After the success of the first Development Marketplace at the World Bank in early 2000, Mari and I began sketching out additional competitions to extend the idea of creating a real marketplace for development.
One concept was to give vouchers to government officials and allow them to "shop" among various Bank teams when the officials came to Washington each year for the annual meetings. The idea was to create a small window outside the usual heavy and bureaucratic planning process by which the Bank's billions of dollars of funding were allocated each year. This new window would enable Bank teams to more directly gauge demand for their products and also to experiment with innovative new ideas that would normally not survive the formal planning process. Over time, we might even allow teams from other aid agencies to compete in this process as well to ensure that Bank teams were exposed to even more new ideas and competitive pressure.
Another concept was to create a small window allowing government officials to pitch innovative ideas to the Bank each time they came to town for the annual meetings. We would set aside maybe $50 million to fund ideas that might not normally make it through the Bank's analysis, design and approval processes. Separately, Bill Easterly proposed that the Bank pilot a program distributing vouchers directly to beneficiaries in select countries, allowing them to choose which projects they wanted.
In the end, we decided that the Bank and other official aid agencies were not yet ready for a marketplace. We felt we could have the most impact by leaving the Bank to create GlobalGiving, a neutral marketplace for community-led development projects around the world. Our goal was to show it could work, and then see if the idea could get adopted more broadly - including eventually by official aid agencies. We started modestly, and decided that actions speak louder than words, so we did not write or speak much about it in the early years (though we did write a chapter on this in the book Reinventing Foreign Aid).
Over the last nine years, we have been gradually implementing the elements of a real marketplace - including open access, transparency, and feedback loops. We have been gratified to see the emergence of many similar marketplaces for development aid and philanthropy. Some have failed to get traction, but a core group has survived and grown. Marketplaces similar to GlobalGiving include DonorsChoose (for education in the US), and Kiva (for microcredit worldwide) and GiveIndia. Important marketplace services such as Guidestar and NetworkForGood are helping create the infrastructure backbone for the market. And there are promising new entrants such as GreatNonProfits and Keystone (and many more).
Much has been written in the press about these various platforms, but few pieces discuss in depth the theory behind them. And, during this time, official agencies have been slow to innovate in this area. In that context, I highly recommend a new paper Beyond Planning: Markets and Networks for Better Aid by Owen Barder, a fellow at CGD. It is an excellent theoretical discussion of the issues and challenges facing official aid agencies, which are waning in influence.
Barden argues that improvements in the official aid system are likely to be the result of evolution rather than intelligent design: "Reform should not focus on a grand new design...but on a set of technical and apparently innocuous reforms which, over time, create strong political pressures for evolutionary improvements in the aid system."
This is a fundamental insight (which incidentally also applies to how the agencies should help stimulate reform in developing countries). There have been a huge number of "grand design" papers written over the years about how to reform the World Bank and IMF. But few of them have had much impact. Barder's approach is much more likely to bear fruit.
But it does raise the question of exactly which "technical and apparently innocuous" reforms to implement. What could we do that is most catalytic?
My own guess is that the greatest impact would come from empowering beneficiaries to have a much more direct say in what they want and how well projects are being run. Any aid agency, local government or project manager would be hard pressed to avoid responding to the voice of the people they are supposed to be helping.
As Owen says, "A particular challenge for aid is that there is a broken 'feedback loop' connecting the intended beneficiaries and decision makers." Recently at GlobalGiving we have started piloting a mechanism whereby beneficiaries in the field can provide feedback on project implementation, and we have demonstrated a rough prototype of a system that would allow beneficiaries to vote ex-ante on the types of projects they most need. I believe creating this feedback loop can fundamentally change the incentives in the aid system and create strong pressure for change.