The Reinhart-Rogoff Debate

The interaction of economics research and ideological politics has once again brought to the forefront the blurring of advocacy and academic integrity that has become so prevalent since the financial crisis. Harvard economists Carmen Reinhart and Kenneth Rogoff published research in 2010 that has been used widely to assert that austerity is needed for economies that are both in a recession and carry a debt to GDP ratio that exceeds 90%. It has been cited for the proposition that at a debt-to-GDP ratio of 90%, high debt causes a dramatic drop off in economic growth.

The validity of the research has been disputed by Thomas Herndon, Michael Ash and Robert Pollin of the University of Massachusetts Amherst. Reinhart and Rogoff published an op-ed in today’s New York Times defending their work. Reinhart and Rogoff claim that some of the criticisms (spreadsheet errors that omitted data for countries whose names begin with A-D) were valid, but not others.

Most importantly, the editorial asserts that the original results, were not so different from the UMass Amherst findings if one looks at them carefully. Reinhart and Rogoff claim that their work should not have been taken to mean that at 90% the relationship between debt and growth becomes “nonlinear” and negatively correlated.

The fact is that the Reinhart/Rogoff work was used for precisely that purpose. For example, it was cited in the Ryan budget proposal. The Economist, which is generally sympathetic with austerity proponents, has this reaction to the Reinhart/Rogoff defense:

Having said all of that, the nonlinearity of the relationship between debt and growth at particular thresholds was clearly the main contribution of their earlier work. That result has not fared well, and while the authors note that other work has also identified thresholds, it seems to me that the difficulty in establishing the relationship and the direction of causality is grounds for a lot more humility than we're currently observing. 

The UMass Amherst piece includes the following excerpt from the Reinhart/Rogoff work:

Table 1: Real GDP Growth as the Level of Public Debt Varies 20 advanced economies, 1946–2009

Ratio of Public Debt to GDP

Let us generously assume that the editorial is accurate in its specifics. In particular, the authors take refuge in the assertion that their calculations using the median rather than the mean in averaging data reach results more in line with the UMass Amherst work. Here is how Reinhart and Rogoff dealt with the mean/median distinction in their paper:

Our main result is that whereas the link between growth and debt seems relatively weak at “normal” debt levels, median growth rates for countries with public debt over roughly 90 percent of GDP are about one percent lower than otherwise; average (mean) growth rates are several percent lower.

Is it plausible that Carmen Reinhart and Kenneth Rogoff failed to appreciate the implications of including this table or by giving equal billing to the mean and median results? Are they so naïve that they did not expect ideological purists to cherry pick the result that suited them?

And the editorial claims that their work is simply a part of a body of research that involves mixed results tending toward a negative correlation between debt and growth at high ratios. But they must have known that their results were being highlighted in the political discourse regarding austerity. They had plenty of opportunities to publicly deny the importance of their work. The story of Pontius Pilate comes to mind.

In reality, the very sloppiness of their research suggests that it was tinged with political motive. Did no one at Harvard take the time to replicate the spreadsheet that failed to cut and paste the data from countries beginning with A to D? This suggests the urgency of advocacy rather than the careful reflection of the academic.

This is part of a disturbing trend in the interface between academic economists and politics. In the past, I have written of studies that have been used by opponents to financial reform in relation to the Volcker Rule, energy derivatives and high-frequency trading. The ugly truth is that economists have an enormous stake in traditional theories that align with conservative ideologies. And in many cases, financial firms are pleased to pay for research that serves their immediate interests.

Issues of monumental importance are being influenced by this research. Overwhelmingly, the forces of deregulation and austerity have deployed the work of “experts” to their advantage. This is structural, reflecting the bias of academics and, tragically, in some cases pure greed. I recommend that readers recall the image of Columbia economist Glenn Hubbard squirming and terminating an interview that questioned his earnings from the financial sector in the film Inside Job.

Ideally, policy makers and opinion leaders should be more skeptical of all research, but it is unlikely that advocates like Paul Ryan will refrain from opportunistically ascribing truth to academic research that is no more than a statistical analysis of extraordinarily complex reality. The more reasonable solution is work like the UMass Amherst critique and a far more robust use of open-minded research by Mr. Ryan’s counterparts.