Dr. Wells teaches in the Department of Economies and Dr. Scruggs in the Department of Finance at Memphis State University, Memphis, Tennessee. This article is based on a paper presented to the Southwestern Economics Association in San Antonio, Texas, on March 20, 1986.
Moving toward a fully deregulated financial system.
Most economists consider money to be a special good that should be controlled by the national government. But advocates of free banking consider money a private good which, as any other good, must meet the market test of acceptability. Let us consider the advantages of free banking, and see how such a system might be implemented in the United States.
Under free banking, no government agency would grant a charter to allow a bank to operate. Banks would be free to begin operations as they saw fit, just as other businesses. They could acquire funds from any source: equity, deposits, banknotes, or other liabilities such as debentures; and pay whatever market conditions dictated on these sources. They could then use these funds in the manner most profitable to the bank as long as they avoided deception or fraud. They would hold whatever reserve they wanted—cash, securities or claims on a clearinghouse—wherever they wanted to hold it. Loans and securities would not be subjected to interest controls, nor would investment in any particular industry be mandated or forbidden. If so desired, banks could even acquire equity positions in other firms. The only role for government would be to prosecute fraud and enforce contracts, including settling disputes in court. Banks would not be subjected to government audits or made to pay any special taxes not levied on other businesses.
Money did not originate through some governmental body authorizing its use, but rather by the public and the market deciding what was mutually acceptable. Free banks could continue to offer to exchange their recognized and highly tradeable liabilities for the less well-known and less marketable liabilities of others. Traditionally, banks have persuaded the public to use banknotes and checkable deposits as money. The right to issue these notes and deposits was never a government prerogative, but the common right of all (Breckenridge 1895, p. 196). In the quasi-free-banking systems of the past, banknotes and deposits were normally convertible; that is, redeemable on demand in some reserve asset that the bank could not create, such as gold or silver. Central banks entered the picture and eventually monopolized the issuance of currency, leaving deposit creation to commercial banks.
Since currency became an outside money which banks could no longer issue, they could not exchange one liability for another—deposits for banknotes—as the public desired, but instead were forced to use up reserves or borrow from the central bank to placate public demand for currency. The ability to issue their own notes permitted banks to ease public fears about bank runs and to reduce funds tied up in till money, since unissued banknotes were not a liability on the bank’s balance sheet, but were instantly available. This was especially helpful to Canadian and Scottish banks in minimizing the cost of operating branch offices (Breckenridge, p. 387; Beckhart 1929, p. 377; White 1984, p. 40).
Banks in an unregulated system would be free to open and close branches wherever they wanted. Branching allows banks to diversify their loans geographically and industrially, since banks would be in a position to acquire loan customers all over the country. An increased need for banking services in a particular area of a nation is met more readily by a new branch of an existing bank than by a new unit bank, which requires new directors and managers. Furthermore, within a given financial center, reserves can be moved among branches of one institution more quickly and cheaply than among institutions.
Branching would not be required, of course, but would face the test of the market. Experience from both Scotland and Canada shows that unit banks found it difficult to compete with larger branch banks and were frequently absorbed by them (Breckenridge, p. 146; White 1984, p. 36).
Many specialized thrift institutions, such as savings banks, savings and loan associations, credit unions, and trust companies, originated because commercial banks were legally barred from, or voluntarily eschewed, a particular field of lending. Free banking implies no restrictions on bank lending or investing; hence specialized thrifts would be necessary only for loan categories systematically shunned by banks.
Limit on Credit Expansion
Unlike the current arrangement, whereby the central bank controls the ultimate expansion of the money supply through its control of the money base, bank loan expansion under free banking would be controlled by the ex change of notes and deposits through clearinghouses. Banks that expanded their lending more rapidly than others would face adverse clearings and would be forced to curtail their lending unless they possessed more cash reserves than their rivals. These reserve assets could be deposits in the clearinghouse, gold, silver or some other “outside” money, such as Federal Reserve Notes (FRNs), which would be frozen in supply as were Greenbacks after the Civil War. If the Federal Reserve (Fed) were abolished, existing FRNs could serve as base money in a free banking system.
An individual bank could increase its reserves by a lending policy more restrictive than that of its rivals. The expanding bank would find its notes and deposits presented for redemption in reserve money; the more restrictive bank would gain reserves, as a smaller volume of its liabilities would be presented, by other banks, for payment through the clearinghouse.
However, the reserve base in an entirely free banking system could expand only if new reserve money entered from the outside. Since no FRNs would ever again be printed, the only avenue open would be an influx of gold or silver from other areas or from mining. But no central bank could expand the money base at will, imparting monetary disturbances to the economy.
Since no bank would be legally required to maintain any specific reserves, each bank would hold only the amount of the reserve assets indicated by its experience and liquidity preferences. Thus, if all banks in the system were to expand their lending by the same percentage, the check on them would not be adverse clearings, since each would be receiving from every other roughly the same amount of notes and deposits. Instead the expansion would be checked by each bank’s desired holding of reserves plus the public’s desired ratio of base money to notes and deposits. If the public trusted banks it would be unlikely to convert notes and deposits to specie or FRNs, but such a conversion would act as a brake on bank expansion.
White (1984, p. 44n) found that Scottish banks, during the late 1700s and early 1800s, were able to reduce their average gold reserves from 10 per cent and higher to 3.2 per cent as these banks gained more public confidence, in any case, credit expansion is limited in a free banking system, since the public will hold only a finite amount of any issuer’s distinctive notes and deposits. But under our current system, there is no limit to expansion by the central bank, because the supply of its monopoly money will create its own demand, as the public has no choice but to hold whatever amount is created to support a higher level of nominal income and prices (Yeager, p. 42).
Transition to a Free Banking System
The first step to achieve a free banking system is to remove all legal obstacles to the production of “outside” base money plus all restrictions on private banking. The legal impediments to the production of outside money, according to White (1984a, pp. 297-98), include: (1) a prohibition on the minting of private coins; (2) a sales tax on the purchase of commodity monies; (3) a capital-gains tax on the holding of non-dollar currencies; and (4) uncertainty regarding the upholding by courts of the payment of a contract in anything but dollars, even when gold is specified. Removal of these barriers would signify that the field is open to any type of innovation that might seem profitable to undertake.
Simultaneously, Congress would have to abolish the Fed, freeing the existing supply of FRNs (the Bureau of Engraving can replace worn bills). The Fed would dispose of its assets, after buying back its stock from the member banks. Termination of the Fed could proceed as follows: (1) cease open-market operations and discounting; (2) require the Fed to buy back its stock from member banks by crediting their reserve accounts; (3) send all government securities and gold certificates to the Treasury for cancellation; (4) move the Treasury account to the commercial banking system; (5) let foreign central banks move their accounts wherever they wish; (6) phase out the Fed’s check-clear-ing system, perhaps over one year, as deregulated commercial banks establish branches nationwide and assume the clearing function.
Private Banking. Equally important, however, would be total deregulation of private banking, concurrent with dismantling of the Fed. The immediate steps should be: (1) allow free entry with no charter needed from any governmental agency; (2) grant freedom to branch anywhere; (3) abolish reserve requirements; (4) remove restrictions on the type of assets held; (5) abolish limits on interest rates paid or charged; and (6) allow banks to issue distinctive banknotes or even fractional token coins.
The freedom to enter without charter should reduce the forced difference between banks and thrift institutions. The unlimited branching will further erode this distinction as mergers occur between banks and thrifts. In addition, unrestricted branching will expedite the replacement of Fed check-clearing with private clearing. Unlimited branching will also lead to the end of correspondent relationships among banks and the holding of interbank deposits, a feature peculiar to the unit banking system. When allowed to branch, as in Canada and Scotland, banks hold insignificant amounts of the liabilities of other banks, and are insulated from contagious bank runs.
The abolition of reserve requirements would permit banks to escape the implicit tax with which they have been burdened since the beginning of the National Bank System during the Civil War; they could then earn profits on all of their assets, rather than about 90 per cent of them. A fixed percentage required reserve is the least liquid asset a bank has. When banks are permitted to hold any amount of reserve they want, their anticipated liquidity needs are the sole determinant of the amount and location of those reserves.
As an example, before the Bank of Canada was founded in 1935, Canadian banks faced no reserve requirements but held as desired reserves: outside money, call loans in New York and London, securities and commercial paper, and deposits in foreign banks (Beckhart, p. 430). Cash reserves were normally 10 per cent of total liabilities, but fluctuated between 8 per cent and 15 per cent. Banks watched the balance sheets of their rivals: the stronger ones kept the weaker in line by refusing to take checks drawn on them or refusing to lend to them in a crisis (Breckenridge, p. 433; Beckhart, p. 485). Most of the outside money (specie and Dominion Notes) was kept at the financial centers; the branches used mostly unissued banknotes as their till money. Since unissued banknotes are costless, this economy measure helped subsidize some of the branches in remote areas.
A similar situation could evolve in the U.S. under free banking. The cash reserves of banks would probably consist of FRNs at first, and then specie if the public displayed a preference for convertible banknotes and deposits. But secondary reserves would undoubtedly be invested in short term securities (such as Treasury bills, commercial paper, and bankers’ acceptances) and call loans. No longer could banks consider themselves liquid merely because they could borrow outside money from a lender of last resort. Each would be responsible for its own liquidity, and any borrowing would have to meet the standards of the market.
It is also very important that banks again be allowed to issue banknotes if unregulated banking is to succeed. This enables banks to exchange one liability for another—deposits for notes—at the demand of the public, without disturbing the bank’s reserves, or any other asset. In addition, these banknotes should be distinctive, not uniform as were the National Banknotes, and issued without pledging any specific asset, such as government bonds. Distinctive banknotes would meet the daily test of convertibility: Each bank would be anxious to issue its own notes; upon receiving notes issued by others, it would return them to their issuers through a clearinghouse. To pay out the notes of another bank would be acting as a broker without fee; to hold them would be lending to the issuer without interest (Breckenridge, p. 407). Therefore, note exchanges act as a check on expansion just as check clearings do.
Allowing an unrestricted issue of banknotes, with no requirement to pledge certain assets such as government bonds, permits banks to supply the exact amount of currency that the public demands at the instant they demand it. Panics and bank runs can be avoided if the public is allowed to exchange one type of money—deposits for bank-notes—when it wants. The panics of 1873, 1893, and 1907 in the U.S. resulted when the public could not convert their deposits to currency (Canadian depositors experienced no such difficulty). Many national banks in the U.S. found it unprofitable to hold the 2 per cent government bonds that were required to back their notes, thus were unable to issue enough notes to satisfy the panicky demand for currency. When banks tried to satisfy this demand by paying out their cash reserve (gold certificates, greenbacks, and other Treasury currency), they depleted their required reserves, and were forced to suspend payment.
Deposit Insurance. Another important step in achieving free banking is terminating the Federal deposit insurance that has existed since 1934. Deposit insurance was part of a system of strict regulation designed to save the unit bank system during the credit implosion of the early 1930s. It is rarely found in nations with a small number of large branch banks, even though Canada adopted a plan in 1967. But it is totally incompatible with free banking and thus would have to be phased out.
There are private options to government deposit insurance that should be explored. One such option could lie with the individual depositor. If uneasy about banking, this person could approach an insurance company about insuring his deposit. The insurer could diversify its risk by covering accounts up to a specified maximum in any particular bank, and then asking future clients to deposit in other banks where the insurer is covering smaller deposit volumes.
Another option lies with the banks themselves. They could alleviate public concern over safety by offering low-interest accounts with preferred claims on bank assets in case of liquidation. Another alternative, similar to the existing tax-and-loan account, would be to offer low-interest accounts backed by pledged securities. Other depositors would receive market rates but would have no special protection.
Banknotes vs. Deposits. Holding private banknotes rather than government fiat money would be a new experience for the current generation. Many may be reluctant to accept “funny money” at first, especially since holders of banknotes, unlike depositors, may not be customers of the issuing bank, but merely recipients of its notes in the course of business.
But banks could overcome the aversion to private banknotes in several ways. Banks could assume the risk for counterfeit notes. Scottish banks honored forgeries presented over the counter by innocent persons, but not those accepted by other banks and returned through clearings. This put the burden on rival banks to be on the alert for counterfeit notes. At any rate, counterfeiting is much more tempting in a fiat-money system than in a free-banking system, in which the notes return quickly to the issuer (White 1984, p. 40). Banks could also offer to pay interest on any note not immediately redeemed in base money on demand.
Free banks could voluntarily offer multiple liability for stockholders if such proved reassuring to depositors and noteholders. Scottish banks operated with unlimited liability for their owners; Canadian banks and National banks in the U.S. formerly imposed double liability on their stockholders. The market might compel smaller, newer banks, but not large, well-established institutions, to impose multiple liability.
Banks could offer a mutual note guarantee exemplified by the Bank Circulation Redemption Fund (BCRF) that protected Canadian banknotes from 1891 until they were replaced by Bank of Canada currency. Banks contributed to a fund to redeem the notes of any failed bank, which notes were to earn interest until redeemed (Beckhart, p. 302). While this measure was successful in that the notes of even weak banks circulated at par all over the country, it put the depositor in a position inferior to that of the noteholder. In addition, this was not a voluntary arrangement but was mandated by the Bank Act of 1891. But in Scotland, when the Ayr Bank failed, other banks voluntarily advertised that they would accept Ayr notes, thereby bolstering confidence and gaining customers (White 1984, p. 32). A similar offer under free banking could include depositors as well as noteholders.
A Temporary Lender of Last Resort?
When a free banking system is established, reliance on a governmental lender of last resort should be avoided, even on a temporary basis. With the Fed abolished, it might be tempting to permit banks to borrow fiat money from the Treasury for a few years during the phasing out of the FDIC. Canadian banks were allowed to borrow Dominion Notes from the Minister of Finance in 1914 as a wartime measure, but this practice was not terminated when the war ended in 1918; it continued until the Bank of Canada began in 1935. Canada did not return to the gold standard until 1926, but this borrowing of base money from the government was incompatible with the gold standard, and the latter was abandoned in 1929 (Curtis 1932, p. 322).
A government lender of last resort is also incompatible with a free banking system, wherein each bank is to be responsible for its own liquidity. Free banks can borrow from one another at market rates, but a lender of last resort is needed only when one bank holds all of the outside money, as did the Bank of England during the 1800s (White 1984, p. 145). Under free banking, no bank would incur liabilities in some money that it could not issue.
Coinage. In addition to banknotes, private banks would also be responsible for new coins: full-bodied coins for reserves, and fractional, token coins for making change. The market gradually developed coins to expedite the assessment of the value of varying amounts of commodity money. Governments intervened to monopolize coinage to extract a profit for themselves. Coinage could have remained private, with inferior coins circulating at a discount (Selgin & White 1985, pp. 4-5). Private mints, faced with existing and potential competition, would have much less incentive to debase coins than would a state with a monopoly on coinage.
Establishment of a free banking system requires that the Treasury cease minting its token coins, just as the Fed must stop issuing currency. However, the Treasury could be allowed one last profit by selling all its gold at Ft. Knox at market prices, ending any conceivable government role in money creation. This seems easier than the Timberlake (1984, p. 189) proposal to give everyone in the country an equal share. The existing quantity of Treasury coins would be frozen, joining the existing FRNs as part of the base, and still used as a medium of exchange. Banks, however, would issue new fractional coins, as well as banknotes. Professor Hayek has hypothesized that local merchants might collaborate to sell through banks a set of uniform coins or tokens to be used in vending machines, perhaps replacing metallic coins with plastic ones bearing electronic markings discernible to cash registers and slot machines (Hayek 1978, p. 48n). if the many U.S. railroads could agree on a standard gauge for tracks, banks and retailers can agree on a substitute for Treasury coins.
Possible Concerns over Free Banking
To those who have experienced only a government monopoly of base money, and central bank manipulation of that base, free banking may seem too radical, even though the inflationary record of central banking must inspire a search for alternatives. If money creation were divorced from the government, politicians would have to pay for their spending with taxes or real borrowing, but not by government monetization of deficits.
Critics may fear free banking would be deflationary, as the money base could grow only through additions to the gold or silver stock, since the FRN quantity would be frozen. But this ignores the possibilities of financial innovations to economize on base money, higher money turnover rates, and increasing confidence in the ability of banks to honor redemptions by the public. One Scottish bank in the early 1800s was able to operate with specie reserves of only 0.5 per cent of assets, revealing that the public may not demand specie when assured of its availability (White 1984, p. 141). Each competitive issuer can observe the demand for his money and adjust the supply thereto, a task no monopoly issuer can accomplish. Each bank must balance the desires of its depositors, who fear a depreciating money, with those of its borrowers, who would object to an appreciating money. But free banking does promise to end the inflationary bias that has crept into most wage negotiations and other long-term contracting. It is a system that would be consistent with more flexibility of prices and wages.
Other qualms about free banking may afflict bankers themselves who have spent their entire careers in a regulated environment. Older bankers may resist the adjustment; but younger, perhaps better educated ones might find a new, unregulated setting to be a stimulating challenge. Many large retailers and brokerage houses already have in place nationwide offices from which a new type of banking business may be conducted. The field may become dominated by those who were not bankers prior to free banking.
Because the U.S. system currently has no nationwide banks, as other countries do, growth in the number of banks to a competitive level may take longer than it would under unlimited branching. However, the system is not burdened with governmentally encouraged banking cartels that would stifle competition, and most interest ceilings have been abolished. But American banks are still subjected to governmental insuring and auditing, practices incompatible with free banking.
Fortunately, no private U.S. bank is a government pet, like the Bank of Montreal (BOM) or the Bank of England (BOE). No American bank would be the government’s fiscal agent: Over 11,000 banks currently hold tax-and-loan accounts of the Treasury; if the Treasury closed its account at the Fed, no single private institution would play the role of the BOM or BOE. Vis-á-vis the banking system, the U.S. government would be neutral, having no independent Treasury funds to deposit at various times, as was the case before 1914.
Finally, free banking is most likely the system that would have emerged from normal market forces, had governments not interfered with banks and money by imposing specific bond-holding for note issue, mandating fixed percentage reserve requirements, forbidding branching across state lines, and creating a central bank to expedite inflationary governmental finance. 
Beckhart, B.H. The Banking System of Canada. New York: Henry Holt & Co., 1929.
Breckenridge, R.M. The Canadian Banking System 1817-1890. New York: MacMillan, 1895.
Curtis, C.A. “The Canadian Monetary Situation.” Journal of Political Economy, 40 (June 1932): 314-337.
Hayek, F.A. Denationalization of Money—The Argument Refined. 2nd ed. London: Institute of Economic Affairs, 1978.
Mises, L. von. Human Action: A Treatise on Economies. 3rd rev. ed. New Haven: Yale University Press, 1949.
Selgin, G.A., and White, L.H. “The Evolution of a Free Banking System.” Paper presented at the 55th conference on the Southern Economic Association, Dallas, Texas, 24 November 1985.
Timberlake, R.H. “Federal Reserve Policy Since 1945.” In Money in Crisis, pp. 177-193. Edited by B.N. Siegel. Cambridge, MA: Ballin-ger, 1984.
Wells, D.R., and Scruggs, L.S. “The Free Banking Alternative.” The Freeman 35(February 1985): 98-103.
White, L.H. Free Banking in Britain: Theory. Experience, and Debate, 1800-1845_ New York: Cambridge University Press, 1984.
_______. “Free Banking as an Alternative Monetary System.” In Money in Crisis. pp. 269-302. Edited by B.N. Siegel. Cambridge. MA: Ballinger, 1984a.
Yeager, LB. “The Medium of Exchange.” [n Monetary Theory. pp. 37.60. Edited by R.W. Clower. Baltimore: Penguin Books, 1970.