Vox on market monetarism

Vox.com did a fairly long article on my views on monetary policy, and market monetarism more generally.  I´m traveling so I´ve only quickly read it once, but I thought it worth linking to.  I was also told that on Sunday the big Frankfurt newspaper (FAZ) may do a piece on me.  Let me know if you see anything.

I´d like to thank Timothy Lee for taking the time to research this topic, and also interview me.  It looks like an excellent piece.

PS.  I see that Tyler Cowen linked to my Kansas tax post.  Rereading the post I realize my claim about a 25% rise in Kansas GDP was too strong.  Tax avoidance might fall, and there are the lower tax rates to consider (which were cut by fairly similar amounts.)  But I stand by the gist of my argument.

Update:  I’m told the FAZ story was delayed.



43 Responses to “Vox on market monetarism”

  1. Gravatar of Michael Byrnes Michael Byrnes
    8. July 2014 at 14:46


  2. Gravatar of benjamin cole benjamin cole
    8. July 2014 at 16:19

    Excellent Vox story! Congrats to all.
    I particularly like the derivative definition of Paul Krugman as a “Sumner critic.”
    That’s the right way to frame the issues!

  3. Gravatar of Sam Sam
    8. July 2014 at 16:51

    Quick question, probably answered on this site before, but I’d appreciate a reference if so:

    Scott advocates NGDPLT, and suggests the target level should be determined by a trend growth (roughly 5%) in a certain quantity NGDP. So in some sense the claim is that in an ideal world, another quantity, the annual rate of change of NGDP, should be held constant at 5%. The annual rate of change is a particular function f(x,y) where x = rate of inflation and y = rate of RGDP growth; namely, f(x,y) = x + y. What is the justification for this functional form? Put differently, NGDPLT suggests that the policymaker should try to target the return on a portfolio containing one instrument pegged to the price level and another instrument pegged to the total production of society. But a priori this seems sort of arbitrary. Why not a portfolio containing, for example, an instrument pegged to the nominal value of total production, i.e. price level times total production? This would be a world in which the policymaker tried to maintain a constant value for (RGDP * inflation) + [price level] * [growth in RGDP].

    Is there a simple justification for why the functional form inflation + (RGDP growth) is the correct rate of change that a welfare-maximizing policymaker should be targeting?

  4. Gravatar of Ricardo Ricardo
    8. July 2014 at 17:31


    A study in the Proceedings of the National Academy of Sciences this week … found that …

    “High earners in a stock market game found to have brain patterns that can predict bubbles and crashes.”


  5. Gravatar of Morgan Warstler Morgan Warstler
    8. July 2014 at 18:17

    Because Reagan was sometimes quaint, let’s remember he thought budget deficits cause inflation:


    (Johnny Carson clip)


    We should also remind ourselves that once in office, Reagan ran MASSIVE BUDGET DEFICITS while he had Volker whip inflation.

    So exactly how much could deficits cause inflation?


    If we admit they do cause a little inflation…

    Then we’d say as MM that they don’t cause ENOUGH inflation, right?

    That if you are trying to cause inflation, printing money is the best possible option.

    Maybe that’s the idea:

    The government can’t cause real growth (fiscal) because it is dumb and full of lazy politicians.

    But the government can cause inflation.

    Whenever the government tries to cause real growth, it will fail.

    Whenever the government tries to cause inflation, it can do so with MP.

    INFLATION: SO EASY, even the government can still f*ck it up. Better to leave inflation to the bankers.

  6. Gravatar of Tom Brown Tom Brown
    8. July 2014 at 18:45

    Someone who agrees here… except for one little thing.

  7. Gravatar of Tom Brown Tom Brown
    8. July 2014 at 18:53

    Morgan, I know you follow Jason’s blog. Today he responded to Vincent Cate (the hyperinflationist). Vincent wasn’t happy with Jason’s hyperinflation model, and has thus not added it to his collection of hyperinflation explanation he maintains on his blog. For one, Jason’s model makes it seem like a CB needs to be willfully stupid to cause hyperinflation (Vincent’s interpretation)… and Vincent’s prior seems to be that hyperinflation should be as easy as spilling a glass of milk… but also one of his complaints to Jason was, “Where’s the debt in your model?” You might find Jason’s response interesting.

  8. Gravatar of Ralph Musgrave Ralph Musgrave
    8. July 2014 at 21:31

    Market Monetarism consists of the Fed printing money and buying private assets. You might as well have the authorities print money and raise spending on some other equally narrow sector/s of the economy: e.g. just on the military and infrastructure.

    Given a need for more spending, there is no reason to concentrate it on any particular area, or any relatively narrow group of people, like the asset rich.

  9. Gravatar of W. Peden W. Peden
    9. July 2014 at 00:10

    Ralph Musgrave,

    How often do symmetry differences between helicopter drops and OMOs have to be raised before you acknowledge them?

  10. Gravatar of Scott Sumner Scott Sumner
    9. July 2014 at 01:25

    Sam, I´m afraid I don´t understand your question, maybe someone else can answer. In general, I regard NGDP as an objectively measurable entity, and inflation and RGDP are merely arbitrary creations of government bureaucrats, with no clear linkages to real world entities. What is the rate of inflation supposed to measure?

    Over time, I´ve come to prefer total labor compensation (not NGDP) as a target. But they are pretty similar for the US.

    You could Google one of my papers advocating NGDP targeting.

  11. Gravatar of Jean Jean
    9. July 2014 at 05:00

    Congratulations on the FAZ piece – it would be great if the members of the Bundesbank read the piece and finally see the light! Having lived in Germany, though, I think the chances of that happening are slim to none.

  12. Gravatar of MG MG
    9. July 2014 at 05:15

    On the marginal issue addressed in the PS…

    I agree with your criticism of Krugman’s strawman and overbroad criticism of supply side economics. I would also note that months before Krugman thought it was safe to rail on Kansas, this very issue was already been covered at the “micro level” by the Tax Foundation. See http://taxfoundation.org/blog/kansas-may-face-budget-problems-senate-again-strips-tax-reform-out-tax-cut-bill
    This analysis, which dates back months, highlighted the real risk that the lack of base broadening (including the opening of a a very crucial loophole concerning pass through businesses). Their analysis, although ostensibly coming from a supply side perspective, highlighs that tax reform details matter. Balance and subtlety versus Krugman’s rants.

  13. Gravatar of Ralph Musgrave Ralph Musgrave
    9. July 2014 at 06:41


    What makes you think that I think that OMOs and helicopter drops are identical (if that’s what you’re saying)? Far as I know I never said that, and certainly never INTENDED to say it.

  14. Gravatar of W. Peden W. Peden
    9. July 2014 at 06:51

    Ralph Musgrave,

    If there are potentially important differences, e.g. it’s easy to send round cheques to every household, but hard to send round bills, then there is a difference between doing helicopter drops and OMOs.

    OMOs, unlike helicopter drops, are symmetrical: buy assets, sell assets.

    You need to at least start off by addressing that concern before you can come out with the sort of arguments you are trying to use.

  15. Gravatar of TravisV TravisV
    9. July 2014 at 10:40

    Prof. Sumner,

    I think you’d find these three recent posts interesting:




  16. Gravatar of Don Don
    9. July 2014 at 11:12

    Congratulations on the VOX piece. That should help bring the message to a new group of people. Keep up the fight, because there is a lot of money aligned against MM.

  17. Gravatar of TravisV TravisV
    9. July 2014 at 18:01

    There’s a new link for Milton Friedman’s famous article: “Reviving Japan”


    The old link doesn’t work anymore:


  18. Gravatar of Hansen Hansen
    10. July 2014 at 00:34

    This is the most succinct summary of your views I’ve read.

    Good article. Great interview. What’s sad is that to get these views published in an econ journal requires so much more pain, but it’s so necessary to give the veneer of sophistication.

    Any chance you’d do a post on Australian monetary policy? The seasonally adjusted unemployment rate just hit 6 per cent. Low interest rates, tight money, apparent fiscal tightening, core CPI targeting with bounds that mimics NGDPLT at times, low NGDP growth since around 2011 or so, forward guidance committed to current stance, monetary offset, receptive audience, conservative government in power: what more could you ask for?

  19. Gravatar of Philippe Philippe
    10. July 2014 at 01:54

    W. Peden,

    “OMOs, unlike helicopter drops, are symmetrical: buy assets, sell assets.”

    Why not do a helicopter drop (print money to spend as Ralph says) and then, if needed, sell assets (government bonds) at a later date?

  20. Gravatar of Brian Donohue Brian Donohue
    10. July 2014 at 05:02

    @Ralph Musgrave,

    Regarding printing money and buying assets. The first part (printing money) causes a diminution of value of every existing dollar out there. This is what makes Austrians apoplectic.

    I think it’s odd to characterize this as being pro-rich.

  21. Gravatar of Dan W. Dan W.
    10. July 2014 at 06:00

    The expansion of private & public debt has expanded the monetary base. A large part of that money has gone into the purchase of assets. It has also gone into elevating demand across the economy which has offset productivity driven deflation.

    Is perpetual growth of the monetary base stable? That would seem to be the most important question if one wishes to use the monetary base to offset natural economic deflation.

    Is the practical answer to the question the same as the theoretical answer? Is it practical to assume a merry band of bankers can manage a multi-trillion dollar portfolio without blowing it up?

    Just curious.

  22. Gravatar of jj jj
    10. July 2014 at 06:37


    It doesn’t matter exactly which function is targeted. The key point is that the future value of NGDP should be predictable and, even better, guaranteed. This condition could be met by defining a constant NGDP from now until forever; but Scott thinks that having NGDP grow every year will overcome some wage and price stickiness.

  23. Gravatar of Daniel Daniel
    10. July 2014 at 06:53

    Is perpetual growth of the monetary base stable?

    No, when it reaches critical mass it starts to emit gamma radiations.

    offset natural economic deflation.

    What does that mean and why does it need to be offset ?

    Is it practical to assume a merry band of bankers can manage a multi-trillion dollar portfolio without blowing it up?

    It’s definitely not practical to assume a “merry band of bankers” manages the monetary base, since it doesn’t.

  24. Gravatar of Garrett Garrett
    10. July 2014 at 07:10

    Good article on the Fed’s inflation target:


    Or lack thereof.

  25. Gravatar of Dan W. Dan W.
    10. July 2014 at 08:02


    I am confused by your logic. On the one hand you advance the thesis that the Central Bank has the ability to accommodate greater expansion of the monetary base. But then you say they do not manage or have influence over it. What do they do then?


    When you write that NGDP should be predictable and, even better, guaranteed how do you suppose that can happen? GDP is an expression of economic activity. What power does it have to guarantee economic activity, whether in real or nominal terms?

    People and the institutions they manage engage in economic activity according to decisions they make. An aggregate function may assume it is representative of the collective decisions of all market participants but such a function cannot predict nor can it guarantee future decisions.

    If consumers recognize higher prices, but they do not yet see higher incomes what do you expect their behavior to be? Does it not matter how timely consumers make purchasing decisions. Does the Central Bank control the temporal mood of consumers?If consumer behavior is not predictable in real terms how is it predictable in nominal terms?

  26. Gravatar of jj jj
    10. July 2014 at 09:54


    First an extreme and much-simplified case:
    1. Fed states that 2015’s NGDP will be same as 2014.
    2. On Jan.1 2015, an asteroid strike destroys 50% of US. Pretend that the productive capacity of the remaining half is unaffected.
    3. On Jan.2, the fed doubles the money supply.
    4. On Jan.3, all prices and wages double.
    5. Life carries on… and at the end of the year, NGDP is the same as 2014.

    Second, a realistic simplified case:
    1. Fed states the level of future NGDP will be NGDP_2014*1.05^n, where n is years.
    2. At the end of every year the fed assess what actual NGDP was. It then estimates what to do to the money supply to reach next year’s NGDP target, and carries that out.
    3. Because the fed isn’t a perfect estimator, it will either over-shoot or under-shoot NGDP every year. However because it’s level targeting, there should not be a persistent bias in either direction, and NGDP will more or less follow the target.

    Third, with futures markets:
    1. same as above
    2. Fed sets up a futures market for NGDP. When market predicts NGDP will be low, fed creates money until market predicts NGDP will be right on target… and vice versa.

    All of the above cases are purely mechanical and don’t rely on the magical “expectations fairy”. However once you add in the effect of expectations you’ll see that the fed’s job becomes even easier, as all market participants will (on average) smooth out their own behaviors secure in the knowledge that NGDP will be on target.

  27. Gravatar of jj jj
    10. July 2014 at 10:02

    Rereading your comment, I think I didn’t clearly address your point. I’ll try again:
    The fed would not directly control any consumer behavior. If all consumers decided that this year they were going to spend only half their incomes, the fed wouldn’t be able to stop that. But by doubling the money supply it could cause wages and prices to double, such that the total spending over the year remained the same!
    In the limiting case, if all consumers decided they would only spend a fixed dollar value, could the fed still control NGDP? Probably not, at least I don’t see how. But in the real world the public’s consumption decisions are pretty smooth and stable, and only minor adjustments in the money supply are required to keep NGDP at exactly the desired level.

    Put another way:
    NGDP = RGDP * price level
    Consumers completely control RGDP, but the fed completely controls price level. Therefore as long as the fed can react fast enough, NGDP is completely under the fed’s control.

  28. Gravatar of Dan W. Dan W.
    10. July 2014 at 10:34

    “the fed completely controls price level”

    I believe this is an economic tautology not manifest in the real world.

  29. Gravatar of jj jj
    10. July 2014 at 12:00

    Please explain. I see it as almost self-evident that if the fed doubles the money supply, in the long run prices and wages also double leaving all else as it was before. The rest follows naturally to me, so I wonder where in the chain of logic is the step you don’t agree with.

  30. Gravatar of Dan W. Dan W.
    10. July 2014 at 16:44

    My review of the data shows growth in the money supply far outpacing growth in nominal wages. For example, since 2000 the money supply has increase more than 100% (M1 or M2) Nominal wages are up less than 50%. In the past 5 years I see a 25% increase in the money supply and a 10% increase in nominal wages.

    Does this disprove the theory?

    Link to nominal wage growth

  31. Gravatar of Coleton Stirman Coleton Stirman
    10. July 2014 at 18:24


    It is a great read, and about damn time the “lamestream media” gave you and market monetarism the credit you deserve.

    Congratulations, keep up the great work, it has truly been a pleasure to watch you to teach the World about proper monetary policy from such a lowly perch.

  32. Gravatar of ssumner ssumner
    11. July 2014 at 01:00

    Hansen, I’ve suggested that they focus on targeting total nominal labor compensation.

    Thanks Jean and Coleton.

  33. Gravatar of jj jj
    11. July 2014 at 06:01

    Dan – To explain that you need to consider Velocity; I had discussed only the Money part of the MV=PQ equation. If Money doubles but we observe there is little change to Prices or Quantity, then Velocity must have fallen by half. This is what’s been happening since 2008. If V now increases and the fed doesn’t reduce M: first Quantity will increase to the economy’s full productive capacity, and then Prices will start to rise.

    In 2008, V fell due to banking problems and the expectation of recession which led to reduced spending. The fed didn’t increase M enough to make up for it, so Prices*Quantity fell. Because Prices are sticky, they didn’t adjust fully and Quantity (or RGDP) fell instead. The expectation of recession became self-fulfilling.

  34. Gravatar of Tom Brown Tom Brown
    11. July 2014 at 09:26

    Scott, I’m not that familiar with Vox, but Noah Smith wrote this about it yesterday:

    “Two thirds of his article harps on the fact that the website Vox has a liberal slant — as if it’s a revelation that a center-left publication is center-left.”

    This made me wonder, what do you think the probability is that Market Monetarism will be embraced by the center left before it is by the center right? Maybe whoever it is in charge of MM marketing should look into this. 😀

  35. Gravatar of Michael Byrnes Michael Byrnes
    11. July 2014 at 09:34

    Interesting post from John Cochrane on rules vs targets:


    If that’s a real dichotomy, put me in the “targets” crowd.

  36. Gravatar of Tom Brown Tom Brown
    11. July 2014 at 09:37

    Do you share (to any degree), Krugman’s growing alarm he expresses in his recent post that the right-wing “monetary fever swamp” (Benjamin Cole calls this crowd the “right-tighties” I think) is on the upswing?


  37. Gravatar of Tom Brown Tom Brown
    11. July 2014 at 09:39

    Michael B., Nick Rowe has a related post up yesterday (at least I thought it was related).

  38. Gravatar of ssumner ssumner
    12. July 2014 at 14:51

    Tom, I’m alarmed by both the left and the right on monetary policy.

  39. Gravatar of Karl Pinno Karl Pinno
    12. July 2014 at 22:11

    I am an instructor at the U of C Economics department. I have recently developed an appreciation of your ideas.

    I read your FAQs. I have the following questions:
    Are there any practical limits to the quantity of assets a central banks may own?
    How does your theory reconcile/account for the stagflation of the 1970s?
    Do you think that sticking to a NGDP target might cause inflationary expectations to become unanchored?
    If the BOJ cancels/monetizes the JGBs what will happen?
    More broadly do you seem limits to NGDP targeting if so what are they.


  40. Gravatar of ssumner ssumner
    13. July 2014 at 01:02

    Karl, There may be legal limits to what central banks can own, but they are not a factor for the US, at least in recent years. It may not be desirable for the central bank to have a very large balance sheet. This issue would play a role in determining the optimal NGDP growth rate.

    The stagflation of the 1970s was mostly caused by rapid growth in the monetary base. As NGDP growth rose to about 11 percent a year, velocity also picked up a bit. RGDP grew at roughly the trend rate of 3 percent, while inflation was close to 8 percent.

    With NGDP targeting, inflation expectations will move inversely to long run trend changes in RGDP growth. Those changes are likely to be small.

    The effect of BOJ monetization of bonds is complicated. The faster the inflation/NGDP target path, the more likely that monetization would caused inflation to overshoot their target.

    I don’t see technical limits to NGDP targeting, But the policy is less desirable when total labor compensation doesn’t closely track NGDP.

  41. Gravatar of TallDave TallDave
    14. July 2014 at 10:58

    Wow, that’s the first worthwhile Vox article I’ve ever read. Congrats!

  42. Gravatar of TallDave TallDave
    14. July 2014 at 11:12

    Karl, “sticking to a NGDP target might cause inflationary expectations to become unanchored” is probably one of the more common objections seen from CBs. But generally speaking NGDPLT isn’t much different in practical terms than inflation targeting — except when NGDP is significantly off-trend, as in 2008-9.

    In fact I think this understates the case — “unanchored” expectations suggest the public does not believe the CB. This certainly happens but cannot really be said to be the result of any particular reliably implemented monetary policy — perhaps more accurately it could be said to be the lack of a reliably implemented monetary policy.

  43. Gravatar of Vivian Darkbloom Vivian Darkbloom
    16. July 2014 at 22:41

    ” I looked at Wikipedia, but it was no help.”

    Economists need an authoritative source of definitions used in their profession. It would greatly facilitate communication among them and avoid needless and confusing debates over what a term means. It would make the academic (and blog) debates much more efficient and would assist in identifying charlatans who, when caught out, frequently insist on their own unique terminology as a way to save face.

    There is no reason a term such as “Aggregate Demand” should not have a fixed definition. That would not mean a particular person would not be able to deviate from the definition, or argue that the definition should be revised; it would only mean that they would be obliged to clearly indicate how they are deviating and why. At least then everyone would have a common reference and starting point.

    I would suggest a good project for economists would be to impanel a working group of economic wise men and women to come up with a draft “Dictionary of Economic Terms”, publish it for comment, subsequently finalize it and revisit it periodically.

    Other professions have it easier: Lawyers can look to the Code, regulations, court decisions and Black’s Law Dictionary. Psychologists and Psychiatrists to the Diagnostic and Statistical Manual of Mental Disorders (DSM), etc. Economists need a DET.

Leave a Reply