Attend Spring Meetings on Development topics from Apr 17-21. Comment and engage with experts. Calendar of Events

Syndicate content

Add new comment

Better Public Sector Projects Which Don't Matter?

Nick Manning's picture

SDM-IN-042 World Bank In last week’s post, I asked whether Governance and Public Sector Management (GPSM) projects are having much large scale impact. It is tempting to reduce this to the question of why don’t development projects which focus on this work more often (although their track record is perhaps not as limited as some reviews of donor assistance might suggest). From this starting point, recent thinking suggests that donor rigidity and project designs which fix the visible form without improving the underlying public management function are the problem.   
The remedy, as set out most prominently in “Problem Driven Iterative Adaptation” and in the World Bank’s own Public Sector Management Approach, suggests that we should focus on the de facto rather than the de jure and adapt the nature of our support as project implementation unrolls. Problem-driven iterative adaptation (PDIA) approaches are referred to in recent reforms of Ministries of Finance in the Caribbean and reform approaches in Mozambique and in Burundi. Bank interventions in Sierra Leone and in Punjab have been cited as examples of this approach in practice.

There are a few difficulties here however. First, we run the risk of declaring this more problem-focused, adaptive intervention style to be a success, when really all that we know is that the more rigid Plan A wasn’t too great. While there are strong theoretical grounds for hope in this more problem-centered approach, there is, as yet, little evidence.
Second, the learning that emerges from this approach is primarily in terms of intervention technique. Since the approach is, intentionally, idiosyncratic and based around the specifics of the context, it will not be easy to add up the findings across a range of cases.
The third problem is perhaps the most vexing. Adaptive approaches, in the way that they have been advocated to date, focus on the relatively small. If they raised the rate of project “success” in small areas of public sector management from, say, 50% to 75%, but reduced the scope of the resulting improvements in the process, would this be a big step forward? We would have a larger scatterplot of mini-successes, but would this add up to anything that really mattered? Surely we don’t want to do things right without making any progress in finding the right thing to do (a friendly critic of the Bank once pointed out that a project I had designed in Afghanistan was not so much like rearranging deck chairs on the Titanic, it was more akin to moving the umbrellas in the cocktail glasses next to the deck chairs). Wouldn’t we be prepared to tolerate a reduction in that success rate if at least some of them were in some way transformative – the country-level public sector equivalents of the Marshall Plan, the “Green Revolution” or the near eradication of smallpox?
How might we edge our way towards more transformative public sector management interventions? 
Taking the adaptive, flexible logic as a reasonable basis for GPSM interventions (but let’s not declare victory – it really is just plan B, and doubtless plans C and D will be announced with accompanying trumpet fanfares at some point) the challenge is to apply it in a way which allows us to build up a picture across time and space, so that we can learn more about what seems to work in public sector reform and where we can see the prospect of transformative change emerging.
Designing reforms that we can learn from…
Large scale improvements in the machinery of the public sector take a very long while to introduce (a reform of the public sector pay system in Afghanistan took 3 years to pass and 8 to implement) and even then don’t show up as improvements in the real world of service delivery until some considerable time later. This is why, when we measure GPSM improvements at the aggregate, country level, we can’t find transformative change. So we need sound metrics of things that can be changed in the short to medium term which we have reasonable grounds for believing will make a significant difference in the overall capacity of the public sector in the long term.
One possibility is that we focus on measuring and improving the different, underlying management systems (public financial management, procurement, human resource management, etc.) within the public sector. This is starting to happen and there are now a variety of system-specific instruments for tracking progress including the World Bank’s CPIA, PEFA, TADAT and CABRI for aspects of the public financial management system and MAPS for the procurement system. 
…and which likely contribute to transformational change
Of course we also need to be comfortable that real change in these management systems is associated with improved development outcomes. There’s a lot to learn from.  Countries such as the Central African Republic, The Gambia, Kyrgyz Republic and Tonga have all made dramatic improvements in their CPIA scores for GPSM (Quality of Budgetary and Financial Management, Quality of Public Administration and Transparency, Accountability, and Corruption in the Public Sector) in the period 2004-2011.   How did these system improvements relate to improved development outcomes in these countries?
There is significant new thinking emerging
The debate initiated by a recent article from Francis Fukuyama has reinvigorated the debate on how to measure state capacity in a useful disaggregated way. The World Bank’s Indicators of the Strength of Public Management Systems (ISPMS) initiative is one of several signs of fresh energy being brought to bear on this topic.
There’s a lot to worry about in the development of such metrics.  Management system indicators could be a Trojan Horse, hiding the reappearance of, now much-reviled, “best practices” and again encouraging a “Washington Consensus”-equivalent approach to public management reform. There’s also the risk that such metrics distract us from important historical lessons about the necessary political conditions for institutional reform or from the significance of learning from positive outliers within countries. These are huge risks, but my sense is that without taking them, public management specialists in development might be doomed to doing ever smaller things right, but without finding the right thing to do.