Can the World Bank do PDIA?

Yesterday I was involved in a conversation about change in the World Bank. A number of officials from the Bank were proposing using a PDIA-type approach to doing governance reform in a particular country. They had done the best of Bank work: excellent consultants had written appropriate and insightful background materials on the country, and a position document clarified why change was needed and what it could look like.

But there were others who asked a question I hear a lot; "can we do PDIA at the World Bank?"

It is a good and important question, but I fear that many who ask it don't intend to get an answer. It is often stated in a rhetorical fashion to shut down the conversation, in my opinion. Those who raise it often say, "We understand that we need to change…but we can't really do it and are wasting our time in this discussion."

So, here is my sense of the problem of change in any organization (the World Bank included). You need four factors to align to make the adjustment happen:

1. A disruption that shows you need to change;

2. Weakened incumbents that agents are willing to move away from (people agree that the systems we have relied on in the past are no longer working);

3. Viable alternatives exist that we have the capacity and authority to work with and pull off;

4. Enough agency support for the change (on balance).

I think there could be an alignment of these factors in the Bank when considering the potential adoption of a new process like PDIA (where the principles of such approach are simple: Reforms should be contextually relevant, hence problem driven; Reforms should be allowed to emerge as locally fitted hybrids, hence the value of iteration and learning; Reforms will need to diffuse and be used broadly, hence the importance of broad agent engagement).

1. There is now at least a ten year history of internal and external work showing that the organization has a performance problem with its work (especially in the area of governance reform); results are not great—especially when one considers the functional impact of reforms in the governance space. We have data showing really poor results, and we have more stories to support this data than we would be able to share.
2. There is a lot of evidence that incumbent project processes are part of the problem, and many bank staff have a good sense of the limits of current management structures in the organization…affecting incentives, reducing flexibility, constraining quality work, etc.
3. There are many alternative processes that people in the Bank introduced over the years to allow processes like PDIA. The spirit behind Adaptable Program Loans (APLs) and Learning and Innovation Loans (LiLs) overlaps incredibly with what we talk about, for instance, given that both mechanisms are designed to allow flexibility, learning, etc.
4. There are many examples of task team leaders in the bank who have used these and other instruments to do PDIA-type work (indeed, Lant Pritchett, Michael Woolcock and I put PDIA together when thinking of the principles evident in projects we thought were successful). These folks represent what I think is a large constituency of people who care deeply about what they do, want to do work that makes a difference, and know that change is needed for this to happen.

So, I have to say that I think PDIA can be done in the World Bank. I believe the tools are there. Even the bureaucratic, project and lending mechanisms exist. There are certainly enough people to start the revolution!

However, there is a different view on the storyline I could tell in respect of all four of the above factors:

1. Whereas there is a lot of evidence that Bank projects (especially in the governance domain) have limited influence, we have also seen growing demand for these projects. As a result, I imagine those folks looking at money flows wonder if there is a problem. "As long as we have a growing portfolio of operations, surely we are doing something right."
2. The Bank culture is pretty robust, and there is something about the focus on 'accountability', 'efficiency' etc. that makes linear project processes and rigid bureaucratic systems look impressive and 'appropriate.' The role of the Board arguably entrenches this situation. Even though people in the bank might think their systems work poorly in helping foster context-specific interventions, therefore, they might find the structures work well in preserving the organization as a large international bureaucracy. Thus the incumbent structures might work well for those in the organization…
3. Whereas alternatives exist, they are not widely used—especially LiLs (which I think may even have been phased out). Where they are used (especially APLs) it is not clear that they always provide the flexibility intended… So they offer the potential of an alternative but this is not always realized.
4. The agents who support change may not be those who have influence and power, and may actually be disempowered by the incentive structures etc. in the organization. It is unclear if the Board and others in senior management really support change that would promote a more flexible, responsive, and dynamic organization.

Though simplified, I hope my views here are not an inaccurate representation of an argument many in the Bank might stitch together. It would lead to the clear answer, "The World Bank cannot do PDIA…"

This is the discussion I have with myself in the final chapter of my book. In the book I suggest that there are some interventions that could tip the conversation in favor of more change (shown in the final column). These ideas are not fleshed out very much in this post but get to some crucial issues that need to be addressed in the Bank: the incentives around the movement of money to countries and within the organization; the importance of going beyond lip service in saying you don't do certain things (like reproducing best practice); the incentives that processes create (and the way these could be fashioned to push task teams towards problem driven reform, more relational engagement, etc.).

Even with the organizational impediments to doing something like PDIA in the bank, however, I think that it is more than possible and there are many examples where people do it routinely. I firmly believe that reforms will continue producing form without function if an approach like PDIA is not introduced. In my opinion, this makes the challenge of change worthwhile for those who care a lot about the development of the world (and maybe for those who care about the efficacy of the World Bank, too).

Comments

[RupertSimons]

Matt, I’m glad you raised the question, we discussed something similar with colleagues at the Africa Governance Initiative and the Budget Strengthening Initiative. I think you’re right that the incentives and processes facing staff of the World Bank and other donors are important. I would suggest that we also need to look at donors' sense of what their money is buying, their willingness to accept risk, and their willingness to trust their grantees.

Some donors act mostly as customers. They spend money to buy a product. That could be influence with a friendly government, tolerance from an unfriendly one, votes from marginal states (food aid) or jobs for their nationals (through tied aid and contractors). They need advance specification of the product they are buying, and if in doubt would rather receive what is in the log frame than what is most useful, because it’s less risky.

Other donors, often the development banks, are debt investors. They lend in support of a project that will pay back in the long term, like a power dam, or better governance. If you want them to invest, you have to show how the project will generate a return that can repay the principal with interest, and minimal risk of default. For a governance program, that means: do we believe this project will improve governance in ways that matter AND that we can measure?

Sometimes donors have enough confidence in a government to give it budget support, or philanthropists give an NGO unrestricted funding. That is an equity investment. The rationale for budget support or unrestricted funding is: we believe in you. We may not like everything you do, but we trust you to spend it better than we could, and expect to reap a dividend at some point in the future. When you want them to fund a project, they need to believe in you, and the country you work in. The details of the project are less important. The same donor might do each of these at different times and different places.

I think the equity investors will welcome PDIA, and the debt investors should too. But I don’t think it’s an appropriate mode for the donors who are buying a product. If the only way they can fund a project is with a log frame, then maybe they shouldn’t be funding governance programs at all. Instead, they should fund programs for which log frames ARE appropriate and useful, like roads and dams. Ideally they would coordinate, so the more flexible funders finance PDIA-style programs and others.

I’d love to hear what people on the donor side think of this issue.

[Matt Andrews]

All great points, Rupert. It is an interesting and important question you ask: if donor agencies are not set up to appropriately support governance work, perhaps they should not be in that business.