Confidence in the Pyramid
External Piece
Ivan Aivazovsky - The Great Pyramid of Giza

Confidence in the Pyramid

Speculation, Calculation, and Stabilization in Hayek’s Monetary Dilemma

Winner of the Mont Pelerin Society’s 2012 Hayek Essay Contest

The debacle of the interwar gold standard left a rising generation of economists wary of inflexible monetary rules, much as reflection on the Great Depression has left central banks wary of any amount of deflation. Hayek in his essay Monetary Nationalism and International Stability takes a more sober look at the former, briefly touches on the latter, and identifies the specific failures of the interwar gold standard – failures which are not, he maintains, to be solved by floating fiat currencies. This paper will first look at how the two institutional alternatives he proposed – a national central bank and free banks – purport to remedy these failures, focusing on speculation as a stabilizing force. Second, it will draw from the larger body of Hayek’s work in evaluating the two alternatives by their effectiveness in utilizing information and promoting stability, revealing as well from this example the indispensible economic foundations of Hayek’s later political philosophy.


The principal problem with the historical gold standard was what Hayek refers to as the perverse elasticity of bank deposits – the fact that the supply of deposits is “inversely affected by the demand for a more liquid type of money” (Hayek 1937, p. 80). Though governments no longer worry about maintaining internationally liquid specie, and though the more acute symptoms of the business cycle have been somewhat alleviated due to confidence-boosting measures such as deposit insurance, perverse elasticity remains a fundamental feature of today’s monetary systems. If at any point a large number of currency holders lose faith in a central bank’s ability to maintain the value of its currency, the bank will be forced to severely contract the money supply.

Speculative pressure, Hayek claims, can be either stabilizing or destabilizing (Ibid., p. 64), depending on the expectation of stability – that is, depending on the credibility of the promise being speculated against. In the case of a bank, that promise is of the ability to convert its liabilities into more liquid media. Distrustful speculation – if sufficient liquidity is thought to be unavailable – then culminates in a run. To the extent that people are aware of the standing of their banks, a bank which would be devastated by a run can certainly expect one, whereas a bank which could weather a run is not likely to have one.1

This is essentially the problem of the perverse elasticity of bank deposits. So long as we enjoy the benefits of a flexible money supply by pyramiding relatively illiquid moneys on top of relatively liquid money, the recall of a portion of the relatively more liquid money beyond what lies in reserve will necessarily cause a greater contraction in the relatively illiquid media. Interestingly, Hayek does not advocate for a 100% reserve system, which would obviate this problem by way of a rather inflexible money supply. Neither of his two solutions – free banking and a global central bank – necessarily do away with pyramiding in principle, yet both come out more stable than his current system of national central banks, because both are capable of instilling sufficient confidence as to evoke stabilizing speculation for individual banks.

Hayek’s proposed solutions approach the problem from two different ways in which national central banks fail to instill confidence: in competition, and in calculation.

Free Banking and Economic Calculation

National central banks are, in their mandates, generally free from the necessity of making a profit and guarding against loss except so far as is necessary to prevent complete insolvency. Economic calculation is therefore not available as a tool to guide policy. There is no mechanism to transmit to their policymakers changes in the demand for liquidity, especially where the central bank has no necessity of maintaining a reserve. So far as they become aware of the necessity of changing policy, their response is already tardy (cf. Selgin, p. 102f). Furthermore, even given the necessary direction of adjustment, the magnitude of adjustment is hardly better than a shot in the dark. And finally, to the extent the bank is (formally or informally) accountable to the people or elected government, the asymmetry of the effects of erring on either side – deflation is much more immediately painful than inflation – leads to a systematic inflationary bias.

This pattern of tardy trial-and-error leads, no less under flexible than fixed exchange rates, to periodic booms and busts. Changes in the demand for liquidity proceed much more quickly than the central bank can become aware, so when this change is drastic enough – at the collapse of a bubble, for example – note holders lose confidence in the value of the relatively illiquid assets, leading to the scramble for liquidity that contracts the total money supply before the central bank can react – or even in spite of its reaction.

A system of free banking, on the other hand, as a competitive market among legally equal banks concerned with profit and loss, enables economic calculation and thereby systematically prevents the missteps associated with monopolistic note issuers. In the first place, freedom of note issue flattens the tier of the pyramid between banknotes and deposits (Selgin, p. 111f).2 Such banks are thus relieved of the necessity of altering the total money supply in response to changes in the demanded ratio of deposits to circulating currency. In addition, with the existence of clearing arrangements, over- or under-expansion of credit relative to demand will harm a free bank’s bottom line much more immediately than its effects become apparent to a central bank.3

A free bank facing solvent local competitors will be hard pressed to survive the actions central banks often take to correct their errors. Devaluations and terminations of convertibility are hurtful, but not necessarily fatal, to a monopolistic central bank. They would, however, drive customers of a competitive bank elsewhere. Thus competitive banks, in addition to being better suited than central banks to prevent the errors that put them in the position to take these measures, are also better motivated to do so. And finally, if a free bank does indeed make a fatal misstep as must sometimes be expected where human knowledge is imperfect, it will not create the sort of systemic national panic or economic downturn that characterizes the failures of national central banks. On the contrary, in the absence of fraud, such failures are much more likely to happen with a whimper than with a bang, and without nearly so much capital misdirection as is caused by the missteps of a central bank.

Economic calculation therefore is the principal tool by which free banks – not indeed individually, but in the context of their competitors – create a regime in which stability is the norm, and is expected to be the norm. Thus invoking stabilizing speculation, free banks are not only better equipped to regulate their respective money supplies, but also need not even work so hard at the task as a central bank.

A Global Central Bank and Competition

A global central bank, on the other hand, is in no better position than a national central bank to manage the money supply by economic calculation. Its mandate is similar to, and its monopoly greater than, that of a national central bank. It cannot prevent similar economic missteps, but by virtue of its global monopoly, it can nevertheless take advantage of stabilizing speculation.

It is a familiar economic principle that demand for a particular good becomes, ceteris paribus, more elastic the easier it is to substitute for it. This is no less true for monies. Thus with more competition, consumers are more sensitive to missteps: fewer are tolerated before speculation becomes destabilizing. National central banks find themselves in the unfortunate position of being unable to prevent missteps, yet facing enough competition via foreign exchange to destabilize them in the event of a great enough misstep. But where a system of free banking would prevent missteps, a global central bank evokes stability by making demand for its currency nearly perfectly inelastic.

Because a global central bank faces no possibility of external monetary shocks – having no competition as well as the legal power to prevent the emergence of alternative media of exchange4 – it is free to regulate the supply of money without the necessity of maintaining foreign reserves. And if it should issue an inconvertible fiat currency, it will also be free from the risk of a run, there being no foreign alternatives and nothing more liquid than its banknotes. Even if other banks should pyramid their own deposits on its banknotes, the demand for currency (and the composition thereof) will be sufficiently inelastic that unless the central bank should manage its currency so poorly as to trigger a widespread flight into goods, individual banks will face a relatively stable demand for currency and liquidity. The global central bank will therefore be largely free from having to make many of the decisions that for national central banks become inevitable missteps.

Furthermore, with neither alternative media of exchange nor a more liquid medium into which the central bank is bound to convert its notes, there is no way to speculate against the central bank short of flight into goods. The confidence that national banks and governments attempt to instill with programs such as deposit insurance and bailouts is therefore perfected in a global central bank. With no reason to expect that a sudden change in demand for liquidity should force a bank to contract the credit superstructure, speculators will be a stabilizing force for any bank that can draw reserves from the central bank.


Hayek’s own philosophy, though largely developed in the decades after the publication of Monetary Nationalism, was certainly not agnostic as to which of these two solutions was ideal. Even without such strong and specific statements as he later makes in A Free Market Monetary System (1977) that “if we ever again are going to have a decent money, it will not come from government,” the lines of his thought clearly point in the direction of free banking.

Central Banking and the Rule of Law

One might be tempted first to answer the question with reference to Hayek’s later political philosophy. Central banks, national or not, require at the very least coercive legal tender laws to be at all effective. The goal of Hayek’s political philosophy was “to minimize coercion or its harmful effects, even if it cannot eliminate it completely” (Hayek 1960, p. 12), clearly giving free banking a prima facie advantage over a global central bank.

However, the second clause of the sentence is telling. Hayek was no anarchist and held no ideals of completely eliminating coercion. Indeed the following chapters are primarily occupied with the question of minimizing the negative effects of coercion by making them general and impersonal – that is, subject to the rule of law. Legal tender laws, though coercive, have historically been for the most part general and impersonal.

There is perhaps a question as to whether a law privileging a private bank (e.g. the Bank of England) would comport with the rule of law according to these criteria. However, this is remedied easily enough and without changing the important economic features by making it an official arm of some international governance organization, even if it should nevertheless be shielded from electoral pressure. In such a case the central bank would be no less capable than any other government branch of operating in accordance with the rule of law. In short, central banking does not present a unique problem with respect to Hayek’s doctrine of the rule of law.

It may of course be objected that the rule of law is a constraint, not a sanction – that Hayek would not support every measure duly passed according to the rule of law. Nevertheless, even a broader constraint based primarily on Hayek’s idea of coercion could not rule out a global central bank and remain recognizably Hayekian. The establishment of a central bank would certainly be a legitimate use of coercion if its economic necessity were clear. The answer must therefore be sought from Hayek the economist, rather than from Hayek the political philosopher.

Money as a Technical Problem

A more fruitful first step in the economic direction is Hayek’s idea of the type of solution to be sought for. It is one thing to theorize on ideal monetary arrangements, and quite another to theorize on a framework in which the ideal may be found.5 From the fact that he hoped for “the discovery of yet unknown possibilities in currency” (Hayek 1990, § XXIV), we can see the esteem in which he held attempts to arrive at the answer a priori, even if he did make a tentative attempt himself (Ibid.).

To say that money is a category of human action is not to say that there are clear lines distinguishing among money, money substitutes, and non-moneys. Rather, various goods are used as media of exchange according to their liquidity, which takes account of heterogeneous criteria of usefulness such as divisibility and stability of supply, much as heterogeneous aspects of a class of items can find a common denominator through the emergence of a market price.6 The pyramid of credit to which Hayek refers (and which might more aptly be called a ziggurat) is unnatural so far as it has taken on a stepped shape due to legal rather than economic factors.7 Thus what initially seem like natural divisions offer no conceptual help in formulating a specific monetary ideal. Indeed, without these divisions it becomes exceedingly unlikely that a satisfactory arrangement can be arrived at without the benefit of experience and experimentation. What twentieth century economist could have foreseen the emergence of Bitcoin, a purely abstract unit of account whose supply is regulated by computational difficulty?

Thus with money, as with most institutional decisions, the question which from our vantage point immediately concerns us is not what is to be done, but how it is to be decided (cf. Sowell, p. 79). Our answers to the first question are necessarily provisional, especially considering the variety of proposals and predictions as to particular outcomes under both systems. However, it will be useful to first consider both alternatives as questions of the second type – as questions of method rather than of outcome.

Public Pressure

Seen therefore as a question of method, the problem of public opinion becomes relevant at this point. When a good is monopolized, bargaining power between its supplier and consumer is highly unequal, and dissatisfaction at outcomes gains a focal point. Whether de facto or de jure, monopolies face the pressure of opinion more strongly relative to market pressure than do competitive firms. For central banks, the mistrust arising from this inequality is somewhat softened by the dual mandate, as opposed to a purely self-serving profit motive. Nonetheless the advantage enjoyed by a global monopolist over monetary policy would constitute a strong focal point for dissatisfaction, from which no distance from the ballot box can entirely shield it.

The introduction of public pressure as a variable changes the optimal institutional arrangement by forcing a tradeoff between stability of expectations and flexibility in policymaking (Selgin, p. 167) – between rules and authority, as the debate went. “We shall,” of course, “never prevent the abuse of power if we are not prepared to limit power in a way which may occasionally also prevent its use for desirable purposes” (Hayek 1940, p. 242). Such inflexible rules were at the same time the primary virtue and the chief shortcoming of the gold standard: they were perhaps economically inefficient, but prevented greater political abuses (Hayek 1937, p. 75). Since its collapse we have traded stability for flexibility, mitigating indeed some of its more obvious flaws, but introducing others by exposure to the inexorably inflationary pressure of public opinion. And without a system of fixed exchanges, we are faced now with “the impossibility of prescribing any fixed rule” (Ibid., p. 91) – or at any rate, stability has become more costly in terms of flexibility.

Competition, however, significantly diminishes8 the bargaining advantage a bank enjoys over its customers, and with that advantage goes the prima facie case for the curtailment of the profit motive by regulation. Inflationary public pressure on individual banks will be further limited by the fact that, as it will be beneficial to spend an inflating currency but detrimental to accept it, no one will desire to hold a currency which no one will accept. And even with some degree of pressure, as we have seen, competitive note-issuing banks will not be able to unduly expand credit without incurring losses to their own reserves.9

It being clear then that political pressure will force a global central bank to be generally inflationary, and competitive pressure will prevent a system of free banking from being inflationary, the question of method reduces into to a question of outcome with respect to inflation, while still allowing us to remain agnostic with respect to many other features about which there is less consensus. If inflation is on balance beneficial to political or economic stability, then the virtue of a global central bank is that it overcomes a collective action problem, and the introduction of public pressure is not so detrimental as the complete removal of inflation as a policy option. Conversely, if inflation’s harm outweighs its benefit, then the virtue of free banking is not only in its systemic prevention thereof, but also in obviating the tradeoff between flexibility and stability.

Inflation and the Knowledge Problem

It need hardly be said that Hayek’s own position on inflation was strongly negative (cf. Hayek 1960, ch. 21). The chief danger of inflation, he warned, is that it makes economic calculation unreliable, and furthermore that it is extremely painful to correct once begun. Yet it is at least conceivable that, as national monetary policy often aims to help its own interest groups by harming the interests of competing segments under other currency areas, the absence of these other areas would put a global central bank in a better position than national central banks to prevent runaway inflation by dampening the parochial element of public pressure that would lead to competitive devaluation.

Political stability is thus, the argument goes, well bought at the price of intermittent capital misdirection, especially being cheaper under a global monetary regime than under national monetary regimes. What public pressure does exist is, if not benign, at least better relieved through monetary measures than allowed to build indefinitely.

This claim, like its opposite, is a prediction about the future. Whether or not the long-term economic instability generated by inflation under a global central bank is preferable to the political instability arising from the immediate dissatisfaction of those segments desiring inflation will depend in large part on cultural factors. The best institutions cannot preserve a well-functioning market in a culture holding strongly anti-market ideals. The difference in perspective is in some measure due to the inflationists taking cultural factors as given, working only within the realm of “political possibility,” whereas Hayek aimed at changing those factors.10

It is impossible to say with certainty whether or not some optimal amount of inflation will be worth the economic disturbance in a given cultural context. It will therefore be likewise as impossible for the central bank to determine the optimal rate of capital misdirection as to eliminate all misdirection. It is the very nature of this misdirection that the missteps which cause it are unknowable for the central bank without the aid of economic calculation. How much less will it be able to arrive at a socially optimal level of misdirection where it missteps deliberately?

Econometric analysis is no substitute for economic calculation either. It is “of use only where we either deliberately ignore, or are ignorant of, the relations between the individual elements” (Hayek 1967, § 4). This is, of course, precisely the problem in determining the effects of inflation, one which is glossed over in treating it simply as a rate. For today’s national central banks, it serves chiefly to lend a veneer of objectivity to their policies, giving the illusion of a rule and thus shielding them to some degree from public opinion (cf. Hasnas 1995).

Where its mandate is stability and it possesses the requisite legal privilege to relieve it of the necessity of making a profit, there is no market mechanism whereby heterogeneous factors such as public opinion and reserve balances can be factored into a measure of profit or loss. As the history of democracy makes abundantly clear, the pull of competing interest groups is no substitute for economic calculation in determining whose interests to prioritize. Even if it were the case that political stability is worth the economic turmoil of some small amount of inflation, it is still politically self-perpetuating, economically destructive, and practically uncontrollable once it has begun. There is no reason to expect even a global central bank not to overshoot whatever optimum may exist, even if it will do so less egregiously than national central banks.


The consonance therefore of free banking with Hayek’s overall philosophy extends far deeper than his aversion to coercion. The preceding discussion of an international monetary system, though, exemplifies the economic foundations of Hayek’s political thought. For indeed, just as coercion would no doubt be justified for Hayek in establishing a central bank if the economic benefits were clear, the political case against totalitarian coercion would be moot if not for the economic case against central planning. Free banking is not an ideal for Hayek because it minimizes coercion; rather, the minimization of coercion is an ideal for Hayek because it allows for the use of vast amounts of vital but often unarticulated knowledge in a spontaneously ordering system. Free banking is simply an instance of this: a microcosm of the larger economic and cultural order embedded within and buttressing it with a stable medium of exchange.


  1. This is not to deny the existence of exogenous psychological factors in bank runs, but real factors constitute the primary trigger: bank runs are not primarily random events (as claimed, for example, by Diamond and Dyvbig). Selgin (p. 193) points to (Gorton) for empirical evidence.
  2. Hayek’s own proposal for free banking in Denationalisation of Money differs somewhat from others such as Selgin’s, but they share broadly stabilizing tendencies for largely the same reasons.
  3. Via what Selgin (p. 40) calls the Principle of Adverse Clearings. Cf. also Hayek 1990, p. 63.
  4. On the importance of legal tender laws for preserving banking monopolies, see Hayek 1990, § V.
  5. “The spontaneous character of the resulting order must therefore be distinguished from the spontaneous origin of the rules on which it rests, and it is possible that an order which would still have to be described as spontaneous rests on rules which are entirely the result of deliberate design.” (Hayek 1973, p. 45f) Thus Hayek did not blush to propose deliberately designed rules under which broad order could emerge.
  6. That is, a good’s liquidity takes account of these qualities in the same way that the market price of a particular car reflects such heterogeneous qualities as mileage and color.
  7. The “smoothness” of free banking with respect to liquidity is emphasized most strongly in Greenfield and Yeager’s (1983) Black-Fama-Hall proposal. Curiously however, they claim this amounts to the abolition of money, calling their proposal an advanced barter system that “avoid[s] the disadvantages both of crude barter and of money as we have known it.”
  8. I do not say eliminates because so long as a market remains imperfectly competitive – that is, so long as there is any differentiation between the products of different firms – public pressure can still be corralled and exercised. See for example the pressure on Apple regarding its use of Chinese labor.
  9. The Principle of Adverse Clearings by which losses are inflicted on overexpanding free banks depends for its operation on consumers differentiating among note-issuers – that is, on an imperfectly competitive market. Thus, as consumers exercise public pressure by differentiating among banks, these banks banks are pari passu more strictly prevented from overexpansion.
  10. The preface of Monetary Nationalism in which Hayek discusses this issue is reminiscent of themes in his later essay The Intellectuals and Socialism (1949).


BankingBargaining PowerCryptocurrencyFree bankingMacroeconomicsMoneyRule of LawF.A. HayekGeorge SelginJohn HasnasThomas Sowell


Facebook Twitter Reddit Threads

Leave a Reply

More Content