Yellowbacks out West and greenbacks back East: social-choice dimensions of monetary reform.

I. Introduction

Oftentimes, in the wake of political developments, monetary questions follow. Should the European Economic Community force adoption of a single money? Or can the job of designating Europe's money be left to the market? Has the rapid dissolution of the Soviet Union left a monetary vacuum? If so, what will fill it? Economists find themselves being pressed to answer these questions and others like them.

In the 1970s, Friedrich A. Hayek [18] proposed safeguarding against governmental inflationary abuses by giving private issuers of money full scope to operate. Under Hayek's scheme, each issuer could adopt a distinct unit of account, and the various units would fluctuate freely against one another. Profit incentives would restrain overissue, because people would not hold units lacking purchasing-power stability. A unit's stable purchasing power would strengthen the demand for that unit, enabling the issuer to have more loans outstanding and earning interest. Even if one unit should come to dominate, the threat of potential competition would check inflationary tendencies.

Sympathy for Hayek's scheme might well lead to recommending that questions of monetary reform be left to the market. Benjamin Klein [21], Milton Friedman and Anna J. Schwartz [16], and Lawrence H. White [29], however, have expressed skepticism about competition as an agent of monetary reform. "So far," Friedman and Schwartz [16, 46] caution, "neither Hayek's belief that privately produced constant purchasing power moneys would become dominant nor Klein's and our skepticism has any direct empirical basis, but derive rather from an interpretation of historical experience under very different monetary conditions than those Hayek proposes."

There now exists, of course, as Friedman and Schwartz note, a large and growing literature on "free banking." The historical studies that we have seen, however, examine banks issuing money convertible into an established base money - gold or silver coins or else fiat money. Thus, the free banking literature does not speak directly to Hayek's proposing free competition in the supply of base money.

"Some direct evidence," Friedman and Schwartz continue, "may emerge in the near future because of developments within the present system...." They cite the 1974 repeal of the prohibition against privately owning, buying, and selling gold. "In principle," they explain [16, 47], "it has been possible since then for individuals in private dealings to use gold as a medium of exchange. And there have been some minor stirrings. The Gold Standard Corporation in Kansas City provides facilities for deposits denominated in gold and for the transfer of such deposits among persons by check. However, this is a warehousing operation - a 100 percent reserve bank, as it were - rather than a private currency denominated in gold and issued on a fractional reserve basis. Unfortunately, there are currently legal obstacles to any developments that would enable gold to be used not only as a store of value or part of an asset portfolio but as a medium of circulation. Hence, the current situation provides little evidence on what would occur if those obstacles were removed."

This paper explores a neglected, American-reconstruction-era experiment that directly tests the feasibility of monetary reform through competition in base moneys. The experiment, which the day's leading financiers and politicians gave their support, had two sides. In the East, the yellow-back dollar, a new, gold-denominated and fractional-reserve currency, challenged the greenback dollar, the entrenched fiat currency. Out West, however, challenger and incumbent exchanged positions. There, the greenback dollar, or, equivalently, the banknote denominated in the greenback dollar and backed by fractional reserves of it, challenged the entrenched gold currency. In both cases, furthermore, the challenger had legal-tender-like properties, making for experimental conditions of a kind that economists rarely see.

II. A Tipping Model of Acceptance and Rejection

To account for the experiment's results - yellowbacks out West and greenbacks back East - we begin with some basic questions. How does a business decide in which money to post prices? How does a worker decide in which money to request wages? How does a person decide in which money to hold transactions balances? Numerous considerations come into play, including the expected bullion value of each money, its legal-tender status, its denominational structure, and the policies that banks and trade associations urge. The position that political parties take representing debtors, creditors, or even producers of monetary metals come into play, too.

The greatest influence that any individual decision maker feels, however, comes from what everyone else does. If most people use a particular money, then everyone else has good reason to use it. If, however, most people refuse to use a particular money, then no one else has much reason to use it. Only two outcomes, general acceptance or general refusal, therefore, are "Schelling points," and the monetary system will normally tip toward one Schelling point or the other [27].

Figure 1 illustrates the tipping tendency. Net benefits accruing to any firm because it posts prices in greenbacks increase as the percentage of all firms posting prices in greenbacks increases - and similarly for posting prices in gold. The greenback has a clear advantage. Even if, to start, fifty percent of all firms post prices in greenbacks, then all firms see that the benefits of posting prices in greenbacks exceed the benefits of posting prices in gold. The system tips to the right; everyone posts greenback prices, everyone receives greenback wages, and everyone holds greenback transactions balances. Tipping toward gold could occur, though it would require that, to start, sufficiently more than fifty percent of all firms posted gold prices.

David Laitin [22, 295-96] employs the tipping model to examine regional language conflicts, focusing on cases in which local political pressures cause the system to tip against using the language that the central government prefers. Perhaps even more effectively with money than with language, however, a self-reinforcing selection process narrows the field. Two parties can converse in whatever language they agree to use. The chain of commerce, however, includes links that extend beyond the two parties involved in any particular transaction. The seller in one transaction, for example, expecting to become the buyer in some later transaction, has to take the second seller's medium-of-exchange preference into account.

A neutral observer might consider the actual Schelling equilibrium the inferior outcome. In terms of long-run price stability, for example, the greenback equilibrium might seem inferior to the gold equilibrium. But once greenbacks become generally accepted, as they had back East, no individual has an incentive to change. Only much altered external pressures, which in Figure 1 means shifts that put the gold-benefits curve everywhere above the greenback-benefits curve, could push the monetary system back, past its initial tipping point. Bearing in mind, then, the social-choice dimensions of monetary reform, we turn to the history of the national gold banks.

III. John Sherman and the National Gold Banks

Usually, monetary historians remember the act of July 12, 1870, as the act that raised the authorized quantity of national banknotes from $300 to $354 million to address currency shortages in the Midwest, West, and South (3, 18; 15, 21, n. 7; 28, 94-96). But the act did something else, too. It authorized establishing a second class of national bank - the national gold bank.

Although historians have neglected the gold-bank provision of the act, Senate finance committee chairman John Sherman, who steered the bill through Congress, took the provision quite seriously. "We have assurances," Sherman [10, 700] said on January 24, 1870, when he brought the currency bill out of the finance committee and to the Senate floor, "that banks will be organized under this system at once on the Pacific coast, thus unlocking . . . a portion of the gold that must necessarily be used now for the ordinary channels of circulation."(1)

California seemed to offer a fertile environment for the national gold banks. After banks and the government had suspended convertibility in 1861 and even after the government had issued legal tender notes, Californians stuck with gold.(2) And though ignored since 1864, the statute that had codified the state's 1850 constitutional language prohibiting corporate commercial banking had just been officially deemed "obsolete" [1, 22]. Though California's aversion to banks had softened, its aversion to "national" banks, which kept their notes convertible into the greenback, had not.(3) In 1870, the country as a whole had 1,612 national banks, but California had none.

"In the city of New York," Sherman [10, 700] continued, "there is now a commerce going on of over six hundred million dollars, all of which is carried on the gold basis; and so great is the necessity for paper money to represent this gold business that they actually deposit $50,000,000 of gold coin in the Treasury of the United States, and receive gold notes without interest, merely to facilitate the ordinary transaction of . . . this business.

"So in the cities of Charleston and New Orleans, where cotton is measured by the gold standard, they can very readily use these coin notes. . . . These banks will furnish to the people a sure currency based on coin, payable in coin, having all the requisites that is possible to have to provide the best national currency."

Most everywhere, the greenback had become the routine medium of exchange. Gold, however, as Sherman reminded the Senate, had not disappeared entirely from financial use. Customs duties required payment in gold, and the Public Credit Act of 1869, though not without some political opposition, had reaffirmed the government's intention to use gold in paying its bonded debt and the interest on it. Fully backed, large-denomination Treasury gold certificates facilitated much of this business, but banks, especially banks in New York, accepted gold deposits and held a large gold reserve behind them [15, 28-29, n. 17].

IV. The Kidder Bank

One national gold bank was launched almost "at once," just as Sherman assured his fellow senators would happen. But it would turn out to be the only one east of the Mississippi, and its fate would not encourage others. "Under the provisions . . . of the act approved July 12, 1870, authorizing the establishment of national gold banks for the issue of circulating notes redeemable in specie," the Comptroller of the Currency reported in 1870 [8, vii, viii], "but one bank has yet been established, the Kidder Bank of Boston. . . . It was not anticipated that specie-paying banks would be established to any considerable extent, at present, in those sections of the country where a paper currency, based upon the legal tender issues of the Government, already prevails; although it was, and still is, supposed that one or more gold banks might be established and successfully conducted in each of those cities on the Atlantic seaboard where a considerable foreign trade is carried on, and in which a certain amount of business is necessarily transacted on a specie basis. . . . If all the business of this kind that is carried on in the cities of Boston, New York, Philadelphia, and Baltimore could be concentrated in one or two banking institutions of this kind, its extent would undoubtedly warrant the employment of a very respectable amount for its exclusive accommodation."

The Kidder Bank had an interest in a business transacted on a specie basis, but that business appears not to have been foreign trade. The bank's president, Henry P. Kidder; vice president, Francis H. Peabody; and cashier, Oliver W. Peabody, were the major figures in the Boston investment-banking house, Kidder, Peabody & Company. Several of the Boston investment houses had become associated with the ailing, western railroads, and in 1870, Kidder, Peabody & Company undertook the rehabilitation of the Atcheson, Topeka, & Santa Fe Railroad [7, 204].(4)

The railroad based itself in Boston, and for a while, Santa Fe securities were owned almost wholly by Bostonians and listed nowhere but on the Boston Stock Exchange. In 1871, Francis L. Peabody, the Kidder Bank's vice president, joined the railroad's board of directors. Soon thereafter, he took the chairmanship of its finance committee, and later the railroad's vice presidency became his. Since the railroad paid its financial consultant in stock, Kidder, Peabody became one of the road's major shareholders, and Oliver W. Peabody, a Kidder, Peabody partner and the Kidder National Gold Bank's cashier, "took an almost proprietary interest in the railroad" [6, 34].

Kidder, Peabody and Company became not only the railroad's agent and owner but also, through the gold bank, its creditor. The bank examiner's report dated June 19, 1871, shows the Kidder Bank holding a $90,000 loan due from the railroad.(5) In fact, the examiner's report cites the bank for having violated the National Bank Act's section 29, which prohibits a national bank's lending to a single borrower a sum greater than ten percent of the bank's paid-in capital.(6)

Typically, investment-banking houses had ties to commercial banks. Its lengthy correspondence with the Comptroller suggests, however, that Kidder, Peabody hoped that its gold-bank would accomplish more than merely providing assured access to bank credit.

Upon organizing, a national bank had the Comptroller prepare plates and use them to print banknotes in specified denominations. A gold bank, besides paying its plate fee, had to deposit $125 par value U.S. bonds for every $100 gold notes delivered to the bank. The Kidder National Gold Bank, in a letter of July 27, 1870, asked the Comptroller to print gold notes - $175,000 to come as $1,000 bills, $50,000 as $500 bills, $10,000 as $100 bills, and $5,000 as $50 bills - totaling $240,000. "Against which," Kidder said, "this institution is prepared to deposit at any time $300,000 of United States bonds."

Especially for 1870, notes of these denominations seem much too large for hand-to-hand use. But its May 8, 1871, letter clarified the bank's intentions. "So far as we have tried the experiment of a gold currency," wrote Kidder, "the only use under existing law will be as a convenient medium of transactions between banks, to save handling the coin. Before the banks will consent to use it for this purpose, they ask whether they can count the notes as part of their reserve, a question which we are obliged to refer to you, being unable to answer it ourselves."

Apparently, the firm considered this privilege critical to the bank's success. Just three weeks earlier, on April 14, 1871, Oliver Peabody, in his own letter to the Comptroller, had written, "I presume that under the law, any Nat'l bank keeping deposits with this bank is authorized to treat the same as part of its reserve. Will you be good enough to let me know whether I am justified in stating that this is the case?" Ordinary national banks had to satisfy reserve requirements by holding "lawful money," whether greenbacks, gold coin, or Treasury gold or silver certificates. Why should the Comptroller not allow national banks to count as reserves their "coin notes" and deposits with gold banks?

Kidder, Peabody & Company had floated Atcheson, Topeka, and Santa Fe Railroad bonds. Like other railroad bonds, Santa Fe bonds promised payment, both interest and principal, in gold.(7) As the railroad's agent, then, the Kidder Bank would pay out these banknotes when it serviced the railroad's debt and, in return, would take accommodation paper from the railroad. The notes would be deposited to gold accounts issued by New York or Boston national banks, and if things went according to plan, those banks would hold the Kidder notes as a reserve against the new gold accounts. Better still, the banks might even make deposits to their own accounts with the Kidder Bank, thereby giving the Bank repeated use of the notes. Perhaps Kidder could even pay out checks that it drew on itself and that the banks would deposit to their Kidder accounts. The Kidder National Gold Bank was to be something of a central bank, in that its monetary liabilities would serve as reserves for other national banks.

G. D. Hancock [17, 610] maintains, "[T]he gold banks could doubtless have found a profitable business . . . if they could have taken all of the gold business which the other banks carried." The gold banks, however, "could offer no facilities or inducements not offered by the other banks," too. Kidder, Peabody recognized this and sought the necessary inducements in having the Comptroller declare the gold deposits and notes a legal reserve for ordinary national banks.

The Comptroller's side of his correspondence with the bank is not available.(8) We cannot say for certain, therefore, that the Comptroller denied the Kidder Bank's request to have its notes and deposit liabilities accorded the status of a legal reserve. Possibly, the Comptroller acceded to the request but the receiving banks presented the Kidder gold notes and checks for redemption in gold nevertheless. In any case, when, on June 19, 1871, the examiner visited the bank, all but $50 of the bank's $120,000 circulation sat in the vault.(9) The Kidder Bank had found no demand for its notes.

The simple truth seems to be that the Kidder Bank, the only attempt in the East to establish a gold bank, miscarried because the notes it issued were not money. In his comments, the bank examiner put the matter this way, "There appears to be little use for a bank of this type in this part of the country." The system had gone past its tipping point.

V. Financial Explanations?

We give a monetary explanation of the gold banks' failure to take hold in the East. Is there a plausible alternative, a financial explanation? It would go this way. Gold was depreciating; between 1870 and 1879 the greenback-dollar price of gold fell from $1.23 to $1.00. This erosion of the premium on gold occurred as a matter of deliberate policy. The Republicans wanted a return to gold at the prewar parity. People would not hold an asset, gold, that they were certain would depreciate. Had people expected gold to appreciate or even hold steady, the gold banks might have taken hold, or so the financial argument would go.

Several readers of an earlier draft of our paper urged this explanation. Two considerations, however, lead us to reject it. First, comes the economic implications of the period's tenuous political balance. Resumption at the prewar parity was highly controversial. The inflationists stood a chance of not only blocking the return to gold but perhaps also gaining the upper hand. And if they did gain the upper hand, large issues of greenbacks would have resulted, taking the price of gold back up.

Reflecting this possibility, the price of gold did not follow a steady downward course. Instead, from 1871 to 1875, the price fluctuated in a band around $1.13. Only after 1875 and passage of the Resumption Act, did it descend steadily. Evidently, passage of the act did contribute to quieting fears that the inflationists would prevail.

As late as 1877, however, inflationists controlled both houses of Congress, and they made a determined effort to repeal the Resumption Act. The repeal bill passed in the House and, according to Davis Rich Dewey [12, 375], failed in the Senate "only through disagreement on details." Rutherford B. Hayes, a hard-money Republican, occupied the White House, true. But as the disputed election of 1876 shows, the Republican hold on the office was by no means certain. If an inflationary regime did not lurk in the wings, then why did the price of gold remain as high as it did? If people holding gold were certain that its price would soon fall to $1.00, they would have dumped gold and immediately driven its price down to $1.00.

Second and more decisive against the financial explanation, however, comes the mirror-image experiment being conducted on the West Coast.

VI. The Yellowback Dollar

In California, the depreciating asset, gold, served as the routine medium of exchange. The appreciating asset, the greenback-convertible banknote, was the government-sponsored alternative. But just as no gold bank took hold in the East, no greenback bank took hold in California. If the expected depreciation of gold kept the gold banks from taking hold in the East, then what kept the greenback banks from taking hold in California? After all, Californians were no less inclined to speculate than their eastern relations.

Here we have the monetary experiment's other side. Before 1871, California had no national banks, even though the greenback, into which national banks kept their own notes convertible, was a legal tender and even though Californians, like easterners, must have expected the greenback to appreciate as the price level continued to drop and specie resumption drew closer. Each of California's first nine national banks organized under the act of July 12, 1870, as a national gold bank. Not until 1880, would California have a bank issuing greenback-convertible national banknotes. And by that time, of course, the greenback itself was convertible into gold.

Table I. Monetary Liabilities of the California Gold Banks

Year          Banks          Notes          Deposits

1871            1          $ 277,060       $ 199,138
1872            3          1,366,175       3,144,398
1873            5          1,988,430       3,193,391
1874            6          2,107,915       5,405,976
1875            9          2,171,877       3,653,602
1876            9          1,414,485       2,499,454
1877            9          1,399,295       2,984,922
1878            9          1,436,545       3,828,136
1879            8          1,450,595       2,869,830

Source: Annual Report of the Comptroller of the Currency [8].

"Since my last report," the Comptroller reported in 1871 [8, vii], "but one bank has been established on a gold basis - The First National Bank of San Francisco. . . . It is presumed that the success of this bank is not so flattering as to induce the organization of others of a similar character, though, in view of the obstacles and the opposition it meets, it holds its own and is gradually winning its way into public confidence."

"The tenacity with which the Pacific States adhere to a gold currency is quite notable," the Comptroller went on.(10) "Whether it is equally praiseworthy is another thing. It is not clear that these states derive any substantial benefit from the course they have pursued, and it is beginning to be manifest that the United States are not at all benefitted by it. The substitution of a paper currency in California and other gold-producing states for their present hard money would probably set free for the use of the Government and the whole country some thirty or forty millions of gold. . . ."

A year and one-half elapsed before the Comptroller could charter another gold bank, The National Gold Bank and Trust Company, also located in San Francisco. Before 1876, however, seven others appeared: The First National Gold Bank of Oakland and banks each with a corresponding name in Stockton, Petaluma, and Santa Barbara; The Union National Gold Bank of Oakland; The First National Gold Bank of D. O. Mills and Company, Sacramento; and The Farmers' National Gold Bank of San Jose [8].

The national gold banks of California, though slow in organizing and relatively few in number, took root. Table I shows their total monetary liabilities, both notes - which Californians called "yellowbacks" [11, 274] - and deposits.

As compared with greenback banks, the gold banks suffered distinct regulatory disadvantages. The gold-bank provision of the July 12, 1870, act stipulated that: a) deposit of an eligible $100 par bond standing at par or better would secure $80 in gold notes, not $90, as for greenback notes and b) the national gold banknote would carry a 25 percent reserve-requirement ratio, which remained in effect even after 1874, when by replacing it with a banknote redemption fund that counted as reserves against deposits, too, Congress effectively relieved the greenback banknote of reserve requirements.

To what degree did these stipulations work against the gold banks? To answer this question, we compare the rate of return on issuing greenback-convertible notes in California (assuming temporarily that they could be circulated) with the rate of return on issuing gold-convertible banknotes. To calculate the rate of return on greenback-convertible notes, we use the method that Cagan [3, 22-23; 4, 86-95] and Cagan and Schwartz [5] present.

That method builds on assuming that banks took full advantage of the opportunity to issue greenback-convertible banknotes. Rather than lending directly to business, then, a bank would have used its notes more profitably, by acquiring additional bonds to secure yet additional notes. The note-issuing operation would then have amounted to the bank's acquiring a $100 par value U.S. bond by investing just the bond's price less the $90 in banknotes that the bank could issue after depositing the bond with the Comptroller ($11, if, for example, the bond sold for $101). To get its net dollar return, the bank would have to deduct from the bond's coupon the costs of holding reserves and the taxes on and expenses of issuing the banknotes. Expressing the net dollar return on a $100 par bond as a fraction of the bank capital invested would give the rate of return on issuing greenback banknotes. Equation (1) is Cagan's equation for the rate of return on issuing greenback-convertible notes, though adapted for the pre-1874 period, when reserve requirements applied.

rate of return = [[i.sub.B][P.sub.B] - [Ri.sub.A] - .01(90)]/[[P.sub.B] - 90] (1)

where we use:

[i.sub.B] to denote the government bond's yield to redemption,

[P.sub.B] to denote the government bond's bond price,

R to denote the required reserves (eliminated by an act of 1874) against notes issues,

[i.sub.A] to denote the yield on an alternative asset, and .01 to denote the tax rate on notes issued.

Column A of Table II shows our calculations, based on equation (1).

Next, we adapt Cagan's formula, equation (1), for use in calculating the return on issuing yellowbacks. For issuing yellowbacks,

rate of return = [[i.sub.B][P.sub.B] - .25(80)[i.sub.A](1 + g) - .01(80)(1 + g)]/[[P.sub.B] - 80(1 + g)] (2)

where we use [i.sub.B], [P.sub.B], and [i.sub.A] as before and where, with g defined as the greenback premium on gold, we express all magnitudes in terms of greenback dollars. Column B of Table II shows these rates of return.

Table II shows that, as far as profitability went, bankers each year had an incentive to switch from issuing gold-convertible notes to issuing greenback-convertible notes. Most years, the incentive was substantial. Had Californians been willing to use the greenback as money, their bankers would have switched to issuing greenback-convertible banknotes. But the banks had no opportunity to exploit the advantages that the regulations gave the greenback-convertible banknote. Californians would not adopt the greenback as their medium of exchange.

Narrative accounts of the period make the greenback's status clear [11]. In California, social pressures reduced the net benefits of posting prices in greenbacks. The system therefore tipped toward gold. Once greenbacks had been rejected, merely keeping gold in circulation required little social pressure. Firms posted prices in gold, workers received wages in gold, and people held money balances in gold because everyone else did.

Table II. Rates of Return on Issuing Notes in California

                      (A)                          (B)
Year          Greenback-convertible          Gold-convertible

1870                10.40%                        7.14%
1871                10.29                         6.57
1872                 7.62                         4.35
1873                 6.85                         3.54
1874                 4.17                         1.12
1875                 4.90                         neg.
1876                 2.48                         neg.
1877                 9.21                         .17
1878                14.20                         2.30
1879                14.02                         1.24

Notes:

(1) Reserve-requirement ratio on greenback-convertible banknotes:
.15 before 1874; .00 after 1874.

(2) Reserve-requirement ratio on gold-convertible banknotes: .25.

(3) Alternative interest rate: .10 [17, 620].

(4) Government bonds: U.S. 6s of 1861-1881, prices and yields to
redemption in Homer [19, 309].

(5) Greenback premium on gold in Mitchell [24, 4].

Some quantitative evidence on this score appears in Figure 2, which compares the ratio of greenbacks to deposits in California to the ratio for the United States as a whole. Evidently, banks in other regions of the country held a far higher ratio of greenbacks to deposits throughout this period. The simplest explanation is that banks outside California were called upon to convert deposits into greenbacks more frequently.

Greenback holdings varied dramatically among California banks. In every year, some banks held zero or negligible amounts of greenbacks: in 1873, Santa Barbara held no greenbacks; in 1874, both Stockton and Santa Barbara held no greenbacks; and in 1875, although each of the nine gold banks held some quantity of greenbacks, six of the banks held under $1000. On the other hand, we see episodic cases of very large greenback holdings. In 1877, for example, when three of the nine banks had greenback holdings of less than $1,000, Petaluma held $112,850, about two thirds of the California state total.

Even as a percentage of deposits, greenback holdings varied among the California banks. We would not expect to see this pattern of greenback holdings if greenbacks had circulated routinely as money. The observed pattern reflects the occasional deposit of large amounts associated with irregular transactions. Perhaps, the gold banks, when they did receive large greenback deposits, credited distinct greenback accounts. We have found no direct evidence of distinct greenback accounts in California banks. In the East, however, where the greenback dominated, gold that came in on deposit went to distinct gold accounts, against which the banks held a very large percentage of gold reserves [15, 28-29, n. 17].

If reported by a larger bank, a very high or a very low greenbacks-to-deposits ratio might unduly influence California's aggregate ratio. To present something more nearly like the state's median ratio but still to weight each bank's ratio by the bank's relative size (and thus maintain comparability with the U.S. aggregate ratio), Figure 2 shows a line marked "high-low." "High-low" gives the state's aggregate reserve ratio computed by dropping the greenback and deposit figures of the bank reporting the highest ratio and the bank reporting the lowest ratio. "High-low produces no substantial change, except in 1877; for 1877, dropping Petaluma's figures reduces the average substantially.

Californians considered the greenback foreign exchange, not money. Expecting the anticipated depreciation of gold to have persuaded Californians of the 1870s to adopt the greenback as money would be like expecting Americans in the 1970s, when the dollar had nowhere to go but down, to have started using yen at the supermarket. Californians, of course, could have speculated on the anticipated greenback deflation. But to do so, they would have invested in greenback-denominated assets; they would not have hoarded actual greenbacks or greenback-convertible banknotes.

The yellowbacks achieved their largest quantity, over two million dollars, as Table I shows, in 1875.(11) The panic of 1873 had engulfed the greenback area but not spread to the West Coast. In California, panic struck in 1875, however, when the Bank of California, the state's first commercial bank, found itself facing difficulties. The 1875 panic seems to have had little to do, directly, with any fear that greenback banks would soon challenge the gold banks. San Francisco's favorite son, William C. Ralston, the Bank of California's founder and president, reportedly had speculated on mining stocks and lost [17, 615]. Word of Ralston's losses touched off a run, which quickly spread to the gold banks. The yellowback fell to a discount against gold coin.

The National Gold Bank and Trust Company, the second established of the California gold banks, was hardest hit. It suspended specie payments for two weeks. Before reopening, it sold $500,000 of its bonds and thereby provided funds for retiring part of its note circulation. The [TABULAR DATA FOR TABLE III OMITTED] other gold banks also curtailed their circulations. Another run, in October of the same year, 1875, further encouraged curtailing the note circulation [17, 615-17].

As Table III (column 6) shows, however, even before the panic of 1875 struck, the yellowbacks had never come close to their legal maximum, 90 percent of the gold banks' paid-in capital. Yellowbacks were better than greenbacks, true, but metal was best.

VII. The Gold Banks after Resumption

Resumption took effect on January 1, 1879, and as a result, all national banknotes became gold notes. An act of February 14, 1880, permitted but did not require the national gold banks to become ordinary national banks. By this time, however, two California gold banks - The National Gold Bank & Trust Company (San Francisco) and The First National Gold Bank of Stockton - had declared voluntary liquidation. The Stockton bank had voluntarily liquidated to take an ordinary charter, since no other conversion mechanism existed before 1880.

All seven remaining gold banks eventually convened under the 1880 act. They did not rush to convert, however, even though conversion would have relieved their notes of reserve requirements and also enabled the Comptroller to deliver an additional $10 in banknotes per $100 par value bond deposited as security. Not until 1884 did the last of the national gold banks disappear, and numismatic evidence suggests that yellowbacks were actually issued as late as April 1883, nearly four years after the notes of ordinary national banks had become convertible into gold [26, 718]. Six of the national gold banks that converted to ordinary national charter - First National of San Francisco, First National of San Jose, First National of Stockton, First National of Oakland, First National of Santa Barbara, and D. O. Mills & Co. (Sacramento) - would survive without further change to become members of the Federal Reserve System, along with all other national banks, in 1913.

VIII. Conclusion

A model that stresses relative returns cannot handle the greenback-yellowback experiment: perhaps the declining return on gold helps explain the gold banks' failure in the East, but the argument runs aground in California. Nor does a model that stresses the cost of issuing notes fare better: perhaps the yellowback's relatively stiff reserve requirements help explain the gold banks' failure in the East, but this argument also runs aground in California. And finally, a model that stresses legal-tender laws can help explain the greenback's success in the East, but not its failure in California.

Exploring the relationship between social pressures and an individual's decision making offers a more promising approach. The Schelling tipping model illuminates the social-choice dimension of monetary experiments. Decisions about institutions as pervasive as the medium of exchange do not occur on an incremental basis, a pinch of this and a dash of that. Institutions come, instead, pretty much on an all-or-nothing and one-or-the-other basis, the scales tipping completely in favor of one base money, for example, or completely in favor of another. Once the scales have tipped, furthermore, considerations of personal profit and loss - as from speculating that one base money will appreciate in terms of another or from issuing banknotes denominated in one base money rather than another - will not likely cause a reversal.

Appendix: The Gold Banks' Descendants (as of 1990)

Even today, two of California's national banks, Wells Fargo and The Bank of America, can trace their ancestry back to the old gold banks. The Wells Fargo family tree has three gold-bank branches: Stockton, First Oakland, and San Francisco. In 1920, The First National Gold Bank of Stockton was absorbed by the newly established American Trust Company. American Trust, in 1926, absorbed The American Bank of San Francisco, two years after it had absorbed the rechartered First National Gold Bank of Oakland. Later, in 1960, American Trust merged with Wells Fargo Bank, which later the same year would convert to national charter as Wells Fargo Bank. The First National Gold Bank of San Francisco, after having changed its name to the First National Bank of San Francisco and later having become the Crocker National Bank, merged with Wells Fargo Bank in 1986. The Crocker merger made Wells Fargo California's third largest bank.

The other California national bank with a link to the national gold banks is The Bank of America. The First National Gold Bank of Petaluma was succeeded in 1894 by the Wickersham Banking Company. Wickersham, in 1923, became part of The Bank of Italy, which in 1927 took a national charter and later, in 1930, merged with The Bank of America of California. Today, The Bank of America is California's largest bank.

Three California state banks also have links to the national gold banks. The First National Bank of San Jose, originally The Farmers' National Gold Bank of San Jose, in 1979 became the possession of the Banque Nationale De Paris. In terms of total assets, the San Jose bank, now The Bank of the West, a state bank, ranks as the state's eleventh largest bank.

The First Western Bank & Trust Company, a state bank that in 1954 had absorbed The First National Trust & Savings Bank of Santa Barbara, known originally as The First National Gold Bank of Santa Barbara, in 1961 merged with California Bank, producing United California Bank. To comply with an anti-trust ruling, United California relinquished 61 of its branches, and they formed a new bank, also called The First Western Bank & Trust. The new First Western became Lloyds Bank California, which in 1968 merged with Sanwa Bank California. Meanwhile, United California had changed its name to First Interstate Bank of California, the state's fourth largest bank. Both of these state banks, then, Sanwa and First Interstate, can claim descendance from The First National Gold Bank of Santa Barbara.

The authors' thanks go to Michael Bordo, Benjamin Klebaner, Roger Koppl, Eugene White, and Leland Yeager for comments on earlier drafts and to the Office of the Comptroller of the Currency and the California State Banking Department for otherwise unavailable information.

1. Friedman [14] argues that Sherman, together with other officials, wanted to ensure resumption on the basis of gold, not silver - that the demonetization of silver, the "crime of 1873," was not inadvertent. Perhaps "unlocking a portion of the gold" (and thus depressing its price) was a measure taken even earlier to give gold an edge in the resumption sweepstakes.

2. Hostility toward the legal-tender notes ran deep. Newspapers sometimes served notice (Cross [11] gives some examples) that a certain named person had, at its face value, forced the depreciated currency upon his creditor, a practice called "greenbacking." California's Specific Contract Act of 1863, which the California Supreme Court and, later, the U.S. Supreme Court upheld, permitted gold-clause contracts [25, 5].

3. We doubt that the assurances Sherman refers to concerning the gold banks came from California's senators. Referring to a proposed amendment, California's Senator Casserly remarked, "I understand the effect of the amendment, if adopted, to be to strike out all the gold note sections of the bill. Is that so?" "That is it," several senators replied, and Casserly proceeded to vote for the amendment [10, 970]. California's other senator did not vote on the amendment, which the Senate rejected. On enacting the bill (S. 378: 41st Cong., Sess. II) itself, however, both men said nay.

4. Wisconsin's Senator Howe, who proposed striking the gold-bank provisions of the bill (see note 3), wanted "to relieve the Government of the United States from indorsing the contracts of those gentleman over in Boston who are dealing in gold" [10, 970].

5. We found the bank examiner's report, as well as the letters cited below, in the National Archives [9].

6. The excess was considerable. The Kidder Bank had capital of $300,000, 10 percent of which is $30,000, and the railroad owed the bank $90,005.07 [9].

7. See any 1870 issue of the Boston Advertiser [2].

8. The Kidder, Peabody Papers [20], available at Baker Library of the Harvard Business School, include neither the Comptroller's responses nor reference to them.

9. Since the bank had deposited bonds of $150,000 par value, not the promised $300,000, the Comptroller had issued but $120,000 of the notes printed [13, 138].

10. Oregon had a greenback bank, in Portland, although both the Comptroller and Lester [23] describe Oregon as a state that remained on gold.

11. Circulation by bank, October 1, 1875: First National Gold Bank, Oakland ($80,000); Union National Gold Bank, Oakland ($31.570); First National Gold Bank, Petaluma ($156,740); National Gold Bank of D. O. Mills & Co., Sacramento ($220,000); First National Gold Bank, San Francisco ($605,575); National Gold Bank and Trust Company, San Francisco ($399,582); Farmers' National Gold Bank, San Jose ($239,940); First National Gold Bank, Santa Barbara ($79,940): and First National Gold Bank, Stockton ($358,710) [8, 756-59].

References

1. Armstrong, Leroy and Denny, J. O. Financial California: An Historical Review of the Origins and Progress of Banking in the State. San Francisco: Coast Banker Publishing Company, 1916. Reprint New York: Arno Publishing Co., 1980.

2. Boston Advertiser, various issues.

3. Cagan, Phillip. "The First Fifty Years of the National Banking System," in Banking and Monetary Studies, edited by Deane Carson. Homewood, Illinois: Richard D. Irwin, Inc., 1963, pp. 15-42.

4. -----. Determinants and Effects of Changes in the Quantity of Money, 1875-1960. Princeton, N.J.: Princeton University Press for the National Bureau of Economic Research, 1965.

5. Cagan, Phillip and Anna J. Schwartz, "The National Bank Note Puzzle Reinterpreted." Journal of Money, Credit, and Banking, August 1991, 297-303.

6. Carosso, Vincent P. Investment Banking in America. Cambridge, Mass.: Harvard University Press, 1970.

7. -----. More Than a Century of Investment Banking: The Kidder, Peabody & Co. Story. New York: McGraw Hill, 1979.

8. Comptroller of the Currency, Annual Report.

9. -----, Record Group 101, Box #24, National Archives.

10. Congressional Globe, 41st Congress, 2nd Session.

11. Cross, Ira B., "Californians and Hard Money." California Folklore Quarterly, January 1946, 270-77.

12. Dewey, Davis Rich. Financial History of the United States. New York: Longmans, Green and Co., 1931.

13. Dillistin, William H., "National Gold Banks and Bank Notes." The Numismatist, March 1950, 133-39.

14. Friedman, Milton, "The Crime of 1873 Revisited." Journal of Political Economy, December 1990, 1159-94.

15. ----- and Anna J. Schwartz. A Monetary History of the United State. Princeton, N.J.: Princeton University Press for the National Bureau of Economic Research, 1963.

16. -----, "Does Government Have Any Role in Money?" Journal of Monetary Economics, October 1986, 37-62.

17. Hancock, G. D., "The National Gold Banks." The Quarterly Journal of Economics, August 1908, 602-25.

18. Hayek, Friedrich A. The Denationalization of Money, 2nd ed. London: Institute of Economic Affairs, 1978.

19. Homer, Sydney. A History of Interest Rates, 2nd ed. New Brunswick, N.J.: Rutgers University Press, 1977.

20. Kidder, Peabody & Co. Papers. Special Collections, Baker Library, Harvard University.

21. Klein, Benjamin, "The Competitive Supply of Money." Journal of Money Credit and Banking, November 1974, 423-53.

22. Laitin, David D., "Language Games." Comparative Politics, April 1988, 289-302.

23. Lester, Richard A., "Retention of the Gold Standard in California and Oregon During the Greenback Inflation," in Richard A. Lester, Monetary Experiments: Early American and Scandinavian. Princeton, N.J.: Princeton University Press, 1939.

24. Mitchell, Wesley C. Gold, Prices, and Wages under the Greenback Standard. Berkeley, California: University of California Press, 1908.

25. Moses, Bernard, "Legal Tender Notes in California." Quarterly Journal of Economics, October 1893, 1-25.

26. Philport, W. R., Jr., "National Gold Bank Notes." The Numismatist, November 1934, 717-19.

27. Schelling, Thomas C. Micromotives and Macrobehavior. New York: W. W. Norton & Company, 1978.

28. Timberlake, Richard H., Jr. The Origins of Central Banking in the United States. Cambridge, Mass.: Harvard University Press, 1978.

29. White, Lawrence H. Competition and Currency: Essays on Free Banking and Money. New York: New York University Press, 1989, pp. 48-69.