Believe those who are seeking the truth. Doubt those who find it. Andre Gide

Sunday, May 9, 2010

Central Banks as Sources of Financial Instability

George Selgin has written an interesting piece here, explaining why the institution of central banking is inherently destabilizing (relative to an institution of free-banking). I have a lot of sympathy for what he has to say here, although I do have some reservations. (Thanks to Prof J for sending this to me).

Selgin begins by describing how a free-banking systems works. Competitive banks issue paper liabilities made redeemable on demand for specie (historically, these paper notes also constituted senior claims against the bank's assets in the event of bankruptcy; i.e., they were backed beyond the available specie held in reserve). Banks would create these notes whenever they wanted to make a money loan. The money loan would be extended against sufficient collateral (hence, the paper issued by the bank was backed, at least in part, by the collateral backing its loan portfolio).

Question: what prevents individual banks from "over-issuing" the paper money that they can evidently create "out of thin air?" Answer: the principle of adverse clearing. What is this? As banknotes circulate, a portion are redeemed for specie (either directly by people, as when you or I visit an ATM, or by other banks who have received deposits consisting of other banks' notes. This continual flow of redemptions is what keeps the supply of credit in check: banks that issue too many notes will have trouble honoring their redemption obligations, leading to bankruptcy.

I have some sympathy for this argument, but I wonder whether it is entirely correct. One might note that Microsoft shares are not made redeemable on demand for anything (although they are backed by capital). What prevents Microsoft, or any other company for that matter, from "overissuing" its debt obligations? There is no principle of adverse clearing at work here.

A central bank is defined by Selgin as an agency that has monopoly control over the supply of small denomination paper notes. As Selgin rightly notes, the monopoly was usually conferred by resource-hungry sovereigns. Private banks would deposit their specie with the central bank, and issue liabilities (checkable deposits) redeemable in central bank money (rather than specie). Or something like this...practices likely varied across space and time. In any case, the point is that the central bank is no longer subject to the force of adverse clearing.

If I understand his argument correctly, all I think he is saying is that the central bank is now in control of the object of redemption (specie in free-banking). There is very little economic discipline on how a central bank manages the base money supply. It can overissue, leading all banks who make their liabilities redeemable in central bank money, to overissue as well. A credit bust inevitably follows the resulting boom in credit.

Well, I suppose. On the other hand, an independent central bank might choose to keep the supply of base money in check. What is the difference now between bank liabilities made redeemable in specie or central bank paper? In fact, the latter might be superior in that it frees the specie from its role as idle reserves...the gold and silver can now be used for other purposes.

Of course, central banks have behaved badly in the past. But this has almost always been because of pressure from the fiscal authority. Even a free-banking system is not free from the grabbing (and destabilizing) hand of the state.


  1. re Microsoft shares: this kind of behaviour would be reflected in the share price, no? In the same way, bad behaviour by a single note issuer would result in its notes being discounted. This prevents the grabbing hand of the state from causing too much trouble, until it starts to regulate all banks. At some point, this would lead to instability and a new central bank (as with the creation of the Fed).

    I suppose your question is which system can remain more 'independent' of opportunistic government meddling. I don't think Selgin and White have given this question much thought.

    BTW, Laidler has argued the Bank of Canada has largely mimicked the behaviour of Selgin's hypothetical free-banking system over the last 20 years. See

    Pedro Bento

  2. Pedro:

    Yes, the presumption is that bad behavior will be reflected in Microsoft share prices. Why can we not make the same presumption concerning bank shares? Demandable debt probably had a role to play as a disciplining device back when the market for bank shares was not well developed. But whether the argument continues to hold water today is debatable. See also Calomiris, Charles W & Kahn, Charles M, 1991. "The Role of Demandable Debt in Structuring Optimal Banking Arrangements," American Economic Review, American Economic Association, vol. 81(3), pages 497-513.

    I think at root the argument has to be that free banking is relatively less prone to government meddling. Presumably, this is because to extract resources from the private sector in this case would require direct taxation, something that is politically costly. The inflation tax, on the other hand, seems to be politically more expedient. Maybe.

    Thanks for the reference to Laidler; I will take a look.

  3. Pedro,

    I don't think you need to make the step to the stock price, although overextension of debt will have an effect on the cost of equity. But extending debt will increase the cost of debt for the firm first. As the firm tries to pile on a greater amount of debt than optimal, the cost of debt and equity will rise for the firm to the point where no projects would have a high enough expected return to be acceptable to the firm. Although, one thinks that prior to this it would become increasingly difficult for the firm to place its debt.

    It is the case, though, that demand deposits have no true counterpart in non-bank debt. The closest might be commercial paper, since it is short term. But the trick is, very little debt can be traded back in to the company at will for redemption prior to the maturity of the debt.

    There is one type of debt that is similar to demand deposits, and I think it shows a good way of understanding demand deposits: puttable debt. This debt can be sold back to the issuing firm (put) at the option of the debt holder. The trick is that the put option typically only becomes available after the debt has been outstanding for a while.

    So, a demand deposit is indefinite-term, no-interest bearing debt that is puttable immediately upon issue.

    I have to think about the other things you mentioned, David. I think this is a really interesting area, and I'm keen to do some work on how a free banking system would affect corporate finance choices.

  4. Prof J:

    I'm glad you noted that demand deposit liabilities are (American) put options (redeemable at par for cash).

    Casual empiricism suggests that the APO is a desirable property to embed in objects that are destined to serve as a medium of exchange. Why this is the case, I am not sure. It may have something to do with fixing the exchange rate of different monies.

    If all banks issue competing monies redeemable at par and on demand for cash, this results in a de facto fixed exchange rate system between different bank monies.

    This is also how currency boards fixed exchanges rates (e.g., by making a Peso redeemable on demand and at par for USD).

    But I haven't yet fully sorted these ideas out.