The Second Wave of Independence: Shopping for Solutions - Working Paper

From time to time there is a paper that aims at the foundational dynamics of the development industry, this time it is Gaarder and Bartsch's "The Second Wave of Independence: Shopping for Solutions"
Abstract:     

"In the 21st century, many developing countries will become emerging markets and will no longer be in need of the carrot-and-stick approach to development assistance most prevalent today: development financing made available conditional on certain policies and interventions. This paper suggests that interactions between development agencies and recipient governments are mostly about inputs deemed (but not known) to contribute to improvements in living standards in recipient countries, rather than outcomes. The paper argues that the development marketplace is beset by market imperfections because of externalities, principal-agent problems, and decision making under uncertainty, which not only make it difficult to achieve the right outcomes, but also take away incentives to learn about outcomes. A fundamental rethink of responsibilities and accountabilities in the development business would make sure that development outcomes are traded in the development marketplace. It would put recipient countries in charge of contracting development agencies to provide these outcomes. Development agencies would commit to and be held financially accountable for outcomes, that is, real improvements in welfare indicators. The paper describes the role of the evaluation function in aligning incentives with the ultimate goal of improving lives and provides examples of emerging solutions."

 

The authors use an economic analysis of the interaction of development agencies, donor and recipient countries, identifying three key issues (market imperfections):

- Externalities. The focus on inputs (funds, resources) and outputs (stuff built, people trained, gadgets distributed) instead of outcomes (improved health, better access) creates a singular kind of externality, not as an unintended consequence but rather an acknowledged but not fully included in the transaction

- Principal-Agent. A classic issue in economics literature, the misalignment of interests between the contracting parties, compounded by the fact that they are entering the agreement on behalf of third parties (be that recipient's poor or donor's rich)

- Uncertainty. In many cases we don't have strong evidence that an specific intervention in a given context will bring the stated outcomes. Theories of change are full of assumptions of assumptions that are unable to address causal relationships, specially within the timelines that many development projects operate

The proposed solution is to internalize the externalities by a tendering process focused on outcomes. The recipient country would contract out the delivery of outcomes to the development agency. This would create an interesting dynamic in relations between development actors and recipient countries... however it requires some assumptions and does raise some questions. While the Principal-Agent issue is sorted in one part of the equation, it still remains that the recipient country in representing the final set of beneficiaries. It also requires an alignment of interest between donor and recipient, and the internal governance of the recipient... or rather, how is the contractee (development agency) is going to interact with different parts of the recipient country that may not have been party to the original negotiation. In an input-ouput model it is easier to deal with institutions as they can be co-opted with more tangible things.

Also the paper tends to make loans and grants equal, while the negotiation dynamics would clearly be different if repayment is on the table or not. I also have trouble defining as donor a provider of a loan, but that may be a personal issue.

Finally there is a call for more rigorous evaluation and improvement of M&E. On face value it is a winning formula, and it would be hard to argue against, but at the same time it feels like the common opt-out 'more research is required'. I would say the key issue is not necessarily M&E but rather to be clear (and humble) on both the intervention objectives and the learning outcomes. RCT is not the magic wand that will answer all the questions.

A good read that keeps the discussion open, whether you agree with the authors or not.