How this Works > Reasons for Caution and Optimism

A Focus on Cost Savings May Undermine the Influence of Outcomes-Based Funding Mechanisms

By Jitinder Kholi, Megan Golden, Samantha Mignotte

Think of a world where the only things you did were the things that saved you money — not the things that helped you succeed in life, not the things that helped others, just the things with quantifiable cash savings. Now imagine if governments did the same, meaning they wouldn’t pay for roads or support early childhood development programs if the monetary costs outweighed savings to the budget.

That might sound outrageous because we all know that the value to society of roads and early childhood development go well beyond just budget savings. However, when governments determine whether or not to fund outcomes-based projects, such as Pay for Success projects, they routinely require that these projects produce real, “cashable” savings in government budgets. This assumed need for cashable savings thus poses a direct risk to the success of Pay for Success — and more broadly outcomes-based funding — as a means for promoting change.


A Pay for Success project — also called a social impact bond — is an outcomes-based funding mechanism where one or more government entities agrees to pay a set amount for specific outcomes. The government’s promise of payment allows organizations to raise money to bring programs to the people who need them to achieve these agreed-upon outcomes. Governments, foundations, intermediaries, banks, and service providers around the world are exploring and using Pay for Success to address pressing problems, including homelessness, crime, education, and health disparities.

In our experience monitoring and assisting with Pay for Success projects across the United States, identifying a source to pay for outcomes is a key bottleneck preventing Pay for Success from reaching its potential. This is because many mistakenly believe that a government should pay for outcomes only if they save money in that government’s budget that can then be used as the main funding source. Why is this belief so widespread? Because when Pay for Success was first introduced in the United States, advocates — including the authors of this piece — presented it that way. At the core of the original pitch for Pay for Success was the idea that outcomes could be paid for out of cashable savings — the amount that an agency would save from the reduced need for costly remedial services.1 Governments could therefore achieve outcomes without committing any new funds and might even “make money” off the project by paying out less than it saved.

We have learned, however, that using savings as the sole basis for determining whether and how much government should pay for outcomes poses two important risks. First, it can lead to missed opportunities to fund programs that can solve some of our nation’s most pressing problems. Second, it risks enshrining the status quo rather than shifting government resources to more effective programs.

Missed Opportunities to Fund Effective Programs

Government agencies make decisions about what to fund every single day — such as patching a road or operating parks. Governments allocate billions of dollars to social services every year. In each of these areas, governments invest because they believe that there are gains to society as a result of the spending (although some commentators find scant evidence of effectiveness for many programs).2 In none of these areas, does government only invest when there are cashable savings. Yet they tend to focus on cashable savings only when contracting for social outcomes.

This emphasis on cashable savings makes it harder to fund social impact outcomes than to fund most other government services. Many projects that provide support to domestic violence survivors, for example, would not be funded if they had to meet the cashable savings standard. Underinvestment may also occur if the savings from outcomes-based projects accrue to another agency or level of government than the one directly undertaking the contract (which some have referred to as the “wrong pockets” problem).3 For instance, if cost savings for asthma prevention accrue to state and federal health care programs, a local housing agency would be less likely to fund cutting-edge ways of reducing home-based environmental health hazards that can provide a healthier environment for its residents.4

Sustaining the Status Quo

In addition to underfunding programs that make a documented impact, basing funding decisions on cashable savings alone makes it difficult to shift funds from less effective programs to new, more effective approaches. If new initiatives cannot be undertaken unless offset by savings, current funding allocations will remain in place as the default, even if these programs are not the most effective way to achieve the government’s ends.


The belief that government budgets are too tight to fund outcomes-based projects without cashable savings incorrectly assumes that government funding allocations cannot and do not change. There is, however, flexibility within budgets that offers alternative ways for governments to find the funds to pay for outcomes beyond just cashable savings.

Almost all city, state, and federal budgets include discretionary funds for political priorities or emerging needs. If paying for outcomes is set as a priority, discretionary funding could also be used to support it. Some jurisdictions have already found ways to use discretionary funds to support outcomes-based projects. Massachusetts, for example, created a Social Innovation Financing Trust Fund that the state uses to support Pay for Success projects funded through annual appropriations.5 Utah, similarly, created a School Readiness Restricted Account, composed primarily of funds appropriated by the state legislature, which pays for the outcomes of its preschool Pay for Success project.6

In addition, government could pay for outcomes by repurposing existing programmatic budgets. Agencies responsible for social outcomes typically have budgets to fund specific types of services to achieve those outcomes. As these programs scale down, end, or are found to be less effective, there may be opportunities to shift some funding to outcomes-based contracts. To do this, an agency could allocate a portion of a programmatic budget to outcomes-based contracts and then issue a solicitation for projects to achieve an outcome for a particular population, instead of issuing a solicitation for a specific type of service. The U.S. Department of Labor, for example, created a Workforce Innovation Fund that allocated $24 million for outcomes as a test of a more effective way to accomplish the agency’s mandate.7 Its solicitation for grant applications called on state and local governments, service providers, and intermediaries across the country to design outcomes-based projects that would improve employment outcomes for high-risk groups in measurable, verifiable ways. As a result, Department of Labor funds will pay for outcomes for both the New York State and the Massachusetts recidivism reduction and employment Pay for Success projects.8

Economic development funds, which often exist even for small cities and counties with limited budgets and areas of government responsibility, offer another possible source of outcome funding. Projects such as those focused on early childhood development often bring jobs and revenue to the community, thus contributing to economic development goals. These economic impacts are often not included in the cost-benefit analyses of social programs that underlie Pay for Success efforts as indirect benefits can be difficult to quantify. The Institute for Child Success, in its work around the country, has seen interest in using economic development funds to pay for outcomes of effective early childhood programs.9

Yet another source of funding for outcomes, especially for local governments, is sales taxes or other assessments. Government officials have told the Institute for Child Success that such taxes are a viable option for funding outcomes.

Though the applicability of these methods may vary, what they show is that cashable savings are not the only way to fund Pay for Success programs, even for cash-strapped governments. Given these alternatives and the risks posed by focusing only on cashable savings, governments should begin exploring alternative ways of determining whether and how much funding to allocate for Pay for Success initiatives.


In our 2015 paper for the Center for American Progress, we recommend that governments consider three factors in determining how much funding to commit for a Pay for Success initiative:10

  1. Wellbeing benefits. To what extent does the outcome lead to improvements for individuals and communities?

    The full benefits of outcome based projects include much more than monetary savings to taxpayers. They also include improvements for individuals and the community at large. The price of outcome payments should thus reflect the value of these broader benefits in all instances where they can be quantified and measured.
  2. Public willingness to pay. Is the outcome important enough to the community that additional public dollars should be allocated to achieving it?

    In some instances, the public may assign additional value to an outcome based on factors beyond wellbeing or savings. The price of an outcome should therefore include public willingness-to-pay for outcomes that reflect their values (such as preventing domestic violence), pressing priorities (such as finding a vaccine for the Zika virus), or moral obligations (such as supporting veterans).
  3. Cashable savings. Will government achieve cashable savings by investing in this initiative?

Despite its potential drawbacks, the amount that government is likely to save should be one of the key factors in determining the price of outcomes, however, it should not be considered in a vacuum.

The goal of this approach is to present key factors that together capture the true value of an outcome to government, individuals, and society as a whole — and, therefore, a better sense of the right price to assign it. New York State’s “Increasing Employment and Improving Public Safety” project did exactly this. It based outcome payments on both estimated savings from reduced prison expenses and the value of benefits to society from reduced crime.11 This combination of factors presents a more holistic, and realistic, approach to valuing Pay for Success outcomes.

While a holistic approach offers many potential benefits, there are still challenges to using it in practice. For instance, wellbeing benefits and public willingness to pay are often difficult to accurately quantify, as noted previously with the example of the economic impacts of early childhood development programs. However, there are good examples of ways to overcome this challenge. The Washington State Institute for Public Policy has developed a rigorous approach to valuing benefits and costs of an intervention over time.12 When analyzing a crime intervention, for example, the institute looks both at the savings to government as a result of a reduction in future crime as well as benefits to potential victims of crime and the improved labor market benefits for participants with improved rates of graduation.

Proxies can also be used to overcome challenges related to the accuracy of public willingness-to-pay. One of the easiest proxies to use is the willingness of legislature or executive branch leaders as their choices often reflect the priorities of their constituents.

Getting better at valuing outcomes is important not only for Pay for Success initiatives, but also for the multitude of outcomes-focused funding strategies spreading across the nation, such as performance-based contracts, evidence-based budgeting,13 and outcomes-based bonuses (to name a few).14 As we shift our thinking from basing funding on cashable savings to a more holistic understanding of outcome value, we can unlock funding for strategies worth far more to society than just the things that save us money.

1 Jitinder Kohli, Douglas J. Besharov, and Kristina Costa, “Fact Sheet: Social Impact Bonds-A Brief Introduction to a New Financing Tool for Social Programs,” Center for American Progress (April 12, 2012), available at
2 John Bridgeland and Peter Orszag, “Can Government Play Moneyball,” The Atlantic (July/August 2013), available at
3 For more information on the “wrong pockets” problem, see John K. Roman, “Solving the wrong pockets problem: How Pay for Success promotes investment in evidence-based best practices,” Urban Institute (September 2015), available at
4 For more information on the Green and Healthy Homes Initiative in Salt Lake City, see
5 Mass. Gen. Laws Chapter 10, § Section 35VV, available at
6 U.S. Government Accountability Office, “Pay for Success: Collaboration Among Federal Agencies Would be Helpful as Governments Explore New Financing Mechanisms” (September 2015), available at
7 For more, visit: U.S. Department of Labor, “What Is the Workforce Innovation Fund,” available at; See also, U.S. Department of Labor, “The Pay for Success Solicitation,” available at
8 U.S. Government Accountability Office, “Pay for Success: Collaboration” (September 2015).
9 Kelly O’Donnell, “More than just Pre-K: The Positive Economic Impact of Preschool in Los Angeles County,” Institute for Child Success (March 2016), available at
10 Jitinder Kohli et al., “From Cashable Savings to Public Value: Pricing Program Outcomes in Pay for Success,” Center for American Progress (September 1, 2015), available at
11 U.S. Government Accountability Office, “Pay for Success: Collaboration” (September 2015).
12 Washington State Institute for Public Policy, “Benefit-Cost Technical Documentation,” Washington State Institute for Public Policy (June 2016), available at
13 Policymakers are integrating evidence of results into the budget process. For example, since 2015, the Mississippi Legislature has required agencies seeking funding for new programs to answer questions indicating whether it is evidence- or research-based, and detailing how measures will be monitored and measured. For more information, see Pew-MacArthur Results First Initiative, “A Guide to Evidence-Based Budget Development: How to Use Research to Inform Program Funding Decisions,” Pew Charitable Trusts (July 6, 2016), available at
14 In Tennessee, outcomes-based bonuses have been used to further improve child welfare outcomes by providing additional rewards to providers that exceeded set performance targets. For more information, see Beeck Center for Social Impact and Innovation, “Funding for Results: How Governments Can Pay for Outcomes,” Georgetown University (May 2015), available at