Notes

• [1]
  Triangle, ENS de Lyon. Email : samuel.demeulemeester@ens-lyon.fr. The author would like to thank the anonymous referees of this article, as well as the participants of the HET workshop between Nice (GREDEG) and Lyon (Triangle) of 6 July 2021, for their helpful comments and suggestions.
• [2]
  According to Aliber and Kindleberger ([2015], pp. 18, 20): “Asset bubbles—most asset bubbles—are a monetary phenomenon … One theme of this book is that the cycle of manias and panics results from the pro-cyclical changes in the supply of credit, which increases rapidly in good times, and then when economic growth slackens, the rate of growth of credit declines sharply” (p. 20).
• [3]
  See also Kotlikoff [2010], Chiarella et al [2012], Krainer ([2013]; [2017]), Goodhart and Jensen [2015], or Fontana and Sawyer [2016]. In the francophone literature, see for example Gomez [2010], Giraud [2012], Grjebine [2015], Quignon [2016], Le Maux [2020], or Gomez and Munier [2020].
• [4]
  The idea of a “100% money” system in which central bank money itself would be directly used by all economic agents had already been proposed by Tolley ([1962], pp. 299-300). Dyson et al. [2016] suggest that this could take the form of a central bank digital currency (CBDC).
• [5]
  See for example Diamond and Dybvig ([1986], p. 65), Benes and Kumhof ([2012] 2013, pp. 4, 79), Wolf ([2014b], p. 210), Turner ([2015], p. 187), or King ([2016], pp. 261-264).
• [6]
  See for example Diatkine ([2002], p. 151).
• [7]
  See for example Friedman ([1992], p.xi), Phillips ([1995], p. 182), Kotlikoff ([2010], p. 132), Chiarella et al ([2012], p. 411n3), or Constâncio [2016].
• [8]
  For an overview of the debates on the “100% money” proposal in the 1930s, the reader can refer, for example, to Barber [1973], Dimand [1993], Phillips ([1995], chap. 11), or Demeulemeester ([2019], chap. 1).
• [9]
  The 1844 Act also imposed an automatic issuing rule—the “currency principle”—on the Bank of England’s Issue Department, allowing it to issue notes only by increasing its metallic reserves by a strictly equal amount.
• [10]
  This group included Garfield V. Cox, Aaron Director, Paul H. Douglas, Albert G. Hart, Frank H. Knight, Lloyd W. Mints, Henry Schultz and Henry C. Simons. The thrust of their plan was similar to that proposed in England by Soddy [1926], whose work had been reviewed by Knight [1927]. This has led some commentators to argue that the Chicago Plan had been influenced by Soddy’s ideas—a claim discarded by Tavlas [2020].
• [11]
  Carter Glass (then a Congressman) had, together with Senator Robert Owen, initiated the bill that became the Federal Reserve Act of 1913, giving birth to the Fed. Even then, he had rejected a clause, introduced by Owen (on a recommendation by Fisher), which required the Fed to promote price level stability (see Dimand [2020]). Under the influence of Glass, the Federal Reserve Act was instead inspired by the real bills doctrine, according to which the issue of banknotes should be based solely on the rediscounting of commercial paper. Adherence to this doctrine, with its pro-cyclical effects, seems to have played a significant role in the passive attitude shown by the Fed during the monetary contraction of 1929-1933 (see Humphrey and Timberlake [2019]). During the 1935 debates on the revision of the Federal Reserve Act, Glass (now a senator) was “violently opposed” to legislative changes that altered the system he had built (Sandilands [1990], p. 64). He was also very upset that Roosevelt had not consulted him on the appointment of Eccles as governor of the Federal Reserve Board (ibid.), as well as not having had the bill in his hands before it was introduced in Congress—which is why, according to Phillips ([1995], p. 120), he did everything to wreck the bill.
• [12]
  Currie remained an adviser to Eccles at the Federal Reserve from 1934 to 1939, where he proposed a new version of his 100% reserve plan in an August 1938 memorandum (Currie [1938]). He then served as President Roosevelt’s administrative assistant for economic affairs from 1939 to 1945. See Sandilands ([1990] ; [2004]).
• [13]
  The authors of this programme included Paul Douglas, Irving Fisher, Frank D. Graham, Earl Hamilton, Willford King and Charles Whittlesey. John R. Commons joined them in a new version circulated in 1940.
• [14]
  Phillips ([1995], pp. 133-135) discusses the reasons why the 100% reserve plan was not ultimately adopted as part of the New Deal reforms. Amongst these, he mentions a widespread misunderstanding of the implications of the reform, which was often wrongly perceived as involving the end of private banking, or as opening the way for state control of credit. The banking profession itself, however, was not systematically hostile to the plan, as illustrated, for instance, by the review of Fisher’s book by banker Harvey E. Fisk ([1935], p. 569).
• [15]
  See Demeulemeester ([2019], pp. 59-61). One may also mention Murray Rothbard [1962], writing in the Austrian tradition of thought, who proposed a system of 100% gold reserves on current account deposits, which would remove all money-creation power not only from the banks but also from the state. This type of reform continues to be supported, for example, by Jesús Huerta de Soto ([1998] 2011).
• [16]
  Conceived as an alternative to deposit insurance to secure the payment system, the narrow banking proposal would separate lending institutions into several categories: narrow banks, on the one hand, which could only invest in government securities or other assets deemed safe, and which alone would be allowed to provide payment services; and other institutions, on the other hand, which could invest in risky assets, but whose liabilities could not serve as means of payment.
• [17]
  Currie and Fisher called “money” all the instruments that commonly fulfil the function of a medium of exchange, i.e. of final means of payment allowing the settlement of transactions, such as cash and current account deposits. The authors of the Chicago Plan, on the other hand, used a broader meaning of the term, including both what they called “effective money”—the means of payment—and “near monies” fulfilling the role of a liquid store of value, but not that of a medium of exchange; this was the case, for example, of savings deposits or other liquid assets, not transferable in themselves but easily convertible into effective money.
• [18]
  See in particular Fisher ([1935a], pp. 180-181). This cumulative interplay was already at the heart of his credit cycle theory developed in 1911 (Fisher [1911], chap. 4), as well as of his debt-deflation theory developed in the early 1930s (Fisher [1932]; [1933]). It was only in his 1935 book, however, that Fisher called for severing the link between money creation and bank loans. Simons et al ([1933] 1994, pp. 45-48) described essentially similar cumulative processes, except that they attributed a prominent, rather than secondary, role to changes in V.
• [19]
  For a discussion of this seigniorage argument, see Demeulemeester [2020].
• [20]
  In Fisher’s plan, the check banks would keep their reserves as deposits (100% covered) at the Federal Reserve Banks, which would themselves have their own checking accounts (also 100% covered) at the Currency Commission (Fisher [1935a], pp. 28, 61). Only the latter would have the power to create money.
• [21]
  See Demeulemeester [2018] for a more detailed analysis of these two approaches.
• [22]
  The emphasis on this distinction is ours. Currie and Fisher, although they clearly rejected the idea of subjecting savings deposits to 100% reserves, often claimed that their respective reform plans would put an end to the “fractional reserve system”, or result in fully liquid banks. However, this was only true for chequable deposits. Such abuses of language on their part could lead to some confusion. With regard to the reserve requirement ratios to be applied to savings deposits, Fisher ([1935a], p. 13) considered that these should not be affected by the reform, even though he considered their strengthening desirable. Currie ([1934b], p. 199; [1938], p. 361) recommended, on the contrary, that they should be lowered to zero.
• [23]
  As Fisher ([1935a], p. 69) stated: “The loan department … would deposit its own cash in the check department and would transfer it by check just like any other depositor”. The movements related to savings accounts would therefore not give rise to any creation or destruction of money, the sums simply passing between the respective current accounts of the loan banks (or loan departments) and their customers.
• [24]
  Both Currie ([1934a], p. 14) and Fisher ([1935a], pp. 168-169) insisted that non-transferable savings deposits, even if convertible “on demand” into means of payment, were of a different nature from deposits transferable by cheque, themselves serving as means of payment. Only the latter fulfilled the role of money and had to be 100% covered. However, they did not exclude that prudential measures, such as withdrawal notices, could be applied to savings deposits (see Currie [1934b], pp. 199-200; Fisher [1935a], pp. 165-166).
• [25]
  Friedman ([1960], pp. 65-70), on the other hand, took up the bulk of the Chicago Plan, in which loan banks would disappear in favour of investment funds; he refused, however, to follow Simons in his recommendation to ban short-term debt contracts (see Friedman [1967], p. 3). As for Allais, he proposed to maintain the intermediation function of lending banks, but to prohibit them from practising maturity transformation: “all lending for a given term would be financed by borrowing of at least the same term” (Allais [1987], p. 525).
• [26]
  “The 100% system would be no cure-all for business fluctuations though it would help reduce them. … [I]t would afford no guarantee that loan banks and savings banks would be completely immune to runs and failures, nor that any such immunity would be enjoyed by investment houses, building and loan associations, insurance companies, commercial concerns, railways or any other persons or corporations except the checking banks” (Fisher [1935a], p. 216).
• [27]
  On this point, Fisher’s interpretation would later be supported by empirical studies, such as those conducted by Warburton ([1949], p. 91) or Friedman and Schwartz ([1963], p. 682). Friedman ([1967], p. 12) will explicitly refute the opposite interpretation supported by Simons: “The movements in velocity—which Simons took as an independent source of instability—come later than the movements in the quantity of money and are mild when the movements in the quantity of money are mild. They have been sharp only when there have been sharp movements in the quantity of money.”
• [28]
  This was Simons’ position until 1934, when he himself came to advocate a policy of stabilising the price level.
• [29]
  This position could make sense in the case of the Chicago Plan, under which lending banks would disappear (see above). Currie, although he would have kept the banks in their intermediation function, initially considered abolishing rediscounting (Currie [1934b], p. 224). He later proposed that rediscounting could be practised by the Reserve Banks (Currie [1938], p. 361), but out of money creation rather than centralised reserves. He therefore did not envisage, as Fisher did, rediscounting independently of the issuance of money.
• [30]
  Fisher ([1935b], p. 41, our translation) stated: “It is not necessary to go so far as Senator Cutting advocates by having the Treasury buy up the Federal Reserve System, which is now a private affair. On the contrary, I would keep the Federal Reserve System as a bulwark for private banking enterprises, but without giving it control of the purchasing power of the dollar, which is a truly governmental function.” Another alternative, which Fisher ([1935a], p. 89) seemed ready to consider, was to dispense with rediscounting from the Federal Reserve Banks, leaving each bank to find another bank to rediscount with on its own.
• [31]
  Fisher ([1935a], p. 203n1) thus noted that, if the Currency Commission had to rediscount, it could always offset (or, in more modern terms, sterilise) the resulting money creation by selling securities on the open market.
• [32]
  Only Currie, after initially rejecting it, actually considered this option in his 1938 memorandum ([1938], pp. 361, 364). In Fisher’s plan, as we have noted, the Federal Reserve Banks would still be able to rediscount, but without this practice resulting in money creation.
• [33]
  Fisher ([1935a], p. 208n3) stated that these transactions could take place in the usual way, with the Treasury selling securities in the primary market (in exchange for pre-existing money), while the Currency Commission would at the same time buy securities in the open market (creating money).
• [34]
  Let us note here that a “100% money” system could potentially reduce the distortionary effects put forward by Richard Cantillon [1755], according to whom an injection of money would impact the relative price structure depending on the specific markets through which it came into existence. By being issued in the way of loans (from commercial banks as well as the central bank), money necessarily comes into existence where the banks lend— mainly, today, in asset markets. By being issued through tax credits or payments to citizens, on the other hand, it could reach all sectors of the economy at once. The potential advantage of a “100% money” reform in this respect, ignored by the 1930s’ authors, has only recently been highlighted in the literature (see for example Baeriswyl [2017]).
• [35]
  One of the referees of this article raises the question of the significance of the “100% money” proposal in light of the evolution of central banks, which in the 1930s were coming under increased state control (see Blancheton [2016]). As illustrated in this section, the “100% money” authors did not adopt a uniform stance in this respect. They all called for a state monopoly on money issuance, but differed on the choice of institution to carry out this function, on the degree of independence to be granted to it, and on the appropriateness of maintaining a bankers’ bank—not to mention the question of whether the latter, if established, should be private or public, independent or not, or even merged with the issuing authority. Their principled positions also allowed for exceptions—Fisher ([1935a], p. 113) pointing out, for example, that in the event of a great war, Congress might have to authorise the Currency Commission to finance the Treasury without regard to price stability.
• [36]
  Fisher ([1935a], pp. 19, 27, 29) himself excessively equated the “100% money” plan with the 1844 reform, without noting their differences. For a detailed comparison between the “100% money” proposal and the reform ideas of the Currency School, see Demeulemeester [2021].
• [37]
  It is notable, for example, that the argument that a “100% money” system would end the cumulative interactions between the volume of loans, the volume of money and the price level, which is still found in Allais ([1947], pp. 278-279, 360-361), does not appear anywhere in Friedman. Friedman pointed out that the public’s choices about the form in which to hold their money balances, under such a system, would no longer bring undesired changes in the total money volume (Friedman [1960], pp. 66-68). But at no point did he observe that the same would be true for the decisions about the granting and repayment of bank loans. Yet, as we have seen, it was mainly the latter aspect that was central to 1930s’ writers.
• [38]
  This has been the case, for example, of Diamond and Dybvig ([1986], pp. 65-66), Williamson and Wright (2010, p. 28), Turner ([2015], pp. 188-190), or King ([2016], pp. 262-264).
• [39]
  This provision of the Chicago Plan, as we have pointed out, was later taken up by Friedman ([1960], pp. 65-66). It is also found in the “revisited” version of the Chicago Plan proposed more recently by Benes and Kumhof of the IMF ([2012] 2013).