The forces of evil easily triumph over Krugman and DeLong

Marcus Nunes already did a post on this, but now that the GDP numbers for 2013 are in it’s official, Greg Mankiw won by a landslide.  Here’s Mankiw back in 2009:

Paul Krugman suggests that my skepticism about the administration’s growth forecast over the next few years is somehow “evil.” Well, Paul, if you are so confident in this forecast, would you like to place a wager on it and take advantage of my wickedness?

Team Obama says that real GDP in 2013 will be 15.6 percent above real GDP in 2008. (That number comes from compounding their predicted growth rates for these five years.) So, Paul, are you willing to wager that the economy will meet or exceed this benchmark? I am not much of a gambler, but that is a bet I would be happy to take the other side of (even as I hope to lose, for the sake of the economy).

Krugman wisely decided to avoid this bet, which suggests he’s smarter than he appears when he is at his most political. In any case, the actual 5 year RGDP growth just came in at slightly under 6.3%. That’s not even close. Mankiw won by a landslide.

A few quick observations.  This data point does nothing to support the market monetarist model. Indeed some would argue that it contradicts any demand-side models. Ed Prescott might consider it a “proven scientific fact” that refutes all demand-side nonsense.  Of course Mankiw shares our belief that demand shocks matter, and I’m convinced by Bennett McCallum’s argument that the unit root problem mostly reflects the fact that GDP is hit by both supply and demand shocks, and that supply (or productivity) shocks tend to be very persistent.  The fall in the unemployment rate during a period of slow growth has helped to convince me that there’s something to Tyler Cowen’s Great Stagnation argument.

FWIW, I have argued that Keynesians underrate the importance of supply-side problems such as European-style labor market distortions.  I’ve argued that the British “output gap” is actually about 3/4th supply side.  So while I would have preferred a faster recovery from what I view as a strong demand shock, it’s at least a bit less embarrassing for us conservative demand-siders than for the Keynesians who focus obsessively on demand shocks.  I can’t speak for Greg Mankiw, but I wouldn’t be surprised if he had similar views.

For a guy who is right about everything Krugman sure seems to have been wrong about an awful lot of recent events.  Over at Econlog I’m about to do an important post documenting Krugman’s consistent attempts to shift position in order to avoid seeming to have been wrong about fiscal stimulus.  Of course on monetary policy nothing can prove him wrong.  If it works, great, he’s a fan. If not, well he was skeptical all along.

[Here’s the Econlog post.]

BTW, Mankiw was replying to this Krugman post.

Roots of evil (wonkish)

As Brad DeLong says, sigh. Greg Mankiw challenges the administration’s prediction of relatively fast growth a few years from now on the basis that real GDP may have a unit root — that is, there’s no tendency for bad years to be offset by good years later.

I always thought the unit root thing involved a bit of deliberate obtuseness — it involved pretending that you didn’t know the difference between, say, low GDP growth due to a productivity slowdown like the one that happened from 1973 to 1995, on one side, and low GDP growth due to a severe recession. For one thing is very clear: variables that measure the use of resources, like unemployment or capacity utilization, do NOT have unit roots: when unemployment is high, it tends to fall. And together with Okun’s law, this says that yes, it is right to expect high growth in future if the economy is depressed now.

But to invoke the unit root thing to disparage growth forecasts now involves more than a bit of deliberate obtuseness.

One point where I do agree with Krugman is the labor market, which does show clear trend reversion. That’s one reason I remain a demand-sider despite the unit root problem. Unemployment doesn’t rise and then stick, as in many European countries.  For those who don’t know, a unit root in a time series like RGDP implies that even after a change in RGDP, the optimal forecast of future RGDP is roughly the trend rate of growth.  No “bounceback” can be expected. Mankiw was pointing out that since RGDP fell sharply in 2009, if the optimal forecast of future growth after 2009 was near trend then the optimal forecast of future growth after 2008 was well below trend.

PS.  If Krugman regards Mankiw as evil, what possible adjective could he use to describe Ed Prescott?

PPS.  My comment about Krugman avoiding the bet was a joke. I actually don’t think academics are required to bet their beliefs.  Public humiliation is more effective in any case.

Update. I forget to mention that Krugman is covered either way.  If the economy bounces back it proves it’s a demand problem.  If not, well didn’t Keynes mention secular stagnation?


Tags:

 
 
 

72 Responses to “The forces of evil easily triumph over Krugman and DeLong”

  1. Gravatar of Morgan Warstler Morgan Warstler
    30. January 2014 at 07:17

    This “Real” GDP # supports my thesis that the shift from atomic to digital is do violent = every new productivity gain is negative GDP.

    The computer industry is staled. Tablet sales stalled, smart phones stalled.

    BUT, the old machines get WILDLY better everyday… because the free software in this here web gets better.

    I’ve been learning a new coding framework a friend is working on, and I haven’t gotten smarter, the tools to teach me have gotten smarter.

    I don’t code for a living, but my skillset to watch the code get done has improved, I can literally watch it get written as the keys are stroked, I can monitor in real time, the communication between my vision and what gets produced goes down, the number of coders you need to build X, falls…

    The speed of spread of X increases, the cost of doing X falls.

    And none of this requires anyone to go buy another computer…

    It does require the malls to be converted into data centers, but again, these new data centers are being built out of cell phone chips in open source designs everything shared, no resource lies dormant.

    If we share driverless cars and need only 25% as many and fire all the teachers, and, and, and, and…

    There is something very, very broken with the REAL #.

  2. Gravatar of PJ PJ
    30. January 2014 at 07:19

    I still think bets are important because they make a ceremony out of the public humiliation.

  3. Gravatar of Wonks Anonymous Wonks Anonymous
    30. January 2014 at 07:28

    If you actually agree to a bet, it’s harder to weasel out later.

  4. Gravatar of John Becker John Becker
    30. January 2014 at 08:06

    If you make a bet publicly and then lose, that’s a much more effective form of humiliation. This especially goes for a slippery snake like Krugman who always tries to leave himself an escape hatch and is pathologically unable to admit that he is ever wrong. You have to pin that guy to a bet to make him admit he is wrong.

    Morgan,

    You’re very right that there’s something very wrong with real GDP. Suppose you marry your housekeeper and she becomes a stay at home mom, real GDP goes down because you aren’t paying her anymore. There is no possible way to measure national wealth. The GDP numbers are as scientific as Gross National Happiness.

    Scott,

    Am I a demand sider if I think that demand shocks only matter in the short run; short run being defined as 1-2 years? After that, if a brisk recovery in the labor market doesn’t happen, there is some supply side issue in the way. Usually the government put it there right after the demand shock.

  5. Gravatar of RSF RSF
    30. January 2014 at 08:21

    It’s not a macroeconomic issue, but Mankiw recently posted on his blog about Krugman’s claims about income inequality and mortality not being supported by the research he was citing. Krugman seems to be quite adept at making explicitly false or disprovable claims, but I think Mankiw seems to have him in this case.

  6. Gravatar of Colin Colin
    30. January 2014 at 08:29

    Let’s also take a moment to laugh at Krugman for some incredible sloppiness in his Monday column which Mankiw highlighted yesterday:

    http://gregmankiw.blogspot.com/2014/01/does-income-inequality-increase.html

  7. Gravatar of Danny Danny
    30. January 2014 at 08:47

    Scott,

    I think your numbers are off, though not far enough to change the point. If I’m reading the BEA data correctly (http://www.bea.gov/national/xls/gdplev.xls), then from 2008Q4 to 2013Q4 RGDP increased by 9.5%. However, since RGDP was contracting for most of 2008 the 6-year change from 2007Q4 to 2013Q4 is even worse, coming in at only 6.5%.

  8. Gravatar of Morgan Warstler Morgan Warstler
    30. January 2014 at 08:52

    John I think RGDP used to work bc the overall effect was more atomic units sold, even as price drops occurred, and even then the govt. could juke the stats a bit

    Now it’s literally, 11 guys give Instagram away to 1B people and Kodak dies and ONLY thing you ever print on paper is something you hang on a wall.

    I can’t see office buildings mattering in 20 years. Certainly, the only government buildings we need are jails, and we shouldn’t need most of those.

    And all the new investment that can generate a return is focus on MOAR INSTAGRAM and less everything else.

    If there are diminishing returns in investments, if there’s too much capital chasing too few opportunities, but the whole time human consumption explodes, it seems like we’ll want to get a nice tight 1% NGDPLT and watch “REAL” GDP plummet while everyone lives better lives.

  9. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    30. January 2014 at 09:05

    He who laughs last; Snidely Mankiw.

  10. Gravatar of Donal Pretari Donal Pretari
    30. January 2014 at 09:31

    Since Sept. of 08, we’ve done some things right and we’ve done some things wrong, but, especially lately, I feel that seeing things this way is not common. A large amount of writing about the Economy is taken up with people talking about themselves, or, often in tandem, what they think about other people. There seems to be a lot of Self-Worth invested in Theories. Not being a big fan of Theories, at least in Political Economy, I’m struggling to find this writing interesting. We’re all guilty of this to some degree, so I’m not making this personal, but I do hope some people agree with me about this trend and will chime in.

    This paragraph from Hyman Minsky sums up a lot of what I feel is wrong now:

    “Between 1937 and 1942, the University of Chicago was a fine place to begin to be an economist. The economists at the University covered a wide spectrum of thought; there was no dominant Chicago School. The emphasis upon intellectual rigor and seriousness was combined with a wide definition of the subject. Only Lange (and perhaps Douglas) of the senior faculty was sympathetic to Keynes, but perhaps this was due to the prior acceptance by the other members of the faculty of the need for a strong expansionary fiscal policy during the depression. Having reached this “Keynesian” policy conclusion by observing the economy, orthodox economists at Chicago felt no strong need to revolutionize economic theory.”

  11. Gravatar of Mike Sax Mike Sax
    30. January 2014 at 09:33

    The ultimate covered either way school is Market Monetaristm. You are holding up low GDP as somehow proving Krugman wrong. it could just as easily be read the other way-that we need more demand and fiscal stimulus-or simply less austerity would have gotten us there.

    The fact that we have lower GDP hardly means the ‘neoliberal formula’-fiscal austerity with easy money-has been a raving success.

    Mankiw shouldn’t really be bragging after his embarrassing embracing of Romney’s budget plan of 2012-where cutting taxes for the rich was going to balance the budget.

    He should take the egg off his face for that one first. Him and Glenn Hubbard were more nakedly political than anything you accuse Krugman of.

  12. Gravatar of Mike Sax Mike Sax
    30. January 2014 at 09:38

    “Mankiw shouldn’t really be bragging after his embarrassing embracing of Romney’s budget plan of 2012-where cutting taxes for the rich was going to balance the budget.”

    I mean to say he shouldn’t be bragging about anything after his embarrassing embrace of that budget plan

  13. Gravatar of benjamin cole benjamin cole
    30. January 2014 at 09:39

    Still…I get the sense if a Krugman or Summers got in at the Fed they would blow the doors open and print more money…which is what we need…

  14. Gravatar of Vivian Darkbloom Vivian Darkbloom
    30. January 2014 at 09:42

    The Mankiw post on Krugman’s misrepresentation of the inequality study was pretty damning. One wonders whether the NYT blogs are subject to the same rules with regard to “corrections” as other parts of the paper. Apparently, Krugman is exempt from the rules. But, this blunder occurred not in the blog but in his regular Op-Ed piece. I do believe it is the policy of the NYT to require factual accuracy in Op-Eds even though otherwise “objectivity” is not required.

    I fully believe that if Krugman is pressed on this by the Times (he won’t be) that his incredible defense will be the last part of the sentence “and mortality” was not part of the hyperlink.

    This reminds me of the last NYT Public Editor, Daniel Okrent, who was the last (and only) one with the moxie to stand up to PK. PK did, reluctantly issue a “correction” with regard to one blatant error regarding tax rates under Reagan, but somehow forgot to note that it was a “correction”.

    http://publiceditor.blogs.nytimes.com/2005/05/31/new-public-editor-hosts-paul-krugman-daniel-okrent-debate/

    Margaret O’Sullivan may want to take a look at that recent post and remind PK of the correction policy, if there is one.

  15. Gravatar of JSeydl JSeydl
    30. January 2014 at 09:43

    “European-style labor market distortions”

    Is lower inequality a distortion? Is giving workers the freedom to collectively bargain a distrotion? Is preserving the communal aspect of production (and thereby of the self) a distortion?

    Maybe try not to use such loaded language when debatable and complicated rights issues are at stake. Maybe say “European-style labor market differences.” Otherwise, you sound like someone who has never read Rawls, Sandel, Sen, Nussbaum, etc etc. – which would be a very problematic lack of understanding for any economist.

  16. Gravatar of Doug M Doug M
    30. January 2014 at 09:48

    Morgan,

    You bring up a few things that I haven’t really thought about since web 1.0.

    Can service productivity be measured? If your doctor starts giving his patients 10 minutes of his time on average instead of 20, he can bill twice as many patients, but has productivity really improved?

    Software has massive fixed costs and zero marginal costs. How does this knock every micro-model on its head? And, what are the macro implications? It creates winner take all scenarios. Large investments are made, some with huge returns, and many which fail without ever raising a dollar of revenue.

    You seem to be into the idea of “creative destruction.” New technologies will disrupt existing businesses. When the economy is in a dynamic phase there will be a high level of churn between growing new business and dinosaurs. But we may hit patches where GDP declines in the short term as the dying dinosaurs outnumber the fast growing start-ups.

  17. Gravatar of Doug M Doug M
    30. January 2014 at 09:56

    On bets, it is a bit of the modern equivalent of the duel with much less dire consequences. When Mankiw challenges Krugman to a bet, he is effectively making the statement that Krugman is a charlatan. If Krugman refuses to “take up the gauntlet.” Mankiw supporters say “Krugman has no honor.” And Krugman supporters say that Mankiw is an uncouth bastard and not worth the effort.

    If Krugman accepts the challenge, then there is some expectation that the loser should grovel and explain the nature of their misguided ways, and how they won’t commit such an error in the future.

  18. Gravatar of John Becker John Becker
    30. January 2014 at 10:13

    Morgan,

    It’s definitely true that technology brings increases in satisfaction that you cannot measure in GDP. One of the problems economists always face is that you can’t measure consumer satisfaction in anything besides a subjective way. You can’t attach a number to how something feels.

    For instance, youtube and torrent downloading have made my life a lot better and I’ve never spent a dollar on any of it. If I had to pay for the stuff, I would have spent many thousands of dollars but since I don’t have to pay, I feel much better off. If I had paid for it, GDP would be higher but my satisfaction from the economy would have been lower.

  19. Gravatar of Reader223 Reader223
    30. January 2014 at 10:16

    Professor, you’re argument of monetary policy is too strong to stoop low enough to use the “Krugman’s beliefs are unfalsifiable” argument. The same exact thing can be said about market monetarism- oh, NGDP growth was below 5%? Money’s too tight! Dropping interest rates to the ZLB and unconventional monetary policy such as QE hasn’t brought us to potential? We didn’t do enough!

    Now, I don’t agree with any of those counter-points to MM, but clearly any crowd can be made to seem unfalsifiable if you resort to those measly tactics.

  20. Gravatar of RSF RSF
    30. January 2014 at 10:18

    Vivian,

    Unfortunately, Daniel Okrent was five Public Editors ago and left all the way back in 2005. I think the Public Editors since him have increasingly had either less of a spineor aren’t even cognizant of the Times’s biases. Okrent was a liberal, but also aware that the NYT can often be totally ridiculous. I don’t think the Times has become more responsive to its critics since then. In fact, it seems that the Public Editor just serves to deflect attention or protect the existing editorial establishment.

  21. Gravatar of Charlie Charlie
    30. January 2014 at 10:20

    There are two ideas being discussed:

    1. GDP always follows a unit root.
    2. The administration’s growth forecast is correct.

    Krugman rejected the first, but never accepted the second. Mankiw is very smart and tries to make PK’s rejection of the first mean he must support the second. He offers him a bet for 2. Most people (including you?) didn’t notice the switch.

    You’ve argued PK is sneaky and often leaves the wrong impression to his readers, but Mankiw plays that game very well also.

  22. Gravatar of Charlie Charlie
    30. January 2014 at 10:37

    Also, Mankiw’s original forecast was wrong:

    “Right now, we are facing a particularly high-variance economy. (Just look at the VIX index.) That means, under the conjecture I just described, that when recovery comes, it will probably be a robust one. But this logic is not necessarily a reason to raise the unconditional expectation of economic growth, because we don’t know when that recovery will begin.”

    Yet, we did get a recovery, but it has not been particularly strong and robust. Rather, it’s been fairly weak and fragile.

  23. Gravatar of TravisV TravisV
    30. January 2014 at 10:41

    Benjamin Cole,

    What on Earth makes you confident that Larry Summers would aggressively push for monetary stimulus? Have you read the things he’s been saying lately?

  24. Gravatar of Vivian Darkbloom Vivian Darkbloom
    30. January 2014 at 10:50

    “Yet, we did get a recovery, but it has not been particularly strong and robust. Rather, it’s been fairly weak and fragile.”

    I wish economists would stop using such precise language. It prevents everyone from later claiming whatever they want to claim.

  25. Gravatar of Charlie Charlie
    30. January 2014 at 11:13

    The Campbell and Mankiw paper is very highly cited, but every cite I open seems to conclude that they were wrong.

    This JPE paper by John Cochrane points out the problems with the Campbell and Mankiw approach:

    “In reconciling these results with previous research, I argued that conventional criteria for time-series model identification and estimation can produce misleading estimates of the random walk component of a series like GNP. The random walk component is a property of all autocorrelations taken together, but conventional procedures concentrate on the first few autocorrelations in order to parsimoniously capture short-run dynamics. When used to estimate the size of a random walk component, they impose identifying restrictions across the frequency range to infer the long-run properties of a series from its short-run dynamics. I argued that, in the absence of credible identifying restrictions, it is best to leave the short run out altogether, as the variance of k-differences or some other spectral window estimator does.”

    “I conclude that if there is a random walk component in GNP at all, it is small.”

  26. Gravatar of ssumner ssumner
    30. January 2014 at 12:32

    John Becker, That makes you a very conservative demand-sider.

    Danny, No, the proposed bet was using annual data, not Q4 to Q4.

    Donal, Good quote. But the Chicago School was still excellent when I was there in the 1970s.

    JSeydl. You asked:

    “Is lower inequality a distortion? Is giving workers the freedom to collectively bargain a distortion? Is preserving the communal aspect of production (and thereby of the self) a distortion?”

    No, and I never said they were. I was thinking of Denmark’s labor market as a sort of flexible ideal. Perhaps you’ve never studied Danish “flexicurity” labor policies, at least your comment makes me think you are ignorant of the subject.

    Reader223, You must be new here, as I am not making unfalsifiable claims.

    Charlie, You missed the point. Mankiw was minding his own business when Krugman called him evil. The burden of proof is on Krugman to justify that claim, and he wasn’t able to do so. And I don’t see Mankiw claiming that GDP always follows a unit root, just that it is hard to forecast and that a forecast of trend growth will generally outperform a forecast assuming lots of trend reversion. Which may we be true even if demand shocks exist (something Krugman seemed to overlook. Mankiw pointed out that Krugman seemed unaware of his research on unit roots.)

  27. Gravatar of Tom M. Tom M.
    30. January 2014 at 12:50

    Paul Krugman will go down in history as the greatest economist ever. And I mean ever. While Austrians were shouting HYPERINFLATION!! and neoliberals like Mankiw were shouting CONFIDENCE! PK was writing a book called “End this Depression Now!”

    By the way, after multiple QE’s, 23,000 people applied for 600 jobs at Walmart. Yes, I said Walmart.

    https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCkQFjAA&url=http%3A%2F%2Fwww.businessinsider.com%2Fwal-mart-receives-23000-applications-2013-11&ei=8LrqUq65KcbmyQGnrYGICA&usg=AFQjCNGRQPxaSFx9kfDuSLVcfNadJfBIRw&sig2=ExZxV7zuuufY8qIV7LhnYg&bvm=bv.60444564,d.aWc

  28. Gravatar of JSeydl JSeydl
    30. January 2014 at 12:54

    Then maybe you should be more precise, and say “European-style labor market distortions (relative to, for example, the non-distoritons in Denmark).” The extra work would probably be worth the effort, considering that the average reader is likely interpret your sentence as comparing labor markets in places like Spain and France to that of the US.

    Also, for what it’s worth, it took me two seconds to find a post in which you previously praised labor-market flexibility in places other than Denmark:

    “In contrast, Australia has one of the smallest, and had no recession. And Hong Kong has a rigid dollar peg, which exposed it to huge demand shocks in 1997 and 2001, and yet it avoided “depressions” despite having the smallest government sector in the entire developed world. Maybe it was those flexible labor markets that the liberals insist will only make a depression worse. But they’d reply that it’s better to have inflexible labor markets, like Greece and Spain.”

    http://www.themoneyillusion.com/?p=21332

    According to my data, Australia’s unionization rate has fallen by 50% and the income share of the top 1% there has doubled since the 1980s. Again, one might see these developments as “distortions” in a framework where economic efficiency isn’t the only thing we care about.

  29. Gravatar of Charlie Charlie
    30. January 2014 at 13:19

    Scott,

    That’s such a load in so many ways. First, PK never called Mankiw evil, that was another leap of Mankiw’s to make PK look bad. He titled his post “Roots of Evil.” Obviously, a play on words between unit roots and the phrase “the root of all evil.” At no time does PK call Mankiw evil.

    Second, you this is just wrong: “I don’t see Mankiw claiming that GDP always follows a unit root, just that it is hard to forecast and that a forecast of trend growth”

    Follow your own links back to where Mankiw says, “The view that Campbell and I advocated is sometimes called the unit-root hypothesis (for technical reasons that I will not bother with here). It contrasts starkly with the trend-stationary hypothesis.”

    So, clearly Mankiw is arguing for the unit root hypothesis. That is what the linked paper argues for as well.

    PK thinks holding this view requires “deliberate obtuseness.” That’s as close to a dig as PK makes at Mankiw personally made in the post. Both he and Delong are pointing to the fact that Campbell and Mankiw is a univariate forecast, rather than a multivariate forecast.

    Of course, this has been studied formally,check out another Cochrane paper “Univariate vs. Multivariate Forecasts of GNP Growth and Stock Returns: Evidence and Implications for the Persistence of Shocks, Detrending Methods,” so your beef with Krugman and DeLong is that they were right, but they didn’t cite all the scholarly papers that defend their view?

    Having looked into this a little, I can’t believe that Mankiw would link to one of his papers that had been so thoroughly debunked. Every citation I look at changes their conclusion with just a slight change in methodology. Even if the unit-root hypothesis is right, it can’t be based on the work of that paper. The problems in their methodology have been highlighted by lots of later work.

    Here is the beginning of the abstract to the second Cochrane paper:

    “Lagged GNP growth rates are poor forecasts of future GNP growth rates in postwar US data, leading to the impression that GNP is nearly a random walk. However, other variables, and especially the lagged consumption/GNP ratio, do forecast long-horizon GNP growth, and show that GNP has temporary components. Labor income and stock prices (using the dividend/price ratio) display the same behavior…”

  30. Gravatar of Charlie Charlie
    30. January 2014 at 13:24

    If you publish an important paper, and then a mountain of other papers point out 1. methodology is flawed, 2. even if it wasn’t flawed, small changes drastically change the results, and 3. extensions of the methodology, for instance, multivariate revers the conclusions. Yet, then you cite the original paper as proof to a lay audience. How can you be called anything other than “deliberately obtuse”?

  31. Gravatar of Larry Larry
    30. January 2014 at 14:01

    How long will it be before goods start moving through Amazon without being touched by a single human hand (even driving a forklift)?

    How long before they’ll reach the front door never having been touched even in the factory or by a truck with a human driver?

    Not that long…

  32. Gravatar of Reader223 Reader223
    30. January 2014 at 15:23

    But Krugman isn’t either and yet you saying,

    “I forget to mention that Krugman is covered either way. If the economy bounces back it proves it’s a demand problem. If not, well didn’t Keynes mention secular stagnation?”

    makes it seem like he is. My point was that MM can be phrased in such a way that looks unfalsifiable as well. I know you don’t make unfalsifiable claims; *that was the whole point I was trying to make*. Don’t try to make Krugman look like he’s making unfalsifiable claims when he’s not- you can twist anyone’s views to seem so.

  33. Gravatar of Kevin Donoghue Kevin Donoghue
    30. January 2014 at 15:34

    “Mankiw was minding his own business when Krugman called him evil.”

    Never mind what Mankiw was doing. Krugman called him evil? Really?

  34. Gravatar of Jim Glass Jim Glass
    30. January 2014 at 16:21

    Vivian, about the Times’s op-ed correction policies…

    Krugman in his early days at the Times made the mega-howler claim that revenue lost to the Bush tax cuts “would have been more than enough to ‘top up’ Social Security and Medicare, allowing them to operate without benefit cuts for the next 75 years”!

    That is a statement that was wrong then by a mere deca-trillion dollars, discounted to present value.

    But of course he couldn’t stop with that — he had to pile on saying Medicare and Social Security weren’t a real long-term problem at all (obviously — because absent the Bush tax cuts they were financed for 75 years) … and all the endless talk that they were came just from lazy pundits who wanted to pretend to be fair and balanced, etc. etc. Read and enjoy:

    http://www.nytimes.com/2003/03/21/opinion/who-lost-the-us-budget.html

    Though while so accusing the pundits of being lazy and superficial he himself was ignoring year-after-year of CBO and GAO reports detailing exactly the opposite — including a recently published GAO 75-year fiscal projection that it ended in only 40 years because Medicare etc. by then created at 20% of GDP annual deficit that crushed the economy, “finis” — based on Clinton policy, with no Bush tax cuts.

    But that made no matter, anyone who disagreed with him was a naif or much worse, plainly.

    How could he pull such a whopper? He read the abstract of a CBPP paper that seemed to say that, but didn’t bother to read the text itself so he didn’t realize it was referring only to the cost of Medicare Part A, the lesser part of Medicare — and he was in a hurry off to his daily Bush bashing and this was a good and righteous new club to use! (As he was accusing other journalists of being lazy).

    Now, Krugman soon knew that he had made this multi-trillion dollar mistake — I pointed it out on DeLong’s blog (DeL was a different person back then, open to contrary views, believe it or not) where it became the subject of a good bit of discussion, and it was picked up at Econlog and other places.

    But, no, there was no correction ever published in the Times, either by it or him. Fuhgetaboutit.

    Moreover, on his personal web site he published a remarkable series of statements denying that he had made any mistake at all: there was no mistake, that’s how everyone talks about Medicare finances … and also the mistake was in the paper, he’d innocently relied on it … and too, everyone makes the same mistake all the time! It was like the Richard Pryor routine about his kid who broke the lamp and spouted six self-contradictory denials at once.

    Then PK ended with a remarkable rant, claiming that those who pointed out the mistake owed an apology *to him*:

    “anyone who accused me of dishonesty on this subject owes me an abject apology. Hey, we can always fantasize. Update: I see that I will nonetheless be berated for immense dishonesty because I didn’t say “part A”… Dishonesty, dishonesty everywhere.”

    Now, I personally found this very revealing about Krugman’s psychology. NOBODY accused him of being “dishonest”, after starting this discussion I monitored it everywhere to the end, and nobody did. People wanted a clarification, explanation, correction. But to PK, “mistake” is “dishonest”, and that tells the story. He won’t tolerate such dishonesty in others, and damn he’s sure not about to let others accuse him of being dishonest. And you know anyone who would even try such a thing is “evil”.

    So that’s what you can expect in the way of corrections re Times op-eds and from Krugman. Okrent understated things.

    The only person ever to get a real published apology from Krugman was Fraga, through his lawyers, after PK libeled him in Slate.

    BTW: Read Krugman’s damning of the US govt’s long-term fiscal position in that column above, and compare it to what he says about the much much worse long term fiscal position of the govt today (with the Bush tax cuts now permanent, of course). If he isn’t the best walking-around example of the practice of what the psychologists call “motivated reasoning” (with a good dose of “projection” supplementing it) tell me who is.

    ~~~
    @ Tom M.

    Paul Krugman will go down in history as the greatest economist ever. And I mean ever.

    That’s right Tom, your idol has feet of gold (apart from a deca-trillion here and there). Platinum! Just ask Fraga. :-)

  35. Gravatar of ChargerCarl ChargerCarl
    30. January 2014 at 17:14

    I don’t see where Krugman called Mankiw “evil.”

  36. Gravatar of philip meguire philip meguire
    30. January 2014 at 18:24

    Let me begin by saying that I too have been disappointed by the lackluster recovery from the Great Recession. But let me explain how that tepid recovery is consistent with there being a unit root in the relevant time series.

    The natural log of real per capita GDP in the USA, appears to be trend reverting, only if the sample includes the data for 1929-47. The reason is the huge decline in 1929-33, the dramatic rise 1933-44 (with a major hiccup in 1938), and the decline in 1945-47. These events yielded the highest variance data ever in the entire time series, which begins in 1869. When the dust settled on all this bipolar behaviour, the American economy found itself in 1947 where a trend forecast made in 1929 would have put it. The economic boom of WWII neatly cancelled out the Great Depression. This fact does not describe any other economy whose real GDP data go back to 1900 or earlier.

    Nelson and Murray (2000, JME) have convinced me that there is a unit root in the natural log of real per capita GDP, when the sample is the postwar quarterly data. This unit root means that the Great Recession will have long lasting adverse consequences for the American standard of living, which I understand to be Mankiw’s position.

  37. Gravatar of Full Employment Hawk Full Employment Hawk
    30. January 2014 at 23:41

    The failure of real GDP to return to the old trend does not require any unit root but can be explained by the fact that the worst demand shock since the Great Depression was not met with a stimulus anywhere near sufficient to restore the economy to this trend. One can argue about whether this was due to the fact that “The Fed did poorly over the past 5 1/2 years,” as market monetarists do, or because the initial fiscal stimulus was too small and then turned contractionary, as many Keynesians would, or due to a combination of the two (my view) and that if the economy had gotten sufficient demand stimulus it would have returned to the old growth path.

    Krugman in making his forecast in 2009 almost certainly did not anticipate that fiscal policy would turn contractionary in subsequent years.

  38. Gravatar of Full Employment Hawk Full Employment Hawk
    31. January 2014 at 00:12

    “If Krugman regards Mankiw as evil” Krugman did not call Mankiw evil. He called Mankiw’s skepticism about the administration’s growth forecast over the next few years is somehow “evil.” Calling a particular view of a person on one particular issue evil in no way amounts to calling the person evil.

    Market monetarism and Krugman style Keynesianism actually have a lot in common. Both agree that deficient aggregate demand is the main cause of the economy being depressed and that policies to increase it would restore the economy to full employment. That puts them on the same side vis a vis people like Barro or Prescott.
    The only disagreement is with respect to the diagnosis of what is the cause of the deficient aggregate demand and therefore the appropriate policy to deal with it. Therefore the hostility between the two camps, when they should be working together where they agree, is unfortunate.

  39. Gravatar of Steve R. Steve R.
    31. January 2014 at 02:07

    I think Charlie nailed pretty much everything that needs to be said on this topic. Nobody triumphed over anything, and honestly Krugman et al. come out looking much smarter…

    Basically, I’ve stopped reading this blog except when it’s linked to somewhere – and posts like this make me question even doing that. You’ve got one good idea – focus on NGDP/targeting – and not in the be all and end all fashion MMs like to think it is, but more in a “well, that’s a fairly interesting thing to add to the toolbox”. Blogging might have been fun, but it’s probably time to retire to Bentley and take up gardening or something.

  40. Gravatar of J Mann J Mann
    31. January 2014 at 05:32

    This is a good example of why Krugman is so frustrating. You would think he would write clearly, but I remember when the Mankiw discussion came up, and at that point I thought he was calling Mankiw evil and disputing his forecast.

    Over the intervening years, it turned out that maybe Krugman didn’t mean what a normal reader might conclude he meant, and didn’t clarify until much later.

    1) I can’t find a cite, but IIRC, “Roots of Evil” was a dry attempt at wordplay about Krugman not liking the unit root hypothesis. Krugman has a habit of inserting dry jokes that detract from clarity. Of course, Krugman’s general snarky tone makes it appear that he’s calling Mankiw evil, and maybe he was, particularly since when Mankiw thought Krugman meant it that way, Krugman didn’t clarify.

    2) When the economy didn’t snap back to trend, Krugman wrote a piece about how he was right all along, and that he had always agreed with Mankiw’s hypothesis that the economy was not likely to snap back to trend any time soon, but only wrote that piece to criticize Mankiw for mentioning the unit root hypothesis in his analysis.

    http://krugman.blogs.nytimes.com/2012/09/24/maybe-another-economist-with-the-same-name/?_php=true&_type=blogs&_r=0

    It’s maddening. Mankiw’s original piece was about whether the Obama growth estimates were reasonable. If Krugman had written “I agree with Mankiw that the growth estimates are probably too high, but disagree that the unit root has any relevance,” then we would have all known what he meant about an issue of some public policy importance. H***, if he’d even responded to Mankiw’s bet offer, we would have known. But instead, we’re apparently all idiots for not cross-referencing his Mankiw piece with all his OTHER blog posts to see what he meant.

    On top of that, the unit root business was a distraction from Mankiw’s piece. If you read it, Mankiw digresses about the unit root for a bit as a possibility, but then concludes that the economy is likely to snap back towards trend, just not as soon as the Obama team was forecasting.

    So Krugman could have written “I agree with Mankiw that the unit root hypothesis probably doesn’t apply, and that the Obama administration growth forecast is excessively optimistic” and saved us a lot of trouble.

  41. Gravatar of J Mann J Mann
    31. January 2014 at 05:36

    Charlie, Krugman seems to take the view that all other economists can be skimmed, but he has to be read with maximum charity.

    See my post above for a discussion of the unit root debate – I read the dispute when it happened, and it honestly never occurred to me that Krugman meant the roots of evil as a joke, or that he agreed with Mankiw’s conclusion, just not his methods.

    At a minimum, if Krugman had clarified when Mankiw thought Krugman was calling him evil, or when Mankiw offered to bet him on the trajectory of the economy, then we would have all known what he meant back in 09 or whenever this whole thing happened. Wouldn’t that have been better?

  42. Gravatar of J Mann J Mann
    31. January 2014 at 07:56

    I apologize, Charlie, for not reading all of your post – I had looked into this a couple weeks ago and was eager to bloviate. (I stand behind all my bloviation, though).

    With regard to whether Mankiw’s piece is dependent on his unit root discussion, I would say not. Mankiw’s conclusion – that we will probably see faster than usual growth when the economy begins to recover – is inconsistent with the unit root hypothesis strongly applying this case, isn’t it?

    For a while, I thought Krugman agreed with the unit root hypothesis, because he’s written that he doesn’t expect a “v shaped recovery,” but on reflection, I think Krugman must mean that he expects a plateau followed by faster than normal growth to catch up to trend. (A “u shaped” recovery?) because otherwise he would be arguing in support of the unit root hypothesis, and we know he doesn’t mean that, I think.

    http://krugman.blogs.nytimes.com/2008/02/10/postmodern-recessions/

    http://krugman.blogs.nytimes.com/2013/09/09/cest-la-v-no-more/

    Mankiw basically says there are two reasons to be sceptical about the 09 growth estimates – (1) maybe there is a permanent shift to the economy’s productive capacity (he doesn’t propose this, but maybe, for example, a number of people have permanently left the workforce and are not coming back or being replaced) or (2) maybe the economy will snap back, but if so, there’s no reason to assume it will do so soon.

    http://gregmankiw.blogspot.com/2009/03/team-obama-on-unit-root-hypothesis.html

    If Krugman thought that Mankiw was wrong on his first hypothesis but right on the second, it would have been more productive to say so, and responding to the bet would have clarified things for Krugman and Mankiw’s readers.

  43. Gravatar of Vivian Darkbloom Vivian Darkbloom
    31. January 2014 at 08:35

    “And I don’t see Mankiw claiming that GDP always follows a unit root, just that it is hard to forecast and that a forecast of trend growth will generally outperform a forecast assuming lots of trend reversion.”

    I’ve read the Mankiw thing a couple of times and that’s the same impression I have.

    Let me try to express this in plain English (As a layperson, I have no other language).

    Mankiw was questioning a fairly rosy administration growth forecast issued in February 2009. That forecast was rather optimistic in part because of the administration reliance on the idea that recessions are followed by robust recoveries such that average growth returns to trend. Important here is that the forecast was for a 10 year period that is customary for budget forecasting.

    Mankiw did not question the idea that there would be a robust recovery following a recession or that average growth would eventually return to trend, he questioned *the forecast* on the basis that while one might predict this will eventually happen, one cannot easily known *when*. Stated differently, when the recession would end affects when the robust growth will occur affects when the return to trend will occur and that affects a growth forecast for a future finite period of time (10 years).

    Update: This seems fairly consistent with J. Mann’s observation.

    In particular, Mankiw questioned the accuracy of the first four years of the forecast (through 2013) and on this basis the proposed bet was made.

    Of course, this position was misstated by Krugman from the start:

    “Greg Mankiw challenges the administration’s prediction of relatively fast growth a few years from now on the basis that real GDP may have a unit root — that is, there’s no tendency for bad years to be offset by good years later.”

    I certainly did not read Mankiw as saying “there’s no tendency for bad years to be offset by good years” but rather that we cannot easily know how quickly those “good years” are going to occur. Notice also the shift Krugman makes here from “prediction of relatively fast growth *a few years from now*” to “there is no tendency (implying *ever*) for bad years to be offset by good”. And, Mankiw’s proposed bet was not even with respect to “a few years from now”, it was from the period 2009 (the then present) to 2013.

    In proposing his bet, Mankiw pointedly did not go out the full 10 years. Consistent with his original point, he picked off the higher probability bet (for him) that ran only through 2013. Essentially, I read him as saying the administration was being statistically too optimistic in their assumptions as to when the recession would end, therefore when this “robust recovery” would kick in and return the economy to trend.

    Actually, this seems a lot like what Krugman closed his blog post on Mankiw with:

    “And yes, we can expect fast growth if and when that capacity comes back into use.”

    “If and when”? If you can’t tell in the short run when that “fast growth” will occur, how can you expect to make an accurate short term forecast that it will occur? That seems to have been precisely Mankiw’s point.

    And, ironically, I recall a Krugman post where he explains what Keynes meant when he wrote “in the long run, we’re all dead”:

    “Keynes’s point here is that economic models are incomplete, suspect, and not much use if they can’t explain what happens year to year, but can only tell you where things will supposedly end up after a lot of time has passed. It’s an appeal for better analysis, not for ignoring the future; and anyone who tries to make it into some kind of moral indictment of Keynesian thought has forfeited any right to be taken seriously.”

    http://krugman.blogs.nytimes.com/?s=%22keynes%2

  44. Gravatar of David Cushman David Cushman
    31. January 2014 at 08:58

    I agree with J Mann’s interpretations, including that “Krugman seems to take the view that all other economists can be skimmed, but he has to be read with maximum charity.” Krugman certainly didn’t seem to have read my paper before snarkily referring to it in his blog entry “maybe another economist with the same name” that J Mann referred to. I later discussed this in

    http://econjwatch.org/articles/paul-krugman-denies-having-concurred-with-an-administration-forecast-a-note

    By the way, contrary to the impression one might get from Charlie, there are lot of prominent papers since the Cochrane JPE paper Charlie refers to that conclude real GDP has a unit root. The Nelson-Murray 2000 paper in the JME mentioned above is but one of them.

  45. Gravatar of Charlie Charlie
    31. January 2014 at 10:16

    JMann and Vivian,

    If Krugman disagrees with a point Mankiw makes in a blog post, it doesn’t imply that Krugman disagrees with the entire blog post. He and Brad picked a fight over the unit-root hypothesis that Mankiw advances. There are lots of OTHER reasons to think that the administration’s growth forecast was too optimistic. Reinhart and Rogoff, for one, that Krugman was trumpeting (another example, PK agreed with RR’s book, but heavily criticized their follow on paper that turned out to have errors).

    Does Mankiw believe that GDP follows a unit root? Yes, it is impossible to come away with the view that he doesn’t:

    The Title of the Post – “Team Obama on the Unit Root Hypothesis” that’s the compare and contrast right there. Mankiw on one side, team Obama on the other.

    “In the language of time-series econometrics, the CEA is premising its forecast on the economy being trend stationary.”

    “The view that Campbell and I advocated is sometimes called the unit-root hypothesis (for technical reasons that I will not bother with here). It contrasts starkly with the trend-stationary hypothesis.”

    I think you are misconstruing the end of the post to be backing off his Unit Root hypothesis. The unit root hypothesis doesn’t mean GDP can never boom and bust, it just means those booms and busts aren’t predictable. Just as stocks can follow a random walk, while still having bull and bear markets. (Bull and Bear are really statements about the past.) That said, stocks are in fact predictable at longer horizons, and it is on that literature that the Cochrane papers I pointed to builds on.

    In Mankiw’s follow up post, he links to a second paper that Campbell and Mankiw wrote:

    “Using post-war quarterly data, it is hard to reject the view that real GNP is as persistent as a random walk with drift.”

    IN CONCLUSION, Mankiw does believe that GDP follows a unit root. BUT HE KNOWS he is on very sketchy ground. That is why he offers a bet about the administration’s growth forecasts. Obviously, he has lots of ways to win that bet, even if he is wrong about the unit root hypothesis. He is very smart, so he tried to move the goal post.

    I read this exchange in real time, and I think one reason so many are coming to the mistaken view is that they must have read Mankiw’s last response first. They have adopted Mankiw’s framing of the exchange, which I think is impossible hold with a fair reading from beginning to end. I can’t imagine anyone that read Mankiw’s first post, then Krugman’s second would say that PK had called Mankiw evil. It just wasn’t the right reading (and evil never modified Mankiw). That so many would think he did after the fact, just shows they adopted Mankiw’s framing.

  46. Gravatar of Vivian Darkbloom Vivian Darkbloom
    31. January 2014 at 10:47

    “He and Brad picked a fight over the unit-root hypothesis that Mankiw advances. There are lots of OTHER reasons to think that the administration’s growth forecast was too optimistic.”

    Charlie,

    Brad DeLong wrote:

    “”O]ur paper, of course, was only one piece of a large literature. The CEA might well be right…” (quoting Mankiw)

    And that is certainly the way to bet.”

    So, both sides were taking positions based on perceived probability. DeLong took the CEA side; Mankiw went the other way. Krugman derided Mankiw and linked to DeLong’s post with approval (sigh).

    Further, subsequent events showed that Krugman concurred with the CEA forecast (see Cushman above).

    Krugman takes great efforts to not get pinned down on anything. If he sees the chance, he’ll take credit. If he takes the wrong side of a bet, he’ll deny, often with very implausible arguments. This is a case in point. DeLong and Krugman lost the bet regarding that CEA forecast. It’s not the end of the world. Live with it.

  47. Gravatar of Charlie Charlie
    31. January 2014 at 12:46

    David,

    The Cochrane JPE paper shows that the methodology in Campbell and Mankiw was faulty. They would get those results if GDP reverted over longer horizons. He then uses the variance tests to show that such an interpretation is consistent with the evidence. That’s why I said, “Even if the unit-root hypothesis is right, it can’t be based on the work of that paper. The problems in their methodology have been highlighted by lots of later work.”

    How can pointing to that paper without drawing attention to the criticism of the methodology be considered anything other than “deliberately obtuse.”

  48. Gravatar of Charlie Charlie
    31. January 2014 at 12:58

    Vivian,

    It’s incredibly clear what Krugman is referring to, “Greg Mankiw challenges the administration’s prediction of relatively fast growth a few years from now on the basis that real GDP may have a unit root.…I always thought the unit root thing involved a bit of deliberate obtuseness”

    Reading the DeLong post is harder than reading the PK, and I think it’s much easier to read DeLong as supporting the CEA forecast. Perhaps, Mankiw should have offered DeLong the bet. That would make some sense. But even with DeLong when he says, “The CEA might well be right…and that is certainly the way to bet.” He’s cut off the “robust recovery over the next few years.” So he may well just think the CEA is right to not assume a unit root, but not necessarily agree that we’ll have robust growth over the next few years.

    That said, I don’t know. To me PK is very clear about highlighting the disagreement, DeLong is not. If Scott’s post had been criticizing DeLong, I wouldn’t have a problem with it. It’d be a fair reading.

  49. Gravatar of J Mann J Mann
    31. January 2014 at 14:19

    Charlie,

    Thanks – I read it in real time too, and frankly until I read David Cushman’s pieces a few weeks ago, I honestly believed that Krugman agreed with DeLong that the 09 growth estimates were “certainly the way to bet.”

    In hindsight, I agree with you that someone who read Krugman’s 08 piece about the end of v-shape recoveries would probably guess that Krugman agrees with Mankiw’s second hypothesis – that the Obama admin was overconfident because there was no reason to believe the economy would snap back now.

    My point as a blog reader is that Krugman doesn’t do much to promote clarity on issues of public interest. He’s not a hostile witness in a deposition — he’s a public intellectual who writes for the NYT in an effort to promote economic literacy.

    Krugman piled on to a discussion between DeLong and Mankiw in which DeLong said that the Obama growth estimates were the way to bet and Mankiw said they weren’t. If Krugman had said “I agree with Brad on the unit root hypothesis, but Greg on his second point – see my no more v shaped recoveries article,” then I would have gotten smarted by reading his blog, instead of apparently stupider.

    Again, if Krugman had responded to the bet, I would have gotten smarter, because I would have learned what he meant in 09, so it honestly frustrates me that he didn’t.

    When it was apparent that Mankiw though Krugman was calling him evil (I thought so too – Krugman’s humor is really subtle, and made more difficult to tease out because it’s not funny) and that Mankiw thought Krugman was arguing in favor of the Obama growth estimates, that would have been a great time for Krugman to say “I meant X and Y but not Z.”

    I agree he was under no moral obligation to do so, but it makes him very frustrating to read, because I’m never sure if I’ve understood him.

  50. Gravatar of Anh Le Tran Anh Le Tran
    31. January 2014 at 14:50

    Well, if Krugman took the bet, both Mankiw and Krugman would probably get arrested by now for illegal gaming. It would be hard for their lawyers to argue their way out since there would be lots of witnesses to the bet, hahaha…

  51. Gravatar of ssumner ssumner
    31. January 2014 at 14:51

    Everyone, I wonder how many of the commenters who say Krugman wasn’t calling Mankiw evil were on my side when I was accused of calling Kaminska a “phony.” In that case I specifically said she wasn’t a phony, and still got massacred by liberals. Oh well.

    JSeydl, Maybe you should assume that I mean what I say, and not read into my posts all sorts of things THAT I DIDN’T SAY.

    Charlie. The claim that the unit root hypothesis fits the data pretty well (which I accept) is very different from the claim that GDP always follows a unit root. If there are time varying changes in trends and deviations from trends, the data may look like a unit root and Mankiw’s proposed bet might be a valid hypothesis, even if demand shocks cause devaluations from the time varying trend line.

    Here’s another way of making the point. There are problems with the unit root hypothesis. But those problems quite probably do not negate Mankiw’s claim that at 5 year horizons there is little trend revision. Maybe the solution is that trend reversion mostly occurs over long time horizons, and is thus hard to detect. But that would support Mankiw’s claim!

    Krugman seems to think that with enough demand stimulus we can get back to the pre-2008 trend line. Every year that goes by gives us more and more evidence that that view is wrong.

    There is an enormous literature on the unit root issue, Mankiw is not the only one who thinks it fits the data.

    Krugman was calling Mankiw’s idea’s evil. I took a little poetic license to add humor to the title of the post. No need to get bent out of shape. I was half joking in the title. No one thinks Krugman views Mankiw as a Voldemort. But he certainly thinks Mankiw is intellectually dishonest.

    Philip, Interesting analysis. My hunch is that the recession happened to occur at a point where growth was slowing anyway. I expect growth to also slow in places that did not have a recession, like Australia. But you raise an interesting hypothesis, and I have an open mind on whether demand shocks lead to long run structural problems.

    FEH, If it was just demand we would not be seeing sub-3% RGDP growth at a time of rapidly falling unemployment.

    Steve, See my reply to Charlie. And just a word of advice—when you don’t have anything substantive to say beyond personal insults, you aren’t helping your cause.

    J Mann and Vivian, Good comments.

  52. Gravatar of ssumner ssumner
    31. January 2014 at 14:56

    Charlie, One other point. Let’s say Krugman wasn’t predicting a rapid recovery. It’s indisputably true that Krugman often takes credit for predicting things where his views are equally vague. If we had gotten 15% growth he would have claimed he was right and Mankiw was wrong. No doubt in my mind.

    Doesn’t Krugman argue that Abenomics vindicates his views on money? And yet many took him to be saying there was not much the BOJ could do, to take just one example.

  53. Gravatar of David Cushman David Cushman
    31. January 2014 at 15:24

    Charlie, you wrote, “The Campbell and Mankiw paper is very highly cited, but every cite I open seems to conclude that they were wrong.” From this I thought one could infer that you thought the literature says CW were wrong about their unit root conclusion, not just their methodology. That is why I wanted to clarify that the real GDP-unit root literature as far as I can tell more often points toward a unit root (though certainly not without exception).

    As for what Krugman meant by his “deliberate obtuseness” remark, it seems unlikely to be Mankiw’s failure to mention methodological flaws. Krugman immediately wrote that he meant “pretending that you didn’t know the difference between, say, low GDP growth due to a productivity slowdown like the one that happened from 1973 to 1995, on one side, and low GDP growth due to a severe recession.”

    (And I think claiming one can identify the permanency or not of shocks that easily is a stretch.)

  54. Gravatar of David Cushman David Cushman
    31. January 2014 at 15:34

    Commenters Charlie and Philip Meguire above call the recovery from the recession “fairly weak and fragile” and “tepid.” I think it’s worth pointing out that for real GDP (which was what CEA forecast concerned) with respect to the former trend line (measured over several periods) the recovery is not weak, it is so far nonexistent, in fact, negative. Here are the annual real GDP growth rates (log differences) for several periods prior to the Great Recession:

    1930-2007: 3.4% (complete annual series from the BEA)
    1948-2007: 3.4% (post WWII)
    1973-2007: 3.0% (post oil shock break point proposed by Pierre Perron)
    1987-2007: 3.0% (Great Moderation period)

    The annual growth rate since the Great Recession: 2.4%
    (This is the annual growth rate using quarterly changes measured starting in either the 3rd or the 4th quarter of 2009 through the end of 2013.)

    To have any recovery so far toward the pre-recession trend you’d need a post-recession growth rate average in excess of 3.0% (or 3.4%). A growth rate <= 3% is therefore certainly suggestive of a unit root, but I will acknowledge that, as discussed by Cochrane in his comment on Campbell and Perron’s 1991 NBER Macro Annual paper, the lack of recovery so far, while suggestive, does not prove the presence of a unit root. Of course, the failure of many unit root tests to reject the unit root does not prove it either. No hypothesis in statistics can be proven or disproven.

  55. Gravatar of Charlie Charlie
    31. January 2014 at 17:09

    Scott,

    For the record, I was on your side on the Kaminiska thing. I didn’t think you said anything wrong in your post or in the linked conversation.

  56. Gravatar of Charlie Charlie
    31. January 2014 at 17:26

    Scott,

    I think you are still confounding

    1. The unit root hypothesis
    2. The Mankiw proposed bet

    It’s clear that PK is picking a fight over one and not two. Remember, the bet didn’t come until after Krugman’s post. Krugman himself in other posts said he didn’t agree with the administrations forecast. How many times did he trumpet the slow growth after financial crisis finding from R&R? I just don’t see how you, who reads Krugman regularly, thought he has been predicting robust growth over the time Mankiw proposed for the bet. In that light, I just can’t understand how you are misreading this argument. It’s just crystal clear to me that PK is talking about the unit root assumption and not the growth prediction.

    I agree with your other points that Mankiw could be right for the wrong reason, I said so in my earlier response. I do think calling him deliberately obtuse for citing the 1987 paper was not unfair, since the methodology won’t work. If he had cited the Nelson and Murray (2000) paper, I wouldn’t have had a problem with it.

    Lastly, when you wrote “Angry Bear needs to be sedated” did you mean that if I saw Robert Randy Waldman, I should shoot him with a tranquilizer dart. I read it as a joke. A funny play on words with “Bear” and “sedative,” but if I’m misreading Krugman, maybe I’m misreading you too. If so, I condemn you for calling for urging your audience to sedate RRW.

    P.S. When you have legitimate criticisms of Krugman, I agree with you. I agree with you about PK and Japan. But you are wrong this time.

  57. Gravatar of Charlie Charlie
    31. January 2014 at 18:08

    Scott,

    “It’s indisputably true that Krugman often takes credit for predicting things where his views are equally vague.”

    It’s just dawning on me how absurd this is. You want me to conclude that Krugman would take credit for this prediction, if it came true. Thus, even though he didn’t do anything wrong, he would have and I condemn him for that. It’s like saying that person has not stolen anything, but if he lost his job, he would steal something, so let’s throw him in jail.

    Let’s criticize and condemn Krugman for the crimes he actually commits (there are plenty), rather than the crimes he has yet to commit or would only commit in unobserved states of the world.

  58. Gravatar of Erik Trygger Erik Trygger
    1. February 2014 at 03:24

    Have Krugman changed his view on Europe again?

    “And yes, Europe is very much in a trap. Inflation is falling because the economy is weak, and the economy is being weakened in part by falling inflation. That’s the Japan syndrome. It leads eventually to actual deflation, but to the extent that there’s a red line (or more accurately, an event horizon), it’s crossed when monetary policy starts being limited by the zero lower bound, which happened years ago.”

    http://krugman.blogs.nytimes.com/2014/01/31/the-low-inflationary-trap/?module=BlogPost-Title&version=Blog%20Main&contentCollection=Opinion&action=Click&pgtype=Blogs&region=Body

    As an aside, this is somewhat funny

    “But while this is a natural human tendency, it’s also a mortal sin. I see it all the time: economists and public intellectuals of all kinds (and pundits of all kinds too) digging into an obviously false position because they refuse to admit that they were wrong — and it’s truly shameful. Folks, we’re talking about real policies that can make or break millions of lives. If you let your ego dictate your position on, say, monetary policy, rather than do your best to get it right, you’re doing something truly vile.”

    http://krugman.blogs.nytimes.com/2014/01/31/i-am-not-a-wise-man/?module=BlogPost-Title&version=Blog%20Main&contentCollection=Opinion&action=Click&pgtype=Blogs&region=Body

  59. Gravatar of ssumner ssumner
    1. February 2014 at 12:18

    Charlie, Well thanks for the support on the Kaminska affair. I think everyone here was half joking. Krugman in his evil comment, Mankiw in playing the victim, me in putting evil into the title of the post. Again, I don’t actually think Krugman regards Mankiw as evil. Perhaps I should have added some smiley faces.

    I do see your point here, but I wonder if you see mine. Let’s say that the unit root hypothesis is roughly correct, but that the RBC types drew the wrong implication, that all shocks are real. Perhaps nominal shocks tend to occur at times when productivity growth tends to slow for other reasons. That might result from interest rate targeting, for instance. Now in that case Mankiw might be right that the optimal 5 or 10 year forecast doesn’t change very much after a shock to RGDP, even one driven by demand side factors. And that’s true even if there is a big output gap for the next 2 years.

    I think Mankiw had a right to respond the way he did. He cites his beliefs about the time series properties of RGDP as a reason to be skeptical of the administration forecast. Krugman scoffs that the unit root stuff doesn’t apply to demand side recessions, and this is demand side recession. Mankiw says OK, lets see if it applies in this case. And Mankiw is right that it does. If you then claim Krugman didn’t think there would be trend reversion for other reasons, then that would be basically admitting that this isn’t a normal demand-side recession, or at least that even if it was there are other things going on that will prevent trend reversion. But that’s perfectly consistent with what Mankiw found in his unit root study. Mankiw is a demand-side economist. So I have to think he believes that supply and demand shocks are entangled in some sort of way, much as I do. he must think “other stuff” plays a role in the difficulty we have in proving trend reversion.

    I don’t know if this makes any sense, but I guess I’m arguing that you can’t accuse Mankiw of being right for the wrong reason. If there was some factor like the R&R study that suggested a slow recovery, that seems entirely consistent with Mankiw’s view that demand matters but trend reversion happens much less than you would expect despite demand mattering. R&R might have provided one reason why.

    You said;

    “Let’s criticize and condemn Krugman for the crimes he actually commits (there are plenty), rather than the crimes he has yet to commit or would only commit in unobserved states of the world.”

    You’ve obviously never seen the movie Minority Report. Seriously, I am fine with that, so let’s once again condemn him for claiming 2013 was a “test” of MM.

    Erik, As I recall Krugman (rightfully) condemned the ECB for raising rates in 2011. He also suggests they’ve been at the zero bound throughout this recession, which is why the 2011 austerity hurt. So he condemned them for raising rates at a time he insisted rates were stuck at zero? Or did I miss something?

  60. Gravatar of I Hope Nobody Took Brad DeLong’s Betting Advice I Hope Nobody Took Brad DeLong’s Betting Advice
    1. February 2014 at 15:34

    […] how did things turn out? Scott Sumner reports that “the actual 5 year RGDP growth just came in at slightly under 6.3%. That’s not even […]

  61. Gravatar of Full Employment Hawk Full Employment Hawk
    1. February 2014 at 16:32

    “Let’s say that the unit root hypothesis is roughly correct, but that the RBC types drew the wrong implication, that all shocks are real.”

    If a demand shock produces a strong hysterisis effect, mean reversion may be very slow or not occur at all. Nevertheless the shock was a demand shock. That is most likely to happen if the shock is exceptionally large and if expansionary demand-side policy is not engaged in quickly and agressively enough to restore the economy to potential output rapidly so that hysterisis does not have time to work its effects.

  62. Gravatar of Full Employment Hawk Full Employment Hawk
    1. February 2014 at 17:05

    “Krugman himself in other posts said he didn’t agree with the administrations forecast.”

    Krugman from the beginning argued that the stimulus was too small, would not do the job of restoring the economy to full employment and, as a result of the economy not recovering, critics of Keynesian economics would argue that this would show that a fiscal stimulus does not work. If he had accepted the optimistic adminstration forecasts, he would not have written such things. On the basis of the views Krugman expressed it made perfect sense for him to not accept Mankiw’s bet even though he did not believe there was a unit root.

  63. Gravatar of Erik Trygger Erik Trygger
    2. February 2014 at 05:45

    Scott, how about writing an “open letter” to Krugman or perhaps to NKs in general asking them to define the zero lower bound?

  64. Gravatar of Charlie Charlie
    2. February 2014 at 10:34

    Scott,

    I do see your point. I read Mankiw making a somewhat stronger statement than you did. Regardless, my main beef is not with Mankiw, but with everyone else interpreting this as PK backing down from a bet of beliefs. To me, it was obvious Mankiw offered Krugman a bet he wouldn’t take, and not taking the bet was perfectly consistent with his public views.

    “I think everyone here was half joking. Krugman in his evil comment, Mankiw in playing the victim, me in putting evil into the title of the post.”

    Yes, perhaps this is all a bit of Kabuki theater, and I can’t help but play the role of the irate blog commenter, pedantically dissecting a several year old blog debate that is basically meaningless by any rational view of the world. We all play our part.

    But I read your blog everyday and only comment so rarely, so think of how often we must agree.

  65. Gravatar of ssumner ssumner
    2. February 2014 at 12:01

    Erik, I tried an open letter to Krugman, it didn’t go well.

    Charlie, I agree Krugman was under absolutely no obligation to do accept that bet. I suppose I’d argue that if he thought growth would be slow it’s odd that he was so dismissive of Mankiw’s claim. Whatever reason Krugman offered for slow growth would be consistent with Mankiw’s claim, as far as I can see. The unit root hypothesis makes no claim as to why there is a unit root. If it have been Prescott using the unit root studies to mock demand side effects, then I could understand the attack. But Mankiw is a demand-sider too.

    Thanks for following the blog despite my occasionally grouchy replies.

  66. Gravatar of Leon Ellsworth Leon Ellsworth
    2. February 2014 at 13:39

    What is remarkable about this post is victory posturing when neither you nor Mankiw are even playing on the same field as Krugman. Krugman did not defend the GDP projection; he attacked Mankiw’s grounds for rejecting it:

    “But to invoke the unit root thing to disparage growth forecasts now involves more than a bit of deliberate obtuseness. How can you fail to acknowledge that there’s huge slack capacity in the economy right now? And yes, we can expect fast growth if and when that capacity comes back into use.”

    Last I looked, US unemployment was still way above normal which implies that there is still slack and the “if and when” condition has not been met. If and when it is, then see if there is fast growth and judge Krugman accordingly.

    Krugman rightly didn’t take the “bet” because it was irrelevant to the point he was making and the prediction he actually made.

  67. Gravatar of ssumner ssumner
    2. February 2014 at 13:57

    Leon, You would have saved yourself some time if you had read the comment section before posting. I never criticized Krugman for not taking the bet. Your other comments have also been addressed.

  68. Gravatar of ssumner ssumner
    2. February 2014 at 13:59

    Leon, One other point, growth since 2009 has been well below trend even as unemployment fell from 10% to 6.7%. So Krugman’s appeal to a falling unemployment rate turned out to be false.

  69. Gravatar of Full Employment Hawk Full Employment Hawk
    2. February 2014 at 23:19

    “asking them to define the zero lower bound”

    It is clear from Krugman’s writings, and the writings of other Keynesians following a similar approach, that the zero lower bound is reached when very short-term interest rates, especially the federal funds rate, have gotten so very close to zero that no further significant reduction in them is possible. Often this condition is described as “THE interest rate has reached the zero lower bound.” This assertion is obviously incorrect because in an economy with many different interest rates THE interest rate needs to be defined as a weighted average of all interest rates that have a significant effect on aggregate demand; and therefore THE interst rate has not reached the zero lower bound because there are many longer term interest rates, especially on mortgage type securities, that have not reached the zero lower bound and can be brought down farther.

    That is why the claim that the economy is in a liquidity trap is a highly ambiguous and confusing assetion. Krugman and similar economists assert that the economy is in a liquidity trap because the very short-term interest rates have reached the ZLB, even though many other interest rates have not. Under such conditions we are clearly NOT in a liquidity trap the way Keynes described it in the General Theory, on page 207, as a situation in which the LONG-TERM interest rate has reached a floor, not at zero, below which it will not go. [The term "liquidity trap" itself was not used by Keynes and was introduced later (by Hicks??)]. Keynes did not believe that his kind of liquidity trap had happened by the time he was wring the General Theory in 1935.

    I would suggest that to end the confusion about whether or not the economy is currently in liquidity trap a distinction needs to be made between a FULL LIQUIDITY TRAP, under which monetary policy totally ceases to be effective, and a SOFT LIQUIDITY TRAP. The economy is in a soft liquidity trap when very short-term interest rates have reached the zero interest floor so that what has become established as conventional monetary policy, pegging the federal funds rate, ceases to work. Under such conditions expansionary monetary policy still works, but because it has to rely on non-conventional monetary policy, it becomes more difficult and uncertain. In addition, in a soft liquidity trap, the central bank is unwilling to use monetary policy agressively enough to offset the effects of changes in fiscal policy, so that fiscal policy affects output and the fiscal multipliers work.

    The current debates about whether or not the economy is in a liquidity trap are therefore about whether or not the economy is in a soft liquidity trap, not about whether or not it is in a full liquidity trap. For example, Krugman’s writings indicate that he believes that the economy is in a soft liquidity trap, but not a full liquidity trap. He asserts that because we are in a liquidity trap the large Keynesian fiscal multipliers apply, but has criticised the Fed for not engaging in a more expansionary fiscal policy to move the economy toward full employment.

  70. Gravatar of J Mann J Mann
    3. February 2014 at 05:25

    Charlie, I’m not sure if I’m included in “Regardless, my main beef is not with Mankiw, but with everyone else interpreting this as PK backing down from a bet of beliefs,” but to the extent that I am, let me clarify:

    I don’t criticize Krugman for “backing down” from the bet, I regret that he didn’t respond.

    Presumably, Krugman’s response would have been to clarify the thing that was obvious to you and opaque to me – that he agreed with Mankiw’s conclusion that the 09 growth estimates were optimistic, but he disagreed with one of the hypotheses that Mankiw used to reach that conclusion.

    That would have been very informative and valuable back in 09. It’s one of the reasons I like bets – they force both parties to clarify what exactly it is that they predict.

  71. Gravatar of Erik Trygger Erik Trygger
    4. February 2014 at 05:38

    Krugman now says something that sounds like he again have changed his views:

    “Some people will argue that the Fed coulda and shoulda and maybe even did make up for its inability to lower short-term rates with other policies. My view would be that making monetary policy truly effective at zero rates, if it was possible at all, would have required more radical action than the Fed was willing or politically able to take.”

    One interpretation could be that Krugman thinks MM, even if correct, is politically infeasable. But does that mean that Krugman believes that fiscal stimulus is more politically feasible? Usually Krugman seems to believe that US politics are highly dis functional so it seems somewhat contradictory.

    http://krugman.blogs.nytimes.com/2014/02/03/changing-circumstances-graphed/?module=BlogPost-Title&version=Blog%20Main&contentCollection=Opinion&action=Click&pgtype=Blogs&region=Body

  72. Gravatar of The Mythology And Reality Of Unemployment Benefits The Mythology And Reality Of Unemployment Benefits
    10. February 2014 at 05:01

    […] unwillingness to admit fault – even after disparaging his opponents’ intentions – is now pretty much a given.)  Krugman’s ultimate conclusions, however, belie such optimism.  He, like many others, remains […]

Leave a Reply