Site Meter

Friday, May 31, 2013

Noah Smith and me

I bring a comment thread debate over here.


Me:
I claim DSGE models generally need one arbitrary constant per phenominon to explain...The number of exogenous unexplained random variables keeps growing...DSGE seems not to have pulled away with this paper.

Smith:
I'd say instead that it's not clear that DSGE has pulled away. A new parameter has been added to explain a previously ignored phenomenon. But that doesn't mean that the model will then fail to "pull away" by explaining more phenomena in the future without the addition of yet more parameters.

me:
hmm I can't resist arguing about grammar.  "DSGE seems not to *have* pulled away" is in some past tense.  I didn't make a claim about what "will then" happen in the future. Well uh tenses uh grammar.  I stand by my claim.

Me: Fitting is not forecasting. It is easy to be wise after the fact.

Smith: Yes, sure. BUT, keep in mind that pseudo-out-of-sample forecasting, while not the same thing as true out-of-sample forecasting, is also not the thing as fitting.

Me:
I think that pseudo out of sample forecasting can be substantially the same as fitting.

The ways this can happen are
A)one tries something find it doesn't fit pseudo out of sample then one tries something else. This is fitting.  the Computer isn't doing all of the work but the process is trying different things till one gets a good fit.

B) Pseudo explaining an ad hoc reduced from model which fits the data.  First fit.  Then figure out a model which implies the coefficients of the reduced form model. In this case, step 1 is look for dramatic extraordinary behavior of time series in 2008 (many are well know and were discussed a lot in 2008-9 especially including huge quality premia).
Step 2 toss one in a VAR along with the DSGE state from some sort of filter. This is fitting not forecasting
Step 3 motivate the variable as an indicator of an otherwise hard to observe rare shock and again filter so that the deep disturbance in the problematic period is basically that newly introduced shock.

I think we have a case B and that step 2 is here

http://libertystreeteconomics.newyorkfed.org/2012/04/forecasting-the-great-recession-dsge-vs-blue-chip.html

my thought from 4/16/12

"'The other difference between this model and SWπ is the use of observations on the Baa-ten-year Treasury rate spread, which captures distress in financial markets.'

is a red flag. How is this incorporated ??? I think it is an ad hoc add on based on regressing outcomes on DSGE forecasts and the spread."


8:45 PM


Thursday, May 30, 2013

Contra the Renegade Noah Smith

Get off of my lawn.

update: No not you Thomists.  Also more here
Please click this link.

Noah Smith has something nice to say about a DSGE model.


I get really really cranky in comments.  this post is pure curmuggeon therapy and not worth reading.  the story is a DSGE model plus the financial accelerator due to Bernanke Gertler and Gilchrist (1999) fits the data from the great recession OK:

I comment.

Wow you were impressed ! I am feeling very very cranky, partly as i feel abandoned on anti-DSGE Island.  I haven't read the paper, so my comment isn't worth reading.

That said, I will now imagine a superficially convincing but invalid article with that title and abstract.

1) Recall the absolutely key fundamental insight  "making predictions is difficult, especially about the future" Yogi Berra  [update spelling corrected thanks to anonymous -- I'm a worse speller than the average angrybear booboo].  What we have is 4 years after the event a working paper in which a DSGE model manages to fit the now long known data.  Fitting is not forecasting.  It is easy to be wise after the fact.

2) I claim DSGE models generally need one arbitrary constant per phenominon to explain.  In this case it is the correlation between a quality premium and subsequent output growth.  Scientific success occurs when the number of stylized facts fit or correct predictions made exceeds the number of degrees of freedom.  A failed approach is neck and neck with the data.  DSGE seems not to have pulled away with this paper.

3. The number of exogenous unexplained random variables keeps growing.  2008 was extraordinary, because a large but not post WWII unprecedented downturn was associated with huge huge yield spreads.  I am pretty sure that they are ascribed to something unobservable.  In Bernanke Gertler Gilchrist 1999 quality premia change endogenously with GDP.  Ascribing 2008 to a financial shock will not shock anyone.  The question is whether we can predict such financial events.  A model with exogenous shifts is the variance of ability across entrepreneurs does not do this.

4)is there anything added by the D or the GE ?

In Bernanke Gertler Gilchrist 1999 the effect of investment decisions by other firms on sales by this firm and so the interest rate it must pay to borrow are modelled.  There is something genuinely general equilibrium there.  I concede that it is a good paper in the (fairly recent) theoretical macro literature (I think this may be my only such concession except for related papers by over-lapping sets of authors).

But you can get a good fit to recent investment with an accelerator and an even better fit if you toss in "loan officer report on lending conditions"
http://www.econbrowser.com/archives/2012/12/investment_why.html

Consumption is always well modelled as a backward looking smoothed series of aggregate disposable income.  Real output is well modelled as aggregate demand.  I think inflation is well modelled with an adaptive expectations augmented Phillips curve.

Models which are, by now, ancient, fit the data very well certainly needing no more fiddling than the DSGE model which fits OK.

What is the point of DSGE ?  The idea is that a micro founded model makes useful predictions following extraordinary events (so just looking what happened last time isn't OK).  That sure didn't work for DSGE in 2009.  Or also that they work across changes in policy regimes.

So a multiplier accelerator Phillips curve model would make the same predictions whether or not the FED were attempting extreme non standard monetary policy.   Uh it did. and it fit the data.  Fancier models made it possible to fantasize about manipulating expectated inflation and therefore getting higher investment than one would guess with an accelerator.  The linked study [update: the Goldman-Sachs study cited by Menzie Chin at econobrowser] shows that this was a step away from reality.

Comment on Footnote

I comment on the comments thread to David Warsh's post here (there I neglected to put quotation marks around the Meegs quote).

  1. Robert Waldmann wrote:
    Your comment is awaiting moderation.
    This is an interesting comments thread. I noticed no attempts to defend the post at all. Since many of the comments point out false claims of fact in the post, defending it would be a challenge. However, our host should read the comments and either defend his claims or retract them.

    Meegs

    "There is an alternate universe–one I really wished we lived in–in which BRAD DELONG AND PAUL KRUGMAN Ball …
    Alas, the Keynesians led a politically INeffective campaign for fiscal stimulus."

    They campaigned for both. In particular, after roughly 2009 they switched their focus from fiscal to monetary policy, basically because they knew there was no way another large stimulus would get through Congress (Krugman by the way, argued in early 2009 that the ARRA was the only chance for fiscal stimulus so this is just another correct Krugman prediction).
    DeLong especially is quite casual about the difficulty of obtaining higher inflation. I have regularly argued with his claim that another QE will work, that the latest QE did work and that of course the latest QE didn’t work, because Bernanke didn’t say the exact right magic words when announcing it.
    Now only in an alternative universe could a DeLong/Krugman campaign be politically effective, since Republicans are against whatever they are for.
    Thursday, May 30, 2013 at 2:58 am | Permalink

Monday, May 27, 2013

Antonio Fatas on QE and Share prices

Antonio Fatas writes 

What will we learn the day Ben Bernanke announces that we are starting that path towards normalization? It might be that we simply learn that he is becoming optimistic about growth in the US. This will be good news. [skip] this has to be good news. 
There is a second and more pessimistic scenario: the day Ben Bernanke announces that QE is ending we learn that the economy is not doing much better but that the FOMC has simply changed their mind. 
[skip] what matters is how the stock market will read the communications of the central bank. The words chosen to communicate their actions at that point  ...  will make a great difference.

 I can't comment there. I comment here.  First I don't see why it has to be good news that Bernanke is optimistic.  It isn't as if traders have such deep respect for his judgment (I do but I have never bought a share of stock).  But more how can Fatas know that the words chosen by the FOMC will make a great difference.  The QEIII and QEIV announcements don't seem to have made a great difference (the S&P 500 index went in opposite directions on the two days if I eyeball FRED graphs correctly).

Also see the post below. Fatas is assuming that traders understand that QE is not pumping up share prices aside from it's effect on future GDP growth and such like.  But many are sure that QE is pumping up share prices somehow.  The assumption is that people don't believe something which makes no sense to Fatas even though they claim to believe it.  I don't think that is a good assumption.

On Dow on Titans of Finance, QE and share prices

Mark Dow writes

So if the banks and shadow banks can just as easily repo their Treasury and mortgage holdings to finance lending, and there is no link between base money and credit creation, why is the Fed doing QE in the first place?
By keeping rates low well out on the yield curve and providing comfort that the Fed will be there to fight the risk of recession and deflation, it creates an environment that enables, over time, a normalization of risk taking in the real economy. Our revealed belief is that the Fed can chop these nastier outcomes off the left-hand side of the distribution. As a result we start feeling better about getting our money back out of the mattress and putting it back to work.
and

This is not a semantic point. I can hear traders saying, “yeah, whatever, who cares, don’t fight the Fed, just buy.” But this concept has huge implications for the phase where the Fed decides to remove the training wheels. If the Fed money is not directly propping up the stock market and the economy underneath has been healing, the much talked about wedge between “Fed-induced valuations” and “the fundamentals” is likely considerably smaller than the consensus seems to think. It’s less “artificial.” In short, what all this means is the day the Fed lets up off the gas might give us a blip, or maybe that long-awaited correction, but ultimately the Policy Bears will end up getting crushed, again.

I comment

You assume that QE must be driving share prices up somehow.  You explain how it isn't as simple as many think (I didn't know about them).  But what if this isn't happening at all ?

So why is the Fed doing QE.  Well we are in uncharted terrritory.  I have great respect for Ben Bernanke, but that doesn't mean that he can tell if QE works without trying it. That woul require not just being a genius (I'm sure he is) but having rational expectations and he is human.  Also one uses the tools one has.  Even if QE doesn't work very well, it's all they have left at the zero lower bound.  It's better to light one small candle than curse against the darkness.  It is also better to buy $600,000,000,000 of Treasury notes (QE II) than curse against the darkness.

In fact, the post as a whole seems to me to suggest that QE has caused lower stock prices.  Convincing titans of finance that something is driving prices above fundamentals should drive prices down as arithmetic says that this belief causes them to estimate a lower fundamental value.  And they have titanic risk bearing capacity.  The fear of a correction when the Fed takes the training wheels off should cause lower prices. What if that is the main effect of QE on share prices.

Correlation is not causation. Especially the argument thta these things happened in the same quarter so one caused the other is very unconvincing.  I agree that the standard event study depends on the false EMH, but the modest announcement effect of QE III and wrong sign daily shift of QEIV should make you worry no ?

On Dean Baker Vs Paul Krugman

In a rare treat Dean Baker disagrees with Paul Krugman.  Sadly, I agree with Paul Krugman (but he just has to write something with which I disagree soon, I am sick of waiting).

Via Mark Thoma who makes a very interesting contribution to the debate



Baker

Paul Krugman misrepresents the central focus of the left-right divide in national politics. He tells readers:
"Start with the proposition that there is a legitimate left-right divide in U.S. politics, built around a real issue: how extensive should be make our social safety net, and (hence) how much do we need to raise in taxes? This is ultimately a values issue, with no right answer."
This is not an accurate characterization of the left-right divide in U.S. politics since there is actually little difference between Republicans and Democrats or self-described conservatives and liberals in their support of the key components of the social safety net: Social Security, Medicare, Medicaid, and even unemployment insurance. Polls consistently show that the overwhelming majority of people across the political spectrum strongly support keeping these programs at their current level or even expanding them. 
Note the bold. Krugman definitely didn't say it was the left-right divide. Krugman went on to discuss illegitimate left right divides, in which one side (hmm always the right) denies reality.  I think that Baker is just noting that, as Krugman went on to argue, reality vs fantasy is the main divide in US politics.

More Krugman

There are, however, a lot of largely empirical questions whose answers need not, in principle, be associated with one’s position on this left-right divide but, in practice, are. A partial list:


I don't think that Krugman claimed that the legitimate debate divided the population roughly equally.  He just said it was legitimate.  The consistent advocates of small government may be few, but Krugman didn't say that they are numerous.  Also, of course, the support for the huge domestic programs is not unanimous, so there is a debate, even if it is not the electorally crucial debate.

In fact, Baker tends to support the conclusion that Krugman reached.  Krugman's main claim is that the political debate is, to a large extent, the reality based vs the fantasy based.  Baker correctly notes that self identified conservatives and Republicans want to expand the really big programs, so when asked program by program they propose higher government spending. However self described moderates (and some liberals) claim to support lower government spending in the abstract.  The real struggle is the vast majority of US adults Vs arithmetic.

I'd add to Krugman's partial list of 5 debates over simple matters of fact 6 more (and there are many more like this)

6. can you increase spending on Defence, Social Security, Medicare, and Medicaid (which is about as popular as Medicare) and cut total government spending roughly in half (the median fraction described as "wasted")
7. Is the Federal Bureaucracy over 5% of total employment (it's about 1% not counting the DOD and the post office)
8  Is the foreign aid budget over 10% of the total federal budget (the median answer in a poll. it was over 20% in another -- it is about two thirds of one percent of the federal budget).
9. Can people on TANF an food stamps live a middle class lifestyle
10.  Is the food stamp budget over 10% of the Federal budget (it is about 2.3%).
11. Does more of the Federal welfare budget go to bureaucrats or to poor people (OK that may be Michelle Bachman vs reality and the majority of her fellow citizens)

The difference between my 6 and Krugman's 5 is that elite conservatives know the answers to my 6 and admit that they do when debating with elite liberals. Also many moderates and liberals make totally incorrect guesses about the federal budget which are wrong in the same direction as conservatives' guesses.

The fact that the programs which many people want to cut are small does not affect the political debate or elections, because those people have convinced themselves that they are large.

I think Dean Baker responded to Krugman's throat clearing (aka his Obamanation) and neglected the point he wanted to get to, to be sure, only after inserting a "to be sure" to be sure he wouldn't be accused (again) of claiming no one can legitimately disagree with him.

update: This post cleaned up when I found Thoma had linked so someone might actually read it

I Fisk Reinhart and Rogoffs Open Letter to Paul Krugman

Cambridge, Massachusetts, May 25 2013
Dear Paul:
Back in the late 1980s, you helped shape the concept of an emerging market debt overhang.  The financial crisis has laid bare the fact that the dividing line between emerging markets and advanced countries is not as crisp as once thought.  Indeed, this is a recurring theme of our 2009 book, This Time is Different:  Eight Centuries of Financial Folly.  Today, the growth bind of advanced countries in the periphery of the eurozone has a great deal in common with that of emerging market economies of the 1980s.

Yes their debts are not denominated in a currency of their own.  The pattern is clear – borrowing in Euros is risky as is borrowing in dollars for countries other than the USA.  Following DeGrauwe (and  I’m sure countless others) Krugman has repeatedly stressed the difference between debts in a currency the debtor can print at will and debts in other currencies.  Let’s see if R&R ever mention the fact that debts in the eurozone are denominated in Euros.


We admire your past scholarly work, which influences us to this day.  So it has been with deep disappointment that we have experienced your spectacularly uncivil behavior the past few weeks.  You have attacked us in very personal terms, virtually non-stop, in your New York Times column and blog posts.  Now you have doubled down in the New York Review of Books, adding the accusation we didn't share our data.  Your characterization of our work and of our policy impact is selective and shallow.  It is deeply misleading about where we stand on the issues. 

The two sentences discuss fundamentally different issues.  The policy impact is not the same as their stand on policy.  This should be obvious to everyone who reads newspapers.

And we would respectfully submit, your logic and evidence on the policy substance is not nearly as compelling as you imply.

You particularly take aim at our 2010 paper on the long-term secular association between high debt and slow growth. That you disagree with our interpretation of the results is your prerogative.  Your thoroughly ignoring the subsequent literature, however, including the International Monetary Fund's work as well as our own deeper and more complete 2012 paper with Vincent Reinhart, is troubling.  

The idea that post WWII OECD experience is what is relevant to post WWII OECD policymakers is not eccentric at all.  More data does not mean more data which are relevant.  Krugman has explained why he considers data from countries which borrow in foreign currencies or which are on the gold standard to be irrelevant.  He has explained why he focuses on the data he uses.  The claim that a paper which ignores these issues is “deeper and more complete” than a paper where they are not relevant is unsupported by logic or argument.

 Perhaps, acknowledging the updated literature-not to mention decades of theoretical, empirical, and historical contributions on drawbacks to high debt-would inconveniently undermine your attempt to make us a scapegoat for austerity.  You write "Indeed, Reinhart-Rogoff may have had more immediate influence on public debate than any previous paper in the history of economics."
Setting aside this wild hyperbole,

If this is hyperbole, it should be possible to come up with an example of a paper which has had more immediate influence on public debate.  R&R don’t, because they can’t. In any case, the assertion that a claim is “wild hyperbole” should be based on something not nothing.  The claim is very likely true.  Strong claims are not   necessarily hyperbole.

you never seem to mention our other line of work that has surely been far more influential when it comes to responding to the financial crisis. 

The claim “surely been more influential” is completely unsupported by any evidence.  Note that R&R seem to ignore the phrase “public debate.” Krugman was not discussing influence on academic economists.  Do R&R claim that the paper was quoted more often in newspapers in congressional and parliamentary debates, in speeches by policy makers ?  In the context of a discussion of “public debate” the claim is plainly obviously false and R&R must know it.

Specifically, our 2009 book (released before our growth and debt work) showed that recoveries from deep systemic financial crises are long, slow and painful.  This was not the common wisdom at all before us, as you yourself have acknowledged  on more than one occasion. 

Krugman also repeatedly noted that recoveries from deep systemic financial crises are long slow and painful before the book was published.  The books is a work of economic history which adds a lot.  But R&R should have noted that Krugman had critiqued the alleged common wisdom before their book was published.

Over the course of the crisis, and certainly by 2010, policymakers around the world were using our research, alongside their assessments, to help justify sustained macroeconomic easing of both monetary and fiscal policy fronts. 

This claim is obviously false.  It asserts that “by 2010, policymakers around the world  … justify sustained macroeconomic easing of … fiscal policy. “  Which policy makers ? What easing of fiscal policy ? The claim is absurd on its face.  There was not fiscal policy easing in much of the world in 2010.  R&R seem to be totally out of contact with planet earth.  Now it may be that policy makers were using their research to advocate fiscal policy easing.  I’d be interested in names, dates and citations, since I know of no such policy makers.  But the claim that there was fiscal policy easing to justify is totally nuts.  Also Krugman made a relative claim. Influential does not mean more influential than the R&R 2010 AEA presentation. 

I google reinhard rogoff ninety danger Circa 585.000 risultati (0,33 secondi) 
I google reinhard rogoff this time is different Circa 400.000 risultati (0,17 secondi) 

Not so surely.

Your desire to blame our later 2010 paper for the stances of some politicians fails to recognize a basic reality:  We were out there endorsing very different policies. 
This is a plainly false assertion on a simple matter of fact  Krugman has repeatedly noted that basic reality
http://krugman.blogs.nytimes.com/2013/04/26/grasping-at-straw-men/they deny having been strong austerity advocates”


R&R’s claim of fact is false.  They are literate and claim to have read Krugmans critiques of them.  .I found the proof of the falsehood of their claim that Krugman failed to recognize a basic reality in a few minutes by googling krugman reinhart rogoff..
Anyone with experience in these matters knows that politicians may float a citation to an academic paper if it suits their purposes.  But there are limits to how much policy traction they can get with this device when the paper's authors are out offering very different policy conclusions. 
This is a claim about history which is plainly false.  Arrow is a social democrat but his work has been used very successfully to argue that markets work an elections don’t.  Phillips was horrified by the use made of his scatter plot (read Zombie Economics).  Uh Godwin’s law violation warning.  Darwin did not advocate genocide but that did not put limits on Hitler.  Rousseau said that France was not capable of Democracy and any effort to establish it in France would fail, but that didn’t limit Robespierre.  The claim is obviously false. The episode of R&R’s AEA presentation is proof enough to anyone who reads the papers, but human history is full of proof that the claim is plainly nonsense.  No sensible person could possibly believe such a thing.  I am sure that R&R don’t except when it is convenient to them.
You can refer to the appendix to this letter for our views on policy through the financial crisis as they were stated publicly in real time.  We were not silent.
Very senior former policy makers, observing the attacks of the past few weeks, have forcefully explained that real-time policies are very seldom driven to any significant extent by a single academic paper or result.
Very seldom does not mean never.  Note that no current policy makers deny that their AEA presentation drove policy debates to a significant extent. I know of no one who has made that claim about that particular paper, because it is false. R&R discuss the general pattern and assert that if something is “very rare” it couldn’t have happened recently  Such plainly invalid arguments made by such smart people are  “very seldom”.
It is worth noting that in the past, polemicists have often pinned the austerity charge on the International Monetary Fund for its work with countries having temporary or permanent debt sustainability issues.  Since its origins after World War II, IMF programs have almost always involved some combination of austerity, debt restructurings, and structural reform.  When a country that has been running large deficits is suddenly no longer able to borrow new funds, some measure of adjustment is invariably required, and one of the IMF's usual roles has been to serve as a lightning rod.   Even before the IMF existed, long periods of autarky and hardship accompanied debt crises. 
It is also worth noting that this has nothing to do with the debate between Krugman and R&R.  They are attempting to tar him with guilt by association with IMF bashers. 
Now let us turn to the substance. The events of the past few weeks do not change basic facts and fundamentals.  
Some Fundamentals on Debt
First, the advanced economies now have levels of debt that surpass most if not all historic episodes. It is public debt and private debt (which often becomes public as a crisis unfolds). Significant shares of these debts are held by foreigners in most cases, with the notable exception of Japan.  In Europe, where the (public and private) external debt exposures loom largest, financial de-globalization is well underway.  Debt financing has become an increasingly domestic business and a difficult one when the pool of domestic saving is limited.

Note no mention of the Euro.  Does the claim apply to European countries outside of the Eurozone ?  R&R completely ignore the issue of domestic denominated vs foreign denominated debt.  They must know it makes a critical difference.

As for the United States: our only short-lived high-debt episode involved WWII debts, which were held by domestic residents, not fickle international investors or central banks in China and elsewhere around the globe.  This observation is not meant to suggest "a scare" in the offing, with bond vigilantes driving a concerted sell-off of Treasuries by the rest of the world and a dramatic spike US in interest rates. 
Even when denying that they are suggesting a scare they neglect to mention that the result of the bond vigilantes could be depreciation of the dollar and increased demand for US made products.  The balance depends on FOMC policy (including future expected FOMC policy).  Rogoff is, to put it very mildly, very familiar with the concept of an exchange rate and the importance of exchange rate regimes (I agree with Krugman that he is the world’s number 1 or so international macroeconomist).  A discussion of a sell-off of Treasuries with no mention of exchange rates is an attempt to hide the reason that their work is not relevant to the US policy debate.
 Carmen's work on financial repression suggests a different scenario. But many emerging markets have stepped into bubble-like territory and we have seen this movie before.  We should not take for granted their prosperity that makes possible their continuing large-scale purchases of US debt.  Reversals are possible.  Sensible risk management means planning for these and other contingencies that might disturb today's low global interest rate environment.
The US Treasury also sells long term debt instruments.  That is a way to plan for those contingencies. There is no reason it has to rely on future decisions by investors.  It can lock in low interest rates now.  Surely R&R know about Treasury bonds. 
Second, on debt and growth.  The Herndon, Ash and Pollin paper, using a different methodology, reinforces our core result that high levels of debt are associated with lower growth.  This fact has been hidden in the tabloid media and blogosphere discourse, but this point is made plain by even a cursory look at the full set of results reported in the very paper they critique. 

Still at U Mass Amherst Dube shows that the timing is low growth then a high debt to GDP ratio except for extremely high growth at extremely low ratios (basically Axis countries whose debt was forgiven growing quickly soon after WWII as noted by Krugman).  I have noted that for the post wwII OECD dataset there is no evidence at all that high debt to GDP granger causes low growth.  Herndon, Ash and Pollin (2013) is an important paper, but it was not the last contribution to the literature.

More importantly, the result was prominently featured in our 2012 Journal of Economic Perspectives paper with Vincent Reinhart on Debt Overhangs, which they do not cite. The main point of our 2012 paper is that while the difference in annual GDP growth between high and lower debt cases is about one percent a year, debt overhang episodes last on average 23 years. Thus, the cumulative effect on income levels over time is significant.

Notice how the data are divided into categories “high and lower debt” this creates a step by assumption. This is the key point of debate with Krugman (see his reply) http://krugman.blogs.nytimes.com/2013/05/26/reinhart-and-rogoff-are-not-happy/   .  The calculation is not useful to policy makers. Also growth is not just brief recessions and the effect of debt (note a literature too huge to cite) so this does not address the issue of causation (as frequently suggested by CR&R and I think VR and no I won’t find the link)).

Third, the debate of the last few weeks does not change the fact that debt levels above 90% (even if one entirely rejects this marker for gross central government debt as a common cross-country "threshold") are very rare altogether and even rarer in peacetime.  From 1955 until right before the recent crisis, advanced economies spent less than 10% of those years at a debt/GDP ratio of higher than 90%; only about two percent of the years are above 120% debt/GDP. If governments thought high debt was a riskless proposition, why did they avoid it so consistently?

This is policy analysis made under the assumption that policy makers know what they are doing. If one is willing to make that assumption, there is no reason to bother with economic research. 

Debt and Growth Causality
Your recent April 29, 2013 NY Times blog The Italian Miracle is meant to highlight how in high-debt Italy, interest rates have come down since the European Central Bank's well-placed efforts to act more as a lender of last resort to periphery countries.  No disagreement there. However, this positive development is meant to re-enforce your strongly held view that high debt is not a problem (even for Italy) and that causality runs exclusively from slow growth to debt.  You do not mention that in this miracle economy, GDP fell by more than 2 percent in 2012 and is expected to fall by a similar amount this year. Elsewhere you have stated that you are sure that Italy's long-term secular growth/debt problems, which date back to the 1990s, are purely a case of slow growth causing high debt.  This claim is highly debatable.
Indeed, your repeatedly-expressed view that slow growth causes high debt but not visa-versa, is hardly supported by the recent literature on the subject. 



The sentence immediately above is totally false.  R&R either know it is false or have neglected to read anything Krugman wrote on the question.  For example he wrote
“I’m also puzzled by the way R-R deal with the reverse causation argument: they admit it can happen, but argue that causation doesn’t always run from growth to debt, but can run the other way. Isn’t that attacking a straw man?”

The claim that the assertion which they definitely claim he made is a “straw man” is the strongest possible proof that their claim is false. 

Of course, as we have already noted, this work has been singularly ignored in the public discourse of the past few weeks.  The best and worst that can be said is that the results are mixed.  A number of studies looking at more comprehensive growth models have found significant effects of debt on growth. We made this point in the appendix to our New York Times piece.  Of course, it is well known that the economic cycle impacts government finances and therefore debt (causation from growth to debt).  Cyclically adjusted budgets have been around for decades, your shallow characterization of the growth-debt connection. 

This is not a complete sentence, but the argument is that difference in growth are nothing but the effects of debt plus brief recessions.  This is nonsense. I am beginning to think that maybe R&R really don’t understand this issue.  It’s as if they have overlooked the entire much too large growth regression literature.  Also it is easy to test if the lower growth due to debt/GDP over 90% (compared to =90%) is due to short term correlation between the previous year’s growth and this yea’ss ratio. For the post WWII OECD sample it is. This is one simple regression.  R&R have no excuse for letting me be the first to report it.

As for ways debt might affect growth, there is debt with drama and debt without drama.
Debt with drama.  Do you really think that a country that is suddenly unable to borrow from international capital markets because its public and/or private debts that are a contingent public liability are deemed unsustainable will not suffer lower growth and higher unemployment as a consequence?

If the result is a sharp depreciation of the currency sure. Rogoff is pretending he doesn’t know what exchange rates are. I was in Italy when there was a sudden collapse of faith in Lira denominated Italian debt (1992).  It was the best thing that happened to the Italian economy in decades.


With governments and banks shut out from international capital markets, credit to firms and households in periphery Europe remains paralyzed. This credit crunch has a crippling effect on growth and employment with or without austerity.  Fiscal austerity reinforces the procyclicality of the external and domestic credit crunch.  This pattern is not unique to this episode.

It is only found when countries borrow in currencies which they can’t print at will and countries which are determined to stay on the gold standard.  R&R pretend that Eurozone problems have nothing to do with the Euro. 

Policy response to debt with drama.  On the policy response to this sad state of affairs, we stress that restoring the credit channel is essential for sustained growth, and this is why there is a need to write off senior bank debt in many countries. Furthermore, there is no reason why the ECB should buy only sovereign debt-purchases of senior bank debt along the lines of the US Federal Reserve's purchases of mortgage-backed securities would be instrumental in rekindling credit and working capital for firms.  We don't see your attraction to fiscal largesse as a substitute. Periphery Europe cannot afford it and for Germany, which can afford it, fiscal expansion would be procyclical.  Any overheating in Germany would exert pressure on the ECB to maintain a tighter monetary policy, backtracking some of the progress made by Mario Draghi.

Here the concern is that Draghi et al will be more austere than Merkel et al.  It is an absurd preoccupation.  German y overheating means inflation in Germany.  Europe needs inflation in the core or deflation in the periphery.  IIRC deflation is always associated with severe recession whether or not there is a debt problem. 

A better use of Germany's balance sheet strength would be to agree on faster and bigger haircuts for the periphery, and to support significantly more expansionary monetary policy by the ECB.

This might or might not be better, but it is much less fun for Germans.  It obviously won’t happen (nor will German fiscal stimulus). It is a fantasy magic pony plan.  To be interested in whether debt forgiveness would be better than tax cuts one has to ignore political realities entirely.  In any case, there is no reason with Krugman Reinhart and Rogoff can’t argue for both debt forgiveness and German tax cuts.  It isn’t as if Germany’s balance sheet strength will be exhausted if they advocate a lot of both. Also my pet obsession. The ECB is now close to the ZLB.  I think it is clear that there isn’t much more they can do (the US experience shows how totally ineffective nonstandard monetary policy is in the absence of support from the fiscal authority).

Debt without drama.  There are other cases, like the US today or Japan since the mid-1990s, where there is debt without drama.  The plain fact that we know less about these episodes is a point we already made in our New York Times piece.  We pointedly do not include the historical episodes of 19th century UK and Netherlands among these puzzling cases. Those imperial debts were importantly financed by massive resource transfers from the colonies. They had "good" high-debt centuries because their colonies did not.  We offer a number of ideas in our 2012 paper for why debt overhang might matter even when there is no imminent collapse of borrowing capacity. 
Bad shocks do happen. What is the foundation for your certainty that as peacetime debt hits new records in coming years, the United States will be able to engage in  forceful countercyclical fiscal policy if hit by a large unexpected shock?  Furthermore, do you really want to find out the answer to that question the hard way?
The US Federal Government has an unlimited supply of dollars.  The idea that it might run out of dollars is obvious nonsense. It is like worrying about a hot air shortage in the US Senate.  This is plainly obvious.  There is no way that R&R can fail to understand that the risk is of a future policy error (resulting in a severe recession or high inflation) not of a binding limit to the US Federal Government’s access to dollars. Also it takes a lack of imagination or empathy to worry more about a possible future problem with the unemployment rate is 7.5% the deep poverty rate is at record levels (party because records don’t go back very far) and long term unemployment is far above the previous post WWII high. I stress that think that in normal times the US Federal Government should run a surplus, and indeed build a sovereign wealth fund. But I believe this because I believe that people aren’t Ricardian so debt creates the illusion of wealth and sub optimal capital accumulation and not because I think the USA might run out of dollars.
The United Kingdom, which does not issue a reserve currency, is more dependent on its financial sector and suffered a bigger banking bust, has not had the same shale gas revolution, and is more vulnerable to Europe, is clearly more exposed to the drama scenario than the US.  And yet you regularly assert that the situations in the US and UK are the same and that both countries have the costless option of engaging in an open-ended fiscal expansion.  Of course, this does not preclude high-return infrastructure investments, making use of the public balance sheet directly or indirectly through public-private partnerships.
Policy response to debt without drama.  Let us be clear, we have addressed the role of somewhat higher inflation and financial repression in debt reduction in our research and in numerous pieces of commentary.  As our appendix shows, we did not advocate austerity in the immediate wake of the crisis when recovery was frail.  But the subprime crisis began in the summer of 2007, now six years ago.  Waiting 10 to 15 more years to deal with a festering problem is an invitation for decay, if not necessarily an outright debt crisis.  The end may not come with a bang but with a whimper.

It may.  Also R&R may write an open letter tomorrow saying Krugman was right about everything and apologizing for this letter.  But it’s not the way to bet.  An argument that a policy proposal may cause problems without any evidence that it would or any explanation of how it could is not worthy of Reinhart and Rogoff. They are appealing to prejudice, because they have no evidence or argument.

Scholarship: Stick to the facts
The accusation in the New York Review of Books is a sloppy neglect on your part to check the facts before charging us with a serious academic ethical infraction.  You had already implicitly endorsed this from your perch at the New York Times by posting a link to a program that treated the misstatement as fact.
Fortunately, the "Wayback Machine" crawls the Internet and periodically makes wholesale copies of web pages. The debt/GDP database was first archived in October 2010 from Carmen's University of Maryland webpage.  The data migrated to ReinhartandRogoff.com in March 2011.  There it sits with our other data, on inflation, crises dates, and exchange rates.  These data are regularly sought and found for those doing research who care to look. The greater disclosure of debt data from official institutions is testament to this.  The IMF began to construct historical public debt data only after we had provided a roadmap in the list of our detailed references in a 2009 book (and before that in a 2008 working paper) that explained how we had unearthed the data. 
Our interaction with scholars and practitioners working on real world questions in our field is ongoing, and our doors remain open. So to accuse us of not sharing our data is an unfounded attack on our academic and personal integrity. 

OK this is an important issue.  Do R&R use sitemeter ? 
Update -- R&R did *not* make their data set available.  Some of their data (debt/gdp) is publicly  available.  The data on real GDP growth are not (except via Herndon and even then there are multiple real GDP time series and I don't know which they used).  
see Herndon on the question.  
http://www.businessinsider.com/thomas-herndon-on-reinhart-and-rogoffs-data-availability-2013-5
He is very polite, but notes that they did not make their spreadsheet available.  I'm pretty sure that it is the only way to find out that they weighted by debt episode and not by country-year.  I am pretty sure that it was impossible to reproduce their 2010 AEA presentation results using publicly available data. I call this one too for Krugman, but it sure isn't as clear as the other issues.
Recap
Finally, we attach, as do many other mainstream economists, a somewhat higher weight on risks than you do, as debts of all measure -- including old age liabilities, public debt, private debt and external debt -- ascend into record territory.   This is not a conclusion based on one or two papers as you sometimes seem to imply, but rather on a long-standing body of economic research and extensive historical experience about the risks of record high debt levels. 
You often cite John Maynard Keynes.  We read Keynes, all the way through.  He wrote How to Pay for the War in 1940 precisely because he was not blasé about large deficits - even in support of a cause as noble as a war of survival. Debt is a slow-moving variable that cannot - and in general should not - be brought down too quickly.  But interest rates can change much more quickly than fiscal policy and debt. 

The interest yield of a 30 year bond as traded on the secondary market is not a problem for the Treasury.  Concerns about future interest rates can be addressed by issuing long term debt.  Are Reinhart and Rogoff familiar with the meaning of the phrase “Treasury Bond” ? They are pretending that they have no idea what it means.

You might be right, and this time might be, after all, different.  If so, we will admit that we were wrong.  Whatever the outcome, we intend to be there to put the results in proper context for the community of scholars, policymakers, and civil society. 

Respectfully yours,