George Selgin, University of Georgia
The proposition known as “Gresham’s Law” is often stated baldly as “bad money drives good money out of circulation.” Ancient and medieval references to this tendency were informed by circumstances in which lightened, debased, or worn coins had assigned to them the same official value as coins containing greater quantities of precious metal. In this context the tendency, which had yet to be elevated to the status of an economic “law,” was one in which “bad” coins alone, that is, coins possessing a relatively low metallic content (“intrinsic value”), continued to be offered in routine payments, while “good” coins were withdrawn into hoards, exported, or reduced through clipping or “sweating” (that is, purposeful erosion by chemical or mechanical means) to an intrinsic value no greater than that possessed by their “bad” counterparts.
Later writers, however, tended to stretch the meaning of Gresham’s Law by invoking it as an argument against bimetallism and the competitive production of money and as the reason for the historical substitution of paper for metallic moneys; and a recent work even goes so far as to treat the law as being nothing more than an instance of the rule that rational agents prefer less expensive means of accomplishing their ends to dearer ones. As we shall see, although some modern interpretations of Gresham’s Law can be regarded as legitimate extensions of the law’s original (and perfectly valid) meaning, others are entirely unwarranted. Indeed, inappropriate applications of Gresham’s Law have caused some economists to err in the opposite direction, by jumping to the conclusion that, because in these applications the law appears to be contradicted by available empirical evidence, it must be altogether fallacious.
Attribution to Gresham and Ancient Status
The expression “Gresham’s Law” dates back only to 1858, when British economist Henry Dunning Macleod (1858, p. 476-8) decided to name the tendency for bad money to drive good money out of circulation after Sir Thomas Gresham (1519-1579). However, references to such a tendency, sometimes accompanied by discussion of conditions promoting it, occur in various medieval writings, most notably Nicholas Oresme’s (c. 1357) Treatise on money, and can even be found in much earlier works, including Aristophanes’ The Frogs, where the prevalence of bad politicians is attributed to forces similar to those favoring bad money over good.
As for Gresham himself, he observed “that good and bad coin cannot circulate together” in a letter written to Queen Elizabeth on the occasion of her accession in 1558. The statement was part of Gresham’s explanation for the “unexampled state of badness” England’s coinage had been left in following the “Great Debasements” of Henry VIII and Edward VI, which reduced the metallic value of English silver coins to a small fraction of what that value had been at the time of Henry VII. It was owing to these debasements, Gresham observed to the Queen, that “all your ffine goold was convayd ought of this your realm.”
Importantly, as Robert Giffen (1891, pp. 304-5) observed, Gresham made no direct reference to bimetallism or to “the analogous case of inconvertible paper when the paper drives the metal out of circulation.” However, Giffen (who appears here to have followed Jevons’s lead) errs in claiming that Gresham was “only responsible for the suggestion that bad coins … drive good ones of the same metal out of circulation,” for Gresham’s letter to Elizabeth clearly points to the debasement of silver coins as a factor leading to the disappearance of gold (Fetter 1932, pp. 490-1).It remains true nonetheless that the treatment of Gresham’s Law as an argument against bimetallism or resort to any sort of paper money are modern extensions of the law’s original meaning, the merits of (and implicit or explicit assumptions underlying which) must be assessed separately from those of early versions.
Correct and Incorrect Interpretations
That bad coins have in fact often tended to drive better coins of the same metal out of circulation is beyond dispute. Yet historical exceptions to this tendency have been observed. Thus even in its narrowest meaning Gresham’s Law must be said to hold only under particular conditions. What are these conditions, and why are they crucial?
These questions may best be answered by first considering those exceptional cases in which good coins appear to have driven out bad ones rather than vice versa. The most notable of such exceptions arose in the context of international trade, where, as Robert Mundell (1998) has observed, “strong” currencies, meaning ones that tended to retain their precious metal value over long periods of time, tended to dominate and drive-out “weak” (that is, less reputable) ones: “The florins, ducats and sequins of the Italian city-states did not become ‘dollars of the Middle Ages’ because they were bad coins; they were among the best coins ever made.” Less well known but equally important exceptions to Gresham’s Law involved relatively rare instances of competitive coin production, one example of which was the competitive production of gold coins by private mints in California in the wake of the gold rush. Here as well it was the higher-quality coins that captured the market, allowing their makers to thrive while less reputable private mints failed (Summers 1976).
The main thing that distinguished these exceptions to Gresham’s Law from other instances in which the law appears to have applied was the lack of any rules or of any authority capable of enforcing rules compelling people to accept particular coins in payment for goods or in the settlement of debts at some officially designated nominal value. Thus while the California private gold coins were, like those produced at the Philadelphia Mint, denominated in dollars, none of them were legal tender, and people were free to value them as they pleased, or to refuse them altogether. In practice only the better coins gained wide employment because others were not considered to be reliable representatives of the pre-existing dollar unit, and because valuing these inferior coins according to their actual gold content was inconvenient. In the market for international exchange media a similar tendency for good coins to be favored over bad stemmed from the absence of government authorities capable of enforcing legal tender laws and other rules compelling the acceptance of official coins “by tale” (that is, at par or face value, rather than by weight) beyond national borders.
Gresham’s Law can hold, on the other hand, where both good and bad coins enjoy similar legal-tender status and where non-trivial sanctions can be applied to persons who insist upon discriminating against bad coin and in favor of good coin. In such cases all coins must be accepted by tale, and the employment of bad coin becomes a dominant strategy in what amounts to a “Prisoners’ Dilemma” game in which both sellers and buyers participate. Buyers, knowing that sellers must accept either good and bad coins at their official face value, offer inferior coins, while hoarding, exporting, or reducing better ones; sellers, anticipating buyers’ dominant strategy, price their wares accordingly (Selgin 1996). As Frank Fetter has observed in a classic paper (Fetter 1932, pp. 494-5), the tendency for good coins (or metal made by melting metal from good coins) to actually leave the country is the result, not of debasement per se, but of the tendency for prices, including the price of bullion, to increase in consequence of an overall excess supply of coins. Debasement, including both official debasement and the unofficial reduction of good coins, contributes to such an excess supply by allowing an increased nominal stock of money to be derived from any given quantity of metal.
Legal tender laws of varying degrees of severity have normally buttressed the “sovereign prerogative” of coinage. In Gresham’s England, for instance, the sovereign assumed the right to reinforce its monopoly of minting by “imposing penalties for the offense of refusing the king’s coins at the value set upon them by the king,” and at least one act of Parliament (passed during the course of the debasements) explicitly “forbade any person to receive or to utter any coin at a price above the current value or proclamation rate.” Penalties for disobeying such laws included fines and imprisonment as well as the forfeiting of unlawfully exchanged sums. It is easy to understand how such laws promoted the use of bad coin over good. On the other hand, occasional changes to the legal status of bad coin, such as when Elizabeth elected (in response to Gresham’s advice) to “decry” (that is, to devalue) the bad shillings then in circulation, can serve to bring good coin back into open circulation.
Failure to recognize the dependence of Gresham’s Law upon laws interfering with the normal course of voluntary exchange has been responsible for some of the cruder misapplications of the law, including the tendency to treat it as describing the inevitable outcome of any sort of currency competition. A particularly egregious example occurs in William Stanley Jevons’s highly influential Money and the Mechanism of Exchange (1882 , pp. 64 and 82), where Jevons argues that Gresham’s Law supplies sufficient grounds for rejecting Herbert Spencer’s (1851, pp. 396-402) arguments for private coinage. Although Spencer was not an economist, his arguments appear to have been more consistent with a proper understanding of Gresham’s Law, as well as with evidence from actual private coinage episodes in California and elsewhere.
Extensions to Paper Money and Bimetallism
To observe that Gresham’s Law originally referred to circumstances where both good and bad coins of the same metal were awarded similar, substantial legal tender status is not to deny that the law may have other applications as well. Thus, legal tender laws may also attempt to compel people to treat coins of different metals, or coins and paper notes, as equivalents, unintentionally driving the more esteemed form of money into hiding. When, for example, the Continental Congress resolved in 1775 to treat any person refusing to accept irredeemable continental bills at their declared (specie) value as “an enemy of this country,” it merely succeeded it putting a stop to any open trading of specie. The French Revolutionary government’s decision to sentence to death persons caught discounting assignats relative to coins bearing the same face value had a similar effect. To describe such episodes as instances of Gresham’s Law at work is to propose a perfectly valid and sensible extension of the law’s original meaning. To insist, on the other hand (as Robert Mundell, among others, does), that Gresham’s Law also accounts for the historical tendency of redeemable banknotes, lacking legal tender status, to take the place of gold or silver coin is to misapply and misunderstand the law, in so far as the redeemable notes must have been regarded by their holders and by others who accepted them, not as “bad” money, but as money that was just as “good” as the coins into which they were readily converted. Properly understood, Gresham’s Law refers to an unintended consequence of legislation the intention of which is to force people to treat a money they view as inferior as if it were not so. The law is not, as Mundell (1998) asserts, simply an instance of the general (free-market) tendency for lower-cost substitutes to replace dearer ones capable of accomplishing the same ends!
Resort to Gresham’s Law in analyzing bimetallism must likewise be done with care. In so far as bimetallic legislation assigns similar legal tender status to both gold and silver coins, coins of one metal may be legally overrated relative to those made of the other metal, and these overrated coins may be treated as being analogous to worn, light, or debased coins, or “bad” money, in a monometallic system. “Good” (underrated) coins will then tend to be driven out of circulation, while only the “bad” (overrated) metal will be brought to the mint for coining.
The tendency just described is, however, limited by the fact that coins of different metals are unlikely to be equally useful in different transactions. In particular, gold coins will generally be of larger denominations and as such cannot supply the need for smaller change (cf. Sargent and Velde 2002). Consequently, even though gold may be legally overvalued relative to silver, and silver may cease to be voluntarily rendered to the mint, silver coins are unlikely to disappear from circulation altogether. Instead, they may circulate at a premium; alternatively, they may be clipped or sweated by private agents until their metallic value no longer exceeds their face value, as happened in Britain during the eighteenth century. Here and in some other bimetallic episodes Gresham’s Law held in its narrowest sense, in that reduced coins made of undervalued metal systematically took the place of full-bodied coins of that same metal. But the law did not hold strictly in the version of it proposed by some critics of bimetallism, in that gold and silver coins continued to circulate together.
While many writers have abused Gresham’s Law by treating it as being more generally valid than is in fact the case, Arthur Rolnick and Warren Weber (1986) err in the opposite direction in calling the law a “fallacy.” Their argument draws on examples involving either bimetallic legislation or the introduction of paper substitutes for gold or silver coin but not, significantly, on episodes involving debasement, to which all early statements of Gresham’s Law refer. With respect to bimetallism, Rolnick and Weber claim that conventional appeals to Gresham’s Law are based on the untenable assumption that government and private agents actually offer to exchange gold for silver and vice versa at some official non-market rate. Such a policy, they observe, “would imply potentially unbounded profits for currency traders at the expense of a very ephemeral mint or a very naive public” (ibid, p. 186). But this reading of conventional arguments is far-fetched: the operation of Gresham’s Law depends, not on persons actually offering to trade moneys having distinct “intrinsic” values at officially declared exchange rates, but on their being subject to sanctions if they attempt to value the moneys otherwise than as prescribed by law. The disappearance of “good” money occurs, if it occurs at all, as the unintended consequence of laws that seek, often quite futilely, to force people to treat differently valued moneys equally. Moreover, the idea that mints might offer to exchange gold for silver at official rates (as implied by the separate mint prices for those metals) is a perfect fiction that no proponent of Gresham’s Law has ever entertained.
A distinct component of Rolnick and Weber’s critique holds that, where both good and bad moneys are available, the good money, instead of disappearing from circulation, will tend to remain in circulation while trading at a premium; alternatively, the bad money may trade at a discount, with the good money serving as the medium of account at hence trading at par. Small change may be an exception to this rule, as it may be prohibitively costly to employ such change at other than its par value. Nevertheless, even with respect to small change Rolnick and Weber regard Gresham’s Law as fallacious, since, according to their view, the small change that disappears from circulation might be either “good” or “bad” money, depending upon which of these happens to be the medium of account.
In support of their argument Rolnick and Weber offer empirical evidence of bad and good moneys circulating together at market-determined exchange rates, including the case of California during the Greenback era, where the gold standard remained in effect, with greenbacks trading at a discount relative to gold. But while such evidence may demonstrate that Gresham’s Law isn’t universally applicable, it hardly succeeds in proving the law a fallacy. As has been noted above, Gresham’s Law, properly understood, applies only to circumstances where people are legally compelled to accept both good and bad moneys at their par or face values, either in spot transactions or in the settlement of debts. Where legal sanctions play no role (as was the case in California, where local authorities refused to enforce Federal legal tender legislation), market-based transactions costs alone may discourage the use of non-par money. However, because market-based transaction costs might systematically favor either good or bad money, depending upon which happens to function as a medium of account, such costs alone cannot account for the overwhelming number of historical instances in which bad money does in fact appear to have taken the place of good.
References and Further Reading
Fetter, Frank W. “Some Neglected Aspects of Gresham’s Law.” Quarterly Journal of Economics 46, no. 3 (1932): 480-95.
Giffen, Robert. “The Gresham Law.” Economic Journal 1, no. 2 (1891): 304-6.
Jevons, William Stanley. Money and the Mechanism of Exchange. New York: D. Appleton and Company, 1882.
Macleod, Henry Dunning. Elements of Political Economy. London: Longmans, Green & Co., 1858.
Mundell, Robert. “Uses and Abuses of Gresham’s Law in the History of Money.” Zagreb Journal of Economics 2, no. 2 (1998): 3-38. (http://www.columbia.edu/~ram15/grash.html).
Rolnick, Arthur J., and Warren E. Weber. “Gresham’s Law or Gresham’s Fallacy?” Journal of Political Economy 94, no. 1 (1986): 185-99.
Sargent, Thomas, and Françcois Velde, The Big Problem of Small Change. Princeton, NJ: Princeton University Press, 2002.
Selgin, George. “Salvaging Gresham’s Law: The Good, the Bad, and the Illegal.” Journal of Money, Credit, and Banking 28, no. 4 (1996): 637-49.
Spencer, Herbert. Social Statics. London: John Chapman, 1851.
Summers, Brian. “Private Coinage in America.” The Freeman 26, no. 7 (1976): 436-40.