sábado, 20 de abril de 2013
Un estudiante encuentra el error en la depresión del Excel
How a student took on eminent economists on debt issue - and won
NEW YORK |
(Reuters) - When Thomas Herndon, a student at the University of Massachusetts Amherst's doctoral program in economics, spotted possible errors made by two eminent Harvard economists in an influential research paper, he called his girlfriend over for a second look.
As they pored over the spreadsheets Herndon had requested from Harvard's Carmen Reinhart and Kenneth Rogoff, which formed the basis for a widely quoted 2010 study, they spotted what they believed were glaring errors.
"I almost didn't believe my eyes when I saw just the basic spreadsheet error," said Herndon, 28. "I was like, am I just looking at this wrong? There has to be some other explanation. So I asked my girlfriend, 'Am I seeing this wrong?'"
His girlfriend, Kyla Walters, replied: "I don't think so, Thomas."
In the world of economic luminaries, it doesn't get much bigger than Reinhart and Rogoff, whose work has had enormous influence in one of the biggest economic policy debates of the age.
Both have served at the International Monetary Fund. Reinhart was a chief economist at investment bank Bear Stearns in the 1980s, while Rogoff worked at the Federal Reserve, passing through Yale and MIT before landing at Harvard.
Their study, which found economic growth slows dramatically when a government's debt exceeds 90 percent of a country's annual economic output, has been cited by policymakers around the world as justification for slashing spending.
Former U.S. vice presidential candidate Paul Ryan, a Republican congressman from Wisconsin, is one influential politician who has cited the report to justify a budget slashing agenda.
Using the two professors' data, Herndon found that instead of a dramatic fall in growth, the decline was much milder, slowing to about 2.2 percent, instead of the slump to minus 0.1 percent that Reinhart and Rogoff predicted.
Things tend to move at a glacial pace in the world of academic research papers, but within 24 hours Herndon and his two teachers, who co-authored the report, Michael Ash and Robert Pollin, found themselves swept up in a global debate.
Herndon's paper began life as a replication exercise for a term paper in a graduate econometrics class. He expected to replicate Reinhart and Rogoff's results, then challenge the idea that high public debt caused growth to slow.
But he never got that far. Repeated failures to replicate the results roused his interest. Pollin and Ash encouraged him to pursue it after he convinced them he was onto something.
"At first, I didn't believe him. I thought, 'OK he's a student, he's got to be wrong. These are eminent economists and he's a graduate student,'" Pollin said. "So we pushed him and pushed him and pushed him, and after about a month of pushing him I said, 'Goddamn it, he's right.'"
Herndon approached Reinhart and Rogoff earlier this year for the spreadsheets they used in their paper. The two professors provided them at the start of April, unlocking the mysteries of the data that had stumped Herndon.
Herndon said only 15 of the 20 countries in the report had been used in the average. He also said Reinhart and Rogoff used only one year of data for New Zealand, 1951, when growth was minus 7.6 percent, significantly skewing the results.
Reinhart and Rogoff have admitted to a "coding error" in the spreadsheet that meant some countries were omitted from their calculations. But the economists denied they selectively omitted data or that they used a questionable methodology.
For Ash, the findings mean the claim that high public debt causes growth to stall no longer holds water.
"Their central thesis has been substantially weakened," he said.
Reinhart and Rogoff, however, say their conclusion that there is a correlation between high debt and slow growth still holds.
"It is sobering that such an error slipped into one of our papers despite our best efforts to be consistently careful," they said in a joint statement. "We do not, however, believe this regrettable slip affects in any significant way the central message of the paper or that in our subsequent work."
Now that Herndon has ably crossed swords with some of the most eminent figures in his field, he is thinking about expanding his work into a Ph.D. thesis.
(Reporting By Edward Krudy; Editing by Stacey Joyce)