Krugman mischaracterizes Lucas’s view of business cycles

Here’s Paul Krugman:

What Davies doesn’t say is that there’s a good reason Lucas won’t even consider the obvious explanation in terms of a shortfall in demand. More than 30 years ago, in a burst of radically premature triumphalism, Lucas and his colleagues declared the “Death of Keynesian economics”. As cited by Greg Mankiw(pdf), Lucas wrote that Keynesian theorizing was so passe that people would giggle and whisper if it came up in seminars.

Since then, as is obvious to everyone but the hermetic inhabitants of the freshwater world, the attempt to explain business cycles in terms of rational expectations and frictionless markets has failed; and Keynesian economics continues to be very useful. But to concede that, to even consider the possibility that we’re in a demand-shortfall slump of the kind Keynes diagnosed, would be an incredible comedown for Lucas.

This creates the impression that Lucas believes demand shocks don’t matter.  Let’s take a look at a few of the Lucas PowerPoint slides that Krugman is commenting on:

Why did the decline in deposits precipitate a depression?

* People, business firms like to hold certain ratio of cash to spending

* Bank failures, decrease in deposits, meant sudden decrease in liquidity-
cash and cash substitutes-held by public.

* To rebuild balances, businesses, households cut back on spending,
trying to restore comfortable ratio of cash/bank deposits to spending

* 48% deposit decline resulted in 58% spending decrease

So Lucas claims that the Great Contraction of 1929-33 occurred because bank failures led to money hoarding and a big fall in “spending.”  Hmmm, that sounds suspiciously like a demand-side problem to me.  Yet Krugman seems to think Lucas is some sort of doctrinaire RBC economist.  And then there is this slide:

What could have been done?
– Get more reserves-more cash-into the system
– Fed could have offset deposit decreases due to bank runs by increasing
bank reserves, permitting/encouraging sound banks to expand deposits
– Instead, Fed stood by and watched
– Milton Friedman, Anna Schwartz argue that this Fed failure was the
crucial policy mistake of the U.S. Great Depression.
– I agree.

I guess Milton Friedman is also a real business cycle economist, who denies demand shocks matter.  Now it’s true that Lucas goes on to argue that some of the New Deal programs delayed the recovery.  Perhaps that view puts Lucas beyond the pale in Krugman’s eyes.  Here’s that fanatical RBC economist, James Hamilton:

I openly confess to believing that government policies that were explicitly designed to limit manufacturing, agricultural, and mining output may indeed have had the effect of limiting manufacturing, agricultural, and mining output.

Then Lucas goes on to make the same argument about the current recession; a severe demand shock in 2008-09, followed by a slow recovery due to government policies that reduced aggregate supply:

– The effects of the credit freeze were also similar to the effects of the
bank runs of the 1930s
– A part of the effective supply of liquidity supply had vanished, other
money-substitutes now became suspect, everyone wanted to get into
government-issued or government-insured assets: reserves, currency,
and insured deposits
– Could see this in the widening spreads between treasury bills and commercial
paper, between government and corporate bonds, etc.
– A flight to currency? Not exactly. But a flight to government promises
of currency, current or future
– All of this similar to earliest stages of the Great Depression

And skipping ahead a few slides:

– Financial panic was over by end of 2008
– Too late to prevent deep spending declines in GDP in 2008-4 and
– But there is world of difference between two quarters of production
declines and four years!

Again, a demand-side explanation.  Although I’m 100% with Lucas on the demand shock/supply shock explanation for the Great Depression, I am somewhat skeptical of his supply-side argument that big government policies have greatly slowed this recovery.  Yes, there are policies that are reducing AS.  This is the first time in American history when we’ve adopted European-style extended UI (now a maximum of 99 weeks, roughly the same as Denmark.)  So the natural rate of unemployment may have risen a bit.  But there’s also lots of evidence that more demand would be helpful.  Stocks are still strongly correlated with TIPS spreads, a pattern you don’t see during normal times when there’s no demand shortfall.

Still, it’s a bit silly to suggest that Lucas is some wild-eyed radical.  After all, you can argue that he’s simply applying the (Keynesian) model that’s now standard in all the textbooks—AS/AD.  Think about the following:

1.  We have a severe AD shock, just like in 1981-82

2.  Inflation levels off at a lower level, just like in 1982-84

3.  But unlike 1983-84, the self-correcting mechanism doesn’t bring a fast recovery.  Why not?

4.  France never recovered from the 1974 and 1981 slumps.  Yet the Keynesians back then thought all France need was more AD.  They were wrong.

Again, I do think we need more monetary stimulus.  But Lucas is working within a standard macro framework.  The vast majority of economists agree with Lucas on the stimulus issue.  A recent poll showed 36 out of 38 oppose more monetary stimulus.  Krugman and I are now in the tiny minority, not Lucas.

In my view 20th century macro made a tragic error in focusing on explaining and controlling inflation, rather than explaining and controlling NGDP growth, which would have clearly shown that we had a severe AD shortfall.  But that’s not the standard model.  The standard NK model says an adverse demand shock causes high unemployment and lower inflation.  Then wages and prices adjust and rapidly bring down the unemployment rate.  It worked in 1983-84.  It didn’t work this time, despite a far smaller reduction in the rate of inflation.  Why not?

In this recovery my views are closer to Krugman.  My views of the 1933-40 recovery are closer to those of Lucas.   But none of this has to do with mythical disputes over “frictionless markets.”   Lucas is one of the best known macroeconomists of the past 50 years.  I’d think an economist who has a column at the New York Times would want to know something about his actual beliefs, before dismissing him with a crude and misleading caricature.  But then we’ve seen what happened to Fama and Cochrane, so I guess I shouldn’t have been surprised.

Et tu, Matt Yglesias:

I’d been dimly aware of Robert Lucas’ argument (PDF) that the Great Recession can be explained without reference to such petty concepts as aggregate demand

Yeah, “dimly aware” sounds about right.

Update; I was trying to be funny, but that probably came off as sarcastic.  Sorry Matt.

PS.  For some strange reason the comments are really piling up.  It will take days to get through the backlog.



25 Responses to “Krugman mischaracterizes Lucas’s view of business cycles”

  1. Gravatar of B B
    21. July 2011 at 20:59

    “Yeah, ‘dimly aware’ sounds about right.”

    Aww, poor Matt. Although he deserves it. He often wades a little too deep into technical discussions. And economists can be combative* (i.e. assholes).

    * I think it’s a self-selection thing. I love conflict.

  2. Gravatar of John John
    21. July 2011 at 21:32

    In the Keynesian tradition, Krugman is great at creating “make work” projects for economists who want to do things like accurately describe history, defend viewpoints from slander, point out miscellaneous distortionary of the truth, or point out simple fallacies Krugman uses to advance his political agenda. Sometimes I wonder if the man is a simple liar of poor moral character.

  3. Gravatar of W. Peden W. Peden
    22. July 2011 at 03:13


    He probably just has his mind on higher things. If one is a leading economist in a huge political movement, then one doesn’t feel the need to bother with the time-consuming niceties of academic debate e.g. working out what the other side is saying.

    Even St. Friedman became lazy in his later years. Logical argument is a lot of WORK and Krugman is likely lacking in time/energy, which usually follows from having the position of a major, influential and highly-respected economist.

    I don’t know what Matt or DeLong’s excuses are, though.

  4. Gravatar of John John
    22. July 2011 at 05:05

    It’s more than that, there are times when he will say things that really strain the truth. For instance he accuses Hoover of savagely cutting spending when Hoover took government spending from 3.1 billion in fiscal year 1929-30 up to 4.6 billion in fiscal year 1932-33. While this was going on, tax receipts fell from 4.1 billion to 2 billion. Yet Krugman will simply accuse Hoover of slashing spending and trying to balance the budget. Then he has the audacity to say that facts have a Keynesian bias! Maybe the numbers you fabricate do Paul.

  5. Gravatar of Brito Brito
    22. July 2011 at 05:50

    I think the main problem is that Lucas, Cohcrane et al are inconsistent. They’ll be clear that there is a demand shortfall in their papers or slides on monetary policy. But then when attacking fiscal stimulus they’ll suddenly use terrible arguments that imply that there is no demand shortfall or any excess capacity.

  6. Gravatar of Brito Brito
    22. July 2011 at 05:51

    John, Hoover was trying to cut spending, he just didn’t get his way in congress.

  7. Gravatar of Scott Sumner Scott Sumner
    22. July 2011 at 05:54

    B, I probably shouldn’t have said that, but it’s 100 degrees and I have no ac in this room–so that’s my excuse. 🙂

    John, He does create a lot of work for others. My view is that highly partisan people create their own reality. I think he’s well intentioned, but lets biases creep in too much. I need to fight against that tendency myself, so I’m not surprised others push the envelope a bit.

    W. Peden, See my reply to John.

    John, He sometimes selects data that gives one impression, when a more complete and balanced look at the data would give another. The Hoover spending is a good example. He had a graph showing the national debt rose much faster under FDR. But so what? I believe the 1930 budget was passed in 1929, before the depression, it isn’t until 31 that Hoover really gets going. I’d guess 1933 also reflects the last Hoover budget.

  8. Gravatar of Scott Sumner Scott Sumner
    22. July 2011 at 05:59

    Brito, I think Hoover did favor public works, just not the bonus bill. But even FDR opposed the bonus bill, didn’t he?

    Yes, there are poor communication skills, plus a tendency to underestimate how long demand shortfalls can persist. Combine the two and you get some statements that look silly at fist glance. I was struck by Cochrane’s paper that they trashed. On page two or three you have that long quotation I provided, and it’s 100% clear that Cochrane understands what they claim he doesn’t understand. But why didn’t he put that up front on page one? Many people don’t read entire papers.

  9. Gravatar of Lee Kelly Lee Kelly
    22. July 2011 at 06:15

    Sarcasm? I thought the headline was sarcastic enough.

    Krugman mischaracterises someone elses views to score political points, also, the bears **** in the woods. Stay tuned for more.

  10. Gravatar of Brito Brito
    22. July 2011 at 06:19

    I was thinking. Nobody can actually believe that the government can NEVER boost output under any context, otherwise how could Cuba and other centrally planned countries exist with more than zero output?

  11. Gravatar of Richard A. Richard A.
    22. July 2011 at 07:46

    “A recent poll showed 36 out of 38 oppose more monetary stimulus.”

    This is a tiny sample size which could be biased as well. How about a random sample of US economic professors who specialize in macroeconomics?

  12. Gravatar of Mark A. Sadowski Mark A. Sadowski
    22. July 2011 at 08:03

    You wrote:
    “In this recovery my views are closer to Krugman. My views of the 1933-40 recovery are closer to those of Lucas.”

    This reminds me of how idiosyncratic my own views are compared to most of what I read.

    My own views are that we are suffering a demand shortfall and that monetary policy holds the key to getting us out. In that sense my views of the current recovery are represented more by you Scott than any other single macroeconomist. (Although there is a growing club of quasimonetarists whose opinions I also geberally share.)

    However, with respect to the Great Depression, I feel very much in the monority.

    Like the left I like to point to the blistering hot recovery of 1933-37. But very much unlike the left I point out that Romer, Temin, Eichengreen etc. credit this rapid recovery almost entirely to monetary policy, not fiscal policy. (Sheesh, give me a break!) I’ve spent many words to little effect trying to persuade progressives that the best part of the New Deal is the part most of them have never seemed to heard of.

    Unlike the right, I don’t think the economy failed to recover, especially in 1933-37. (Anyone up for an average 9.5% RGDP growth rate?) And thus I don’t think NIRA and other policies, as counterproductive as they might have been, really interferred that much with the recovery. In a depression the quality of demand side policies easily trumps the quality of supply side policies. (In short, Lucas is completely and absolutely wrong about the recovery from the Great Depression.)

    My views are in part tempered by my own research, which is supply side in nature. Supply side policies simply cannot have dramatic short run effects. Their effects are more like the long term effects of rain and wind in eroding rocks. Over time the effects can be quite significant, but don’t expect anything great in the short run. Demand side policies on the other hand can have quite dramatic effects in the short run.

    This recession and this blog have made me more interested in doing research into macroeconomic stabilization policy (as soon as I finish defending my dissertation). I’d like to think I have a well grounded respect for, and proper understanding of the proper place of, demand and supply side policy.

  13. Gravatar of Benjamin Cole Benjamin Cole
    22. July 2011 at 08:35

    My apologies in advance but–this blog proves what I have said: No one in the USA will ever convince the “opposition” about the Great Depression, the Great Recession and what caused it and how to fix it, if they use the USA as the platform for argument. You think John Taylor or Krugman is going to suddenly see the light?

    Okay, bear with me, I ask Scott Sumner and NGDP crowd to consider this document from the Hong Kong Monetary Authority:

    It is a study of mainland China’s monetary policy. It concludes:

    “Our estimated monetary reaction functions for the PBoC’s show that the
    key factors of monetary policy are economic growth and inflation, which are in fact the PBoC’s stated mandate. Of the two main policy objectives, the PBoC appears to react more strongly to deviations of growth from targets.”

    Okay, so the Hong Kong Monetary Authority says that mainland China tilts towards growth with their monetary policy. And let’s look at Japan: We know the Bank of Japan puts inflation-fighting first.

    And what has been the result? How have the two nations fared in the last 20 years?

    Answer: China has boomed, while Japan is shriveling.

    This is the argument we need to bring to the public. This is the insight we need at this time.

  14. Gravatar of marcus nunes marcus nunes
    22. July 2011 at 11:36

    And now he´s baaack into denying any role for MP!

  15. Gravatar of TylerG TylerG
    22. July 2011 at 12:34


    I understand your view of Krugman as the intellectual champion of the progressive-Keynesian camp. Although Krugman does articulate left-wing arguments well, I would like to see him put his mouth where his money is first before you take him more seriously. Recall in 2009 where after implying that Mankiw was ‘evil’ for his skepticism of the administration’s growth forecast, he never accepted Mankiw’s proposed bet that RGDP in 2013 would be less than 15% higher than it was in 2008 – which the forecast called for.

    While blogging has been a nice internet innovation, it’s also been a platform for smart economists to unproductively slander eachother. I think that’s one big external cost to the economics profession.

    I follow this blog because of your perspective on monetary issues and promotion of Fed NGDP futures targetting, but in particular I like that you’ve been willing to escalate this issue.

  16. Gravatar of B B
    22. July 2011 at 12:51


    Krugman says:
    “So here’s a picture to illustrate the first point. It shows monetary base and nominal GDP since the recession began; if any kind of quantity theory applied, these should have gone up in tandem.”

    I thought the baseline of these conversations has always been that we were destined for the Next Great Depression. Now he’s arguing the larger monetary base + NGDP did NOT plummet catastrophically = failure?

    I guess by that logic we can shout down anyone that says, “without the stimulus act things would have been much worse.”

  17. Gravatar of marcus nunes marcus nunes
    22. July 2011 at 13:01

    He´s very wrong. BmV=Py. And m and V plummeted!

  18. Gravatar of Scott Sumner Scott Sumner
    22. July 2011 at 17:08

    Brito, But they may believe the government can’t boost it from current levels.

    Richard, I agree it’s biased, but still a striking result, as it’s obvious to me we need more stimulus.

    Mark, Annual GDP growth rates are very misleading. industrial production rose 57% between March and July 1933–we were half way out of the Depression in 4 months. Then FDR raised nominal wages 20%, and IP fell, not regaining July 1933 levels for another 2 years. That policy was a disaster. He tried it 5 times, and every time it killed the recovery. When he wasn’t screwing around with wages, the recovery was explosive, like 1922.

    The idea that fiscal policy was really tight in 1937-38 is somewhat of a myth. I did a recent post on that. The wage shock of 1937 was the biggest problem (along with gold hoarding. In mid-1940, FDR had been in office 7 years, and we had about 15% unemployment. That’s not good.

    You said;

    “Supply side policies simply cannot have dramatic short run effects.”

    Check out the July 1933 wage shock, before you make that claim.

    Benjamin, Yes, people need to consider Japan.

    Marcus, Well at least he’s consistently inconsistent.

    Tyler, I did a post earlier in the year pointing out that Mankiw almost certainly would have won that bet.

    B, Of course if they are paying interest on reserves, the relationship breaks down. In fairness, even without IOR the demand for base money rises at near-zero rates. We all know that, but it’s not the issue. No one is still calling for the old monetarist k% rule.

  19. Gravatar of Scott Sumner Scott Sumner
    22. July 2011 at 17:12

    Marcus, I actually don’t read that as denying any role for MP, but 99.9% of his readers will infer that. And he knows that. BTW, check out the Forbes article Matt links to–I get a lot of good publicity.

  20. Gravatar of Mark A. Sadowski Mark A. Sadowski
    22. July 2011 at 17:30

    You wrote:
    “Mark, Annual GDP growth rates are very misleading. industrial production rose 57% between March and July 1933-we were half way out of the Depression in 4 months. Then FDR raised nominal wages 20%, and IP fell, not regaining July 1933 levels for another 2 years. That policy was a disaster. He tried it 5 times, and every time it killed the recovery. When he wasn’t screwing around with wages, the recovery was explosive, like 1922.”

    I’ve got the numbers right in front of me. True, industrial production did not exceed July 1933’s level until August 1935 but by May 1937 they were up by another 40% for a total gain of 121%. I don’t care how bad the supply side policies were, 121% in 4 years and two months burns. We should be so luck now.

    You wrote:
    “The idea that fiscal policy was really tight in 1937-38 is somewhat of a myth. I did a recent post on that. The wage shock of 1937 was the biggest problem (along with gold hoarding. In mid-1940, FDR had been in office 7 years, and we had about 15% unemployment. That’s not good.”

    I never mentioned fiscal policy. Like F&S (and Romer) I ascribe the cause of the 1937 recession primarily to monetary policy (upping the reserve requirements and the subsequent gold hoarding).

    And just to nitpick, counting the FERWs as employed, unemployment was 9.5% in mid 1940 and by early 1942 we were way more than back to trend growth.

    You said:
    ‘”Supply side policies simply cannot have dramatic short run effects.” Check out the July 1933 wage shock, before you make that claim.’

    I’ll take a second look, but I think I’ve already addressed that effect in this comment.

  21. Gravatar of Skip Skip
    22. July 2011 at 18:38

    I think a short interview from Snowdon and Vane’s 2005 book “Modern Macroeconomics” holds the key to Lucas’s views on recessions. Sadly I cannot find a free copy of the interview anywhere. The book essentially divides economics into several schools- I believe there are chapters on old Keynesians, Monetarists, New Classicals, Real Business Cyclists, and New Keynesians. Interviews with proponents of each are included, with folks such as Tobin, Milton Friedman, Lucas, Prescott and Mankiw making appearances.

    In his interview when asked about his thoughts on Real Business Cycle theory he essentially says that he thinks RBC theory is basically right WHEN MONETARY POLICY IS GOOD.

    For further illustration of his divergence from Prescott here’s a bloomberg interview with both of them:

    Prescott chastises (in a friendly way) Lucas’s focus on monetary factors in the recent recession/historical recessions.

  22. Gravatar of Jim Glass Jim Glass
    22. July 2011 at 20:12

    …what has been the result? How have the two nations fared in the last 20 years? Answer: China has boomed, while Japan is shriveling.

    While it’s true the countries have followed the different money policies and have had the different results, there are some pretty darn big real differences in their economies that might have rather more to do with their different growth rates. It’s a lot easier to get rapid GDP growth from a very low base with a surging workforce than it is from a very high base with a shrinking workforce, etc.

    Monetary policy is supposed to be neutral in the long run. With Japan we are talking about 20+ years now.

  23. Gravatar of Lorenzo from Oz Lorenzo from Oz
    22. July 2011 at 20:37

    One of the striking things about American politics (in the broadest sense, including policy debate) is how “alive” FDR and the 1930s are. Part of that is because, more than any other developed democracy, the 1930s are in the US stand-out period for increase in the size of government.

    If we divide the US into the “Southern Republic” (1787-1860), the “Northern Republic” (1860-1932) and the “Welfare Republic” (1932-?) then FDR and the New Deal was as big a transformation of US governance and politics as the Civil War.

    The US did not experience much of the 1950s government expansion that most Western democracies did (except in matters military), and the 1960s expansion has remained contested (hence the welfare reform of the mid-90s). A history that both came out of, and reinforced, the contested nature of the 1930s.

    That the 1930s still looms so large in framings of US politics makes it harder to get agreement on but also makes it a much more alive area of analysis. Getting folks to accept that monetary policy was central to the history of the 1930s threatens all sorts of ideologically fundamental framings.

  24. Gravatar of Mark A. Sadowski Mark A. Sadowski
    22. July 2011 at 21:19

    Lorenzo from Oz,
    It’s nice to see familiar names of people who’s opinions I value (and usually agree with) still here. I reckon Scott’s “sabbatical” couldn’t kill a good thing after all.

  25. Gravatar of Scott Sumner Scott Sumner
    23. July 2011 at 11:11

    Mark, Yes, Growth from mid-1935 to early 1937 was very fast. It was triggered partly by the Supreme Court throwing out the NIRA in May 1935. A lucky break for FDR.

    Skip. Thanks, I got part way through, and will do the rest later.

    He’s right that the RBC theory is true when you have good monetary policy.

    Lorenzo, That’s a good point. I’m working on a post now that makes a similar point.

Leave a Reply