Pay for Systems Change

It’s still early days for governments experimenting with pay-for-success (PFS) projects, in which governments ask private funders to foot the bill for social programs, and then agree to repay those funders if the programs achieve agreed-on results. As public budgets tighten and social challenges remain stubbornly persistent, policymakers see PFS contracts as a new way to get more out of taxpayer money by focusing on outcomes, rather than funding services that may or may not work.

Results so far have been mostly positive. The world’s first PFS project, a program to reduce reoffending among short-sentence offenders in the United Kingdom, reduced reoffending by 9 percent. A few dozen others have shown strong outcomes, and many have made repayments to funders. And, as was the case with the very first project in the United States, when a program focused on juvenile justice failed to produce any impact, funders lost money, but the city paid nothing.

Largely, though, it’s too early to tell what impact PFS will have over the long term. There are nearly 100 PFS projects in the works worldwide, and the verdict is still out on most of them.

But here’s the thing: even if they’re all successful, growth for PFS will be slow. It will require that governments craft every contract with painstaking detail and cultivate a large pool of risk-tolerant investors. The bespoke nature of PFS projects today suggests a halting path to real scale.

The approach can be too narrow in its focus. Most commentators on PFS focus on the financing. The intermingling of worlds—of government spending and nonprofit providers and investment capital—somehow captures the imagination. But the financing is not what’s special. In our view, the concept’s real potential lies elsewhere: as a model for how governments should fund social services.

Local, state, and federal governments in the United States spent $5.4 trillion in 2014, with more than half flowing to education, healthcare, criminal justice, workforce development, and child welfare. Few of the things governments spend money on require strong existing evidence that they work or rigorous ongoing measurement of their impact. The bi-partisan US Commission for Evidence-based Policy Making is the latest salvo in a long-running series of efforts to bring better data into the way government allocates resources.

Imagine what would happen if governments adopted a more performance-based approach to social service contracting. By linking payment more closely with performance, we could build a stronger market for outcomes that we all agree matter—and governments could get more for their, and our, money.

Doing this would require that governments clearly define and measure success, build data-driven program designs, and invest seriously in performance management. It would also require that governments get energized and creative about unsexy things like collaboration, data, and procurement. In particular, they would need to:

• Buy longer term. Most government services contracts last only one year; behavior change takes longer. It takes a few years, for example, to show the payoff of a maternal home visiting program that helps prevent childhood injuries. The best government agencies develop cooperative, multiyear contracts, creatively working under their authority to plan into the future. They design incentive structures that ramp up over time, allowing budgets of both public and private partners to adapt.
   
• Smash budget siloes. The benefits of an intervention often accrue across agencies and levels of government; hence, a single agency may dramatically undervalue a great program unless others join in. Moving chronically homeless individuals into permanent supportive housing, for example, isn’t just about reducing the number of nights they spend in a shelter. It also reduces emergency department visits, in-patient psychiatric care, and jail stays. Understanding and accounting for each of these elements helps incite action. Cross-agency budgeting and decision-making aren’t easy, but they help to get the most out of spending.
   
• Zone in on outcomes. Most governments pay for inputs, such as enrollment into a workforce program. Paying for outcomes—say, earnings increases over the 24 months after finishing that workforce program—means doing more: setting clear goals; being careful about measurement, benchmarks, and tracking over time; and closely managing performance.
   
• Price differently. Money—in the form of increased tax revenue or avoided costs—is a factor in funding decisions. But policymakers and taxpayers also value social and community benefits—things like lower infant mortality or better child literacy. These benefits are often implied but not explicit. A good cost-benefit analysis, and hence a good performance-based contract, puts numbers to the value of social outcomes and thereby puts them on the same foot as fiscal outcomes.
   
• Make data useful. Governments collect hugely valuable administrative data, but much of it is inaccessible, inaccurate, or incompatible with other sources. Only cutting-edge jurisdictions have integrated their systems so that they can understand, for example, which homeless individuals frequently go to jail or need emergency care. Even when good data do exist, agencies feel hamstrung by antiquated IT systems and inflexible privacy requirements. The next generation of performance-based contracting starts with better data and data sharing—both within governments, and between governments and service providers.
   
• Get creative with financing. Even the best-run nonprofits are notoriously cash-strapped. They can’t wait to get paid until they’ve measured long-term outcomes. Governments need to get creative with financial incentives—making interim payments, offering working capital loans, or carefully phasing in performance incentives—to head off budget squeezes during the transition to performance contracting.

In case it isn’t obvious by now, the real limiting factor to all of this is government capacity. Resources are already scarce; there’s hardly time to double down on procurement reform, data analysis, and ongoing performance management. This is where public leaders need to step up. They need to partner with philanthropy to outsource in the short term, and then carve out resources to do this work right going forward. It’s an investment in efficiency well worth the cost.

PFS took root as an innovation in nonprofit finance, but its biggest benefit will come from helping governments revamp the way they spend public money on social programs. Projects underway today are letting governments build the muscle for this kind of performance-based contracting. As the excitement transforms and clarifies into a track record, governments should learn from PFS, adopt what works, and adapt it to their priorities. Doing so will lead us toward a future that’s more successful in getting results for our communities.

Read more stories by Jeff Shumway, Michael Etzel & Jake Segal.