Simon Johnson on development strategies:

Michael Spence is in favor of an manufactured export-led strategy. Larry Summers points out that it would be hard for all countries in the world to do this at the same time. And Robert Rubin emphasizes good governance.

Here’s a potential way to put it all together. If your governance is good, you can sustain rapid growth even based on primary commodities — Botswana would be the leading example (but if you want to go back in time, I would suggest Norway and Sweden were mostly commodity producers at key early phases of their development). But if your governance is problematic, then commodities may be more tricky for political reasons; it’s just too tempting to try to grab power and get all those “rents”.

If your governance is an issue, then a strategy based on manufactured exports may make more sense. Why? After all, manufacturing is more footloose than mining (which is obviously tied to mineral deposits). It’s the footloose nature of manufacturing that makes this approach work. If you expropriate some factories, the rest will leave. So you can build an equilibrium in which entrepreneurs believe they will have reasonably good protection for their property rights, even though the laws on the books or the courts or something else about the political environment is not ideal.

And it is striking that almost all countries that have grown fast for the past 30 or so years have done so with manufactured exports as a major focus. It’s also encouraging that exporters of commodities more recently have moved to diversify their exports — this is a key point from chapter 5 of the WEO.