of any potential payments in: • Massachusetts • Utah • Peterborough In NYC, Bloomberg Philanthropies assumed the liability of the government by providing $7. 2 million to “be held by MDRC in a guarantee fund to back the loan”

Can SIB borrowing grow the pie? Legislative budget offices: • California Legislative Office analysis of Assembly Bill 1920 found “Unknown, major future cost pressure potentially in the millions of dollars to fund [SIB] contracts” • Maryland Department of Legislative Services fiscal note on House Bill 517: “It is assumed that the state will at least initially have to provide upfront funding for some or all of the SIB contract costs” and “general fund expenditures may increase. ” • NJ fiscal note on Assembly 3289 “If there are no savings realized by the departments and no repayments made to the lenders, the authority may be responsible for the full $15, 000”

Can SIB program outcomes grow the pie? In Peterborough, payment amounts were set “on the basis that the SIB was innovative. ” –RAND Europe “The Peterborough SIB is too small to deliver substantial ‘cashable’ savings (monetised benefits)” –RAND Europe “There is no guarantee that a reduction in re-offending will reduce demand for prison places. ” –Fox and Albertson, Journal of Criminology and Criminal Justice “This tool is a more expensive way to scale programs than if government simply contracted directly with a service provider. ” –Mc. Kinsey & Company

What kind of innovation can we expect from investor selected program? “Profit-seeking investors will be most interested in social programs or models that are proven. ” –MDRC Staff “Investors appear to be sticking to models that have already been extensively evaluated. ” –CLASP Surveys of impact investors: • “Lack of successful investments” was the most commonly cited challenge to impact investment – 2011 JP Morgan Survey • 62% of respondents would sacrifice financial returns for greater impact, but 60% thought this was not necessary – 2011 JP Morgan Survey • “ 65 percent principally target ‘market rate returns. ’” There was also a “clear client interest in risk-mitigating features. ” – 2013 JP Morgan Survey

How confident can we be in the results? “Interventions have to be delivered within systems. . . When a program achieves positive results, the success may be attributable to a wide range of factors, and it is often difficult to identify exactly why the program worked. ” –MDRC staff “The political imperative to demonstrate the success of this new financing scheme can create incentives for weaker evaluation designs that are more likely to show positive results, but that are spurious. ” –MDRC staff

How confident can we be in the results? “Overall, the results suggested that those released from HMP Peterborough were of lower risk of reoffending than those released from other prisons. ” –Qineti. Q, University of Leicester “It is debatable whether the use of PSM in the current context violates the CIA assumption. This is because data held on the Police National Computer (PNC) are not very useful in controlling for selection of being released from HMP Peterborough. ” –Qineti. Q, University of Leicester The Peterborough SIB “does not have a defined intervention model or theory of change. While this was perceived to facilitate individualised services, it poses some challenges for evaluation. ” – RAND Europe Evaluations always have caveats. Value of research accrues over time with additional evidence.

Summary “There is currently no conclusive evidence substantiating the success of this model. ” –Austin/Travis County Reentry Roundtable, Feb 2014 Extensive evidence in history of performance-based contracting suggests a number of significant challenges and risks. Governments and stakeholders should carefully evaluate not only the intended benefits but the risks and opportunity cost of the model before starting a pilot. • Comparison should be between direct financing and operations via public agencies or contracts, rather than “no intervention. ” SIBs are a means to implementing a program and should be evaluated compared to other means of implementation. Important to remember that no program is a policy unto itself.

Resources “Social Impact Bonds: Overview and Considerations, ” Elizabeth Lower-Basch, CLASP Massachusetts Contracts & SIB RFIs maintained by the Harvard Social Impact Bond Lab “Evaluating Social Impact Bonds as a New Reentry Financing Mechanism: A Case Study on Reentry Programming in Maryland, “ Maryland Department of Legislative Services “Payment by results and social impact bonds in the criminal justice sector: New challenges for the concept of evidence-based policy? ” Chris Fox and Kevin Albertson, Journal of Criminology and Criminal Justice

Kelly Walsh, Ph. D kwalsh@urban. org John K. Roman jroman@urban. org Kyle Mc. Kay kyle. mckay@lbb. state. tx. us Keshia Cheeks@njjn. org J U S T I C E P O L I C Y C E N T E R The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. Photo: Moriza