Mar 06, 2009

Stimulus Funding for Effectiveness Research Worth Following

by Laura Starita

In times of limited resources, it is more important than ever to be confident that money spent is producing desired results. President Obama apparently thinks so, since more than a billion dollars of the economic stimulus bill has been designated for researchers to do comparative studies on the effectiveness of certain medical treatments, in order to ascertain which approaches produce the best results. Reform of the US health care system depends on getting more bang for the buck—the idea behind the funding is to produce evidence for doctors and patients on what treatments actually work (and their relative cost-effectiveness). It’s no big stretch to see great hope (sorry) for identifying inexpensive approaches—ice and physical therapy for lumbar pain, for example—that often produce results that are just as good as expensive, specialized, and risky interventions—such as invasive surgery.

Unfortunately, aid agencies and nonprofits are not getting their own billion dollar investment in comparative effectiveness research, but if ever there was a time for private philanthropy to follow government’s lead, this is it. In lean times, the tendency is to cut back on everything possible in order to keep programs running. That instinct is self-defeating to the best interests of both the organization and its beneficiaries, however. There are plenty of things we don’t know about what works to alleviate social ills. When money is dear, we want to be especially sure it is doing the most good: Should funds go toward the free distribution of bed nets (thereby ensuring maximum access), or should we simply lower the user’s cost of buying (thereby stretching dollars further and ultimately making more bed nets available) ? Should disadvantaged kids be given cash as a reward for school performance?

The good news is that many of these questions are of interest to academics as well, and both of the questions noted above are being tackled by scholars. We have written before about conditional cash transfers;they have received much attention in the past year, particularly as they have been applied in Mexico as incentives for adults to attend professional training or send their children to school. Mexico’s program, called Oportunidades, was developed by Mexican economist Santiago Levy while he was working in the finance ministry, and is indebted to behavioral economics ideas about choice. In brief, people often say they want to do things that are in their own best interest—such as go back to school, or lose weight—but when presented with a choice to take a job or eat a cookie, many sacrifice the long-term goal for short-term benefit. Conditional cash transfers are geared around switching the incentives, so that there is a short term gain for the action with long-term benefit; sending kids to school instead of putting them to work in the fields, for example. Empirical evidence on CCTs in Mexico show that the children of CCT-recipient families have better school attendance and higher immunization rates.

Those results have inspired some in the United States, New York City for example, to implement CCTs both broadly, for impoverished families, and granularly, such as cash for school performance. But the ‘cash for performance’ programs raise alarms for some psychologists, reports the Times, who say that paying kids to do well in school replaces internal incentives with external ones. This might work for people who have no internal motivation to do well. But for those who do intrinsically want to succeed, paying them to do it effectively kills the intrinsic desire that will need to be in place for a lifetime. And what happens if program participants don’t meet the program requirements for a pay-out? And what does it do to the motivation of kids who attend the same schools but don’t qualify for the program? Many reform advocates argue that we need to try anything to improve school performance among disadvantaged youth. But it is not clear that paying actually helps, or that any gain outweighs the losses.

Questions such as these abound in nearly every social area, from education to health to economic development. Behavioral economics is also at work on donors and funders as they make choices about what to invest in. The impulse—as with that cookie—is to earmark money directly for recipients, because the gain seems immediate. But absent any evidence that programs work, it is a false gain. Instead, we should remember to support the research projects that can tell us for sure whether the gain is sustaining or not, and be willing to act on the evidence, even if we don’t like it.