Wednesday, April 16, 2008

Governance stuff that makes me think harder 

I've mentioned the book, which on Monday got its cover design (thank God my co-editor is a visual thinker). I'll reveal it when we get back the mock-up in a month or two, because I think it's a neat idea that Bryan had. My role was "looks good!"

One of the chapters is on the various ways in which we measure governance and what these mean. A basic theme of the book is that the measurements are amalgamations, atheoretical and often ill-considered. Measuring governance seems to be one of those areas.

So two posts around a World Bank conference this week got my attention. First, Larry Summers is provocative:
Mr. Summers�s predecessor at Treasury, Robert Rubin, ventured that �effective government� was the key for growth. But Summers wasn�t buying that one either. Government are often seen as effective because the countries are growing fast, he said. Then when they stop growing, the governments are dismissed as lame and corrupt � Indonesia under Suharto, for instance.
The first three words out of my mouth were "Lee Kuan Yew". How does one classify Singapore's governance? If you say "must be good, look at its growth", Summers is right. If you say "horrible, it's authoritarian", you have a harder time making the link between governance and growth. Take a sheet next to your desk and write a 2x2 matrix. Along one dimension write "growing" and "stagnant"; along the other write "good governance" and "poor governance". Pick about 20 countries at random and put them in one of the four boxes based on what you know. Now since many readers will know some economics they probably get the growth/stagnation classifications roughly right. But watch how you decide the governance question.

At the same conference I think, and somewhat relatedly, the maddeningly brilliant Dani Rodrik writes,

So good governance is both an end and a means. It is a key goal of development, broadly construed, and it is also an instrument for achieving better policymaking and improved economic outcomes. Any sensible discussion of governance must be clear about the distinction. And it must clarify in which of these two senses governance is �the problem� we are trying to fix.

I make the following points below. First, economists have very little useful to contribute to governance-as-an-end. Second, while they have more to say about governance-as-a-means, what they do say is often not what they should say. Where economists can be useful is in designing institutional arrangements for specific policy reforms targeted at relaxing binding growth constraints--what one might call �governance in the small.� This agenda differs quite a bit from the broad governance agenda on which much ink is being spilt. And third, there are sometimes tradeoffs between governance-as-an-end and governance-as-a-means which policymakers and advisors need to be conscious of.

I have to read the paper to see what those tradeoffs are, but I'm going to guess that they relate to Summers' observation.