Research follows one rhythm, policy another. Researchers arrive at study questions, design fieldwork, collect and analyze new data, figure out what it means, write up technical articles, and then – and only then – may translate their new knowledge into policy recommendations. Sometimes they have to get funding along the way, which slows the whole business down even more. Months melt into years.
Policymakers, on the other hand, act with alacrity, responding to problems they observe or to pressure from constituents regardless of whether there are facts to draw on. Simply because of the need for speed, “get it done” dominates “get it right.”
Or so we seem to like to believe.
One of the most common excuses for the failure of policymakers to use high quality evidence is that they can’t wait for, say, the cost-effectiveness analysis of a job training program or the randomized controlled trial of a new intervention to keep girls in secondary school. But I think that assumption needs to be challenged. With the evidence.
Exhibit A: In some domains, policy waits (a long time) for analysis.
In health and medicine, certainly, we are all accustomed to the fact that rigorous testing is needed before promising interventions are widely available. From the start of clinical testing of a new medicine, it’s typically a dozen years until a product appears on pharmacy shelves – and many drugs that start clinical trials don’t demonstrate safety and efficacy, so never get approved.
But beyond that obvious example, research and development lead time is long when governments are creating new weapons systems, and when they undertake feasibility studies and design work for bridges, roads and dams. The governors of Central Banks look at long trends in inflation and investment before moving interest rates by even a quarter of one percent. And in many other areas of public policy, from environmental regulations to curriculum reform, decision makers don’t just permit analytic work to be done and digested; they require it.
Exhibit B: Policymaking can be agonizingly slow.
Politicians and agency heads may like to think of themselves as acting with urgency, but – let’s be honest here – they often do not. To take one example: For at least 15 years, critics of food aid have pushed for reform of a system that does some good, but often inefficiently and in ways that distort local food markets in recipient countries.
Policy analysts have systematically documented the problems, and proposed several alternatives, highlighting the costs and benefits of each. For at least 10 years, the U.S. Congress has debated food aid reform, and the current Administration has proposed modest changes in the way food is purchased and transported. However, major actions – the actions for which evidence has been amassed – have yet to be taken. In this area and in many others, policymaking is, in fact, far slower than the research. It is not until the weight of the evidence is so strong, and the advocacy around it sufficiently powerful, that decision makers are able to overcome inertia.
Exhibit C: There are enduring questions in many fields.
Is it better to pay health care workers on the basis of services rendered or outcomes achieved? How can welfare and other social protection programs for the poor be well targeted without creating burdensome paperwork or stigmatizing recipients? Why do people in need refuse free services? What are the ways to accelerate changes in harmful social norms, like those that underlie race or gender discrimination?
These are questions that can be answered, in part, through thoughtful empirical research, generating a body of evidence about how people live and how they respond to different services or incentives – empirical research that can provide insights when specific policy initiatives arise. Investments in these studies don’t pay off immediately, but they can contribute to better policy over a long time horizon.
For these reasons, I’ve stopped giving much weight to the shibboleth that the slow pace of research diminishes its value for policymaking. That’s the good news. The bad news is that approaches touting “rapid” research may not be the solution at all. We’ll have to think harder about the real constraints to the use of evidence, and work harder to solve them.
