'Paying for Results' at a tipping point?

It is fascinating to watch new ideas spread. Researchers who study the diffusion of innovative ideas and practices refer to an "S curve" of adoption (first studied literally decades ago in looking at how the use of new high-yield seeds spread among farmers).

At first, only a few venturesome and bold people try something new. Maybe then a few others join them. Many innovations never get beyond these early adopters, and die shortly after birth. But for others, adoption continues, with the use-curve still sloping gently upward over time, still limited to fairly venturesome people willing to go a bit out on a limb, but not quite so bold as the first adopters. Then at some point -- which researchers call an "inflection point," "tipping point" or "take-off point" -- use starts to skyrocket. People start adopting a change, not because they are particularly bold, or even particularly convinced about the change on its merits, but because they see so many others doing the same thing. Soon, a change has taken over.

I am wondering whether we are approaching a tipping point in the interest around government in the idea often known as "pay for results." The pay-for-results mantra is straightforward and (in my view) appealing: If government wants to get another party to do something, stop paying them just for effort, or for producing stuff that doesn't necessarily accomplish an outcome. Don't pay job-training providers for providing training courses, pay them for actually getting people jobs.

Not surprisingly, this idea began in the arena of government procurement -- the classic (and still by far the most important) way government tries to induce private parties to promote government purposes.

In some sense, the idea is as ancient as what in contracting is called a "completion contract," where a vendor is not paid until the product (be it a pencil or a supercomputer) is actually delivered. But the version closer to today's idea of paying for results began to appear 30 years ago under the rubric of "performance-based contracting" (PBC), especially for services.

PBC included a bunch of ideas we still associate with paying for results -- tell the contractor the result you want, let them decide how to do the work, and link payment (including rewards and penalties) to the outcomes. The George Bush administration worked on this in the late 1980s, and it was one of the main issues I personally spent time on as administrator of the Office of Federal Procurement Policy in the 1990s. At the time, we felt very much like those bold early adopters.

Fast forward to 2014. I started thinking about the S curve of innovation diffusion as I opened a new report published by Georgetown University's Beeck Center for Social Impact and Innovation called Funding for Results. The report discusses five efforts in three countries using different techniques -- such as contracting and performance-tied national government grants to local governments -- that all fall under the broad "paying for results" rubric. The report doesn't discuss some other approaches that blog readers have read about here over the years, like procurement contests and social impact bonds, but the emphasis on outcomes is a common theme.

I see two signs we may be reaching a take-off point for this kind of innovation. One is the sheer number of different forms the basic idea takes, and the growing number of disparate jurisdictions where it is being tried. A second -- which we see very much in the Georgetown report -- is the beginning of efforts to generalize from experiences so far to derive lessons learned. (This report has a number of them, such as to keep performance measures for these arrangements few and simple, and to pair increased responsibility for results with a lowering of process requirements the producer must provide the payer.)

If we're indeed nearing a take-off point, that's good news for government and the public.