Tuesday, April 23, 2013

Mistakes

Several people have asked me to comment on the coding error found in one of the Reinhart-Rogoff papers. I have avoided the topic, since I don't think I have a lot to add to the discussion.  But because so many people have asked, here are a few observations:

1. Everybody makes mistakes. I once made an analytic error in one of my published papers and, after it was pointed out to me, subsequently wrote a correction (published version).  Finding and correcting errors is a part of the research process.  Sure, errors are embarrassing, but there is nothing dishonorable about making them.

2.  Policy should not be based on the results of a single study.  And my experience is that it never is.

3. I believe that high levels of debt and deficits are a negative for the economy in the long run.  My views on this issue have not changed substantially since I wrote about it with Larry Ball almost twenty years ago.

4. I never thought there was a magic threshold for the debt-to-GDP ratio above which all hell breaks loose.  The world is more continuous than that.

5. The coding error in Reinhart and Rogoff has gotten a lot more media attention than it deserves.  Some people on the opposite side of the policy debate have taken advantage of this opportunity to pound the drum for their views.  But just because someone in Team A makes an inadvertent excel error does not mean that everything Team B believes is true.  To suggest otherwise would be a truly egregious mistake.

No comments:

Post a Comment