Innovation in public services must be Darwinian

How do you tear down an oppressive system? Creatively. Another name for innovation is creative destruction. By pioneering a better way, innovators can undermine what existed before.

Peeling apart the concepts of policy innovation (trying something new) and program evaluation (rigorous tests of what works) helps us see how they operate differently and serve different purposes. Policy innovators try something new, while program evaluators legitimize existing practices.

For anybody with extreme views of a public service—like abolishing child protective services—the goal should be to create such compelling alternatives that program evaluators cannot ignore them. Evaluation should usually be saved for later, for reasons discussed below.

Evidence-based Practices

Government grants for privately-operated public-serving organizations typically require those organizations to use evidence-based practices. The evidence is compiled by respected research consultancies like Rand Corporation, the Urban Institute, Chapin Hall, Child Trends, Abt Associates, Mathematica, MDRC, or J-PAL.

This practice makes sense. Rule-of-law requires that public grantmaking be transparent, rules-based, and somewhat blunt. Thus, requiring grantees to pick from a menu of accepted practices is a way to ensure nothing too whacky gets approved.

However, the devil is in the details of any operation. Evidence-based practices are too high-level to ensure grantees are uniformly effective. Also, generating evidence-based practices is a slow, expensive, and blunt process. It is no way to innovate. Evaluation studies legitimize established practices; but they don’t push the bleeding edge forward.

How Innovation Works

In the for-profit sector, formal evaluation studies practically don’t exist. Entrepreneurs try a business and it either succeeds or fails. Private investors judge a business on a host of factors, both concrete and implicit. “Fairness” is irrelevant; they only want winners. Successful models emerge via natural selection.

For critics of the status quo in human services, more innovation is what’s needed. Large literatures exist about how to foster innovation in any field. Some of the high points are:

  • Easy entry into market, not blocked by a monopoly.
  • Low startup costs
  • Autonomy to try new things, not blocked by rules.
  • Sink or swim, so bad ideas die out.
  • Availability of investment capital and skilled workers.
  • A community of practice to learn from other innovators.

Innovation emerges when upstarts have the freedom, motivation, and ability to pursue whatever methods they think will succeed. Many or most of these upstarts are wrong—but the few whose ideas are right will push the frontier forward.

This is not a novel argument. Private philanthropies in human services already play the role of discerning investors, trying to pick innovative winners. These include the Annie E. Casey Foundation, Gates Foundation, Ford Foundation, and Walton Family Foundation. (Not coincidentally, these foundations are named after the founders of successful private businesses: UPS, Microsoft, Ford Motor Company, and Walmart.)

Sometimes, policy innovations are pre-designed alongside rigorous evaluation. Doing this can sometimes make sense. If the innovation is particularly expensive (e.g., basic income guarantee programs) you may only get one shot. But doing this constrains the innovator. Everything becomes less flexible, harder to adjust midstream, more expensive, and more difficult to plan.

Equipping Innovators

Policy entrepreneurs are everywhere. Every non-profit leader wishes they could try a new approach. But they are bound by the same endemic problem—they’re spending other people’s money. This agency problem is like the original sin of innovation in policymaking. It’s too hard to try risky approaches when everything must be justified to a funder.

For an exception that proves this rule, see how far billionaire Charlie Munger got in building a dormitory at University of California Santa Barbara with no windows…because it was his money.

Igniting non-profit innovation requires the same conditions needed in for-profit innovation, a few of which are listed above. How can it be cheaper to get started? How can the entrepreneurs gain more autonomy by spending their own money? How do you attract more venture funding? Are there regulations blocking innovation? How can the entrepreneurs form community to learn from each other?