The Keynesian bubble

I’m increasing inclined to believe that the Keynesian way of thinking about the world (that spending shocks drive the economy) is so intuitively appealing that it becomes hard to visualize any other worldview.  As a result, Keynesians start to think that any piece of empirical evidence, or logical argument, somehow supports the Keynesian model.

1.  One obvious example from a while back is the expansionary effect of tax cuts.  Yes, but how do we know whether they are expansionary for demand or supply-side reasons?

2.  A much more interesting example occurred recently, when a series of studies showed positive “regional multipliers.”  More federal spending in a given area tends to increase regional GDP.  I would have thought that rather obvious, as all the anti-Keynesian models that I am aware of would predict the same result.  But people like Paul Krugman and Brad DeLong latched on to this as evidence in favor of the Keynesian model.  How can it be evidence in favor, if it would also expect positive regional multipliers if the national multiplier was zero?   Take the famous Alaskan bridge to nowhere.  Does anyone serious doubt that the GDP of that tiny Alaskan hamlet would have increased had the $100 million bridge been built?  Even Chicago economists would say yes.

So it seems to me that Keynesians sometimes make the mistake of simply assuming the Keynesian model is true, and then seeing any and all real world events as somehow confirming that belief.  I’m not claiming they are stupid; indeed I’m sure if they thought about the alternative explanations for a few minutes they see that those are also logically possible.

Recently we saw another example, when Wren-Lewis claimed that consumption smoothing somehow refuted Cochrane’s claim that the balanced budget multiplier is zero.  The problem, of course, is that it doesn’t refute Cochrane, as consumption smoothing is also expected to occur in anti-Keynesian models.  Milton Friedman’s Permanent Income Hypothesis is a great example of a consumption smoothing model, and by the mid-1990s Milton Friedman had reached the conclusion that fiscal stimulus is ineffective.

There are of course many reasons why the balanced budget multiplier might be zero.  Many of them I find weak, although one (the monetary offset argument) seems pretty persuasive to me.  But the important point is not whether you agree or disagree with these non-Keynesian models, what’s important is whether you can understand them on their own terms.  Let’s take three quick examples:

1.  The supply-side argument:  Supply-siders claim that higher tax rates discourage work, and thus lower GDP.  They also tend to be skeptical of the efficacy of public sector output.  If tax increases do discourage private sector output (C+I), then there is no particular reason why you won’t expect consumption smoothing to occur, with consumption falling by only a modest fraction of the total decline in private sector output.

2.  The crowding out argument:  In this view the decline in after-tax income will reduce saving, and hence crowd out private investment.  Again, this is completely consistent with consumption smoothing, indeed in a sense it assumes consumption smoothing.  Investment is assumed to be a big part of the decline in private sector output.

3.  My favorite argument is the monetary policy offset argument.  Under modern monetary regimes such as inflation targeting, any government attempts to boost AD are offset by less monetary stimulus, and hence AD is unaffected.  Again, in that case it’s easy to envision consumption smoothing occurring, as consumer theory predicts it will occur any time after-tax income falls.

So why do Wren-Lewis and Krugman think that consumption smoothing shows Cochrane was wrong?  What could they have possibly meant by this argument?

Both make the same simple error. If you spend X at time t to build a bridge, aggregate demand increases by X at time t. If you raise taxes by X at time t, consumers will smooth this effect over time, so their spending at time t will fall by much less than X. Put the two together and aggregate demand rises.

Read these over several times, especially the “put these two together” and the “so their spending at time t will fall by much less.”  I’m sorry, but I simply can’t accept the convoluted explanations being offered by their defenders.  Let’s do this with numbers to make it easy to see.  Wren-Lewis is basically saying that if you raise taxes by $100, then C will fall by much less than $100 (consumption smoothing), so their spending will fall by much less than $100.  Assume, for example, that C falls by only $20.  Then he says “put these two together.”  Correct me if I’m wrong, but isn’t the straightforward interpretation that he was putting together the $100 increase in G and the $20 fall in C, and noting that the fall in C is smaller than the increase in G?  Isn’t he implicitly assuming that a fall in C is equivalent to a fall in “spending?”  I can’t read it any other way. Otherwise why use the term “so” in explaining the impact on “spending?”  I’ll bet 99.9% of readers read it this way.

The best counterargument I’ve seen is from Andy Harless, who suggests that consumption smoothing somehow underlies the Keynesian result.  Even if he’s right I’d still claim they made a bogus criticism of Cochrane, is it doesn’t weaken his (non-Keynesian) argument.  But is he right?  Isn’t the balanced budget multiplier one in the Keynesian model regardless of the MPC?  And isn’t the regular multiplier actually higher with a bigger MPC, i.e. a smaller amount of consumption smoothing?  Andy understands the Keynesian model better than I do, so I’ll keep an open mind on that issue.  But it’s going to be pretty hard to convince me that their “consumption smoothing” criticism of Cochrane makes any sense.

Of course both Wren-Lewis and Krugman could say; “we simply assume the Keynesian model is right, and if you assume it is right, then it is right.”  And Cochrane could simply say “I simply assume the Keynesian model is wrong, and if you assume it’s wrong, then it’s wrong.”  I hope we don’t fall into that sort of debate in the comment section.  Let’s actually evaluate the quality of the arguments, not whether they happen to be true or not. [Update:  I obviously meant whether the conclusions happened to be true.]  Are they convincing?  I say both sides are completely unconvincing. I say both sides simply assume income changes or doesn’t change according to their preconception.  And I think many find the Keynesian model so intuitive appealing that they conflate “Keynesian reasoning” with “logical reasoning.”


Tags:

 
 
 

47 Responses to “The Keynesian bubble”

  1. Gravatar of math math
    16. January 2012 at 09:19

    Scott Sumner said in the previous post about ‘consumption smoothing’: “Math, Sure growth might have occurred, but so what?”

    OK, it’s good that you agree. Now, look at it in dynamics.

    Consumers can assume that government will repeat that trick to increase the growth further. Therefore, they can increase the current consumption in anticipation of the future growth.

    It’s similar to the TBTF problem. Banks are behaving risky in the anticipation of the future bail-outs.

  2. Gravatar of Kevin Donoghue Kevin Donoghue
    16. January 2012 at 09:27

    Andy Harless is right and it’s not a bogus criticism of Lucas and Cochrane. A perfectly reasonable interpretation of their remarks is that dY/dT is at least as great as dY/dG in any respectable model, which is not true.

  3. Gravatar of SG SG
    16. January 2012 at 10:02

    I’m curious to know what Cochrane thinks of all this.

  4. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    16. January 2012 at 10:05

    I’m glad I’m not Paul Krugman having to respond to his fans’, ‘Say it isn’t so, Paul.’

  5. Gravatar of Alex Godofsky Alex Godofsky
    16. January 2012 at 10:09

    Scott:
    “Let’s actually evaluate the quality of the arguments, not whether they happen to be true or not.”

    I think you meant to say “just happen” there, otherwise the meaning isn’t what you intended.

  6. Gravatar of Jason Jason
    16. January 2012 at 10:19

    On point 2, I found it was statistically unlikely that the regional multipliers would add up to a zero (or negative) effect in the Feyrer and Sacerdote study than a positive one. It is true F&S did not show this explicitly, but there is a 1-sigma significant increase in the national employment based on summing up the county and state level effects.

    http://www.themoneyillusion.com/?p=9213#comment-62530

    In a sentence, the individual regional effects are too big and too biased towards net benefit to add up to no national effect.

    It is true that F&S did not show this explicitly nor did Delong or Krugman; however on the other side, no one showed explicitly that the numbers did not add up.

  7. Gravatar of MI MI
    16. January 2012 at 10:48

    Really hoping Nick Rowe will clear this debate up for me.

  8. Gravatar of grooft grooft
    16. January 2012 at 10:51

    Could you explain why consumption smoothing would act the same in times of financial crisis / housing wealth melt-down / Lesser Depression and times of 10% unemployment as during times of 7.5% unemployment.

    Krugman writes clearly about how the model is aligned with individual decisions at the micro level and how these roll up to macro effects. I am confused why “consumption smoothing” would work at the individual level during an economic crisis where individual wealth disappears overnight and the long-term income prospects at the individual level go from “reasonably certain” to “OMG, I’m gonna lose my job”.

  9. Gravatar of Patrick Patrick
    16. January 2012 at 10:56

    1. It’s true that consumption smoothing doesn’t rule out a balanced budget multiplier of 0. If, for whatever reason, G and C are perfect substitutes (eg. Gov decides to provide school lunches so parents stop making lunches) then increased G will be offset by reduced C even with consumption smoothing.

    2. Krugman, Delong etc. never claimed that consumption smoothing means the bbm can’t be zero. Rather, they were pointing out that even with full rational expectations, Ricardian equivalence, consumption smoothing, perfect information etc. government stimulus can still be effective. They were pointing this out because Lucas/cochrane were ridiculing the notion that stimulus could be effective. It wasn’t up to krugman etc. to show that stimulus would always be effective, but rather to show stimulus could be effective even in the rational expectations world.

    3. Delong sums it up nicely here:

    http://delong.typepad.com/sdj/2011/12/ricardian-equivalence-is-a-claim-that-tax-cuts-are-ineffective-stimulus-not-that-spending-increases-are.html

    ” I learned this from Andy Abel and Olivier Blanchard before my eyes first opened: increases in government purchases are ineffective only if (a) “Ricardian Equivalence” holds and (b) what the government buys (and distributes to households) is exactly what households would buy for themselves. RE by itself doesn’t do it.”

  10. Gravatar of Morgan Warstler Morgan Warstler
    16. January 2012 at 11:36

    math

    “Consumers can assume that government will repeat that trick to increase the growth further. Therefore, they can increase the current consumption in anticipation of the future growth.”

    this is EXACTLY WRONG.

    What happens is the Tea Party.

    The RATIONAL EXPECTATION is that we become more libertarian on the time horizon, and less likely to extend the safety net, or do gvt. stimulus.

    The right is winning since 1980. DeKrugman is an outlier.

    Silly rabbit, trix are for kids.

    —-

    Note: this is exactly why so much rides on Scott and I’s bet – if Obama loses for not making like Bill Clinton, then SCOTT was wrong for couching his policy on ying side (print money now), and instead should have focused on the yang side (how his policy = less inflation and smaller govt.)

  11. Gravatar of math math
    16. January 2012 at 11:49

    Morgan Warstler,

    I don’t care if it is right or wrong in practice from the moral point of view. My point is that “consumer smoothing” does have an economic effect during the financial stimulus. Thus, both Krugman and Wren-Lewis are theoretically right and Sumner is theoretically wrong here.

    Btw, this “anticipation of growth” is somewhat equal to the “rational expectations” doctrine of Chicago school from the conceptual point of view.

  12. Gravatar of BW BW
    16. January 2012 at 12:35

    What Patrick said.

    Scott, Krugman isn’t arguing that consumption smoothing implies a positive multiplier. He’s arguing that Ricardian equivalence does not imply a zero multiplier.

    Here’s how to look at it. Suppose Krugman was completely ignorant of the monetary offset. Suppose also the monetary offset argument is completely right. What Krugman said about Cochrane could still be 100% true.

  13. Gravatar of Rien Huizer Rien Huizer
    16. January 2012 at 12:48

    Scott,

    There is so much to agree with on your blog which makes comments redundant but this

    ” I hope we don’t fall into that sort of debate in the comment section. Let’s actually evaluate the quality of the arguments, not whether they happen to be true or not. Are they convincing? ”

    What do you mean?

  14. Gravatar of Morgan Warstler Morgan Warstler
    16. January 2012 at 12:50

    math,

    me personally, I’m making a larger argument – what happens is congress gets drummed out and terrified of such a tax hike.

    it is “wrong in practice” because it doesn’t happen. when taxes are raised, the public seeking to increase their own personal standard of living and THROWS OUT Congress…. they EXPECT to lower their taxes.

    You make the same mistake that DeKrugman makes:

    1. The top 20% of the population consumes the vast majority of the stuff.
    2. The top 20-1% earn significantly more than even the top 1%.
    3. The top 20% all VOTE. We call them the Tea Party.
    4. More than just the top 20% identify as being in top 20% because many people spend part of their earning lifetime there.
    5. And those in #4 own the huge % of 225M guns.

    My point: The folks described in #4 are the hegemon in the US, what in bi-polar game theory you would call the A power.

    THEY DECIDE. Not you. Not the Fed. Not the rousted up proles.

    That isn’t a moral argument. It is just a fact.

    And the difference between now and say 1930, is that the A Power is now a gvt. crushing beast. Happy to see money wasted on wars precisely because they don’t want to buy butter for poor people…. since the left won’t let them keep the tax dollars if they end the wars.

    Let me say it this way – economists/economics/money/monetary policy are bishops at best in a LARGER GAME.

    The game is power. And the A Power doesn’t get tricked into behavior by the Fed or GVT. they literally get whatever the hell the want, the the other folks (even the top 1%) are wise to not get in their way.

    You made an assumption, that when GVT acts people will predict it will happen again – Corzine just learned this betting on bailouts.

    This is my radical assumption: DeKrugman would be more effective at achieving his ends if he began with saying, “beggars can’t be choosers” and then argued for stimulus meant to convince the top 20% that the bottom folks were being given eggs, flour, butter, pans, and cooking lessons, and not food stamps – because he’d ACTUALLY GET MORE SPENDING for the stimulus and…

    A less hostile tax payer = better consumption smoothing

    —–

    What Scot said on points carries.

    Wren-Lewis said consumption smoothing is why Cochrane is wrong. That was wrong.

  15. Gravatar of math math
    16. January 2012 at 13:12

    Morgan Warstler,

    Nope, I still disagree.

    Consumption smoothing is only one of the reasons why Cochrane is wrong, but it is not the only reason. He is wrong for another reason because fiscal stimuli can themselves initiate the economic growth (even without taking into account the idea of consumption smoothing).

  16. Gravatar of Tommy Dorsett Tommy Dorsett
    16. January 2012 at 13:17

    In Krugman’s Investment/Savings diagram, he ‘shifted’ an upward sloping savings schedule ‘up’ about an upward sloping investment schedule, denoting a drop in GDP (income). But would this simply not be income (GDP) falling, occasioning the upward/backward shift in the savings schedule? The starting point thus had to be a fall in Y(GRP, income), otherwise we’d have to expect an outward shift in the savings schedule, no? So he’s assumed a fall in income and drop in S,I but I don’t see how we get there without a decline in the money supply or its velocity. Thus, the assumption is ‘savings and investment fall’. But why? Because GDP falls….because MV falls. He leaves this out of course because he’s advocating a large rise in G as the primary offset the fall in S,I and C.

  17. Gravatar of John Schultz John Schultz
    16. January 2012 at 13:30

    I agree that Wren-Lewis pointing at “consumption smoothing” was incorrect and sloppy thinking / speaking at best. However, I think you are latching onto his one incorrect argument and ascribing it to a wider population than the actual people who believe it.

    For example, Krugman had been posting on this topic, which was actually whether full Ricardian equivalence implies that fiscal stimulus is ineffective, long before Wren-Lewis chimed in and Krugman gave him a thumbs up. Krugman almost surely read Wren-Lewis’s conclusion which agrees with his analysis and gave it a thumbs up giving little consideration to the term “consumption smoothing” that Wren-Lewis threw in there. The arguments on their face sound similar even though they aren’t really the same at all.

    http://krugman.blogs.nytimes.com/2009/04/06/one-more-time/

    Here’s Krugman’s core argument:

    “If the government introduces a new program that will spend $100 billion a year forever, then taxes must ultimately go up by the present-value equivalent of $100 billion forever … wiping out any expansionary effect of the government spending … But suppose that the increase in government spending is temporary, not permanent “” that it will increase spending by $100 billion per year for only 1 or 2 years, not forever. This clearly implies a lower future tax burden than $100 billion a year forever, and therefore implies a fall in consumer spending of less than $100 billion per year. So the spending program IS expansionary in this case, EVEN IF you have full Ricardian equivalence.”

  18. Gravatar of math math
    16. January 2012 at 13:46

    Morgan Warstler,

    Moreover, I would even go one step further by saying that direct effect of financial stimuli on the economic growth is the primary reason of why Cochrane is wrong and consumer smoothing (as an “indirect effect”) is the secondary reason.

    So, Krugman and Wren-Lewis’s is the only partial fault that they have not explicitly listed the aforementioned primary reason in that piece.

  19. Gravatar of Morgan Warstler Morgan Warstler
    16. January 2012 at 14:20

    “I would even go one step further by saying that direct effect of financial stimuli on the economic growth”

    Yeah this is wrong on its face. You get Solyndra and Obama’s weatherization joke, and no new high-speed trains. Nothing else. Nada.

    You’d again do BETTER to follow my lead… use fiscal stimulus to drive down wages and end unemployment (productivity gains). Forget “investment” – leave that to the real people.

    See my Guaranteed Income plan:

    http://biggovernment.com/mwarstler/2011/01/04/guaranteed-income-the-christian-solution-to-our-economy/

    math, it is easy to make systematically better pro-gvt. policy than you and yours because I don’t actually want to give away free stuff to buy votes. That’s the underlying argument of everything DeKrugman and Simon do.

  20. Gravatar of Noah Smith Noah Smith
    16. January 2012 at 14:28

    This argument against “regional multipliers” is not correct if the regressions weight states by population.

  21. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    16. January 2012 at 14:33

    Here’s Krugman on MLK, jr, using the same fallacious logic as he did criticizing Scott:

    http://www.nytimes.com/2012/01/16/opinion/krugman-how-fares-the-dream.html?_r=1

    ‘…King “” who was campaigning for higher wages when he was assassinated “” would surely have considered soaring inequality an evil to be opposed.’

  22. Gravatar of math math
    16. January 2012 at 14:39

    Morgan Warstler,

    OK, I don’t have to convince you. Let me try one more time.

    Yup, capitalists usually are savvier than government with investments. But sometimes they just don’t want to undertake the projects. For example, it may be more profitable for capitalists to engage in some kind of trouble-free or even societally destructive (e.g., casino operations) activities than undertake a heavy manufacturing or infrastructural project.

    I’m not sure if I don’t want to advocate that kind of fiscal stimuli here from the practical point of view but I do recognize that those possibilities exist from the theoretical point of view.

  23. Gravatar of Rob Rob
    16. January 2012 at 15:23

    I think Wren-Lewis is possibly saying that that even if you assume the balanced budget multiplier is zero then a non-Keynesian concept (consumption smoothing) still indicates that Cochrane’s claim that tax-financed govt spending would have no effect on AD is false.

    I just read Cochrane’s article and it is a pretty unbalanced attack on Keynesianism so he can hardly be surprised at the response he got.

  24. Gravatar of DM DM
    16. January 2012 at 15:33

    Professor Sumner,

    I am a law student who enjoys reading macro-economics blogs regardless of the authors’ political affiliations. Naturally, this debate has been enjoyable to follow because so many bloggers have weighed in. Before your response here, I commented on one of Professor Krugman’s post looking for guidance. I wanted to give you the same opportunity, because I think it may help the non-econ-PHD readers in the blogosphere that have followed this debate.

    Here is an excerpt of the question:

    “To facilitate, I assume the following: (1) Lucas/Cochrane argue that fiscal stimulus will not increase AD because it will reduce private spending by the amount of the fiscal stimulus. (2) [Krugman’s] camp has responded that the “cost” of paying for the fiscal stimulus will be spread over future years, allowing AD in the year the stimulus is allocated/used to increase.

    . . .

    Assume the government introduces stimulus in the amount of $100 billion. They pay for it with a tax increase over the next 10 years. In year 1, the stimulus spending is far greater than the increased taxes, which you and Professors Delong and Wren-Lewis have made clear. If we stop here, it is apparent that AD increases in that year.

    But where did the $100 billion come from in the first place? If bonds, then didn’t the private sector pay $100 billion that it might have otherwise spent elsewhere?

    In essence, my question is: Where does the $ come from? And how does it impact the aggregate AD analysis?”

    Thanks,
    DM

  25. Gravatar of Morgan Warstler Morgan Warstler
    16. January 2012 at 16:38

    math,

    ya know saying “theoretical point of view” doesn’t validate your personal preferences.

    It isn’t, “usually are savvier,” it is “are savvier” – when you use weird qualifiers like this it shows your panties.

    If you would just READ MY PLAN, you’ll find that “theoretically,” when ALL US LABOR is priced correctly against global labor – you won’t niggling little doubts about whether we engage in heavy manufacturing.

    The trick is to remove the relationship between:

    1. what we think all US citizens need to live in the US.
    2. what each worker in the US is worth.

    Remove the mental link, and you will go far.

  26. Gravatar of Nick Rowe Nick Rowe
    16. January 2012 at 16:57

    I haven’t followed who said what. But here’s the quick version of consumption-smoothing and the Keynesian multiplier:

    No consumption-smoothing: If consumption depends on current income only, the balanced budget multiplier is one, regardless of the mpc, and regardless of the duration of the increase in G. The bond-financed government expenditure multiplier exceeds one.

    Full consumption-smoothing: the government expenditure multiplier is less than one, and approaches one in the limit as the duration of the increased government expenditure approaches zero.

    Therefore, consumption smoothing almost always makes the government expenditure multiplier smaller, and never makes it bigger.

    Is that what you are saying Scott?

    As an aside, there is one interesting case however where you can get a sort of multiplier of two with consumption smoothing. That’s where you don’t *increase* G. Rather, you “prepone” G (increase G today and cut G next period). If you have full employment next period, the cut in future G allows an equal increase in future C, so current C rises to smooth.

  27. Gravatar of Nick Rowe Nick Rowe
    16. January 2012 at 17:27

    Actually though, now I’ve skimmed them, I’m not sure I read Simon Wren-Lewis and Paul Krugman as saying that consumption smoothing makes the multiplier bigger. Did either of them say that? You could read them as saying that consumption smoothing makes the effect of shortening the duration of G on the multiplier bigger. Which would be correct.

  28. Gravatar of Nick Rowe Nick Rowe
    16. January 2012 at 17:30

    All of the above is of course premised on a Keynesian model in which the Fed has a horizontal LM curve, and does not “move last”.

  29. Gravatar of I Figured Out What Scott Sumner Is Talking About « Uneasy Money I Figured Out What Scott Sumner Is Talking About « Uneasy Money
    16. January 2012 at 18:01

    […] the sort of cyberspace fireworks triggered by his series of posts (this, this, this, this, this and this) about Paul Krugman and Simon Wren-Lewis and their criticism of Bob Lucas and John Cochrane? In my […]

  30. Gravatar of ssumner ssumner
    16. January 2012 at 18:33

    math, Yes, growth might increase.

    Kevin, If dY/dT is at least as great as dY/dG, then isn’t the multiplier zero, or less? Taxes reduce output in the Keynesian model.

    Alex, Thanks, I corrected it.

    Jason, They did a cross sectional study because there’s all sorts of problems with national studies. If you aggregate you’ve suddenly got all those problems. In any case, last time I looked one sigma was regarded as no effect.

    MI, Me too.

    grooft, I certainly agree that consumption smoothing is a reasonable theory, I agree it occurs. I just don’t see what it has to do with Cochrane’s claim of no effect.

    Patrick, They may have a good argument, but I have yet to see it. Wren-Lewis certainly didn’t provide a good argument. He implied the change in total “spending” equals the change in G + C.

    Regarding Abel, Blanchard, et al, the real issue is monetary policy, not RE.

    math, You said;

    “My point is that “consumer smoothing” does have an economic effect during the financial stimulus. Thus, both Krugman and Wren-Lewis are theoretically right and Sumner is theoretically wrong here.”

    I’m not saying it has no effect, I’m saying it doesn’t have the effect Wren-Lewis claims.

    BW, I’ve never once criticized anything Krugman said about RE. That’s not the issue here, it’s the balanced budget multiplier.

    Rien, That was poorly worded, and I corrected it. I meant we shouldn’t focus on whether the conclusions are true. Obviously you’d like the reasoning to be accurate.

    Tommy, He’s implicitly assuming a fall in M*V. My only complaint there is I wish it was an explicit assumption.

    John, Yes, but notice how he simply assumes that I is unchanged. That’s really the key Keynesian assumption, but I think it’s a very doubtful assumption. Recall that even cars sales are actually investment. When you start thinking in those terms you begin to realize how little the Keynesian model has in terms of theoretical foundation. And what monetary policy is being assumed?

    I agree he may have glanced at Wren-Lewis, and doesn’t necessarily agree with the sloppy reasoning. But I wish he’d said so, instead of the insulting reply he actually posted. He wrote as if I was a first-grader.

    Noah, I strongly disagree. I’ve discussed this issue with many people, including one of the authors. And I’m pretty confident I’m right. It doesn’t show national multipliers, and all plausible models predict positive regional multipliers, Keynesian or not.

    Rob, you said;

    “I think Wren-Lewis is possibly saying that that even if you assume the balanced budget multiplier is zero then a non-Keynesian concept (consumption smoothing) still indicates that Cochrane’s claim that tax-financed govt spending would have no effect on AD is false.”

    Yes, and that’s wrong.

    I agree with your criticism of Cochrane’s article.

    DM, I haven’t agreed with your premise, I was assuming a balanced budget multiplier, and so was Wren-Lewis. If not, you need to bring in Ricardian equivalence.

    Nick, Thanks for that info, you know much more about it than me. But again, I was objecting to the reasoning on the quotation above. His example implicitly assumes I changed, otherwise it makes no sense, and yet he excludes I when considering the impact on “spending.” He just adds C and G.

  31. Gravatar of math math
    16. January 2012 at 19:26

    Scott Sumner said above : “I’m not saying it has no effect, I’m saying it doesn’t have the effect Wren-Lewis claims.”

    First, thanks for your reply.

    Second, please reread what exactly you have said about “consumption smoothing” in your previous post. Any unbiased reader can presume from your lines that it doesn’t play any role in refuting the Cochrane’s argument.

    So, perhaps both you and Krugman were not accurate in formulating their positions.

  32. Gravatar of Benjamin Cole Benjamin Cole
    16. January 2012 at 19:40

    OT. but if you like looking at property returns by sector, the following site is a lot of fun.

    It pretty much destroys the idea that it was Fannie and Freddie that caused a housing price collapse. Even timberland went negative 2009, after decades of positive returns. Offices, industrial–all went negative.

    http://www.ncreif.org/property-index-returns.aspx?region=O

  33. Gravatar of Patrick Patrick
    16. January 2012 at 22:07

    Scott

    Once again, your whole argument is based on the false notion that Krugman etc. thought that they were establishing that consumption smoothing means that the bbm can never be zero. They were doing no such thing, and it is perverse to maintain that they were.

    Again, they were simply giving an example of how government stimulus can be effective even when you make extreme assumptions about rational expectations.

    Sure, you can give an example where an increase in G is entirely offset by a fall in C and private I, but so what? Krugman etc, can and do give you an exmpale where it isn’t.

    This just goes to prove Krugman’s orginal point – using identities will get you nowhere. You need to provide an account of the actual adjustment mechanisms, which is precisely what Cochrane failed to do in his orginal piece. Instead, he appeared to argue that a rise in government spending would NECESSARILY lead to an offsetting decrease in savings available to the private sector and hence to a reduction in private spending. His supporters, realising that this would be a stupid argument, tried to explain it away by filling in the necessary assumptions. His detractors, going by what Cochrane actually wrote and not by what he might have had in his head, called it what it was – a stupid argument, and provided an example where it didn’t hold.

    I once had an international trade lecturer who would give his students a zero if they didn’t state their assumptions (usually along the lines of “assume the economy is in competitive equilibrium at full employment”). It didn’t matter if all the subsequent diagrams and reasoning were correct – it was all worth 0 if the key assumption was not stated (harsh, but it sharpened us up a lot). In this case, Cochrane hasn’t followed that kind of philosophy, and even his supporters are still guessing as to what he meant.

  34. Gravatar of John Schultz John Schultz
    16. January 2012 at 22:22

    “John, Yes, but notice how he simply assumes that I is unchanged. That’s really the key Keynesian assumption, but I think it’s a very doubtful assumption.”

    I agree that Krugman restricts that particular RE example to consumption — “Assume that consumers want to reduce consumption by the same amount every year to offset this tax burden” — to most clearly demonstrate his point (to his lay audience). However, I see little reason to assume that his argument isn’t more general and can easily be rephrased in terms of C+I rather than only C.

    What he’s really talking about is the private sector pulling back to counter balance the expected increase in taxes, so it’s pretty obvious to me, at least, he’s really talking about Y = P + G rather than the more specific Y = C + I + G in his example and to which you hold him. He just restricted his consideration to consumption in that example because it is easier for people to grasp in their heads.

    “Recall that even cars sales are actually investment.”

    I’m pretty sure Krugman knows the categorical definitions of national accounting. Your point just reinforces my point that Krugman’s actual argument is almost surely more general than for which you are giving him credit.

    “And what monetary policy is being assumed?”

    Most likely inflation targeting since that is what most central banks have been doing in the real world for a while now?

    “I agree he may have glanced at Wren-Lewis, and doesn’t necessarily agree with the sloppy reasoning. But I wish he’d said so, instead of the insulting reply he actually posted. He wrote as if I was a first-grader.”

    Yeah, there is a lot of quick reading, quick assumptions, talking off the cuff and talking past one another going around these days. The lines of communication aren’t really functioning very well, unfortunately. You’d think having all of this written down and read broadly would drastically improve the conversation, but I think some people (e.g. – Krugman) don’t have or are unwilling to spend the time to think through their “opponent’s” position and are quick to jump to uncharitable interpretations of what they read.

    I give you credit for definitely having thought through your “opponents'” positions and you did catch Wren-Lewis making a mistake (that Krugman apparently didn’t spot and made a mistake in endorsing). I’m not sure if Wren-Lewis’s mistake was semantic (i.e. – misusing the particular phrase consumption smoothing) or a deeper one (i.e. – misunderstanding consumption smoothing in the first place and wrongly applying it here).

    Two minor criticisms though: (1) I think you could have been clearer / more direct / more plain spoken about how Wren-Lewis was wrong to rely on consumption smoothing. It took a good bit of reading and thinking before I understood your counter argument and I’m not sure most of the lay people reading your blog get it at all (if they even know what consumption smoothing is in the first place). (2) I’m not sure you are giving Krugman’s (and DeLong’s) arguments on this topic an entirely fair shake (see above).

    PS – I thought Krugman’s position that RE doesn’t imply that temporary fiscal stimulus can’t increase output in the short term was broadly accepted?

  35. Gravatar of D R D R
    16. January 2012 at 22:35

    “Cochrane hasn’t followed that kind of philosophy, and even his supporters are still guessing as to what he meant.”

    BINGO.

  36. Gravatar of Morgan Warstler Morgan Warstler
    16. January 2012 at 22:59

    John Schultz inadvertently backs my position…

    “I give you credit for definitely having thought through your “opponents'” positions and you did catch Wren-Lewis making a mistake (that Krugman apparently didn’t spot and made a mistake in endorsing). I’m not sure if Wren-Lewis’s mistake was semantic (i.e. – misusing the particular phrase consumption smoothing) or a deeper one (i.e. – misunderstanding consumption smoothing in the first place and wrongly applying it here).”

    Ultimately, Scott is going to “win” this on points because he’s talented enough to stand in the ring, keep his elbows in and never deviate from an equally compelling (both are ideologically consistent) anti-Keynes narrative, and RIGHT NOW, he’s the underdog, so chinking the DeKrugman armor is VALUABLE.

    DeKrugman making smallish mistakes at his bread and butter is news.

    —-

    I’m predicting more of this, not less. Scott’s hours pay off. And ultimately, DeKrugman is not an economist. He’s a guy who wants social justice – that’s not compatible with economics.

  37. Gravatar of John Schultz John Schultz
    16. January 2012 at 23:15

    “And ultimately, DeKrugman is not an economist.”

    Sorry, but that’s just plain laughable …

  38. Gravatar of tom tom
    17. January 2012 at 01:44

    Scott I’m not sure you read any of Krugman’s textbooks.
    This is quote from him (via wikipedia):

    “Note that this is an “identity”, meaning it is true by definition. This identity only holds true because investment here is defined as including inventories. Thus, should consumers decide to save more, and spend less, the fall in demand would lead to an increase in business inventories. The change in inventories brings savings and investment into balance without any intention by business to increase investment.[2]

    Note, that as such, this does not imply that an increase in savings must lead directly to an increase in investment. Indeed, business may respond to increased inventories by decreasing both output and intended investment. Likewise, this reduction in output by business will reduce incomes, forcing an unintended reduction in savings. Even if the end result of this process is ultimately a lower level of investment, it will nonetheless remain true at any given point in time that the S=I identity holds”

    Note: “this does not imply that an increase in savings must lead directly to an increase in investment”.
    So without inventories savings are not equal to investments and decrease in savings may not lead to decrease in investment spending. Simon and Krugman is 100% correct. You and Cochrane are wrong.

  39. Gravatar of tom tom
    17. January 2012 at 01:47

    Here is direct link
    http://en.wikipedia.org/wiki/Savings_identity

  40. Gravatar of Mike Sax Mike Sax
    17. January 2012 at 02:00

    Scott when you say “I think Wren-Lewis and Krugman both made the same mistake as those angry Austrians that complained I was trying to goose up “spending”. They momentarily forgot that saving is not money down a rat hole, but rather represents spending on capital goods. They were so overconfident that anti-Keynesian arguments are bogus that they momentarily let down their guard, and employed a bogus argument themselves.”

    How do you explain Uneasy Money also not buying what you’re selling here? You can’t accuse him of any “Keynesian bias” To make the lincpin of your argument speculations that Krugman only criticized your post because he didn’t read the whole thing is a pretty shaky foundation.

    I will say more about this later-in my own blog as well-but for now it is clear that you are just as guilty of putting words in Krugman and Wren’s mouth as you claim they are.

    You keep trying to claim that they deny S=I because they disagree with your gloss that “savings is spending on capital goods.”

    Many people disagree with you here, including some who are not Keynesians-Uneasy Money and Bill Woolsey has left some commments over the last week which show he’s not so happy with your definition either.

    I think Noahpinion makes a valid point about you when he says, “Principle 4: Argument by accounting identity almost never works.”

    “Example: “But your theory is wrong, because Y = C + I + G!”

    No mattter how many times you state the obvious and claim that there’s all this disagreement among “Keynesians” -though others disagree as well- you cant prove that Lucas and Cochrane are wrong or that criticism of them is based on some speculative bias.

  41. Gravatar of tom tom
    17. January 2012 at 03:01

    Scott it is pretty simple (using your numbers):
    Income will fall by 100, consumption will fall by 20 savings will fall by 100 but investment spending will not fall because of stock liquidation (altough still S=I). This will increase total spending in the economy and this will increase output and income. Because of higher income savings will rise and still S=I (even if stocks were emptied and higher investment spending is maintained).
    I dont know why it is so difficult to understand (of course this logic is correct only if we have significant output gap like today).

  42. Gravatar of Kevin Donoghue Kevin Donoghue
    17. January 2012 at 03:06

    Scott asked me:

    “Kevin, If dY/dT is at least as great as dY/dG, then isn’t the multiplier zero, or less?”

    The balanced budget multiplier is non +ve in that case, yes. Presumably you just misread what I wrote, since I reall can’t see any other reason why you would be asking.

    Here’s what I wrote again, responding to your remark about Andy Harless:

    “Andy Harless is right and it’s not a bogus criticism of Lucas and Cochrane. A perfectly reasonable interpretation of their remarks is that dY/dT is at least as great as dY/dG in any respectable model, which is not true.”

    God only knows what Lucas and Cochrane really had in mind, but their words do suggest that they think dY/dT must be at least as great as dY/dG in any model which is not a “schlock model”.

    As DeLong, Krugman and now Wren-Lewis have now pointed out this is simply untrue. The balanced budget multiplier for a temporary increase in G is positive in any model where Ricardian equivalence holds and the G-good is not a good substitute for the C-good. Chicago economists make extensive use of such models. Are they “schlock models” when they give a result that Cochrane and Lucas don’t want? Also, consumption-smoothing is the underlying reason for that result — but you may have your own definition of consumption-smoothing for all I know; you tend to use a language which is all your own.

    AFAICT Andy Harless is simply echoing what these three have said more than once.

  43. Gravatar of tom tom
    17. January 2012 at 03:19

    of course savings will fall by 80

  44. Gravatar of tom tom
    17. January 2012 at 04:35

    Scott:
    “His example implicitly assumes I changed, otherwise it makes no sense”

    Investment spending may not change but because of inventories total I may change. So in this case S=I even if savings fall and investment spending doesnt (so in effect total spending in economy increases).
    You dont seem to understand S=I identity.

  45. Gravatar of Morgan Warstler Morgan Warstler
    17. January 2012 at 06:35

    John Schultz,

    Its ok man, I don’t think macro is real either.

    To have macro, you have to assume multiple currencies run by govts. for the purposes other than aggressively protecting the owners of money.

    And those are two really shitty assumptions. What are we cavemen?

    Note: you didn’t respond. I’m saying social justice even at the end of a mob vote, is NOT an acceptable goal of economics. There needs to a clear brutal science called economics, one that doesn’t change, have feelings, or care – just describe.

    DeKrugman should use some other term to describe what he does, like preachy-thievery or something.

  46. Gravatar of ssumner ssumner
    17. January 2012 at 14:57

    Everyone, I have way too many old comments going back to previous posts to answer them all. Let’s move the action to the most recent post, where people finally seem to be accepting that I am right. I will read all comments, but won’t respond to stuff like whether S really equals I anymore. Read any intro text if you want to know why.

    I’ll make scattered remarks, bring anything important that I missed up on the new post.

    Tom, Perhaps he is distinguishing between saving (the amount) and savings (the schedule) If so, it’s true that an increase in the savings schedule need not raise investment. But everyone agrees more saving means more investment.

    Mike Sax, Read all the comments in my new post and watch me slowing winning over my skeptics.

    Kevin, I did misread that. I believe Lucas and Cochrane were assuming constant M*V, which was not wise.

  47. Gravatar of Must Read Articles, Blog Posts, and Studies #1 « Evidence Based Investing Must Read Articles, Blog Posts, and Studies #1 « Evidence Based Investing
    18. January 2012 at 01:45

    […] (2)http://www.themoneyillusion.com/?p=12716 […]

Leave a Reply