This is a revised excerpt from White (2015), and the first item in our “What You Should Know” series offering essential background information on various alternative money themes.
Historical monetary systems that are properly classified as free banking systems, in Kevin Dowd’s (1992, p. 2) words, have involved “at least a certain amount of bank freedom, multiple note issuers, and the absence of any government-sponsored ‘lender of last resort.” There were 60-plus episodes around the world of plural private currency issue in the 19th century (Schuler 1992). Dowd (1992) has compiled studies of 9 of these episodes, and Ignacio Briones and Hugh Rockoff (2005) have surveyed economists’ assessments of 6 relatively well-studied episodes: Scotland, the United States, Canada, Sweden, Switzerland, and Chile. Because none of the six systems they review enjoyed complete freedom from legal restrictions, they suggest that “lightly regulated banking” is a more accurate label than “free banking.” All these nineteenth-century episodes had another feature worth mentioning: banknotes and deposits were denominated in and redeemable for silver or gold coins.
When we look into these episodes, we find a record of innovation, improvement, and success at serving money-users. As in other goods and services, competition provided the public with improved products at better prices. The least regulated systems were not only the most competitive but also by and large the least crisis-prone.
Case Studies
Scotland. The Scottish free-banking system of 1716 to 1845 combined remarkable stability with competitive performance. To quote my own earlier work on it (White 1995, p. 32), there were “many competing banks, most of them were well capitalized,” while in its heyday after 1810 “none were disproportionately large, all but a few were extensively branched,” and “all offered a narrow spread between deposit and discount rates of interest.” Briones and Rockoff (2005, pp. 295–96) find “considerable agreement that lightly regulated banking was a success in Scotland.” They note that some writers have given at least partial credit to “unlimited liability, or the presence of large privileged banks acting as quasi-central banks.” After 1810, however, the three chartered banks (the only banks with limited liability) were no larger than the nonchartered banks (which had unlimited liability) and did not play any special supervisory roles, while the system continued to perform successfully. Scottish banking exhibited economies of scale but not natural monopoly. The banks mutually accepted one another’s notes at par. A few writers have expressed doubt that Scotland was a good example of free banking on the grounds that the Bank of England backstopped the system, but such claims are mistaken (White 1995, ch. 3).
United States. Banking restrictions differed dramatically among states in the antebellum United States. The least restricted, most openly competitive, and best-behaved system was in the New England states, where the Suffolk Bank of Boston, succeeded by the Bank for Mutual Redemption, operated a banknote clearinghouse that kept most notes at par throughout the region. Many other states, led by New York, enacted what were called “free banking” laws. These acts opened up entry to all qualifying comers (in contrast to chartering systems that required a special act of the state legislature), but also imposed collateral restrictions on note issue (banks had to buy and hold state government bonds or other approved assets to provide a redemption fund for their notes) and maintained geographical branching restrictions. Briones and Rockoff (2005, p. 302) reiterate a point that Rockoff emphasized in his own pioneering work on the state free-banking systems, namely that these legal restrictions were fairly heavy. The less successful experiences in some states “appear to have been the result of restrictions imposed on the American free banks—restrictions on branch banking and the peculiar bond security system—rather than the result of freedom of entry.” On the positive side, freer entry enhanced competition, and the “stories about wildcat banking” that some historians took to be the natural consequence “although not baseless, were exaggerated.” In New York and some other early-adopting states, the system “worked well,” which explains why it spread to more and more states.
Canada. The Canadian system, Briones and Rockoff (2005, p. 304) note, “like the Scottish system and parts of the American system, was clearly a successful case of lightly regulated banking.” Canada did not suffer the financial panics that the United States did in the late 19th century. Its banks did not even fail in the Great Depression. The Canadian banking system “did so well that a central bank was not established until 1935,” and even then the reason was not dissatisfaction with the existing banking system but some combination of nationalism and wishful thinking about what a central bank could do to end the Great Depression (see Bordo and Redish 1987).
Sweden. Sweden had a system of competitive private note issue by “Enskilda” banks while at the same time having the official Riksbank as banker to the state. The Enskilda banks’ record for safety was remarkable. Briones and Rockoff (2005, pp. 306-7) report that, “Although one could debate the relative contributions of the Riksbank and the Enskilda banks, it is clear that the combination of the two maintained convertibility and provided an efficient means of payment for the Swedish economy.”
Switzerland. Switzerland’s system ended in a crisis, but Briones and Rockoff (2005, p. 310) doubt that this reflects poorly on lightly regulated banking because, “at least after the federal banking law of 1881, the Swiss experience seems to have been less free than other experiences in many important dimensions such as the existence of privileged cantonal banks and restrictive collateral requirements for private banks.” Moreover, the law diminished “the capacity of the public for differentiating notes,” which created a common-pool problem, weakening the effectiveness of the clearing system against overissue. (For a harsher assessment of Swiss free banking, see Neldner (1998); for a rebuttal to Neldner see Fink (2014).)
Chile. Briones and Rockoff (2005, p. 314) also consider Chile’s experience a poor test because the system was skewed by government favoritism, “With a small ruling elite and concentrated economic power, Chile had great difficulty creating note-issuing banks that were completely independent of the government.” Nonetheless a free-banking law was “successful in developing the financial and banking industry.” A new volume of studies on Chile’s free-banking experience is under way by economic historians at the Universidad del Desarrollo in Santiago (Couyoumdjian forthcoming).
Australia. Operating with few restrictions, Australian banks were large, widely branched, and competitive, and they practiced mutual par acceptance, making the system resemble Scotland’s. The Australian episode is of special interest for suffering the worst financial crisis known under a free-banking system. After a decade-long real estate boom came to an end in 1891, some building societies and land banks failed, after which 13 of 26 trading banks suspended payments in early 1893. George Selgin (1992a) finds that the banks’ reserve ratios do not indicate any overexpansion of bank liabilities during the boom, though some banks clearly made bad loans. The boom was rather financed by British capital inflows, which suddenly stopped after the Baring crisis of 1890. Kevin Dowd (1992) adds that the banks were not undercapitalized. He argues that “misguided government intervention” in the first failed institutions “needlessly undermined public confidence” in other banks, while other interventions boosted the number of suspensions (all but one of the suspended banks soon reopened) by providing favorable reorganization terms for banks in suspension. (For a different view, see Turner and Hickson (2002).
Colombia. The free-banking era in Colombia lasted only 15 years, from 1871 to 1886, during the period of a classical liberal constitution. Thirty-nine banks were created, two of which did about half the business. The system survived a civil war in 1875 with only a few months’ suspension and appears to have been otherwise free of trouble. It ended when the government created its own bank and gave it a monopoly of note issue for seigniorage purposes (Meisel 1992).
Foochow, China. George Selgin (1992b) reports that the banking system in the city of Foochow (or Fuzhao) in southeastern China operated under complete laissez faire in the 19th and early 20th centuries, being left alone by the national ruling dynasty. The successful results resembled those of free banking in Scotland or Sweden. Banknotes were widely used and circulated at par, bank failures were rare, and the system provided efficient intermediation of loanable funds.
Postrevolutionary France. The end of the French Revolution, the economist Jean-Gustave Courcelle-Seneuil later wrote, “left France under the regime of freedom for banks.” New banks began issuing redeemable banknotes in 1796. In Courcelle-Seneuil’s evaluation, the banks operated “freely, smoothly and to the high satisfaction of the public.” After only seven years, in 1803, Napoleon Bonaparte took power and created the Bank of France with a monopoly of note issue to help finance his government (Nataf 1992).
Ireland. In 1824, after poor results with plural note issues by undersized banks, the British Parliament deliberately switched Ireland from the English set of banking restrictions (the limitation of banks to six or fewer partners) to the Scottish free-banking model (joint-stock banks with an unlimited number shareholders, each with unlimited liability) and thereafter enjoyed results like Scotland’s. Howard Bodenhorn (1992) considers it “not surprising” that “free banking in Ireland should rival the success of the Scottish. After 1824, restrictions on banking were repealed, except unlimited liability, and joint-stock banks were formed based on the Scottish mould. Failures were infrequent, losses were minimal … and the country was allowed to develop a system of nationally branched banks.”
Why then did central banking triumph over free banking?
As Kevin Dowd (1992, pp. 3-6) fairly summarizes the record of these historical free banking systems, “most if not all can be considered as reasonably successful, sometimes quite remarkably so.” In particular, he notes that they “were not prone to inflation,” did not show signs of natural monopoly, and boosted economic growth by delivering efficiency in payment practices and in intermediation between savers and borrowers. Those systems of plural note issue that were panic prone, like those of pre-1913 United States and pre-1832 England, were not so because of competition but because of legal restrictions that significantly weakened banks.
Where free banking was given a reasonable trial, for example in Scotland and Canada, it functioned well for the typical user of money and banking services. Why then did national governments adopt central banking? Free banking often ended because the imposition of heavy legal restrictions or creation of a privileged central bank offered revenue advantages to politically influential interests. The legislature or the Treasury can tap a central bank for cheap credit, or (under a fiat standard) simply have the cental bank pay the government’s bills by issuing new money. Economic historian Charles Kindleberger has referred to a “strong revealed preference in history for a sole issuer.” As George Selgin and I have noted (Selgin and White 1999), the preference that history reveals is that of the fiscal authorities, not of money users. In some places (e.g., London) free banking never received a trial for the same reason. Central banks primarily arose, directly or indirectly, from legislation that created privileges to promote the fiscal interests of the state or the rent-seeking interests of privileged bankers, not from market forces.
References
Bordo, Michael D., and Angela Redish. 1987. “Why did the Bank of Canada Emerge in 1935?,” Journal of Economic History 47 (June 1987), 405-417.
Briones, Ignacio, and Hugh Rockoff. 2005. “Do Economists Reach a Conclusion on Free-Banking Episodes?,” Econ Journal Watch 2 (August), 279-324.
Couyoumdjian, Juan Pable. Forthcoming. Editor, Instituciones Económicas en Chile: La banca libre durante el siglo XIX.
Dowd, Kevin. 1992a. Editor, The Experience of Free Banking. London: Routledge.
Dowd, Kevin. 1992b. “Introduction” to Dowd 1992a.
Dowd, Kevin. 1992c. “Free Banking in Australia,” in Dowd 1992a.
Fink, Alexander. 2014. “Free Banking as an Evolving System: The Case of Switzerland Reconsidered,” Review of Austrian Economics 27 (March), 57-69.
Hickson, Charles R., and Turner, John D. 2002. “Free banking Gone Awry: The Australian Banking Crisis of 1893.” Financial History Review 9 (October), 147–67.
Meisel, Adolfo. 1992. “Free Banking in Colombia,” in Dowd 1992a.
Nataf, Phillipe. 1992. “Free Banking in France (1796–1803),” in Dowd 1992a.
Neldner, Manfred. 1998. “Lessons from the Free Banking Era in Switzerland: The Law of Adverse Clearings and the Role of Non-issuing Credit Banks,” European Review of Economic History 2 (Dec.), 289–308.
Selgin, George. 1992b. “Bank Lending ‘Manias’ in Theory and History,” Journal of Financial Services Research 6 (Aug.), 169-86.
Schuler, Kurt. 1992. “The World History of Free Banking,” in Dowd 1992a.
Selgin, George A., and Lawrence H. White. 1999. “A Fiscal Theory of Government’s Role in Money,” Economic Inquiry 37 (Jan.), 154–65.
White, Lawrence H. 1995. Free Banking in Britain, 2nd ed. London: Institute of Economic Affairs. Available online at
White, Lawrence H. 2015. “Free Banking in Theory and History,” in Lawrence H. White, Viktor Vanberg, and Ekkehard Köhler, eds., Renewing the Search for a Monetary Constitution. Washington, DC: Cato Institute.