The grad student who exposed Reinhart and Rogoff: They still can’t get their facts straight

April 29, 2013
April 29, 2013

Thomas Herndon is on a tight schedule. The University of Massachusetts-Amherst doctoral student of economics has a steady stream of media interviews lined up with global VIPs to discuss the far-reaching impact of his paper refuting the work of renowned Harvard economists Carmen Reinhart and Kenneth Rogoff.  But his mind is already looking ahead to more pressing matters: his looming microeconomics final next week and the piles of homework waiting for when the media attention dies down.

Herndon has become the unexpected symbol for proponents of anti-austerity measures. His paper, co-authored by UMass-Amherst professors Michael Ash and Robert Pollin, blows the lid on the widely accepted economic policy touted by Reinhart and Rogoff that has been at the heart of the global austerity movement. What started as a grad school assignment in replication unearthed fundamental errors in the spreadsheets used by Reinhart and Rogoff—altering their findings on debt-to-GDP ratios enough to call into question the validity of the measures that have been sparking global protests.

While the economists have responded, Herndon notes “in their public response, they admit the spreadsheet error that was in Table 2 but then they point to Table 1 and they say the spreadsheet error is not in Table 1. But in fact, those errors are present in Table 1. That is problematic.” Herndon sat down with Quartz to discuss his unexpected celebrity and the impact he hopes his paper will have in shifting economic policy going forward. Edited excerpts:

Why do you think your paper struck such a chord globally?

I was wholly unprepared for this. I did not think at all that the world would be so interested in my econometrics replication paper. But this policy of austerity affects people the world over. It’s really controversial and I think it’s really good to get more discussion about it. Economic policies impact the world in terms of people’s ability to have a job, a house, food on the table. It’s the human aspect of it that really drew me to economics. But bad economic policies can cause a lot of pain.

You are an outspoken critic of austerity measures. How does your paper support these beliefs?

I think austerity is really counterproductive because of the historical circumstances we’re in. Nowhere in the paper do we say public borrowing is always good. But we do think that with public borrowing you have to look at the tradeoffs. In the current time, where we have massive unemployment because of the recession and financial crisis, this is exactly the time where we would expect public borrowing to have really large positive effects and comparably fewer costs. It’s also the time we would expect austerity to have really large negative effects with comparatively few benefits, if any. Reinhart and Rogoff’s claims were more or less that in all times and for all countries, when debt-to-GDP rises above 90%, growth notably falls. That’s why they looked at two centuries and 20 advanced economies and, I believe, 44 emerging markets. That was their claim but we think that’s inaccurate. We need to look at the tradeoffs involved with public borrowing and I think if you do that, it’s really clear that right now is specifically the time when public borrowing can have very large positive effect on the economy.

If Rogoff and Reinhart’s fundamental assertion, supporting the use of austerity measures, was flawed, what alternatives should have been considered in its place?

The stimulus needed to be larger and more continued. It needed to be more targeted in to infrastructure projects, especially into the transformation into a green economy. We do research (at UMass-Amherst’s Political Economy Research Institute) that green investment creates a lot of jobs. Retrofitting buildings, for instance, to be energy efficient has to be done in the US and that would create a lot of jobs. There should be emphasis on that type of targeted infrastructure investment and also supporting state and local budgets. A lot of the state and local governments can’t borrow like the federal government can. We should support them so that there aren’t budget shortfalls at universities, so that you don’t have to layoff police, teachers, firemen. When public sector employees become unemployed, that has very negative effects on the economy. It’s hard for them to get another job because we’re in the middle of a recession. They can’t participate in the economy anymore. They’ll lose their homes, they can’t put food on the table. It is a devastating thing to do for an economy in recession.

In your views then, pushing up the public debt for the greater good will actually yield long-term benefits that will counteract the impact of higher debt?

Yes, that’s really the message. The debt-to-GDP ratio may not be the most informative measure. It’s much more likely that you can grow your way out of debt. Often times we’ve seen that when you actually do the austerity plans, it is counterproductive because it contracts the economy. So the governments receive less in taxes and debts go up anyways. The message is that austerity processes can actually raise debt loads.

Do you believe that if we had more closely examined the data supporting global austerity measures that it might have prevetned the resulting protests and hardship we’ve seen?

I think it could have been prevented. There were a lot of criticisms of austerity within the profession already. However, they weren’t taken as seriously as the austerity side of the debate. We had good criticisms and historical experience to show the balance of evidence even before our paper showed austerity to be a lot more painful and have very few, if any benefits. That was there but for one reason or another, those criticisms weren’t given as much weight as the calls for cutting the public budget. My hope is that we can renew a lot of debate and do it in a way where the criticisms that were already there can be given more weight. I think they were right and they weren’t being listened to. Hopefully we can be more critical of these policies because they’ve just been so painful for people. I hope we can stop the austerity policies and that we can really start to think of intelligent ways to use public policies to get out of the recession.

Do you think it’s possible at this stage to implement alternative solutions to austerity?

I don’t think it is ever too late to do the right thing. If bleeding a patient is bad, then stopping that is a good thing. That may be too crude of a metaphor but I don’t think it is ever too late to do the right thing. And I think we should do the right thing.

What is your view to Reinhart and Rogoff’s rebuttal, namely that they admit some errors in the spreadsheet but stand by their conclusions?

That’s certainly not our interpretation at all. We showed that the negative correlation above 90 is not statistically significant. I don’t see how that confirms their results. We challenged the claim that this (debt-to-GDP ratio of) 90% line exists for all time and space. That’s their claim, looking at two centuries and for all countries ever. But we show that the negative correlation that does exist–and that’s not causation at all–reduces the closer we get to the present time. If you look at the most recent time period of 2000 to 2009, growth in the over 90% category was actually higher than growth in the 60 to 90 category. That also is not at all consistent. One more thing that I’d like to point out–and these are points that I raised in a blog posting–in their public response, they admit the spreadsheet error that was in Table 2 but then they point to Table 1 and they say the spreadsheet error is not in Table 1. But in fact, those errors are present in Table 1. That is problematic. Also, they claim that we tried to say they were dishonest and I’d like to categorically say that this is not the message of our paper at all. We did not include a single message in the paper that implies any dishonesty. We do use the words selective and unconventional but those were merely descriptive. There could be good reasons for not using a conventional method. However, these reasons weren’t discussed or justified.  That’s a problem because then it makes it very difficult for people to replicate someone’s work if reasons for methodological work aren’t justified and transparent.

UMass-Amherst’s economics department is often called a Keynesian-Marxist hybrid. Have you received any criticism from conservative circles?

Overwhelmingly, the response has been really positive. People have been able to separate whatever intellectual influences we’ve had from the fact that there was just a gross error.  The fact that we read really fascinating books has nothing to do with the fact that Rogoff and Reinhart made a mistake. That was independent of us. There is really a role in economics for critical departments. This is a good example of that. We didn’t just buy the claim. We investigated it. That’s science, right? I really enjoy the introduction to the world that our department has gotten in the last week or so.

What is next for you?

I would really just like to pass my micro final which is in a week or so because I lost about two weeks to media interviews. I think I’ll be okay. I’d like to get my homework done and wrap the semester up. But after that, I’m finishing coursework this year and I’ve finished my two comprehensive exams. So the next step is really doing the prospectus and doing my dissertation. I do think I will continue on some type of this topic. My original plan was to replicate their work right off the bat and use some of the more advanced statistical techniques to really ask the questions about causality that hadn’t really been answered. I couldn’t get to that point because I couldn’t replicate their paper. Now I think it’s the appropriate time to look at some of those questions. But, for now, I’m just trying to keep my head above water.

We welcome your comments at ideas@qz.com.

Top News

Powered by WordPress.com VIP
Follow

Get every new post delivered to your Inbox.

Join 26,752 other followers