This archived content originally appeared at, the predecessor site to, and does not carry the sponsorship of the Cato Institute.

Market Monetarism and the Monetary Constitution

An important post from Lars Christensen at The Market Monetarist this morning.  Lars argues that many folks have misunderstood the argument for monetary easing in order to target NGDP growth by framing it in terms of discretionary policy.  The market monetarists, he argues, are not imagining a world in which the central bank is constantly fiddling with the money supply, nor does their view reject the idea that sometimes monetary policy can be too expansionary.  They are neither "doves" nor "hawks," because both of those are, again, couched within a framework of discretion.

Instead, he argues that what they want is what Buchanan wanted:  a monetary constitution.

Buchanan was a constitutionalist. He was concerned about one thing and one thing only and that was how to define the rules the game – also in monetary matters. This to me is what Market Monetarism to a large extent is about (or at least should be about).

We want central banks to stop the ad hoc’ism. In fact we don’t even like independent central banks – as we don’t want to give them the opportunity to mess up things. Instead we basically want as Milton Friedman suggested to replace the central bank with a “computer”. The computer being a clear monetary policy rule. A monetary constitution if you like.

The problem with today’s monetary policy debate is that it is not a Buchanan inspired debate, but a debate about easier or tighter monetary policy. The debate should instead be about rules versus discretion and about what rules we should have.

Obviously Market Monetarists have been arguing in favour of monetary easing in both the US and the euro zone, but the argument is made within the framework of NGDP level targeting. We not always “dovish”. In fact most of us would probably have argued that monetary policy in the US and in most Europe have been overly easy for the last 40 years. But that is besides the point. The point is that we really should not have a discussion about easier versus tighter monetary policy. We should debate the rules of the game – James Buchanan would have told us that.

Free banking types might wonder whether NGDP targeting really does operate "like a computer" give the complexities of ensuring that a given change in the monetary base will have the desired effect.  However, Lars does make an important point in arguing that NGDP targeting is a rule and that debates over the desirability of further monetary expansion (which Lars, I believe, does not favor at this time) should be framed in terms of what we would want a good "monetary constitution" to generate, rather than our preference over what sort of discretionary path the Fed should take.

I have written on free banking, Buchanan, and the monetary constitution in "Do We Need a Distinct Monetary Constitution?" which is part of the Buchanan symposium in JEBO.  The ungated SSRN version is here.

Cross-posted at Coordination Problem.

  • Mike Sproul

    We should no more favor targeting the money supply based on NGDP than we should favor targeting the production of apples based on NGDP. Free market economists understand that the production of goods is best left to free markets. Unfortunately, those same economists fail to see that what is true of goods is also true of money.

    One bright spot: While the monetary socialists debate how many Federal Reserve Notes our monetary commissars should issue, the free market is quietly developing substitute moneys like credit cards, debit cards, gift cards, etc. The monetary socialists, thankfully, haven't expressed much interest in regulating the substitute moneys–so far.

  • Bill Stepp

    Constitutionalism defines the "rules of the game" from the top down, which is why all constitutions, including a monetary constitution, are con jobs. Social institutions, including law, money and banks, develope spontaneously (or spoontaneously) on the market. Mor Spooner, less Buchanan.

  • Steve Horwitz

    Both Lars and I agree that a free banking system is to be preferred to a rule-constrained central bank. If you actually read the paper of mine I linked, you'll find me making more or less exactly the argument you just summarize, though without the language of "con jobs." And you do misunderstand what Buchanan means by "constitutions." He doesn't necessarily mean a literal constitution, but rather the set of rules that frame the game. That is what he thinks is necessary. Yes, he tended to believe that they could not rise endogenously, but demonstrating that he was wrong about that has been a core project of a number of younger economists influenced by Austrian ideas. I also think it doesn't destroy his more fundamental point about the importance of the rules.

    Finally, I love Spooner, but in no universe are his contributions to social theory more important or better than Buchanan's. Just sayin' – as the kids today say.

  • Thank you for an interesting exchange, but I wonder if Buchanan would really have supported nominal GDP targeting – I may have missed something but is there any evidence? Whatever the theoretical case for regarding it as a rule-based discipline, or monetary constitution, make no mistake, the attraction to policy makers right now is that they can use it to free themselves from any remaining inhibitions that conventional inflation targeting imposes on them. That is why the UK government loves Mark Carney's floating of the idea – who imagines that the likes of George Osborne cares a fig for the academic arguments? What they want is an old-fashioned, 1970s style "dash for growth".

    • Lars Christensen


      I think you are missing the point here. Both Steve and I are pretty clear that Buchanan did not advocate NGDP level targeting. However, the point is that it is important to get a rule.

      As a consequence it is meaningless to talk about advocating easier or tighter monetary policy. It is a question about whether or not you favour a ruled based monetary regime or not. Therefore, it is also a great mistake to think that NGDP level target in the UK would be a 1970s style "dash for growth" monetary policy. If you think so then you simply hasn't studied what NGDP level targeting is about. Both the UK and the US would have had "tighter" (and lower inflation) monetary policy over the last 40 years (and the last 20 years for that matter) if the countries had had a 4 or 5% NGDP level targeting rules. See this post:

      Can we question George Osborne's motives? Sure – James Buchanan would remind us that he is a politician so his motives are likely to be far from noble. But of course neither Steve nor I need to be reminded about that – that is exactly why we are advocating a monetary constitution in the form of a NGDP level targeting rule (if we cannot get free banking).

  • Steve Horwitz

    No one is arguing Buchanan would have supported this particular rule. Lars makes that point explicitly in his post. The concern with HAVING a rule, however, is very Buchananesque.

  • But in present circumstances it would open the door to a very un-Buchananesque discretionary, politically-driven outcome….

    • BillWoolsey

      Last spring, Buchanan advocates a constitutional rule mandating a stable price level. He wasn't focused much on a formula for controlling base money and tying it to the price level. The constitutional rule would be for a stable purchasing power for money.

      Nominal GDP level targeting is similar, but a different target. Slow, steady growth for spending on output rather than a stable price level.

      Nominal GDP targeting allows or implies that the price level shifts with changes in supply.

      From Lars' (and my) perspective, we agree with Buchanan, we just think the price level is the wrong nominal anchor to constitutionalize. Spending on output provides the benefits of a price level rule without some of the disadvantages.

    • Lars Christensen

      Robert, why is that?

      Isn't what we have had over the past four years both in the euro zone and the US extremely discretionary monetary policies???

      That said introducing a rule is in itself a discretionary act. Maintain the status quo on the other hand can hardly be described as a rule – at least then it is an insanely bad rule. But see here on that topic:

  • MichaelM

    Given a serious enough attempt at setting an actual, predictable rule-based monetary policy, you would eventually find this monetary policy becoming no policy at all. That is, the rule would become the new neutral, markets would adapt and eventually begin behaving as if there was no monetary authority at all.

    • Lars Christensen

      MichaelM, Spot on! This is exactly what we want. We want a rule that is "neutral" monetary policy in the sense that the central bank has no impact on relative prices and the markets behave as if there was no central bank at all. I think this is the key reason why an NGDP level targeting rule is the preferable monetary policy rule as it is the only monetary policy that in fact is "neutral" in the sense of not distorting relative prices.

      • MichaelM

        Such an environment might be a good 'first step' towards an actual free banking system. It is, at least, a promising step away from the arbitrariness of discretionary policy.

        • I wish it were so, MichaelM but I can't see a good reason to expect that; even given saintly guardians – and in the shadow of Buchanan how can or should one even imagine such an animal? -the genuine difficulties and dilemmas of implementing such a rule renders it even more vulnerable to political meddling.

  • In response to the political economy point made by Lars, certainly I agree, we need to make a public choice analysis, and it is precisely the reason why I doubt that NGDP targeting is likely to provide a more binding or effective rule in practice than IT. The collapse of credibility of IT should lead us straight to more radical alternatives; see – "Goodhart's cry of despair" and an earlier column see

  • Rick

    Steve, in your conclusion of the SSRN version of "Do We Need a Distinct Monetary Constitution?," you state:

    "A freely competitive banking system with bank-issued money redeemable in an actual money commodity, resting on the more general constitutional protections for private property, contract, and exchange, will keep money out of the political arena, ensure that relative prices consistently reflect underlying variables of tastes, technology, and resources, and provide the predictability necessary for economic calculation and long-term planning by households and firms. Generating those results requires, as Brennan and Buchanan (1981: 65) note, that we engage the constitutional question and recognize that an unrestrained political monopoly over money is the source of much trouble. It need not, however, mean that explicit constitutional treatment of money is the best solution."

    Although the mind automatically repels the idea of physically dealing with current U.S. coin, why can't this be the "actual money commodity" you recommend? Then, it seems to me, there'd be no need for an implied constitutional treatment of money (because the Coinage Clause (I:8:5) explicitly authorizes Congress to produce any kind of coin it chooses, so long as it's metallic. Of course, I'm not suggesting depositors take actual possession of the coin (though it's their right to do so), but I am suggesting banks would have an obligation to actually store the stuff and treat those coin-based deposits differently, and more responsibly, than Federal Reserve note deposits.

  • Rick

    Another quote I thought was significant from Professor Horwitz's "Do We Need a Distinct Monetary Constitution?," which refers to money's central role as a tool for accurately measuring economic value (which it hasn't been doing for a long time):

    "One oft-drawn comparison (e.g., Buchanan 1962) is to weights and measures. Because weights and measures are used for comparative purposes, their users need confidence that the scale or ruler they are using is accurate both longitudinally and cross-sectionally. The user of a ruler, for example, needs to be sure that both the ruler and the very definition of an “inch” or a “foot” do not change through time as he uses the ruler. What the ruler measures as six inches today must be equal in real terms to what it measures as six inches tomorrow or a year from now. In addition, rulers used by different people at one point in time must also be stable and consistent. If I measure a length of pipe at three feet and someone else uses a different ruler measuring a length of pipe that will fit it and that should also be three feet long, we must both believe that our rulers are measuring three feet the same way. Without such assurances, weights and measures are much less useful, just as money that constantly changes in value is similarly less useful for the assessment of comparative value that is at the heart of economic calculation."

    This reminded me of the quote by Alexander Del Mar, on page 58 of "The Science of Money" (1885):

    "When money shall be recognized in the law, when it is defined, when its volume, magnitude, dimensions, limits are set forth as precisely, fixed as unchangeably, and protected as securely from alteration, as are now the dimensions of the
    yard-stick, the pint-pot, and the pound-weight, then, and then only, will money perfectly resemble other measures; for then only will it become a concrete thing of known dimensions. When this comes to pass, Aristotle's defini- tion of its function will resume its original correctness, and money will be as fit in fact, as it is now only in theory, to measure the relation called value."