Kenneth P. Thomas is Professor of Political Science and Fellow in the Center for International Studies at the University of Missouri-St. Louis. He is the author of "Competing for Capital: Europe and North America in a Global Era" and "Investment Incentives and the Global Competition for Capital." He blogs at Middle Class Political Economist.
We have witnessed an amazing week and a half in economics where one of the most famous and influential papers in recent memory has been shown to have spreadsheet errors and more significant problems in data analysis that greatly weaken the political case for austerity policies of the kind currently plaguing Eurozone periphery countries like Greece, Cyprus, Spain, Portugal, Ireland, and Italy.
I am talking, of course, about the famous 2010 paper by Harvard's Carmen Reinhart and Kenneth Rogoff, "Growth in a Time of Debt" (free version here). The supposed moral of their story was that if a country's ratio of debt to gross domestic product (debt/GDP) exceeded 90 percent, economic growth would cease.
While this was greeted at the time by some skepticism in the profession, it also supported a powerful political narrative saying we needed to cut deficits even though we were deep in a recession and jobless recovery. Hence. the paper achieved a totemic status as proof of the need for austerity.
The unmasking of this paper's aura benefited from the David-and-Goliath aspect of the story, wherein a graduate student at the University of Massachusetts-Amherst, Thomas Herndon, exposed the errors of the two famous Harvard economists. While Reinhart and Rogoff had previously published their data, they had never released the spreadsheet with the actual (erroneous) calculations until Herndon asked them for it.
Herndon discovered that they had left five countries out of their calculations entirely by a simple Excel error (Mike Konczal at Rortybomb has the screen shot and the best early analysis). As Herndon writes in Business Insider, not only had they left out the five countries in the 1946-2009 analysis, but the same five countries were omitted in an analysis covering 1800-2009.
The spreadsheet error's fundamental sloppiness has led to the popular discrediting of the paper. However, there were bigger problems. First, for reasons nowhere explained in the paper, Reinhart and Rogoff omitted several years where countries had over 90 percent debt/GDP but reasonable levels of growth. This was especially significant in the case of New Zealand, for which they had only one year, 1951, when the growth was -7.6 percent; but including all the postwar years in which New Zealand had a debt/GDP ratio over 90 percent gave an average growth rate of 2.58 percent.
This was then compounded by weighting each country equally, so that one year of New Zealand (at the erroneous average of -7.6 percent) was counted as significant as 19 years of reasonable growth by the much larger UK economy. Reinhart and Rogoff thus reached an average growth rate for countries above 90 percent debt/GDP of -0.1 percent. After correcting these many errors, Herndon, along with co-authors Michael Ash and Robert Pollin, found in their paper that the correct figure was 2.2 percent.
It actually took Reinhart and Rogoff two responses before they admitted the spreadsheet error. However, the bottom line of both of these responses was that it didn't matter: Even at 2.2 percent, there was still a trend for lower growth above 90 percent debt/GDP, even if it wasn't negative growth.
But that means 90 percent debt/GDP is no longer a bright warning sign. Pollin and Ash point out that their analysis showed that the effect of high debt weakened over time; in fact, for 2000-2009, real GDP growth is higher above 90 percent debt than between 60 percent and 90 percent.
Moreover, Konczal carried two subsequent analyses which show that it is not high debt which causes slow growth, but slow growth which causes high debt. In other words, Reinhart and Rogoff had the causation exactly backwards. The posts, by Arin Dube and Deepankar Basu, are both worth reading, though Dube's is more accessible to the lay reader.
What does this mean for the politics of austerity? As Paul Krugman points out, there is actually a double whammy against the advocates of austerity. First, this paper's reputation is now floundering. Second, austerity has now been tried in lots of places, notably the United Kingdom and the Eurozone, and both have returned to recession. While neither of these guarantee the success of anti-austerity politics, both are welcome developments if they bring an end to austerity's misguided reign.