Torsten Persson and Guido Tambellini ask some questions about democracy and economic development in a new NBER paper (an earlier version here):

Does democracy promote economic development? We review recent attempts to address this question, which exploit the within-country variation associated with historical transitions in and out of democracy. The answer is positive, but depends – in a subtle way – on the details of democratic reforms.

First, democratizations and economic liberalizations in isolation each induce growth accelerations, but countries liberalizing their economy before extending political rights do better than those carrying out the opposite sequence. Second, different forms of democratic government and different electoral systems lead to different fiscal trade policies: this might explain why new presidential democracies grow faster than new parliamentary democracies. Third, it is important to distinguish between expected and actual political reforms: expectations of regime change have an independent effect on growth, and taking expectations into account helps identify a stronger growth effect of democracy.

Huh. This debate somtimes seems a bit muddled to me—when you start looking at things on the microscopic level, it's quite rare that any country will carry out all of its economic reforms (which, to NBER, tends to mean "liberalize early and liberalize often"), finish that up, and only then work on extending political rights. Usually these things happen in tandem. Usually they're interrelated. Little movements towards one might lead to little movements towards the other. Both "democracy" and "development" are complicated processes, and it's not always clear whether the question of "which should come first?" is as helpful as it should be. Still, it's an interesting set of findings, no doubt.

In other NBER news, Joseph Stiglitz and Sergio Godoy offer new evidence that rapid liberalization isn't all it's cracked up to be:

In particular, we look at whether speed of privatization, legal institutions or initial conditions are more important in explaining the growth of the transition countries in the years since the end of the Cold War. In the mid 90s a large empirical literature attempted to relate growth to policy measures. A standard conclusion of this literature was the faster countries privatized and liberalized, the better. We now have more data, so we can check whether these conclusions are still valid… Our results suggest that, contrary to the earlier literature, the speed of privatization is negatively associated with growth, but is confirms the result of the few earlier studies that have found that legal institutions are very important.

So much for "shock therapy," then.