Poverty Reduction Strategy Papers are nearly ten years old. It is nearly nine since the Enhanced Highly Indebted Poor Countries Initiative  placed on the international agenda a new type of donor conditionality, requiring governments to prepare, and hold public consultations about, comprehensive three-year plans for improving their performance in reducing poverty.

And it is eight years since the first wave of PRSP processes was completed. This makes it a good moment to ask: what have we learned from this experiment?

I use the word experiment deliberately. It is intended to suggest a particular perspective on the experience initiated in 1999, one that may be contrasted with two other approaches: the cynical one that sees in PRSPs only as an attempt by the World Bank and IMF to refurbish their image and reinforce their position as dominant actors in the world’s poorest countries; and the naïve one that accepts at face value the claims commonly made about country ownership of PRSP policies and the alignment of aid around those policies.

Two central points are captured by the concept of PRSPs as an experiment, and missed by the other approaches.

On the one hand, the core PRSP idea – namely, that it might be possible to elicit greater commitment to equitable and efficient development policies by obliging governments to debate their policies openly with other actors in their countries …  did not arise casually.

It was the response to an objective, longstanding and fairly well-understood problem. With a few significant exceptions, the policies supported by concessional lending institutions and development cooperation agencies in the poorest countries had not been implemented in sustained and consistent ways, and a massive body of research suggested the root problem to be lack of local commitment.

Policies were not owned, even by the governments that nominally approved them and certainly not by organized political and interest groups in the wider society.

On the other hand, changes in the thinking of political leaders and their supporters are not usually the sort of thing that can be engineered by actors who are external to the country in question, except perhaps in situations of extreme economic or political breakdown.

Therefore the core PRSP idea must be regarded as uncertain …. worthy of being described either, pessimistically, as a last ditch gamble or, more optimistically, as a rather bold venture into uncharted territory.

Calling the PRSP initiative an experiment seems a reasonably neutral way of communicating the observation that this was a process whose results could not be fully known in advance.

The PRSP experiment is not yet over. The second generation processes that have occurred in a number of countries are in several respects more promising than those of the first round.

This is partly because the direct link to HIPC approval processes has gone, but also there has been some genuine learning and innovation in some countries, notably about the form of public consultation that supports critical policy thinking.

It is good, in this sense, that the 2005 Review of the PRSP Approach by the World Bank and IMF provides renewed support for the initiative, suggesting improvement within the broad framework of concepts and procedures that has been established since 1999.

At the same time, it is not too early to ask what the experience so far tells us about the original hypothesis, and to try to extract the policy implications.

This needs to be done boldly and without too many concessions to the current etiquette of the aid business.

It is not clear that the Bank and Fund staffs, with their heavy involvement in the mechanics of the process have sufficient critical distance to see and state what is not working.

Other donors, now preoccupied with their commitments, renewed in Paris in February-March 2005, on harmonisation of aid programmes and alignment country-led policies and systems, are also not in a frame of mind where hard questions about the fundamentals of policy ownership are easily faced.

It is true that some donor agencies have become more interested than they were in understanding domestic politics and the political economy of state formation in the countries where they work.

But it remains still to be seen how far even the more politically informed agencies will allow the conclusions of this type of analysis to drive what they do.

With those preliminary remarks in mind, the rest of this paper tries to address several fundamental questions: What has been learned about the feasibility of ‘engineering’ commitment to poverty reduction?

And what does this imply? Could development cooperation do better within the PRSP framework, and if so how? What else might be worth a try? In particular, what might be done by developed world governments, outside the PRSP framework, to improve the results from aid-supported PRSPs?

What has been learned?

The broad picture: There is not exactly a consensus among the various studies, reviews and evaluations of PRSP processes and outcomes that have appeared in the last few years.

Anyway, I would not pretend to be able to summarize the nuances and points of agreement and disagreement in this literature as a whole.

However, I believe that, cutting across the different judgments on specifics, there is an important measure of convergence on two points: (1) PRSPs have been associated with some worthwhile improvements in policy processes, especially in countries that were already moving in the direction of greater results’ orientation and accountability; (2) these improvements are modest in character even in the best cases, and still fall far short of what is really needed: namely, local generation of high-quality policy thinking around poverty-reduction goals and arrangements for ensuring the corresponding action.

If this is a fair summary, it is already quite a sobering judgment on the gains made nearly ten years into the experiment. What are the factors that seem to account for the limited character of the progress?

Not all of the recent studies attempt to answer the deeper questions, but those that do tend to agree that the buy-into the PRSP process is mostly technocratic …. that is, restricted to quite a small number of strategically placed officials.

Fundamental political interactions and change processes have not been much affected, even in countries such as those of East Africa where political rhetoric at election times does not steer entirely away from the topic of poverty.

Several factors have been found to influence the degree to which the PRSP becomes an important factor in the allocation of resources and the provision of incentives within government systems: whether there are sector working groups or something equivalent that set priorities and guidelines based on experience and evidence of what works and what doesn’t in particular fields of policy; whether the budget process has been reformed so that it is a) trustworthy, b) based on policy objectives, and c) linked to the PRSP by definite criteria and procedures; whether incentives in the public service and local government have shifted in such a way that capacities to make and carry out public policies have stopped declining and begun to develop in a positive way.

There are plenty of examples suggesting that improvements in any of these areas help a PRSP to give new direction to country efforts and to provide, as a consequence, a focus for greater and more harmonized donor support.

There are also some suggestions that the existence of a PRSP has, in its turn, been a motivator of better sectoral policy thinking, and renewed efforts at reform of public financial management.

On the other hand, research and evaluation work suggests that lack of political buy-in affects all of these processes negatively. The reforms needed to give real life to PRSPs have almost everywhere suffered severe slippages.

The root cause is that those who exercise real power in a country are not interested in promoting them.

What are the implications of this assessment? Is the PRSP idea still viable? I believe this needs to be answered in two parts.

(To be continued)