Krugman vs. Ferguson: Letting the Data Speak

When giants like Paul Krugman and Niall Ferguson start to argue, they both sound compelling. Ferguson says that interest rates are rising because of the deficit, and Krugman retorts that Ferguson has forgotten his first-year economics. Fortunately, the data can speak, and it’s time to give them a voice. This is why I turn to my frequent collaborator, Eric Zitzewitz, who has an incredibly handy knack for getting financial data to speak clearly. Eric’s verdict? You’ll have to keep reading.

Quantifying the Nightmare Scenarios II: Why Have Long-Term Interest Rates Risen?
By Eric Zitzewitz
A Guest Post

There’s no debating that long-term interest rates on government debt have risen. But there’s a pretty fierce debate about what it means. Harvard historian Niall Ferguson interprets this as indicating that the bond market is worried about the U.S. deficit and the prospect of inflation.

Princeton economist Paul Krugman thinks it indicates that worries about deflation have eased.

It’s a high-stakes debate: Professor Ferguson is arguing that the stimulus package has counterproductively stimulated inflation fears, while Professor Krugman thinks the stimulus has worked as intended by reducing the likelihood of deflation. In fact, Krugman has argued for another dose of fiscal stimulus.

So who is right? Standard measures of expected inflation — such as the difference between 20-year nominal and inflation-protected Treasury yields — have risen sharply, from an almost implausibly low 0.9 percent in December to 1.5 percent throughout February and March to 2.25 percent last week. But an increase in expected inflation can come from an increased probability of high inflation (bad news; Ferguson’s story) or a decreased probability of deflation (good news; Krugman’s story).

Resolving their debate requires measuring the likelihood of different inflation scenarios. Let’s do it.

The graph below plots the probability of different outcomes for the yield on 25-year Treasuries on two different dates — late February and the end of last week. I calculated these probabilities using the technique I discussed in my last post, which extracts the probabilities implied by option prices on those dates. (Wonkish detail: I used the January 2011 options on the iShares Barclays 20+ Year Treasury Bond exchange-traded fund (TLT) and converted bond prices to yields using the portfolio average data reported by iShares.)


The blue line shows that there was a lot of uncertainty about future Treasury yields in February, including a very large chance of very low interest rates, as in Krugman’s deflation scenario. But the green line shows that this deflation risk appears to have receded. In fact, the recent increase in Treasury yields is almost entirely due to a reduction in the probability of the deflationary (low nominal interest rates) scenario. Score this round for Krugman.

While Ferguson wrongly diagnosed the cause of the rise in interest rates, he is right that the markets are spooked about the risk of an inflationary breakout. There’s about a 7 percent chance that 25-year interest rates will exceed 10 percent, although surprisingly, this risk was slightly higher back in February. This is a fairly extreme scenario: long-term interest rates have not been above 10 percent since inflation was tamed in the mid-1980’s. So there’s a chance that Professor Ferguson may be right about the broader issue: now that deflationary worries seem to have eased, it might be time to start turning the fiscal policy battleship around.

Finally, let me suggest that Professors Ferguson and Krugman settle their dispute like gentlemen. No, I don’t mean pistols at dawn, or the modern equivalent. Instead I suggest that Professor Ferguson purchase from Professor Krugman a January 2011 put on TLT (the iShares ETF) at a strike price of 55 (which would profit if future yields were over 8 percent) at the current market price. If the two good profs would put their money where their mouths are, the whole debate may well become clearer, more civil, and more credible than the usual cheap talk, which may be our largest remaining national surplus.

the Gooch

Nice post. Well written. I expect few snarky comments from the peanut gallery.


I like the idea of your bet. It puts some skin in the game so to speak. (and neither bettor could rig the outcome)

Robot Mistake

As a debt laden middle class american consumer how does high inflation hurt in the long run?

I have housing/ student loans/ cc debt - all will be cheaper to repay in an inflationary environment. I already bought my house so what to I care if future barrowing is more expensive?

I have grandparents that lived out thier retirement on 10-15% bond yeilds from the mid-70's

The American dollar is being intentionally devalued why else have they been changing the designs so much.

p.s. I am no economist but I believe the argument above is just two sides of the tipping point, they disagree on where the point is, but agree on the direction of the data.


Is it possible that the market believes that deflation will be the primary threat for the next several years, followed by a switch to inflation?

I'll explain:

Household deleveraging has only barely begun. So far, increased government borrowing has replaced the decreased household borrowing. But what if the market believes the US can't keep this up forever. When total private + public deleveraging starts in earnest, we may see deflation for awhile, but eventually the US will have to monetize a large portion of its debt which will then lead to high inflation. The mixture of near term deflation and long term inflation might average out to a normal looking yield in 2011. It would be interesting to see this analysis done on options for shorter term treasuries for different times in the future to see what the expected trajectory of yields will look like.


pebbles and slingshot

the greatest chess player ever, gary kasparov, got whupped by a computer.

some dang computer somewhere is messin with our markets and our heads.



"The blue line shows that there was a lot of uncertainty about future Treasury yields in February, including a very large chance of very low interest rates, as in Krugman's deflation scenario"

I think it should be"as in Fergusson's deflation scenario". Otherwise, it does not make sense to me.


what kinda debate is this?- krugman is a nobel economist, and ferguson is ?- not an economist!- so giving equal airtime to the two views is ridiculous- krugman/roubini makes more sense to me


There are an infinite number of possible histories, and we'll end up mapping only one. Meaning a bet's outcome doesn't tell you who was correct. Nor, I'm afraid, does this analysis. It's like debating the Ninja vs Spartan. The attempt is excellent for discussion however.


I would poll investment bankers on which scenario they believe will play out. Then I would bet on the other outcome.


This looks like a Caltech plot, only one axis is labeled.

Sundar Srinivasan

I don't think there is any fear of inflation among serious economists. We are in the middle of a classical Hicksian liquidity trap. Even at zero nominal interest rate, consumers are willing to save resulting in commodity prices going down. How can you assume hyperinflation in near future in this scenario?
The US government is trying its best to make a commitment for future raise in commodity prices. Why would it do so, if hyperinflation is looming? We have to check and control the inflation immediately once we are out of the liquidity trap, but not now.
Paul Krugman makes more sense and wins this debate hands down.

Dan M

Your analysis is flawed by using options. Options are priced using a model that includes an implied volatility of the total market that cannot be calculated beyond an educated guess. Normalize the options in terms of the VXV and VIX and I think you will end up with a different result.

Jeff Reisberg

Eric, a little knowledge of reserve accounting is needed to make sense of the interest rates:

If the government deficit spends, (taxes


Is it too late to double down on Krugman?


#14: Yes, given how interest rates have developed since this post I certainly hope Paul is gentle with Niall when they debate this week in Seoul. Somehow I don't expect Ferguson to admit he was wrong. But then again, he isn't really interested in the evidence, is he?


I really appreciated Eric's metric based approach and wish the results were more widely circulated.

This just shows the results you can get when you can intelligently use an actual economic model connected to reality, rather than a hip shot or an untested model you made up back in 2009 and have since discarded for the theory du jour.

I hope Krugman is STRONG with Niall. We have too many punters making up stuff and need more economists like Paul who actually take the time to analyze the data and validate their model


I think we have a clear winner. Can we get an update on how much Professor Krugman would have won.


Looks like Paul Krugman was entirely correct. The debate was not, and is not, cheap talk. And your analysis added little value.