A 28-year-old economics graduate student has rewritten a study led by Harvard economists Carmen Reinhart and Kenneth Rogoff that has been widely cited as the intellectual basis for worldwide government austerity measures. The Harvard study argues that higher public debt slows down economic growth when the GDP rises above a 90% threshold. But after an attempt to duplicate the Harvard study’s findings, Thomas Herndon, a Ph.D student at the University of Massachusetts-Amherst, ended up debunking Reinhart and Rogoff’s economic theory and found that the Reinhart-Rogoff study was incorrect due to spreadsheet coding errors and selective data. Herndon did not attribute motive; he focused only on the statistical and computational inaccuracies of the influential paper.
Released April 17 by University of Massachusetts researchers, Thomas Herndon and his two economics professors, Michael Ash and Robert Pollin, published a paper pointing out several inaccuracies in Reinhart and Rogoff’s paper. The “RR” study has been repeatedly used as an argument for pushing austerity and for the view that government deficits are economically threatening.
Herndon’s study finds that their “results are not consistent with and do not confirm their findings” after uncovering flaws in their data analysis and computing method. Herndon clarified the Harvard study’s “selective omissions and unconventional weighting” on Monday.
“We did use the terms ‘selective’ and ‘unconventional’ to describe the problems we saw with their paper, and we believe these are accurate characterizations. ‘Selective’ is an appropriate description because the data were ‘selected’ for exclusion,” Herndon writes.
In terms of Reinhart and Rogoff’s “unconventional” weighting system, Herndon points to a Excel spreadsheet error that compounds the growth-rate error. Herndon says, “It was the combination of the weighting system with the exclusion–for whatever reason–that combined to cause the most significant fall in average GDP growth. There is nothing inherently wrong with their weighting system. However it is unusual and it is their obligation to be open and clear in explaining why they used this unusual methodology.”
Additionally, Herndon also uncovered a transcription error with Spain’s average GDP growth. In one of Reinhart and Rogoff’s tables, Spain’s average GDP growth was entered at 2.8% instead of 2.2%. Two other samples showed five countries–Australia, Austria, Belgium, Denmark, and Canada–were removed, adding to the amount of computational errors in the Harvard report.
Herndon concluded, “Contrary to RR, average GDP growth at public debt/GDP ratios over 90 percent is not dramatically different than when debt/GDP ratios are lower.”
An addendum to Herndon’s paper also defines a stronger causal relationship between economic and public debt. A contribution by his professor Arin Dube provides evidence that the causality runs the other way around — from slow growth to high debt.








