NGA PPP-SIB Lessons Learned
- 1. 1 Social Impact Bonds: Lessons Learned from the Growth of Public- Private Partnerships in Infrastructure Summary Social Impact Bonds (SIBs) are receiving increasing attention as a low-risk way for governments to finance innovative social service programs and pay only for desired outcomes. SIBsare currently being considered in 36 states. However, only six SIBs are currently operating in the United States, including three state-level SIBs in New York and Massachusetts, and only one has reached maturity.1 Because of the limited history of SIB use in the United States, government leaders are unable to adequately assess the costs and benefits of SIB delivery of social services to determine if the model is appropriate for their state. The expansion of interest in SIBs mirrors the development of the Public-Private Partnership (PPP) model for infrastructure delivery.2 The use of PPPs grew rapidly in the United States from 1989, when the first PPP enabling legislation was passed in California, to the present day. Between 1989 and 2015, $24.3 billion has been invested in transportation PPPs alone, including $22.7 billion since 2007.3 Today, the PPP markethasa robust track recordand a setof best practiceshas emerged to guide government leaders considering PPP delivery of infrastructure services. SIBs can be viewed as a specific type of PPP that delivers social services. Proponents of SIBs and PPPs share many misconception about the tools, arguing that their use helps provide new funding for social and infrastructure projects, they can be costless to the public sector, and they help curtail public sector waste. SIBs, like PPPs,are financing mechanisms to push payment of services into the future. They can realize significant value through the transfer of risk from the public to private sector. However,this risk transfer comes with increased costs due to the complexity of the contract processand expectations by private partners. Using SIBscanencourage the scale-up and innovation of social services.However,SIB use can also constrain current and future spending and raise public concerns around privatization. Officials interested in SIB delivery should undergo extensive quantitative and qualitative analysis for each project to ensure that SIB delivery will realize public value over traditional public sector delivery. Exploring PPP implementation can also point to possible pitfalls of using innovative delivery methods in state social services,including significant transaction costs that reduce savings, costly renegotiations, and optimism bias in project planning. What Are Social Impact Bonds? The public sector utilizes a variety of contracting methods for private delivery of services. SIBs should be seenaspartof a larger public sectorshift from standardcontractmethods to performance- based contracting. The basic types of standard contracts are fixed-price and cost-plus contracts. In a fixed-price contract, the public sector holds a competitive procurement and selects the lowest- price bid. In a cost-plus contract, the public sector pays for pre-agreed project costs along with a negotiated fee for service.4 In these types of contracts, the government procures specific services 1 Pay for Success Learning Hub, “Pay for Success U.S. Activity,”Nonprofit Finance Fund, http://payforsuccess.org/pay-success-deals-united-states#sc(accessed June 3, 2015) 2 For thepurposes of this paper PPPs refers to public-private partnerships focused on infrastructuredelivery. 3 Eno Center for Transportation, Partnership Financing: Improving Transportation Infrastructure Through Public Private Partnerships (Washington, D.C.: Eno Center, 2014). 4 Federal Acquisition Regulation, “Part 16 - Types of Contracts,”FAR 2005-83/07-02-2015, https://www.acquisition.gov/?q=/browse/far/16 (accessed July 15, 2015).
- 2. 2 with a focus on inputs, such as materials and labor. The government pays for negotiated inputs, regardless of whether the project achieves its intended goal. In this way, the government retains all performance risk and the project’s successis largely determined in the planning and contractdesign stages. A standard contract for education may engage a service provider for a set amount of hours of education services in a specific location over a pre-agreed period of time. Performance-based contracting focuses on outputs rather than inputs. In performance-based contracts,the government designs measurable performance standards,such as quantity and delivery time, and remunerates service providers based on their success in meeting those standards.5 The government’s role shifts from prescribing services and monitoring inputs to monitoring outputs.6 A performance-based contract for education may provide a set payment for each hour of education provided or each student enrolled. The government shifts some performance risk onto the private sector by tying payment to performance measures, although the public sector is responsible for whether the services achieve the intended goal. In 1994, the federal government initiated a pilot- program to encourage the use of performance-based contracts, with 15 agencies converting 26 contracts to performance-based contracts. The Office of Management and Budget found a 15 percent reduction in price and an 18 percent improvement in agency satisfaction.7 SIBs represent an extension of the performance-based contract model called “Pay-for-success” (PFS). In a PFS contract,the government delineates a specific set of performance outcomes for the services procured. The government only pays if those outcomes are met and verified by an outside organization.8 PFS focuses on performance outcomes rather than service outputs. A PFS contract for ESL services may tie payment to results of a language test administered to students after the program ends. In a PFS contract, performance risk is shifted fully to the service provider, but the public sectorretainsfinancial risk. Because the provider bearsthe performance risk, they have more freedom to design appropriate services to meet outcomes, and the contract design process focuses more on negotiating outcome measures rather than prescribing services.. In an SIB, performance risk is shifted to the service provider as in a PFS contract, but the financial risk is shifted to outside investors. An SIB involves a contract between the government and an intermediary to institute a program that will seek to achieve a specific set of outcomes, with a repayment structure tied to achieving those outcomes. The intermediary then subcontracts with a service provider to provide the treatment program and finds investors to provide upfront capital for the service provider. An independent evaluator is hired to assess the program. If the treatment population meets the negotiated outcomes as a result of the program, the government compensates the intermediary, who is then responsible for compensating investors. 9 Proponents of SIBs point to a number of potential benefits in using the model, including: Transfers risk from the public to the private sector. The risk shift away from government reduces the cost of failure and encourages governments to support innovative services, as they only pay for programs that achieve the desired outcomes.10 5 Federal Acquisition Regulation, “Part 37 – Service Contracting Subpart 27.6 – Performance-Based Acquisition,” FAR 2005-83/07-02-2015, https://www.acquisition.gov/?q=/browse/far/37 (accessed July 15, 2015). 6 William Eggers, Performance Based Contracting (Washington, DC: Reason Foundation, 1997). 7 Office of Management and Budget, Best Practices for Performance-Based Contracting, (Washington, DC: The White House, October 1998). 8 Steven Godeke and Lyel Resner, Building a Healthy and Sustainable Social Impact Bond Market: The Investor Landscape, (New York City, NY: TheRockefeller Foundation, 2012). 9 Laura Callanan, Jonathan Law, and Lanny Mendonca, From Potential to Action: Bringing Social Impact Bonds to the US, (McKinsey &Company, May 2012). 10 Jeffrey Liebman, Social Impact Bonds: A Promising New Financing Model to Accelerate Social Innovation and ImproveGovernmentPerformance (Washington, D.C.: Center for American Progress, 2011)
- 3. 3 Solves the “wrong pockets” problem by aiding in value capture across agenciesand among levels of government. A disincentive of prevention spending is that the benefits do not always accrue to the agency that pays. Therefore, agencies are reluctant to spend resources on prevention, regardless of the overall cost-savings. That is referred to as the “wrong pockets” problem.11 Similarly, state governments have little incentive to undertake prevention efforts if the federal government accrues the majority of the savings.12 SIBs rectify the wrong pockets problem by using outside funding for programs, with payments coming directly from top-line budgets where savings are accrued.13 Builds evidence for innovative or promising programs. State and federal leaders are increasingly interested in funding programs that have a track record of success.14 As a consequence, small programs must secure funding for evaluations in order to expand.In an SIB, providers undergo rigorous evaluation throughout the program to assess outcomes, helping to build an evidentiary-base around programs that encourages program adoption.15 Provides reliable funding for providers to scale. Non-profits are often unable to predict cash flow because they rely on the appropriations process or donors for their funding. Organizations can face solvency challenges because of funding unpredictability that act as a disincentive to growth. SIBs use investor funds to capitalize programs at the outset, removing the potential of insolvency from service providers participating in an SIB.16 Creates incentives for additional monitoring. In traditional contracting, the government is the only monitor of service providers. SIBs increase outside monitoring by introducing intermediaries, investors, and an independent evaluator to act as additional monitors. Enhances coordination among stakeholders on specific policy ideas and issues. SIBs can offer a way for leaders to coalesce support around key policy issues inside and outside of government. Successful SIBs require input from multiple agencies, and the process of identifying canhelp improve inter-governmental coordination.17 Furthermore, currentSIBs require support from the philanthropic community and canalign stakeholder groups around policy issues that government leaders have identified as high priorities.18 What Are Public Private Partnerships? The Organization for Economic Cooperation and Development (OECD) defines a PPP as an agreement between the government and private partners so that sufficient risk transfer occurs to align the objectives of the public and private sectors. 19 PPPs are increasingly used by government in the public provision of infrastructure services from Design-Build (DB), where a private firm is 11 John Roman et al., Five Steps to Pay for Success: Implementing Pay for Success Projects in the Juvenile and Criminal Justice Systems (Washington, D.C.: Urban Institute, 2014). 12 Jeffrey Liebman, Response to the U.S. Department of TreasuryRequest for Information, “Strategies to Accelerate the Testing and Adoption of Pay for Success Financing Models (Boston, MA:Harvard Kennedy Social Impact Bond Technical Assistance Lab, 2013), http://siblab.hks.harvard.edu/files/siblab/files/sib_lab_response_to_federal_rfi_v3- 1.pdf (accessed June 24, 2015). 13 Roman et al., Five Steps to Pay for Success. 14 2014 Economic Report of the President, “Chapter 7:Evaluation as a Toolfor Improving Federal Programs,” March 2014, 269-298. 15 David Butler, Dan Bloom, and Timothy Rudd, “Using Social Impact Bonds to Spur Innovation, Knowledge Building, and Accountability,” Community Development Investment Review (Federal Reserve Bank of San Francisco, 2013): 57. 16 KyleMcKay, Evaluating Social Impact Bonds as a New Reentry Financing Mechanism: A Case Study on Reentry Programming in Maryland (Annapolis, MD:Department of Legislative Services, January 2013). 17 Hannah Azemati et al., “Social Impact Bonds: Lessons Learned So Far,” Community Development Investment Review (Federal Reserve Bank of San Francisco, 2013): 22. 18 Steven Godeke and Lyel Resner, Building a Healthy and Sustainable Social Impact Bond Market. 19 OECD, Public-Private Partnerships: In Pursuit of Risk Sharing and Value for Money (OECD Publishing 2008).
- 4. 4 responsible for designing and constructing an infrastructure facility to Design-Build-Finance- Operate-Maintain (DBFOM), where a private firm is responsible for all construction, operation, and maintenance aspects of infrastructure delivery. Those models stand in contrast to traditional delivery, where the government contractswith individual firms for design, construction, operations, and maintenance, while at the same time holding public responsibility for service delivery. There are a number of potential benefits from using PPP delivery, including: Encourages risk transfer from the government to the private sector. Infrastructure projects entail a significant amount of construction, financial, and political risk. Academic literature points to a long history of risk mismanagement by the public sector.20 PPPs help transfer some risk to the private sector, where profit-maximization motives encourage better risk management. The degree of risk transfer depends on the type of PPP sought and the allocation of risk between public and private stakeholders during contract design. Access experience and innovation in the private sector. Although the public sector is capable of delivering infrastructure services, they often lack access to the same technologies, labor, and management capabilities as the private sector. Utilizing PPPs allows government agencies to take advantage of private sector expertise.21 Builds life cycle costing into upfront project costs.Life cycle cost analysis is a tool that seeks to determine all costs for the expected life of an asset. Those include design and capital costs, operating and maintenance costs, and end-of-life or disposal costs. With life cycle cost analysis, public officials can make decisions based on long-term rather than short-term cost projections.22 PPPs that combine construction and maintenance can force the private partner to internalize the life cycle costs of a project and seek efficiencies by reducing future costs through better design and construction.23 Increases on-time and on-budget delivery. Construction risk is especially costly in infrastructure delivery. The public sector bears the costs for budget overruns or construction delays in traditional delivery. PPPs shift some of that construction risk to the private sector to create incentives for on-time and on-budget delivery.24 A 2006 study of 62 design-build projects in the United States found that construction cost was 2.6 percent lower than estimates and reduced the construction time by an average of 14 percent.25 Similarities between SIBs and PPPs SIBs can be viewed as a public-private partnership used for the delivery of social services.26 Both SIBs and PPPs are designed to shift a significant portion of risk away from the public sector to the private sector. Efficient allocation of risk attempts to transfer each risk to the party best able to manage that risk at the lowest cost.27 The amount of risk transfer is related to the return expected by the private sector. As additional risk is transferred to the private sector, the public sector pays 20 Bent Flyvbjerg, “Survival of the Unfittest:Why theWorst InfrastructureGets Build – And What We Can Do About It,” Oxford Review of Economic Policy 25(3) (2009): 344-367. 21 Patrick Sabol and Robert Puentes, Private Capital, Public Good (Washington, D.C.: The Brookings Institute, December 2014). 22 American Society of Civil Engineers and theEno Center for Transportation, LifeCycle Cost Analysis (Washington, D.C.: ASCE and Eno, 2014). 23 Eduardo Engel, Ronald Fischer, and Alexander Galetovic, Public-Private Partnerships to Revamp U.S. Infrastructure (Washington, D.C.: The Brookings InstituteHamilton Project, February 2011). 24 Deloitte Research, Closing American’s Infrastructure Gap: The Role of Public-Private Partnerships (Deloitte, 2007). 25 SAIC, AECOM Consult, and University of Colorado at Boulder, Design-Build Effectiveness Study – As Required by TEA-21 Section 1307(f): Final Report (prepared for theUS Department of Transportation and the Federal Highway Administration, January 2006). 26 John Roman et al., Sharing Risk: How Pay For Success Can Make GovernmentMore Efficient (Washington, D.C.: The Urban Institute, June2014). 27 OECD, Public-Private Partnerships.
- 5. 5 an increasing “risk premium” to compensate private partners for that risk transfer. Government leaders must be aware that although risk transfer can be attractive, the increased risk premium, manifested in higher private sector rates of return, is a cost not present in traditional delivery. SIBs and PPPs are tools that expand budgetary horizons. In PPPs, procuring all infrastructure services from a single firm places life cycle costs at the forefront and encourages service delivery that may be more expensive in the short-term but cheaper in the long-term. Similarly, SIBs shift the budgetary focus forward by investing in preventative measures that deliver long-term cost savings, a portion of which may be monetized to pay for the upfront services. Although both place a focus on long-term costs, PPP dealterms are structured to last for the expected design-life of an asset (50 to 99 years), as private partners capture tax benefits born from accounting rules around asset depreciation and public partners pay for savings as realized.28 SIBs are structured to last only for a treatment period and observation period (typically 3 to 7 years) and the majority of public sector cost-savings will occur long after an SIB is concluded. SIBs have a short term because private sector partners are reluctant to commit capital to high-risk, illiquid investments and the public sector is unable to provide the rates of return required for use of long-term capital.29 Shared Misconceptions of PPPs and SIBs Because oftheir similarities, PPPsandSIBs share a number of misconceptions about their benefits: Both are tools of financing, not funding: Although both models allow governments to defer payments for current services, neither provides a new source of funds. Furthermore, the public sector must pay a rate of return to defer those payments and a risk premium for the possibility repayment will not occur. Therefore, PPP and SIBs both increase the cost of services above traditional delivery. Public leaders may justify the increased cost by pointing to user fees and economic growth (from infrastructure assets) or future savings (from preventative care). However,governments with the budgetary flexibility to pay for infrastructure or preventative social services on an ongoing basis will have difficulty justifying the increased costs associated with financing methods such as PPPs and SIBs. Both are unlikely to realize enough savings to make projects cost-neutral. A cost- neutral program is one whose revenue or savings matches the cost to implement the program. PPPs and SIBs can realize significant government savings, but rarely enough to pay for services in their entirety. PPPs realize savings from reduced construction costs, shorter construction time, and reduced maintenance expenditures, but those savings representonly a fraction of the total project cost.30 SIBs realize savings if treatmentsreduce the need for long-term mitigation services. But SIB deals are short and unlikely to realize enough savings during the deal term to make a project self-sustaining.31 Similarly, a sizable portion of savings may come from non-monetizable public goods, such as increased public safety, which undercuts the notion that SIBs can be cost-neutral in the long-term.32 Private sectorinvolvementis notasafeguardagainstwaste. Apurported benefit of PPP delivery is that it selects only viable infrastructure projects, as unviable projects will not produce enough user fees to attract private investors, thus filtering out wasteful projects from consideration.33 A similar benefit is ascribed to SIBs. Investors will only pursue SIBs 28 John Roman et al., Sharing Risk: How Pay For Success Can Make GovernmentMore Efficient. 29 Steven Godeke and Lyel Resner, Building a Healthy and Sustainable Social Impact Bond Market.; and Center for American Progress, Frequently Asked Questions: Social Impact Bonds (Washington, D.C.: Center For American Progress, December 2012). 30 SAIC et al., Design-Build Effectiveness Study. 31 Payforsuccess.org, “Fact Sheet: The Utah High Quality Preschool Program,” http://payforsuccess.org/sites/default/files/utah-fact-sheet-pdf.pdf (accessed June 26, 2015). 32 KyleMcKay, Evaluating Social Impact Bonds as a New Reentry Financing Mechanism. 33 Eduardo Engel et al., Public-Private Partnerships to Revamp U.S. Infrastructure.
- 6. 6 for treatments with strong evidentiary success,so SIBs guarantee better, more effective social delivery.34 In reality, if a contractis designed to have sufficient profits, investors will participate regardless of public value. Government officials bear the sole responsibility for determining which projects are the best use of public resources and whether public value is achieved by using innovative service delivery models. What Lessons Can Be Learned From PPPs? Significant Risk Transfer Aligns Public and Private Incentives The level of risk transfer between public and private partners determines a large part of the value PPPs represent for tax payers. When an appropriate level of risk transfer occurs, public and private incentives are aligned to help realize that value. Some examples include: Transfer of construction risk creates incentives for on-time and on-budget delivery. When sharing the cost of budget overruns or project delays, private partners better manage the construction period in order to reduce the likelihood of those outcomes.35 Transfer of maintenance risk creates incentives for lower lifecycle cost in project design. Transferring the cost, frequency, and effectiveness of asset maintenance encourages private partners to assess and minimize life cycle costs during project design.36 Transfer ofdemand risk creates incentives for better service. Demand risk refers to the demand by users for infrastructure services. When demand risk is held by the private partners,they have an incentive to ensure high-quality service to attractand sustain users.37 The lack of adequate risk transfer often leads to increased costs for service delivery or failed PPP deals. For instance,Virginia entered into a design-build contract in 2012 to construct SR-460, a 55- mile tolled highway. The state agreed to issue toll revenue bonds on behalf of the private partners and make regular payments during the construction period. However, the environmental permits for the project had not been obtained and the state did not transfer permitting risk to the private partners. Despite three years of permitting delays because of potential impacts to surrounding wetlands, the state made $250 million in payments to the private partners and incurred $43 million in costs to the DOT.38 The state was able to terminate the PPP contract in July 2015 and recover $40 million from the private partners, but no roadway has been built to date and the state is still obligated to pay bondholders for the outstanding debt.39 The limited transfer of risk incurred steep costs to Virginia taxpayers without any delivery of infrastructure services. A centralquestion of SIBs is whether the risk transfer that occursencourages better service delivery from partners. In a PPP,partners have direct financial incentives to provide more efficient service delivery. In an SIB, service providers are separated from financial rewards,as they receive upfront funds from investors and bear no financial risk for underperformance. Investors have a financial incentive for better service delivery, but may not have sector- or program-specific knowledge to 34 John Roman et al., Sharing Risk: How Pay For Success Can Make GovernmentMore Efficient. 35 Federal Highway Administration Innovative Project Delivery, Risk Assessmentfor Public-Private Partnerships: A Primer (Washington, D.C.: U.S. Department of Transportation, 2014). 36 Federal Highway Administration Innovative Project Delivery, Risk Assessmentfor Public-Private Partnerships. 37 Federal Highway Administration Innovative Project Delivery, Risk Assessmentfor Public-Private Partnerships. 38 Virginia Business, “Dead End for theNew Route 460,” http://www.virginiabusiness.com/news/article/dead-end-for- the-new-route-460 (accessed July 29, 2015). 39 Roanoke.com, “Virginia Settles Legal Fight Over U.S. 460 Project,” http://www.roanoke.com/news/virginia/virginia-settles-legal-fight-over-u-s-project/article_5a172f05-6c0c-5a01-86b1- 07918cc02605.html (accessed July 28, 2015).
- 7. 7 improve service delivery. The SIB model assumes that simply by making investors an additional monitor, more efficient or effective delivery will occur. In theory, service providers could bear some financial risk by taking on an equity position in an SIB or deferring a portion of their fees. In a Massachusetts SIB focused on recidivism, service provider Roca deferred 15% of their service fees (or $3.26 million) at the program outset. Roca will only receive full repayment if the intervention is a success. 40 Service providers who share in the financial risk transfer have an incentive for more successfulprogram delivery. However, SIBs are appealing to service providers precisely because ofguaranteed funding streams,and fewservice providers will be able to take on sizable financial risk and forgo program fees at the outset, especially for interventions that are being implemented at a new size or scale. Some SIBs involve investors with the ability to influence outcomes. Many SIBs have attracted program related investments from non-profit organizations focused on the sector the SIB is designed to serve. Service providers in those SIBs may benefit from the involvement or support of those organizations that can intelligently monitor progress and share best practices.41 A working paper from the National Bureau of Economic Research on SIBs in non-profit health care asserts that SIB investors with experience in the program area are able to encourage betterservice delivery, but inexperienced investors will likely have no marked effect on treatment outcomes.42 Unfortunately, many SIBsreceive support from foundations and banks with no sectoralexperience. Furthermore, SIB investors have been reluctant to accept a large amount of financial risk. Because SIBs are new, the risk of losing principal funds is difficult for investors to assess. SIBs also carry significant political risks. The appropriations processoffersfewguaranteesof government payment even if target outcomes are reached,especially if a change in government administration occurs. 43 Cost-savings in SIBs come less from small savings like reduced food costs in prison but instead from major reductions like reduced prison staffing or prison closure.44 To realize those savings, governments must contend with interest groups, which may reduce the motivation for governments to pay if these cost-savings are not realized. Investors have demanded more counterparty risks in early SIBs via government or philanthropic guarantees of capital, graduated payments for small increments of treatment success, recurring payments throughout the program, and wind-down provisions for investors.45 All of those steps increase investor interest in SIBs, but reduce the level of risk transfer to the private sector. Competitive Procurement and VfM Analysis Realize Value PPPs are typically procured in a competitive fashion, wherein the government announces project requirements and entertains bids from multiple firms. Competition encourages private partners to offer the best possible terms to win the contract by sharing efficiency, technology, and cost-savings with public sector.46 Government agencies typically undertake a multi-phase procurement process 40 Payforsuccess.org, “Fact Sheet: The Massachusetts JuvenileJustice Pay for Success Initiative”, http://payforsuccess.org/sites/default/files/ma-jj-pfs-fact-sheet.pdf (accessed July 23, 2015). 41 Steven Godeke and Lyel Resner, Building a Healthy and Sustainable Social Impact Bond Market. 42 Mark Pauly and Ashley Swanson, Social Impact Bonds in Nonprofit Health Care: New Product or New Package? (National Bureau of Economic Research, 2013). 43 Steven Godeke and Lyel Resner, Building a Healthy and Sustainable Social Impact Bond Market. 44 Hannah Azemati et al., “Social Impact Bonds: Lessons Learned So Far.” 45 Timothy Rudd et al., Financing Promising Evidence-Based Programs: Early Lessons From the New York City Social Impact Bond (New York City, NY: MRDC, December 2013). 46 Congressional Budget Office, Using Public-Private Partnerships to CarryOut Highway Projects (Washington, D.C.: Congressional Budget Office, January 2012).
- 8. 8 to maximize competition. That process can include a Request for Qualifications to identify qualified bidders, a Request for Proposals to develop a short-list of bidders, and a Best and Final Offer Round where the public sector negotiates final terms directly with bidding teams. Agencies also use a “best value” rather than “low cost” criteria to evaluate PPP proposals. “Best value” criteria allows bidders to submit proposals that may have higher initial construction costs but lower life cycle costs from reduced maintenance or longer expected life of the asset.47 Currently, SIBs are not awarded via competitive procurement because there are few qualified intermediaries and deals must be uniquely structured for each SIB.48 PPPs face similar constraints, as the PPP market is dominated by a small number of firms and each contract is designed around local and state laws. Agencies encourage competitive procurement for PPPs by compensating bidders in exchange for ownership of the bid proposals. Often, losing proposals can help shape the final project. The House Committee on Transportation and Infrastructure found that compensation encourages competition and recommends states use federal highway funds for PPP bidder compensation.49 A similar compensation system during SIB procurement could help encourage bidders, strengthen final proposals, and increase the number of intermediaries in the market. Value-for-Money (VfM) analysis helps governments determine if entering into a PPP results in more value than traditional delivery. Before procurement,the government produces a Public Sector Comparator (PSC) estimate, an assessment of the life cycle cost if the project is delivered solely by the public sector, and a “Shadow Bid,” a life cycle cost estimate if the project is delivered by a private bidder.50 That process gives the agency a rough comparison of traditional and PPP delivery and a cost-estimate of public delivery that can be compared with private bids during the procurement process. The VfM process ensures that agencies consider traditional delivery alongside innovative delivery. The VfM process is a central tool of countries like the United Kingdom, who require VfM analysis to be done at multiple stages of the procurement process.51 Agencies interested in SIBs rely on Cost-Benefit Analysis (CBA) to determine the value of SIB delivery. However, CBA determines whether a program is a good use of public resources,but is not appropriate in examining possible delivery methods.52 Conducting a CBA is essential to determine potential benefits of preventative programs, but a VfM analysis can help keep public delivery of social services a comparable option throughout the SIB procurement process. Enabling Legislation is Necessary and Can Help or Hinder Success PPPs cannot be used for infrastructure delivery without the passage of state-level enabling legislation, which grants the legal authority for agencies to enter into a PPP and provides the legal foundation for PPP contracts. State-levelenabling legislation addresses project eligibility, funding regulations, procurement requirements, PPP approval and review, and PPP contract provisions.53 California was the first state to pass PPP enabling legislation in 1989.54 Since then, 33 states,along with Puerto Rico and the District of Columbia, have passed some form of PPP enabling legislation. 47 Federal Highway Administration Innovative Program Delivery, Establishing a Public-Private Partnership Program. 48 Social Finance, A New Tool For Scaling Impact: How Social Impact Bonds Can Mobilize Private Capital To Advance Social Good, (Boston, MA:Social Finance, 2012). 49 Special Panel on Public-Private Partnerships, Findings and Recommendations of the Special Panel on Public-Private Partnerships (Washington, D.C.: House Transportation and InfrastructureCommittee, 2014). 50 Federal Highway Administration Innovative Program Delivery, Value for Money Assessment for Public-Private Partnerships: A Primer (Washington, D.C.: U.S. Department of Transportation, December 2012). 51 PricewaterhouseCoopers, Public-Private Partnerships: The US Perspective (PricewaterhouseCoopers, 2010). 52 Federal Highway Administration Innovative Program Delivery, Value for Money Assessment. 53 Eno Center for Transportation, Partnership Financing. 54 Jamie Rall, James B. Reed, and Nicholas J. Farber, Public-Private Partnerships for Transportation: A Toolkit for Legislators (Denver, CO: National Conference of State Legislatures, October 2010).
- 9. 9 States must pass similar enabling legislation to use an SIB. PPP legislation will not allow SIB delivery because most PPP legislation names the state agencies able to use PPP contracts,few of whom would also pursue SIBs. Twenty-seven states have home rule, which allows municipalities to enter into PPP contracts without state legislation. Large municipalities in many of those states have explored PPPs for infrastructure delivery.55 Those same municipalities may explore SIB delivery if state legislatures do not move forward with legislation to authorize SIBs. Not all legislation encourages PPPs.Manylaws include discouraging provisions such aslegislative approval of each project, limits on PPP contract length, provisions on the use of state and local funds, and project-specific legislation.56 Government leaders should explore how the design ofPPP enabling legislation serves to encourage or discourage the PPP market. State legislatures must balance between oversight to ensure public value and flexibility to design projects that attract investors. Mandating legislative or gubernatorial approval at a late stage is a deterrentfor investors, as they are reluctant to negotiate a PPP deal if there is an absolute veto point late in the process. 57 State legislatures should consider including provisions in SIB legislation to increase investor confidence that mandated payments will be made. Massachusetts passed enabling legislation that backsSIB contractswith the full faith and credit of the state.The legislation requires appropriations for the maximum potential payments eachfiscal year.Those funds are deposited into a sinking fund to guarantee the state has the ability to make expected payments to investors.58 PPP Units Help Build Public Sector Expertise and Best Practices PPP units are public sector teams that support the development, evaluation, implementation, and monitoring of PPP contracts. PPP units provide legal, technical, and financial expertise to agencies interest in PPP delivery. They can be housed within a specific department or created as a separate public corporation operating on a fee-for-service basis.59 Some experts advocate for two distinct PPP units: one responsible for project planning, selection, and awarding; and one for contract monitoring and enforcement. Separation reduces the likelihood that contract enforcement is weakened to encourage more private sector involvement in the procurement process.60 PPP units provide a number of benefits. They increase the in-house capacityto evaluate and execute PPPs and reduce the need for outside consultants. They minimize transaction costs by building human capital and institutional knowledge. PPP units can help to speed decision time when evaluating proposals, which can encourage more competitive bidding.61 PPP units enhance transparency because they are subject to government transparency laws.62 No state has established a SIB unit to date. States with existing PPP units may consider using the same team for SIB procurement, as those teams have many of the skills necessary to evaluate and structure SIBs. Guarantees Can Help Attract Private Investment Governments often include a variety of guarantees to attract investors for potential PPPs. The centralguarantee is the compensation mechanism. Investors may demand rights to collect tolls and 55 Eno Center for Transportation, Partnership Financing. 56 Eno Center for Transportation, Partnership Financing. 57 Eno Center for Transportation, Partnership Financing. 58 Jeffrey Liebman and Alina Sellman, Social Impact Bonds: A Guide for State and Local Governments (Boston, MA: Harvard Kennedy Social Impact Bond Technical Assistance Lab, June 2013). 59 Federal Highway Administration Innovative Program Delivery, Establishing a Public-Private Partnership Program. 60 Eduardo Engel et al., Public-Private Partnerships to Revamp U.S. Infrastructure. 61 Patrick Sabol and Robert Puentes, Private Capital, Public Good. 62 PricewaterhouseCoopers, Public-Private Partnerships: The US Perspective.
- 10. 10 set rates for a given facility. States that wish to retain rate-setting ability may choose to compensate investors based on an “availability pay” structure, where investors receive payments from the state based on performance metrics.63 Investors may also receive rights to revenues from an existing or proposed tax levy.64 States have awarded non-compete clauses to investors that guarantee revenue by preventing competing public or private investments.65 Stateshave also used structured financing through a state infrastructure bank to take on subordinate debt in a project.66 All six SIBs in the United States have used guarantees to attract investors. Each has some form of structured credit with senior and subordinate debt obligations. In Cuyahoga County, Ohio, the SIB program was entirely capitalized by non-profit investors. The Reinvestment Fund has a senior debt position, while the George Gund Foundation, the Nonprofit Finance Fund, the Cleveland Foundation, and the Sisters of Charity Foundation of Cleveland all have subordinate debt positions.67 Most programs with private investors have a principal loss guarantee from participating foundations. In the New York City SIB, Bloomberg Philanthropies guaranteed $7.2 million of the $9.6 million provided by Goldman Sachs.68 Other potential guarantees have been proposed to attract SIB investors including guaranteed cash flows, insurance on principal payments, or a state reserve fund.69 Federal Government Can Help Drive Adoption A number of federal programs have catalyzed the infrastructure PPP market in the United States. The Transportation Infrastructure Finance and Innovation Act (TIFIA) loan program was established in 1998 to provide low-cost financing for surface transportation projects. The Department of Transportation manages TIFIA, which provides direct loans, loan guarantees, and standby lines of credit.70 TIFIA loans are subordinate to other project debt and cannot comprise more than 49 percent of total project financing. Since 1998, TIFIA has made $22.1 billion in loans for 52 projects, and had $46 billion in project requests in fiscal year 2013.71 Other federalprograms include private activity bonds and 63-20 corporations, which allow private issuance of tax-exempt debt, Build America Bonds, which provide an interest subsidy to state and local governments, and TIGER grants, which are federal discretionary grants for infrastructure projects. The federal government has made some efforts to promote SIBs in state and local governments. The Department of Labor and the Department of Justice have held funding competitions for SIB pilots, which have supported SIBs in Massachusetts and New York. The Corporation for National and Community Service Social Impact Fund awarded $11.2 million in 2014 for SIB feasibility studies atthe state andlocal level.72 The 114th Congress is considering the Social ImpactPartnership 63 Federal Highway Administration Innovative Program Delivery, Establishing a Public-Private Partnership Program. 64 Patrick Sabol and Robert Puentes, Private Capital, Public Good. 65 Eno Center for Transportation, Partnership Financing. 66 Robert Puentes and Jennifer Thompson, Banking on Infrastructure: Enhancing State Revolving Funds for Transportation (Washington, D.C.: TheBrookings Institute, September 2012). 67 Payforsuccess.org, “Fact Sheet: The Cuyahoga Partnering For Family Success Program”, http://payforsuccess.org/sites/default/files/141204_cuyahoga_pfs_fact-sheet.pdf (accessed June 30, 2015). 68 Timothy Rudd et al., Financing Promising Evidence-Based Programs. 69 Social Finance, A New Tool For Scaling Impact. 70 Roger C. Altman, Aaron Klein, and Alan B. Kruger, Financing U.S. Transportation Infrastructure in the 21st Century (Washington, D.C.: TheBrookings InstituteHamilton Projects, May 2015). 71 Roger C. Altman, Aaron Klein, and Alan B. Kruger, Financing U.S. Transportation Infrastructure in the 21st Century; and Transportation.gov, “TIFIA Background Slides,” http://www.transportation.gov/sites/dot.gov/files/docs/ TIFIA%20Background%20Slides%20%2805-20-2015%29_1.pdf (accessed June 30, 2015). 72 Corporation for National and Community Service, State of the Pay for Success Field.
- 11. 11 Act, which would appropriate $300 million for a Federal Interagency Council on Social Impact Partnerships to provide funding and technical expertise for governments interested in SIB.73 The federal government can accrue a significant share of cost-savings from SIBs, which may warrant more federal support for SIBs. The federal government could provide grants for feasibility studies, cost-benefit analyses, or program evaluations to lessen the administrative costs of SIBs. Federal agencies could allow access to federal data sets to reduce evaluation costs in SIBs. Subordinate federal loans or loan guarantees could be made to attract investors.74 Lastly, the Community Reinvestment Act of 1977 (CRA) requires banks to meet the lending needs of their community. The federal government could allow SIB investments to fulfill CRA requirements.75 How PPPs Highlight Potential Concerns Around SIBs SIBs Have a Very Limited Track Record PPPs and SIBs are complex arrangements designed to address unique infrastructure and social service needs. The history of innovative financial tools shows that until a number of deals are completed, it is difficult to have a full understanding of the risks and benefits involved. With PPPs, many of these risks were revealed by failed deals. Best practices for PPPs were mostly developed by examining the failures of specific aspects of PPP deal structures. An example of this is the limited use of non-compete clauses after SR-91 in California. In 1995, Caltrans entered into a 35-year DBFOM to construct 10 miles of express lanes on SR-91. The agreement contained a non-compete clause barring public or private investment within a one and a half mile area around SR-91, which served as a guarantee to attract investors. In 1999, SR-91 faced congestion issues and Caltrans attempted to add additional lanes, but was sued for violation of the non-compete clause. The parties entered into four years of litigation and Caltrans was unable to make any upgrades to the surrounding roadways. In 2003, the court allowed Caltrans to purchase SR-91, freeing them to make delayed maintenance upgrades. Since that litigation, non-compete clauses have become very uncommon in PPPs and,when used, include compensation and buy-out clauses to give public officials flexibility to undertake appropriate upgrades as needed.76 The PPP world has made a concerted effort to share best practices to prevent those issues. The Federal Highway Administration Office of Innovative Program Delivery has developed an extensive library of resources and guidelines for structuring transportation PPP deals.77 The field has also benefitted from significant academic research by professors of public policy, economics, and civil engineering on the use and effectiveness of PPPs for infrastructure services.. Despite the proliferation of interest in SIBs, only one SIB has closed in the United States. The New York City SIB focusing on reducing juvenile incarceration at Rikers Island resulted in no measurable reduction in recidivism rates afterthree years.Asa result, the program wasclosed early during an exit window in year three of the four year program. The SIB has been celebrated as a success because the government did not pay for failed services, but the SIB resulted in a $7.2 million loss from the first SIB in the United States. Although, Goldman Sachs invested all $7.2 73 Social Impact Partnership Act (H.R. 1336), 114th Congress, 1st Session (March 4, 2015); Social Impact Partnership Act (S. 1089), 114th Congress, 1st Session (April27, 2015). 74 Jeffrey Liebman, Response to the U.S. Department of TreasuryRequest for Information. 75 Steven Godeke and Lyel Resner, Building a Healthy and Sustainable Social Impact Bond Market. 76 Eno Center for Transportation, Partnership Financing. 77 FHWA.DOT.gov, “Public-Private Partnerships,”Office of Innovative Program Delivery, http://www.fhwa.dot.gov/ipd/p3/(accessed July 1, 2015).
- 12. 12 million, Bloomberg Philanthropies guaranteed $6 million of that investment. As a result, Goldman Sachs lost $1.2 million of their principal investment, while Bloomberg Philanthropies absorbed a $6 million loss.78 Evaluators of the New York City SIB should make all treatment,evaluation, and financial reports public to fully understand this outcome. All government leaders should commit to making documents public via resources like the online “Pay For Success Learning Hub” developed by the Nonprofit Finance Fund to facilitate information sharing among practitioners.79 Doing so allows interested parties to learn and to conduct research necessary to develop the SIB field. High Transaction Costs May Minimize Savings PPPs are complex programs that involve technical and legal coordination among a number of stakeholders. As a result, implementing a PPP incurs significant transaction costs that can reduce projected savings. Examples of common transaction costs include legal fees, consulting and advisory fees, and engineering and environmental reports. PPP transaction costs are estimated to comprise ten percent of the total project value.80 That total does not account for the significant time and resource costs of government officials to design, manage, and negotiate those contracts. As a result, PPPs are rarely the cheapest delivery option when judged by dollar value, and instead derive their competitive value from the risk transfer that can occur. SIBs are much smaller deals in dollar value and length, but they incur a significant amount of transaction costs to execute. A feasibility study conducted by Maryland revealed that fixed costs to implement an SIB focused on recidivism reduction for 500 parolees totaled $450,000, an amount that would negate any projected government savings.81 SIB contract design is estimated to take between nine months and two years depending on the experience of contracting officials.82 Public staff may be required to provide assistance throughout the SIB such as coordinating data collection and monitoring budgetary effects. The Harvard SIB Technical Assistance Lab estimates that the equivalent of one full time public employee is required for the duration of the entire SIB contract.83 SIB transaction costs are likely to decrease as best practices are identified, program replicability emerges,and contracting officials develop experience in SIB arrangements. However,government officials should be aware of how transaction costs reduce the amount of savings. Understanding the overall transaction costs can help officials minimize those costs among stakeholders or seek assistance from federal governments or nonprofit partners. Renegotiation of Contracts is a Common Occurrence PPPs are often renegotiated. Between 1991 and 2010, nine out of twenty PPPs implemented in the United States were renegotiated, on average seven years after financial close. Renegotiations are problematic for PPPs because they often are settled in favor of the private partner, increase the costs of the project borne by the public partner, and undue the advantages gained via competitive 78 The original SIB was a four-year $9.6 million dollar program, with Bloomberg Philanthropies guaranteeing $7.2 million of theoriginal investment. Because it was ended in year three, only $7.2 million of the principal was spent. Urban Institute, “PuttingEvidence First:Learning From theRikers Island Social Impact Bond” http://www.urban.org/urban-wire/putting-evidence-first-learning-rikers-island-social-impact-bond (accessed July 8, 2016); and James Anderson and Andrea Phillips, “What We Learned From the Nation’s First Social Impact Bond,” The Huffington Post, July 2, 2015, http://www.huffingtonpost.com/james-anderson/what-we-learned-from-the- _1_b_7710272.html?ir=Impact&utm_hp_ref=impact (accessed July 15, 2015). 79 Payforsuccess.org, “Pay For Success Learning Hub,” Nonprofit Finance Fund, http://payforsuccess.org/(accessed July 1, 2015). 80 Patrick Sabol and Robert Puentes, Private Capital, Public Good. 81 KyleMcKay, Evaluating Social Impact Bonds as a New Reentry Financing Mechanism. 82 Jeffrey Liebman and Alina Sellman, Social Impact Bonds: A Guide for State and Local Governments;and Jeffrey Liebman, Social Impact Bonds (Washington, D.C.: Center for American Progress, 2011). 83 Jeffrey Liebman and Alina Sellman, Social Impact Bonds: A Guide for State and Local Governments.
- 13. 13 procurement. 84 Renegotiations are causedby both public and private partners.Private partners may submit unrealistic bids to win a contract with the intention of renegotiating later. Public partners may overestimate demand for infrastructure projects to attract investors or gain political support. 85 Renegotiation may be less likely in SIBs because the deal terms are shorter and the dollar values smaller, but it is still a possibility. Modern PPPs include termination plans for the government, and private partners can sell their investments at any time.86 Conversely, SIBs are designed so partners are unable to exit freely. External organizations may have a financial incentive to immediately cease treatment if repayment is unlikely, and the same holds true if governments have not negotiated maximum repayment amounts. However, abrupt cessation of services may be detrimental to the treatmentpopulation. SIB contractsare designed to have exit windows atdefined points or with specific notification timelines.87 The Massachusetts SIB has exit opportunities for lenders after the second and third year of the program.88 The New York City SIB was ended during an exit window in the third year of the four year program.89 The negotiated exit constraints may increase or decrease the likelihood of SIB renegotiation or litigation later on. Renegotiations may be particularly costly for SIBs. The SIB model assumes a bilateral contract betweenthe government and anintermediary. 90 In reality, many governments enterinto multi-party SIB contractsto retain authority over provider and intervention selection. 91 The Massachusetts SIB involved a multi-party contract with both the intermediary and the service provider.92 Multi-party SIB contracts will undoubtedly face higher costs in the event of a renegotiation. PPPs Can Suffer From Optimism Bias and Misaligned Incentives Actualdemand for infrastructure candiffer vastly from early forecasts,reflecting the optimism bias of government officials who use ambitious projections to build political support for a project.93 SIBs may suffer from the same optimism bias as PPPs. Feasibility studies project large cost- savings, but many of the savings come from long-term changes that may be politically unfeasible or subject to externalfactors. Furthermore, a significant portion of those savings come from public benefits that cannot be monetized. The gross savings from a SIB program often appears high, but that may reflect public sector optimism that overstates benefits to gain support for the program. The metrics used to determine payment may also cause misaligned incentives in an SIB. Payment metrics are often proxies identified to reflect potential long-term benefits of treatment. One of the early education SIBs uses the Peabody Picture Vocabulary Test before the education program to identify preschoolers who may need special education services in the future. Those students are 84 Eduardo Engel et al., Public-Private Partnerships to Revamp U.S. Infrastructure. 85 Bent Flyvbjerg, MetteK. Skarnris Holm, and Soren L. Buhl, “How (In)accurate Are Demand Forecasts in Public Works Projects: The Case of Transportation,” Journalof the American Planning Association 71(2) (2005): 131-146. 86 Fitch Ratings, Global PPP Lessons Learned (Fitch Ratings, 2013). 87 Jitinder Kohli, Douglas J. Besharov, and KristinaCosta, Inside a Social Impact Bond (Center for American Progress, 2012). 88 Pay for Success Contract among the Commonwealth of Massachusetts, Roca, Inc. and Youth Services Inc. (January 7, 2014) http://payforsuccess.org/sites/default/files/final_pay_for_success_contract_executed_1_7_2013.pdf (accessed June 5, 2015). 89 The Bond Buyer, “No Success Like Failure: N.Y. Sees Social Impact Bond Pluses” http://www.bondbuyer.com/news/regionalnews/ny-city-officials-social-impact-bond-big-plus-1077971-1.html (accessed July 8, 2015). 90 Jitinder Kohli, Douglas J. Besharov, and KristinaCosta, Inside a Social Impact Bond. 91 Mildred E. Warner, “PrivateFinance for Public Goods: Social Impact Bonds,” Journal of Economic Policy Reform (October 2013): 11. 92 Pay for Success Contract among the Commonwealth of Massachusetts, Roca, Inc. and Youth Services Inc. 93 Bent Flyvbjerg et al., “How (In)accurate Are Demand Forecasts in Public Works Projects: TheCase of Transportation.”
- 14. 14 tracked after the program throughout grades K-12, with payments made for each year they do not consume special education services.94 The public sector assumption is that the Peabody Test can be used as a proxy to identify future special education needs in students, but the program design may create a perverse incentive. Investors could seek lower initial scores to increase the number of tracked students and therefore raise the amount of possible repayment. As Campbell’s Law states, “The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor.”95 The public sector may also have a perverse incentive to set repayment targets low or make SIB evaluation methods weak so the SIB is seen as a success. Doing so risks renegotiation by the public sector if government administrations change during the SIB program. Relational contracting helps to constrain opportunistic behavior from perverse incentives in a bilateral contract if there are economic benefits from having a positive, long-term relationship between parties.96 However,entering into those relationships can result in misaligned incentives in PPPs and SIBs. PPP equity investors may accept unfavorable conditions to establish a long-term relationship with the public partner. That can create conflicts if equity investors agree to fees or payments that cannot pay debt holders, resulting in a misalignment betweendebt and equity holders that can cause significant problems for the project.97 SIBs could potentially have similar incentive misalignment. Project evaluators could misreport or manipulate findings to maintain a relationship with public or private parties in an SIB. Intermediaries may be encouraged to “cream-skim,” or select treatment recipients most likely to meet targets, by the promise of shared profits from investors. Any of those scenarios would trigger renegotiation or contract termination. Furthermore, the small number of intermediaries and interested investors compared to the large public sector interest in SIBs means relational contracting may do very little to constrain opportunism. Entering Into a PPP or SIB Can Constrain Future Spending PPPs and SIBs are both financing tools that can help deliver current services and push payments into the future at a higher overall cost. Government leaders should be aware of the opportunity costs of that decision. PPPs and SIBs require governments to enter into long-term contractual agreements that can bind future administrations to pay for current services. Furthermore, PPP and SIB investors often require mandated appropriations or sinking funds for potential government payments. Although the traditional appropriations processallows governments to continue or cease program funding on an annual or biennial basis, future administrations have far less control over funding PPPs or SIBs than traditional programs. In that way, SIBs could be used a tool to bind social services spending of future administrations. SIBs may alter spending decisions in less obvious ways. In an SIB, governments only pay if the treatment population reaches negotiated metrics that determine payment. Depending on the evaluation design, governments may be less inclined to provide additional social services to the treatment population because those services could have spill-over effects that cause increased government repayment. The public sector could demand that evaluation methods be redesigned to account for additional services, but that may initiate a costly renegotiation, as private partners are better served by leaving additional social spending unaccounted. Although it may seem unethical for governments to withhold additional services for a vulnerable treatment population, the decision will be driven by the repayment structure and the level of fiscal constraints facedby public partners. 94 Payforsuccess.org, “Fact Sheet: The Utah High Quality Preschool Program.” 95 Donald T. Campbell, “Assessing theImpact of Planned Social Change, Evaluation and Program Planning 2(1) (1979): 67-90. 96 Oliver Williamson, The Economic Institutions of Capitalism (New York, NY: TheFree Press, 1985). 97 Fitch Ratings, Global PPP Lessons Learned (Fitch Ratings, 2013).
- 15. 15 If a treatment does not show results, governments may find themselves unable to end SIBs because of expensive exit provisions or because ceasing services may be detrimental to the treatment population. Alternatively, governments may make full SIB payments because payment targetswere reached,but later find that projected budgetary-savings are unrealized. In that case,the government has doubled spending, as they must pay for both prevention and mitigation services for the same population. Some may argue that governments would be required to pay for mitigation services regardless, but those increased payments may exacerbate budget constraints. SIBs could alter overall spending priorities in the social sector. New capital for SIBs may redirect current spending towards programs attractive to SIB investors and away from existing areas, thereby cannibalizing rather than increasing spending for social services. That potential shift represents an increased cost of government services and an underinvestment in programs that may hold social value but prove inappropriate or undesirable for SIB delivery. 98 Concerns Around the Source of Public Value Critics of PPPs have raised concerns about the source of their cost-efficiencies. A CBO report of the 2005 lease of the Chicago Skyway to private operators showed that although average maintenance expenditures decreased by an annual average of 10 percent, the majority of those savings came from reduced labor costs, as municipal employees were replaced by new employees earning 25 percent to 40 percent less. 99 PPPs may also derive value from circumventing prevailing wage laws and Buy America provisions. Government leaders should be aware of similar concerns with SIBs. Savings in an SIB may derive from lower labor costs or decreased regulation on service providers. Although that would represent savings for a taxpayer, it may come with political costs that may not be initially apparent. Conversely, labor and procurement costs may represent less of a cost-driver in the social service sector targeted, which would result in less cost-savings realized by SIB delivery. User Fees Represent the Public Voice in a PPP PPPs often utilize direct user fees such as tolls to compensate investors. Those user fees can be seen as a market mechanism that reflects consumer sentiment in a project. If citizens feel a toll is unjustly expensive, a certain population will stop using the facility. If they are happy with the price and service quality, they will use that facility more often. The user fee creates an incentive for private partners to deliver betters services and maximize total user fees collected. SIBs do not have a similar mechanism. In fact, the “user” in an SIB has no voice. The treatment population in an SIB has no control over program design or service provision. Instead, intermediaries or investors represent the voice of treatment groups in the SIB model during service provision. Early SIBs have offered options for recipients to “opt-out,” but that is unlikely to happen with an SIB focused on vulnerable populations, especially if no alternative services are offered.100 Government officials may wish to retain more control in an SIB to representvulnerable populations by entering into multi-party contracts with intermediaries and service providers, but that comes with increased costs, reduced risk transfer, and the potential for more expensive renegotiations. PPPs Can Be Used as a Tool to Implement Unpopular Policy 98 John Roman et al., Sharing Risk: How Pay For Success Can Make GovernmentMore Efficient 99 Congressional Budget Office, Using Public-Private Partnerships to CarryOut Highway Projects. 100 Mildred E. Warner, “PrivateFinance for Public Goods: Social Impact Bonds.”
- 16. 16 PPPs are often used to implement politically unpopular decisions, such as toll increases. The Chicago Skyway had a flat toll rate that declined 25 percent in real dollars between initial construction in 1989 and 2004. After being leased to a private consortium in 2005, toll rates increased 60 percent over a six-year period.101 By providing private partners control over rate- setting, a much-needed rate increase occurred without political consequences for public officials. PPPs have also been used for off-balance sheet spending. When PPP debt is issued by an SPV or backed by specific revenue streams like user fees,it is not counted on government balance sheets, which allows governments to circumvent debt constraints, budgetary limitations, and credit rating issues for large projects.102 Using PPPs in that way tarnishes the model for the public. There is no evidence to suggest SIBs have been used to circumvent budget constraints. In fact,that scenario is currently unlikely because investors are requiring pre-appropriation of SIB payments, which guarantees they appear in public budgets. However,as the SIB market grows and investor requirements ease,governments may be tempted to misuse SIBs for off-budget spending increases. There are Strong Public Concerns Around PPPs PPPs draw strong criticism from the public around a number of aspects. Many view PPP delivery as a form of privatization that results in the sale of public assets. Critics became especially vocal after the 75- and 99-year leases in the Chicago Skyway and Indiana Toll Road contracts.103 PPPs are more common internationally than in the United States, and foreign companies are often involved in domestic PPPs. Critics have expressed concern that PPPs amount to foreign control of vital infrastructure with the potential to limit state sovereignty over public assets.104 PPPs with user feesinspire concern over how tolls are set,who controls rate-setting,and whetheror not tolls should expire afterdebt is retired. Labor unions have beenactive participants in discussion of PPP delivery because of concerns that value stems from reduced labor costs.105 Public opposition to PPPs has stopped projects and cooled the PPP market within a state. In Texas,legislators proposed a 4,000 mile Trans-Texas Corridor that would use PPP delivery for highway, rail, and utility infrastructure. Concern over tolls, non-compete provisions, and the contract length caused legislators to cancel the project and pass a state moratorium on PPP development.106 Public concern around SIBs has been minimal. Opposition against SIBs has focused on private profiteering from core social services. As the SIB market grows, increased concerns around privatization, transparency, and contract terms are likely to emerge. The PPP market has shown that significant investment in public engagement is essential to address stakeholder concerns, especially in the early stages. Successful PPPs have engagement campaigns that continue throughout the processand provide stakeholders an opportunity to raise issues thatcan be addressed to ensure the public benefit of a project. Project transparency must be highlighted in the project planning and contract design phase to assuage concerns of cronyism. Conclusion SIBs and PPPs are innovative delivery methods that provide much needed government services. By involving the private sector, both models help align participant incentives to deliver services that can be more cost-effective and less risky for the public sector. However, neither PPPsnorSIBs should be viewed asthe preferredmode of service delivery by a state.Both PPPsandSIBs are most 101 Congressional Budget Office, Using Public-Private Partnerships to CarryOut Highway Projects. 102 Eduardo Engel et al., Public-Private Partnerships to Revamp U.S. Infrastructure. 103 Eno Center for Transportation, Partnership Financing. 104 PricewaterhouseCoopers, Public-Private Partnerships: The US Perspective. 105 Eno Center for Transportation, Partnership Financing. 106 Eno Center for Transportation, Partnership Financing.
- 17. 17 useful for high-risk services that can incur substantial costs to the public sector. Predictable and routine services procured using innovative delivery methods will result only in increased costs for standard services. As such, PPPs and SIBs should be viewed as part of an extensive contracting toolkit available to the public sector. States must work to implement regulations and processes that place public value at the forefront of SIB procurement to maximize their success. States interested in SIB delivery should focus on developing an analytical framework that can be used to assess each potential SIB on a case by case basis. A CBA can help determine if the project will result in a net benefits for the public, but it does not help weigh traditional versus innovative delivery. Developing a VfM tool for SIBs can help officials decide amongst potential SIB bids and weigh bids against traditional delivery. As the SIB market grows, governments should institute a competitive procurement processand require public CBAand VfM analyses to ensure SIB delivery represents savings over traditional delivery of services. States must also weigh public and private concerns when designing enabling legislation and business practices to ensure that there is project flexibility and that public value is at the forefront of SIB procurement. Building a base of institutional knowledge around the assessment, implementation, and evaluation of SIBs can help make them an integral part of service procurement for state and local governments. Recommendations to Move SIB Project Forward State leaders that have determined that SIB delivery may realize significant public value can take a number of policy steps to help build the pipeline of potential SIB bidders. By implementing these policies, government leaders can help create a robust SIB market in their state. Pass Robust SIB Enabling Legislation Well-designed SIB legislation serves as a positive signal to investors and can help to catalyze the SIB market. Policy makers may be tempted to pass project-specific legislation with a sunset provision. However, project-specific legislation can force parties to move faster than desired to meet imposed timelines. Government leaders should support SIB legislation that provides flexibility and reflects governmental values to protect the public interest in a project. Strong SIB legislation should not include project constraints. Limiting contractlength or total dollar value of deals often disqualifies projects that may realize public value. SIB legislation should act as a guide to implement projects that benefit the public. Legislators should consider the approval process for SIBs. Legislative and gubernatorial approval provides public protection but can slow the processdown significantly. Approval at the end of a project canactasa project veto that negates years of project development and dissuades future investors.107 Legislative or gubernatorial approval should occur early in the process and final approvals should be limited to key personnel. SIB legislation should have strong requirements for project transparency. All RFIs,RFPs,and bids should be publicly available. Final contracts should be made public after project close along with all periodic and final program evaluations. All SIBs should be open to state audits at the discretion of the governor or legislature. Transparency laws can assuage concerns around the abuse of SIBs. Begin Engagement Plan for Public Understanding of SIBs Widespread usage of SIBs will only occur if the public is adequately educated about the potential benefits. State leaders that see SIBs as a potential delivery model should begin to engage stakeholders around the legislative process. Early engagement will identify key stakeholder groups and help design SIB legislation that addresses public concerns. Officials should hold public RFIs 107 Eno Center for Transportation, Partnership Financing.
- 18. 18 early on to identify service providers that are interested in SIBs and non-profit organizations that may wish to participate. Engaging local organizations early will ensure the design of strong, well- developed SIBs that are better able to attract private investment. Create a Dedicated SIB Unit SIB units help build institutional knowledge and streamline evaluation, implementation, and monitoring of SIB contracts. States should consider utilizing existing PPP units to execute SIB delivery, or embed SIB units alongside PPP units. Government leaders can exercise discretion in designing SIB units to act as consultancies to state agencies interested in SIBs or as autonomous agencies with decision-making power to evaluate and execute SIBs. Early on, SIB units should outline procurement rules based on existing state laws and SIB legislation. Clear procurement rules with defined selection criteria encourages investors and can help build a project pipeline. SIB units may consider compensating bidders to increase competition in procurement. Unsolicited proposals represent a way to generate a number of early projects. However, they can cause problems by diverting scarce departmental resources to evaluating proposals and raise concerns about favoritism if no-bid contracts are awarded. SIB units must develop a protocol early on to address unsolicited proposals. Unsolicited proposals should be reviewed on a clear timeline. Unsolicited bidders may be remunerated in some way, as long as it does not alter the competitiveness and transparency of the established procurement process. 108 Government leaders may also wish to give SIB units some responsibility around data management and data sharing. Effective data management can reduce a number of SIB transaction costs including preparation of a CBA, determination of appropriate payment metrics, and preparation of progress and evaluation reports. SIBs should be empowered to work with agencies to implement strong data practices, streamline information sharing, and identify efficiencies in data collection. Federal Government Incentives Can Catalyze SIB Market State leaders may wish to encourage federal support to promote SIBs. Direct grants or subsidized financing can help make SIBs more attractive to investors. Federal support allows states to share in cost-savings that may accrue to the federal government as a result of SIB programs. Federal agencies should also allow SIB investments by banks to qualify as a CRA obligation to encourage more participation in that new type of delivery. States may also wish to consider providing incentives for SIBs such as capital guarantees,subordinate credit positions, or financial incentives in procurement. Government leaders should be aware thatwhile incentives may attract more private investors to a project, they can reduce risk transfer,create perverse incentives,or reduce the public value of SIB delivery for those services. 108 Eno Center for Transportation, Partnership Financing.