While researching the Taylor Rule, I came across an old Paul Krugman post that responded to a Nick Rowe essay.

Interestingly, only one day after Krugman brushed me off like a mosquito, he posted the following reply to Nick:

Hey, who you callin’ neo-Wicksellian? (wonkish)

Nick Rowe argues that the neo-Wicksellian approach of much modern macroeconomics “” by which he means the tendency to define monetary policy in terms of the choice of a short-term interest rate “” leaves economists ill-equipped to think about monetary policy once interest rates have hit the zero bound.

Up to a point, that’s true. But those of us who started worrying about liquidity traps in the face of Japan’s experience were well aware of that pitfall. In fact, I wrote down my original liquidity trap model starting from a firm belief that the liquidity trap was nonsense: even if the interest rate is zero, I thought, increasing the money supply must raise demand. So I set out to write a model with all the i’s dotted and t’s crossed, so as to demonstrate that point “” and found, to my shock, that the model actually said the reverse.

What comes down to is this: once you’ve pushed the short-term interest rate down to zero, money becomes a perfect substitute for short-term debt. And any further increase in the money supply therefore displaces an equal amount of debt, with no effect on anything. Period, end of story.

Now, maybe the central bank can do other things, like buying long-term debt or risky assets, which will have an effect. But that’s because the central bank is taking some risk off the private sector’s hands. It has nothing to do with increasing M, per se.

Reading this made my jaw drop, as I knew the paper he cites reaches conclusions that are quite different from those in the post.  You may be thinking that I have already found better examples of how Krugman’s views have changed over time.  Yes, but there is something especially amusing about seeing someone cite their own research, when the paper doesn’t even support the argument being made in the post.  Here is the final portion of the 1998 paper Krugman cites (italics added):

Fiscal policy: The classic Keynesian view of the liquidity trap is, of course, that it demonstrates that under some circumstances monetary policy is impotent, and that in such cases fiscal pump-priming is the only answer. The framework here is rather different in its implications for monetary policy, but it does suggest that fiscal expansion could work. Obviously the model is subject to Ricardian equivalence, so that tax cuts would have no effect. But government purchases of goods and services in the first period, while they would be partly offset by a reduction in private consumption expenditures, could indeed increase demand and output.

While this policy could work, however, is it the right one for Japan? Japan has already engaged in extensive public works spending in an unsuccessful attempt to stimulate its economy. Much of this spending has been notoriously unproductive: bridges more or less to nowhere, airports few people use, etc.. True, since the economy is demand- rather than supply-constrained even wasteful spending is better than none. But there is a government fiscal constraint, even if Japan has probably been too ready to use it as an excuse. And anyway, is it really true that it is impossible to use the economy’s resources to produce things people actually want?

Monetary policy: It may seem strange to return to monetary policy as an option. After all, haven’t we just seen that it is ineffective? But it is important to realize that the monetary thought experiments we have performed have a special characteristic: they all involve only temporary changes in the money supply.

This point needs enlarging upon. Because the traditional IS-LM framework is a static one, it cannot make any distinction between temporary and permanent policy changes. And partly as a result, it seems to indicate that a liquidity trap is something that can last indefinitely. But the framework here, rudimentary as it is, suggests a quite different view. In the flexible-price version of the model, even when money and bonds turn out to be perfect substitutes in period 1, money is still neutral – that is, an equiproportional increase in the money supply in all periods will still raise prices in the same proportion. So what would a permanent increase in the money supply do in the case where prices are predetermined in period 1? Even if the economy is in a liquidity trap in the sense that the nominal interest rate is stuck at zero, the monetary expansion would raise the expected future price level P*, and hence reduce the real interest rate. A permanent as opposed to temporary monetary expansion would, in other words, be effective – because it would cause expectations of inflation.

Let us now bring this discussion back to earth, and to Japan in particular. Of course the Bank of Japan does not announce whether its changes in the monetary base are permanent or temporary. But we may argue that private actors view its actions as temporary, because they believe that the central bank is committed to price stability as a long-run goal. And that is why monetary policy is ineffective! Japan has been unable to get its economy moving precisely because the market regards the central bank as being responsible, and expects it to rein in the money supply if the price level starts to rise.

The way to make monetary policy effective, then, is for the central bank to credibly promise to be irresponsible – to make a persuasive case that it will permit inflation to occur, thereby producing the negative real interest rates the economy needs.

This sounds funny as well as perverse. Bear in mind, however, that the basic premise – that even a zero nominal interest rate is not enough to produce sufficient aggregate demand – is not hypothetical: it is a simple fact about Japan right now. Unless one can make a convincing case that structural reform or fiscal expansion will provide the necessary demand, the only way to expand the economy is to reduce the real interest rate; and the only way to do that is to create expectations of inflation.

Of course, it is not necessary that Japan do anything. In the quasi-static IS-LM version of the liquidity trap, it appears as if the slump could go on forever. A dynamic analysis makes it clear that it is a temporary phenomenon – in the model it only lasts one period, although the length of a “period” is unclear (it could be three years, or it could be 20). Even without any policy action, price adjustment or spontaneous structural change will eventually solve the problem. In the long run Japan will work its way out of the trap, whatever the policy response. But on the other hand, in the long run …

Oops, Mr. Krugman must have linked to the wrong paper.  This paper suggests that the old distinction between conventional and unconventional OMOs is wrong, what matters is whether the monetary injection is temporary or permanent.  A simple once and for all swap of cash for zero interest T-bills will raise the long run price level, reduce real interest rates, and hence boost AD.  What a great idea!  It doesn’t involve fiscal policy with its “bridges more or less to nowhere.”  (Did he coin this term—it was nearly 10 years before the Alaska bridge.)  In addition, monetary policy has less effect on the “government’s fiscal constraint” and is more capable of creating “expectations of inflation.”

I concede there are lots of good arguments on both sides here.  It may be hard to create inflation expectations (although I doubt it.)  But why cite a paper saying cash for zero interest T-bills can work if expected to be permanent, and might even be better than fiscal policy, in a post where you said it cannot work “Period. End of story.”  That simple Keynesian model of the liquidity trap was discredited in 1998, and it is still discredited.

The 2007 B-film entitled Grindhouse (a Rodriguez/Tarantino production) has a hilarious trailer for an imaginary film entitled “Machete:”

The highly skilled Federale Machete is hired by some unsavory types to assassinate a senator. But just as he’s about to take the shot, he notices someone aiming at him and realizes he’s been set up. He barely survives the sniper’s bullet, and is soon out for revenge on his former employers, with the reluctant assistance of his old friend Cheech Marin, who has become a priest and taken a vow of nonviolence. If you hire him to take out the bad guys, make sure the bad guys aren’t you! Written by rmlohner

To paraphrase the last line of this plot synopsis, if you are going to use research by a brilliant Nobel-Prize winning new Keynesian economist in order to attack discredited ideas, then makes sure the discredited ideas being attacked are not . . . your own.

OK, enough fun and games.  But I do have a semi-serious point here.  Over the last few months I have had a chance to closely examine many of Krugman’s recent and past writings on fiscal and monetary policy.  One thing that I notice is that Krugman is very skilled at making an argument.  He can use the same basic model to make either monetary or fiscal policy seem like the only reasonable option.  All that is required is that one tweak the assumptions in such a way that the less favored policy seems either undesirable or infeasible.  So my message is this—don’t let anyone (including me) bully you into thinking that your policy instincts are wrong, until you have closely examined the assumptions that underlie each side of the debate.  All this drivel in the blogoshere about velocity, multipliers, liquidity traps, Ricardian equivilence, budget constraints, etc, etc, are merely verbal tools that can be wielded to achieve any desired outcome.  It’s an insiders game.  The real battle is elsewhere.



19 Responses to “Machete”

  1. Gravatar of Kevin Donoghue Kevin Donoghue
    14. May 2009 at 00:21

    But why cite a paper saying cash for zero interest T-bills can work if expected to be permanent, and might even be better than fiscal policy, in a post where you said it cannot work “Period. End of story.”

    I don’t read Krugman’s blog post as doesn’t saying “it” cannot work, if by “it” you mean an action which drastically changes the reputation of the central bank. He makes no mention of permanence in the blog post you link to, nor does Nick Rowe in the post Krugman is responding to (see David Pearson’s comment on Nick Rowe’s blog). In Krugman’s Japan paper, conventional open market operations won’t do the trick. Of course if the Bank of Japan starts dropping yen from helicopters, or playing the stock market with money which it prints for the purpose, that’s going to have an impact on inflationary expectations. Krugman doesn’t deny that now any more than he did in 1998.

    That simple Keynesian model of the liquidity trap was discredited in 1998, and it is still discredited.

    Maybe it’s discredited in your eyes, but not in Krugman’s? He says he looked at it and found, to his surprise, that he needed to take it seriously. I believe him. If he wrote that paper as a joke he certainly took a lot of trouble over it.

  2. Gravatar of ssumner ssumner
    14. May 2009 at 04:18

    Kevin, I still don’t think you understand what Krugman was doing in his 1998 paper. Krugman discovered (for himself–others already knew this) that what really mattered was not whether the assets being bought had zero interest rates or not, but whether the monetary injection was temporary or permanent. Obviously temporary currency injections will have little or no impact, that was known even before Krugman’ paper.

    In basic monetary economics courses the baseline “thought experiment” that they always start with is an X% permanent increase in the money supply. That’s is the experiment that leads to the quantity theory, etc. A temporary monetary injection is a footnote. And now you seem to be saying that if Krugman dismisses the possibility of conventional OMOs doing any good (Period, end of story), we are to read that as only applying to temporary currency injections? I can tell you that when I made that point on Brad DeLong’s comment section, it was news to his readers. Krugman is a NYT writer, do you think those readers would understand that distinction?

    Your claim that Krugman is saying (in 1998) that “conventional OMOs won’t do the trick” is simply false, as you can see from the first italicized portion of the quote above. I gave you the entire last couple of pages of Krugman’s paper. Not some obscure footnote. Also note that while Krugman distorts what people mean by taking quotes out of context, I quote him fully, so that in context people can see that not only does he think cash for T-bills might work, but he thinks it’s such a promising option that he quickly dismisses fiscal policy and spends a lot of time discussing the way to make it work. That it is Japan’s best option. He was right then, I just wish he was still making this argument.

    Lot’s of people like Greg Mankiw have suggested price level targets with 3% inflation. If you have that sort of policy, then OMOs will permanently increase the money supply, or at least that portion required to achieve higher future prices will be permanent. Nothing at all exotic about Mankiw’s proposal. You can do it by buying T-bills. It will raise inflation expectations. Krugman doesn’t seem to think it will work, but he is the one who gives the weird convoluted explanation (in recent posts) of why this wouldn’t work. It has zero to do with the technical aspects of the model he cites here.

    Also, I think you overlooked the point in my post where I said that there were respectable arguments for Krugman’s current policy position. Then I said the main point was that people shouldn’t be bullied into thinking they had to defer to the policy views of a skilled macroeconomist. That is, when someone says:

    “What comes down to is this: once you’ve pushed the short-term interest rate down to zero, money becomes a perfect substitute for short-term debt. And any further increase in the money supply therefore displaces an equal amount of debt, with no effect on anything. Period, end of story.”

    One should not infer that a policy of increasing the money supply will be ineffective.

    I’m sorry, but I think most readers would infer that Krugman is saying an increase in the money supply (via T-bills) would be ineffective. You say he isn’t, that he is only referring to the special case where money is injected temporarily and then pulled out. Sorry, but I just don’t think people would read his statement in such a limited way. Especially when his later “exception” (unconventional OMOs), makes no reference to the temporary/permanent distinction.

    Right now the BOJ does not have a positive inflation target; were they to announce the 3% target that Mankiw suggests for the U.S., the policy might well work, that is Krugman’s (1998) argument in a nutshell. Why does he no longer make that argument? He somehow came to the conclusion that if the BOJ were to announce a 3% inflation target, people wouldn’t believe them. Fine. I don’t see a shred of evidence for that political/psychological argument, but don’t pretend it has anything to do with the logic of his paper.

    Regarding your second comment, his 1998 paper doesn’t validate the old Keynesian IS-LM model, which made no distinction between temporary and permanent injections. It shows that if the injections are temporary, one may get the same policy implications (ineffectiveness) as that older model.

  3. Gravatar of David Pearson David Pearson
    14. May 2009 at 05:01

    I would also argue that Krugman is referring to a temporary injection. The reason lies in the context for the debate: aren’t you and other QE/targeting proponents arguing that stimulus should be removed in the future? I’m having trouble reconciling your view that QE is not inflationary because the Fed can shrink the base later; with your view that it is “obvious” that temporary stimulus has no impact on AD.

    Perhaps I misunderstand your exit plan. Will the base shrink in the future? Stay the same? How will the Fed control rising inflation expectations without shrinking the base? I’m confused…

    I give Krugman credit. He is willing to trade off higher inflation for lower unemployment. I don’t agree with his Keynesian perspective, but there is a certain honesty in recognizing the costs of policy. Lots of QE proponents (including a majority on the FOMC), on the other hand, will only admit that there will be a temporary, virtually costless, inflation “blip” caused by QE, and that we after that we will be back on our merry way.

  4. Gravatar of Thruth Thruth
    14. May 2009 at 06:40

    A cynic might suggest that for Krugman the only difference between Japan and the US is that he likes US bridges.

    A conspiracy theorist might suggest that the Fed members have been persuaded to like US bridges too.

  5. Gravatar of Bill Woolsey Bill Woolsey
    14. May 2009 at 07:15


    Suppose the target growth path for nominal income is 3%. The trend inflation rate is 0%. The central bank is credible and people believe that the inflation rate will remain 0%.

    The shadow banking system collapses and people panic and sell off commerical paper and buy T-bills and increase deposits in FDIC insured accounts. Interest rates on T-bills fall all the way to zero. Banks allow reserve balances at the Fed to increase rather than bother to buy zero interest T-bills. Again the demand for FDIC insured deposits have increased. The overnight lending rate falls to zero, because if banks can earn interest on lending to other banks, that is better than zero interest reserves. And with banks accumulating reserve balances anyway, none of them need to borrow.

    Suppose that the increases in interest rates on commerical paper depresses spending one way or another. Other things being equal, this will cause nominal income to fall below its 3% growth path.

    It is impossible for the Fed to “lower” interest rates on T-bills or federal funds. It can purchase more T-bills and increase base money.

    suppose the Fed was willing to change its policy and say that no longer will nominal income grow at 3% with a 0% trend inflation rate, but rather it will grow at 8% with a 5% trend inflation rate, so that now the zero nominal interest rate on T-bills, federal funds, FDIC insured deposits, and reserve balances at the Fed is a -5% real interest rate. I don’t deny that this would cause nominal income to rise.

    But that is undesirable, in my view. Krugman, I think, is saying that the only way to get of the liquidty trap is to increase the money supply enough to create anticipations that nominal income will rise enough (above target) to create inflation high enough, so that when it is anticipated it will lower real interest rates enough, to raise nominal inocme to this extra high level.

    There is no way to raise the money supply a more modest amount so that nominal spending remains on target and the expected inflation rate remains zero.

    Now, if nominal income falls enough to generate enough deflation, so that returning back to trend involves inflation, then that might work. But that means nominal income and the price level must first fall. And, once we are back to target the inflation rate is again zero and the real interest rate is again zero.

    To me, the “problem” is obvious.

    If there were really one interest rate and it were at zero, this would be a problem. Because in reality, only some nominal interest rates are at zero, it is possible for the Fed to purchase other kinds of assets and their nominal interest rates will fall. (I think Krugman is wrong when he claims this is all about fixing risk and not about monetary policy. It is monetary policy, but it also does involve the Fed bearing risk.)

    Or, of course, currency payments can be suspended, the nominal interest rates on FDIC insured deposits, reserve balances at the Fed, and T-bills can become negative. The other nominal interest rates will become lower (less positive.)

    The advantage to this approach is that it doesn’t necessarily impact risk premia or the yield curve. I expect it would, but the market still determines the difference between short and long rates and between low and high risk securities.

    Of course, the disadvantage is the currency no longer trades at par. That makes currency transactions difficult.

    If all interest rates hit the zero bound (as unlikely as that is,) then suspending currency payments is the only answer. (Consistent with keeping nominal income on its 3% zero inflation rate trajectory. Oh, other than fiscal policy.)

    If we assume that the only reason why any interest rate ever hits the zero bound is because people believe nominal income is going to fall and that there will be lower real income and/or deflation in combination, then breaking that expectation will fix the problem.

    If it is possible to some interest rates to hit the zero lower bound for some other reason, and those are interest rates that the Fed manipulates, say, the yields on the “only” assets the Fed buys, then that sort of monetary policy becomes ineffective.

    With index futures convertibility, the Fed would buy up all the T-bills. And still, there would be people shorting the contract, and the Fed would have to buy something else. It would. And it would lower other nominal interest rates.

    Fiscal policy, of course, raises the interest rates on T-bills by having the Treasury sell a bunch of them to build bridges to nowhere. Banks hold the T-bills rather than base money. People hold the T-bills rather than FDIC insured deposits, reducing the induced demand for base money. The excess demand for base money is fixed. The goverment builds bridges rather than the private expenditure that would have been financed by the commercial paper that people won’t buy.

    Or we can give up on the 3% nominal growth rate and zero inflation rate target and create inflation.

    I support quantitative easing. I hope it doesn’t work by causing higher expected inflation and reducing real interest rates. I don’t think it can work by purchasing more zero interest T-bills that are already creating a “second best” demand for base money. The demand for base money is going to rise to match the increase in the supply. It can work by purchasing other assets. But that simultaneously shrinks the yield curve and risk premia (one or both, depending on what the Fed buys.) In my view, that is an undesirable market intervention. But it is better than adding to the national debt.

    I will grant that currently, expectations of depressed nominal income is depressing all nominal interest rates, and so, fixing that problem would raise the needed real and nominal interest rate. But, that isn’t the only possible problem.

  6. Gravatar of JTapp JTapp
    14. May 2009 at 12:52

    This is a bit off-topic. Dr. Sumner, what textbook do you use, or would recommend using, for undergraduate macro? If you don’t want to use IS-LM, are there any texts out there you would approve of teaching from?
    Is there a Chicago-view textbook somewhere?

    I would rather not use a Krugman text since Mankiw often posts examples of how Krugman contradicts his textbook vis-a-vis his columns(there, I found a way to make this almost relevant).

  7. Gravatar of ssumner ssumner
    14. May 2009 at 16:37

    David, I understand why you have this confusion; people mix up two very different and seemingly contradictory issues:

    1. Temporary currency injections have little or no impact on prices.

    2. If the current bloated base stayed in circulation forever, eventually we’d return to double digit inflation.

    Both of these are true, but they don’t conflict on close examination. Let’s use some simple examples to illustrate the point. I’ll make a long run quantity theory assumption, but nothing of substance would change if I relaxed the QT assumption. (Obviously the QT doesn’t hold while in the liquidity trap.)

    Start with a price level target: 100, 103, 106, 109, …

    Now assume you are in a liquidity trap with zero rates, real money demand doubles. If the increase is viewed as being temporary, there will be no increase in the price level, just as Krugman says. If it is permanent, then prices will double when you exit the liquidity trap, and prices would even rise a lot right now in anticipation of that high inflation. So how do you get the goldilocks economy? You promise to always adjust the monetary base in a way such that prices rise according to the target. For instance, suppose the liquidity trap is expected to exist in year two (when the target price level is 103), but not in year three, when money demand goes back to normal. In that case (abstracting from trend changes in money demand) then the public would expect the base to go 100, 206, 106, 109, 112.

    So if you ask whether the money supply increase is permanent, there are two answers:

    1. Most of it is not, or you are in big trouble.
    2. 6% if it is expected to be permanent (i.e. over two years). Otherwise you won’t get any inflation at all right now. If the price level is expected to return to 100 in year three, then lots of luck getting out of your liquidity trap. If the price level is expected to be at 106 in year three, then you are in good shape.

    Regarding your temporary inflation “blip” I just don’t see it. I see a temporary deflation blip in late 2008, but no inflation blip at all. When with this “blip” occur? And why? The Fed still hasn’t adopted an expansionary monetary policy, and until they do I think it is premature to talk about inflationary blips. Yes the base has gone up a lot, but that means nothing as long as they are paying banks to hoard reserves.

    Krugman knows full well that people like Mankiw are proposing a 100, 103, 106, 109 price trajectory. Krugman argues it won’t work because the public won’t believe the central bank’s promise to adhere to that path. Why, I don’t know. I don’t recall previous examples of central banks promising to inflate and then going back on their word, do you? The BOJ never promised to inflate–they promised not to inflate, and didn’t.

    Thruth, Bernanke’s a Republican. I really have trouble believing he favors fiscal stimulus because he likes big government. But I do think that whether he likes it or not, the Fed policies are delivering big government.

    Bill, there is a lot there, so let’s take this in stages, and then come back for a second round.

    First, I think my answer to David might have at least a small bearing on your question.

    Second, I think there are at least 3 very different issues embedded in your question.

    1. Can we hit a 3% NGDP target when in a liquidity trap?
    2. Can we do it if we are limited to buying only T-bills?
    3. If we are in a liquidity trap, is a 3% NGDP target a bad idea? Should it be raised?

    As far as question one is concerned, I think the answer is “yes” if you have NGDP futures targeting. However, there may not be enough T-bills in the world to make this happen, so central banks might have to buy something else.

    As an aside, I have never understood the fixation of people like Krugman on T-bills. At times he seems to suggest that cash for T-bills is the only real monetary policy. Everything else is some kind of weird, unconventional hybrid of monetary and fiscal policy that is very risky. First of all, as Jon says, the Fed normally buys lots of T-notes. But more fundamentally, if someone wants to argue that only T-bill OMOs are real monetary policy, and that the Fed may not be able to target inflation with this “real” monetary policy, my answer would be “duh.” It is exactly like saying the Fed might not be able to hit a 3% inflation target under the gold standard—duh. So I want to completely get away from the T-bill question. There is no doubt that in a liquidity trap the demand for cash at a stable price level might exceed the entire stock of T-bills. Indeed the stock of T-bills could be very low, if we are just talking about abstract theory. So here is where I would start. The Fed can increase the money supply as much as it wants, by buying all sorts of assets.

    Now let’s say the demand for base money is so great that to hit a 3% NGDP target, the Fed must buy an undesirable amount of assets. How is undesirable defined? I don’t know–maybe there is a risk they would be resold in the future at a loss. Well then the Fed must decide what it wants to do—stick to the 3% and take the risk, or go to a higher NGDP target. I don’t know what more to say.

    Now let me go to another part of your question, do OMOs have an effect if you are in the liquidity trap? In my view it is wrong to focus on what you buy, the issue is what expectations the the public has about future monetary policy. If you can create expectations of 3% NGDP growth it is much easier (in the sense of less OMOs required) to hit your target.

    Regarding your agreement with Krugman that no amount of OMOs might be effective unless you raised the target, I don’t quite agree. What I would say is if the policy was not credible, then you’ve got big problems. You can still hit the 3% without going over, but you might have to buy the entire national debt, and then start buying other assets. And I think Krugman would entirely agree. My only difference is that I don’t think that scenario would be likely to occur. (BTW, I don’t think that means an inflationary time bomb, as long as you watch futures markets and unwind quickly of necessary.)

    Now suppose that your 100, 103, 106, 109 NGDP target path was credible. By that I mean that it was believed that if after 2 years you were only at 101 (and the target was 106) then you would shoot for 8% NGDP growth (to catch up to 109.) Just the expectation that you would adhere to this target path in the long run, would get you out of the expectations trap (not necessarily the liquidity trap, but the expectations trap.) Nominal rates might stay at zero, but NGDP would rise more than 1% in the first two years. Just the threat of 8% NGDP growth between year 2 and 3 would make for close to 3% annual NGDP growth in the first two years. That is the basic point of his 1998 paper. And that works even if all you buy is T-bills.

    I have never focused on QE. I agree that swapping cash for T-bills right now does virtually nothing if the T-bill injections are viewed as temporary. Like Krugman I have always believed that you need a credible inflation trajectory. I believe the Fed can credibly promise to inflate, Krugman does not. Krugman believes the public will not believe what the Fed says, or maybe he believes the Fed is too conservative to make those promises, I’m not quite sure. But for whatever reason his is skeptical about monetary policy despite the fact that his 1998 paper clearly implies monetary stimulus has great advantages over fiscal stimulus. Of one thing I am sure, his skepticism has nothing to do with the notion that swapping cash for T-bills is always ineffective–it is very effective if permanent.

    JTapp, I have never taught intermediate macro. I use Mishkin for money. I use Mankiw for principles. I think the fact that Krugman contradicts his book in his columns should be viewed as a point in the book’s favor (if you catch my drift.) I would probably just pick a regular IS-LM text, and skip over that lightly. Focus much more on monetarist transmission mechanisms, etc. If you can’t find a book that covers the non-interest rate transmission mechanisms, and you are not afraid to do something unconventional, then use Mishkin’s text, but you’d have to have something else for growth theory and fiscal policy etc.
    Focus on long run growth, business cycle theory is pretty worthless now, in my view. If we were teaching useful stuff, we wouldn’t be in this mess.

    I really don’t know if there is a good non-Keynesian undergrad book out there. Barro had one, but it was pretty high level.

  8. Gravatar of Bill Woolsey Bill Woolsey
    15. May 2009 at 02:41


    Krguman (and you) generally describe the expectations as being in terms of inflation. Higher expected inflation lowers real inerest rate (to a more negative level) and this motivates more spending.

    Now, if I expect 8% inflation between years 2 and 3, I don’t want to be stuck holding a bunch of zero nominal interest reserve balacnes at the Fed, FDIC insured deposits or T-bills. I will take an 8% real loss on them. So, I spend now. This increase in demand raises inflation now.

    I don’t see how this works if it nominal income (and so, real output) that is changing. I am holding short, low risk assets because I don’t want to take a loss on uncertain projects. Unfortunately, that results in lower nominal income and real output. I dismiss that effect as temporary, beleiving that, somehow, it will be reversed. Why do I stop holding the low risk, short term assets? Because after nominal income falls, having nominal income go back to trend implies it grows rapidly? I don’t see why this causes be to consume more or purchase capital goods. I was buying the low risk, short securities when nominal income was on target.

    I would like nominal income to stay on target always with a growth path consistent with zero inflation. Of course, it is likely to really deviate away from target. Having these deviations dampened and self-reversing is great. So, a process by which nominal income (and real income) drop a lot, and can be reversed is better than nothing. But I would hate for there to be a class of shocks, that even if anticipated, can only be corrected by nominal income first falling below target.

    I think you want to say that if nominal income falls, then it will be expected to rise, which will make it rise. So, falling nominal income causes rising nominal income. The solution to that contraction is that nominal income can’t fall in the first place. (The decreases are greatly damped anyway.)

    As I said in my first comment, if the reason for nominal income falling (and the extra low nominal and real interest raes) is expectations of nominal income falling, then credible nominal income targeting should fix that up well. I still am a bit concerned that there be an actual process that will make nominal income rise despite pessimistic attitudes… but we aggree that the Fed just has to buy more stuff.

  9. Gravatar of ssumner ssumner
    15. May 2009 at 06:14

    Bill, I see your point, and I can’t completely disprove it, but let me throw out two pragmatic observations:

    1. I think inflation and NGDP growth have effects that are more similar (on velocity or money demand) than people think. We all know about the Fisher effect, so the part of NGDP growth that is inflation produces similar effects on nominal rates. But in addition, real GDP growth correlates with the real rate of interest. When RGDP is high, there are more profitable investment projects, and thus the real interest rate tends to be higher. Of course this is less precise than the Fisher effect, but I think as a practical matter 8% NGDP growth would raise velocity and reduce base demand in a way similar to 8% inflation.

    2. I also think it is very likely that we would never face a situation where the public expected 1% NGDP growth over two years, and then 8% NGDP growth the third year. I would point to the current situation here. I think the big problem last October wasn’t that people became pessimistic about the next three months, I think it was because they suddenly became pessimistic about NGDP growth over the next three years! Or even longer. If the fear had merely been a bad quarter, then stock and commodity prices would not have collapsed.

    To me, it is like driving a car. I was taught not to look 50 feet ahead, but to steer for a point several hundred feet down the highway. I was told the car would actually swerve less than if I steered 50 feet ahead. Obviously this analogy doesn’t prove anything, but I think if people have confidence that NGDP will be 9% higher three years from how, that confidence will help support AD in years one and two. I don’t know enough about dynamic models to prove the point, but my intuition comes from observing how devastating it is when long term expectations suddenly get much more bearish (late 1929, late 1937, late 2008.) But short term problems (Y2K, Katrina, LTCM, etc), don’t seem to have any significant effects.

  10. Gravatar of RebelEconomist RebelEconomist
    18. May 2009 at 05:54


    The link to Krugman’s 1998 paper was useful. While I was sceptical that Japan’s problems had much in common with the present US situation, I think that Krugman’s paper provides the answer, though he may have been too committed to his own solution to see it.

    In explaining why zero interest money is preferable to income-generating assets like real investments, stocks and overseas assets, Krugman argues that expectations of capital losses reduce their return over the period of interest. It seems to me therefore that the problem is that asset prices (and in Japan’s case the yen too) are too expensive. This does seem to be a common feature of both the Japanese and the present US busts. If asset prices were allowed to fall, there will not be an excess demand for money for inter-temporal substitution purposes.

  11. Gravatar of Charlie Charlie
    18. May 2009 at 11:09

    Dear Scott,

    In the second comment, you asked “Why does he no longer make that argument?” I don’t know if you were being rhetorical or actually wanted to know the answer, but I remembered that Krugman had answered this when Mankiw blogged about it. From April 19th here is the post. You may reject his argument for changing his view, but he has been open about it and discussed it in the past.

    “Greg Mankiw says yes. Since that was the answer I arrived at for Japan more than a decade ago, I have to say that it makes sense in principle.

    But here’s why it won’t work now, at least not yet: we’re talking about making a credible commitment to fairly high inflation over the medium term, yet you still have distinguished central bankers appalled at the Fed’s 2 percent inflation target.

    And again, the inflation commitment has to be credible. So I don’t think we’re ready for this, not yet.”

    Krugman may not have crossed all his t’s and dotted all his i’s in that post, but I will also note that I read Krugman’s blog because it is short and pithy (and usually don’t read this blog). There are costs to length as well as brevity. Long-time readers, like me, generously read posts by Mankiw, Krugman, and Cowen against the backdrop of their other writings and posts and do not require each post to stand on its own.

    Maybe it is not right or fair or good that reading some blogs accurately (if accurracy is discovering what the author means) requires a body of knowledge, but that is. Relevant body of knowledge reading this post requires knowing things like:

    Krugman divides monetary policy into conventional (changing fed funds rate) and everything else.

    Krugman is making political assumptions about what the Fed can feasibly do.

    Right or wrong, pithy blogs have to be read as a whole or they cannot be pithy.


  12. Gravatar of ssumner ssumner
    18. May 2009 at 14:07

    Rebeleconomist, I am afraid that you may have confused cause and effect. Asset prices aren’t high, they are low. They need to be high. Why are they low? They are low because we have a very contractionary monetary policy. If we had expectations of strong economic growth, stock prices would be much higher AND the expected return on stocks would also be higher. A win-win situation. Anyway, asset prices are set in markets. The Fed can only influence asset prices by affecting the economy. If you want lower stock prices then the Fed needs to screw up the economy with tight money. Last time the did that? 1929. And it worked. I hope this doesn’t sound too sarcastic–your premise about low expected returns is right, and the error that (I think) you made is very common, indeed most people probably look at things the way you do. But I feel strongly that that is not the right way to think about the problem. With an appropriate monetary policy asset prices would be much higher.

    Charlie, Although I do understand your point, I disagree for a number of reasons:

    1. First, it doesn’t respond at all to my paragraph that began “I concede”. I did understand his argument. My point was to ask why link to a paper that contradicts your argument, as a way of making that argument. Yes he’s changed his mind since 1998. I knew that. But then why link to a 1998 paper that makes the opposite argument.

    2. By conceding his argument could be defended, and by saying it was all “fun and games” I signaled to readers that it was just a light-hearted dig. In contrast, consider how Krugman and DeLong treated John Cochrane for his implicit assumption of constant velocity in the face of fiscal expansion. As I recall they treated him like a moron. I’ve never suggested Krugman was a moron (maybe just a bit inconsistent and disingenuous at times.) Even worse, Cochrane later relaxed the constant velocity assumption in the same paper. In Krugman’s case you had to go to other posts. So Krugman’s error was no different from Cochrane’s, and yet I treated him with kid gloves by comparison. And by the way, I consider Krugman’s assumption that the Fed couldn’t credibly inflate to be more far-fetched than velocity being unaffected by fiscal stimulus. If his assumption is that they could credibly inflate but won’t, then see point 4.

    3. I know you understand what Krugman was getting at, but I believe 99% of his readers do not. I’ll go even further, I believe at least 80% of those readers who are professional economists don’t understand that Krugman was not saying monetary policy becomes ineffective at zero rates.

    4. Krugman forcefully criticizes Congress for inadequate fiscal stimulus, but rarely criticizes the Fed for not doing enough. Why? The Volcker anecdote is irrelevant–he’s no longer on the FOMC. The guy now in charge of the Fed has published articles with very similar views to Krugman about the need for credible inflation expectations. And very similar views on Japan. Why be so defeatist and assume the Fed won’t adopt an inflation target? Why not recommend such a target? Why isn’t Krugman doing what I am doing. He is making my job harder because people who read him come criticize my argument with mistaken impressions of what he is saying.

    5. I’m afraid you’re defense of Krugman just made things much worse. If I had read that before my “What happened to the Keynesian SRAS” post last week, I would have been much tougher on Krugman. We do not need high inflation expectations today. If we had 2% inflation expectations it would be a big help. Given how flat the SRAS is in deep recessions, 2% inflation would probably imply at least 5% NGDP growth, perhaps even more. I gave Krugman a break in that column, and assumed he wasn’t making the same error in interpretation as the Fed researchers, looks like I was being too generous.

    6. His style of writing is sort of asking for trouble. If you know that monetary policy can work at the zero bound, and then develop a convoluted theory of how for some reason markets wouldn’t find a central bank’s promise to inflate to be credible, even though a normal, garden variety inflation targeting regime would still work fine, and then say in a post that monetary policy cannot work “Period. End of story,” it seems to me you are trying to be provocative. That’s fine, but then don’t be surprised when other people (like me) are provoked. I also try to be provocative, as I hope my views will get more attention. I want to get the Fed to be more expansionary, so we don’t have to rely on $1.8 trillion dollar deficits, which are not in the country’s long term interest.

    Despite these reservations you make a good point. I hope you will take a look at my SRAS post, it is from just a few days back.

  13. Gravatar of RebelEconomist RebelEconomist
    19. May 2009 at 00:09


    Well, because the Fed has neither followed my advice or yours, asset prices have fallen a bit, so we probably need to backtrack to your supposed policy mistake last autumn. I do not think that many people would say that asset prices (houses, stocks, MBS etc) were too low then, assuming that inflation was going to remain low. So I suppose that what you mean is that asset prices even then would not have been unreasonable if the economy was expected to continue expanding one way or another (ie inflation if not real growth). If that is what you are saying, then maybe you are right; the present problem is that asset prices should not have been allowed to fall. But I would say that solving that problem in the way you suggest would simply create a bigger problem, say ten years down the track. It would reinforce the “stocks for the long run”, “safe as houses”, etc culture that is surely familiar in the US as it has been here in the UK. In short, the US is in a moral hazard trap. I do not pretend to be able to offer my own way out yet, although I suspect that it would involve beginning with Andrew Mellon’s advice and ending with Keynesian policies.

  14. Gravatar of RebelEconomist RebelEconomist
    19. May 2009 at 00:44

    By the way, I am not talking about the Fed deliberately engineering a crash by “screwing up the economy with tight money”. I am talking about simply allowing asset prices to fall, supplying money as it is demanded at some low interest rate, but certainly no TARP, TALF etc etc. Plus rigorous accounting practice that makes the process transparent.

  15. Gravatar of ssumner ssumner
    19. May 2009 at 18:14

    Rebeleconomist, I agree on one point—if we start with Andrew Mellon’s conservative program, we will end up with left wing Keynesian policy. I don’t know your politics, but many of my readers are right wing. I get frustrated when right wingers recommend deflationary policies that put left wingers into office. It has happened many times, and it is self-defeating. We need a progressive right wing macroeconomics, not a right wing macro that only seems concerned about everything being too expensive, and seems oblivious to unemployment. I want to produce a stable macro environment—I hope that makes my stocks worth a lot more. It won’t store up trouble for 10 years down the road if we continue stable policies indefinitely.

    I also oppose TARP, TALF etc. I want NGDP targeting with ordinary OMOs.

    There is no such thing as “supplying money as it is needed at a low interest rate.” Such a policy leaves the price level indeterminate, resulting in hyperinflation or hyperdeflation. You need a nominal anchor.

  16. Gravatar of RebelEconomist RebelEconomist
    20. May 2009 at 09:11


    My view is that economists should simply try to identify the best policies, regardless of politics. I would, for example, do nothing to stop banks from failing, and then nationalise them, albeit temporarily – does that make me right wing or left wing? I suspect, however, that a truly stable macro environment would be one in which you are more interested in the income your stocks generate than in their value.

  17. Gravatar of ssumner ssumner
    21. May 2009 at 03:46

    Rebeleconomist, I agree regarding bank failures. I’m not really sure whether stock market behavior would change all that much, but perhaps at the margin.

  18. Gravatar of Matt Yglesias on environmental impact regulations, Scott Sumner | EconLog | Library of Economics and Liberty Matt Yglesias on environmental impact regulations, Scott Sumner | EconLog | Library of Economics and Liberty
    15. June 2015 at 12:17

    […] that bias toward inaction can be very environmentally destructive. In an earlier post entitled "Machete" I quoted from a trailer for the film of the same […]

  19. Gravatar of Matt Yglesias on environmental impact regulations, by Scott Sumner | Citizens News Matt Yglesias on environmental impact regulations, by Scott Sumner | Citizens News
    15. June 2015 at 21:50

    […] an earlier post entitled “Machete” I quoted from a trailer for the film of the same […]

Leave a Reply