UMass economists take on Harvard

The economics department at the University of Massachusetts, Amherst, has long been known for producing unconventional research—and this month a trio of professors there made headlines with a new paper debunking arguably the most influential economics study of the last several years.

That influential study is “Growth in a Time of Debt,” by Harvard economists Carmen Reinhart and Kenneth Rogoff (whose research about inflation was featured in Ideas in 2011), which argued that national debt hurts economic growth when the ratio of debt to GDP exceeds 90 percent. The paper was published in 2010, in the midst of economic debates on both sides of the Atlantic about the proper balance between fiscal stimulus and austerity. It was cited seemingly everywhere, including by Congressman Paul Ryan and high-ranking officials in the Eurozone, as proof that the only way out of our economic malaise was to start cutting government and paying down debt.

But the UMass economists—Thomas Herndon, Michael Ash, and Robert Pollin-- found significant errors with Reinhart and Rogoff’s methods. These included data omissions, questionable decisions about how to weight data, and a coding error in the Excel spreadsheet Reinhart and Rogoff used to calculate their final results. The UMass team revised Reinhart and Rogoff’s calculations in light of these issues and found that when the adjustments are made, “average GDP growth at public debt/GDP ratios over 90 percent is not dramatically different than when debt/GDP ratios are lower.”

Reinhart and Rogoff responded quickly, on the Wall Street Journal’s economics blog and with a long op-ed in the New York Times. They acknowledged the coding error but disputed the other charges and said that their fundamental result remains in tact. For lay readers it can be hard to figure out which side has the stronger case. It helps to know, then, that both liberal economist Paul Krugman and libertarian economist Tyler Cowen have written that they find the UMass critique convincing.

The whole dust-up has also highlighted a basic issue present all along with Reinhart and Rogoff’s 2010 paper: It can’t speak to causality; that is, it can’t tell whether high debt slows economic growth or whether slow economic growth leads to high debt. Reinhart and Rogoff have acknowledged this limitation all along, but it was largely ignored in the initial rush around their paper. Now it’s getting more attention.

For the UMass economics department, this moment in the limelight is consistent with a long tradition of iconoclastic thinking. On April 24, Dylan Matthews of the Washington Post had a great piece on the history of the department. He explained how in the 1970s it was remade around a coterie of Keynesian and Marxist economists. Today, faculty members still concentrate on understudied research areas like feminist and LGBT economics, and on using empirical analysis to question tenets of economic thinking—which is just what they’ve done in this case.

For another good piece on the controversy, see John Cassidy's post at the New Yorker.