To end poverty, we need to get better at improving institutions


This page in:

Good institutions matter for development. Institutions enable societies to address challenges – from managing irrigation and schools systems, to raising and spending revenues. In the terms of Nobel Prize laureate Elinor Ostrom, the right institutions enable effective collective action, while poor or missing institutions hinder problem-solving.

As a consequence, countries with better institutions are more likely to experience sustained and equitable growth (see also Robinson and Acemoglu 2012; Fukuyama 2011; Rodrik 2000). As the New Consensus for "More Effective" Institutions for Development sets out, including institutional issues in the post-2015 development agenda will be essential to end poverty. Not least of all, good institutions are key to effective lending. While fiduciary safeguards help where institutions are weak, the development effectiveness of aid ultimately hinges on the complex set of sector and core institutions in a country.

In a number of instances, institutions have improved: in many places, education systems have expanded and begun to improve quality. A number of countries have succeeded in improving their customs systems – which were previously often regarded as hot-beds of corruption, and hence as difficult to reform. These institutional improvements have come about within the time span of a decade or less, and at least in part, they were induced by deliberate efforts involving external assistance.

But many attempts at institutional reform have not achieved their aims. Many countries still have shoddy Ministries of Agriculture, dysfunctional judiciaries, or decentralization arrangements that fail to contribute to better service delivery – despite significant attention in national development strategies and external efforts.

Recent experience has also led to growing discomfort with easy prescriptions. A number of countries have made significant development progress despite governance weaknesses or with institutional features that depart from standard prescriptions. China achieved its impressive infrastructure expansion with a rather complex set of institutions and stakeholders involved, including party cadres. A different example is Bangladesh which has made significant gains on social development – despite many deep-seated institutional dysfunctions.

Research and practitioners’ debates are grappling with what institutions really matter, when and where. This includes debates about the role of property rights (Khan 2002, Meisel and Ould 2008, Haber 2003) whether to emulate institutional models or focus on home-grown institutions (Pritchett et al. 2010, Krause 2012), the proposal to aim for ‘good enough’ rather than ‘good governance’ (Grindle 2007), and guidance on how to identify a good institutional fit (see the World Bank’s Public Sector Management Approach 2012).

At the same time, for the practical level of day-to-day institutions – how to organize administrative courts or drugs procurement, regulate power sectors, or conduct civil service hiring – systematic evidence of what has worked in what contexts is often still missing. Despite emphasis on the need for a ‘good fit’ in principle, in practice advice often still rests on notions of best practice or standards; and at times involves wishful thinking, such as suggesting regulatory arrangements that will work “if the judiciary can be strengthened” – while the latter may be years away.

Two further insights are important. One is that institutional change depends on decision-makers. But decision-makers typically represent established elites -- which are often served well by existing arrangements. This raises the question of why these ‘entrenched interests’ would ever come to favor progressive reform. Yet such changes happen at least in some instances – be it because underlying incentives shift, or because certain ideas, such as the need for greater government transparency become very powerful and increasingly hard to reject.
Another part of the puzzle is the chicken-and-egg problem: institutions enable growth and poverty reduction, but economic and social progress also facilitates better institutions. In countries such as South Sudan or Somalia building good institutions may be exceedingly difficult as long as the majority of citizens face high levels of physical and economic insecurity, and literacy remains very low. In many middle income countries in turn, much hope for better institutions rests with an expanding middle class that can demand less corruption and institutions that provide a more level playing field for all.

Several efforts are ongoing to understand how better institutions emerge – from the researchers associated with the institutional profiles database, to authors such as Fukuyama and Acemoglu and Robinson, as well as successive major research programs funded by DfID on Africa Power and Politics and Effective States. For fragile states, the World Development Report 2011 on Conflict, Security, and Development conceptualized how more resilient states emerge and required time periods. The challenge is how to connect the dots between debates on fundamental issues such as the role of property rights, and the practical issues relevant for operational work – such as whether a ‘one-stop shop’ or more distributed institutional responsibilities for allocating valuable urban land produces better results.

Getting better at improving institutions will be essential for accelerating poverty reduction and achieving other goals such as mitigating and adapting to climate change. For the post-2015 development agenda, four approaches stand out as important:

  • Embracing across all practice areas of development that the what and how of institutional reform are uncertain and truly moving away from relying on preconceived notions rather than evidence;
  • A broader and more systematic effort at generating evidence about recent institutional transformations and their impacts – drawing on the experience of development agencies and NGO partners as well as directly on the experience of developing countries to make this tangible, specific, and ‘actionable’;
  • More deliberate experimentation with institutional forms – testing what works best for different environments and development challenges. This can facilitate breaking away from perceived ‘best practice’, and towards institutional solutions that are robust even in difficult environments;
  • Raising the game on knowledge diffusion: information on what institutional reforms have worked where, how and why need to be accessible for development practitioners across sectors, whether they are agronomists, climate change experts, macro economists, public sector specialists, or working on private sector development or natural resource management.


Verena Fritz

Senior Public Sector Specialist, Global Governance Practice, the World Bank

Join the Conversation

March 01, 2013

A lot has been said and done about alleviating poverty in the world. Every year new methods, programs, approaches are deployed, and yet the rich everyday get richer while the poor slump even further.

Perhaps yet again its time for yet another new approach and yes, aaargghh, a paradigm shift .... again ....

Anyway my point is that this has been going on for more than 50 years, but with very little to show for it as the number of poor continue to increase.

Maybe we need to go back to the basics.

Take a look at that picture. This is representative of the poor in most poor and developing countries. It could be in South America, Asia, or Africa and it remains the same.

The developed economies also have their share of poor. The percentage is much lower, but more importantly, the image changes significantly

Whereas the poor economy, the image of a poor person is that of a man, woman or child, scavenging in dump heaps, and typically either carrying or with a toddler/infant in tow.

In contrast, a poor person in a developed economy is depicted slouched on a worn out couch, surrounded by hoarding articles and junk, probably with a beer can etc. They may still scavenge but more for recycle materials and will usually have tools and devices that help them along.

The point is that the developed economies have over time factored in social structures that allow the poor in their societies to still maintain some form of dignity even in their poverty. What is dignity you ask? Some kind of shelter, access to food, clothing, some health care and school for children. Because they are not struggling to obtain these for themselves and their families, they can afford to have some dignity and channel their energy elsewhere

In countries that do not have these facilities, the dirt poor have no option but to do everything possible to obtain whatever little value they can harvest from whatever the activity in order to feed,clothe and shelter themselves and their families just enough for that day and enough to survive to the next day.

I'm suggesting that by providing structures for the poor to have access to the very basic human needs for them and their depend-ants, they may be just able to have sufficient latitude to focus their energy on other activities that will lift them from their vicious poverty cycle.

Not all will be success stories, but am pretty sure, many pleasant surprises await such interventions.

March 01, 2013

As much as I agree with the sentiment of the post, I believe we need to improve contemporary understanding of what constitutes an institution before the listed priority approaches will yield fundamentally.

Paraphrasing Avner Greif, our current understanding of institutions tends to define institutions in the same way as saying that a car transports people would - rather than define it as a vehicle moving on wheels as the dictionary does.

Do we really understand what it is we're seeking to improve - not just the visible outcomes?

Related, Matt Andrews from the Kennedy School often blogs in the same vein. E.g. Here and here.