Cross-posted from Middle Class Political Economist.
This story has rapidly made the rounds in the blogosphere for two days, and it is indeed a big deal. One of the most significant economics papers underlying the argument for why high government debt (especially over 90% of gross domestic product) is bad for growth was published in 2010 by Carmen Reinhart and Kenneth Rogoff, "Growth in a Time of Debt" (ungated version here).
The basic finding of this paper was that if debt exceeds 90% of GDP, then on average growth turns negative. But as Thomas Herndon, Michael Ash, and Robert Pollin report in a new paper (via Mike Konczal at Rortybomb), there are substantial errors including data omitted for no reason, a weighting formula that makes one year of negative growth by New Zealand equal to 19 years years of decent growth by the UK, and a simple error on their spreadsheet that excluded five countries from their analysis altogether (see Rortybomb for the screen shot).
The authors say that with these errors corrected, the average growth rate for 20 OECD countries from 1946 to 2009 with debt/GDP ratios over 90% is 2.2%, not the -0.1% found by Reinhart and Rogoff. This is a huge difference. We still have a negative correlation between debt/GDP and growth rate, but it is much smaller, as we can see from Figure 3 from their paper:
Debt/GDP Ratio R/R Results Corrected Results
Under 30% 4.1% 4.2%
30-60% 2.8% 3.1%
60-90% 2.8% 3.2%
Over 90% -0.1% 2.2%
As Paul Krugman (link above) argues, what we are likely seeing is reverse causation: slow growth leads to high debt/GDP ratios. That is certainly what EU countries are finding as they implement austerity measures and slip back into recession. But even if high debt/GDP did cause slower growth, we can see it is nowhere near the crash that Reinhart and Rogoff's paper made it out to be.
The bottom line here is simple: the focus on deficits and debt that have dominated our political discourse is completely misplaced. We need to do something about the unemployment crisis by increasing growth, something that is even truer in the European Union where the unemployment rate in Spain and Greece exceeds 26%.
Update: Reinhart and Rogoff have responded in the Wall Street Journal. They emphasize that there is still a negative correlation, and that having debt/GDP above 90% for five years or more reduces growth by 1.2 percentage points in developed countries, which is still substantial for developed economies.
Update 2: Paul Krugman's response to Reinhart and Rogoff is here. He pronounces it very disappointing, saying they are "evading the critique."
Update 3: Reinhart and Rogoff have a new response in the Financial Times (registration required). Here, they admit they committed the Excel error, but claim there was nothing nefarious in their disputed data choices:
The 'gaps' are explained by the fact there were still gaps in our public debt data set at the time of the paper. Our approach has been followed in many other settings where one does not want to overly weight a small number of countries that may have their own peculiarities.This is a very odd response from two authors who equated one year of New Zealand to 19 years of the far larger UK economy. Worse still when you add the fact that by excluding several years when New Zealand had a debt/GDP ratio over 90%, they got an "average" (actually only one year) growth rate of -7.6%, when the correct average, with all relevant years over 90% included, was 2.58%, a 10.18 point swing!
It's obvious the austerity crowd is still going to stand by this paper, but it doesn't mean anyone else should be taken in by it.