One needle in a haystack, or many?

As I suspected, the event earlier this week on Industrial Policy and the Role of the State in Promoting Growth attracted a standing-room only crowd. Although the event was billed as a "panel discussion", the structure ended up being much more of a friendly debate, with Justin Lin and Ann Harrison sitting on one side of the table and Bill Easterly on the other. The topic, as I discussed before, was the question of whether there is a role for industrial policy in the developing world.

Easterly, Harrison, and Lin each had a chance to give a 15 minute presentation, which was then followed by a round of questions combined with closing arguments. Although the whole thing was entertaining, I didn't get the feeling that the proceedings changed the minds of anyone who was already leaning in one direction or another.

Easterly gave a presentation that covered many of the points you would be familiar with if you've read any of his previous work on development. He hews closely to the Socratic motto: "I know that I do not know." Experts should be well aware of the limits of their own knowledge, and instead trust in the wisdom of a decentralized search process whereby entrepreneurs discover new, more efficient ways of doing things. Simply put, industrial policy=bad idea.

On the other side of the table, Harrison and Lin supported the notion that we do, in fact, know enough to make intelligent choices about industrial policy. Lin provided an answer to the question I posed on the blog before the event: "How can a government agency know where a country's comparative advantage lies?" Lin argues that in advanced industrial economies, government agencies can't answer this question because they are at the technological frontier. The next discovery is unknowable until it is discovered.

But for developing countries, the path is not so uncertain. For countries inside the technological frontier, governments should look to countries with similar endowments of capital and labor but at a somewhat higher income level. A government agency could look to see what industries have thrived in this somewhat richer country and formulate an industrial policy around the path the richer country followed.

Is this feasible? The crowd seemed split, and Easterly certainly wasn't convinced. He pointed to one particularly curious example: the little island nation of Fiji dominates the U.S. market for women's cotton suits (some 42% of the market). Easterly likened Fiji's development of this industry to finding a needle in a haystack. How would any development expert or government agency have known in advance that Fiji would be so good at producing women's cotton suits? Isn't it better to have many dispersed searchers (Easterly's term for entrepreneurs) all looking for the needle? 

Although I am sympathetic to Easterly's view, I am left with a question: What if there are many needles in the haystack? Perhaps then a government agency doesn't need to be quite so omniscient to find at least one needle in the haystack?

Update: Easterly provides his own summary of the event. Money quote:

In the end, the question boils down to: does a poor country government have a comparative advantage in discovering a poor country’s comparative advantage?