Select Page


  • In our previous report, we described 40 years of stagnation in key areas of social policy— e.g., little change in earned income of low/moderate income Americans or in educational achievement of high school students.
  • Given the stagnation’s endurance and seeming immunity to myriad policy interventions, some may question whether even the exceptional, proven-effective programs that we cite as a potential solution can really change the trajectory—especially given their limited number.
  • However, we propose a fundamental reinvention of social spending so as to build and deploy proven-effective programs on a major, government-wide scale, rather than, as now, to do so at a snail’s pace by waiting for the occasional policymaker to recognize the value of evidence and decide to advance it in his or her area.
  • Such reform would be in the tradition of Franklin Roosevelt’s call for “bold, persistent experimentation” in government to identify strategies that work. We discuss how to systematically incorporate this concept into social spending.
  • Evidence-based medicine has produced amazing advances in human health over the past half-century. Similar progress is possible in social policy.

In our previous Straight Talk report, we described the great stagnation in key areas of social policy since the 1970s. To recap:

(i) The earned income of low and moderate income Americans is virtually unchanged over 40 years, after adjusting for inflation; and (ii) U.S. student achievement at the end of high school has been flat for 45 years, with more than a quarter now scoring below basic competency in reading or math. As a result, the American dream of rising social and economic well-being for all citizens, regardless of birth circumstance, is at great risk.


The policy response has been “government by guesswork”—i.e., launching of major new initiatives based on expert opinion or the like (rather than credible evidence). Guesswork usually fails because, as we know from the history of rigorous evaluations, surprisingly few program and policy initiatives produce the hoped-for effects on education, poverty, and other key outcomes. But rigorous evaluations have also identified exceptional social programs that do produce important improvements in people’s lives, showing success is possible.

Some observers may wonder if the situation is hopeless, given the stagnation’s endurance over four decades despite myriad policy initiatives aimed at increasing economic opportunity and improving U.S. schools. They may question whether even proven-effective programs such as those we cite can really change the trajectory—especially given their limited number.

However, in this report we call for a fundamental reinvention of government social spending to make evidence, rather than guesswork, the new defining principle in what gets funded. The goal would be to build and deploy proven-effective programs on an industrial scale (so to speak), rather than, as now, to do so at a snail’s pace by waiting for the occasional policymaker to recognize the value of evidence-based approaches and decide to advance them in his or her jurisdiction or program area.

In 1932, Presidential candidate Franklin Roosevelt, responding to a pervasive sense of helplessness about our country’s ability to reverse a prolonged economic collapse, called for “bold, persistent experimentation.” He said, “It is common sense to take a method and try it; if it fails, admit it frankly and try another.” We agree with this common sense. Adapting it for the 21st century, we call on Congress and the President to enact reforms across the broad array of federal social spending initiatives to incorporate the following three principles:

Principle #1: Innovation. State and local agencies that receive federal funds, and federal agencies that disburse the funds, should be strongly incentivized and encouraged to pilot innovative new approaches to addressing the funding’s targeted purpose (e.g., improving education, training workers, reducing the cost of healthcare, preventing opioid abuse, and so on). The incentives might include, for example, waivers from federal law/regulation or larger funding awards for entities proposing pilots with compelling logic or initial evidence to suggest they might produce a meaningful effect. Many, many, many such pilots would be launched, but each would be funded initially only on a small or modest scale.

Principle #2: Rigorous Testing. The federal government would require each pilot to be evaluated in a rigorous study—ultimately, a randomized controlled trial (RCT), where feasible—to determine whether  and under what conditions it produces the hoped-for improvements in participants’ lives.

Principle #3: Expansion of only the proven-effective approaches. For the subset of approaches and strategies found effective, the federal government would fund their expansion to improve the lives of many thousands or even millions of people nationally. The evidence threshold for such expansion would be high: Strong scientific evidence, replicated across multiple studies or sites, showing sizable effects on ultimate outcomes of policy importance.

In many areas, these evidence-based principles could be implemented not with new government funding, but by reforming existing billion-dollar spending initiatives (specifically, their authorizing statutes) to direct them to allocate funds according to these principles. In this way, evidence-based criteria would be integrated into the machinery of government funding decisions.

The specific means of incorporating these principles into government social spending initiatives would depend on the initiative. Large federal grant programs such as Head Start and Job Corps that seek to improve participants’ educational or workforce outcomes—and which, RCTs show, often do not produce the hoped-for effects—could phase in a tiered-evidence grantmaking approach as we’ve described in previous reports. Under this approach, the $1.7 billion Job Corps program for disadvantaged youth, for instance, would now fund a broad array of innovative pilots, coupled with rigorous testing, to identify program strategies that produce substantially better education, employment, and/or earnings outcomes than the standard program. The proven-effective strategies (and only the proven strategies) would then receive scale-up grants from Job Corps to serve large numbers of youth.

But the principles could also be incorporated into many other types of government spending initiatives, such as (i) Medicare and Medicaid, to build and scale proven-effective strategies for reducing healthcare costs while improving—or at least not compromising—patient health (such as the Transitional Care Model); (ii) the federal Violence Against Women Act programs, to build and scale effective approaches to preventing sexual assault and other violence (such as the EAAA program); and (iii) federal substance abuse prevention and treatment programs, to build and scale effective interventions in this area (such as Lifeskills Training and PROSPER). If Congress and the President were to demand “bold, persistent experimentation,” then with creative effort these and many other government initiatives could become engines of innovation, evidence-building, and scale-up of proven strategies that will finally move the needle on major national problems.

This reinvention of U.S. social spending would be analogous to the transformation of medicine over the past half-century.[i]  In 1962, Congress enacted legislation that, as implemented by the Food and Drug Administration (FDA), required the effectiveness of any new pharmaceutical drug to be demonstrated in high-quality RCTs before the FDA would approve it for marketing. This change in the drug licensing mechanism—analogous to the change in funding mechanism that we propose for social spending—created an enormous incentive for both the development and widespread use of proven-effective medical treatments.

It was a turning point in medicine.[ii] Since the early 1960s, RCTs required by the FDA or funded by the National Institutes of Health or Veterans Administration have produced the conclusive evidence of effectiveness behind most major medical advances, including vaccines for measles, hepatitis B, and rubella; interventions for hypertension and high cholesterol, which helped cut the incidence of coronary heart disease and stroke by more than 50 percent over the past half-century; and cancer treatments that have dramatically improved survival rates from leukemia, Hodgkin’s disease, breast cancer, and many other cancers.

In 1945, President Roosevelt died of a stroke caused by malignant hypertension. Earlier in his life, he was crippled by polio. Today, because of evidence-based medicine, both conditions are easily treatable or preventable, and it can truly be said that the average American has access to far better medical treatments than the President of the United States did 74 years ago.[iii]

This amazing transformation of medicine occurred despite the fact that, as in social spending, only a small fraction of potential medical treatments, when rigorously tested, are found to produce the hoped-for effects. That is exactly the reason why guessing at solutions is a dead end in both medicine and social spending. The alternative—Roosevelt’s “bold, persistent experimentation”—is the only viable path to progress. We propose it as the new standard operating procedure in social policy, so as to build and deploy a body of exceptional programs that will end 40 years of stagnation and ignite progress in education, economic opportunity, and many other areas.


[i] Baron, Jon 2018. A brief history of evidence-based policy. The Annals of the American Academy of Political and Social Science 678 (1): 40-50.

[ii] Chassin, Mark R. 1998. Is health care ready for six sigma quality? The Milbank Quarterly 76 (4): 574-576.

[iii] Gifford, Ray W. 1996. FDR and hypertension: If we’d only know then what we know now. Geriatrics 51 (1): 29–32.