• Reinhart and Rogoff: another one bites the dust


    Although I am not a macroeconomist, I thought it would be worthwhile posting some links to what is currently agitating the blogo-, twittosphere, namely the debate about the potential debunking of a very influential paper by researchers Carmen Reinhart and Kenneth Rogoff on the adverse growth effects of high debt levels (above 90% of GDP).

    This review on the Bruegel blog, dubbed “the Reinhart and Rogoff debacle“, offers a pretty neat overview of the facts of the matter, including links to the original contributions and the recent criticisms.

    Difficult also to avoid mentioning Paul Krugman’s reaction to the story. The latest NYT OP-ED is probably a good place to start, and he has also had several blog posts in recent days, both on the academic controversy, and on the broader policy issue. It’s true that it is hard to think of a hotter economic policy topic right now, and this comes at a time when austerity is being increasingly challenged in many political, academic and intellectual circles (or at least it seems to me).

    Beyond this, a few thoughts come to my mind:

    – Most economists know very well the limits of the type of empirical exercise that is based on cross-country aggregate data, which is the case of the one conducted by Reinhart and Rogoff. It’s certainly not completely useless, but we have 2 decades of experience telling us about the fragility of such results, in particular when it comes to causality. How is it possible that such papers become so prominent and their lessons taken so seriously (other recent examples include of course the Alesina-Ardagna paper on taxes versus spending already mentioned in the Krugman part above, but also for more development oriented debates, the Burnside and Dollar (2000, American Economic Review) paper on the positive effect of aid on growth among others)? (on a personal note, if I, modest and unknown researcher, were to present today a cross-country exercise in a seminar in front of a knowledgeable academic audience in some high-level university, I would probably not reach the end of the seminar alive. The rules must not be the same for everybody!) In any case, I see a lot of commentators keen to use the conclusions of such papers in opinion pieces, presenting these views to the general public as if there were eternal truths, while in fact all we know is that we don’t know for sure. We need to be a bit more cautious!

    – Should papers proven to be factually or technically wrong be retracted by economic journals, as is common in general science journals? Flawed publications that are maintained are often still quoted as true long after they’ve been proven wrong (well, retracted papers also for that matter, but at least retraction allows this to be addressed). Another example that comes to my mind from the development literature is Emily Oster’s Journal of Political Economy paper on the link between hepatitis B and missing women in India (see this column by Stephan Klasen for a summary). Again, the economic profession is under such pressure to deliver correct diagnostic and to prove it’s free of conflict of interests, that we should probably think harder about how to improve our internal rules, our publication process, etc.

    Leave a comment | Share on Facebook | Share on Twitter

  • last posts

  • Archives

  • Frontières

  • links

  • Paraguay


    © Design graphique et développement : www.mrcam.fr