Illustration by Mitch Blunt 

The 2016 US presidential election season has put a spotlight on the deep and growing anger that Americans have toward their public sector institutions. Indeed, Americans’ trust in government has reached an all-time low: In a survey taken in late 2015, only 19 percent of respondents said that they trust the federal government to do what is right “always” or “most of the time.” Trust in the ability of government to address urgent social needs is also low. A large majority of Americans (61 percent) said that the government is doing a bad job at confronting poverty. In this environment, what can government leaders—especially a new president—do to make progress on solving important social problems?

One way to rebuild Americans’ faith in government is by focusing on “what works”—by finding and funding solutions backed by evidence and data. Early next year, a new administration will arrive in Washington, D.C. That administration should move to create a What Works Accelerator, a new national entity that will promote results-based decision making at all levels of government.

In 2008, I wrote an article for Stanford Social Innovation Review that advocated the creation of an office in the White House that would help build “a policy environment that fosters greater social innovation and investment in high-impact social entrepreneurial efforts.” Later that year, I joined the transition team of President Barack Obama and in that role helped launch the White House Office of Social Innovation and Civic Participation, as well as the Social Innovation Fund (SIF), an investment vehicle run by the Corporation for National and Community Service. Partly through these efforts, the Obama administration has built a strong foundation for “what works” policymaking.

In his first inaugural address, President Obama promised to restore trust between Americans and their government by subjecting policies and programs to closer scrutiny. “The question we ask today is not whether our government is too big or too small, but whether it works. … Where the answer is yes, we intend to move forward. Where the answer is no, programs will end,” he said. To meet that promise, his administration created several evidence-based initiatives. Among these efforts are SIF and Investing in Innovation (i3), a grant program within the US Department of Education. Together, these two initiatives have funded more than 100 program evaluations, and those evaluations show that i3 and SIF grantees have generated strong results in areas such as boosting student achievement, workforce training, personal income savings, and childhood health.

And yet, as promising as these initiatives are, they remain one-off, “boutique” experiments. Evidence-based approaches have not, for the most part, spread broadly enough to change government practice in a fundamental way. Even when rigorous evidence is available, policymakers still rarely use it to inform their decision making. Instead, they base their decisions on factors such as instinct and intuition, pressures from constituent and special-interest groups, the need for compromise to advance legislation, and—most commonly—previous funding choices.

I continue to be surprised by the limited use of evidence-based practices in US policymaking. In April 2015, my organization—Results for America—partnered with Bloomberg Philanthropies to launch What Works Cities, a three-year initiative to help midsized American cities enhance their use of evidence and data. When we reviewed applications for the initiative, we found that only about one-quarter of cities in the initial applicant pool reported that they modify programs on the basis of evaluation results.

At the same time, the robust response to What Works Cities indicates that city officials are ready to embrace a new approach: In the first six weeks after we launched this initiative, more than 40 percent of eligible cities had applied to participate in it. In recent years, people throughout the public sector have been charting a course for improving the way that government invests in social change. Now it’s time to accelerate those efforts.

A New Architecture for Policy

A What Works Accelerator, housed in the Office of Management and Budget (and thus part of the Executive Office of the President), would build on the achievements of the past eight years. Yet it would move beyond the use of experiments and pilot projects. In effect, it would create a new architecture for policymaking—one that transforms how policymakers direct public resources. Its mandate would be to minimize the chance that officials will base decisions on politics and to maximize the chance that they will base decisions on evidence of what works.

The accelerator would help lead a government-wide effort to use public data to analyze the impact of various government investments. It would use technology to reduce the costs of program evaluation. And it would give policymakers at all levels of government the tools that they need to pursue evidence-based change. The What Works Accelerator would have three key elements.

Support for federal agencies | The accelerator would enlist and enable all parts of the federal government to build the infrastructure necessary to use evidence and data in budget, policy, and management decisions. According to the Federal Invest in What Works Index—an assessment issued annually by Results for America—there is a set of leading federal agencies that have gone roughly 75 percent of the way toward putting this infrastructure in place. The accelerator would continue that process by taking these steps:

  • Build evidence of what works by encouraging agencies to allot a portion of their funding for evaluation efforts. At Results for America, we advocate setting aside 1 percent of program funds for this purpose. A few agencies have hit this bar. In the 2015 fiscal year, for instance, the US Agency for International Development (USAID) spent about 1.1 percent of its program budget on evaluation. But large agencies have been slow to move in this direction. The US Department of Housing and Urban Development, for example, has a budget for research, evaluation, technical assistance, and technology that equals just 0.19 percent of the $45.5 billion that it receives for program funding.
  • Use that evidence to make funding and policy decisions. USAID, for example, has a policy that sets forth quality standards for research, and it has guidelines that require the use of evidence and data to assess its various country strategies. At the Department of Education, similarly, officials have developed evidence standards for its competitive grant programs and used those standards to steer dollars toward more proven solutions.
  • Repurpose funds to direct federal dollars away from low-performing programs and toward programs that deliver positive results. The Head Start Reauthorization Act of 2007 includes a mechanism of this kind. That legislation makes all Head Start grants renewable but requires that the lowest performing 10 percent of providers recompete for funding.

A focus on opportunity | The accelerator would design and help launch a Social Mobility Challenge Corporation. Economic inequality plays a major role in driving the anger and distrust that many Americans feel today, and it has become one of the most pressing issues of our time. This new entity would address that issue by supporting evidence-based programs at the state or local level that promise to increase mobility and opportunity for low income individuals and families. One model for the Social Mobility Challenge Corporation would be the Millennium Challenge Corporation. This agency, created with bipartisan support under President George W. Bush, funds efforts to fight global poverty, and it gives priority to evidence-based approaches. A second model would be the Center for Economic Opportunity (CEO), an office in the government of New York City. Created in 2006 under Mayor Michael Bloomberg, CEO tests poverty-fighting interventions, helped design and implement those interventions, and conducts evaluations to judge their effectiveness.

Promotion of learning | The What Works Accelerator would work with federal agencies to help states and local governments—as well as nonprofits, social enterprises, and research institutions—to form partnerships to strengthen the use of evidence and data. Cities and states generate large amounts of data but often lack the resources to use those data effectively. Nonprofits and social enterprises, meanwhile, need access to government data in order to test and improve their solutions, and researchers need access to such data for their work as well. By partnering to share data about what works, and by using new tools for evaluating impact, these groups can reduce the lag time between program development and program adoption. The accelerator would support this process by spreading ideas about effective policies and practices.

A Foundation for Progress

Despite partisan divisions, the under pinning of a commitment to creating a What Works Accelerator is already in place. In late 2015, a large bipartisan majority in the US Congress passed the Every Student Succeeds Act (ESSA). Among other provisions, ESSA will shift more than $2 billion in federal education funding toward evidence-based efforts in each of the next four years. Congress also recently passed a measure—championed by Speaker of the House Paul Ryan, R-Wis., and US Sen. Patty Murray, D-Wash.—that created a new Evidence-Based Policymaking Commission. And in both the Senate and the House of Representatives, legislators from both parties have collaborated to introduce bills that would create a fund at the US Department of the Treasury to support pay-for-success projects—projects that entail the use of evidence and data.

The next administration will have an opportunity to channel this bipartisan energy into the launch of a new vehicle for results-based policy. A What Works Accelerator will not restore Americans’ confidence in government overnight, but by helping officials to address problems more effectively, it will move public sentiment in the right direction. It also will transform how policymakers at all levels envision their role in driving social change