Rick Ronsavelle
September 25th, 2010, 02:00 PM
"For almost as long as money has been the object of systematic study, Gresham's law--bad money drives out good--has been recognized as one of money's governing principles. Now, however, doubts have arisen concerning the validity of this venerable principle. In their provocative, theoretical reappraisal of Gresham's law, Rolnick and Weber (1986) maintain that bad money and good money will typically circulate together, bad money at its face value and good money at a price exceeding its face value. Bad money, they say, rather than driving good money out, drives it to a premium.
This paper examines the leading cases in nineteenth-century U.S. monetary history, looking for evidence of cocirculation. No such evidence appears. Money circulated by tale, not weight.
1. GRESHAM'S LAW AND ITS RIVAL
What, according to Gresham's law, makes a money "bad"? What enables it to "drive out good money"? A bad coin contains less bullion than does another coin carrying the same denomination. The mint, for example, might begin paying a better price for bullion. Naturally, people will take advantage of an opportunity to have bullion struck as coin more valuable than the bullion itself. Access to the mint thus makes bad money cheap money, too, and its supply grows. Prices rise, including the price of bullion. It becomes more valuable than the heavier coin, which people remove from circulation. Cheapness, not badness per se, then, enables bad money to drive out good.(1)
Gresham's law embodies two distinct propositions: (1) two intrinsically distinct moneys will not both circulate at once, except when, because the quantity of one money is limited, that one remains a subsidiary money, and (2) of the two moneys, only the bad money will circulate. We will mainly concern ourselves with demonstrating that (1) held in nineteenth-century America. Such a demonstration, though it does not itself constitute empirical verification of Gresham's law, suffices to reject its rival.
Gresham's law assumes that a coin circulates, as long as it does, at face value. Rolnick and Weber question this assumption. Despite legal-tender laws, people would supposedly continue using good coin as money, but at its bullion value. Consider a creditor, for example, with a $100 claim. If the debtor offered bad money denominated $100, then its legal-tender status would force the creditor to accept the offer; the creditor could not win a suit for nonpayment. But suppose that the debtor offered good money denominated $95 but, as bullion, worth $100. The legal-tender law would not stop the creditor from accepting the offer, and neither party would have reason to seek legal redress.(2) Rolnick and Weber thus focus on price, not quantity, as the magnitude that adjusts to keep the good coin worth no less as coin than as bullion. They do allow for quantity adjustment, but "only when the costs of using good money at a premium are significant" (pp. 195-96).
What costs are these? When do they become significant? The costs are the costs of rounding, and they become significant when applied to small-denomination coins. A 25 percent premium on a nickel, for example, is 1.25 cents. If people round 1.25 cents to a penny, then in a small transaction, a customer using nickels would lose part of the premium. People would therefore bundle nickels for use in a large transaction or else melt nickels for sale as bullion.
Actually, rounding would not be the main cost of using the good money at a premium. The value of the nonpar coin would fluctuate constantly, forcing people to monitor its value and then negotiate in every transaction.(3) Perhaps today, we can imagine that, at virtually every transaction site, a computer network would make a quotation on the nonpar money available instantaneously and inexpensively. Under nineteenth-century conditions, however, using the nonpar coin would have required negotiating its value despite considerable uncertainty about its value in central markets. Most transactors, therefore, would have preferred the par money and left negotiation of the nonpar coin's bullion value to specialists. Irving Fisher made the same point: "The better money might conceivably be used at a premium, i.e., at its bullion value; but the difficulties of arranging payments in it, which would be satisfactory to both parties, are such that in practice it is never so used in large quantities" (1922, p. 114).
We shall thus frame the issue in empirical terms. Before turning to the historical record, however, we should note two points that will play an important part in evaluating it. First, a premium might develop in the bullion market, giving specialists incentive to cull good money from circulation. Historical references to a premium thus require careful scrutiny. Do they refer to a premium in the bullion market, a necessary adjunct of Gresham's law? Or do they refer, instead, to a premium in ordinary transactions, evidence for Gresham's rival?
Second, even with Gresham's law in operation, a small quantity of good money might see use in sheltered markets, perhaps at a premium. Gold, for example, offered key advantages in nineteenth-century international trade. Importers and exporters, then, might have continued using gold even after it had become the dear money. Only if business in unsheltered markets stood as much chance of involving the nonpar money as involving the par money would the case under investigation support the dual-currency hypothesis.
2. THE AMERICAN EXPERIENCE IN THE NINETEENTH CENTURY
When it achieved independence, the United States had no coins of its own manufacture. By the end of the nineteenth century, however, the country had a sophisticated coinage system based on full-bodied gold coins and subsidiary coins of silver and other metals. The path from one point to the other was far from smooth. Changing economic and political forces led repeatedly to the introduction of new dollars intended to circulate alongside or to supplant the incumbent dollar. This process produced numerous tests of the dual-currency hypothesis. Here, we examine most of them, including all that Rolnick and Weber urge as violations of Gresham's law.(4) The test is simple. Did an increased supply of cheap dollars leave both dollars circulating at a fluctuating exchange rate, confirming Gresham's dual-currency rival? Or did the increased supply of cheap dollars drive dear dollars from circulation, con- firming Gresham's law?
Foreign Coins before the Civil War
Foreign silver coins, notably the Spanish milled dollar (the peso), circulated in the United States during most of the antebellum era.(5) Officially, the Spanish dollar contained 373.5 grains of pure silver, whereas the U.S. dollar, established in 1792, contained only 371.25 grains. Rolnick and Weber (1986, p. 187) point to two pieces of evidence suggesting that a heavier Spanish dollar might have circulated at a premium alongside a lighter U. S. dollar. First, in 1830, according to an estimate that J. Lawrence Laughlin reports, the Spanish dollar accounted for 22 percent of the coins circulating in the United States. Second, references exist to a premium of 0.25 to 1 percent on Spanish dollars, especially from 1792 to 1811 and as late as 1819.
In fact, the typical Spanish and Latin American coins in circulation contained less than the official 373.5 grains. Changing mint practices, defacement through gouging, clipping and sweating, and ordinary wear and tear produced the shortfall. In percentage terms, the shortfall was greater for smaller denominations and pieces of the Spanish dollar. (It was common practice to cut up the Spanish dollar to make small change; one eighth of a dollar was known as a bit.
In setting the weight of the U.S. silver dollar, Congress followed Alexander Hamilton's recommendations, as set out in his Treasury Report of January, 1791. Hamilton (1791, pp. 164-65, 186) explained that the Spanish mint had reduced the silver content of the Spanish dollar and that only the most recently minted Spanish dollars still circulated. Heavier Spanish dollars had circulated before the Revolution, but they had long since disappeared. Hamilton cited conflicting assays of the newest Spanish dollar. The assay that he considered the most reliable put the weight of the Spanish dollar then in circulation at about 371 grains, and this assay became the main basis for choosing the weight of the U.S. dollar.
Underweight Spanish and Latin American silver remained a substantial component of the currency throughout the early national period. Laughlin reports that as late as 1830 the U.S. coinage comprised $14 million in U.S. coins, $5 million in Spanish dollars and parts of dollars, and $4 million in other dollars and parts, mostly from Latin America. He (1896, pp. 54-55) describes the Spanish and Latin American silver as "much worn and reduced in weight" and adds that these coins "being in practice current with other coins, without regard to weight, naturally acted to drive out our own coins." In the same vein, Neil Carothers (1930, p. 81) reports: "It was feasible to clip Spanish coins to the legal weight of 415 grains [373.7 fine grains] and coins far below this weight were accepted without question. The silver coins of Mexico and Spain entered into circulation in a degenerate condition and once there they could not be withdrawn. . . . Storekeepers habitually accepted coins in any state of depreciation."
The 1830 Senate report that Laughlin and Carothers relied on claimed that foreign dollars were typically depreciated about 1 percent, half-dollars about 3 percent, eighths of a dollar 11 percent, and so on (New American State Papers, Public Finance, vol. 27, pp. 219-20). A weighted average of the foreign coins in circulation put them about 6 percent below their official weight. 6 The report (p. 22 1), moreover, noted that, "All these coins are equally current, without regard to actual weight," and it concluded that underweight foreign coins were driving out American coins.
The Senate committee had taken aim at laws making foreign coins legal tender. The committee may thus have exaggerated the amount of foreign silver in circulation and the extent of its depreciation.(7) But the committee would not likely have misrepresented the facts of interest here, namely, whether the foreign silver was full weight or depreciated and whether it circulated by weight or by tale, facts anyone reading the report could have verified easily.
In 1842, a book that the U.S. mint's assay office published reported on the Spanish dollar: "Its divisions, though greatly depreciated by wear, continue to circulate largely in this country . . . " (Eckfeldt and Du Bois, p. 117). Legislation passed in 1857 led, finally, to the removal of most foreign coins.
But if the circulation consisted of light foreign silver circulating at par with U.S. coins, what do references to a premium on Spanish coins mean? Merchants in the China trade paid the premium because the Chinese heavily discounted the newer U.S. dollar. Assays made in China showed the true worth of the U.S. dollar, but it lacked the credibility of the Spanish dollar, which had become the medium of exchange in certain areas of China. The premium compensated merchants in the China trade for culling acceptable coins. The premium therefore meant that Gresham's law was working, not that it was being violated. When the premium on the Spanish dollar rose to about 4 percent, merchants in the China trade switched to the U.S. dollar because the Chinese undervalued U.S. silver by about that much.
The confusion over the premium commanded by better specimens of the Spanish peso traces to a footnote of Laughlin's that quotes Secretary of the Treasury Crawford in 1819. "Spanish milled dollars compose the great mass of foreign silver coins which circulate in the United States," Crawford (quoted in Laughlin 1896, p. 53) wrote, "and generally command a premium when compared with the dollar of the United States . . . " The full sentence ends with the phrase, though Laughlin omitted it, "especially for exportation to China" (American State Papers, Finance, vol. 3, p. 395).
Further on, Crawford noted that the Spanish dollar circulated at anywhere from par to a 10 percent premium. Crawford's remarks suggest that bullion dealers paid the premium to acquire full weight coins for use in the China trade and that, depending on local demands and transportation costs, the premium varied from port to port. But a literal reading of Crawford's report would not rule out a premium on coins used in ordinary transactions. Other observers, however, were unambiguous. A House committee examining the currency laws, for example, reported, "Within the United States, it [the Spanish dollar] has an equal value with the U.S. dollar, and in many foreign countries a much higher value" and also, "The dollar of Spain and that of the United States are of exactly the same value within the United States" (American State Papers, Finance, vol. 3, p. 398).
Better specimens of Spanish and Latin American dollars thus disappeared from circulation because they were heavier and useful in the China Trade. The coins left in ordinary circulation, although issued by several governments, circulated at par.
Bimetallism before 1834
Another alleged violation of Gresham's law concerns the circulation of gold before 1834 and of silver afterward. In 1792, the United States set its mint ratio at 15: 1, the dollar containing 371.25 grains of pure silver or 24.75 grains of pure gold. That ratio almost equaled the ratio of world market prices. But soon the market ratio began to climb. According to Soetbeer's figures, it stood at 15.17:1 in 1792, rose to a peak of 16.25 in 1813, and then, after a brief fall, climbed to 15.95 in 1821 (Laughlin 1896, pp. 289-90).
Gold disappeared. According to a House committee report of 1819, "Gold is estimated below its fair relative value, in comparison with silver, and . . . can scarcely be considered as having formed a material part of our money circulation for the last twenty-six years" (American State Papers, Finance, vol. 3, p. 399). In 1821, another House committee reported, "On inquiry, they find that gold coins, both foreign and of the United States, have in great measure disappeared" (American State Papers, Finance, vol. 3, p. 646). In 1830, a Senate select committee stated flatly: "The coins held by the banks and in circulation among the people are silver. All the coins in common circulation, and most of those held by the banks, are half dollars and the minor silver pieces. We have no gold coins in circulation" (New American State Papers, Public Finance, vol. 27, p. 292). And in 1832, the director of the mint, Samuel Moore, noted, "Gold at present constitutes no part of our currency; and not having, within any recent period, performed in the United States the office of coin, it has not been the standard of value assumed in existing contracts" (New American State Papers, Public Finance, vol. 27, p. 325).
The congressional committees and the mint may have erred in terms of causes and cures. But, again, they would not likely have misrepresented the state of the circulation, something that anyone reading their reports would have been familiar with. Laughlin (1896, p. 30) thus stood on solid ground when he concluded, "Gold coins were seldom seen during the largest part of this period from 1792 to 1834."
Rolnick and Weber (1986, p. 188) point to the continued minting of gold--between 1793 and 1833, gold accounted for about 25 percent of the mint's output--as evidence against Gresham's law. Much of the gold coinage was exported (1986, p. 196), however, and much of the rest was sold as bullion. But why would someone have bullion (often obtained by melting foreign coins) struck as U.S. coins and then export the coins or sell them domestically as bullion?
Apparently, dealers converted bullion or foreign coins into U.S. coins to certify weight and purity. The mint took no seigniorage, and the resulting stamp of approval carried more authority than a private assay. The Senate committee of 1830, quoted above, expressed its indignation: "We prohibit and punish all private coinage of gold; we coin this metal at the mint, upon a principle which does not permit it to circulate as money; and we pay the expenses of this useless coinage. In practice, this coinage affords a facility to the possessor of gold bullion; since it enables him to employ the mint to weigh and assay his bullion, and to divide it into convenient portions, without expense to himself. When the coins are received from the mint, they are sold for their value as bullion; some of them are used in manufactures, and the greater part are exported" (New American State Papers, Public Finance, vol. 27, p. 292).
Bimetallism after 1834
In 1834, Congress reduced the weight of the gold dollar from 24.75 grains to 23.2 grains, thereby raising the mint ratio from 15:1 to 16.002:1.(8) This change, given prevailing market prices, made gold the overvalued money. Some silver flowed out and some gold flowed in, but the circulation did not change greatly. The small difference between the mint ratio and the market ratio, a difference that the worn state of the silver coinage made even smaller, permitted both gold and silver to circulate. Around the legal ratio, that is, stood a range of ratios that would not stimulate bimetallic arbitrage, a range analogous to the range the gold points set off under a gold standard (Friedman 1990, p. 90). In the West, relatively high transportation costs kept silver preeminent. Hugh McCulloch, at the time a banker in Indiana--he later served as Comptroller of the Currency and Secretary of the Treasury--wrote (1889, p. 119), "I had been a banker for fourteen years before I handled or saw a dollar in gold except the ten-thaler pieces which were brought into the country by German immigrants."
Gold discoveries in the late 1840s and early 1850s, however, quickened the expulsion of silver. Between 1834 and 1848, the premium on full weight U.S. silver coins, the difference between their internal and external gold values, averaged 1.2 percent. Between 1849 and 1859, it averaged 3.6 percent (Laughlin 1896, p. 297). By the Civil War, gold dominated the metallic circulation.
Greenbacks in the East
The Greenback era (1862-1879) has played an important role in shaping views of Gresham's law. In December 1861, the banks suspended paying specie (gold or silver) for notes and deposits. The Treasury reacted by suspending payment on its "old demand notes," issued the preceding summer (Mitchell 1903, pp. 40-41). In the East, where the money supply consisted largely of notes and deposits, gold immediately went to a premium. Apparently, merchants decided to post prices in terms of the most common medium of exchange, thus minimizing the number of transactions that required negotiating the price of the nonpar money.
The greenback, a fiat paper currency created by the Legal Tender Acts of 1862 and 1863, swiftly filled the channels of trade, serving both as a bank reserve arid as a hand-to-hand currency. In July 1864, the price of gold climbed to its high, $2.85 (Mitchell 1903, p. 424). After the War, a combination of slow growth of money and rapid growth of real income produced deflation. Resumption at the prewar parity occurred on January 1, 1879.
Most historians of the Greenback era consider it an example of Gresham's law. According to Laughlin (1896, p. 87), "The result of the depreciation of the paper money which became manifest by a premium on gold in June [1862] to the extent of 5 percent, and in July of 20 percent, naturally brought Gresham's law into operation, by which the cheaper paper was substituted for the more valuable gold. Gold disappeared before the depreciating paper, and it was not until January 1, 1879 that it again appeared. " Don Barrett (1931, p. 73) says, "Gold practically disappeared from circulation for a period of eighteen years." Francis A. Walker, a superb economist who lived through the period, says that gold saw no use (except by special contract) other than to pay customs duties and interest on the national debt (1883, p. 508).
Wesley C. Mitchell, the leading authority on the period, believed that, after the greenback appeared, gold no longer circulated in the eastern United States. His table (1903, p. 179) of "The Circulating Medium" shows no gold circulating in the loyal states, except on the Pacific coast.
A second piece of evidence counting against the dual-money view of the Greenback era is the fate of the national gold banks. Legislation of 1870 permitted the organization of national banks that could issue notes redeemable in gold. Had a large quantity of gold remained in circulation, gold banks would have succeeded throughout the country. Gold banks succeeded nowhere, however, but in California. The one eastern gold bank, the Kidder National Gold Bank of Boston, quickly liquidated, having found no demand for its notes (Greenfield and Rockoff 1995).
Mitchell and the others used a medium-of-exchange definition of money. Even if we define money more broadly, however, we find that specie played a minor role during the Greenback era. In face value terms, the stock of specie held by the public fell from about $149 million in 1859 to about $55 million in 1867. Comparisons with the stock of money make the point even more dramatically. In 1859, 26 percent of the money stock consisted of specie; by 1867 the figure had fallen to about 6 percent. And this figure includes gold circulating in California, where the greenback never circulated. Eliminating California's gold reduces the ratio of specie to total money to a little more than 3 percent. Subtracting specie holdings tied to paying customs duties cuts the ratio even further.
Figure 1, which shows the ratio of specie to money between 1850 and 1890, provides a longer-term perspective.9 The lower lines adjust the gross ratio to eliminate both specie circulating in California and gold held to pay customs duties. Specie used in California is Mitchell's estimate, based on Treasury opinions. Gold used in customs duties is based on the aggregate currency-income ratio and the amount of customs duties.
The ratio exhibits a falling trend. During the Civil War, however, the ratio falls well below trend. By 1875, probably less than I percent of the money stock, after the adjustments, consisted of specie. Figure I thus confirms Gresham's law and contradicts its rival.
Extensive hoarding began, apparently, when only a small increase in the amount of cheaper money had actually occurred. People expected the war to produce even larger increases. The government's fiscal position, which depended on the course of the war, conditioned these expectations. Richard Roll's (1972) and Charles Calomiris's (1988) finding that war news determined the premium on gold is thus consistent with a Gresham's-law interpretation of the period.
Gold did not actually leave the country until a few years after disappearing from circulation. Annual net gold exports for the years 1862 to 1865 averaged $54.75, as compared with $34.25 million in the preceding four years. Net gold exports peaked at $90 million in 1864. No other year in the nineteenth century recorded larger net gold exports, though 1893 came close (U.S. Bureau of the Census 1975, p. 885).
Silver disappeared along with gold. The silver did not wind up in the melting pot. Instead, it went to Latin America, and in lesser amounts to Canada, to serve as currency (Carothers 1930, p. 219). Later, when the greenback price of silver fell, a substantial reflux occurred (Carothers 1930, p. 259). During the Civil War, the U.S. mint continued to produce silver coins, but they did not circulate. According to Carothers (1930, p. 219), "The mints were, so far as silver was concerned, merely establishments conducted for the benefit of New York and San Francisco bullion dealers." The mints certified the billion's weight and fineness and produced coins suitable in foreign circulations.
A sheltered gold market remained active throughout the Greenback era. Servicing and paying both the government's and the railroads' bonded debt required gold, as did paying customs duties and buying things abroad. Even Laughlin and the other writers who took this period to confirm Gresham's law recognized the existence of this market.
Greenbacks in California
The legal-tender status of the greenback gave it a strong starting point in California. But the social pressure to use gold, a major export, proved insurmountable. California had shown a preference for hard money even before the War. The constitution, written in 1849, prohibited paper money. A few private banks had issued notes, defying the law, but these notes had difficulty swimming against the tide of public opinion.
Greenback inflation, of course, did nothing to warm Californians toward paper money. Social preferences became private costs. Debtors who tried repaying in greenbacks found themselves denounced for "greenbacking," their names published in the newspapers. The greenback remained foreign exchange, quoted at a discount from gold that mirrored the eastern premium on gold. In July 1864, for example, a greenback dollar sold for $.35 gold in California (Wright 1916/1980, p. 87). Paper money did circulate in California after 1870, when legislation established national gold banks. But the "yellowback dollar" was convertible into gold (Lester 1939, and Greenfield and Rockoff 1995).
Does the greenback's failure in California represent a defeat for Gresham's law? One interpretation would regard this case as lying outside the scope of Gresham's law because Californians considered the greenback foreign exchange, not money. Interpretations, however, may vary. Since the greenback had legal-tender status, critics could insist that it qualified as "bad money" and should have driven out gold, even in California. But whether or not the greenback in California counts against Gresham's law, this case cannot count for Gresham's rival. Domestically, gold and the greenback did not circulate alongside one another.
The Trade Dollar
In 1873, hoping to encourage the China trade, Congress authorized the Trade dollar, which contained 378 grains of silver. (10) The Trade dollar, according to Rolnick and Weber (1986, p. 190), circulated at its bullion value, about 93 cents. Again, we have something that looks like Gresham's rival, though, oddly, the cheap money appears to have carried a discount.
The full story of the Trade dollar is indeed curious. Although intended for the China trade, it circulated at par in certain areas of the United States, and large amounts were minted. Congress, apparently inadvertently, had made the Trade dollar legal tender for sums under 5 dollars (Garnet 1917, p. 96).(11) The Trade dollar, moreover, looked very much like a coin intended for domestic circulation and actually contained more silver than the old Spanish dollars or U.S. half-dollars. Merchants hesitated to refuse the coin. In 1876, however, Congress revoked the Trade dollar's legal-tender status, and in 1878 the mint discontinued production, Even so, the Trade dollar continued to circulate and, but for some form of collective action, might have continued to do so for a long time.
The federal government and business trade associations, however, set out to demonetize the Trade dollar. The government took a variety of actions against it, such as prohibiting the Post Office from accepting it. The trade associations objected to the banks' unwillingness to accept the coin, except for large deposits by favored customers and even then at its bullion value. Although we haven't found a specific regulation, the Trade dollar might not have fulfilled reserve requirements because it was not a legal tender.
In California, the Trade dollar disappeared by 1878 (Willem 1959, p. 126). In New York, however, the matter did not come to a head until 1883. The New York Times, in June 1883, reported that, at a meeting of produce dealers, a strong disagreement arose over whether to ban par acceptance of Trade dollars (Willem 1959, p. 133). Bankers Magazine (cited by Taxay 1966, p. 284) reported that pressure to refuse Trade dollars had succeeded completely in certain lines of business and partially (credit but not change for Trade dollars) in others.
In June 1883, posters appeared in New York City, probably placed there by the trade associations, warning against accepting Trade dollars. According to Willem (1959, pp. 133-34), the Trade dollar immediately ceased to circulate and headed toward the bullion dealers. After 1883, the Trade dollar was used in the China trade but not in the United States.
Pressure to redeem the Trade dollar grew. The argument, at least for public consumption, focussed on ordinary citizens who had accepted Trade dollars at face value. Formulating a redemption bill that would satisfy both the silver and the gold interests in Congress, however, proved difficult. Legislation authorizing redemption was finally passed in 1887. For six months, the government redeemed unstamped, full-weight Trade dollars at face value, paying standard silver dollars, the Bland dollars. (12)
The Bland Dollar
Congress, under pressure from silver interests, had authorized the Bland dollar, which contained the usual 371.25 grains, in 1878. It had unlimited legal-tender status and, when delivered to the Treasury as a bundle of ten, could be exchanged for $10.00 in silver certificates. Like gold, the silver certificates were accepted at par for customs duties. Because the market price of silver stood below the mint price, coining Bland dollars offered considerable profit. Under free coinage, the Bland dollar might have driven out the gold dollar. But the Bland-Allison Act limited the treasury's silver purchases to between $2 million and $4 million per month, and generally, purchases stood nearer the lower limit (Friedman and Schwartz 1963, p. 132). Because the supply of cheap money was limited, the dear money continued to circulate. The Bland dollar, in other words, was a subsidiary coin.
3. CONCLUSION
We have investigated seven cases in which bad money could have driven good money to a premium: (1) the U.S. silver dollar versus the foreign silver dollar in the early national period; (2) the silver dollar versus the gold dollar when, before 1834, the mint overvalued silver; (3) the silver dollar versus the gold dollar when, after 1834, the mint overvalued gold; (4) the greenback dollar versus the gold dollar in the east during and after the Civil War; (5) the greenback dollar versus the gold dollar in California during and after the Civil War; (6) the Trade dollar versus the gold dollar, 1873-1887; and (7) the Bland dollar versus the gold dollar, 1878-1890. None of these seven cases offers evidence that bad money drove good money to a premium.
Nor, on balance, does the evidence speak badly for Gresham's law. In cases (1)-(4), either the bad money was not bad enough or it did drive out the good money. In case (6), the bad money, the Trade dollar, lost monetary status before Gresham's law could operate. And in case (7), the restricted supply of the cheaper money, the Bland dollar, kept it a subsidiary coin.
Only case (5) presents any ambiguity concerning Gresham's law. In California, the greenback, despite its legal-tender status, never gained currency. One interpretation would put this case outside the scope of Gresham's law; another interpretation would consider this case a contradiction of Gresham's law. Whatever its verdict concerning Gresham's law, however, even case (5) gives no support to Gresham's rival.
A dual-currency system, which the rival view envisages, never materialized. Evidently, the transactions costs of using a nonpar money were substantial. (1.) Both Fisher (1922 [1971], pp. 115-22) and Fetter (1932) stress this point. (2.) This conclusion depends on how the authorities write and enforce the law. A referee points out that the French revolution mandated death for a seller who asked what type of money the buyer would use. We have the less drastic, American-type law in mind. (3.) Warren Weber points out that monitoring the supply of bad money also entails costs. In some circumstances, therefore, some people might prefer to use, say, gold coins at a premium despite the costs of monitoring the bullion market. (4.) The most important omissions occurred in the South during the Civil War. Gary M. Pecquet's (1987) finding that gold circulated at a premium in the Trans-Mississippi Confederacy supports Gresham's rival. But the cases in which he finds the Confederate currency to have lost monetary status resemble two cases that we discuss below: the greenback in California and the Trade dollar. Pecquet, furthermore, cites an interesting example of Gresham's law: the Confederate currency reform of 1864 gave an advantage to small bills, which were then hoarded while the large bills continued to circulate. (5.) Schilke and Solomon (1964) give an excellent numismatic history of this
(6.) According to McCullough (1984, p. 105), actual weights fell substantially below official weights in Lower Canada.
1830 1840
Spanish Dollar -.6% 1.1%
U.S. Dollar NA .5
Spanish Quarter -4.7 -6.4
U.S. Quarter NA .9(7.) The committee criticized the Mint's 1826 study, which reported much less depreciation (American State Papers, Finance, vol. 5, pp. 585-92). (8.) In 1837, the weight of the gold dollar was increased to 23.22 grains for technical reasons, making the ratio 15.988 to 1. (9.) Although the ratio is conceptually simple, calculating a consistent series requires many steps. The details appear in an appendix, which we will (10.) Willem (1959) gives an excellent numismatic history of the Trade dollar. (11.) Carothers (1930, p. 234), however, maintains that the decision to make the Trade dollar a legal tender was taken deliberately. (12.) Chinese merchants stamped coins (leaving "chop" marks) to show that they were genuine."
LITERATURE CITED
Barrett, Don C. The Greenbacks and Resumption of Specie Payments, 1862-1879. Cambridge: Harvard University Press, 1931.
Calomiris, Charles W. "Price and Exchange Rate Determination during the Greenback Suspension." Oxford Economic Papers 40 (1988), 719-50.
Carothers, Neil. Fractional Money. New York: John Wiley and Sons, Inc., 1930.
Eckfeldt, Jacob R., and William E. Du Bois. A Manual of Gold and Silver Coins of All Nations. Philadelphia: The Assay Office of the Mint, 1842.
Fetter, Frank Whitson. "Some Neglected Aspects of Gresham's Law." Quarterly Journal of Economics (May 1932), 480-95.
Fisher, Irving. The Purchasing Power of Money. New York: Augustus M. Kelly, Reprints of Economic Classics [The Macmillan Company], 1971 [1922].
Friedman, Milton. "Bimetallism Revisited." Journal of Economic Perspectives (Fall 1990), 85-104.
Friedman, Milton, and Anna J. Schwartz. A Monetary History of the United States. National Bureau of Economic Research, Princeton, N.J.: Princeton University Press, 1963.
Garnet, Porter. "The History of the Trade Dollar." American Economic Review (March 1917), 91-97.
Greenfield, Robert L., and Hugh Rockoff. "A Tale of Two Dollars: Currency Competition and the Return to Gold, 1865-1879." In Money and Banking: The American Experience, edited by Clifford H. Thies. Fairfax, Virginia: George Mason University Press for the George Edward Durell Foundation, 1995.
Hamilton, Alexander. "Treasury Report on the Establishment of a Mint, January 28, 1791." Reprinted in Documentary History of Banking and Currency in the United States, edited by Herman E. Krooss, vol. 1, 162-92. New York: Chelsea House Publishers, 1969 [1791].
Laughlin, J. Lawrence. A History of Bimetallism in the United States. New York: D. Appleton and Company; reprint, New York: Greenwood Press, 1968 [1896].
Lester, Richard A. "Retention of the Gold Standard in California and Oregon during the Greenback Inflation." In his Monetary Experiments: Early American and Scandinavian. Princeton, N.J.: Princeton University Press, 1939.
McCulloch, Hugh. Men and Measures of Half a Century. New York: Charles Scribner's Sons,1889.
McCullogh, A. B. Money and Exchange in Canada to 1900. Toronto: Dundurn Press Limited, 1984.
Mitchell, Wesley C. A History of the Greenbacks. Chicago: University of Chicago Press, 1903.
Pecquet, Gary M. "Money in the Trans-Mississippi Confederacy and the Confederate Currency Reform Act of 1864." Explorations in Economic History 24 (April 1987), 218-43.
Roll, Richard. "Interest Rates and Price Expectations during the Civil War." Journal of Economic History 32 (1972), 476-98.
Rolnick, Arthur J., and Warren E. Weber. "Gresham's Law or Gresham's Fallacy?" Journal of Political Economy 94 (1986), 185-99.
Schilke, Oscar G., and Raphael Solomon. America's Foreign Coins with Legal Tender Status in the United States, 1793-1857. New York: Coin and Currency Institute, 1964.
Taxay, Don. The U.S. Mint and Coinage. New York: Arco Publishing Company, Inc., for the U.S. Bureau of the Census, 1966.
U.S. Bureau of the Census. Historical Statistics of the United States, Colonial Times to 1970. Washington, D.C.: GPO, 1975.
Walker, Francis A. Money. New York: Henry Holt and Company, 1883.
Willem, John M., Jr. The United States Trade Dollar. New York: Marchbanks Press, 1959.
Wright, Benjamin C. Banking in California, 1849-1910. San Francisco: H.S. Crocker Co. [Arno Press, New York], 1916 [1980].
The authors thank Michael Bordo, Stanley Engerman, Milton Friedman, John M. Kleeberg, Roger Koppl, Angela Redish, Arthur Rolnick, Anna J. Schwartz, Warren Weber, Eugene White, and Leland Yeager for helpful comments on earlier drafts. Also, the University of California, Berkeley, the National Bureau of Economic Research Summer Workshop on the Development of the American Economy, and Rutgers University gave us much appreciated opportunities to present the paper.
Robert L. Greenfield is professor of economics and finance at Fairleigh Dickinson University. Hugh Rockoff is Professor of economics at Rutgers University.
http://www.allbusiness.com/finance-insurance/credit-intermediation-related-activities/539901-1.html
This paper examines the leading cases in nineteenth-century U.S. monetary history, looking for evidence of cocirculation. No such evidence appears. Money circulated by tale, not weight.
1. GRESHAM'S LAW AND ITS RIVAL
What, according to Gresham's law, makes a money "bad"? What enables it to "drive out good money"? A bad coin contains less bullion than does another coin carrying the same denomination. The mint, for example, might begin paying a better price for bullion. Naturally, people will take advantage of an opportunity to have bullion struck as coin more valuable than the bullion itself. Access to the mint thus makes bad money cheap money, too, and its supply grows. Prices rise, including the price of bullion. It becomes more valuable than the heavier coin, which people remove from circulation. Cheapness, not badness per se, then, enables bad money to drive out good.(1)
Gresham's law embodies two distinct propositions: (1) two intrinsically distinct moneys will not both circulate at once, except when, because the quantity of one money is limited, that one remains a subsidiary money, and (2) of the two moneys, only the bad money will circulate. We will mainly concern ourselves with demonstrating that (1) held in nineteenth-century America. Such a demonstration, though it does not itself constitute empirical verification of Gresham's law, suffices to reject its rival.
Gresham's law assumes that a coin circulates, as long as it does, at face value. Rolnick and Weber question this assumption. Despite legal-tender laws, people would supposedly continue using good coin as money, but at its bullion value. Consider a creditor, for example, with a $100 claim. If the debtor offered bad money denominated $100, then its legal-tender status would force the creditor to accept the offer; the creditor could not win a suit for nonpayment. But suppose that the debtor offered good money denominated $95 but, as bullion, worth $100. The legal-tender law would not stop the creditor from accepting the offer, and neither party would have reason to seek legal redress.(2) Rolnick and Weber thus focus on price, not quantity, as the magnitude that adjusts to keep the good coin worth no less as coin than as bullion. They do allow for quantity adjustment, but "only when the costs of using good money at a premium are significant" (pp. 195-96).
What costs are these? When do they become significant? The costs are the costs of rounding, and they become significant when applied to small-denomination coins. A 25 percent premium on a nickel, for example, is 1.25 cents. If people round 1.25 cents to a penny, then in a small transaction, a customer using nickels would lose part of the premium. People would therefore bundle nickels for use in a large transaction or else melt nickels for sale as bullion.
Actually, rounding would not be the main cost of using the good money at a premium. The value of the nonpar coin would fluctuate constantly, forcing people to monitor its value and then negotiate in every transaction.(3) Perhaps today, we can imagine that, at virtually every transaction site, a computer network would make a quotation on the nonpar money available instantaneously and inexpensively. Under nineteenth-century conditions, however, using the nonpar coin would have required negotiating its value despite considerable uncertainty about its value in central markets. Most transactors, therefore, would have preferred the par money and left negotiation of the nonpar coin's bullion value to specialists. Irving Fisher made the same point: "The better money might conceivably be used at a premium, i.e., at its bullion value; but the difficulties of arranging payments in it, which would be satisfactory to both parties, are such that in practice it is never so used in large quantities" (1922, p. 114).
We shall thus frame the issue in empirical terms. Before turning to the historical record, however, we should note two points that will play an important part in evaluating it. First, a premium might develop in the bullion market, giving specialists incentive to cull good money from circulation. Historical references to a premium thus require careful scrutiny. Do they refer to a premium in the bullion market, a necessary adjunct of Gresham's law? Or do they refer, instead, to a premium in ordinary transactions, evidence for Gresham's rival?
Second, even with Gresham's law in operation, a small quantity of good money might see use in sheltered markets, perhaps at a premium. Gold, for example, offered key advantages in nineteenth-century international trade. Importers and exporters, then, might have continued using gold even after it had become the dear money. Only if business in unsheltered markets stood as much chance of involving the nonpar money as involving the par money would the case under investigation support the dual-currency hypothesis.
2. THE AMERICAN EXPERIENCE IN THE NINETEENTH CENTURY
When it achieved independence, the United States had no coins of its own manufacture. By the end of the nineteenth century, however, the country had a sophisticated coinage system based on full-bodied gold coins and subsidiary coins of silver and other metals. The path from one point to the other was far from smooth. Changing economic and political forces led repeatedly to the introduction of new dollars intended to circulate alongside or to supplant the incumbent dollar. This process produced numerous tests of the dual-currency hypothesis. Here, we examine most of them, including all that Rolnick and Weber urge as violations of Gresham's law.(4) The test is simple. Did an increased supply of cheap dollars leave both dollars circulating at a fluctuating exchange rate, confirming Gresham's dual-currency rival? Or did the increased supply of cheap dollars drive dear dollars from circulation, con- firming Gresham's law?
Foreign Coins before the Civil War
Foreign silver coins, notably the Spanish milled dollar (the peso), circulated in the United States during most of the antebellum era.(5) Officially, the Spanish dollar contained 373.5 grains of pure silver, whereas the U.S. dollar, established in 1792, contained only 371.25 grains. Rolnick and Weber (1986, p. 187) point to two pieces of evidence suggesting that a heavier Spanish dollar might have circulated at a premium alongside a lighter U. S. dollar. First, in 1830, according to an estimate that J. Lawrence Laughlin reports, the Spanish dollar accounted for 22 percent of the coins circulating in the United States. Second, references exist to a premium of 0.25 to 1 percent on Spanish dollars, especially from 1792 to 1811 and as late as 1819.
In fact, the typical Spanish and Latin American coins in circulation contained less than the official 373.5 grains. Changing mint practices, defacement through gouging, clipping and sweating, and ordinary wear and tear produced the shortfall. In percentage terms, the shortfall was greater for smaller denominations and pieces of the Spanish dollar. (It was common practice to cut up the Spanish dollar to make small change; one eighth of a dollar was known as a bit.
In setting the weight of the U.S. silver dollar, Congress followed Alexander Hamilton's recommendations, as set out in his Treasury Report of January, 1791. Hamilton (1791, pp. 164-65, 186) explained that the Spanish mint had reduced the silver content of the Spanish dollar and that only the most recently minted Spanish dollars still circulated. Heavier Spanish dollars had circulated before the Revolution, but they had long since disappeared. Hamilton cited conflicting assays of the newest Spanish dollar. The assay that he considered the most reliable put the weight of the Spanish dollar then in circulation at about 371 grains, and this assay became the main basis for choosing the weight of the U.S. dollar.
Underweight Spanish and Latin American silver remained a substantial component of the currency throughout the early national period. Laughlin reports that as late as 1830 the U.S. coinage comprised $14 million in U.S. coins, $5 million in Spanish dollars and parts of dollars, and $4 million in other dollars and parts, mostly from Latin America. He (1896, pp. 54-55) describes the Spanish and Latin American silver as "much worn and reduced in weight" and adds that these coins "being in practice current with other coins, without regard to weight, naturally acted to drive out our own coins." In the same vein, Neil Carothers (1930, p. 81) reports: "It was feasible to clip Spanish coins to the legal weight of 415 grains [373.7 fine grains] and coins far below this weight were accepted without question. The silver coins of Mexico and Spain entered into circulation in a degenerate condition and once there they could not be withdrawn. . . . Storekeepers habitually accepted coins in any state of depreciation."
The 1830 Senate report that Laughlin and Carothers relied on claimed that foreign dollars were typically depreciated about 1 percent, half-dollars about 3 percent, eighths of a dollar 11 percent, and so on (New American State Papers, Public Finance, vol. 27, pp. 219-20). A weighted average of the foreign coins in circulation put them about 6 percent below their official weight. 6 The report (p. 22 1), moreover, noted that, "All these coins are equally current, without regard to actual weight," and it concluded that underweight foreign coins were driving out American coins.
The Senate committee had taken aim at laws making foreign coins legal tender. The committee may thus have exaggerated the amount of foreign silver in circulation and the extent of its depreciation.(7) But the committee would not likely have misrepresented the facts of interest here, namely, whether the foreign silver was full weight or depreciated and whether it circulated by weight or by tale, facts anyone reading the report could have verified easily.
In 1842, a book that the U.S. mint's assay office published reported on the Spanish dollar: "Its divisions, though greatly depreciated by wear, continue to circulate largely in this country . . . " (Eckfeldt and Du Bois, p. 117). Legislation passed in 1857 led, finally, to the removal of most foreign coins.
But if the circulation consisted of light foreign silver circulating at par with U.S. coins, what do references to a premium on Spanish coins mean? Merchants in the China trade paid the premium because the Chinese heavily discounted the newer U.S. dollar. Assays made in China showed the true worth of the U.S. dollar, but it lacked the credibility of the Spanish dollar, which had become the medium of exchange in certain areas of China. The premium compensated merchants in the China trade for culling acceptable coins. The premium therefore meant that Gresham's law was working, not that it was being violated. When the premium on the Spanish dollar rose to about 4 percent, merchants in the China trade switched to the U.S. dollar because the Chinese undervalued U.S. silver by about that much.
The confusion over the premium commanded by better specimens of the Spanish peso traces to a footnote of Laughlin's that quotes Secretary of the Treasury Crawford in 1819. "Spanish milled dollars compose the great mass of foreign silver coins which circulate in the United States," Crawford (quoted in Laughlin 1896, p. 53) wrote, "and generally command a premium when compared with the dollar of the United States . . . " The full sentence ends with the phrase, though Laughlin omitted it, "especially for exportation to China" (American State Papers, Finance, vol. 3, p. 395).
Further on, Crawford noted that the Spanish dollar circulated at anywhere from par to a 10 percent premium. Crawford's remarks suggest that bullion dealers paid the premium to acquire full weight coins for use in the China trade and that, depending on local demands and transportation costs, the premium varied from port to port. But a literal reading of Crawford's report would not rule out a premium on coins used in ordinary transactions. Other observers, however, were unambiguous. A House committee examining the currency laws, for example, reported, "Within the United States, it [the Spanish dollar] has an equal value with the U.S. dollar, and in many foreign countries a much higher value" and also, "The dollar of Spain and that of the United States are of exactly the same value within the United States" (American State Papers, Finance, vol. 3, p. 398).
Better specimens of Spanish and Latin American dollars thus disappeared from circulation because they were heavier and useful in the China Trade. The coins left in ordinary circulation, although issued by several governments, circulated at par.
Bimetallism before 1834
Another alleged violation of Gresham's law concerns the circulation of gold before 1834 and of silver afterward. In 1792, the United States set its mint ratio at 15: 1, the dollar containing 371.25 grains of pure silver or 24.75 grains of pure gold. That ratio almost equaled the ratio of world market prices. But soon the market ratio began to climb. According to Soetbeer's figures, it stood at 15.17:1 in 1792, rose to a peak of 16.25 in 1813, and then, after a brief fall, climbed to 15.95 in 1821 (Laughlin 1896, pp. 289-90).
Gold disappeared. According to a House committee report of 1819, "Gold is estimated below its fair relative value, in comparison with silver, and . . . can scarcely be considered as having formed a material part of our money circulation for the last twenty-six years" (American State Papers, Finance, vol. 3, p. 399). In 1821, another House committee reported, "On inquiry, they find that gold coins, both foreign and of the United States, have in great measure disappeared" (American State Papers, Finance, vol. 3, p. 646). In 1830, a Senate select committee stated flatly: "The coins held by the banks and in circulation among the people are silver. All the coins in common circulation, and most of those held by the banks, are half dollars and the minor silver pieces. We have no gold coins in circulation" (New American State Papers, Public Finance, vol. 27, p. 292). And in 1832, the director of the mint, Samuel Moore, noted, "Gold at present constitutes no part of our currency; and not having, within any recent period, performed in the United States the office of coin, it has not been the standard of value assumed in existing contracts" (New American State Papers, Public Finance, vol. 27, p. 325).
The congressional committees and the mint may have erred in terms of causes and cures. But, again, they would not likely have misrepresented the state of the circulation, something that anyone reading their reports would have been familiar with. Laughlin (1896, p. 30) thus stood on solid ground when he concluded, "Gold coins were seldom seen during the largest part of this period from 1792 to 1834."
Rolnick and Weber (1986, p. 188) point to the continued minting of gold--between 1793 and 1833, gold accounted for about 25 percent of the mint's output--as evidence against Gresham's law. Much of the gold coinage was exported (1986, p. 196), however, and much of the rest was sold as bullion. But why would someone have bullion (often obtained by melting foreign coins) struck as U.S. coins and then export the coins or sell them domestically as bullion?
Apparently, dealers converted bullion or foreign coins into U.S. coins to certify weight and purity. The mint took no seigniorage, and the resulting stamp of approval carried more authority than a private assay. The Senate committee of 1830, quoted above, expressed its indignation: "We prohibit and punish all private coinage of gold; we coin this metal at the mint, upon a principle which does not permit it to circulate as money; and we pay the expenses of this useless coinage. In practice, this coinage affords a facility to the possessor of gold bullion; since it enables him to employ the mint to weigh and assay his bullion, and to divide it into convenient portions, without expense to himself. When the coins are received from the mint, they are sold for their value as bullion; some of them are used in manufactures, and the greater part are exported" (New American State Papers, Public Finance, vol. 27, p. 292).
Bimetallism after 1834
In 1834, Congress reduced the weight of the gold dollar from 24.75 grains to 23.2 grains, thereby raising the mint ratio from 15:1 to 16.002:1.(8) This change, given prevailing market prices, made gold the overvalued money. Some silver flowed out and some gold flowed in, but the circulation did not change greatly. The small difference between the mint ratio and the market ratio, a difference that the worn state of the silver coinage made even smaller, permitted both gold and silver to circulate. Around the legal ratio, that is, stood a range of ratios that would not stimulate bimetallic arbitrage, a range analogous to the range the gold points set off under a gold standard (Friedman 1990, p. 90). In the West, relatively high transportation costs kept silver preeminent. Hugh McCulloch, at the time a banker in Indiana--he later served as Comptroller of the Currency and Secretary of the Treasury--wrote (1889, p. 119), "I had been a banker for fourteen years before I handled or saw a dollar in gold except the ten-thaler pieces which were brought into the country by German immigrants."
Gold discoveries in the late 1840s and early 1850s, however, quickened the expulsion of silver. Between 1834 and 1848, the premium on full weight U.S. silver coins, the difference between their internal and external gold values, averaged 1.2 percent. Between 1849 and 1859, it averaged 3.6 percent (Laughlin 1896, p. 297). By the Civil War, gold dominated the metallic circulation.
Greenbacks in the East
The Greenback era (1862-1879) has played an important role in shaping views of Gresham's law. In December 1861, the banks suspended paying specie (gold or silver) for notes and deposits. The Treasury reacted by suspending payment on its "old demand notes," issued the preceding summer (Mitchell 1903, pp. 40-41). In the East, where the money supply consisted largely of notes and deposits, gold immediately went to a premium. Apparently, merchants decided to post prices in terms of the most common medium of exchange, thus minimizing the number of transactions that required negotiating the price of the nonpar money.
The greenback, a fiat paper currency created by the Legal Tender Acts of 1862 and 1863, swiftly filled the channels of trade, serving both as a bank reserve arid as a hand-to-hand currency. In July 1864, the price of gold climbed to its high, $2.85 (Mitchell 1903, p. 424). After the War, a combination of slow growth of money and rapid growth of real income produced deflation. Resumption at the prewar parity occurred on January 1, 1879.
Most historians of the Greenback era consider it an example of Gresham's law. According to Laughlin (1896, p. 87), "The result of the depreciation of the paper money which became manifest by a premium on gold in June [1862] to the extent of 5 percent, and in July of 20 percent, naturally brought Gresham's law into operation, by which the cheaper paper was substituted for the more valuable gold. Gold disappeared before the depreciating paper, and it was not until January 1, 1879 that it again appeared. " Don Barrett (1931, p. 73) says, "Gold practically disappeared from circulation for a period of eighteen years." Francis A. Walker, a superb economist who lived through the period, says that gold saw no use (except by special contract) other than to pay customs duties and interest on the national debt (1883, p. 508).
Wesley C. Mitchell, the leading authority on the period, believed that, after the greenback appeared, gold no longer circulated in the eastern United States. His table (1903, p. 179) of "The Circulating Medium" shows no gold circulating in the loyal states, except on the Pacific coast.
A second piece of evidence counting against the dual-money view of the Greenback era is the fate of the national gold banks. Legislation of 1870 permitted the organization of national banks that could issue notes redeemable in gold. Had a large quantity of gold remained in circulation, gold banks would have succeeded throughout the country. Gold banks succeeded nowhere, however, but in California. The one eastern gold bank, the Kidder National Gold Bank of Boston, quickly liquidated, having found no demand for its notes (Greenfield and Rockoff 1995).
Mitchell and the others used a medium-of-exchange definition of money. Even if we define money more broadly, however, we find that specie played a minor role during the Greenback era. In face value terms, the stock of specie held by the public fell from about $149 million in 1859 to about $55 million in 1867. Comparisons with the stock of money make the point even more dramatically. In 1859, 26 percent of the money stock consisted of specie; by 1867 the figure had fallen to about 6 percent. And this figure includes gold circulating in California, where the greenback never circulated. Eliminating California's gold reduces the ratio of specie to total money to a little more than 3 percent. Subtracting specie holdings tied to paying customs duties cuts the ratio even further.
Figure 1, which shows the ratio of specie to money between 1850 and 1890, provides a longer-term perspective.9 The lower lines adjust the gross ratio to eliminate both specie circulating in California and gold held to pay customs duties. Specie used in California is Mitchell's estimate, based on Treasury opinions. Gold used in customs duties is based on the aggregate currency-income ratio and the amount of customs duties.
The ratio exhibits a falling trend. During the Civil War, however, the ratio falls well below trend. By 1875, probably less than I percent of the money stock, after the adjustments, consisted of specie. Figure I thus confirms Gresham's law and contradicts its rival.
Extensive hoarding began, apparently, when only a small increase in the amount of cheaper money had actually occurred. People expected the war to produce even larger increases. The government's fiscal position, which depended on the course of the war, conditioned these expectations. Richard Roll's (1972) and Charles Calomiris's (1988) finding that war news determined the premium on gold is thus consistent with a Gresham's-law interpretation of the period.
Gold did not actually leave the country until a few years after disappearing from circulation. Annual net gold exports for the years 1862 to 1865 averaged $54.75, as compared with $34.25 million in the preceding four years. Net gold exports peaked at $90 million in 1864. No other year in the nineteenth century recorded larger net gold exports, though 1893 came close (U.S. Bureau of the Census 1975, p. 885).
Silver disappeared along with gold. The silver did not wind up in the melting pot. Instead, it went to Latin America, and in lesser amounts to Canada, to serve as currency (Carothers 1930, p. 219). Later, when the greenback price of silver fell, a substantial reflux occurred (Carothers 1930, p. 259). During the Civil War, the U.S. mint continued to produce silver coins, but they did not circulate. According to Carothers (1930, p. 219), "The mints were, so far as silver was concerned, merely establishments conducted for the benefit of New York and San Francisco bullion dealers." The mints certified the billion's weight and fineness and produced coins suitable in foreign circulations.
A sheltered gold market remained active throughout the Greenback era. Servicing and paying both the government's and the railroads' bonded debt required gold, as did paying customs duties and buying things abroad. Even Laughlin and the other writers who took this period to confirm Gresham's law recognized the existence of this market.
Greenbacks in California
The legal-tender status of the greenback gave it a strong starting point in California. But the social pressure to use gold, a major export, proved insurmountable. California had shown a preference for hard money even before the War. The constitution, written in 1849, prohibited paper money. A few private banks had issued notes, defying the law, but these notes had difficulty swimming against the tide of public opinion.
Greenback inflation, of course, did nothing to warm Californians toward paper money. Social preferences became private costs. Debtors who tried repaying in greenbacks found themselves denounced for "greenbacking," their names published in the newspapers. The greenback remained foreign exchange, quoted at a discount from gold that mirrored the eastern premium on gold. In July 1864, for example, a greenback dollar sold for $.35 gold in California (Wright 1916/1980, p. 87). Paper money did circulate in California after 1870, when legislation established national gold banks. But the "yellowback dollar" was convertible into gold (Lester 1939, and Greenfield and Rockoff 1995).
Does the greenback's failure in California represent a defeat for Gresham's law? One interpretation would regard this case as lying outside the scope of Gresham's law because Californians considered the greenback foreign exchange, not money. Interpretations, however, may vary. Since the greenback had legal-tender status, critics could insist that it qualified as "bad money" and should have driven out gold, even in California. But whether or not the greenback in California counts against Gresham's law, this case cannot count for Gresham's rival. Domestically, gold and the greenback did not circulate alongside one another.
The Trade Dollar
In 1873, hoping to encourage the China trade, Congress authorized the Trade dollar, which contained 378 grains of silver. (10) The Trade dollar, according to Rolnick and Weber (1986, p. 190), circulated at its bullion value, about 93 cents. Again, we have something that looks like Gresham's rival, though, oddly, the cheap money appears to have carried a discount.
The full story of the Trade dollar is indeed curious. Although intended for the China trade, it circulated at par in certain areas of the United States, and large amounts were minted. Congress, apparently inadvertently, had made the Trade dollar legal tender for sums under 5 dollars (Garnet 1917, p. 96).(11) The Trade dollar, moreover, looked very much like a coin intended for domestic circulation and actually contained more silver than the old Spanish dollars or U.S. half-dollars. Merchants hesitated to refuse the coin. In 1876, however, Congress revoked the Trade dollar's legal-tender status, and in 1878 the mint discontinued production, Even so, the Trade dollar continued to circulate and, but for some form of collective action, might have continued to do so for a long time.
The federal government and business trade associations, however, set out to demonetize the Trade dollar. The government took a variety of actions against it, such as prohibiting the Post Office from accepting it. The trade associations objected to the banks' unwillingness to accept the coin, except for large deposits by favored customers and even then at its bullion value. Although we haven't found a specific regulation, the Trade dollar might not have fulfilled reserve requirements because it was not a legal tender.
In California, the Trade dollar disappeared by 1878 (Willem 1959, p. 126). In New York, however, the matter did not come to a head until 1883. The New York Times, in June 1883, reported that, at a meeting of produce dealers, a strong disagreement arose over whether to ban par acceptance of Trade dollars (Willem 1959, p. 133). Bankers Magazine (cited by Taxay 1966, p. 284) reported that pressure to refuse Trade dollars had succeeded completely in certain lines of business and partially (credit but not change for Trade dollars) in others.
In June 1883, posters appeared in New York City, probably placed there by the trade associations, warning against accepting Trade dollars. According to Willem (1959, pp. 133-34), the Trade dollar immediately ceased to circulate and headed toward the bullion dealers. After 1883, the Trade dollar was used in the China trade but not in the United States.
Pressure to redeem the Trade dollar grew. The argument, at least for public consumption, focussed on ordinary citizens who had accepted Trade dollars at face value. Formulating a redemption bill that would satisfy both the silver and the gold interests in Congress, however, proved difficult. Legislation authorizing redemption was finally passed in 1887. For six months, the government redeemed unstamped, full-weight Trade dollars at face value, paying standard silver dollars, the Bland dollars. (12)
The Bland Dollar
Congress, under pressure from silver interests, had authorized the Bland dollar, which contained the usual 371.25 grains, in 1878. It had unlimited legal-tender status and, when delivered to the Treasury as a bundle of ten, could be exchanged for $10.00 in silver certificates. Like gold, the silver certificates were accepted at par for customs duties. Because the market price of silver stood below the mint price, coining Bland dollars offered considerable profit. Under free coinage, the Bland dollar might have driven out the gold dollar. But the Bland-Allison Act limited the treasury's silver purchases to between $2 million and $4 million per month, and generally, purchases stood nearer the lower limit (Friedman and Schwartz 1963, p. 132). Because the supply of cheap money was limited, the dear money continued to circulate. The Bland dollar, in other words, was a subsidiary coin.
3. CONCLUSION
We have investigated seven cases in which bad money could have driven good money to a premium: (1) the U.S. silver dollar versus the foreign silver dollar in the early national period; (2) the silver dollar versus the gold dollar when, before 1834, the mint overvalued silver; (3) the silver dollar versus the gold dollar when, after 1834, the mint overvalued gold; (4) the greenback dollar versus the gold dollar in the east during and after the Civil War; (5) the greenback dollar versus the gold dollar in California during and after the Civil War; (6) the Trade dollar versus the gold dollar, 1873-1887; and (7) the Bland dollar versus the gold dollar, 1878-1890. None of these seven cases offers evidence that bad money drove good money to a premium.
Nor, on balance, does the evidence speak badly for Gresham's law. In cases (1)-(4), either the bad money was not bad enough or it did drive out the good money. In case (6), the bad money, the Trade dollar, lost monetary status before Gresham's law could operate. And in case (7), the restricted supply of the cheaper money, the Bland dollar, kept it a subsidiary coin.
Only case (5) presents any ambiguity concerning Gresham's law. In California, the greenback, despite its legal-tender status, never gained currency. One interpretation would put this case outside the scope of Gresham's law; another interpretation would consider this case a contradiction of Gresham's law. Whatever its verdict concerning Gresham's law, however, even case (5) gives no support to Gresham's rival.
A dual-currency system, which the rival view envisages, never materialized. Evidently, the transactions costs of using a nonpar money were substantial. (1.) Both Fisher (1922 [1971], pp. 115-22) and Fetter (1932) stress this point. (2.) This conclusion depends on how the authorities write and enforce the law. A referee points out that the French revolution mandated death for a seller who asked what type of money the buyer would use. We have the less drastic, American-type law in mind. (3.) Warren Weber points out that monitoring the supply of bad money also entails costs. In some circumstances, therefore, some people might prefer to use, say, gold coins at a premium despite the costs of monitoring the bullion market. (4.) The most important omissions occurred in the South during the Civil War. Gary M. Pecquet's (1987) finding that gold circulated at a premium in the Trans-Mississippi Confederacy supports Gresham's rival. But the cases in which he finds the Confederate currency to have lost monetary status resemble two cases that we discuss below: the greenback in California and the Trade dollar. Pecquet, furthermore, cites an interesting example of Gresham's law: the Confederate currency reform of 1864 gave an advantage to small bills, which were then hoarded while the large bills continued to circulate. (5.) Schilke and Solomon (1964) give an excellent numismatic history of this
(6.) According to McCullough (1984, p. 105), actual weights fell substantially below official weights in Lower Canada.
1830 1840
Spanish Dollar -.6% 1.1%
U.S. Dollar NA .5
Spanish Quarter -4.7 -6.4
U.S. Quarter NA .9(7.) The committee criticized the Mint's 1826 study, which reported much less depreciation (American State Papers, Finance, vol. 5, pp. 585-92). (8.) In 1837, the weight of the gold dollar was increased to 23.22 grains for technical reasons, making the ratio 15.988 to 1. (9.) Although the ratio is conceptually simple, calculating a consistent series requires many steps. The details appear in an appendix, which we will (10.) Willem (1959) gives an excellent numismatic history of the Trade dollar. (11.) Carothers (1930, p. 234), however, maintains that the decision to make the Trade dollar a legal tender was taken deliberately. (12.) Chinese merchants stamped coins (leaving "chop" marks) to show that they were genuine."
LITERATURE CITED
Barrett, Don C. The Greenbacks and Resumption of Specie Payments, 1862-1879. Cambridge: Harvard University Press, 1931.
Calomiris, Charles W. "Price and Exchange Rate Determination during the Greenback Suspension." Oxford Economic Papers 40 (1988), 719-50.
Carothers, Neil. Fractional Money. New York: John Wiley and Sons, Inc., 1930.
Eckfeldt, Jacob R., and William E. Du Bois. A Manual of Gold and Silver Coins of All Nations. Philadelphia: The Assay Office of the Mint, 1842.
Fetter, Frank Whitson. "Some Neglected Aspects of Gresham's Law." Quarterly Journal of Economics (May 1932), 480-95.
Fisher, Irving. The Purchasing Power of Money. New York: Augustus M. Kelly, Reprints of Economic Classics [The Macmillan Company], 1971 [1922].
Friedman, Milton. "Bimetallism Revisited." Journal of Economic Perspectives (Fall 1990), 85-104.
Friedman, Milton, and Anna J. Schwartz. A Monetary History of the United States. National Bureau of Economic Research, Princeton, N.J.: Princeton University Press, 1963.
Garnet, Porter. "The History of the Trade Dollar." American Economic Review (March 1917), 91-97.
Greenfield, Robert L., and Hugh Rockoff. "A Tale of Two Dollars: Currency Competition and the Return to Gold, 1865-1879." In Money and Banking: The American Experience, edited by Clifford H. Thies. Fairfax, Virginia: George Mason University Press for the George Edward Durell Foundation, 1995.
Hamilton, Alexander. "Treasury Report on the Establishment of a Mint, January 28, 1791." Reprinted in Documentary History of Banking and Currency in the United States, edited by Herman E. Krooss, vol. 1, 162-92. New York: Chelsea House Publishers, 1969 [1791].
Laughlin, J. Lawrence. A History of Bimetallism in the United States. New York: D. Appleton and Company; reprint, New York: Greenwood Press, 1968 [1896].
Lester, Richard A. "Retention of the Gold Standard in California and Oregon during the Greenback Inflation." In his Monetary Experiments: Early American and Scandinavian. Princeton, N.J.: Princeton University Press, 1939.
McCulloch, Hugh. Men and Measures of Half a Century. New York: Charles Scribner's Sons,1889.
McCullogh, A. B. Money and Exchange in Canada to 1900. Toronto: Dundurn Press Limited, 1984.
Mitchell, Wesley C. A History of the Greenbacks. Chicago: University of Chicago Press, 1903.
Pecquet, Gary M. "Money in the Trans-Mississippi Confederacy and the Confederate Currency Reform Act of 1864." Explorations in Economic History 24 (April 1987), 218-43.
Roll, Richard. "Interest Rates and Price Expectations during the Civil War." Journal of Economic History 32 (1972), 476-98.
Rolnick, Arthur J., and Warren E. Weber. "Gresham's Law or Gresham's Fallacy?" Journal of Political Economy 94 (1986), 185-99.
Schilke, Oscar G., and Raphael Solomon. America's Foreign Coins with Legal Tender Status in the United States, 1793-1857. New York: Coin and Currency Institute, 1964.
Taxay, Don. The U.S. Mint and Coinage. New York: Arco Publishing Company, Inc., for the U.S. Bureau of the Census, 1966.
U.S. Bureau of the Census. Historical Statistics of the United States, Colonial Times to 1970. Washington, D.C.: GPO, 1975.
Walker, Francis A. Money. New York: Henry Holt and Company, 1883.
Willem, John M., Jr. The United States Trade Dollar. New York: Marchbanks Press, 1959.
Wright, Benjamin C. Banking in California, 1849-1910. San Francisco: H.S. Crocker Co. [Arno Press, New York], 1916 [1980].
The authors thank Michael Bordo, Stanley Engerman, Milton Friedman, John M. Kleeberg, Roger Koppl, Angela Redish, Arthur Rolnick, Anna J. Schwartz, Warren Weber, Eugene White, and Leland Yeager for helpful comments on earlier drafts. Also, the University of California, Berkeley, the National Bureau of Economic Research Summer Workshop on the Development of the American Economy, and Rutgers University gave us much appreciated opportunities to present the paper.
Robert L. Greenfield is professor of economics and finance at Fairleigh Dickinson University. Hugh Rockoff is Professor of economics at Rutgers University.
http://www.allbusiness.com/finance-insurance/credit-intermediation-related-activities/539901-1.html