Lessons from the field: pay for success outcome areas

Pay for success (PFS) allows governments to pay on outcomes, rather than inputs, which incentivizes the expansion of evidence-based programs, improvements in service delivery, and reductions in costs, among other benefits. The outcomes governments have selected span multiple policy areas, reflecting the diversity of PFS target populations and goals. However, as the figure below shows, some policy areas are more prevalent than others. Recidivism/criminal justice and housing/homelessness are currently the most common outcome policy areas across PFS projects, while health-related outcomes are the least common.

Existing and past PFS projects suggest that the most widespread outcomes tend to share at least three important qualities. First, the outcome represents a large cost for governments. Second, the costs tend to be localized, typically within city or county government budgets. Finally, the potential end payors face few legal and practical restrictions to paying on that outcome.

Figure 1: Outcome Policy Areas, by number of PFS projects

Source: Data on PFS project outcomes came from the Nonprofit Finance Fund

Recidivism/Criminal Justice: Recidivism/criminal justice, the most common policy area, represents a large expense for local governments. Over half of all PFS projects pay for a criminal justice outcome, primarily days in jail or arrests. A Vera Institute of Justice report found that local jails across the United States held more than 730,000 people per day, on average, and had a direct estimated cost of $22.2 billion in 2011. Indirect costs can be expensive as well. New York, for example, spent $1.1. billion through its Department of Corrections and an additional $1.3 billion for employee benefits, health care, and education programs for employees and people incarcerated through other city agencies.

Housing/Homelessness: About a quarter of all PFS projects have an outcome related to housing or homelessness. Homelessness is expensive for local communities; Santa Clara County, for example, spends an average of $80,000 in services per person per year among its highest utilizers. However, capturing these potential cost savings is slightly more complicated than for a criminal justice-focused project because individuals who are homeless tend to incur costs in multiple systems, such as shelters, detox facilities, hospitals, and jails. As a result, the housing stability outcome rests on the idea that if an individual is stably housed, they will no longer cycle through those different systems, generating cost savings. Since housing stability is not directly tied to cost reductions on shelters, jails, hospitals etc., some PFS projects adjust the housing stability metric to account for days spent in those other services.

Housing stability also differs from jail days or arrests because it is measuring an output instead of an outcome, making it more difficult to estimate impact. Typically, projects focused on homelessness simply count the number of days or months a person remains in housing and pays a fixed amount. Jail days or arrests, on the other hand, are often measured as a reduction compared to a control group through a randomized control trial. Measuring the outcome through a comparison allows for better estimation of the project’s impact.   

Less common outcomes tend to have legal or practical restrictions that complicate incentives and repayment structures. 

Physical Health: Interestingly, physical health is not a common outcome (two projects have mental health outcomes and two have substance abuse outcomes). Only the South Carolina Nurse Family Partnership focuses on physical health, and several other projects focused on health have yet to get off the ground. Although there is a large potential for cost savings, health projects face several unique obstacles. First, health-related data faces a lot of restrictions around its use, which may discourage communities from pursuing PFS projects in this area. Second, although many projects are billing Medicaid for services, no project has been able to figure out how to overcome legal restrictions to use Medicaid as an end payor. Finally, incentives within the healthcare system may discourage other potential end payors. Hospitals make money by providing services and may not have a direct financial incentive to reduce costs. Insurance plans, which would benefit from reductions in costs, may not have an incentive to invest in preventative programs for their customers because customers frequently change plans, and any benefit would then go to a competitor. 

As the field matures, more governments may pursue projects with a focus on quality of life improvements that carry nonfinancial, or difficult to calculate, benefits. But for now, cost savings are a driving consideration for many project stakeholders. 

Have a Pay for Success question? Ask our experts here!

As an organization, the Urban Institute does not take positions on issues. Scholars are independent and empowered to share their evidence-based views and recommendations shaped by research. Photo via Shutterstock.