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

Monetary Policy: Is There a Prudent Second Best?

It's usually a good thing when representatives of the Von Mises Institute and George Mason ends of the Austrian-economics spectrum agree with each other. But I can't say I found it much to my liking when Joe Salerno and Pete Boettke agreed with each other concerning a CNBC interview I gave a week ago. For both held that, by declaring during that interview that "a case existed" for quantitative easing back in late 2008 and 2009, I seemed to forget my own case for free banking, joining ranks instead with apologists for monetary central planning.

Having defended myself against this particular accusation at the two sites where it came up (and in Joe's follow-up post), I don't intend to repeat my defense here, except to observe again that "a case existed" isn't the same as "a slam-dunk case existed," or something equivalent. I do, however, want to further engage the broader question concerning the pitfalls of the second best. For while I agree that the pitfalls are real, it hardly follows that what might be called the "intransigent first best" strategy, which is to say the strategy that treats second best prescriptions as slippery slopes to be rejected out of hand, is a more prudent alternative.

In considering whether it is, let's not conflate the question at hand with questions concerning the merits of the Fed's actual, recent undertakings. It seems to me relatively easy to find fault with those undertakings, including all three rounds of quantitative easing, even considered merely as second-best means for encouraging recovery, and I have for that reason not found it especially difficult to resist endorsing them. Instead, let's suppose that it is the autumn of 2008, that the wholesale credit market has shut down (thanks in large part, admittedly, to the combined shenanigans of the Fed and the FDIC), and that aggregate demand has consequently begun its downward spiral. Let us also assume (again, for the sake of argument, but also because there is, I believe, plenty of evidence for it) that the collapse of demand, if allowed to go on, is bound to lead to falling sales and unemployment and other sorts of hardship, beyond the hardship made inevitable by the end of any unsustainable boom, in part because prices, and wage rates especially, can only be expected to adjust downward in response to fallen demand very gradually, if at all.* Let's assume, further, that a free banking system would tend automatically to counter the tendency for demand to shrink, even despite a fixed monetary base, by allowing banks to operate on lower reserve ratios as the real demand for their deposits and notes increases. Finally, let's suppose that we are fully aware of the mischief that discretionary central banks can do, and that we are all anxious to avoid making any statement that might be construed as approving of, much less further empowering, such institutions.

The relevant question then is: given the existing, centralized arrangement, what should the Fed do? Suppose the question is asked of you on a popular, live TV broadcast. How to respond prudently?

Will merely insisting on the first-best solution suffice? I don't think so. In the present context that would mean saying something like, "The Fed should freeze the monetary base; but that isn't all: Congress should then wind it up, while allowing other banks complete freedom to meet the public's monetary needs, including the freedom to issue their own notes. This would also require doing away with deposit insurance and…" etc. Even supposing one could elaborate the argument in a few sound bites, or that one could go on for hours, the obvious problem remains that the steps required would take months to accomplish even if they could be instantly and universally agreed upon. The answer, therefore, begs the question, "What should the Fed do in the meantime?" We remain more or less where we began.

Nor, as some commentators (myself included) have noted elsewhere, is saying that the Fed should do "nothing" any better. So long as it exists the Fed is, of necessity, doing "something." So "nothing" here must actually refer to some particular Fed policy. Most often (I gather) it means that the Fed should refrain from any further lending or open-market activities, or from otherwise expanding its balance sheet. It amounts, in other words, to recommending that the Fed maintain a constant monetary base. But is asking the Fed to maintain a constant base really indicating any more principled opposition to monetary central planning than one indicates by calling upon it to alter the base by some particular amount, or by whatever it takes to achieve some particular target? I don't see why. Indeed, what some have wrongly taken to be the more principled of the two alternatives seems to me to be hardly more principled, though rather less prudent, for it calls, not for the avoidance of monetary central planning, but for the implementation of a monetary central plan that is likely, according to "our" theory, to be particularly lousy. Nor will it do to say that the "keep the base constant" alternative is superior in that it might be proposed, not just as a suggestion for the present, but as a hard-and-fast monetary rule, for the same might be said concerning the suggestion that the Fed expand the base only when doing so serves to keep spending stable.

The option of recommending that the central bank "do nothing" is, by the way, precisely the one Hayek chose to take back in the early 1930s when, despite having recognized in print the desirability of a constant "money stream," and despite the fact that the money stream had dried up dramatically, he campaigned against expansionary monetary policy. As he explained many years later, with regrets, Hayek's reason for departing from his own theoretical ideal had to do, not with any slippery-slope considerations, but with his belief that allowing the collapse of demand to go on could be just the thing needed to wrench the British labor market free from labor unions' stranglehold.** Still it is difficult to see why the results would have been any different if, instead of having opposed monetary expansion because he hoped by doing so to help thaw a rigid labor market, he did so on other political-economy grounds. To say that Hayek's strategy backfired is putting it mildly, for it was not labor unions but Hayek's own theoretical legacy that suffered.

Must one, then, simply endorse monetary expansion, setting aside the first-best alternative altogether, while also neglecting those political-economy considerations as point to the inherent dangers of such a compromise? I don't believe so. It seems to me that making a case for monetary expansion is not the same as making one for monetary central planning. It is a matter of choosing one's language carefully. The prudent answer to the interviewer's question may be something like this: "For better or for worse, we are forced to rely on the Fed to prevent such a collapse in demand, so the Fed should do what it is supposed to do. But the Fed is a deeply flawed system and continuing to rely on it is asking for trouble. There are better alternatives, and now is as good a time as any to start taking them seriously."

Perhaps putting it so would still make one an apologist for central banking. In that case, my plea to Pete and Joe is, "guilty as charged." But why settle for that, fellows, when you can throw the whole first-best book at me? After all, I also believe that, until better (free market) alternatives are in place, the USPS ought to deliver my first-class mail, and the FAA ought to keep the airplane I'm flying in from attempting to land on a busy runway.


*I understand that the Salerno-Austrians will regard these assumptions as proof of my being a "Keynesian" and of my failing to understand Say's Law. In answer I can only observe (1) that if I'm a Keynesian so, too, are Milton Friedman and Leland Yeager, among others; and (2) that my "problem" is not that I don't understand Say's Law; it is, rather, that, like most neoclassical (and some classical) economists, I happen to think that Say erred in assuming that people (or banks) never attempt to accumulate reserves or broader-money balances. I know not how anyone can look at recent statistics regarding banks' (and other firms') accumulations of cash reserves and still treat Say's position as if it were irrefutable.
**See Lawrence H. White, The Clash of Economic Ideas (Cambridge, UK: Cambridge University Press, 2012, p. 95.

  • Lars Christensen


    I can understand that you are somewhat puzzled by Pete and Joe's critique, which I frankly speaking find extremely misplaced.

    The fact is that the fed exist and we do not have a free banking system even though that would be desirable. So as long as the fed exist we must demand it do the right thing – which in 2008-9 would be monetary easing within a rule based framework.

    I have a counter example. Lets assume that we think provision of protection of property rights is best taken care of private competing "protection agencies", but there instead is an government monopoly on this service. Imagine then we have a crime wave. Would Pete and Joe then argue "the police should 'do nothing' and instead the protection of property should be left to the market"? Or should they argue that the police – given its monopoly obviously should do its job and protect private property?

  • Rob R.

    This is a great article with which I wholeheartedly agree.

    I have a question on something in your footnote: You say "I happen to think that Say erred in assuming that people (or banks) were never tempted to accumulate money balances". My question is why will free banks be better than a CB with a policy rule at delivering a constant "money stream" if they sometimes choose to accumulate money balances at the same time as the non-banking sector is doing the same. A CB with a policy rule would seem better placed to deliver the constant money flow in this scenario.

  • Bill Stepp

    In response to Lars's example, if there were a crime wave (I mean other than the ongoing permanent crime wave of State action), absent the workings of government "defense," private defense services would spring up to protect people. In 2008, if we could have pressed the button and ended the Fed (as well as the other criminal entities involved in creating the real estate bubble), the economy would have recovered eventually.
    We should be Rothbardian on this and "press the button."
    I guess that would lead to not being invited back to do TV interviews.

  • Olav

    Isn't Say's Law correct as long as you view money as any other good? That is, isn't hoarding money (or increasing cash balances) the same as hoarding apples? Isn't hoarding or taking provisions an act of production?

    • George Selgin

      Olav, your interpretation of Say's Law, though a common one, isn't consistent with the arguments Say himself makes in the famous chapter in his Treatise (Book I, chapter XV) that is the law's locus classicus. There Say goes to great lengths to try and prove that there can never be a general glut of goods (not goods + money) stemming from a shortage of money, because people who acquire money in exchange for goods do so only in order to spend the money on goods.

      Later (I believe in reaction to the Crisis of 1825) Say himself relented, acknowledging the possibility of a money shortage and consequent glut of goods. Concerning this see Thomas Sowell's excellent Say's Law: An Historical Analysis.

  • BillWoolsey


    The better analogy is a government monopoly on electricity. If the electricity monopoly allocated electricity to promote the production of the most important goods (this steel plant can have this much electricity because it is important to produce this much steel so it can be sent to this car factory and produce this many cars and over and over) then this would be central planning.

    The status quo is something like having the government monopolist adjust the price of electricity every six weeks and supply the amount demanded to whoever wants to buy it. If the price is set too high or too low, problems develop. In fact, set it low enough and we can see that it will be impossible to both produce all of the electricity demanded and all of the other goods and services that it would be profitable to produce with low energy costs. And when all of those new factories are built based upon the signal of these unfeasibly low energy costs, they can be deemed "malinvestments."

    Those Austrians like Boettke and Salerno (as different as they are in many ways) insist that the electricity monopoly should always keep the quantity of electricity constant. And even when there is a surplus of electricity, when the monopoly lowers the price, they begin telling stories about malinvestment.

    Superficially, we market monetarists are proposing having the monopolist auction off electricity to the highest bidder and supply a quantity such that spending on output (much of it produced with electricity) is growing at a slow stable rate. Of course, this confuses money and credit. Market monetarists favor letting the market determine interest rates and generally think of monetary instruments as trading at a fixed face value.

    However, suppose that our revolutionary government generations ago wrote a new constitution that not only nationalized the electricity industry but also fixed the price. At the time, the price was fixed too low and so we have a history of electricity shortages. Happily, improved technology has turned the shortages into surpluses. Or rather, the marginal and average cost of electricity production is well below the fixed price.

    Many Market Monetarists are quite open to the possibility of denationalizing electricity. Unfortunately, our key spokesperson is really convinced that it is a natural monopoly. But we all agree that given the current set up (and cost conditions,) the government electricity monopoly should supply the quantity of electricity demanded at the fixed price.

    Boettke and Salerno insist that even though electricity shortages have caused all sorts of disruptions (reductions) in other sorts of economic activity, it is essential that no more electricity be produced.

    Really, the problem is one of relative prices. If all of the other prices and wages in the economy were much higher, then the fixed by constitution price of electricity would be a lower relative price and the electricity market would clear.

    OK, the analogy doesn't work perfectly. We all agree that adjustments in the price and wage level will bring the real quantity of money in line with the demand. Leaving the quantity of electricity the same when more or less should be produced cannot be solved by price adjustments alone.

    And, of course, central banks are really manipulating hookup fees for power, or more exactly, the price of some joint product of electricity that I can't think of right now. They supply that joint product to meet the demand, and that is what determines the amount of electricity they produce at the fixed price. In some rough sense they try to clear the electricity market, but because firms need less electricity when they produce the lower amount of output they can produce when they can't get enough electricity, it is clear that the electricity market always clears. So, they adjust the price of the joint product according to some formula related to the output of the products of electricity and the prices of those products.


  • George Selgin

    (Posted on behalf of Toby Baxendale):

    I am a creator of wealth.

    Wealth is only every created when an entrepreneur , seeing peoples needs and requirements not being fully satisfied , seeks to do this , by using the savings from prior productive activities , be that of his or of others , then getting together the available factors of production and putting them together in better combinations , to provide better products / goods and services / solutions to peoples needs and requirements . This all takes place over time.

    More money units or less does not change this process . It cant speed it up or slow it down in the long run. It can cajole and fool the entrepreneur into believing things that his customers don’t really believe in by causing mis signalling etc , it can bring forth production from the future if you like , but only temporary.

    I don’t think I have said anything controversial thus far , even to a very mainstream , booked based economist with little knowledge other than his books ( I am not suggesting this is you BTW) .

    If we hold substantial cash balances , we are refraining from consumption . In the UK corporate balance sheets have some £1.5tr of cash I believe and in the USA some double again . I know fund managers sitting on billions not knowing what to do with it. When a government bond in a "safe" jurisdiction is bought on 50-70 times in effect forward earning and what is more, you are promised faithfully you will get your principle back (as we can mint up as much as we like) , but you sure will get back a hell of a lot less purchasing power as well, AND when the cyclically adjusted PE ratio on the S&P is close to 24 x forward earnings with no noticeable increase in companies performance to warrant this, you can see why it is prudent to hold a cash balance . I submit there will be a correction . This might be the largest correction in history just around the corner. Only then will people lower their demand for money as prices will eventually re adjust. If we keep trying to hold them up by more nationalisation of large sectors of the economy , we can keep hanging on in there and maybe we will be able to do another 5 years before a correction – who knows!

    In the Promised Land of Free Banking of the fractional variety , no bank in their right mind would be accommodating more issue of media to buy more of this junk . I don’t doubt some would , but on the whole , they are a prudent bunch, especially if they are of the Kevin Dowd school of open ended liability style bankers.

    Now today , with banks impregnated with cash , they cant / won't lend it anyway , despite the fact they can issues many , many more loans on a vast and scary scale if they really wanted to max out the use of their reserves etc , which they could do. This is in the full knowledge , that you and me, the silly idiot called the taxpayer , will bail them out as well and they still don’t loan out more. You can take a horse to water , but you cant make it drink. This is what we have here. This problem would exist in the Free Banking Promised Land as it would in the world we occupy today.

    I would urge you at all points in time to resist the temptation to advocate second best when it will not even do what you suggest .

    The most important thing I ever say to professional economist re large scale falls of aggregate demand , the total emphasis on all policy should be with the new line of revenue you are faced with, be it a business or a government , is you must bring your costs down faster than your revenue line has come down. Restoration , not of the aggregate demand , which is not in your control, but your cost line to insure profitability and more of it than before , is the key measure, and the only thing the entire business and government enterprises should be focused on. That is a positive pro active policy response. More profitability is actually what causes more aggregate demand . In the final analysis productivity gains is the only solution.

  • A comment by Nicolás Cachanosky in Punto de Vista Económico

    Salerno, Selgin and Monetary Policy

  • Richard Schulman

    For an excellent defense of Say's Law, see Steven Kates, _Says Law and the Keynesian Revolution: How Macroeconomic Theory Lost its Way_ (2009).

    For a critique of early Sowell, in particular his affection for Malthus, Sismondi, and Marx — vs. Say — see

    • George Selgin

      Thanks for those references, Richard. The debate is certainly an important one. For my part, though, I must say that, if it is a matter of being or not being convinced by Say's arguments in chapter XV of the Treatise, then I count myself among those who remain unconvinced! In any event, though, there were many prior to Keynes, and not all of them lefties, who also weren't convinced, including many neoclassical economists.

  • John S

    "John, OK, let me explain my reservation about free banking with words. It seems to me that under any monetary regime the demand for the medium of account (gold or cash) will be inversely related to the nominal interest rate. If nominal interest rates change, so does the demand for the medium of account. Why should profit maximizing banks have an incentive to adjust the supply of the MOA in precisely the amount that would be needed to offset changes in the demand for the MOA due to interest rate variations? That doesn’t make any sense to me."

    Prof. Selgin, forgive me for playing the role of a tattler here, but perhaps you could answer Prof. Sumner's question better than me. Why would free banking tend to lead to roughly stable nominal income, in your opinion? Despite my best efforts, I'm having a hard time understanding the theory of free banking myself.