Richard H. Timberlake is professor of economics, University of Georgia. This article originally appeared in the Journal of Money, Credit, and Banking, Vol. 19, No. 4 (November 1987) and is reprinted here with permission. Copyright © 1987 by the Ohio State University Press.
“In the company town, or mining camp, . . . United States coin and currency were not in good supply . . . . During the heyday of the old company town, scrip circulated more freely than U.S. currency and was in deed the coin of the realm . . . . Eleanor Roosevelt . . . in the mid-thirties, during [one of] her humanitarian crusades, attacked the use of scrip by coal mining companies as a very evil thing . . . .
Although many mourn the days of a bustling and active coal economy, little can be said to support the . . . issuance of scrip.” (Truman L. Sayre, “Southern West Virginia Coal Company Scrip,” in Trade Token Topics, reprinted in Scrip, Brown, 1978, pp. 343-344)
1. The Possibility of Free Market Money
Ever since the abolition of the operational gold standard in the early 1930s, the federal government through its agent, the Federal Reserve System, has been almost the sole creator of the monetary base, and has also been the licensing agent for the banks that create most of the demand deposits used in the United States. No money of any significant mount can be created today without some sanction or act of the Federal Reserve System.
This condition has encouraged the notion that government is a necessary, or at least desirable, regulator of any monetary system—that without government involvement any monetary system quickly degenerates into “chaos.” If this supposition were valid, the evolution of money could hardly have occurred. The barter system that preceded early monetary systems, in which government had no part, would not have been superseded if the resulting monetary systems were destined to be chaotic. This logic suggests the possibility and perhaps the feasibility of a non-government money. However, the practical efficacy of such a system cannot be deduced from a theory that merely suggests its possibility, but must be sought from historical evidence of monetary arrangements that have developed spontaneously in the private sector.
This paper examines one such incidence of private money creation—the issue and use of scrip, which occurred primarily in the isolated economic environments of mining and lumbering company towns during the first half of the twentieth century. Fortunately, numismatic collections and records reflect the operational character of the scrip systems in these communities so that some evaluation of their monetary properties is possible.
Much of the recent research on the creation of private money has focused on that issued by private banks in the presence of a dominant legal money such as gold. (White 1984, Sylla 1976, Rolnick and Weber 1982) The issue of scrip, however, had nothing to do with banks. It was issued by private mining and lumbering enterprises. While it, too, was redeemable in a dominant money, its issue and acceptance were not critically dependent on any dominant money. For this reason, the phenomenon of scrip issue is especially revealing.
2. Legal Restraints Against the Issue of Private Money
Proscriptions against the arbitrary or casual issue of money appeared at the very beginning of this country’s political formation. First, the Constitution stated: “No state shall . . . emit bills of credit, [or] make anything but gold and silver coin a tender in payment of debt.” (U.S. Constitution, Art. 1, Sect. 10) No money except gold and silver was to be the legal tender issue of any governmental unit.
Money to be money, however, does not have to be legal tender. It can be what one might call common tender, i.e., commonly accepted in payment of debt without coercion through legal means. Indeed, privately issued money to exist at all would have had to be common tender, and would have had to earn its acceptability in a market environment.
Even though the states and Congress were constrained to monetizing only gold and silver, the general laws of contract and commercial instruments sanctioned the appearance of moneys issued by privately owned commercial banks. (Hurst 1973) In addition, “Nothing in the Constitution barred private manufacture of coin, and through the first half of the nineteenth century Congress did not act against private coinage . . . . General contract law allowed any contractor to issue his notes and coins and circulate them so far as the market would take them.” (Hurst 1973)
Free enterprise in the issue of common tender money was accidentally encouraged in practice by the federal government’s ineptness in establishing a useful denominational spectrum of fractional currency during much of the nineteenth century. (Carothers 1967) Private transportation companies—canals, turnpike companies, and rail- roads—issued significant amounts of such currency between 1820 and 1875. Municipal and state governments did likewise. Redemption of transportation currency when called for was in services rendered, while state and local government currency was redeemed as tax payments. (Timberlake 1981)
The paucity of government-issued fractional currency was catastrophically aggravated by the first issues of greenbacks during the Civil War. The metallic values of subsidiary coins rose rapidly above their monetary values in the summer of 1862, and the coins disappeared from circulation. These circumstances provoked not only the ill-conceived issue of postage stamp currency, but also extensive private issues of minor coin. (Carothers 1967, Faulkner 1901) The act that authorized postage stamps as currency in 1862 also outlawed the private issue of notes, memoranda, tokens, or other obligations “for a less sum than one dollar intended to circulate as money or to be received or used in lieu of lawful money of the United States.” (Act of Congress, 12 Statutes at-Large, 592, July 17, 1862) Then in 1864, even the private issue of gold and silver coin was forbid den, again, “when the coins were intended for use as current money.” (Hurst 1973)
3. The Appearance of Scrip as an Economizing Medium
The lack of adequate denominations in gov-ernment-produced money was not the only factor that stimulated the private production of money. Shortly after fractional coinage was stabilized around 1885, coal mining and lumbering became major industries. Both coal mining and lumber-ing enterprises had to be organized in the vicinity of the contributory resources, so were often located in isolated areas with low population densities significantly distant from commercial centers. Coal-producing regions were hilly or mountainous areas where agriculture had been marginal and other commercial development had lagged. “The ‘Main Street,’” noted one observer in describing a coal mining community “was often railroad tracks.” (Brown 1978) Coal mining en trepreneurs, therefore, had unique problems to contend with in organizing their enterprises.
Their common problem was what is known today as a lack of infrastructure—no streets, no churches, no schools, no residences, no utilities, and no banks or financial intermediaries. The specialized industries that might otherwise have provided these services were dissuaded from doing so by the high start-up costs and the enduring uncertainties of dealing with low-income communities that might be there today and gone tomorrow. Alternatively, the coal mining companies could deal with such conditions because they were in a better strategic position to change un-calculable uncertainties into calculable risks. (Fishback 1986, Johnson 1952) Mining companies, therefore, built residences, churches, schools, and water works, and opened company stores or commissaries. In so doing, they became both buyers of labor from, and sellers of corn- modifies to, the coal miners and their households. This kind of organization invited an economy in the community’s payments system—the use of scrip in lieu of ordinary money.
“Scrip” has become a generic term for the issue of a localized medium of exchange that is redeemable for goods or services sold by the issuer. Originally printed cards or “scraps” of paper, scrip evolved into metallic tokens with many of the physical attributes of official coins. Indeed, scrip in the very beginning was more in the nature of a trade credit, or demand deposit, at the single local general store. Ledger credit scrip, however, gave way to scrip coupon books, which “eliminated the tedious bookkeeping chores that were incident to over-the- counter credit (day book or journal entries followed by ledger entries).” (Brown 1978)
The use of scrip not only implied an issuer—the mining company—and a demander—the miner, it also required a supplying industry. The institutions that supplied coupon scrip were companies already in business printing tickets, tokens, and metal tags for various other kinds of enterprise. They advertised extensively in mining catalogues during the first half of the twentieth century touting the advantages of their own scrip systems. The Allison Company of Indianapolis, for example, noted that when one of its coupon books was issued to an employee, “He signs for it on the form provided on the first leaf of the book, which the storekeeper tears out and retains for the [company] time-keeper, who deducts the amount from the man’s next time-check.” Then, when the employee buys goods from the company store, “he pays in coupons, just as he would pay in cash, and the coupons are kept and counted the same as cash . . . . The coupon book is a medium of exchange between the company employees and the company store.” (from 1916 Mining Catalog, Brown 1978) Other scrip-producing ticket companies emphasized the safety of the scrip coupon system in coal mining communities “where little or no police protection is afforded.” (adv. of the International Ticket Co., in the Keystone Catalog of 1925, Brown 1978)
The Arcus Ticket Company of Chicago advertised a list of advantages of scrip to both the employer and employee, one of which for the employer was the fostering of employee good-will by avoiding misunderstandings on charge accounts. The advantages to the employee included keeping the “‘head of the house’ better informed as to the purchases made by his family from day to day . . . . This frequently puts a check to extravagance and debt.” (Keystone Catalog, 1925 in Brown 1978) Local scrip of this type was very similar to modern day travelers checks. The costs of travelers checks were also the costs of coupon scrip: each unit could be used only once. It had to be signed out when it was issued and signed when it was spent. (Brown 1978)
The transactions costs of coupon scrip eventually encouraged the increased use of metal scrip. This medium became cheaper overall than coupon scrip, in spite of metal’s higher initial cost, largely due to the invention and development of • the cash register after 1880. Pantographic machines also were instrumental in reducing the unit costs of metal tokens. (Brown 1978)
Instead of receiving cash, the scrip-issuing “cash registers” paid out metal tokens, made a record of the pay- out and to whom it had gone, and kept a grand total of the amount issued. The scrip registers would eject a specified “dollar” amount of scrip when a lever like that on a slot machine was pulled. In a 1927 advertisement, the Osborne Register Company (ORCO) of Cincinnati pictured a 10-year-old child who, in a demonstration, issued $600 worth of metal scrip in various amounts to 200 hypothetical employees in 55 minutes, implying an average emission of $3 per employee every 16.5 seconds. (Brown 1978)
4. The Positive-Sum Benefits of Scrip
The economics of scrip issue, as with all exchange between economic agents, required that both the issuer (the coal mining company) and the acceptor (the employee) benefit from the transaction. The company necessarily had contact with the outside world. It bought machinery and other resources and sold coal in a national market. All these activities required the use of standard money.
Scrip was used essentially as a working balance of money with which the coal operator could make advances to his impecunious employees between paydays. It was issued at the request of the miner to the extent of the wages he had already earned, and it was redeemable in standard money on the next payday. The amounts were usually small—five or ten dollars, or even less. To the worker it amounted to an interest-free, small-sum loan that he could get with almost no effort. It enabled him to buy ordinary household goods at the company store. To those workers who had “gone out and got drunk” on the previous weekend, or who had suffered some kind of household emergency, scrip was a blessing only measurable by the cost of its common alternative. (Clark 1980, Johnson 1952)
Its alternative in a conventional urban setting without scrip was the pawn shop, loan shark, or installment peddler. (Johnson 1952) An industrial worker in the same unfortunate position in, say, Detroit, Pittsburgh, or Chicago, had access to money between paydays only by borrowing against his household capital at a pawn shop where he paid exorbitant interest rates if he reclaimed his pawned goods.
The scrip system could be abused in such a way that a discount would also appear in some scrip transactions. Since the company store did not sell liquor—for the obvious reason that its sale would encourage absenteeism and worker inefficiency—workers would at times obtain scrip from the company clerk and sell it for conventional currency in order to buy liquor. The bootlegger (during Prohibition) or other liquor vendor, whose shop was not likely in the neighborhood of the company store, faced significant costs in redeeming the scrip for conventional money, thus giving rise to a discount. (Brown 1978, Caldwell 1969)
In spite of the obvious advantages of the scrip system to both worker and mine owner, scrip, the company store, and the company town have been universally bemeaned. (Brown 1978) The accounts of their operations include contradictions that appear sometimes in the same paragraph. (For example, see quote of Sayre used as an epigraph, p. 1, Brown 1978.) All accounts, while critical of the scrip system, acknowledge, first, that it was issued at the behest of the miner; second, that its issue cost the miner nothing; and, third, that it was redeemable in standard money on payday. The dogma of scrip’s critics was that the company store, in which the scrip had to be spent, raised prices to monopolistic levels and thereby exploited the defenseless miner. (Dodrill 1971) Fishback’s and Johnson’s studies of prices in company stores versus those in independent stores refute this popular prejudice. Prices were four to seven percent higher, but so were costs. (Fishback 1986; Johnson 1952)
The advantage of scrip issue to the mine operator was that it was one worker perquisite he could offer to attract labor into a somewhat unattractive environment. He already offered housing and mercantile services; by issuing scrip against future wages he also provided commercial credit with virtually no interest charges to the borrowers. (Johnson 1952) The practice, indeed, was so widespread that it can only be viewed as a traditional perquisite of the trade. A company that did not offer the scrip privilege would have been at a competitive disadvantage.
The mine operator thus became a quasi banker. His cost for metal scrip during the 1920s varied from slightly less than 1 cent to 5 cents a unit for scrip tokens of simple design made in aluminum. In brass or nickel silver and with scalloped edges and more intricate designs, costs could run as high as 11 cents a piece. (All these values are unit costs in thousand-unit lots, and are from advertisements of several different scrip manufacturers between 1925 and 1940, in Brown 1978.)
Scrip sales information from the Ingle Company sales journal of 1928 reveals that the average denomination issued was about $ .25. (Brown 1978) Since the average cost per token was only about 3 cents and could have been even less, an investment by the coal company bank in, say, 5,000 pieces cost it about $150 for the scrip coin, and perhaps $100 more for a scrip-issuing machine. To carry out this same banking function with regular U.S. Currency would have required an investment in cash alone of $1,250, as well as substantially greater security costs to protect the money. One observer noted, “The mining company could pay almost its entire payroll in company scrip, disturbing only a few dollars of actual working capital.” (Sayre, in Brown 1978) Of course, paying out scrip gave workers some additional claims on the working capital of the company stores. So the monetary economy of using scrip was in part offset by higher costs of merchandising goods.
The difference between the payment system costs of scrip and of real money was a form of seigniorage revenue the coal mine operator realized and shared with his employees. They received interest-free loans; he was able to offer a fringe benefit that tended to reduce what would have been a higher working capital requirement.
While scrip was usually specialized to one company in a particular community, many coal mining companies had mines in different regions. Their scrip was good in all the different locations where their mines operated. As the scrip-using communities gradually came to experience more extensive commercial relations with each other, their localized scrips became interchangeable. Even some independent stores accepted coal company scrip. (Brown 1978)
Given the proscriptions against the private printing or coining of money by the Acts of 1862 and 1864, one may wonder how scrip could have been issued and used legally. The key is the word “intended” in the proscriptive laws. The courts ruled that scrip was not intended to circulate as money: first, bemuse it was redeemable only in merchandise until payday; and, second, because it resembled money only superficially and was clearly distinguishable from standard money. (The coin under the court’s scrutiny was a 50-cent token, but weighed only one-fifth as much as a standard 50- cent piece.) Any token that was redeemable in lawful money on demand was construed to be illegal, and whether the token in question was coin or pasteboard did not matter. (Brown 1978)
5. The Environments in Which Scrip Appeared
The extent of scrip use has many dimensions—temporal, geographical, and industrial. Its most notable occurrence in the twentieth century was in the coal mining regions of West Virginia, in part because the state government passed a “wide open” scrip law some time before 1925. However, it was extensively used in other states as well. The Tennessee Coal Iron and Railway Company, for example, ordered 547,500 pieces between 1933 and 1937 from the Ingle-Schierloh Company of Dayton, Ohio. (Brown 1978) Another source lists 20,000 coal company stores in the United States, Canada, and Mexico all of which used scrip between 1903 and 1958. (Dodrill 1971)
Numismatic records indicate that scrip was also used extensively in several other industries—fishing canneries, agriculture (to pay crop-pickers), fruit canneries, logging and lumbering companies, and paper companies. (Brown 1978, Trantow 1978. Trantow’s index lists over 1,100 companies that issued scrip currency in 40 states.) One scrip numismatist cites a Chicago newspaper of 1845 that regularly quoted the discounted prices of coal scrip, city scrip, canal scrip, railroad scrip, Michigan scrip, Indiana State scrip, and Indiana land scrip, as well as the notes of private and chartered banks. Private businesses issuing such scrip numbered in the thousands. (Harper 1948) Furthermore, as Brown observed, “The use of paper scrip was much wider than the use of [coin] scrip . . . [but] only a comparatively small amount [of the paper] has survived.” Therefore, the extent of scrip use must have been much greater than the vestiges in metallic collections would indicate. (See also Caldwell 1969.)
Just as Brown in his work seemed unaware of scrip that had preceded the issues by coal companies, Harper in his study of Scrip and Other Forms of Local Money thought that intensive use of scrip only appeared in the United States during the depression years, 1932-1935. His research uncovered several sources of “depression” scrip: (1) issues by local governments due to decreases in tax revenues; (2) issues by chambers of commerce after local bank failures as a means of “corralling as large a proportion of the depression diminished volume of business as possible for their membership”; (3) issues by “home-owned stores as a weapon against . . . chain- store competition”; (4) issues by “barter groups as a means by which the unemployed could more conveniently exchange services”; and (5) issues by charitable organizations to needy persons as “commodity orders” for foodstuffs. “Local money in some form,” he concluded, “is likely to recur in response to a public demand under substantially similar circumstances.”
Most of this “depression” scrip had appeared in earlier times—for example, municipal scrip that was redeemable as tax payments. The depression scrip, however, was usually linked tO a dated stamp scheme that required the holder to fix low denomination (2- or 3- cent) stamps to the scrip at specified times. The stamps were to provide the revenue to redeem the scrip and to encourage spending, but they added an undesirable burden that greatly reduced the efficacy of the scrip’s use. They also detracted from the scrip’s effectiveness as an addition to the existing stock of ordinary money. (Harper 1948)
6. Implications of the Scrip Episode
The phenomenology of scrip issue has significant implications. First, no one had any incentive to leave scrip behind for monetary researchers to count or to analyze. Demanders of such currency would not regard it as a store of value for any time longer than the period between paydays. Suppliers, to whom the scrip was an outstanding demand obligation, would redeem it first if they liquidated, merged, or closed clown their enter prises. In addition, everyone who used it and benefited from it was aware of its questionable legality. Archival records of its outstanding quantities, therefore, are almost nonexistent. (Timber-lake 1981)
Scrip’s unrecorded existence is emphasized as well by the research that has uncovered its former use. Each scholar who has unearthed one of the diverse scrip appearances has treated the phenomenon as unique, and with good reason. Each one was widely separated in time, place, and circumstance from the others. Yet, each one had characteristics similar to the others. All episodes combined emphasize the feasibility of the spontaneous production of money in the private sector.
The coal mining scrip episode adds significantly to the total scrip experience for a number of reasons. First, it lasted for over 50 years, so it was not just a temporary happenstance. Second, it appeared in a wide range of independent communities. In West Virginia alone, almost 900 coal mining companies employing about 120,000 miners issued scrip in one form or another. In other areas of Appalachia—southern Virginia, eastern Kentucky, eastern Tennessee and southwestern Pennsylvania—the experience was similar.
Third, scrip’s tenure was not dependent on the previous existence of standard legal tender money. True, the coal company was bound to redeem the scrip on payday, but this guarantee was only a flourish that enabled scrip issuers to avoid violating the proscriptive laws against the issue of private moneys. As it was, many children living in coal mining communities did not see a dollar of “real” money until they grew up and left the area. (Caldwell 1969)
The self-sustaining nature of the scrip system, without recourse to standard money, stemmed from the fact that both the demander and supplier of scrip were active participants in both the labor market and the household goods market at the company store. This intimacy in two markets by both participants enabled them to evaluate wages paid and received in real terms, that is, by the quantity of household goods that the scrip wages could purchase. A decline in the purchasing power of scrip at the company store would simply have indicated to the miner that the real value of his services to the company had de-dined. He thereupon would have moved to an other location or occupation. If the decline in real wages was due to an industrial depression or the competitive decline of the coal industry, as occurred simultaneously in the 1930s, both mine workers and mine operators would realize reduced real returns in the mode of any resource owners under similar circumstances.
A fourth important result of the scrip system was its reflective emphasis on the returns to the capital structure of the payments system. In the scrip system the money was supplied endogenously: the coal company banks, the borrowing miners, and the scrip suppliers were all parts of an economy of private ownership. Scrip money was not dependent on any outside money, but was produced under the same conditions and incentives as any common commodity. The mining companies rather than the workers produced the scrip because in working without wages until payday, the workers were implicitly extending credit to the company. Scrip issue was a means of clearing this debt before the regular payday. In addition, the coal mining company had the collateral value of the mined coal to secure the “loan.”
Both the companies and the workers realized the seigniorage returns from its existence. While the scrip system was small-scale and had a low profile, the government could ignore it because it posed no threat to the government’s monopoly over the production of money. However, if scrip issue had shown any tendency to become a national practice, the proscriptive laws against private coinage would surely have been interpreted and enforced much more rigorously.
An observer of the scrip system might conjecture that the experience of the isolated communities could have ramified into an intercommunity system using some kind of scrip clearinghouses (i.e., scrip banks) if the laws restraining the private issue of money had not existed. Over time, technological and organizational developments Could have led to economies of scale and enterprise. Probably as few as three or four or as many as two dozen issuers of scrip money might have appeared. Some of the minters of scrip—Ingle-Schierloh, Osborne, Insurance Credit, Adams, Dorman, and others—would have expanded their enterprises to include management of inter-community scrip systems and ultimately their probable evolution into credit card systems. Such an extension of function would have been analogous to automobile dealers expanding into the car leasing business—a sort of horizontal integration to reap certain economies of scale.
Had the scrip system become intercommunal and given rise to scrip-on-deposit in scrip banks necessitating bank reserves and clearing operations, some high-powered scrip into which local scrips could be converted would probably have appeared. The experience of the ages seems to confirm this evolution. (Friedman and Schwartz 1986) Less clear is why the high-powered money has to be issued or regulated by the state. The question of whether or not the market system could, alternatively, produce a private monetary base that would prove to be both stable and serviceable has not been attempted or allowed, and will remain unimaginable until a general belief in market efficacy becomes pervasive. That time as yet seems nowhere near.
1. This comparison must be qualified. Many travelers checks, as well as other U.S. currency, are currently used as hand-to-hand media in foreign markets. Sometimes travelers checks return from abroad with more than a dozen endorsements on them. They are called “checks,” but like food “stamps,” they are a quasi currency.
2. Scrip was frequently advertised as redeemable only to the worker to whom it was originally issued. This condition applied in mines. However, for metallic scrip, it could hardly have been enforce, and would have detracted from the utility of any scrip if it were enforced.
5. In a thought-provoking paper, David Glasner argues convincingly that governmental assumption of a monopoly role over money enabled governments to enhance their fiscal powers, particularly during war emergencies (Glasner, “Economic Evolution and Monetary Reform,” especially the section: “A Rational for Government Monopoly over Money”). In short, not only is seigniorage an important revenue 40 the state, but capital expropriation through debasement of money’s function as a unit of account may be even mote lucrative.
6. However, the commercial bank clearinghouse system in the United States during the second half of the nineteenth century is an example of a private lender of last resort that produced base money efficiently at critical times. (Timberlake 1984)
Brown, Stuart E., Jr. Scrip. Berryville, VA.: Virginia Book Company, 1978.
Caldwell, Walter. Coal Company Scrip. Montgomery, W. Va.: War Printing Co., 1969. Carothers, Neil. Fractional Money (1930), reprint. New York: Kelley, 1967.
Clark, C. R. Florida Trade Tokens. St. Petersburg, Fl.: Great Outdoors Publishing Company, 1980.
Dodrill, Gordon. 20,000 Coal Company Stores in the United States, Mexico and Canada. Pittsburg: Duquesne Lithographing Company, 1971.
Faulkner, Roland F. “The Private Issue of Token Coins.” Political Science Quarterly 16 (1901), 320-22.
Fishback, Price. “Did Miners Owe Their Souls to the Company Store? Theory and Evidence from the Early 1900′s.” Journal of Economic History 46 (December 1986), 1011-29.
Friedman, Milton, and Anna J. Schwartz. “Has Government Any Role in Money?” Journal of Monetary Economics (January 1986), 37-62.
Glasner, David. “Economic Evolution and Monetary Reform,” (unpublished).
Harper, Joel W. Scrip and Other Forms of Local Money. Ph.D. dissertation, University of Chicago, 1948.
Hurst, James Willard. A Legal History of Money in the United States, 1774-1970. Lincoln: University of Nebraska Press, 1973.
Johnson, Ole S. The Industrial Store, Its History, Operation and Economic Significance. Atlanta: Foote and Davies, 1952.
Rolnick, Arthur J., and Warren E. Weber, “Free Banking, Wildcat Banking, and Shinplasters.” Federal Reserve Bank of Minneapolis Quarterly Review 6 (Fall 1982), 10-19.
Sylla, Richard. “Forgotten Men of Money: Private Bankers in Early U.S. History.” Journal of Economic History 36 (March 1976), 173-88.
Timberlake, Richard H. “The Significance of Unaccounted Currencies.” Journal of Economic History 41 (December 1981), 853-66.
_____________. “The Central Banking Role of Clearinghouse Associations.” Journal of Money, Credit, and Banking 16 (February 1984), 1-15.
Trantow, Terry N. Catalogue of Lumber Company Store Tokens. Ellensburg, Wash.: Trantow, 1978.
White, Lawrence H. Free Banking in Britain. Cambridge: Cambridge University Press, 1984.
The author is indebted for support and suggestions to the sponsors and participants of the Manhattan Institute Monetary Conference of 1986, especially to David Glasner and Anna Schwartz. My colleague, Price Fishback, and Milton Friedman also made valuable suggestions, as did Hus-ton McCulloch and two referees for the Journal of Money, Credit, and Banking.