COD Aid: Can It Be Evaluated? Compared to Traditional Aid? Systematically?

Cash on Delivery Aid has gone viral in the blogosphere.  (For a sample see a few of my recent posts.)In this post Maurits van der Veen of AidData proposes a systematic comparison of a COD Aid project to a traditional one with similar goals.  That’s a good idea except that it would require measuring outcomes in the traditional one, and then it would not anymore be traditional!  Still the fundamental point is a good one.  COD Aid programs need to be evaluated. In Chapter 5 of the Cash on Delivery book, Bill Savedoff and I propose a protocol for research and “process evaluation” of COD Aid that we hope provides a starting point for evaluating and learning from the approach.Van der Veen also suggests doing a survey to see whether citizens in the donor and recipient countries are better able to understand a COD Aid project than a traditional one.  We have been assuming that COD Aid would be easier than traditional aid for donor country taxpayers to understand or assess; after all the outcomes are measured and verified as a basis for the transfer of taxpayer money. Taxpayers know exactly what they paid for and how much. And the New York Times’ Tina Rosenberg apparently agreed.  My guess is that for citizens in recipient countries it is also easier to understand and assess COD Aid, since COD Aid involves an annual report of progress on just one single indicator with a pre-announced payment for each unit of progress (for each reduction in infant mortality, each increase in average test scores of schoolchildren, each decrement inthe deforestation rate).  Surely that is easier to assess than the multiple objectives, inputs, indicators, and policy steps associated with typical project-based aid and with typical program aid and budget support.But the key hypothesis worth testing has to do with whose performance citizens in recipient countries in the end judge: the donor’s or their own government’s? Our idea is that COD Aid will encourage citizens to focus on their own government’s performance and not mostly or solely on the donor’s performance. That requires sufficient transparency about the annual payments the government can and does get per unit of outcome, as well as annual publication of the reported outcomes, and the report of the outside verifiers.  (Publication of the contract, the annual government reports, and the verification are the sole conditions of the contract we propose.) Making recipient governments accountable to their own citizens instead of outside donors seems like a good step in the direction of broader systemic change and social transformation. It may be the best way to ensure aid that pays for better lives in the short-term helps trigger systemic and institutional change over the long-term.