Emmanuel noticed the Dani Rodrik op-ed I quoted the other day, and went off:
It honestly bothers me that I was the first to notice this factual error. On the aforementioned webpage, there are supposedly 134 Facebook shares and 41 retweets, yet nobody bothered to flag this up. What worries me is that few folks are really that familiar with the literature or do not read closely enough.
This subject matter--the relationship between democracy and economic growth--has been researched to the point of becoming a cliche, but overall, econometric analysis yields a null result. There is, statistically speaking, no evidence that democracy has a direct impact on economic growth. None. Nada. Zip. Zilch. A paper that I can suggest for those unfamiliar with this area is Doucouliagos and Olubasoglu's meta-analysis--or an econometric study of studies--that provides a similar conclusion. (There's also a non-gated earlier version.) At best, democracy only has positive "indirect" effects, but it alone doesn't make it significantly more likely that economic growth will occur.
As for the freedom-and-growth shtick, save it for Wolfowitz, Perle, and Feith. To state things correctly, authoritarianism may not necessarily be conducive to economic growth, but neither is democracy. There are many political recipes for economic growth, period.
I should say that I too noticed the part he's talking about, but didn't comment on it for two reasons: first, it wasn't related to the substance of my post; second, that the relationship between democracy and economic performance is controversial. While it may be true that democracy is not "directly" responsible for economic growth as Emmanuel says above, that does not preclude it from having an indirect effect. If growth comes from stable institutions that emphasize things like property rights, free markets, and rule of law. Consider this influential paper by Barro:
Growth and democracy (subjective indexes of political freedom) are analyzed for a panel of about 100 countries from 1960 to 1990. The favorable effects on growth include maintenance of the rule of law, free markets, small government consumption, and high human capital. Once these kinds of variables and the initial level of real per capita GDP are held constant, the overall effect of democracy on growth is weakly negative.
So after controlling for other effects, democracy does not tend to have an independent effect on growth. But where do those other things -- rule of law, free markets, small government consumption, high human capital -- come from? In general, they are strongly associated with democracy. Democracy could be working indirectly through those other variables to have an effect on growth. Perhaps that is why there is little evidence that democracy directly impacts economic growth rates but ample evidence that democracy and levels of economic development are very highly correlated. (In other words, richer countries tend to be the most democratic.)
There are other ways to read this, including dependency theory and other variants of Marxism, but the correlation remains. Perhaps to that end Rodrik amended the wording in his argument (see above link), but left this part in:
[Democracies] provide much greater economic stability, measured by the ups and downs of the business cycle. They are better at adjusting to external economic shocks (such as terms-of-trade declines or sudden stops in capital inflows). They generate more investment in human capital – health and education. And they produce more equitable societies.
If democracies generally produce those kinds of institutions, and those institutions produce economic growth, then is it really right to conclude that there is no relationship between democracy and growth? Probably not. In fact, that is the precise argument of Baum and Lake:
Democracy is more than just another brake or booster for the economy. We argue that there are significant indirect effects of democracy on growth through public health and education. Where economists use life expectancy and education as proxies for human capital, we expect democracy will be an important determinant of the level of public services manifested in these indicators. In addition to whatever direct effect democracy may have on growth, we predict an important indirect effect through public policies that condition the level of human capital in different societies. We conduct statistical investigations into the direct and indirect effects of democracy on growth using a data set consisting of a 30-year panel of 128 countries. We find that democracy has no statistically significant direct effect on growth. Rather, we discover that the effect of democracy is largely indirect through increased life expectancy in poor countries and increased secondary education in nonpoor countries.
This is also the conclusion of the meta-analysis Emmanuel cites above. None of that says that authoritarian governments cannot promote public health, property rights, rule of law, etc. Some authoritarian regimes, like Singapore, have done very well in this regard. China has obviously moved quite far in that direction as well. But if nothing else it appears that democratic regimes have been better able to build and maintain stable growth-promoting institutions over times. As a result, the richest countries in the world are all democratic.
At least that's my take. As I said I before, this stuff is controversial among social scientists. There are multiple ways to read the massive literature. But to claim without caveat that there is no relationship ("None. Nada. Zip. Zilch.") between democracy and economic growth is misleading at best.
2 comments:
Sir, I object!
Note that I said there was "no evidence that democracy has a direct impact on economic growth." That's exactly what the authors you've cited say.
Next, the Barro studies are cited in the meta-analysis, so that's not a point of contention from me. Statistical power suggests the latter work is more definitive.
Your post left the impression, I imagine deliberately but perhaps not, that there was no relationship b/t growth and democracy at all. ("None. Nada. Zip. Zilch.") If you meant to say that there were indirect effects or that readers should take a more nuanced perspective then you should have done so.
Post a Comment