Skip to content

Evidence-Based Policy: Prove It


One of the most curious aspects of politics is our refusal to see reality, to demand policies regardless of how well they work or even whether they work at all. Why do we damage ourselves with counterproductive approaches?

I like to start with the big questions. Why do we have a government at all? When you get right down to it, governments exist to solve problems that cannot be effectively addressed at a lower level. If two families cannot resolve their differences, the city government will. When a city is in trouble or multiple cities are in conflict, the state can step in. When states are caught off-guard by disasters or face other problems they lack the resources to solve, the federal government gets involved. And problems involving multiple countries are negotiated to solutions through treaties and international organizations.

That is, of course, the ideal. Each tier should be equipped to tackle issues facing the tiers below. It doesn’t always work that way.

For one thing, determining just what is a problem can be a matter for debate. More precisely, whether a given problem is something the government should be involved in solving can be subject to considerable disagreement. A conservative, for instance, may see poverty as unfortunate, but not an issue that the government should put forth any effort (especially monetary) to correct. A liberal would probably think the opposite: government should jump in and do something about it.

Who’s right? No prize for guessing what I think, but that’s not the point. Let’s say we can agree that something is a problem and that the government must be involved in the solution. What then? How do we determine the best solution?

Experimentation can offer some answers. By this, I don’t mean putting needy people in a lab and poking them to find out what they need and why. Instead, governments at different levels can be empowered to experiment with policies that are either currently nonexistent or are not functioning as well as would be desired. This actually happened in the US in the 1980s, when welfare-to-work programs were tried in various forms around the country, and produced results that surprised people across the political spectrum. According to the nonprofit MDRC:

Although the [welfare-to-work] programs did not help everyone, these findings (combined with those from studies begun by other organizations) had a profound impact on the political debate about welfare reform. Now both the right and the left had to argue within the bounds of the evidence. And the evidence challenged long-held beliefs of both groups. Conservatives learned that social programs could work and that the benefits could exceed the costs. Liberals learned that work mandates and requirements could produce positive effects, that children were not harmed by welfare-to-work programs, and that participants thought the programs were fair and preferred work to welfare. No longer could policy be based only on anecdote and ideology. After two decades of failed attempts to reform the welfare system, this new consensus — based on experience and evidence — led to passage of federal legislation, the 1988 Family Support Act. As a Congressional staffer explained at the time: _In all the years I worked on welfare reform, we never had a body of data that showed what worked…For the first time, we could characterize reform as an investment._ Importantly, while the results of experimentation told the nation what worked, it also showed what did _not_work, and what we still did not know. Remarkably, the 1988 welfare reform law included funding for a next generation of experiments that would use random assignment research methods.

It is a shame, then, that these lessons were discarded when Republicans retook Congress in the ‘90s and passed their welfare reform package in 1996, which dramatically reduced federal obligations in this area and put much stricter limits on who could get help, how much, and for how long. This welfare “reform” was not an evidence-based policy approach, but a partisan move undertaken in the name of slashing government deficits. It is tragic that only a few years later, that same Republican-controlled government found the money to massively expand our military budget in order to carry out multiple wars around the globe. The idea of experimenting to find effective welfare policies regardless of their partisan implications seemed to fall by the wayside as our existing welfare system had been gutted and our national priorities shifted.

The news is not all bad, though. The Affordable Care Act, passed in 2010, made evidence-based care a central feature. Medical providers would have to gather much more information about their patients in order to track their long-term progress as well as responses and adherence to treatment regimes. The ACA itself was not radical in terms of its policy prescriptions, either, even if some of its specific mechanisms are uniquely American. Near-universal health coverage improves health outcomes in every country where it is implemented–it’s a no-brainer with ample evidence behind it. That we piggybacked such coverage on the existing health insurance system, on the other hand, had no basis in evidence. Rather, it was viewed as politically possible in ways a public option or Medicare-for-all proposals were not–and transition dynamics can often override what would ultimately be the best practice.

The War on Drugs is another domain where we have, for decades, ignored what the evidence has told us, and are only recently beginning to take evidence-based approaches. Harm-reduction policies are all about what is known to work in terms of reducing disease transmission and drug-related deaths.

Good policy should not be about what aligns with our moral sensibilities, but what actually works. If it is the government’s duty to keep people safe, to keep them healthy, to help them when they are in need, then the policies implemented to carry out those duties should be based on the evidence for what actually achieves those goals. If the evidence doesn’t exist, gather it! We should not be afraid to experiment with new, radical policies, first in small-scale pilot programs, and then in broader programs once the most effective policies have been identified.

It’s also entirely possible that a policy which works well in rural Maine will work very poorly in southern California. This doesn’t mean the policy is bad, only that different areas with different populations will have different needs. This is ultimately the best argument for devolving some functions to the states: a state government is closer to the issues faced by its citizens than the federal government is likely to be. Solving problems should then be a combined effort, with the federal government providing funding and resources, allowing the state to use those tools as they see fit, with the federal government holding the state accountable for results. The ACA, it must be noted, has provisions for this sort of approach, as well. In this way, the federal government can ensure that overall duties to citizens are being served even without direct control over policies.

It is deeply unfortunate that our political environment tends to stress ideology and sentiment over data. Data cannot solve everything, either–I’m well aware of its limitations and am no technocrat, myself. But we should not allow research and evidence to be supplanted with gut feelings and moralizing. Our problems have solutions, if we’re willing to listen to what the evidence tells us.

Photo by KamiPhuc