by Milton Friedman
Mont Pelerin Society
September 1961

International monetary arrangements have held a consistently important place among the topics discussed at the meetings of our [Mont Pelerin] Society. This is eminently fitting, since there is probably no other major facet of economic policy with respect to which liberals (in the sense of our Society) reach such divergent conclusions from the same underlying principles.

One group, of which Philip Cortney is a distinguished member, favors a continuation of the formal linking of national currencies to gold, rigid exchange rates between different national currencies, a doubling or more than doubling of the official price of gold in terms of national currencies, and an abandonment of governmental measures designed to evade the discipline of gold. This group is apparently indifferent about whether gold circulates as coin; it is satisfied with a gold bullion standard.

A second group, represented by the Economists’ National Committee on Monetary Policy, also favors a continuation of the formal linking of national currencies to gold together with rigid exchange rates between different national currencies. But it emphasizes the importance of gold coinage and of a widespread use of gold coin as money in national as well as international payments. Apparently, this group believes there is no need for a change in present official prices of gold, or, at least, in the Unites States price.

A third group, of which I count myself a member, favors a separation of gold policy from exchange-rate policy. It favors the abandonment of rigid exchange rates between national currencies and the substitution of a system of floating exchange rates determined from day to day by private transactions without government intervention.

With respect to gold, there are some differences, but most of us would currently favor the abandonment of any commitment by governments to buy and sell gold at fixed prices and of any fixed gold reserve requirements for the issue of national currency as well as the repeal of any restrictions on private dealings in gold.

I have stated and defended my own policy views elsewhere at some length. Hence, I would like to use this occasion instead to explore how it is that liberals can reach such radically different conclusions.

My thesis is that current proposals to link national currencies rigidly to gold whether at present or higher prices arise out of a confusion of two very different things:

• the use of gold as money, which I shall call a “real” gold standard;
• governmental fixing of the price of gold, whether national or international, which I shall call a “pseudo” gold standard.

Though these have many surface features in common, they are at bottom fundamentally different – just as the near identity of prices charged by competitive sellers differs basically from the identity of prices charged by members of a price-ring or cartel. A real gold standard is thoroughly consistent with liberal principles, and I, for one, am entirely in favor of measures promoting its development, as, I believe, are most other liberal proponents of floating exchange rates. A pseudo gold standard is in direct conflict with liberal principles, as is suggested by the curious coalition of central bankers and central planners that has formed in support of it.

It is vitally important for the preservation and promotion of a free society that we recognize the difference between a real and pseudo gold standard. War aside, nothing that has occurred in the past half-century has, in my view, done more to weaken and undermine the public’s faith in liberal principles than the pseudo gold standard that has intermittently prevailed and the actions that have been taken in its name. I believe that those of us who support it in the belief that it either is or will tend to be a real gold standard are mistakenly fostering trends the outcome of which they will be among the first to deplore….

The Distinction between a Real and a Pseudo Gold Standard

Because of its succinctness and explicitness, Cortney’s numbered list of prerequisites for the restoration of “monetary order by returning to an international gold standard” forms an excellent point of departure for exploring the differences between a real and a pseudo gold standard. His point number (6) concludes “the price of gold will have to be raised to at least $70 an ounce.” His point number (7) is “Free markets for gold should be established in all the important countries, and trading in gold, its export and import should be absolutely free.” Here is the issue in a nutshell. Can one conceive of saying in one breath that worldwide free markets should be established in, say, tin, and in the next, that the price of tin shoiuld “be raised” to some specified figure? The essence of a free market is precisely that no one can “raise” or “fix” price. Price is at whatever level will clear the market and it varies from day to day as market conditions change. If we take Cortney’s point (7) seriously, we cannot simultaneously take his point (6) seriously, and conversely.

Suppose we follow up the logic of his point (7) and suppose a free market to prevail in gold. There might then develop, as there has in the past, a real gold standard. People might voluntarily choose to use gold as money, which is to say, to express prices in units [ed. note: ounces or grams] of gold, and to hold gold as a temporary abode of purchasing power permitting them to separate an act of barter into a sale of goods or services for money and the purchase of goods or services with money. The gold used as money might be called different things in different languages: “or” in French, “gold” in English; it might be measured in different units: say, in grams in France and ounces in the United States; special terms such as “napoleon” or “eagle” might develop to designate convenient amounts of gold for use in transactions, and these might differ in different countries. We might even have governments certifying to weight and fineness, as they now inspect scales in meat markets, or even coining “eagles, double eagles,” and the like. Changes in nomenclature or in units of measure, say, the shift from ounces to grams, might be made by legislation, but these would clearly have no monetary or income or redistributive effects; they would be like changing the standard units for measuring gasoline from gallons to liters; not comparable to changing the price of gold from $35 an ounce to $70 an ounce.

If such a real gold standard developed, the price of commodities in terms of gold would of course vary from place to place according to transportation costs of both the commodities and of gold. Insofar as different countries used gold, and used different units, or coins of different size, the price of one kind of gold in terms of another would be free to vary in accordance with preferences by each country’s citizens for the one kind or the other. The range of variation would of course be limited by the cost of converting one kind of gold into another, just as the relative price of commodities is similarly limited.

Under such a real gold standard, private persons or government might go into the business of offering storage facilities, and warehouse receipts might be found more convenient than the gold itself for transactions. Finally, private persons or governments might issue promises to pay gold either on demand or after a specific time interval which were not warehouse receipts but nonetheless were widely acceptable because of confidence that the promises would be redeemed. Such promises to pay would still not alter the basic character of the gold standard so long as the obligors were not retroactively relieved from fulfilling their promises, and this would be true even if such promises were not fulfilled from time to time, just as the default of dollar bond issues does not alter the monetary standard. But, of course, promises to pay that were in default or that were expected to be defaulted would not sell at face value, just as bonds in default trade at a discount. And of course this is what has happened when a system like that outlined has prevailed in practice (e.g., in much of the pre-Civil War period in the United States).

Such a system might and I believe would raise grave social problems and foster pressure for governmental prohibition of, or control over, the issue of promises to pay gold on demand. But that is beside my present point, which is that it would be a real gold standard, that under it there might be different national names for the money but there would not be in any meaningful sense either national currencies or any possibility of a government legislating a change in the price of gold.

Side by side with such a standard, there could, of course, exist strictly national currencies. For example, in the United States from 1862-1879, greenbacks were such a national currency which circulated side by side with gold. Since there was a free market in gold, the price of gold in terms of greenbacks varied from day to day, i.e., in modern terminology, there was a floating rate of exchange between the two currencies. Since gold was in use as money in Britain and some other countries, its main use in the United States was for foreign transactions. Most prices in the United States were quoted in greenbacks but could be paid in gold valued at the market rate. However, the situation was reversed in California, where most prices were quoted in gold but could be paid in greenbacks at the market rate. No doubt, in this historical episode, the expectation that greenbacks would some day be made promises to pay gold had an effect on their value by expanding the demand for them. But this was not essential to the simultaneous coexistence of the two currencies, so long as their relative price was freely determined in the market, just as silver and gold, or copper and silver, have often simultaneoulsy circulated at floating rates of exchange.

If a government abjured a national currency, it might still borrow from the community in the form of securities expressed in gold (or bearing gold clauses), some of which might be demand obligations and might be non-interest bearing. But it would thereby surrender everything that we now call monetary policy. The resources it could acquire by borrowing would depend on the interest it was willing to pay on interest-bearing securities and on the amount of non-interest bearing demand securities the public was willing to acquire. It could not arbitrarily issue any amount of non-interest bearing securities it wished without courting inability to meet its promises to pay gold and hence seeing its securities sink to a discount relative to gold. Of course, this limitation in governmental power is precisely what recommends a real gold standard to the liberal, but we must not make the mistake of supposing that we can get the substance by the mere adoption of the form of a nominal obeisance to gold.

The kind of gold standard we have just been describing is not the kind we have had since at least 1913 and certainly not since 1934. If the essence of a free market is that no one can “raise the price,” the essence of a controlled market is that it involves restrictions of one kind or another on trade. When the government fixes the price of wheat at a level above the market price, it inevitably both accumulates stocks and is driven to control output – i.e., to ration output among producers eager to produce more than the public is willing to buy at the controlled price. When the government fixes the price of housing space at a level belowe the market price, it inevitably is driven to control occupancy – i.e., to ration space among purchasers eager to buy more than the sellers are willing to make available at the controlled price. The controls on gold, like the related cotnrols on foreign exchange, are a sure sign that the price is being pegged; that dollar, pound, etc. are not simply different names for different sized units of gold, but are national currencies. Insofar as the price of gold in these currencies and the price of one currency in terms of another are stable over considerable periods, it is not because of the ease of converting one qualtity of gold into another and not because conditions of demand and supply make for stable prices, but because they are pegged prices in rigged markets.

The price of $35 an ounce at which gold was supported by the United States after January 1934 was initially well above the market price – like the price at which wheat is currently being supported….

But the war intervened, which stopped the inflow of gold and brought a major rise in the stock of money. The resultant rise in other prices with no change in the price of gold has altered the character of the fixed United States price. … The restriction on the ownership of gold abroad by United States citizens is a first, and feeble, step toward still tighter rationing of demanders. The gentlemen’s agreement among cental banks not to press for conversion of dollar balances into gold is a more far-reaching if still rather weak additional step. The history of every attempt at government price fixing suggests that if the pegged price is far below the market price for long, such attempts are doomed to fail.

Doubling the price of gold would no doubt reverse the situation and raise the pegged price again above the market price. Gold production and United States gold stocks would no doubt rise. But to what avail? Gold would still be simply a commodity whose price is supported; countries would continue with their separate monetary policies; fixed exchange rates would freeze the only market mechanism available under such circumstances to adjust international payments; foreign exchange crises would continue to succeed one another; and direct controls of one kind or another would remain the last resort, and one often appealed to, for resolving them.

This kind of pseudo gold standard violates fundamental liberal principles in two major respects. First, it involves price fixing by government. It has always been a mystery to me how so many who oppose on principle government price fixing of all other commodities can yet approve it for this one. Second, and no less important, it involves granting discretionary authority to a small number of men over matters of the greatest importance; to the central bankers or Treasury officials who must manage the pseudo gold standard. This means the rule of men instead of law, violating one of our fundamental political tenets. Here again, I have been amazed how so many who oppose on principle the grant of wide discretionary authority to governmental officials are anxious to see such authority granted to central bankers. True, central bankers have on the whole been “sound money” men with great sympathy for private enterprise. But since when have we liberals tempered our fear of concentrated power by trust in the particular men who happen at a particular moment to exercise it? Surely our cry has been very different – that benevolent or not, tyranny is tyranny and the only sure defense of freedom is the dispersal of power.

Conclusion

Let me close by offering a proposal, not for reconciling our views, but at least for possible agreement among us on one part of the gold problem. Can we not all agree with Mr. Cortney’s point (7): the establishement of a thoroughly free market in gold, with no restrictions on the ownership, purchase, sale, import, or export of gold by private individuals? This means in particular, no restrictions on the price at which gold can be bought or sold in terms of any other commodity or financial instrument, including national currencies. It means, therefore, an end to governmental price fixing of gold in terms of national currencies.

The major problem in achieving such a reform is, as for the United States wheat program, the transitional one of what to do with accumulated government stocks. In both cases, my own view is that the government should immediately restore a free market, and should ultimately dispose of all of its stocks. However, it would probably be desirable for the government to dispose of its stocks only gradually. … This seems to me a matter of expediency not of principle.

A worldwide free market in gold might mean that the use of gold as money would become far more widespread than it is now. If so, governments might need to hold some gold as working cash balances. Beyond this, I see no reason why governments or international agencies should hold any gold. If individuals find warehouse certificates for gold more useful than literal gold, private enterprise can certainly provide the service of storing the gold. Why should gold storage and the issuance of warehouse certificates be a nationalized industry?
________________________
Reprinted from The Journal of Law and Economics, vol.4 (Oct.1961), by permission of the University of Chicago Press. This text is taken from Research Report 81CNC-04, August 31, 1981, issued by the U.S. Choice in Currency Commission, a non-profit association in Washington, DC, of which I was the executive director.