Can Government Play Moneyball?

David Bornstein in the New York Times: “…For all the attention it’s getting inside the administration, evidence-based policy-making seems unlikely to become a headline grabber; it lacks emotional appeal. But it does have intellectual heft. And one group that has been doing creative work to give the message broader appeal is Results for America, which has produced useful teaching aids under the banner “Moneyball for Government,” building on the popularity of the book and movie about Billy Beane’s Oakland A’s, and the rise of data-driven decision making in major league baseball. (Watch their video explainers here and here.)
Results for America works closely with leaders across political parties and social sectors, to build awareness about evidence-based policy making — drawing attention to key areas where government could dramatically improve people’s lives by augmenting well-tested models. They are also chronicling efforts by local governments around the country, to show how an emerging group of “Geek Cities,” including Baltimore, Denver, Miami, New York, Providence and San Antonio, are using data and evidence to drive improvements in various areas of social policy like education, youth development and employment.
“It seems like common sense to use evidence about what works to get better results,” said Michele Jolin, Results for America’s managing partner. “How could anyone be against it? But the way our system is set up, there are so many loud voices pushing to have dollars spent and policy shaped in the way that works for them. There has been no organized constituency for things that work.”
“The debate in Washington is usually about the quantity of resources,” said David Medina, a partner in Results for America. “We’re trying to bring it back to talking about quality.”
Not everyone will find this change appealing. “When you have a longstanding social service policy, there’s going to be a network of [people and groups] who are organized to keep that money flowing regardless of whether evidence suggests it’s warranted,” said Daniel Stid. “People in social services don’t like to think they’re behaving like other organized interests — like dairy farmers or mortgage brokers — but it leads to tremendous inertia in public policy.”
Beyond the politics, there are practical obstacles to overcome, too. Federal agencies lack sufficient budgets for evaluation or a common definition for what constitutes rigorous evidence. (Any lobbyist can walk into a legislator’s office and claim to have solid data to support an argument.) Up-to-date evidence also needs to be packaged in accessible ways and made available on a timely basis, so it can be used to improve programs, rather than to threaten them. Governments need to build regular evaluations into everything they do — not just conduct big, expensive studies every 10 years or so.
That means developing new ways to conduct quick and inexpensive randomized studies using data that is readily available, said Haskins, who is investigating this approach. “We should be running 10,000 evaluations a year, like they do in medicine.” That’s the only way to produce the rapid trial-and-error learning needed to drive iterative program improvements, he added. (I reported on a similar effort being undertaken by the Coalition for Evidence-Based Policy.)
Results for America has developed a scorecard to rank federal departments about how prepared they are to produce or incorporate evidence in their programs. It looks at whether a department has an office and a leader with the authority and budget to evaluate its programs. It asks: Does it make its data accessible to the public? Does it compile standards about what works and share them widely? Does it spend at least 1 percent of its budget evaluating its programs? And — most important — does it incorporate evidence in its big grant programs? For now, the Department of Education gets the top score.
The stakes are high. In 2011, for example, the Obama administration launched a process to reform Head Start, doing things like spreading best practices and forcing the worst programs to improve or lose their funding. This February, for the third time, the government released a list of Head Start providers (103 out of about 1,600) who will have to recompete for federal funding because of performance problems. That list represents tens of thousands of preschoolers, many of whom are missing out on the education they need to succeed in kindergarten — and life.
Improving flagship programs like Head Start, and others, is not just vital for the families they serve; it’s vital to restore trust in government. “I am a card-carrying member of the Republican Party and I want us to be governed well,” said Robert Shea, who pushed for better program evaluations as associate director of the Office of Management and Budget during the Bush administration, and continues to focus on this issue as chairman of the National Academy of Public Administration. “This is the most promising thing I know of to get us closer to that goal.”
“This idea has the prospect of uniting Democrats and Republicans,” said Haskins. “But it will involve a broad cultural change. It has to get down to the program administrators, board members and local staff throughout the country — so they know that evaluation is crucial to their operations.”
“There’s a deep mistrust of government and a belief that problems can’t be solved,” said Michele Jolin. “This movement will lead to better outcomes — and it will help people regain confidence in their public officials by creating a more effective, more credible way for policy choices to be made.”