The government had been shut for more than three weeks when President Donald Trump decided to strike a blow for sensible policymaking. (Go figure.) On Jan. 14, he signed the Foundations for Evidence-Based Policymaking Act, legislation that aims to improve data analysis and program evaluation.
And good for him! But much more needs to be done to measure the effectiveness of government programs.
The law, which passed the Senate unanimously and the House of Representatives by a vote of 356 to 17, requires each government agency to appoint a chief evaluation officer as well as an official with statistical expertise to assist colleagues. It also pushes agencies to provide more data to the public and to do so in machine-readable format.
Many of these ideas were proposed in “Moneyball for Government,” a book I co-edited with Jim Nussle, who ran the Office of Management and Budget under President George W. Bush. As we noted in that book, too much of what the federal government does is not informed by empirical evidence. Good progress toward rigorous evaluation was made under President Barack Obama, as Greg Margolis and Ron Haskins of the Brookings Institution argue in their book, “Show Me the Evidence: Obama’s Fight for Rigor and Results in Social Policy.” But there’s a long way to go.
For example, 10 large decades-old social programs, which together cost more than $10 billion a year, were subjected to randomized controlled trials, the highest standard of evaluation. Nine of them were found to have had “weak or no positive effects” on their participants. Many programs are simply not evaluated at all. The Government Accountability Office recently reported:
In a 2017 government-wide survey, GAO found that most federal managers lack recent evaluations of their programs. Forty percent reported that an evaluation had been completed within the past five years of any program, operation, or project they were involved in. Another 39 percent of managers reported that they did not know if an evaluation had been completed, and 18 percent reported having none.
It makes no sense from a good-government standpoint that more than half of federal managers have not had their programs evaluated within the past five years or were unaware of the evaluation if it had been conducted. As Harvard University economist Jeffrey Liebman has argued, “Spending a few hundred million dollars more a year on evaluations could save tens of billions of dollars by teaching us which programs work and generating lessons to improve programs that don’t.”
For this reason, I support a proposal from Results for America, a nonprofit organization that promotes evidence-based decision-making, to reserve 1 percent of program spending for evaluation. In other words, out of every $100 spent, the government should dedicate $1 to making sure the other $99 actually works.
As the cost of obtaining data and analysis continue to decline, Washington should be leading the effort to design and carry out cutting-edge program evaluation. The Foundations for Evidence-Based Policymaking Act represents a surprising moment of success after years of dogged effort by Results for America, in particular.
Now Americans need to build on this glimmer of hope with a broader commitment to making sure that taxpayer dollars are effectively spent. Perhaps all the legislators who voted for the new law would also favor requiring government departments to conduct more evaluations so those chief evaluation officers have more to work with?