Inside Obama's effort to use science — and find policies that actually work

A couple of years ago, former Obama and Bush officials estimated that only 1 percent of government spending is backed by any evidence at all. 1 percent. Perhaps unsurprisingly, then, evaluations of government-sponsored school and work programs have found that some three-quarters of those have no effect.

Over the last six years, a group of evidence-based nerds in the White House have been trying to change the status quo by way of a quiet evidence-based revolution. By only funding programs — from home visits to low-income families, to pregnancy prevention and K-12 education — that have high-quality studies demonstrating they work, the White House has been trying to push government officials in a more science-minded direction.

In his new book, Show Me the Evidence, Brookings Institution fellow and policy analyst Ron Haskins documents what is probably one of President Obama's most important and least-understood achievements.

I talked to Haskins about how the White House accomplished the feat of getting science into the policymaking process, what governments everywhere can learn from Obama, and whether the evidence-based approach will stick in Washington.

Julia Belluz: You’re a Republican and former George W. Bush White House staffer. Why praise President Obama?

Ron Haskins: We spend a trillion dollars a year on programs for poor and low-income people and we have very little idea about whether these programs are successful. Peter Orszag, the former director of the Office of Management and Budget (OMB), says that less than 1 percent of federal spending goes to programs and activities that we have any evidence that they work. It’s essential we learn a lot more about our programs. If they don’t work, we need to change them or end them, and spend the money elsewhere.

So Obama deserves a lot of credit here. Unfortunately Republicans are completely blinded by their rage against him and I’m afraid that they will end some of the Obama evidence-based programs.

programs that get money have to use evidence-based models. There has seldom been a requirement like that.

JB: How is what this administration is doing with research evidence different from what previous administrations have done?

RH: The Bush administration deserves credit for moving in the same direction on evidence-based policy as the Obama administration. But the difference in scale is the first difference, and the second difference is the approach: programs that get the money have to be using evidence-based models. There has seldom been a requirement like that to get federal dollars. Here and there, agencies have had high quality evaluations of their programs, but it’s still a very, very small minority of federal spending that there’s any attempt to make sure is being spent on programs that work.

JB: This isn't unique to the US. Using science to inform policy is something governments around the world struggle with. Why was Obama able to experiment so successfully?

a lot of these initiatives were hidden in major pieces of legislation.

RH: I think the most important thing is that OMB provided leadership. OMB worked in harmony with the agencies that ran the programs.

They planned the initiatives, the evidence requirements, and — another important characteristic — they spent new money, over $5 billion, on programs. [The six Obama administration evidence-based initiatives include teen pregnancy prevention, i3 investing in innovation fund, home visiting, the social innovation fund, community college and career training, and the workforce innovation fund.] So they didn’t have to take money away from other programs to pay for the initiatives.

They were also able to do that in part because a lot of these initiatives were hidden in major pieces of legislation like the Affordable Care Act. I would not call that a trick but it often happens that you can pass smaller bills in legislation that has a lot of moving parts.

Also, these six evidence-based initiatives got relatively little attention from congress. Only one of them — the home visiting initiative — had a hearing that brought it to the center of attention. So the administration was very clever. I could not document that the President intervened in Congressional passage of any of the initiatives, but his chief of staff Rahm Emanuel did, and so did several other senior officials from the White House and OMB.

JB: We’ve seen great progress reducing teen pregnancy across the country. Can you tell me about how the administration used evidence to create related programs and policies?

RH: The teen pregnancy rate has fallen more than 50 percent since 1991. It’s not exactly clear why. There’s really a methodological issue here: when you have hundreds and thousands of programs all over the country, plus media messages, and cultural changes, how do you separate out the effects of those things? But we have very good studies showing a number of these programs work for the relatively small group of kids who participate in them.

The administration started with two overall goals: to increase the number and share of federal dollars that go to evidence-based programs, and to increase the share of agencies using model programs that use rigorous designs to evaluate the programs.

So that means at some point they identified kids in the community — or in schools — and then randomly assigned them to experimental and control groups, followed the kids in both groups, and compared them on outcomes like age of initiation of sexual activity, frequency of sexual activity, and number of different partners.

It didn’t take until after this administration for Republicans to kill the workforce innovation fund.

JB: Which programs were most successful?

RH: Interestingly, they are not the ones that regale kids with things like, "Don’t get pregnant because you’ll burn in hell." They are the ones that involve the kids in community activities. They try to convince kids that pregnancy could have a damaging impact on them and the baby. They make a point of making sure the males understand it takes two to tango. So what works isn’t just lectures to use birth control or be careful. That’s part of the best programs but another essential part is involving the kids in constructive activities like community service and tutoring.

JB: What’s the post-Obama outlook for science in the White House?

RH: I’m worried about it. It didn’t take until after this administration is over for Republicans to kill the workforce innovation fund as part of the ‘cromnibus.’ They’re trying to kill i3. If that turns out to be the case, it’ll be the second of the six evidence-based initiatives that have been terminated by Congress. I’m worried that by ending the initiatives prematurely we’re going to learn less than we could. We should be expanding this approach, including to current grant programs to make sure people who get federal dollars have to show they are producing effects. This is called accountability.