www.JoeCobb.com

Joe Cobb
Staff Economist, Committee on Banking (1983-85)
U.S. House of Representatives

[ this paper was presented at the Center for the Study of Public Choice, George Mason University, February 15, 1984 ]

The public choice literature has dealt for some years with the issues of monetary policy and the case for a monetary constitution.[2] In this paper we shall take for granted the desireability of a “monetary constitution,” in contrast to an institutional framework in which the king, or his omniscient economic advisor, or the prudent and cautious governors of the central bank, have full discretion to carry on “with firmness in the right as God gives [them] to see the Right.”[3]

The purpose of this paper is to describe a model that could form the basis for a monetary constitution in today’s context. The central idea is to make use of what we know about the free market process to constrain existing insitutions. It relies particularly upon judgments of “opportunity cost” by individuals in a micro-economic framework to suggest an equilibrium solution to the question:

How can the supply of real balances be exactly and continuously matched with the demand for real balances in such a way that exogenous shocks originating either from deliberate monetary policy exploitation by politicians or mistakes by their economic advisors would have no disturbing consequences for employment, nor any tendency to expropriate savings?

The present, unconstitutional system is characterized by uncertainty and inflation, as well as the frequent deviation of economic growth from its long-term trend. It seems much more costly than some ideal alternatives we might design, and this task has indeed fascinated economists for over 200 years.

Two Misleading Proposals for a Constitutional Rule

Discussions of a monetary constitution should properly focus on some process that can ultimately govern monetary policy in a political environment. The seach for a monetary constitution should start with the question, “What are the opportunities or alternatives available to us?” If there are none, then Gordon Tullock’s challenge - that our present system is the best possible result of a “political market” and may be the least-cost alternative - will be validated.

Economics measures the opportunity cost of flying from New York to Washington by an individual’s alternatives of taking bus or train, driving, walking, or not going at all. The opportunity cost spectrum, however, cannot include the obviously superior methods of transportation used by Peter Pan and Tinkerbell (thinking happy thoughts) or as described in science fiction stories, broadcasting oneself through a “matter transmitter.” Unfortunately some proposed models of a monetary constitution, particularly the two most familiar to us, suffer from exactly this problem. They do not represent “paths not taken” because they are not paths that could be taken: they do not represent genuine alternatives.

One proposal for a monetary constitution that receives much attention by economists is the Henry Simons/Milton Friedman version, in which the monetary authority is mandated by some rule to maintain the stock of money, or its growth, to a fixed (or knowable) formula. In its most pure version, the formula alone is the monetary authority; there are no human judgments in the system - except perhaps the initial research to compute a fixed growth rule and to choose whether “society would prefer” an unchanging price level, or one that mildly declines or rises over time.

I like to characterize this proposal for a monetary constitution as the “Buddhist Heaven” version, because it is suppose by its faithful to be perfect in every way; there is no deity; and the elected officials who inhabit such an economy have no earthly, political desires to play around with the monetary system. Economists are good at analyzing human behavior in markets, but are they guilty of “fighting the last war,” or of using oversimplification of complex phenomena as a basis for policy recommendations?

Certainly politicians are excessively concerned about short-run effects, and enshrining a fixed formula in place of the Federal Open Market Committee would lengthen the time horizon. Yet, would it guarantee monetary stability? Or would the formula itself become an object for political attack, with one party demanding it to tilted up, another that it be reduced, leaving businessmen throughout in an uncertain monetary regime? How frequently should the formula be re-constructed to adapt to institutional change, or international events?

Another version of a monetary constitution is commonly discussed. It is proposed by the school of supply-siders who would like to see a new Bretton Woods conference fixing international exchange rates, or the adoption of a classical gold standard to regulate the money supply “automatically.” I call this kind of monetary constitution the “Garden of Eden” version, because its claim is that life was so much better in the old days that we must purge ourselves of monetary original sin.

Yet, was there ever an “age of innocence” in banking and finance, when politicians and central bankers played acording to the rule of the “gold stardard game”? Many students of money argue the classical gold standard never really did operate according to the Hume-Ricardo model, not to mention over 2,000 currency devaluations during the Bretton Woods era. Much of the controversy over the role of gold in the monetary system is based on the vehemence with which it is asserted that the classical system must be the solution to the defects of the modern one.

Both the “Buddhist Heaven” and the “Garden of Eden” models have to be ruled out as possible monetary constitutions today, not because implementation would be “politically difficult” but because both models deny the reality that what one man designs, other men can change. A constitution has to be more durable than the next election or the next trendy intellectual fad.

The Case Against the Economist-Utopians

In the previous century there was a belief among bankers and financiers that money had to be connected to real reserves - real bills or real commodities. The financiers believed in a commodity standard because the common people believed in it. Of course the “unsophisticated” still seem to put a lot of faith in gold and silver coins today, which is the principal reason tangible assets, particularly those with some liquidity, are so popular during periods of financial instability.

Economists from Jevons through Keynes - including virtually all the famous ones - systematically attacked the idea that fixing a par value for national currency in gold need either constrain the monetary authorities against error or guide them to optimal actions. It has become a common joke among economists that gold was merely an inconvenient decoration.

It can take a long time to move so-called “informed opinion,” but it can be overturned. At the Genoa conference in 1922 and the Bretton Woods cofnerence in 1944, the old banker’s theory was still strong enough to produce a semblance of the classical gold standard, but by 1971 the economists had persuaded the people who control monetary institutions - those who have something to say about managing large pools of capital, and therefore the most to lose - that scientific economics offered a better way to conduct monetary policy: in the style of Henry VIII, who was the first king to debase English money.

Advocates of the “Garden of Eden” proposal, however, seem to expect almost every academic economist and Wall Street professional to acknowledge that he knows much less than he believes. Every money manager has a college degree today. They have been taught that gold is a silly anachronism; they firmly believe there is a benefit in scientific planning or management; and they casually assume this is what the boys at the Treasury or the Federal Reserve mean by the words “monetary policy.”[4] Mark Twain is supposed to have said once, “it is not the things you don’t know that can kill you - it’s the things you ‘know’ which ain’t so.” Like Adam and Eve, we can’t spit out a bite of the apple, even though the situation we now have to cope with may be deadly. A myth dispelled cannot be resurrected, no matter how healthful.

As for the “Buddhist Heaven” proposal, it is certainly not useless or foolish for professors to teach students about the relationships among monetary policies, price indexes, business cycles, and economic welfare. Students, after all, can go on to become investment advisors and help people escape the vicissitudes of monetary policy, but it seems curious to expect any of that knowledge to lead to institutional reforms or a “monetary constitution.” We have gotten into the present dangerous, unconstitutional situation because economists convinced lawmakers that the science of money is technical. Ipso facto, as long as technicians can keep a debate going, the institutions will not be significantly modified or placed under a monetary constitution. The debate plays every night on television.

One of the unavoidable details of democracy (and more so, bureaucracy) is political activity in pursuit of self-interest. It is in nobody’s self-interest to lock up the monetary power, never to abuse it again. There is no decision process in a democratic society capable of creating a dispassionate unity in favor of the monetarist rule. In “Buddhist Heaven” there is no resident deity, but in our mundane situation there is a temple on Constitution Avenue in Washington with many active sons-of-god in residence, and there are monasteries in New York, London, and every other world capital. These religious leaders are not going to convert to any form of Buddhism requiring them to subscribe to poverty and abnegation. When the Federal Reserve is criticized, it becomes cautious to hide its power - but it does so to preserve its power, not to relinquish it.

Constructing an Opportunity Cost Model

The difference between the proposal presented below and the discussion above is that the process constraining monetary policy - once it is set into motion - would be a free-market phenomenon that would continue to operate unless the government violently suppressed it. There is reason to hope that a potential combination of political and economic resistance to any subsequent government attempt to suppress “monetary freedom” would instead bias the policy in favor of respecting and preserving it, just as with freedom of speech and religion. Thus, the opportunity costs to those politicians who would exploit monetary policy hopefully could be raised to unacceptably high levels.

The proposal in this paper is a parallel standard with two forms of base money. One form is exactly what we have today: liabilities of the central bank, both in the form of circulating banknotes of various denominations and clearing reserves held on deposit for commercial banks. The second form of base money would be gold coins denominated only by untis of weight and not in terms of the other currency.[5]

There would be a floating exchange rate between gold ounces and Federal Reserve accounting units. There is no particular reason to assume that all gold ounce coins should be issued by the government. The marginal costs of mining and minting would set a constraint on supply. The structure of the banking system would be the same as we have today, with whatever additional elements of deregulation might be needed to accommodate demand deposits and other financial services in units of “gold ounces.”

Assume that this parallel standard has been in operation for a few years, sufficient for any transitional factors to have disappeared. For example, initially there would be “announcement effects” and a period during which the Treasury would be primarily a seller of bullion or an issuer of gold coin from newly minted supplies. During this phase, the price of gold would not fluctuate significantly and there would undoubtedly be a depressing effect upon gold mining. This last detail should draw support for the proposal from critics of South Africa and Soviet Russia.

As a reserve asset, gold ounces ought to be as widely dispersed as possible.[6] When gold holdings of the Treasury were significantly reduced, it should be willing, on a passive basis, to make or accept payments in gold ounces instead of dollars on a freedom-of-choice basis, at the election of the transactor. Since the dollar value of gold coins could change, some policy should be formulated to allow independent, tax-free determination of the market exchange rate applicable in each case. More than any other de-regulation, this tax treatment would assure the high elasticity of substitution between gold coins and dollars. There would be some technical questions about how the government itself should keep its ledgers, and how Congress should specify appropriations.

The public’s demand for cash balances would be partially satisfied by Federal Reserve accounting units and partially by gold-ounce accounting units. The major share of the total money supply in each denomination would probably be, as today, in the form of deposit accounts (some would bear interest and some would not) with the balance in the form of base money - gold coin and Federal Reserve notes - in the physical possession of the public.

The Demand and Supply of Money

One departure from conventional analysis in this model is to avoid a scalar “monetary aggregate” approach. Intuitively, in the real world “money supply” is not strictly a mathematical linear combination of the various denominations, except as economic models make it so. The demand for cash balances can be expressed as a vector, [x, y, z, ...], with each element representing a utility function for some specific type of monetary instrument.

Consider that each individual in our model has a demand for cash balances structured like a diversified portfolio. In portfolio theory, individuals hold a blend of assets with different perceived benefits or the same benefits in different degrees. It is not always the case a one-dollar banknote is an alternative for four 25-cent coins. Demand deposits are poor substitutes for banknotes in some cases, as when you are far from your local bank.

The form of a monetary instrument has a utility component independent of the additive property of money. The demand for money is not just a demand for an amount of consumable value in liquid form, but also a demand for particular forms of liquidity, satisfying different expectations. For example, we might want dimes and quarters for public telephones, banknotes for taxicabs and toiletries, and checkbooks for payments by mail.

Let the money supply also be represented as a vector, [X, Y, Z, ...]. For example, let X be cupronickel dimes in physical possession, Y be gold ounce coins in physical possession, Z be demand deposits at zero interest in a local bank, etc. All of the components have a nominal fixed value, but in terms of utility the values of components have a floating exchange rate, governed by the opportunity cost of time and travel to some place, like a bank, where an exchange can occur. With deregulation, some money supply components now have explicitly floating rates - in the form of various yields on deposits or transactions fees on bank services.

We will assume the relative utility configuration of the money demand vector arises out of an individual’s wealth, circumstances of daily life, age, and tastes - factors not subject to whim or volatility. The extent to which a component of the money supply vector, however, satisfies the demand can change quickly. Just as an individual might watch a stock market portfolio and adjust his holdings as the market changes to keep his relative utilities for risk and return in balance, he will handle his money the same way. As an example, if two banks offer checking accounts but one pays interest on your daily balance and the other doesn’t, you might go for the yield; but if there were a run on that bank, you would want to put your hands on cold cash and open an account in the other bank.

Exogenous Shocks and Initial Market Response

Now consider that the Federal Reserve Open Market Committee purchases Treasury bills from the public and expands the total supply of base money denominated in Federal Reserve accounting units. Assume the public’s relative demand for Federal Reserve accounting units has not changed. Individuals first receiving the new money will find they have more dollars to spend and save, but not more gold. Since there was previously a marginal utility balance in their demand vectors, some prices or quantities must be changed to restore it.

If only one kind of money existed in the model, the entire excess supply would be bid for consumption or investment, but with a parallel monetary system, part of the excess supply of dollars will bid up the price of gold. The demand schedule for gold coins as well as gold-ounce denominated deposit accounts and securities will shift outward. With floating exchange rates, there would be a reciprocal drop in the value of the dollar. Even if no other prices had initially changed, the market would be immediately aware of new information about cash balances: one ounce of gold would be worth more dollars, and one dollar would be worth less gold.

Obviously the central bank might take the opposite initial action, and constrict the supply of dollars. A movement in the dollar value of a gold ounce would reveal the action as individuals in the market started to sell gold to replace the dollar-denominated liquidity the Fed had removed from the system. The relative value of gold in terms of dollars would fall.

The response of the public to an excess supply of dollars would be transmitted by first-recipients to their bankers by (1) depositing the check received from the FOMC, and (2) instructing the banker to diversity the increase in their balances. To avoid exposure to exchange risk on its balance sheet, the bank would find it prudent to initiate offsetting adjustments in its own investments. It would want to acquire additional gold-denominated assets and reserves. It would drop interest rates on gold-ounce loans, and raise interest rates on dollar loans, to encourage a shift in the proportion of the bank’s assets in each accounting unit. The directions of interest rate movements is dominated by a “Fisher effect” that is determined by the dollar/gold price movement.

In conventional macroeconomic models, we know that some prices respond more slowly than others to monetary shocks. This is due largely to an information problem. The effect of an excess supply, or an excess demand, for money has to be transmitted through its impact on real output and employment. Managers have to make explicit decisions to increase or decrease particular prices, or to increase or decrease employment. They have no way of knowing whether apparent changes in market conditions are real or just monetary.

In this parallel currency model, the immediate transmission of information by means of relative price changes about monetary supply and demand is a principal advantage. If the relative price of two alternative base monies signalled policy changes, every market participant would be in a better position to evaluate whether an apparent increase in demand were real or nominal and to respond accordingly.

The Monetary Adjustment Mechanism

We might expect to observe part of any increase or decrease in demand for gold ounces to take the form of transactions at the Treasury, disbursing or receiving coins in exchange for dollars. This activity should be essentially passive on the part of the Treasury, as the public adjusted to changes in the supply and demand for dollars. The Treasury’s disbursement of coins would entail a decrease in the volume of dollars in circulation. These open market operations would partially offset actions taken by the Federal Reserve to increase the dollar supply, just as repurchases of gold coins by the Treasury paying out dollars, would offset the problem of dollar-supply contraction. This implies that the gold price in dollars would fluctuate less than if the Treasury did not participate. One research issue, therefore, is the apportionment of the shift in demand for gold ounces between price movements and quantity movements.

There would be an impact on the behavior of the central bank from the flow of Federal Reserve accounting units back into the Treasury. On the quantity side of the process, the Treasury could sterilize any dollars obtained from the public or use them to repurchase the T-bills and bonds the Federal Reserve had acquired in the first place when it expanded its monetary base. The information about the dollar/gold price would signal whether public demand for any FOMC increase in the supply of dollar base money was greater or less than the amount supplied.

To the extent the Treasury chose to play this offsetting role, the Federal Reserve’s open market operations would become circular. If it took no such action, the public would continue to shift cash balances and security holding among gold and dollars to achieve a marginal utility equalization in the face of anticipated price changes for goods and services. In other words, the supply of Fedeal Reserve base money, or gold coins, in the hands of the public could be monitored directly so as never to exceed - or fall short of - the public’s demand to hold them. This is another way of stating the condition of real-balance equilibrium.

To achieve this equilibrium result, of course, one has to assume a certain kind of repurchase activity on the part of the Treasury, as well as its willingness to act as an exchange agency for gold coins and Federal Reserve dollars; but clearly the mechanism for coordinaton would exist - and most important, it would not depend upon any kind of statistical data gathering or centralized planning by monetary authorities.

If the Federal Reserve makes policy mistakes, the Treasury’s behavior can neutralize them; if the Treasury makes policy mistakes, the Federal Reserve can neutralize them. If both branches of government make policy mistakes, the public can neutralize them with price adjustments among components of the money supply. Finally, and perhaps most important, if the exogenous shock is not from a policy mistake but from some real disturbance, the entire system can shift in search of a new equlibrium - it is not necessary to have a central planning committee adopt one policy for the entire economy, based on possibly wrong information and guesswork. Their ignorance and mistakes can be quickly seen by market participants, who can adjust portfolios quickly.

There is a similarity, but also an important difference, between these transactions at the Treasury and the process that was supposed to prevail under the classical gold standard. When the price between banknotes and coins is fixed, but the banknotes or deposits are in excess supply, supposedly the public can restore a balance in its portfolio by withdrawing deposits or turning in banknotes one-to-one for coins (an “internal drain” of central bank reserves). This process was supposed to limit the overissue of banknotes. The reason why it never quite worked that way is due to the fixed price: secure banknotes are almost perfect substitutes for coin, but banknotes and deposits in excess of prudent reserves were never so identified until it was too late.

A periodic report of changes in central bank gold reserves is a very inefficient relay of information about monetary policy; it is delayed, subject to conflicting interpretations, and has no immediate relevance to the cash balances of people in the market. The classical gold standard provides no information mechanism like the relative exchange rate movements of monetary units under a parallel standard.

The public’s expectations about inflation or a trend in the value of dollars to gold would also play a dynamic role in the process of substituting gold ounces for dollars in the public’s cash balances. Under a parallel standard, as under floating exhcange rates internationally today, an appreciating currency unit attracts speculators. These “capital flows” tend to “overshoot” the strict purchasing power parity ratios between two different monetary units, as the experience internationally has shown.

Except for government restrictions on banking and tax policies, a parallel currency system already exists in much of the world - which uses the U.S. dollar for certain kinds of transactions. Its benefits are not particularly visible to Americans, because the size of our economy tends to isolate us - in much the same way that explicit controls would do. One complication of present U.S. law, the tax treatment of assets that fluctuate in price, discriminate against the use by Americans of any accounting unit except U.S. dollars. If a flat-rate income tax were adopted, as many are advocating, this barrier to the use of gold ounces as money could disappear. Moreover, the Federal Reserve enforces an informal policy against any U.S. banks offering foreign-currency demand deposit accounts to U.S. residents, although these services are available to anyone who walks in and gives a foreign address. [Author's note (2008): I have been told this "informal policy" was relaxed in the 1980s after this was written.]

In spite of this view by some economists that anyone can use gold as “money” in the U.S. today, its use as a parallel currency is fully circumscribed by tax laws and banking regulations that would have to be modified to permit “monetary freedom” to become a constraint on monetary policy.

The Price Adjustment Mechanisms

During the collapse of the peso in 1982, in northern Mexico there was a massive rush of consumers from Texas who emptied the shops before the Mexican government intervened. Commerce was diverted from shops in southern Texas, which suffered an increase in unemployment. Some prices controlled in pesos made it easy for the Texans to exploit Mexican shopkeepers, and Mexican consumers were in some places left with shortages of bread and milk. In other instances, the ability of the Texans to pay in dollars instead of pesos was attractive to the Mexican shopkeepers, so relative prices were reduced for payment in hard currency.

Under a parallel monetary system, a depreciating unit may be what the buyers want to tender, but the appreciating unit is what the sellers want to receive. Without a legally mandated exchange ratio of price controls, a process of agreeing quickly on contractual values would emerge to minimize transactions costs between buyer and seller.

It is difficult, perhaps impossible, to predict the precise form the market’s adaptation would take, but the awareness of the need for some adaptive mechanism itself is beneficial. At the retail level, this might take the form of a dual price system - where every item is quoted in two prices. Or, if electronic cash registers are in use, similar to the ones today that confirm whether your credit card is authorized, the exchange rates could be computed instantaneously depending upon what means of payment the buyer offered. In certain parts of Europe, near national frontiers, people are never unaware of current exchange rates.

Longer-term obligations could probably by stated as “X dollars or Y ounces,” constituting a form of indexing; or, more subtly, they could be stated as “X dollars plus Y ounces.” This second formula would represent a method of hedging any exchange fluctuation with equal risk for both debtor and creditor. Another possible pattern could be “dollar” pricing of some things like groceries with a short turn-over and “ounce” pricing of durables. The difference in size of the two units might influence pricing conventions, but milli-ounces would always be transferred by check, credit card, or certificates of deposit.

In a society with only one form of money, people suffer under money illusion because of the “field-ground” problem (i.e., is a glass of water half-empty or half-full). We make economic decisions in terms of relative prices of things, and prices are quoted in monetary units. Inflation is preceived therefore as an increase in prices, rather than a decline in purchasing power of the unit. Governments often mke the problem worse by price controls or restrictions on foreign currency.

What is important to notice in this discussion of market adjustments to a parallel currency is that neither buyer nor seller is necessarily penalized by changes in monetary policy due to “money illusion.” First, the market would be alert to changes and prepared to cope with them. The problem we experience in our present system due to inflexible prices will have been explicitly modified by the market to promote flexibility. Second, the bad habit of thinking about value over multiple time periods in terms of depreciating monetary units would be curbed as people began to pay more attention to relative real market values: the field-ground problem would be eliminated because people would be aware of it. Even if a person is not normally aware of economic news, others will be, and ask sensitizing questions, like “How do you want to pay for this?”

Conclusion

In our opinion, the major point that can be made against a parallel currency is the accusation of “chaos” in the pricing system. If general transactions costs are increased by a parallel monetary arrangement, these must be weighed as opportunity costs against a system with less attention paid to units of pricing and payment, but more “money illuision” and price rigidity as well. To the extent that price flexibility were able to be institutionalized, stabilization policy on the part of governments or central banks would become unnecessary and irrelevant. The degree of “chaos,” moreover, cannot be any different in substance than the situation in world trade today, which most economists endorse.

The major benefit claimed for a parallel currency system is greater overall stability in the monetary system, with the supply of cash balances more likely to coincide with their demand. The interplay between the central bank, the Treasury, and the public would replace the current system where the public is merely victimized. Because the “gold ounce” would not be the coin of any single realm, it could perform this beneficial role in every nation’s economy and improve the coordination of economic decision makers in many different countries.

There are a number of research issues presented in this model. Economists will want to formulate hypothetical cases and test the relationships we have described. There are historical examples, as in the U.S. when “gold dollars” and greenbacks were both current at a floating rate.[7] There is empirical evidence: Parts of the world today have parallel currency systems (e.g., the West Bank of the Jordan, or the capital markets in Eurocurrencies).

In Hayek’s Denationalisation of Money (1978), he proposed a similar model but started with an assumption that international banks would create their own units of account, like IMF Special Drawing Rights consisting of baskets of currencies. His model has a lot in common with money market mutual funds today. Reviews of Hayek’s proposal found it wanting primarily due to the problem of an eventual currency unification under his system.[8] Competition among issuing banks would find some superior to others, and the market would shift progressively to just one bank’s currency unit.

The incentives that lead to disappearance of competition, however, are due to the profit inherently accruing to any issuer of fiat units. The capturing of this profit (rent) for use by the government exclusively is the main non-political argument in favor of central banking. The monopoly profit is otherwise called seigniorage, but there is no seigniorage if the second form of base money is strictly a unit of weight of gold or silver; a cartel arrangement with a central bank would not be monopoly-profitable.

Our departure from Hayek’s plan is based on a consideration of the liquidity of his proposed units. The enduring attraction that precious metals have to “unsophisticated” investors gives gold and silver a quality of liquidity that new or unusual units would not possess. The use of the lable “ounce” is already part of the price-quoting system when precious metals and national currency units are compared, so one important process of acculturation for this kind of monetary constitution is already in place, and no single issuer could trademark a “gold ounce” unit and expect to command a premium over anybody else’s similar monetary unit.

The specific Hayek proposal’s lack of durability poses a serious objection to its consideration as a monetary constitution, but his general argument in favor of monetary freedom is valid. Since a world with floating exchange rates and multiple currencies has become both “political reality” and the focus of research regarding its impact on trade, competitiveness, and price levels, it seems reasonable to look at a simple extension of the existing system. The basis for a further step, to enhance the liquidity of gold-ounce units by deregulation of financial services, would seem to be a logical extension of a current trend.

The adoption of a parallel standard by the United States would undoubtedly influence the policies of other governments. The number of countries that already have established legal-tender gold-ounce coinage include South Africa, Canada, Singapore, China, and Mexico. The United States Mint also produces a one-ounce medallion today [1984].[9] This parallel currency model merely introduces the “gold ounce” as a U.S. legal tender unit of account. Even if most economists still wince at the four-letter word, it provides one way to look for a monetary constitution consistent with our prevailing opportunities.

Endnote References

[1] Joe Cobb is a member of the professional staff of the Committee on Banking, Finance, and Urban Affairs, U.S. House of Representatives. This paper was presented at the Center for the Study of Public Choice, George Mason University, February 15, 1984.

[2] One of the early landmarks is Leland B. Yeager, In Search of a Monetary Constitution (1962), but note Henry Simons, “Rules versus Authorities in Monetary Policy,” Journal of Political Economy (1936). Even Sir William Blackstone, Commentaries on the Laws of England (1769) devoted a number of pages to an analysis of constraints on the monetary powers of the sovereign.

[3] Few arguments in favor of this presumption are as persuasive as “The Little Alchemist: Two Versions of a Fable,” in H.Geoffrey Brennan and James M. Buchanan, Monopoly in Money and Inflation (1981).

[4] Se F.A. Hayek, “The Pretense of Knowledge,” Nobel Lecture (1974).

[5] Either gold or silver coins would be suitable - the choice of metal is not important and should probably be left to the market. We will discuss below why the second form of money cannot be fiat, as proposed by F.A. Hayek, Denationalisation of Money (1978).

[6] Cf. the observations of Walter Bagehot, Lombard Street (1873), on the “unnatural system” of centralization.

[7] Cf. this comment by Milton Friedman and Anna J. Schwartz, A Monetary History of the United States (1963): “The period from the end of the Civil War to the resumption of gold payments in 1879 is of unusual interest to the student of money. . . . The price level fell to half its initial level in the course of fifteen years and, at the same time, economic growth proceeded at a rapid rate. . . . [This] casts serious doubt on the validity of the now widely held view that secular price deflation and rapid economic growth are incompatible.”

[8] Cf. Roland Vauble, “Free Currency Competition,” Weltwirtschaftliches Archiv (Kiel, 1977).

[9] The U.S. Mint began issuing legal tender one-ounce bullion coins in 1986. See Public Law 99-185