When Thomas Herndon, a third-year graduate student at the University of Massachusetts Amherst, began a class assignment last fall, he had no inkling that he would end up in the middle of a furious debate over international economic policies.
But the results of Herndon’s work have sparked an academic and political firestorm over one of the most famous economic papers in decades, and over whether austerity or stimulus is the best route to long-term prosperity. On the evening of April 15, Herndon posted a paper that reported data flaws and other potential errors in an influential 2010 study by Harvard University professors Carmen Reinhart and Kenneth Rogoff. By the next morning, the paper had become an Internet and media sensation.
“It was really overwhelming,” recalled Herndon, 28. “By noon, the story had gone viral and I started getting calls from the media from all over the globe. It was incredible.”
Herndon’s paper, authored with two of his UMass advisers, created an international furor not only because a graduate student had discovered mistakes by two prominent economists, but also because Reinhart and Rogoff’s work has been used by some policy makers to justify slashing social and other government spending after the recent recession.
Reinhart and Rogoff’s study, “Growth in a Time of Debt,” argued that countries suffer low growth when government debt exceeds 90 percent of total economic output. But citing the data errors, the UMass paper argued that this connection between debt levels and economic growth was overstated, a point which critics of austerity polices, including New York Times columnist and Nobel laureate Paul Krugman, quickly seized.
“There is no threshold at 90 percent,” Krugman wrote on his blog. “The appearance of such a threshold . . . was an artifact of missing data.”
Herndon, a native of Austin, Texas, is considered a hard-working and persistent researcher who, as the events of the past few months show, isn’t afraid to take potentially controversial stands.
A 2007 graduate of Evergreen State College in Olympia, Wash., with a bachelor’s degree in economics, Herndon said he eschews political labels, but added he was attracted to the iconoclastic, left-leaning reputation of the economics department at UMass Amherst.
Herndon’s research began last fall as a class assignment to replicate the work of other economists. Herndon chose Reinhart and Rogoff’s “Growth in a Time of Debt,” but wasn’t able to duplicate the results before the semester came to an end. “Everyone thought I had just made a mistake,” he recalled, “and I felt really bad about it.”
But Herndon kept working on his own. He e-mailed Reinhart and Rogoff for their data, and when it arrived in early April, he immediately checked the figures supporting their contention that economies slow when public debt reaches 90 percent of economic output.
Herndon quickly discovered problems, including a spreadsheet error that failed to include historic economic growth rates for five countries, a typographic error on New Zealand’s economic growth for one year, and the omission of some data on New Zealand’s positive economic growth.
His advisers, Robert Pollin and Michael Ash, reviewed Herndon’s findings and ultimately helped write the working paper that was published on the website of the UMass Political Economy Research Institute. The title: “Does High Public Debt Consistently Stifle Economic Growth? A Critique of Reinhart and Rogoff.”
“I had absolutely no idea this would go global,” said Pollin, calling the ensuing furor the biggest controversy he’s seen over an economics working paper during his 31 years as an economist.
While admitting that Herndon, Pollin, and Ash correctly identified a spreadsheet coding error that led to a miscalculation in the growth rates of indebted countries, Reinhart and Rogoff argued in a New York Times op-ed that the UMass paper didn’t overturn their main finding that growth is about 1 percent lower when public debt is 90 percent or more of economic output.
They have vehemently denied the UMass paper’s assertion that they had made serious errors as a result of omissions and employed questionable weighting of data. They’ve also bemoaned the “politically charged discussion” following release of the UMass study.
Rogoff declined to comment for this story, and Reinhart couldn’t be reached.
Pollin said he has warned Herndon he has about three more years before he gets his PhD and needs to keep working hard to avoid becoming “a flash in the pan” graduate student.
“This obviously gives him a reputation and it should help [his career],” Pollin said. “But he’s still got a long way to go.”
Herndon said he isn’t sure if he’ll turn his research into a doctoral thesis, though he plans to continue to study the effects of public debts on economic growth.
“It has been an incredible experience,” he said of the past few months. “It’s given me renewed enthusiasm for the work I do.”