Data for Effective Partnerships: The Case of Social Impact Bonds in Canada OMSSA, December 2014

Transcription

1 Data for Effective Partnerships: The Case of Social Impact Bonds in Canada OMSSA, December 2014 Content copyright Finance for Good. All rights reserved

2 Agenda Introduction and Background on Social Impact Bonds How Social Impact Bonds (SIBs) operate The Data Story and Collaborators : Design and Contracting Which Issue? Is it worth doing? Is a SIB the appropriate mechanism? Section 2: Implementation and Wrap-up Are we improving? What would have likely happened? Now that we know this, what next? 1

3 What are Social Impact Bonds? INTRODUCTION Two Contractual Arrangements: 1. An Outcomes-Payment Contract engaging a willing payer for social outcomes, for example government or a philanthropy. This can either be directly with a given service provider or via an intermediary which is enabled to sub-contract to supplement with relevant services 2. A financing agreement to fund service providers, usually charities or nonprofits, to deliver a service targeting the realization of the agreed outcomes. This usually will take place through a special-purpose vehicle to protect service providers liabilities. 2

4 What are Social Impact Bonds II INTRODUCTION SIB Steering Committee Finance for Good SIB Project Partnership 1 Commissioning Ministry Investors 2 Service Provider Beneficiaries Independent Auditor 3

5 Social Impact Bond (SIB) Process Flow INTRODUCTION 4

6 Steps to Implement SIBs INTRODUCTION Pre-Structuring Issue Identification Service Provider Feasibility Assessment Identify high-potential issue areas that provide mutual shared value to commissioner and community Structuring and Design Program Design Investment and Payment Structuring Documentation Design program particulars including details of service for target population and geography, timeline, and evaluation/performance measurement Capital Raising and Implementation Capital Raising Service Delivery Market the project to investors securing necessary commitments for operations and other costs Project Wrap-up Reporting and Measurement Payment and Wrap-up Monitor project performance relative to outcomes, reporting to investors and service providers 5

7 Stages and Collaborators INTRODUCTION Pre-Structuring Issue Identification Service Provider Feasibility Assessment Structuring and Design Program Design Investment and Payment Structuring Documentation Capital Raising and Implementation Section 2 Capital Raising Service Delivery Project Wrap-up Reporting and Measurement Payment and Wrap-up 6

8 : Design and Contracting 7

9 Incidence Per 1000 individuals Which Issue: Identification Targeting Beneficiaries: Hypothetical Incidence Year Entire Population 8

10 Incidence Per 1000 individuals Which Issue: Identification II Targeting Beneficiaries: Hypothetical Incidence II Year Low Group Medium Group High Group 9

11 Incidence per 1000 Which Issue: Identification III Incidence Distribution Percentile 10

12 Collaborators on Issue Identification: Who has data on the state of local issues in your community? Is that information continuous, in intervals, or infrequent? Is there administrative data we can use or match to support this effort? 11

13 Is This Worth Doing? The Overall Economic Case 1 $ Public Sector Savings Retained Savings Reactive Spending Impact of SIB Reactive Spending Total Spending Preventative Spending Investor Returns Preventative Spending Outcome Payments Status Quo With SIB Service Cost Saving 1 Adapted from: Social Finance UK, 2014, Social Impact Bond Financial Model. 12

14 Digging Deeper: Economic Logic Ave. Prosecution Costs per case Ave. Administration Costs per offense Ave. Costs per arrest Cost of Victimization Expected Length of Incarceration Ave. Legal Aid Costs per case Ave. Court Costs per case Ave. # of Rearrests Ave. Rate of Re-arrest Average Severity of Crime Average O&M of Incarceration Offense Processing Costs (B2) Total Policing Costs (B1) Social Costs (B3) Prison Costs (B2 + B4) Number of Individuals Total Justice System Costs Expected Reoffense Baseline Reoffense Total Justice System Benefits Input Legend Output 13

15 Value Estimate to Simulate Business Case Outcomes Reduced Social Costs ($) B3 Additional Investment Cost ($) C2 Researched or Jointly Determined Ranges Reduced Provincial Cost ($) B1+B2 F = f (A, B, C, D,..) FTE Cost per Year ($) C1 Rate of Return (%) 14

16 Cumulative Cost (Thousands) On Estimating Value: Cost Structures Cost Structures Dominated by Fixed Costs $14, $12, $10, $8, $6, $4, $2, $ Number of Individuals 15

17 Willingness to Pay Distribution of Value Attachment of Value Social Provincial Local Federal Social Included Provincial Included Local Social Excluded Provincial Excluded Federal 16

18 Collaborators on Building a Business Case: Who has data on the cost of your delivery and the value of success? Where does that value accrue if we re successful in making positive social impact? When there s competing evidence around cost which do you consider the best sources? 17

19 Is a SIB appropriate?: Prevention and Risk SIBs generally come from the premise that there is value creation through earlier intervention and/or prevention than is occurring in current systems. The literature identifies two primary causes: 1. Time-delay There is a significant delay between spending and the outcome prevention targets which generally stretches past both budgetary and political cycles 2. Risk of Outcomes We can pay for both prevention and acute services through direct contracting if preventative outcomes fail to materialize 18

20 Value for Money $3,000, Value for Money Case Comparison $2,500, $2,000, $570, $143, $176, $50, Value for Money $1,500, Investor Premiums Additional Contracting Costs $1,000, $2,000, $2,200, Cost of Outcomes Risk Direct Cost of Operations $500, $- Direct Contracting Comparator Social Impact Bond Contracting 19

21 Public Sector Value for Money Value for Money II 20% 15% Public Sector Value for Money and Investor Returns in a SIB 10% 5% 0% -5% 14% 13% 12% 11% 10% 9% 8% 7% 6% 5% 4% -10% -15% -20% -25% -30% Blended Investor IRR Remedial/Preventative Spend Ratio

22 Cash flow ($1000s) Percent of Total Government Savings Sample Outputs I All Party Cash Flow $2,000 $1,500 $1,000 $500 $- -$500 -$1,000 30% 20% 10% 0% -10% -20% -$1,500 Investors Government Service Provider Government Cashflow as % of Total Savings -30% 21

23 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100% Total Net Amount Per Client Sample Outputs II $30, $25, $20, $15, $10, $5, Illustrative Public Sector Savings and Benefits from Performance of Hypothetical SIB $- Percent Reduction in Rate of Recidivism / Gain in Rate of Post- Release Employment Value of Additional Employment Cost Savings from Reduced Recidivism Value of Direct Avoided Service Costs Average Cost of Delivery Performance Payment 22

24 Collaborators on SIB Contracting and Deal Structuring: Who has a willingness to pay for the value created? What is an acceptable return if we re creating that value? Who are local investors who might be willing to discuss a deal before it s launched? How might we engage them in that discussion early? 23

25 Section 2 Section 2: Implementation and Wrap-Up 24

26 Performance Management Section 2 Data systems must allow for accurate and timely reporting on progress. While analyzing outcomes continuously is often cost in-effective, creating opportunities for interim monitoring and reporting allows for improvement as we learn from the front lines of operations. Consider where there are opportunities for shared measurement platforms Case 1 Intervention Effect Case 2 Intervention Effect Case 3 0 Intervention Effect 25

27 Performance Management Section 2 Gatekeepers will regularly be within the public sector regardless of a SIB, in order for ongoing referral to function, it is essential that they are bought-in, and ideally incentivized. This requires relationships with referring agencies, which need to be built as part of any project. 26

28 SIB evaluation Section 2 To get to the point of demonstrating impact, we should be asking ourselves What would have likely happened?. This is where counterfactuals are used In order for any counterfactual to be robust, it should to be effectively populated as the intervention is delivered and the process for this population identified ahead of time. Usually we rely on administrative datasets to keep costs down, so larger sets expand the scope of what is possible. 27

29 SIB Completion Section 2 We should be asking ourselves from the outset: What level of evidence do we think is necessary for this intervention to be better understood? What level is necessary for it to attract direct public investment? What level is necessary for it to become public policy? A SIB is likely to ultimately conclude with a particular program winding down, rolling into another SIB, or attracting further investment The capacity building instigated by a SIB can be thoughtfully nurtured throughout to endure post conclusion 28

30 Collaborators on SIB Implementation Who will control data on outcomes? Who refers the beneficiary population? Where are there shared tracking and monitoring systems that we would like to have, regardless of whether we engage in a SIB? 29

31 TITLE HERE Takeaways SIBs are part of a broader shift towards outcomes-accountability within social investment. Where concerns exist regarding activities, these can be addressed through specifying standards of behaivour rather than activities. The process of evidence-building around a SIB is one in which significant capacity-building takes place. Which skill-set(s) are anticipated to be gained enduringly could be made explicit and targeted. Part of the interest in SIBs is in continuous improvement within the social sector. The process of raising the bar on evidencing impact is not unique to SIBs and should be considered a building block in all government contracting. 30

32 Thank you, questions? Lars Boggild E: W: FinanceforGood.ca 31