DEBUNKING REINHART & ROGOFF

I am not sure if one single paper has ever received so much press as the Reinhart & Rogoff paper that concluded that the 90% threshold for public sector debt:GDP was the death knell for growth.  They also co-authored a book called “This Time Is Different” which had the same conclusions.  There are some basic misunderstandings of the monetary system that are clearly evident in the research.  The apples to oranges comparisons between the USA and Europe are most evident and renders the paper highly flawed in my opinion.  Additionally, this research has been debunked by Randall Wray and Yeva Nersisyan, but just yesterday Robert Shiller of Yale decided to take a crack at them also (via Project Syndicate):

“A paper written last year by Carmen Reinhart and Kenneth Rogoff, called “Growth in a Time of Debt,” has been widely quoted for its analysis of 44 countries over 200 years, which found that when government debt exceeds 90% of GDP, countries suffer slower growth, losing about one percentage point on the annual rate.

One might be misled into thinking that, because 90% sounds awfully close to 100%, awful things start happening to countries that get into such a mess. But if one reads their paper carefully, it is clear that Reinhart and Rogoff picked the 90% figure almost arbitrarily. They chose, without explanation, to divide debt-to-GDP ratios into the following categories: under 30%, 30-60%, 60-90%, and over 90%. And it turns out that growth rates decline in all of these categories as the debt-to-GDP ratio increases, only somewhat more in the last category.

There is also the issue of reverse causality. Debt-to-GDP ratios tend to increase for countries that are in economic trouble. If this is part of the reason that higher debt-to-GDP ratios correspond to lower economic growth, there is less reason to think that countries should avoid a higher ratio, as Keynesian theory implies that fiscal austerity would undermine, rather than boost, economic performance.

The fundamental problem that much of the world faces today is that investors are overreacting to debt-to-GDP ratios, fearful of some magic threshold, and demanding fiscal-austerity programs too soon. They are asking governments to cut expenditure while their economies are still vulnerable. Households are running scared, so they cut expenditures as well, and businesses are being dissuaded from borrowing to finance capital expenditures.

The lesson is simple: We should worry less about debt ratios and thresholds, and more about our inability to see these indicators for the artificial – and often irrelevant – constructs that they are.”

So it appears the 90% figure is a nice piece of datamining and it makes for even better TV (and political theater), but we have to be more careful drawing our conclusions from such data before allowing it to influence public policy and the direction of the discourse.  It appears to me that this research has garnered an unjust amount of attention.

-------------------------------------------------------------------------------------------------------------------

Got a comment or question about this post? Feel free to use the Ask Cullen section or leave a comment in the forum.

Cullen Roche

Mr. Roche is the Founder of Orcam Financial Group, LLC. Orcam is a financial services firm offering research, private advisory, institutional consulting and educational services.

More Posts - Website

Follow Me:
TwitterLinkedIn

Comments

  1. Wow wonderful interpretation, totally agree. I think the productivity side is way more important than debt to GDP ratio, right? One think I still do not understand, most of the S&P companies in fact doing better than expected, with outsourcing and insist on investing heavily in emerging market because of their still growth prosper, however where their profits go to..

  2. What they fail to separate is private debt from public debt.

    Private debt to GDP ratios matter for all countries because those debts have to be funded out of output. Public debt is only added to that if it is genuine and not an illusionary relic of a long dead gold standard.

  3. Good post. Good point for Shiller. He is one of those super-smart guys whose smarts are not intimidating. He comes down as a regular folk.

  4. It’s amazing how observers don’t differentiate between productive self-liquidating debt versus un-productive non self-liquidating.

    Here’s an example:

    BMW sells $500 MM in 20 yr notes to fund construction of a new plant in Spartanburg, SC for domestic production of the X3 and 3-Series.

    GM sells $10 BN in 20 yr notes to fund its massively underfunded UAW pension plan, in which GM assumes an 10% return on plan assets thus generating huge EPS accretion due to the spread between assumed ROPA and cost of funds

    Is this debt the same? Is one debt productive and the other not?

  5. Boys and girls – always look where the money comes from.

    Anyone, anywhere at anytime can be prevailed upon to say what ever their sponsors ask them to say.

    Debt levels don’t matter. Oh really?!

    If the Greeks go from their current 175% to 500%, it does not matter. It is only numbers on a line somewhere.

    Right!

    So why aren’t all of you buying Greek debt hand over fist? Talk really is cheap.

  6. Cullen,

    I do not quite understand, why you are calling this piece debunking. Let’s look at this situation in more common terms.
    We have a heroin addict taking heroin (or a fat person eating hamburgers). His health level deteriorates as he takes more heroin/eats more hamburgers per day. Reinhart at all state that after level of consumption is above X the health deteriorates substantially. Shiller states that level X is not critical – health deteriorates below it and above it? Looks like debunking? Not to me!
    Also, he states it is not clear about causation…. Ok, the addict’s health deteriorates, and to compensate it he takes more heroin; or, he takes more heroin and after that his health deteriorates… Big deal? Not to me!
    These points here do have importance to me: the process is self-sustaining, deterioration accelerates with time, it leads to very serious problems, the earlier it stopped the easier the stopping process and the consequences to health. Here, I think, Reinhart and Shiller do agree.

  7. As said hundreds of times, debt is not the origin of the crisis/change (as media means now with their focus on immediate instead of important), it is consequence of… So what R&R say is something like when you have problem you get debts… ok, thanks for that… but you can solved assuming more debts? That is something they don’t know… I don’t believe is worthy to go further in these counter facts.
    What is important is to define how to fill the global gap existing between developed countries and others in the way less people suffer or we add political instability. Now the most are going to austerity to push those countries to support they growth by internal expending that can add global demand… that macroeconomic view is true, but my question is the timing.. How long it will take?

  8. Shiller never “debunks” Reinhart and Rogoff – in fact, he freely admits that their data and analysis of the data are correct. What he does is criticize their use of a 90% threshold as being arbitrary, which R and R clearly admit to in their paper – they themselves indicate that it needs more study. What he is really criticizing is the way the media use the 90% threshold, as if it was some sort of bright-line trigger (and note that R and R never claimed that it was). Basically he is agreeing with the data and analysis of the data, just disagreeing with the conclusions drawn by the media and others.

    As far as reverse causality, clearly debt-to-GDP ratios tend to increase for countries that are in economic trouble. However, R and R’s observations for over 90% include all years in which they are over 90%, not just the years in which they became over 90%, so it would include years in which countries are paying down debt to get below 90%. This would include years after the years of economic trouble that got them into the mess in the first place.

    My question is whether we really want to test the idea behind their title “This Time Is Different”. That is what a lot of the countries in R and R’s study thought as well. I know the U.S. is different in some ways than other countries, but in the end will these differences continue, and will they be enough?

  9. A repost from TC comment:
    http://pragcap.com/debunking-reinhart-and-rogoff/comment-page-1#comment-57028

    R&R cherry picked data. This isn’t really even a dispute any more. Most of the high debt/low growth effect came from an 18 month period just after WW II ended – and when I say just after, that’s exactly what I mean. Throw out those huge outliers and the rest of the data doesn’t show anything like that.

    Now you could debate that the R&R wanted to include all the data. But any honest appraisal of the data would have been far more upfront about what caused them to get their results. If you have a few data points after the greatest war in history that are the sole reason you are able to make a claim about economic history, you should let every reader know.

    R&R did not do that. Instead, they made a bold sweeping claim about debt load, and it turns out to be closer to a lie than to what most people would consider the truth.

    I wish TC or anybody else elaborates on this with some more meat.

  10. Holy smokes, even I(non-economist) can respond to this one:

    Greece is in debt to others that control the mutual currency(EURO) leaving one Greek option: Austerity. The US and the Japanese are in debt to themselves(and a few others) and CAN control their sovereign currencies.
    How can you improve an economic system when you impose austerity? Greek austerity is an externally imposed shrinkage on an already shrinking economy and a hugely expensive debt burden imposed at the worst possible time. Greece is a little like California, but not like the US.
    You will learn a lot here if you keep examining the differences between Greece and the US and stop assuming that they are the same. It’s about flexibility in our economic system and a rigid trap in Greece’s. A lowering of debt cost with low interest rates here is much better than forcing Greece to swallow high debt costs at a bad time. It’s as if Europe wants Greece to fail. Who here would buy into that?

  11. I wouldn’t consider this article a strong rebuttal, by any means. Sure, by itself the Debt/Annual GDP ratio is just an arbitrary fraction. But as an investor, does that mean you wouldn’t consider a Price/Annual Earnings ratio just because its an arbitrary fraction? The bond markets surely wouldn’t solely use a Debt/GDP ratio as their only indicator of a country’s economic health, in the same way that an investor wouldn’t solely use a P/E ratio to determine if a company is trading at a good valuation. P/E and Debt/GDP are two of many tools that are used in making economic/investment evaluations. And if the correlation is strong enough, who cares about the causation? Is there any causation between a stock’s return and it’s P/E ratio – or only correlation? Who cares? The bond market, just like the equities market, is a herd. You can sit on your Greek Debt screaming that Greece is fine all you want, but if the bond markets say otherwise (which it has), you do so at your own peril.

  12. Furthermore, there’s absolutely no theoretical rationale, even within their own paradigm, for using gross debt instead of net debt. For the US, the difference between the two is the SS and Medicare trust funds, mostly–so, what they’re essentially arguing is that as those two programs become more “solvent” (using the actuarial definition that most use–fundamentally flawed as it is, but just for the sake of making this one point) economic growth slows. They don’t even realize that’s what they’re saying.

  13. You’ve missed the point. The relation to P/E is not relevant. First, statistically, it’s very poor analysis by R/R; traders lose money if they don’t have robust correlations to go on, and not getting the causation right can matter quite a bit, too (just ask LTCM).

    For a nation’s debt ratio, “is the debt ratio big because the economy’s in trouble (US now) or is the economy in trouble because the debt ratio’s too big” makes all the difference in the world in terms of how you approach policy. The policy responses are completely opposite depending on which is true. You don’t ask the same question about a P/E ratio–it wouldn’t make any sense to ask “is the company in trouble because it’s P/E ratio is too low or is the P/E ratio too low because the company is in trouble?”

  14. Well, guys, can you come up with a debt ratio, that on average, could indicate trouble?
    By the way,I like Neil’s comment about differetiating between private and public debt.
    The media needs to put out more figures on that.
    I guess if private debt is too high, the only logical answer is more public debt, right?
    Don Levit

  15. “Well, guys, can you come up with a debt ratio, that on average, could indicate trouble?”

    Don, if I told you that a speed of 60mph is unsafe, “indicates trouble”, would you take me seriously? I don’t think so, because the speed on its own tells you very little. Under some circumstances it is very unsafe and under others it is entirely OK. It depends what vehicle you are riding, the road conditions and so forth.
    You cannot concentrate on one number and say it indicates “trouble”. I guess if you know you’re driving a car, then we can agree that usually 80mph and up seems too high. Fine, for debt-to-GDP 300% and up is most probably too high. Deal?

  16. Nice Post, got to read it in full to understand what he’s criticizing and where he’s coming from.

    Mr. Roche would like to hear your opinion on this.

    Few years ago during my graduate school Mr. Shiller gave a couple of seminars and during Q&A I’ve asked about warren buffet’s returns. His reply “It’s much harder to do it the way seth klarman does than warren buffet. As WB has a bit of financial engineering ( especially with the free float and the way Berkshire Hathaway is structured). I’ve asked for clarification which i’ve got from his research assistant. Since regular HF have to liquidate for cash flow to go for new investments or get new funds but for WB it’s more like an annuity and a $100 with a $5 annual payment @ 20% generates what a $100 generates at @23% (approx. (his explanantion for the annuity part is the free cash flow thrown by the subsidiaries completely owned by BH that can be used with out liquidating the investments where as regular HF’s they just get dividends if held long enough but are not close to the FCF and the float) Seems reasonable explanation.

    Just as stated by shiller lot of misinformation get’s passed around as great wisdom, in the markets. Great work debunking the misinformation, what newbies like me need to better understand the markets.

    Keep up the good work Mr. Roche

  17. David Walker was futured on finance.yahoo listing 5 consequences that would occur if the debt ceiling was not increased one of which was the following.

    5. Interest rates will rise. For every 1% rise in interest rates, taxpayers will be on the hook for an additional $150 billion in debt payments.

    If treasury yields react mainly to inflation, would this not work to lower yields as the government reduces the amount of net financial assets it emits into the private sector? Would this ulimately be disinflationary as the private sector receives less USDs from the public sector over time? Also, a dramatic enough reduction in government deficits would lead to a deteriorating nominal GDP proving positive for treasuries. Do I have this right?

    As long as the government continues its promise to convert treasuries into USD balances over time I don’t see how this would lead to higher rates. Best case for treasury holders would be for the government to cut as much as possible as long as they continue to pay interest on treasuries.

    With that said, I bet the tax payer would be on the hook, but not in the way he suggests. Those who need government assistance will have to resort to more debt in order to make up for the loss. In this case the public sector would net save more while the private sector would dissave. What if they were both to save at the same time? This would have to be positive for the dollar and negative for inflation rates. Do I have any of this correct?

  18. oops featured.. that is what I get for rushing to complete this post before my lunch break is over

  19. Debt is “bad”– economically undesirable–unless it is undertaken to fund productive activity, the proceeds of which can be used to repay the debt. Otherwise the debt must be repaid, reducing income or, in the case of public debt, via taxes or inflation. The problem with government debt is that it almost never finances productive activity, so increased taxes or inflation are inevitable, unless of course an economy is growing faster than debt is accumulating, in which case a Ponzi-like scheme can occur in which increased debt loads don’t seem to matter. But the game ultimately runs out when debt starts rising faster than GDP, as is the case with most developed countries.

  20. Right. One thing you have to keep in mind with Buffett is that he’s running a company and not just a portfolio. I think the portfolio holds like Coke, etc represent 25% of the firm’s assets. BRK is really a huge option writing company with several diversified arms. He’s writing insurance, sucking in the premiums, and reinvesting the cash flows into his other businesses. It’s genius. And it’s also 10X more complex than anyone thinks.

    You might be interested in reading this piece:

    http://pragcap.com/the-many-myths-of-warren-buffett

    Enjoy the weekend!

  21. But the game ultimately runs out when debt starts rising faster than GDP, as is the case with most developed countries.

    You are esentially talking about the Intertemporal Government Budget Constraint, which is unobservable (see this, for example: http://traderscrucible.com/2011/05/03/the-concise-way-to-destroy-the-igbc-and-why-to-destroy-it)
    Only ex post you can know whether your debt grew faster than GDP or not.
    And where is the evidence that debt rises faster than GDP in most developed countries? For US, for example, via Krugman (http://krugman.blogs.nytimes.com/2011/07/22/debt-and-forgetfulness/):

    Between 1993 and 2001, federal debt held by the public fell from 49.2 percent of GDP to 32.5 percent of GDP.

  22. Thanks Mr. Roche.I’ve been reading BH annual letters which are very illuminating and Mr. Gannon’s take on Value investing and BH ( use of leverage), wanted to make sure that i understood it correctly.

    WB might have lost close to 30% -40% of his net worth based on rolfe winkler’s calculations ( of course assuming that he will not lose the entire $7 bn as illustrated by Mr. Winkler & assuming if he was asked to post collateral like banks ask every body (sic AIG) on his derivative contracts which means he would not have made the GS & GE deals)

    Actions do speak louder than words ( Lose money i’ll be understanding, lose reputation i’ll be ruthless – WB).He sounded hallow with David Sokol & lubrizol deal and his contradictory statements on that affair.

  23. “BRK is really a huge option writing company with several diversified arms. He’s writing insurance, sucking in the premiums, and reinvesting the cash flows into his other businesses. It’s genius.”

    Bingo! And it’s not just the cash flows from the premiums and dividends. BRK loves to purchase capital intensive companies, who then benefit enormously from the access to cheap capital that comes from huddling under the umbrella of BRK’s credit ratings. MidAmerican, Burlington, etc. benefit from financial and operational advantages their competition can’t reasonably expect to match.

  24. So I am sure you personally are buying lots of Greek debt.

    Like I said – talk is cheap.

  25. Before we start celebrating we may want to look around and notice that this economy right now sucks, and it is going nowhere fast ( and that is after throwing 5 trillion dollars or more at the problem). Ray Dalio, John Taylor, Jim Rogers, George Soros, and Marc Faber to name just a few fully believe that the US and the rest of the world countries will continue making poor choices ( as in money printing) until there is a final collapse. Does anyone really believe the EU bought more than 2-3 months relief with this latest and greatest settlement? That is about how long their other bailouts have lasted.

  26. I agree this time with TPC. The best example to debunk Carmen and Rogoff is Japan. it has a debt to GDP of 200% and it still isn’t belly up.

  27. I wouldn’t touch Greek debt with a ten foot pole, but I buy US debt all the time. And I would if the US was ‘even more in debt’. You just didn’t and obviously can’t grasp the very simple point that Greek and US economic systems are very different.

  28. Ha! Bill Gross, whom I have the greatest respect for, cited Reinhart and Rogoff the other night on Charlie Rose.

  29. No one reading about that particular crisis will find any of it strange or unfamiliar least of all the 100-million-sesterces interest-free loan the emperor had to provide without even having read Bagehot in order to end the panic..So although I am not smart enough to tell you who will or won t default I have my suspicions however based on my historical reading and experiences I think there are two statements that I can make with confidence. Now at least some investors are treating it as the crisis that wasn t..The article goes on to quote Jean-Claude Trichet sniffing over the tendency among some investors and market participants to underestimate Europe s ability to take bold decisions. Of course I d be more impressed with Trichet s comments if pretty much the same thing hadn t been said before nearly every previous crisis. Before the decade ends I am pretty convinced there will be several countries including European struggling with the process of debt restructuring and some of the victims will surprise us..The second statement I think I can make with some confidence is that there is no threshold debt level that indicates a country is in trouble.

  30. The US is more different from Japan and more similar to Greece than many folks like to think (including TPC). Although Japan has giant problems it’s a large creditor nation (with a current account surplus) whereas the US is tha largest debtor nation (with a current account deficit). On top of that the US has the world’s reserve currency. And those three combined WILL come back to haunt the US in the very near future. I expect the next six months but I could be wrong.

  31. Look, the point of Rogoff’s work is that the increasing debt load of a nation reaches a point where it is very deleterious. No one cares if certain people do not understand this. Personall I don’t care if you don’t get gravity so long as you are not piloting the airplane I am traveling on.

    We are broke and we will default on our debt one way or another. Bill Gross has already said we are being defaulted on through negative real rates. Marc Faber, Felix Zulauf, Peter Bernholz, Kyle Bass and anyone understanding the Austrian School has already figured this out. Paul Krugman, Nancy Pelosi, Bernanke, Obama, that “Gang of Six” (3 Republicans and 3 Democrats in the Senate) et. al. have not and clearly never will.

    You are disagreeing with people who totally understand the problem and agreeing with people who could not run a child’s lemonade stand profitably.

  32. Trade is always a two party deal. The US carries import inflation risks for being such an import dependent nation; BUT all of the nations that are export dependent on the US carry unemployment risks tied to the very same trade. Having an export surplus is not necessarily a benefit to any nation.

  33. “We have a heroin addict taking heroin (or a fat person eating hamburgers)”

    More like… a heroin addict eating hamburgers. Did you read to the end?

    “We should worry less about debt ratios and thresholds, and more about our inability to see these indicators for the artificial – and often irrelevant – constructs that they are.”

  34. Like Lord Balfour said, nothing matters very much and few things matter at all. The size of the debt doesn’t matter at all, net interest cost does matter, but not very much (after all, 100% of GDP debt at 5% interest is a bigger burden than a 500% of GDP debt at 0.25% interest).

    If Members of Congress considered for 10 seconds that their fiduciary duty is towards taxpayers and not bondholders, they’d simply lock Fed Fund rate at 0 and would only allow Tsy to issue T-bills (<1yr), repealing Tsy authority to issue new T-notes (1 to 10 yrs) or T-bonds (more than 10 yrs). Net interest costs (especially in the out years of CBO budget estimates) would drop line a rock and a non-corrupt Congress could allocate the 2% to 3% of GDP ($300 to 450 bill.) budgetary windfall to permanent tax cuts.

  35. The original point made by Mark was about debt.
    Similarities can be broken down into debt and ability to deal with debt.
    Ability to deal with debt can be further broken down to economic system and import/export ratio.
    Then there is the whole ‘Reserve Currency’ issue.

    Japan & Greece roughly similar in amount of debt.
    US and Greece are similar in import/export ratio.
    Japan and US are systemically similar with sovereign currencies.
    Japan and Greece are policy constrained (Greece externally)(Japan internally)
    Then there is the whole ‘Reserve Currency’ issue.

    I’ll take US debt over Greece any day and THAT is where I have put my money.
    Where is yours??

  36. Of course the debt could be broken down into public and private and then cultural issues concerning private debt IE:
    Japan is moving more towards Greece in private debt.
    Americans are getting out of debt.

    But this breaking down into constituent parts brings to mind a previous mathematical discipline that I was working in back in the late 70s early 80s. It is called linear programming.(multiple equations with multiple unknowns) In LP you can assign characteristics to the constituent parts and then quantify the characteristics. Then you can blend them to get the desired final product outcome. LP is used in the manufacture of food/drink, perfumes etc. In this case you could characterize a country’s economic parts; put them together to define the whole ‘situation’ and then possibly derive from that adjustments that could be made or even the likelyhood of being able to make those adjustments given resources, population dynamics etc and use it as a tool for FOREX etc. Maybe this technique already exists. I’m not that far into econ 101 yet.