Why Reinhart and Rogoff Results are Crap
Kudos to UMass economists for exposing the flaws in R&R’s empirical work. The conclusion that there is some magic debt ratio beyond which growth slows was never substantiated by R&R’s data–they fudged and flubbed the empirical work. True enough.
However, the whole “research” method as well as the “theory” behind it was bunk anyway, as Yeva Nersisyan and I argued from the get-go. You cannot just add up across centuries, exchange rate regimes, public and private debt, and debt denominated in foreign currency as well as domestic currency. Garbage-in and garbage-out research even if they had not made statitiscal errors.
Further, they had no theory of sovereign currency. Not only is the theory in their papers and book fundamentally unsound, I know from quizzing Reinhart during one of her presentations that she’s clueless. (And you all recall Rogoff’s attempt to explain on-camera what a credit default swap is–you would probably conclude he doesn’t know much about finance, anyway. It was almost as funny as Freddie Mishkin’s attempt to explain away his fudging of the title of his research on Iceland as listed on his CV.) Like many economists, they do not understand that a sovereign government that issues debt denominated in its own floating currency cannot be forced into involuntary default. And so they make no distinction in their book–and you cannot find out from the book which supposed “sovereign default” was actually a sovereign default. I suspect that almost all of the defaults they list were by governments that issued debt in foreign currency, but they do not provide enough information to tell.
In any case, I did an interview for Canadian Business; See here for the story:
(If clicking does not work, copy and paste into your browser.)
Here’s the full interview:
1. Q: You critiqued the Reinhart-Rogoff study very early on (2010) in another paper, but there was little or no reaction then. Why has there been a reaction now to this latest critique from UMass?
A: I think there were two factors. First there was the will to believe. Everyone wanted to believe that government debt is bad. Reinhart and Rogoff seemed to have generated evidence–no matter how shoddy their research was–that supported what they wanted to believe. Second, since they had refused to provide the data, no one could check. Hence our main argument–that the method was flawed because they had no distinction between sovereign debt and nonsovereign debt–did not resonate. In other words, our critique was a bit more sophisticated–we did not accuse them of outright fraud but rather of doing something no historian would take seriously. That is, aggregating up data obtained across various types of countries (different exchange rates, different currency regimes) over hundreds of years, just leads to nonsensical results. So our critique required thinking about the subject at hand. The critique from Umass is easy to understand: they fudged the data to fit their theory. Everyone understands that. But really, if you think about it, that is a much weaker critique.
2. Q: Will the revelations about the RR study have any impact in terms of reversing or slowing the drive toward austerity? Why or why not?
A: Oh, I think so. This was the main empirical work trotted out to support austerity. It is fundamentally flawed in numerous ways. They will not be able to overcome the fact that they fudged the data. I hope people will now rethink the silly approach. Please tell me this. Suppose they really are able to “prove” that countries on gold standards 400 years ago got into trouble if their government issued too much debt that they had promised to convert to gold. Would anyone seriously believe that has any implications for a country like the USA that issues a sovereign, floating currency? You’d have to be incredibly dumb to be swayed by that. Anyone with any understanding of modern monetary relations knows that the US government (and the Japanese, UK, Australian, and Canadian governments) can always “afford” to pay interest. There is zero risk of involuntary default on government debt of such sovereign nations. Greenspan has said this; Bernanke has said it; Samuelson has said it. Rogoff and Reinhart ought to get out of their university offices now and then and confront the real world.
3. Q: Do you think the oversights/errors made by RR were innocent or planned in order to push a particular agenda? If they were deliberate, what is the agenda?
A: I do not know if they intentionally fudged the results, but the evidence points that direction. Motive? To push austerity. Can we prove intentionality? I do not think so nor do I think that is necessary. The research is crap. Ignore it and them on this topic. They do not understand sovereign debt–if they did, they would never have conducted this “research” in the first place.
Let me say that there is one piece–and only one piece–of their research that holds up. When you have a serious financial crisis–such as the Global Financial Crisis of 2008–that does lead to slow growth. However (and this is key) the crisis was set off by private debt, not sovereign debt. Learn that lesson. Forget sovereign, floating currency, debt. It is not the problem. Euro-using countries gave up their currencies and so they can get in trouble. But most of them got high government debt ratios when they tried to bail-out their private sector. That just indicates the set-up of the EMU was flawed; it tells us nothing about a country like the US that retains its own currency.
21 Responses to “Why Reinhart and Rogoff Results are Crap”
Dr. Wray, the link to Canadian Business isn't working.
True, EMU is a bad construct. Not helped by Mr Rehns advocacy of austerity, based on the deeply flawed study of R&R. The problems for each country in the Eurozone reverting to the sovereign would be extreme. The elephant in the room is Germany. German economic strength leads to a relatively over valued currency throughout the zone,except Germany. Therefore money flows in but not out. Time to return to the Deutschmark ?
Nice job, Randy. IMO, more than most places, Canada is the country that needs to hear this. Canada's experience in the 90s has converted nearly all 'progressives' toward the pro-austerity camp.
In many ways, other than the last of the Old Keynesians here and the U of Ottawa guys, Marshall Auerback is the only person worth listening to on TV up here.
Speculation here and elsewhere on possible motives of Reinhart and Rogoff distracts (imo) from the vital issue, which I haven't yet seen aired (if I missed it, apologies).
"No decision can be better than the information on which it is based"
– either the paper was not peer-reviewed, in which case economic decisions resulting in frightening permanent losses between publication and the present apocalypse were influenced by the opinions of two persons;
– or it was peer-reviewed, so a review review is appropriate – and one not only by economists, as rigour is rigour across science. By general quality assurance norms, it appears this paper should never have been published as such.
Would proper review have caused disclosure of the key data promptly?
the crisis was set off by "Sudden Debt" not by some mystical "private vs public debt." since none of the monetary authorities believed a housing bubble existed once "the error was realized" (the Great Moderation was total b.s.) Greenspan was out and Bernanke and the Plunge Protection Team were in. so while the debate here sounds interesting enough the problem still exists on how to restore growth to the economy and get tax receipts heading higher. the answer of course is restoring job growth…debt is the obvious inhibitor (just look at how well Germany is/was? doing). complex problem…seems like we're not moving the ball forward here with this simplistic argumentation.
Unfortunately, the academic review process normally does not catch “errors” in the empirical work. Indeed, it is not at all uncommon to simply make up the data. Replication of empirical results is almost never attempted, and when attempted is almost never successful. The data is almost always “massaged” and authors usually are not willing to supply it. Always read empirical works with the proverbial “grain of salt”. It was obvious to us (Yeva and me) that the results were driven by a handful of cases–so while we did not know they had “neglected” to include contrary cases, we were sure the results were not robust.
cut and paste into your browser.
Wonderful straight talk! I wonder if Supply Side Economics (magic trickle-down) was b.s. too. It has that whiff of falling short of reality just to cheer lead for a certain political agenda.
yep. Look at it this way. If you “free” markets, do you expect wealth to trickle up (to those who have power) or down (to those who have none)? Supply Siders supposed it would trickle down from the powerful to the weak, I suppose on the belief that the rich are philanthropists or fools.
How many times have Brasil, Argentina and Greece defaulted on their sovereign debt?
The occasional currency peg notwithstanding, it is factually incorrect to assert that "a sovereign cannot be forced into defaulting on their floating rate currency denominated debt".
The EXPonential function does in fact "force" sovereigns into default, even if they call "Cruzeiros" "New Cruzeiros" and the "Cruzados" before calling them "Real"….
Whether or not Rogoff and Reinhardt were sloppy does not alter the parabolics of this mathematical function.
Thanks and thanks. Is there a regular tv show Marshall Auerback appears on.
Does he have a regular blog? And thanks.
stanley grouch: check your facts. Argentina was on a currency board; greece is on euro. No idea what you have in mind with regard to Brazil.
Marshall is now at Inet. He doesn’t have regular TV show but is often interviewed. Go to Inet to see his interviews of others (including Moi).
He appears on Canadian cable business news BNN and Fox Business in the US. He writes a blog here:
You should monitor the New Economic Perspectives blog also (they cover his TV appearances)
Cause of this economic slump is "Wealth Shock". US households lost about 8 trillion dollars in housing wealth and that got them saving. They have become too poor to consume at levels that are consistant with full employment.
Government spending gives us money and therefore increases private sector wealth. Government taxing does the opposite and decreases private sector wealth. A budget that is balanced is wealth neutral.
We can call that partion of the government spending that is not neutralized in taxes "Wealth Enhancing Spending", because that is what it does, it increases wealth in the private sector.
Solution to our economic mess is to have government spend more than its income so long that wealth in the private sector is increased to a point where households feel they can afford to spend as much as is required for economy to operate at full employment.
I hate to jump on the Schadenfreude-bandwagon, but several of us have noted similar problems with Reinhart and Rogoff's other studies, including her "Fear of Floating" article (misunderstanding the IMF's de facto categories) and their 2002 exchange regimes work (used the wrong exchange rates).
Hence, a larger question for me is why practically no econ journal has blind reviews. Yes, having famous authors and snazzy factoids increases the Impact Factor, but there's a huge conflict of interest and a high likelihood of failure. I hope some impetus comes out of this to align us with other academic disciplines.
Great read. I have one point to make regarding your statement:
"When you have a serious financial crisis–such as the Global Financial Crisis of 2008–that does lead to slow growth. However (and this is key) the crisis was set off by private debt, not sovereign debt."
I would argue the private debt MAY have been the result of the government running deficits that were too small. First we had the Clinton surpluses, then back to small deficits and an increasing current account deficits that started back in 1998. For example, from 2005-2007 the deficit as a percent of GDP was (-2.52%), (-1.86%), (-1.15%) respectively. The Trade deficit was (-5.3%), (-5.9%), (-6.0%) respectively. We should have been running much higher deficits in those years to offset the huge trade deficits..
Now, the above also requires some acknowledgement of other financial shenanigans that needed to be corrected as well around private debt.
Funny, it took 3 years to discover all those flaws. Who were the reviewers of that paper? I scientific reviewing still working?
"Garbage-in and garbage-out research even if they had not made *statitiscal* errors."
Please correct this typo!
Who saved the huge problematic private debt? The state. How? By lending money, on as well, huge scale – still not enought as it should…, and partly (in prety big portion…) incrising the public debt to certain entities (state borrowers departments, and related international private lenders as lending contractors). Furthermore, tax payers, pay and garantee it all!, and they are in significant extent affected by public (as well private…) debt. Could be said a lot more, but reserch, and you'l get the point.
World is not black and white, but "grey coloured". As well, you can not difirentiate the public and private debt, there is a more latent and sinister relation than You try to convince/persuade ppl.
Rightus: Yep. $29 trillion and counting. Probably won’t save the banksters, tho.