It wasn’t so much the firestorm that erupted in economic blogs and across the national business media.
It wasn’t the wave of interview requests.
It wasn’t even the guest appearance on the “Colbert Report,” the popular Comedy Central program with Stephen Colbert.
No, it was the 8 a.m. start time that Thomas Herndon really worried about when it came down to his microeconomics final last week.
“It’s hard to remember all the proofs I need at that hour,” he said with a laugh. “I did well enough … but I don’t think I was as stellar as I could have been.”
One can excuse Herndon if he didn’t spend quite as much time in the books as he might have otherwise. Since April, he’s had a few distractions.
The brouhaha erupted last month, when Herndon — who grew up in Austin and still has a 512 area code on his mobile phone — and two of his professors at the University of Massachusetts Amherst published a paper that exposed some key mistakes in one of the most influential economic studies produced in recent years.
That influential study, authored by Harvard University professors Kenneth Rogoff and Carmen Reinhart, suggested a country’s economic growth will slow when its debt hits 90 percent of its gross domestic product. Their findings were used by U.S. and European policymakers to justify tighter restrictions on public spending, despite the stuttering economies on both sides of the Atlantic Ocean.
Critics of those austerity measures suggested times of weak economic demand and high unemployment require greater government spending to drive demand and create jobs.
Austerity proponents countered that public spending would crowd out private spending, potentially increase inflation and ultimately curb economic growth. Not long after it was published, the Rogoff and Reinhart paper became the academic underpinning for their viewpoint.
As part of one of the final courses in his Ph.D. program, Herndon had to replicate a major economic study. Skeptical of its conclusions, he started looking into the Rogoff and Reinhart paper.
By reconstructing some of the leading works in economics, Herndon said, doctoral candidates can learn cutting edge techniques and thinking in the field. Yet, other economists had struggled to re-create the Rogoff and Reinhart results.
Herndon’s breakthrough came when the Harvard duo agreed to share their data and worksheets with him. As he parsed through the data, he discovered a few red flags – not the least of which was a basic spreadsheet coding mistake that skewed the results.
Fair or not, the idea that such a basic error could impact a study of such national and global economic importance sent the blogosphere and national business media into a frenzy. And, of course, it gave fresh ammunition to proponents of more governmental stimulus to pull the economy out of the doldrums.
Rogoff and Reinhart admitted the spreadsheet error, but they have subsequently argued that the basic findings of their paper – that high debt levels slow economic growth – remain valid. In a recent column for the Financial Times, they urged policymakers to consider all the tools at their disposal to get the economy humming, including targeted but limited public spending.
Given current debt levels, they wrote, “enhanced stimulus should only be taken selectively and with due caution.”
For his part, Herndon disagrees with their continued suggestion that higher debt necessarily means slower growth. In his paper, he argued that from 2000 to 2009, developed countries with higher debt-to-GDP ratios actually enjoyed greater economic growth than those with moderate ratios.
“That’s what a debate is about,” he said. “We think our results refute theirs.”
But Herndon’s paper did more than add fuel to the debate about how to lift the economy and reduce the stubbornly high unemployment rates. It also put a spotlight on an ongoing discussion at the core of the economics field itself.
Unlike in medicine and the physical sciences, where findings typically are deemed valid only if they are repeatable, few of the research papers in economics go through a rigorous review and replication process. That hasn’t become a norm in the field, Herndon said.
First, it takes a lot of time to replicate such complex studies, especially when the data and models can be closely guarded secrets, often for perfectly legitimate reasons. Herndon said it took him a semester and a half to complete his study, and his breakthrough came only after Rogoff and Reinhart shared their work.
Furthermore, few academic publications will publish replication studies. In a profession that’s as much “publish or perish” as any field on college campuses, the incentive for this kind of work just doesn’t exist.
“It’s like playing bass in a rock ‘n’ roll band,” Herndon said. “It’s so important, but it’s not as glamorous. Everyone wants to be Slash or (Eddie) Van Halen, the lead guitarist.”
Of course, with this paper under his belt, Herndon is looking to play a little lead guitar, too. In the coming months, he plans to work with another professor at UMass Amherst to take a deeper look at the Rogoff and Reinhart paper. Together, they want to take a closer look at causality – do the higher debt levels lead to slower economic growth, or does the slower economic growth lead to higher debt levels?
It’s probably safe to say it won’t result in another guest appearance on the “Colbert Report” — the clear highlight of the whole affair, Herndon said. But then he never expected this paper would generate this kind of reaction to begin with.
“We’d all love to think policy makers … base their policy moves on our papers, but it’s really hard to say what the effect of any one paper or one series of papers would have,” he said. “And you’d hope there are limits to the effect any one paper could have. That was one of the lessons of this whole story.”
In the end, he said, “I’m just really happy to have facilitated the conversation.”