Paul Krugman Synthesizes a Little More MMT

Paul Krugman, in “The Simple Analytics of Invisible Bond Vigilantes (Wonkish)”, has lent his support to the view, emphasized by modern monetary theorists, that bond vigilantes cannot impede a sovereign currency issuer. Krugman has made this important observation in the past, but this time he spells out his reasoning, step by step, and the key points align closely with that of modern monetary theory (MMT). To illustrate his reasoning, Krugman links to a short paper, which includes a simple model he put together while in the air, perhaps somewhere over water between lands Neoclassical Synthesis and Modern Monetary Theory. The model is of the IS-LM variety, so some may prefer to look away while reading. Nevertheless, the potential for an advance in public understanding seems significant. Some people may be more inclined to adopt this better understanding of currency sovereignty thanks to Krugman lending his weight to it.

These are the points Krugman identifies as critical in the case of “the United States (or for that matter the UK)”:

Greece didn’t have its own currency, and therefore didn’t have its own monetary policy or its own exchange rate. We do. …

… an attack by bond vigilantes has very different effects on a country with a fixed exchange rate (or a shared currency) versus a country with a floating exchange rate.

In short, “we have our own currency and a floating exchange rate.”

By way of explanation, Krugman writes:

Think about it this way: with the Fed setting interest rates, any loss of confidence in US bonds would cause not a rise in rates but a fall in the dollar – and a fall in the dollar would be a good thing, helping make US industry more competitive.

He then adds an important caveat that is very familiar to modern monetary theorists:

Things will be different if you have large debt denominated in foreign currency – but we don’t.

Krugman concludes:

So what are the fiscal fear types thinking? Basically, they aren’t. But to the extent that they do have a model … they’re imagining that American macroeconomics are just like those of a country on a fixed exchange rate with no independent monetary policy.

All this is both interesting and heartening. Clearly, Krugman has based his argument on the following observable facts:

1. The US government is a currency issuer, not currency user.

2. The US government allows the exchange rate to float.

3. The rate of interest under a flexible exchange-rate sovereign currency regime is set as a matter of policy rather than being market determined.

4. The US government does not have large debt denominated in foreign currency.

These four points are precisely the points emphasized by modern monetary theorists, 1 and 2 taken together being the meaning of currency sovereignty and the characteristics of what modern monetary theory (MMT) describes as a “modern monetary system”.

From a heterodox perspective, of course, the IS-LM is not an ideal choice of model, it having been disowned long ago by its creator (Hicks), so Krugman’s working will not win the hearts of many Post Keynesians and modern monetary theorists. Still, the IS-LM model remains a stalwart in the mainstream, and we are not the ones in need of convincing. Perhaps Krugman’s argument will have better persuasive power with insiders, or at least with the many in the general public who understandably, if unfortunately, place their faith in the authority of mainstream thinkers in prominent positions rather than attempt to reason for themselves.

In any case, Krugman’s view strikingly resembles the MMT position on what is a fundamental point: markets cannot dictate the terms on which a currency sovereign issues its own liabilities. (For more on this point, see here and here.) This consistency with the MMT position follows recent research coming out of the BIS that seems to be moving the mainstream view closer to Post Keynesian and MMT understandings of monetary policy. For example, previously I have linked to a 2009 paper, Unconventional monetary policies: an appraisal, which rejects money-multiplier reasoning and corrects common misunderstandings of quantitative easing. (For further discussion of recent shifts in orthodox monetary research, see a couple of old posts, here and here.)

Despite differences in analytical framework, hopefully Krugman’s post on the “invisible bond vigilantes” will help to move the policy discourse in the right direction.


43 thoughts on “Paul Krugman Synthesizes a Little More MMT

  1. “4. The U.S. government does not have large debt denominated in foreign currency.”

    5. The US dollar is not convertible, only exchangeable.

  2. True, but PK didn’t explicitly state 5. I double checked to make sure he hadn’t been “more MMT than heteconomist”.

    Still, he is very close to the MMT position on the bond vigilantes, wouldn’t you agree?

    On a related point, I just noticed this old post by Edward Harrison:

    Krugman, Tabarrok, Cowen and the bond vigilante fallacy

    Interest rate as policy variable seems to be the additional step forward in Krugman’s presentation this time round, which makes transparent that the exchange rate rather than the interest rate is, as Harrison puts it, the “relief valve”.

  3. “which makes transparent that the exchange rate rather than the interest rate is, as Harrison puts it, the “relief valve”.”

    He now just needs to move onto the next stage and realise that the external sector isn’t some Old Testament God able to wreak wrath with no consequences to itself.

    Instead it is full of ‘export led’ economies desperate to flog their wares and a set of central banks supporting that policy by ensuring there is always sufficient ‘liquidity’ in their own currency.

  4. Think about it this way: with the Fed setting interest rates, any loss of confidence in U.S. bonds would cause not a rise in rates but a fall in the dollar – and a fall in the dollar would be a good thing, helping make US industry more competitive.

    I am hard-pressed to think of any conclusion Krugman draws that I don’t agree with. I get Point A, and I get Point B. It’s just how he gets from Point A to Point B that leaves me scratching my head (not an unusual state of affairs). Especially IS-LM. And don’t even get me started on ‘Sam and Janet’ stories. So let’s start here:

    I’ve probably read that post a dozen times over the months. And I also know it is rejected by PKE and MMT, but that’s not my point. I just don’t understand it! Without trying to be stupid about it, here is what I think I know:

    IS = Loanable funds
    LM = Fed controls money supply through quantity

    No, really, THAT is what I have come up with, as absurd as it is. Please hold your applause for later. And while I would reject both of those theories (not the point), I just don’t know how they are related — to like — ANYTHING. Much less how it would work in practice.

    Back to the point, in the event I had one. So Krugman admits that the Fed controls short-term interest rates, but the ISLM theory says the Fed controls the quantity (my analysis). Surely he knows that the Fed can’t control both?

    My question? I don’t KNOW.

    And don’t even get me started on ‘Sam and Janet’ stories.

  5. Good comment, Trixie. I’ll attempt a brief rundown.

    The IS is meant to show all combinations of income, Y, and interest rate, i, consistent with goods-market equilibrium in which planned investment, I, equals planned saving, S. The LM is meant to show (Y, i) combinations consistent with money-market equilibrium in which “money supply” equals “money demand”. The BP, when the external sector is included, is meant to show (Y, i) combinations consistent with balance-of-payments equilibrium in which the current account deficit equals the capital account surplus.

    IS is usually drawn downward sloping (lower i associated with higher equilibrium Y) because it is assumed that investment is inversely related to i. So if i falls, investment is supposed to increase, which through the multiplier expands Y, with saving rising in accordance with the marginal propensity to save, maintaining I = S but at a higher level of Y.

    LM, under the assumption of exogenous money, is drawn upward sloping (higher equilibrium Y associated with higher i) because it is supposed that the central bank can fix the “money supply”, implying that if Y increases, and therefore people desire to hold more money for transactions purposes, i will rise, inducing people to hold less money for speculative purposes, keeping overall money demand (transactions demand plus speculative demand) constant and equal to the fixed money supply.

    In Krugman’s short paper, he is also taking into account the external sector. This introduces another term in the IS schedule, involving the exchange rate, e, which is taken to cause exogenous changes in Y, with currency depreciation considered expansionary.

    It also introduces the BP. This is constructed by allowing higher Y to induce higher imports, M, while considering exports, X, unchanged due to world income and the exchange rate being taken as given. So higher Y under these assumptions causes a larger CAD. If capital mobility is imperfect, then, for a given exchange rate, i must rise to induce sufficient capital inflow to maintain KAS = CAD. So, with imperfect capital mobility, higher Y will be associated with higher i on the BP curve, meaning the BP will be upward sloping. If, instead, there is perfect capital mobility, the BP will be horizontal.

    Okay, that was a brief rundown, withholding judgment on the virtues of the model.

    Without getting into Hicks’ reasons for eventually rejecting his own model, a few observations.

    1. Krugman is discussing interest-targeting monetary policy, which implies endogenous money. So, if he had drawn an LM, it would have been horizontal. In this case, exogenous changes in i, conducted by the central bank, will simply shift the horizontal LM up and down, with the central bank accommodating whatever money demand eventuates at its chosen interest target.

    2. The underpinnings of the IS are dubious, to say the least. The inverse relation between i and investment, and therefore i and Y is unfounded in light of the capital debates. If an IS must be drawn, it would probably be either vertical (no systematic impact of variations in i on equilibrium Y ) or strangely shaped: a more or less vertical but squiggly line, with (Y, i) equilibrium combinations that exhibit no discernible pattern.

    3. With a floating exchange rate, if higher Y and M widen the CAD, the exchange rate can adjust to whatever level is consistent with the corresponding capital inflow at a given i. This is one of Krugman’s observations in his short paper. It is also Edward Harrison’s point when he writes that the exchange rate rather than the interest rate is the “relief valve” in a “modern monetary system”.

  6. Doesn’t the model assume that there is only an overnight rate, that the cb is paying interest on reserves and there are no longer bonds, and that is how it ‘works’.

  7. I’m not sure which version of the model you have in mind, Neil. Any number of assumptions can be made about the different components. My brief outline was of the way the model has usually been introduced to students, which traditionally assumed exogenous money. But endogenous money and interest targeting would be represented by a horizontal LM, insensitivity of investment to interest rates by a vertical IS, and so on.

    Often, but not always, the interest rate considered in models is the general rate of interest (i.e. an average over all rates).

  8. I tend to find a problem on fact 3 (central banks exogenously sets the interest rate). The problem is that government securities mostly consist of medium/long-term notes rather than short-term T-Bills.

    The central bank short-term interest rate (which usually involves the overnight interbank rate) can only be projected and arbitraged for a shot time span, which corresponds to 3/6/12-month T-Bills lifetime. On the other hand, the central bank also uses long-term notes as a monetary policy instrument (they are the main securities used in repos) but cannot dictate their specific yield unless it desires to set long-term rates (which it usually does not).

    As a result, long-term rates are a bit endogenous. Large capital inflows in the government securities market by dollar-pegging central banks (BoJ, PBoC) can move bond yields and relevant instruments such as interest rate swaps (see for example These movements impact on business/consumer decisions (since they change their long-term lending rates) and on the exchange rate. The domestic central bank can certainly counter these effects but that would make its policy rate an ‘endogenous’ variable with inflows in the bond market acting as ‘explanatory’ variables.

    The above could also just suggest that government bonds are not the perfect instrument for central bank policy but other alternatives should be considered such as (for example) short-term central bank bills as a reverse repo instrument for foreign central banks FX holdings.

  9. Once again, you tricked me. Go ahead and read your rundown with British Petroleum or Bernadette Peters as “BP”.

    Now I will think. But this only reaffirms my position:


    Hope you’re happy.

  10. Thanks, Ryan! Hopefully others like it as well.

    Surely, at the very least, “they” will appreciate the “new” capability (new at heteconomist, that is) to edit their posts? But “they” are hard taskmasters, so “we” will see.

    In any case, considering we are almost at the end of 2012, I thought perhaps it was time to give wordpress twenty eleven a burl. 🙂

    PS: I have been following your new blog with much interest. Google translate is getting quite a work out.

  11. O-O, I have to be careful what I am writing now. 🙂
    Luckily Pavlina Tcherneva, whose first language is Bulgarian, will keep an eye on my Bulgarian MMT blog from time to time.

    You don’t know how much my poor fellow Bulgarians need MMT macroeconomics. It is just criminal there…

  12. New site design looks fantastic, Peter!

    OH! And spell check! And pop-out comment box!

    Very fancy! So much so, I’m wondering if you’ve been hacked. 😉

    Great job!

    (And edit feature!)

  13. And a brief bio too! Peter, are you feeling okay?

    Also, I didn’t know you had a PhD! Does this mean I have to stop heckling you now? And swearing?

    Because holy shit!

  14. Nice post, Magpie. Crystal clear and to the point. Thanks for the link.

    Trixie, I would be devastated if your behavior improved in any way!

    Glad you both approve of the altered look. 🙂

  15. Trixie, on reflection, I am not sure I entirely answered your question from earlier in this thread. I will put up a new post that hopefully connects the dots a bit better between A and B.

  16. It seems to me that for this vigilante theory to hold any water, these vigilantes would have to be a bunch of super ninjas with this un parallelled ability to know when exactly to attack a currency and precipitate the type of interest rate moves that they can profit magnificently on. Why havent they attacked the US in the last 4+ years if they are capable of such an attack? What are they waiting for. They attacked Europe early. Obviously they didnt wait to make a fortune in Euros. Why wait on dollars? Is there more to be made by waiting longer? They are just going to have to pay more taxes soon!!!! Are they not interested in maximizing return today? Right now?

    If we dont see them do their vigilante *thing* (whatever it is they do) by the time a new tax plan is enacted, cutting their potential take home gains, Im gonna assume they are just a bunch of sissies, not worthy of being called vigilantes, or capitalists. Arent capitalists supposed to capitalize ?


  17. Looks like the scales are finally falling from eyes.

    I suppose it’s always been obvious to me, since my work is all about replacing people with automated processes operated by ‘smart’ machines.

    I remember in my second job interview in 1993 naively waxing on about how improved technology can increase productivity by reducing manpower required from five to one. The interviewer listened politely and then asked ‘and what are those four people now going to do for a living instead?’.

    Every time I see another bank of self-service tills go in at the supermarket, that same question comes back to me.

  18. I myself have been interested in that subject (and I’ve written about it in my own blog and commented on it here). For example:

    Inequality and Technology.

    But that’s not what I found extraordinary in Krugman’s piece. This is:

    “I think our eyes have been averted from the capital/labor dimension of inequality, for several reasons. It didn’t seem crucial back in the 1990s, and not enough people (me included!) have looked up to notice that things have changed. It has echoes of old-fashioned Marxism — which shouldn’t be a reason to ignore facts, but too often is. And it has really uncomfortable implications.

    “But I think we’d better start paying attention to those implications.”

  19. So, Magpie, who has the best chance of recruiting PK, the Marxists, the Kaleckians or the Sraffians?

    And where does this leave MMTers?


  20. @Pete

    Who knows! Or, as the old song went: Que será, será. Whatever will be, will be, The future is not ours to see. Que será, será.

    Thank you. Thank you.

    Thanks for the applause. You’ve been a wonderful public. 🙂

  21. That song always reminds me of Heathers. I love that film.

    You know, Magpie, I’ve been more critical of Krugman than you in the past, but I’m starting to think this heterodox / orthodox or outsider / insider dynamic may work in some ways to all our advantage. Outsiders are free to argue a point if they think it correct, and it is important that they do so. Their role seems to be to yell from the sidelines until they can’t be ignored. Then it is up to any insiders who may find themselves essentially in agreement with one or more outsider positions to synthesize what they can without losing their positions as insiders. This usually seems to require maintaining a distance from the outsiders while incorporating what they can of their arguments without giving due credit.

    It must be irksome for outsiders when ideas are plagiarized, but I suppose it must also be irksome as an insider for arguments to be off limits unless they can be forced into a flawed mainstream analytical framework. If the combined impact of having a few insiders plus some outsiders on the side of truth exceeds the impact of having only some outsiders and no insiders, it may be serving some kind of purpose. My own instinct is always to be with the outsiders, but perhaps it is just as well that some have different instincts.

    Just as a president or prime minister can only pursue public purpose that is in opposition to elite interests if there is sufficient pressure exerted from below — many yelling, striking, protesting, etc., from the political sidelines — perhaps economic insiders can only do as much as is necessary to quieten the heterodox hubbub.

    (Or maybe I am just going off my rocker.)

  22. “It must be irksome for outsiders when ideas are plagiarized,”

    Only if you believe you can own ideas. If you believe that ideas are part of the commons to be shared, then you are glad when people take them up regardless of how it happens.

  23. Good point, Neil. I very much agree that ideas should be part of the commons. However, in academic circles, as you know, it is generally understood that any prior work drawn upon will be duly cited. This does not tend to occur in the case of orthodox economists drawing upon work of heterodox economists. The exceptions are pretty rare and usually involve only the most obvious cases (e.g. Keynes, more recently Minsky).

    One reason for this seems to be that an “acceptable” critique or modification of a mainstream position must appear to be generated from within the orthodoxy. For example, in a working paper (later published in the Journal of Post Keynesian Economics), Matias Vernengo provides an example in which a prominent mainstream economist, having synthesized quite prominent heterodox work on inequality produced by David Gordon and James Galbraith in the 1990s, credits a couple of orthodox economists from the 2000s for making the seminal contributions. As Vernengo puts it, “the acceptable critiques of orthodoxy must come from mainstream insiders” (See p. 7).

    By normal academic standards, this practice is not on, but is endemic in economics. In these neo-liberal times, this has ramifications for research funding and career prospects, impacting for example on the number of citations an academic receives in highly ranked (i.e. mainstream) journals. Heterodox economists, who in some cases have done far more to advance the common pool of economic ideas, including mainstream ideas, fail to attract research funding while careerists with nothing of substance to contribute (who are legion in academia) thrive.

    Once the current generation of leading heterodox economists retires, it is hard to see there being many younger heterodox economists in academic positions from which yelling from the academic sidelines will be effective. Ranks have been well and truly closed by now. And if the academic voices are absent, any impact bloggers have also goes by the wayside, because our credibility hinges on their being peer reviewed academic output to cite in our support.

  24. @Pete

    The way I see it, is a good thing just to have mainstream and high-profile people like Krugman openly admit that there may be something to Marxism.

    I mean, in our societies, Marxism and/or socialism have endured over a century of largely unopposed rubbishing by all sorts of charlatans (and, let’s face it, lots of this rubbishing have been self-inflicted). That’s why people otherwise utterly ignorant of what socialism (or Marxism) is, feel as if they knew and feel it their right to go around parroting that. “I don’t know what ‘commodity’ is; but I don’t need to know. I’ve heard a gazillion intellectually respectable people saying Marxism is bullshit; therefore, I say it is bullshit and hopefully some of that intellectual respectability rubs off on me. Besides, Marxism doesn’t even explain why I prefer blond women to red-headed ones!”.

    It’s the same attitude you see here:

    At least Krugman’s use of the word may hopefully contribute to lift the taboo character on it.

    Would this last? Who knows.

  25. ” because our credibility hinges on their being peer reviewed academic output to cite in our support.”

    I don’t think it does.

    Economics is a religion and the current texts are religious texts. The standard religious texts might have been ‘peer reviewed’ by other believers in the cult (L Ron Hubbard springs to mind for some reason), but that doesn’t make them any more scientific than any other religious text.

    I see parallels with the non-conformist movement. Wesleyianism has more to teach us I think.

  26. Magpie, I had to chuckle. Your post prompted me to put up a brief post linking to an old post of mine about a mechanized economy. In doing so, I saw your response to that post, which is a very nice precursor to your latest contribution at Magpie’s Asymmetric Warfare. History, in this case, not only rhymes but repeats!

  27. Hey Pete,

    So, were you laughing with me, or at me?


    I was thinking about asking Trixie that same question, but on second thoughts, better not. After all, I might not like the answer!

  28. I was laughing with us but not at us. 🙂

    I was thinking about asking Trixie that same question, but on second thoughts, better not. After all, I might not like the answer!

    Wise move … (ducking for cover)

  29. Paul Krugman is still a Bastard Keynesian, in my view. By that, I’m using the term used by Post-Keynesians to describe those still stuck in the neoclassical synthesis established after WWII.

  30. Agreed, Jacob. He has moved quite a bit in the last few years, but ultimately he still adheres to an imperfectionist explanation of involuntary unemployment (liquidity trap). If he ditched the natural rate of interest, that would be something. He really should considering he now appears to accept the view that the short-term interest rate is set as a matter of policy. Doing so would strengthen his own theoretical positions, but I think this is about internal politics of academia in the sense of needing to remain within the orthodox framework. They are allowed to argue just about anything (other than questioning private property, and the monetary system, it seems) as long as they can do so within the standard framework.

  31. Peter, what I get from Krugman is that he thinks that a model is needed, and I think that while he may like and understand the PKE/MMT position, his response is that without a model, it’s not “serious” economics. I’m wondering what you make of that, whether you think it is a correct reading of his view and if so, what could be done to overcome it.

  32. Tom, that could be true. There is rigor in the heterodox approaches, but the models (rightly) tend to be open.

    For example, the simple Keynesian multiplier model treats investment as exogenous, not as a function of the rate of interest. There are, of course, good theoretical and empirical reasons for this. This choice leaves space for competing — but compatible — theories of investment. It also leaves space for competing theories of the rate of interest. Two different theorists might choose to combine the multiplier model with different theories of investment and interest. Each could be equally rigorous in their approach.

    Similarly, the simple multiplier model treats output determination as independent of the price mechanism. At the micro level, there are good theoretical and empirical reasons for this. At the macro level, also, there are good reasons for this under normal circumstances where the economy is operating below full employment. This leaves space for competing — but compatible with Keynesian output determination — theories of price. So, at the micro level there are Sraffian, Kaleckian and other Post Keynesian perspectives on price theory. Different theorists might choose to combine different theories of output, investment, price. And so on.

    There is also much needed space for institutional and historical analysis, both of which undoubtedly bear on output, accumulation and growth, as well as on distribution, pricing and the rate of interest.

    Rather than starting from the orthodox baseline model that bears little relation to reality or logical considerations (e.g. spontaneous convergence to full employment via the price mechanism) and then making ad hoc assumptions (introducing imperfections) in an effort to get the model’s results closer to reality, heterodox approaches tend to start with a basic model that is consistent with perceived reality, but agnostic on certain aspects, and conceptually open on any aspects that are contestable. Greater complexity and greater specificity is then added as appropriate to the problem being examined.

    Example 1: The simple multiplier model holds that output, below full employment, is demand determined, but is agnostic on an important component of that demand (investment), prices (e.g. interest rate), and also distribution.

    Example 2: The Sraffian model of prices is agnostic on output determination. This leaves space for alternative (e.g. Keynesian) explanations of output, accumulation and growth.

    Example 3: The sectoral financial balances (SFB) model combined with circuitist and neo-chartalist monetary theories is compatible with Keynesian output determination and competing theories of price.

    In other words, there is plenty of scope for rigor outside the standard framework. But there tends to be more of a “building blocks” approach. You start with a simplified model of what is considered the most essential aspect of the problem being addressed, then add complexity and additional (compatible) theories as appropriate.

  33. Thanks for explaining that clearly, Peter. Based on that, it seems that Krugman is unwilling (yet) to step out of the mold (pun intended). He has to know that Hicks dissed the ISLM model and that the loanable-funds/crowding out assumption is untenable based on the way the monetary/financial system works.

Comments are closed.