Monday, November 8, 2010

Development 3.0

Very nice summary of the developments in the field of development by Shanta Devarajan!


In the old days—that is, the 1950s and 1960s—development was about correcting market failures.  Influenced by the “big push” theories of economists like Rosenstein-Rodan, post-war Keynesian economics and the apparent success of the Soviet Union, policymakers in developing countries saw the role of government as providing public goods (bridges, roads and ports), addressing externalities (protecting “infant industries”) and redistributing income to poor people (by, for instance, keeping food prices low).  Donors supported these countries by financing some of the public goods—a bridge, say.  Knowledge assistance consisted of helping to identify the market failure, and then designing the “optimal bridge”.

The incentives of poor-country politicians and rich-country donors were aligned.  Politicians could take credit for correcting market failures—government was doing what it was supposed to do—while donors could make sure their money was well-spent.  This was Development 1.0.

Starting in the 1970s, it became clear that these government interventions were not delivering the intended results.  Protected industries were so insulated from world markets that they never produced efficiently (the Morogoro shoe factory in Tanzania never exported a single pair of shoes).  Roads were built but not maintained to the point that they were not passable.  Low food prices led to food shortages and increased poverty in rural areas.  In correcting market failures, we created a set of government failures: well-intentioned interventions that fail to deliver the intended results.

To rescue these economies from distress, donors made their financial assistance “conditional” on governments’ reversing these policies.  The previously harmonious relationship began to fray.  Politicians resisted—mostly because their friends and family were benefiting from the distorted policies—and complied half-heartedly—often blaming the donors went things went awry.  And knowledge assistance focused on estimating the costs of the previous interventions (as if the only thing standing in the way was the politician’s lack of knowledge about these costs).  You might call this Development 2.0.

Today, although many of the egregious distortions have been removed, we find ourselves still faced with government failures, but in a more insidious form that directly hurts poor people.  Many of the failures are in infrastructure, education and health—the sacred cows of government intervention.  Correcting them invites the criticism that we are trying to undermine government, harking back to the days of conditionality.

A classic example is water tariffs.  Subsidized or free water leads to water scarcity.  Politicians, who control the utilities through these subsidies, ensure that the scarce water goes to neighborhoods where their clients live.  Poor people meanwhile have to pay 5 to 16 times the meter rate to buy water from vendors.  But no politician can run on a platform of raising water tariffs (even if it will help the poor) and hope to get elected. Other examples include absenteeism of teachers in public primary schools—25 percent in India, 27 percent in Uganda.  Or the leakage of public funds in health—that reaches a staggering 99 percent in Chad.

These government failures do not happen by accident.  Rather, they arise from two kinds of imperfections in the public sector (much like market failures arise from imperfections in the private sector).

  • When they don’t use market incentives, governments have difficulties in monitoring and enforcing performance by frontline service providers.  The result is absentee teachers, clinics without drugs, impassable roads.
  • A second, more pervasive imperfection is in the political system.  Even in democracies where the median voter is poor, politicians who advocate anti-poor policies (such as some of the government interventions above) continue to get elected.  One reason is that politicians are able to control the flow of information to the electorate, convincing them to vote for policies that are, in fact, not in their interest.  In the water tariff example, politicians run on a platform of maintaining free water—and get re-elected.

In this situation of government failure, the traditional instruments of financial and knowledge assistance are not very effective.  Politicians will resist conditionality, and refuse the financial assistance if it could lead to electoral defeat.  Providing financial assistance without conditionality makes it easier to continue with distorted policies.  Reports about the costs of distortions are of little value (not to say irritating) to the politician who is the cause of the distortions.  Even if he is not the cause—and is instead a reform champion—then by definition he already knows the costs.  The reports are still of little value.

So what can we do?  Our understanding of government failure has coincided with two other developments.  One is the rise of civil society’s voice in public discourse.  The second is the technology revolution in poor countries.  There’s a message here.  Can we use technology and the voice of civil society to address these government failures?  Rather than imposing conditions, we can empower poor people to monitor service providers.  With some 80 percent of Africans having access to a cell phone, it is not difficult to have parents (or the students themselves) send an SMS message if the teacher is not in school, or there are no drugs in the clinic or the purported road maintenance program is not happening.  This could do more for helping governments and donors get value for money than all the fiduciary controls we put in place.  While we are at it, why don’t donors (including the World Bank) use technology to have the beneficiaries monitor and supervise development projects?

We can also use technology to alleviate the information problem.  Rather than writing reports on the costs of distortions (and whispering them in the Finance Minister’s ear), we could disseminate these results—in digestible form—to poor people through their cell phones.  Get the information out about who benefits from infrastructure subsidies, which districts have the highest teacher absentee rate, etc.  This is information about poor people’s daily lives; they should be the first to receive it.  As better informed voters, they may then start voting for politicians who advocate in their interest.  Going further, why not prepare these reports in collaboration with poor people?  After all, the analysis is about them.

Each year, the World Bank produces a World Development Report.  While there is an extensive consultation process with the draft, the Report is essentially written by a core team of Bank staff.  Why not produce the report like Wikipedia, and invite the whole world to write it?  As one of my colleagues put it, “Then it will be the World’s Development Report.”

And a fitting symbol of Development 3.0.


It sounds like Dani Rodrik’s Capitalism 3.0


“Just as Smith’s minimal capitalism was transformed into Keynes’ mixed economy, we need to contemplate a transition from the national version of the mixed economy to its global counterpart.

This means imagining a better balance between markets and their supporting institutions at the global level. Sometimes, this will require extending institutions outward from nation states and strengthening global governance. At other times, it will mean preventing markets from expanding beyond the reach of institutions that must remain national. The right approach will differ across country groupings and among issue areas.

Designing the next capitalism will not be easy. But we do have history on our side: capitalism’s saving grace is that it is almost infinitely malleable.”


More here

R&D Exploration vs. Exploitation

“We study how exploration versus exploitation innovations impact economic growth through a tractable endogenous growth framework that contains multiple innovation sizes, multi-product firms, and entry/exit. Firms invest in exploration R&D to acquire new product lines and exploitation R&D to improve their existing product lines. We model and show empirically that exploration R&D does not scale as strongly with firm size as exploitation R&D. The resulting framework conforms to many regularities regarding innovation and growth differences across the firm size distribution. We also incorporate patent citations into our theoretical framework. The framework generates a simple test using patent citations that indicates that entrants and small firms have relatively higher growth spillover effects.”

Interesting finding: exploration R&D does not scale as strongly with firm size as exploitation R&D. The full paper by Akcigit and Kerr is here.