Making aid payments conditional on results has intuitive appeal and is an idea that has been given a lot of attention in aid-world in recent years, with the Center for Global Development leading the charge. The tide now seems to have turned, with early enthusiasm replaced by doubts and questions about why, as Duncan Green puts it, there has been so much hype about the use of Payment by Results.
The case against Payment by Results is: first, that the evidence that it produces results is patchy at best; second, that Payment by Results is poorly suited to supporting change in complex systems (also known as, actually existing society); and third, that Payment by Results is motivated by donors’ wish to manage risks and strengthen upward accountability.
I’ve had mixed feelings about Duncan’s post. My initial reaction was positive, glad that the warnings about Cash-on-Delivery Aid that Paolo de Renzio and Ngaire Woods outlined several years ago, particularly around the possible incentive effective of paying by results, were now being heeded. But that feeling quickly dissipated to be replaced by a concern that the criticism of Payment by Results was too black and white; either it (as if there is one way of doing Payment by Results) works or it doesn’t. Development and development assistance is much more shades of grey.
The results agenda, in its various forms, isn’t going away. That’s a good thing. Neither is donors’ need to manage their risks. That is the political reality in donor countries right now. Improving the effectiveness of aid requires taking account of the political constraints to doing things differently, in donor countries as elsewhere. Payment by Results takes account of the political constraints, and, done right, could provide governments with flexible funding that supports their efforts to try, learn and adapt their way to solutions that work in their context. And in some cases – see the comments on Duncan’s blog, for instance from Michael O’Donnell, Jake Allen, Donald Menzies and Florian Schatz – that does seem to be happening.
Duncan’s blogpost provides a potential way forward when he poses the key question: to what extent are the results being paid for any use for learning and improving, or for increasing downward accountability? The challenge then is to design support that is conditional on results and which:
- has the flexibility needed to enable and allow for adaptive experimentation and learning, and the emergence of locally-owned approaches that are effective at solving problems and delivering results in particular contexts; and
- values and incentivizes learning outcomes and improvements in the capacity of country systems for learning, for downward accountability, and for adaptation, as well as about how many kids are vaccinated.
This is a challenge, but it’s a challenge that a number of donors are grappling with, including USAID through its Learning Lab (Collaborate. Learn. Adapt. For Better Development Results).
So, let’s not treat PDIA and Payment by Results, or positive deviance and Payment by Results, as mutually exclusive (see Chris Underwood’s blogpost, welcoming the “take-down” of Payment by Results). And let’s not assume that a focus on results is incompatible with an approach that aims to provide flexibility and space for countries to try, learn and adapt their way toward solutions that work in their context. Instead, let’s experiment and explore whether and how there are ways of putting Payment by Results into practice in ways that allow for and encourage the adaptive learning that is needed to deliver results.
Taking context and complexity seriously, and making the case for doing development differently, must involve engaging with the results agenda, injecting complexity into our understanding of results as Dave Algoso puts it, and shifting the focus to results that are useful for learning and downward accountability.
Thanks to Dave Algoso, Kathy Bain, Owen Barder, Leni Wild and Michael Moses for comments.