Why do we persist so long with a reform approach that does not solve problems?

written by Kate Bridges and Michael Woolcock

In Malawi, efforts at institutional reform have been numerous, earnest and longstanding. Since 1966, there have been more than three times as many World Bank projects with ‘institutional reform’ content  as there have been in any other thematic or sectoral category.

In a recent paper, we argue that these efforts have largely failed. Public scandals such as “Cashgate” – in which about US$ 32 million in government funds was misappropriated between April and September 2013 – are the tip of the iceberg: high profile cases reflecting a logic of corruption that remains unchallenged by reform efforts. Continue reading Why do we persist so long with a reform approach that does not solve problems?

BSC website gets a makeover

As you may have noticed, our website was antiquated, to say the least. The task of giving it a makeover has been on the back burner for a while now. We are proud to finally announce that our new website is live!

The site provides an overview of PDIA, links to our blog and lists our publications by policy area. We have a page for each of our workshops which include videos and slides: Untying Development and Doing Development Differently. We have also included a map of the BSC video series – a set of short 2-5 minute videos highlighting the key elements of our approach. We hope this will provide an overview of how the videos fit together.

Take a look at it and let us know what you think!

Why many development initiatives have achievement gaps…and what to do about this

written by Matt Andrews

Yesterday I blogged about Hirschman’s Hiding Hand. As I interpret it, a central part of his idea is that many development projects:

  • focus on solving complex problems, and
  • only once they have started does a ‘hiding hand’ lift to show how hard the problem is to solve,
  • but because policy-makers and reformers are already en route to solving the problem they don’t turn away from the challenges, and
  • so they start getting creative and finding ways to really solve the problem. Initial plans and designs are shelved in favor of experiments with new ideas, and after much muddling the problem is solved (albeit with unforeseen or hybrid end products).

I like the argument. But why do I see so many development projects that don’t look like this?

I see projects where solutions or projects are introduced and don’t have much impact, but then they are tried again and again–with processes that don’t allow one to recognize the unforeseen challenges, and rigid designs that don’t allow one to change or experiment or pivot around constraints and limits. Instead of adjusting when the going gets tough, many development projects carry on with the proposed solution and produce whatever limited form is possible.

I think this is because many reforms are not focused on solving problems; they are rather focused on gaining short-run legitimacy (money and support) which comes through simple promises of quick solutions. This is the most rank form of isomorphism one can imagine; where one mimics purely for show… so you get a ‘fake’ that lacks the functionality of the real thing…

Let me use Public Financial Management (PFM) reforms as an example.

What problems do these reforms try to solve? Quite a few, potentially. They could try to solve problems of governments overspending, or problems of governments not using money in the most efficient and effective manner (and ensuring services are delivered), or of governments using money in ways that erode trust between the state and citizens (and more).

Now, let me ask how many reforms actually examine whether they solve these problems? Very few, actually. Mostly, reforms ask about whether a government has introduced a new multi-year budget or an integrated financial management system. Or a new law on fiscal rules, or a new procurement system.

Sometimes the reforms will ask questions about whether fiscal discipline is improved (largely because this is something outsiders like the IMF focus on) but I seldom see any reforms–or any PFM assessments (like PEFA or even the assessments of transparency) asking if services are better delivered after reforms, or if reforms enhance trust between citizens and the state. I don’t even see efforts to systematically capture information about intermediate products that might lead to these ‘solved problems’. For instance:

  • Do we have evidence that goods are procured and delivered more efficiently (time and money-wise) after reform?
  • Do we have any systematic data to show that our new human resource management systems are helping ensure that civil servants are present and working well, and that our new payment systems pay them on time (and do a better job of limiting payments to ghost workers)?
  • Do we have any consistent evidence to show that suppliers are paid more promptly after reforms?
  • Is there any effort to see if IT systems are used as we assume they will be used, after reforms?
  • Does anyone look to see if infrastructure projects are more likely to start on time and reach completion after costly project management interventions?
  • Do we have records to show that infrastructure receives proper maintenance after reform?
  • Is there any effort to see if taxpayers trust government more with their money?

This is a long list of questions (but there are many more), and I am sure that some reforms do try to capture data on some of them (if you’ve measured these in a reform, please comment as such…it would be interesting and important to know). Most reforms I have observed don’t try to do it at all, however, which was the focus of a recent discussion on the role of PFM and service delivery Time to Care About Service Delivery? Specialists from around the world were asked whether PFM reforms improve service delivery and the answer was “we think so…we expect so…we hope so…BUT WE CAN’T TELL YOU BECAUSE WE DON’T ACTUALLY ASK EXPLICIT QUESTIONS ABOUT THIS.”

My concern with this is manifold: (i) Does the failure to ask if we are solving the problems suggest that we as a community of reformers don’t really care about the problems in the first place? (ii) Does it mean that we will not be sensitive to the situations Hirschman speaks about when he discusses unforeseen challenges that undermine our ability to address problems (simply because we don’t focus on the problems)?  (iii) Does this also mean that we will not have any moments where we explore alternatives and experiment with real solutions that help to overcome hurdles en route to solving problems?

Unfortunately, I think the observations of gaps after reforms speak to all of these interpretations. And this is why many reforms and interventions do not end up solving problems. In these cases, we get the half-baked versions of the pre-planned solution…with no adjustment and no ‘solved problem’. PFM systems look better but still don’t function–so payments remain late, wages are unpaid to some and overpaid to many, services are not delivered better, and trust actually declines. Most worrying: we have spent years doing the reforms, and now need to pretend they work..and have no learning about why the problems still fester.

The solution (maybe): In my mind this can be rectified–and we can move towards producing more projects like those Hirschman observed–by

  • focusing reforms on problems, explicitly, aggressively, from the start;
  • measuring progress by looking at indicators of ‘problem solved’ (like improved levels of trust after PFM reforms) and intermediate indicators we think will get us there (better payment of contracts, more efficient procurement, etc;
  • regularly monitoring this progress;
  • being on the lookout for expected unexpecteds (things that we didn’t know about that make our initial solutions less impactful); and
  • being willing to adjust what we started with to ensure we produce real solutions to real problems–functional improvements and not just changes in form.

For more, read This is PFM which advocates a functional approach to thinking about and doing PFM reform.

Hirschman’s Hiding Hand and Problem Driven Change

written by Matt Andrews

I referred to Albert Hirschman’s work on the “Principle of the Hiding Hand” in my class today. It is a great principle, and has real application when thinking about PDIA and problem driven change.

In his essay, “The Principle of the Hiding Hand” Hirschman argues that creative solutions most frequently come from adapting to tasks that turn out to be more challenging than we expect.

In Hirschman’s words, “men engage successfully in problem-solving [when] they take up problems which they think they can solve, find them more difficult than expected, but then, being stuck with them, attack willy-nilly the unsuspected difficulties – and sometimes even succeed.”

It’s really beautiful, because it takes as a given some facts that we often think stand in the way of doing flexible, PDIA-type development. Hirschman expects that decision makers will tackle problems, often adopt solutions that look attractive but are hard to pull off (perhaps like big best practice type initiatives), and will overestimate the potential results.

He argues that they wouldn’t try to do the challenging things that development demands if they didn’t think this way. So, he advises to ‘go with it’ …. but then wait for the unexpected… in the form of complexities, constraints, hidden difficulties, etc.

When these unforseen difficulties emerge, Hirschman argues, we have the opportunity to become creative–and to iterate and experiment and find and fit ways to solve the problems that initiated the work in the first place … building on the sunk costs already incurred in pursuing the big, best practice, perfect solution. (saying something like “we’ve come so far…let’s now iterate to ensure we actually solve the problem we set out to solve.”)

Beautiful: Start where you are, focus on solving problems, try the big best practice (but hard to actually do) solution, and become creative when you hit the challenges…

What he assumes is that you have space for flexible change and PDIA-type innovation because of the sunk costs associated with past (or current) reform. An interesting assumption, that I think we can look at academically and reflect on practically.

Required and fundamentally vital reading for anyone in development.

Doing Development Differently: Day 2 Summary

Yesterday was the last day of Doing Development Differently (#differentdev). A group of about 40 development professionals from around the world met to discuss positive cases where development initiatives (call them projects, interventions, activities or whatever) have led to real results and impact. It was another full day with two DDD Exchange Sessions, a PDIA example and another wind tunnel meeting. View the storify to see all the content, including videos, tweets, photos and blogs (Duncan Green, Alan Hudson).

We are delighted to share the rest of 7:30 presentations.

You can also watch Matt Andrews closing remarks below. Stay tuned for the upcoming Manifesto from the workshop!

Doing Development Differently: Day 1 Summary

Today was the first day of Doing Development Differently (#differentdev). It was a full day with two DDD Exchange Sessions, a design thinking session and a wind tunnel meeting. View the storify to see all the content, including videos, tweets and photos.

When we designed this workshop, we wanted to maximize the opportunity to hear from as many people as possible. Specifically, we wanted

  • to show that it is possible to do development differently;
  • the participants to discern key principles and cross-cutting modalities or tools;
  • to explore whether we could promote a vibrant Community of Practice for those trying to do development differently.

To facilitate this, we asked our presenters to prepare a 7:30 minute talk —with no powerpoints or visual accompaniments. The talk had to address the following questions:

  1. What problem were you trying to solve?
  2. How had you/your organization/others addressed this problem in the past?
  3. What did you do?
  4. How did you manage the politics of your work?
  5. How did you ensure learning in the process?

We are delighted to share the first set of 7:30 presentations: Michael Woolcock, Zack Brisson, Tim Williamson, and Kay Winning. Here are some key principles that cut across all the presenters:

  • Humility: We don’t know the answers
  • Articulate principles that can scale
  • Donors role: broker, convenor, facilitator, adviser
  • Understand context: listening, relationships and personal networks are central
  • Need feet on-the-ground to support the process
  • Create space for local solutions and local ownership
  • Embrace and navigate politics: work with what you have
  • Building and sustaining broad coalitions: middle/low level bureaucrats, many stakeholders at all levels
  • Iterative messy process: one that evolves over time, problems change, solutions change
  • Built-in rapid cycles of learning
  • Refine problem definition: focus on what really needs to be solved
  • Take advantage of windows of opportunity (shocks, critical junctures, etc)
  • Adaptability: thinking strategically but building on flexibility

Follow #differentdev and storify for live coverage of Day 2.

Doing Development Differently 2014

Last October, we hosted a one-day workshop entitled, Untying Development: Promoting Governance and Government with Impact. The day brought together different voices to discuss the challenge of creating a governance agenda that focuses on solving country-specific problems, involves local people through flexible and context-fitted processes, and emphasizes learning in the reform process.

We are proud to announce a two-day follow up workshop entitled, Doing Development Differently to be held on October 22-23, 2014 at the Harvard Kennedy School. This event is an opportunity to share practical lessons and insights, country experience, and to experiment first hand with selected methodologies and design thinking.The aim of the event is to start to build a shared community of practice, and to crystallize what we are learning about what doing development differently really looks like in practice. This event is co-hosted with the Overseas Development Institute (ODI), with funding from the Governance Partnership Facility.

Follow #differentdev on Twitter for live coverage.

World Bank uses PDIA in Sierra Leone

written by Salimah Samji

International development experts often tell us that they cannot do PDIA because the project processes within their organizations do not allow for flexibility. The truth however, is that all development agencies have some sort of instrument that does allow for experimentation and flexibility. Here’s an example of how a Pay and Performance project in Sierra Leone explicitly used PDIA principles.

Civil service reforms are complex in and of themselves. If you add, a lack of capacity to implement programs, multiple reporting lines, demoralized civil servants, a lack of coordination amongst key agencies, and a low-level of trust, the potential for success of such a reform decreases significantly. Recognizing this, the World Bank team decided to use the key principles of the PDIA framework with support from the Leadership for Results (LforR) program for their Pay and Performance Project in Sierra Leone. The rationale for this was to bring a broad range of stakeholders together and facilitate a process of collective problem and solution identification, as well as to introduce experimentation and adaptability during implementation.

They began with some short-term results-focused Rapid Results Initiatives (RRIs) in Year 0 and Year 1. The pilot was instrumental in building the confidence of the local civil servants by demonstrating that progress was possible in their context and gave them a sense of ownership. In addition, the short feedback loops facilitated rapid experiential learning about what results were actually achieved for both government and the World Bank staff – in PDIA terminology, we call this strategically crawling the design space.

Specifically, they used a two-pronged, learning-by-doing process, which included:

  1. Structured team coaching throughout the implementation process: A locally based rapid results coach who had an in-depth understanding of government and public sector reform was hired to provide support to teams on a daily basis. The coach:
    • Facilitated problem solving at multiple levels in the system with team-level work,
    • Helped create action plans by breaking a huge daunting task into smaller easier to digest chunks,
    • Motivated the teams despite the challenges, and
    • Created an opportunity for the teams to learn from each other and to see how their work fit within the larger picture.
  1. Facilitated leadership fora for dialogue: One-day strategic leadership convenings between leaders and implementation teams were held at critical points. These retreats served to review progress and learning, problem-solve, facilitate reflection, make strategic decisions, and course-correct where needed. In PDIA terminology, we call this maintaining the authorizing environment.

After 20 months of implementation (February 2014), they had several hard results. More importantly, there was stronger inter- and intra-agency collaboration and increased trust and communication. The teams actually had the capacity to do things themselves. The flexibility at the design stage allowed more politically and technically feasible solutions to emerge.

So, large bureaucracies can do PDIA and it doesn’t take forever. Bottom-line: the mundane matters and cannot be ignored for a project to succeed.

Roberto O. Panzardi and Kay Winning are in the process of publishing a paper with more details on this project. You can read about the preliminary results here.

 

Hello Organization Man: the importance of old (and boring) administration in a new (and exciting) world

written by Matt Andrews

Earlier this week, the New York Times ran the following great Op-ed on the lack of administrative focus in governance systems. This is an important article. We need to be saying ‘Hello’ when it comes to learning about organization… no matter how mundane it seems. This inspires me to spend even more time teaching about bureaucracy. For more see my blog on mundane and my course entitled Getting things done in development.

Here’s an excerpt from the Op-ed entitled Goodbye organization man.

Imagine two cities. In City A, town leaders notice that every few weeks a house catches on fire. So they create a fire department — a group of professionals with prepositioned firefighting equipment and special expertise. In City B, town leaders don’t create a fire department. When there’s a fire, they hurriedly cobble together some people and equipment to fight it.

We are City B. We are particularly slow to build institutions to combat long-running problems. …

A few generations ago, people grew up in and were comfortable with big organizations — the army, corporations and agencies. They organized huge construction projects in the 1930s, gigantic industrial mobilization during World War II, highway construction and corporate growth during the 1950s. Institutional stewardship, the care and reform of big organizations, was more prestigious.

Now nobody wants to be an Organization Man. We like start-ups, disrupters and rebels. Creativity is honored more than the administrative execution. Post-Internet, many people assume that big problems can be solved by swarms of small, loosely networked nonprofits and social entrepreneurs. Big hierarchical organizations are dinosaurs.

The Ebola crisis is another example that shows that this is misguided. The big, stolid agencies — the health ministries, the infrastructure builders, the procurement agencies — are the bulwarks of the civil and global order. Public and nonprofit management, the stuff that gets derided as “overhead,” really matters. It’s as important to attract talent to health ministries as it is to spend money on specific medicines.

As recent books by Francis Fukuyama and Philip Howard (the rule of nobody) have detailed, this is an era of general institutional decay. New, mobile institutions languish on the drawing board, while old ones are not reformed and tended. Executives at public agencies are robbed of discretionary power. Their hands are bound by court judgments and regulations.

When the boring tasks of governance are not performed, infrastructures don’t get built. Then, when epidemics strike, people die.

The boring bureuacrat

What is Action Learning?

written by Matt Andrews

Action learning is a key part of PDIA. It is “a hybrid technique that allows participants to use what they learn to tackle priority problems within their companies under actual work conditions. Action learning is a social process for resolving the difficulties managers increasingly confront, where history offers no solution.

At its heart, action learning is a systematic process that increases participants’ organizational learning in order to help them respond more effectively to change. Originated by Reg Revans (1983), action learning is based on the underlying premise that there is no learning without action and no action without learning. Action learning is inextricably linked with action science. Action science (Argyris, Putnam, and Smith, 1985) provides a conceptual framework and a methodology for facilitating action learning, while Revan’s work establishes the actual form. The following processes of action science are implicit in action learning:

  • Critical reflection: bringing underlying assumptions to consciousness; testing those assumptions to determine if they are appropriate for attaining the desired goal
  • Reframing: altering assumptions that don’t accomplish desired goals
  • Unlearning and relearning: developing new sets of learned skills based on reframed assumptions; replacing old with new skills until new ones are automatic.

Action learning methodology has three main elements:

  1. Problems that people identify;
  2. People who accept responsibility for taking action on a particular issue; and
  3. Colleagues who support and challenge one another in the process of resolving the problems.

Using real tasks as the vehicle for learning, individuals, groups, or teams develop management and leadership skills while working on organizational problems and testing their assumptions against real consequences. By taking a real problem, analyzing it, and implementing solutions derived with colleagues, individuals monitor results and can be held accountable for their actions. Revans believes that if we are to cope with accelerating and turbulent change, then we must place our confidence in the lived experiences and insights of others in order to be successful.” from Experiential Learning, Past and Present  Lewis, L.H. and Williams, C.J. (1994).

Reflection Graphic