I am sure many of you have seen the recent study by Thomas Herndon, a 28-year-old economics grad student at UMass Amherst. If you have not, it is titled "Does High Public Debt Consistently Stifle Economic Growth? A Critique of Reinhart and Rogoff." (At the cite you can read the article, download the data, see responses to it and so forth.)
For a very good quick overview click here.
Here is an even shorter summary:
The article is a devastating critique of the recent austerity approach advocated in the states and other countries, which cynically seeks to cut and slash federal and state government into fiscal submission, while big business continues its massive growth into highly dynamic global markets, making more money than ever before.
What makes it so devastating is that Henderson uses Reinhart and Rogoff's own data: it appears that the initial authors made a major, major, glaring database error, forgetting to properly code information correctly; the initial authors also left out three of the countries (Canada, New Zealand, and Australia) where a charitable and relaxed approach to spending resulted in significant success.
What also makes the paper interesting from a complexity sciences perspective is the issue of curving fitting and Herndon's approach to nonlinearity versus the approach taken by the initial authors.
(The Figure to the right was taken from Herndon's paper.)
Anyway, I think it is a very good paper to read; and, for those of us who teach statistics and method, it is a wicked reminder to share with our students: always, always, always, check your data; and then check your data again!!!!!!! Out of all the things my mentor, Galen Buckwalter taught me, that is one of those things burned onto the back of my brain.