Nozick and Fractional Reserve Banking
Mar27
Money & Macro

Nozick and Fractional Reserve Banking

Rothbardian critics of fractional reserve banking (FRB) tend to use natural-rights-esque arguments, even when not explicitly invoking natural rights. That is, they take for granted not only the perspicuity of some definition of property, but also the obviousness of its application to any situation. Hülsmann, for example, argues that, “on the economic level, [FRB] implies right now a disequilibrium, because . . . [t]he members of our little community thus behave as if they could rely on more resources than really do exist.” This is not an argument from the economics of FRB, that it is disequilibrating; it is a claim that, because FRB violates the absoluteness of property (whose indivisibility is taken for granted), and regardless of the actual economic effects, it is by definition a disequilibrium.

He argues,

The main problem of fractional reserve banks is, of course, that they are virtually bankrupt, because at any point of time they have more cash liabilities than they have cash in their vaults. If too many customers demand redemption of their titles, the bank is doomed. One might grant that the bank tries to keep a sufficient amount of cash on hand at all times to satisfy demands for redemption. But exactly which amount of cash is “sufficient”? Because of the uncertainty inherent in all human undertakings, there is no cognitive way by which the bank can answer this question.

Whether Hülsmann is being deliberately obtuse here or whether he is sincerely unaware of the discipline of statistics, the fact remains that there is a way to determine how much cash is “sufficient”. Given the variance of adverse clearings, an x percent risk of a bankruptcy implies a definite reserve ratio. In Knight’s terms, this is not uncertainty at all, but quantifiable risk. And since trust is self-reinforcing, a bank with a proven record can manage a very low risk of bankruptcy with a relatively low reserve ratio.

The Rothbard-Block-Hoppe tradition of natural rights is oddly hostile to to the sort of probabilistic thinking necessary to deal with risk (which was the thrust of Caplan’s critique). But this is peculiar to this particular flavor, not a characteristic of natural rights in general. Robert Nozick, in chapter 4 of Anarchy, State, and Utopia, lays out the argument: let us suppose that the Lockean natural rights are indeed valid and perfectly perspicuous. Because of uncertainty, we still have the problem of their application to particular circumstances. Should I be allowed to engage in an activity that has a 50% chance of depriving someone (I do not know who beforehand, so I cannot contract with him) of his life, liberty, or property? If not, how about a 1% chance? Or a 0.001% chance? “This construal of the problem cannot be utilized by a tradition [e.g. the Rothbardian] which holds that stealing a penny or a pin or anything from someone violates his rights” (p. 75).

Nozick arrives on compensation for the imposition of risk as the solution.1 It should not be difficult to see the connection to FRB (viz. interest payments), but the application is far wider. “Mining or running trains,” Nozick argues (p. 74), “…presents risks to the passerby no less than compulsory Russian roulette with one bullet and n chambers (with n set appropriately).” This is at least comparable to, if not greater than, the danger of all depositors coming for their money at once by chance. And indeed, what activity at all does not carry some minuscule chance of causing harm to someone? “People would be willing to agree to a system that allows them to impose ‘normal’ risks of death upon each other, preferring this to a system that forbids all such imposing of risk.” (p. 76) Such a system is preferable precisely because it allows such innovations as mining, running trains, and fractional reserve banking.

It is possible, of course, to fall on the wrong side of a low risk. Indeed, the flipside of self-reinforcing trust is self-reinforcing distrust. Like in any industry, some banks will find themselves in crisis due to mismanagement and some due to unfortunate systemic events for which they should not be held liable. Hülsmann claims to be unable to see a difference between illiquidity and bankruptcy, as if it takes a systemic crisis to “unmask” the true fraudulent nature of FRB. The difference is between negligence and due diligence. A legal distinction rather than an economic one (but which will certainly take into account the economic distinction), as the normative question is what occupies us here. Naturally, a lower planned risk will be acceptable the graver the potential harm. But even if we admit that the difference is one of degree, it is still a significant and legally determinable difference, here as in thousands of other cases which could be overturned only at the expense of the entire common law paradigm.

The Rothbardian will object that his system can account just as well for contracting to deal with uncertainty. After all, Hülsmann concludes, “no law should suppress any foolish activity just because it is foolish.” But this is almost an afterthought to the Rothbardian system, something which makes allowance for risk and uncertainty only incidentally. “Well of course it should be allowed, if anyone wants to do it, but I still don’t like it.” By contrast, the sort of manageable risk exemplified in FRB, insurance, and really any activity at all if you get down to it, is at the core of the Nozickian system, where experience suggests it belongs. Whatever other gaps and problems it has, this fact alone strongly commends it over the Rothbardian system.

In the end, whether or not natural rights are a useful way of working through normative problems, the proponent must deal with them – first, explicitly and without smuggling them in under the guise of positive economics, and second, in a manner at least sophisticated enough to incorporate Nozick’s insights about their application under risk and uncertainty. Fractional reserve banking, as it turns out, is a good barometer for these.

Footnotes

  1. Which, one would think, should accord just fine with Rothbard’s own reconstruction of welfare economics along revealed preference lines.

Tags

Banking, Money, Natural Rights, Bryan Caplan, Guido Hülsmann, Murray Rothbard, Robert Nozick

SHARE

Facebook Google Plus Twitter Reddit StumbleUpon

Leave a Reply

More Content

About »

Hi, I'm C. Harwick, an economics PhD candidate in Virginia with an interest in monetary theory, institutional evolution, and folk music.

Care to know more? Read on »

Twitter »

    • Apr2622:15
      Bill Nye is no eugenicist. No, it’s worse than that: he’s a dysgenicist.
    • Apr2622:07
      RT @Faris_dream: I’m going to translate some tweets in this hashtag #مرتد_حفرالباطن the world must see this.


Design By Cameron Harwick Powered By Wordpress Hosted By Nearlyfreespeech No Copyrights