The financial crisis in America is really a moral crisis, caused by the series of proofs which the American public has received that the leading financiers who control banks, trust companies, and industrial corporations are often imprudent, and not seldom dishonest. They have mismanaged trust funds, and used them freely for speculative purposes. Hence the alarm of depositors, and a general collapse of credit (The Economist, November 2, 1907, p. 1856).
I. INTRODUCTION
The 2007–09 economic crisis has led economic historians and historians of ideas alike to look for episodes in the history of the American economy that could shed some light on current events. At first, some parallels were drawn with the Great Depression sparked off by the stock exchange crash of 1929, but attention has gradually moved to another episode, which also started in New York City: the bank panic of October and November 1907 (see, e.g., Krugman Reference Krugman2009, pp. 155, 160). Business contraction had begun in May 1907, but it was only after October that a sharp drop in output and employment took place, until the lower turning point in June 1908 (see Friedman and Schwartz Reference Friedman and Schwartz1963, pp. 156–168). The immediate cause of the crisis was the failure of the third-largest trust company in New York, the Knickerbocker, which was followed by contagious panic, runs on other trusts, and pressure on banks’ minimum reserve requirements, culminating with a concerted refusal by the banking system to convert deposits into currency (Goodhart Reference Goodhart1969, chapter 4; Wicker Reference Wicker2000, chapter 5; Bruner and Carr Reference Bruner and Carr2009, chapters 9–11). Trust companies had grown swiftly since their emergence as a financial innovation in the 1890s (Sprague Reference Sprague1910, p. 227; Conant [1927] Reference Conant1969, p. 707, that is part of chapter 25, introduced from the 4th [1909] edition of Conant’s book). The lower reserve requirement and looser regulation of trust companies, as compared to other commercial banks, led them to holding portfolios of riskier assets, using, among other sources, collaterized loans. Similar to modern-day investment banks, mortgage companies, and hedge funds, the trust companies concentrated their portfolios on large positions in risky assets. Depositors’ perception of increasing risk of insolvency led to intense withdrawals from the trust companies, which ignited the panic of 1907 and brought about the failure of seventeen trust companies and twenty-five banks in less than two months (Moen and Tallman Reference Moen and Tallman1992).
The crisis of 1907 was widely discussed by contemporary American economists (the best-known analysis was provided by Oliver Sprague [1908; Reference Sprague1910, chapter 5]), leading eventually to the creation of a central reserve of lending power in 1913, the Federal Reserve System. Panic conditions and the suspension of cash payments induced the export of a considerable amount of gold from London into New York, which pushed the English bank rate to its highest level since 1873 and helped turn the panic into an international crisis (Conant [1927] Reference Conant1969, chapter 25; Goodhart Reference Goodhart1969). Hence, the American crisis became a matter of concern for European economists as well, particularly Knut Wicksell. In May 1907, Wicksell had delivered in Oslo a lecture on “The Enigma of Business Cycles,” his main piece on the topic of economic fluctuations. He would refer for the first time to the “present world crisis” in a German version of that lecture written later that year and published only recently (Wicksell [1907] Reference Wicksell2001, p. 336). In the following year, Wicksell wrote three articles in the Ekonomisk Tidskrift, motivated by the (monetary) causes and (real) effects of the 1907 crisis. Two of them, about price stabilization and unemployment, respectively, have been translated into English (Wicksell [1908a] Reference Wicksell1997; [1908b] Reference Wicksell and Sandelin1999) and referred only indirectly to the crisis in the United States. The other one, published in Swedish in February 1908, discussed at length the main features of the American banking system that had led to the panic of 1907.
Wicksell’s (Reference Wicksell1908c) “Lesson on Banking Legislation” was a reflection on what could be learned from the American events concerning the efficient regulation of banking. Although the theme of the organization and reform of banking and monetary systems is conspicuous in Wicksell’s writings (especially in connection with his concept of a pure-credit system; see Boianovsky Reference Boianovsky1998; Boianovsky and Erreygers Reference Boianovsky, Erreygers, Fontaine and Leonard2005), the 1908 essay is the main source of his views on the basis of a “rational” banking system (Wicksell Reference Wicksell1908c, p. 42). Wicksell advanced, probably for the first time in the literature, a clear distinction between the “solvency” and “liquidity” of banks, and discussed its implication for the interpretation of crises.Footnote 1 Moreover, he pointed out the role of available reserves of legal means of payment, especially when the demand of agents for liquidity in the form of cash (as opposed to credit instruments) increases during economic crises.
II. SOLVENCY AND LIQUIDITY
The first lesson drawn by Wicksell from the American 1907 crisis was that a good bank should have three attributes: it should be solvent, liquid, and flexible (1908c, p. 42; the original Swedish words are “solvent,” “likvid,” and “kulant,” respectively). A bank “is solvent if it offers full security for its deposits or—in the case of a bank that issues notes—for the conversion of its notes under any circumstances” (ibid), even if it must temporarily suspend its payments, permanently liquidate its business, or go bankrupt. “Whether its share holders are able to get their money back is only relevant for the bank’s prosperity as an enterprise, not for its solvency as a bank.” This is a further elaboration of the definition in Interest and Prices that banks are solvent “in the ordinary sense of the word,” if their “assets exceed their liabilities” (Wicksell [1898] Reference Wicksell1936, pp. 115–116). A bank is liquid if it can “fulfill its obligations on demand or within a day,” and pay, against a rate of discount, obligations that are not due yet (1908c, p. 42). Finally, a bank is flexible if it is “under any circumstances prepared to satisfy all reasonable credit requests, against suitable guarantee and at an adequate—that is, not too high or too low—rate of interest” (ibid).
Wicksell pointed out that although the flexibility requirement may look like a paradox, it is the “core and utmost basis” of both solvency and liquidity, because of its implications for aggregate economic stability, as elaborated further below. Bank legislation had usually aimed at just one of those three characteristics; that is, liquidity.Footnote 2 In this connection, banks were often forbidden to lend money for a long period, and were requested by law to keep as cash balance a certain amount of currency (see also Wicksell [Reference Wicksell1906, 1915, 1929] 1935, p. 80). As observed by Wicksell (Reference Wicksell1908c, p. 42), in that regard the American legislation was stricter than the European practice. The American national banking law of 1863 required a minimum amount of legal tender—inaccurately called “reserves,” according to Wicksell, as they could not be freely used—of 25% to deposits for banks in New York and other central reserve cities (Chicago and St. Louis) banks, whereas the country banks and reserve city banks could hold part of their—respectively 15% and 25%—reserve in deposits in central reserve city banks (see also Sprague Reference Sprague1908, p. 366; Hawtrey Reference Hawtrey1913, pp. 171–172).
Although liquidity may be important for “psychological” reasons, severe requirements “may easily jeopardize solvency itself.” The higher the fraction a bank must keep as cash without interest, the higher the temptation to risk the remaining assets (in order to get enough interest income) and to hold a lower amount of own capital. Moreover, liquidity ultimately depends on the bank’s solvency, or better, on the depositors’ confidence on the existence of such solvency (1908c, p. 43). When confidence is lacking, the most severe regulations will not prevent a loss of liquidity, especially if distrust becomes general and entails a run on several banks simultaneously. Wicksell (p. 44) referred to a statement often repeated by The Economist in the last months of 1907 about the fact that the monetary system and the economic life of any country cannot depend on gold but on confidence.Footnote 3 “This is true, but it requires, of course, that banks are worthy of the public’s confidence, that is, that they are undoubtedly solvent. There cannot be suspicion in that regard—a bank must be like Cesar’s wife” (ibid; italics in the original).
The conditions for a bank’s solvency are not generally easy to define or control. However, one of these conditions—sufficient own resources in relation to the bank’s business—can be measured through a “barometer” expressed by the level of its profit payments. If those are exceptionally high, it indicates that the bank’s operations are either too large or too risky in relation to its capital. According to Wicksell, that was exactly what happened in the American panic of 1907. Whereas the capital of regular banks in the US represented on average one-quarter or one-fifth of their loans—a number similar to European banks—the situation was different for the trust banks, which featured high leverage. The capital of the Knickerbocker trust bank was $1,200,000, against deposits of $62 million (see also Sprague Reference Sprague1908, p. 360). Moreover, in the months before the panic, the bank paid profits to its shareholders as high as 50% (Wicksell Reference Wicksell1908c, p. 45). Because of their original features as hedge funds (“trustees of shares,” a translation of “förmynderskap på aktier,” the original Swedish expression used by Wicksell, ibid), trust banks were not as regulated as the other banks—except in part for the reserve requirement of 15% instituted in 1906, of which only one-third needed to be in the form of currency (see Moen and Tallman Reference Moen and Tallman1992, p. 614). In contrast with the banks that formed the national banking system, trust banks did not have access to the New York Clearing House, except indirectly through a clearing-house member bank.
As noted by Wicksell (Reference Wicksell1908c, p. 45), the relatively small capital of trust banks made them an “easy prey” for speculators, who acquired a substantial amount of the shares of one or several trust banks in order to use the public’s deposits in bold enterprises. Indeed, the 1907 panic began with the failed stock manipulation scheme to corner the market in the United Copper Company in mid-October, in which many banking institutions had become deeply involved. On October 21, one of the directors of the Knickerbocker trust was forced to resign because of his ties to the copper speculators. On the same day, the trust—whose assets were, to a large extent, locked up in enterprises that could not be immediately liquidated—applied for aid to the Clearing House. The National Bank of Commerce, which attended to the clearings of the Knickerbocker trust, refused to clear for it because of a heavy unsettled balance against the company, which was followed on October 22 by a bank run and suspension of operations by the trust bank (Conant [1927] Reference Conant1969, chapter 25; Wicker Reference Wicker2000, chapter 5; Bruner and Carr Reference Bruner and Carr2009, chapters 10 and 11). The run spread to other trust banks (including the Trust Company of North America, the second largest), which caused severe strain upon the clearing-house banks. As reported by Sprague (Reference Sprague1908, pp. 363–364), everywhere in the US “the banks suddenly found themselves paying out money in response to the demand of frightened depositors, and were in turn forced to draw upon their reserves deposited with other banks.... Banks as well as individuals throughout the country were evidently hoarding money, and trying to collect still more.” As a result, specie payments were suspended, and a premium of 3% to 4% was paid for money (Wicksell Reference Wicksell1908c, p. 46). The connection between the solvency problems and the general strive for liquidity was clear enough.
III. FLEXIBILITY, MONETARY POLICY, AND RESERVES
The crisis was made worse by the fact that the sudden demand for legal tender money had reduced most of the banks’ reserves below their statutory limit, and the banks in that position were “according to the law prescription, forced to reduce their supply of credit, just when an increase was needed” (Wicksell Reference Wicksell1908c, p. 46; italics in the original).Footnote 4 This is related to the third characteristic of an efficient banking system listed above: flexibility. Wicksell (ibid) criticized the fact that the flexibility of banks—that is, the “unimpeded but not unconditional granting of credit”—was largely overlooked in the banking and monetary literature, with the partial exception of Bagehot (1873). The topic was part of the broad Wicksellian theme of price-level changes as the link between monetary capital (credit) and real capital (saving). Wicksell (Reference Wicksell1908c, pp. 46–48) restated his theory of the cumulative process and its policy corollary that the banks’ rate of interest should be raised as soon as the price level starts to increase. If the banks’ interest rate is not raised promptly (cf. Wicksell [1898] Reference Wicksell1936, p. 189), the inflationary process will induce agents to become too optimistic about their profit prospects, and form false expectations about the natural rate of interest (1908c, p. 48; see also Boianovsky and Trautwein Reference Boianovsky and Trautwein2001, section 4). When those expectations are proved wrong and the boom comes to an end,
General confidence is replaced by general distrust, loans to buy shares are rejected, underwritten shares must be paid, liquid cash is demanded where before goods were easily bought on credit, etc. Now comes a difficult and dangerous time for banks, which get credit requests from all over. Such credit is increasingly granted in the form of hard currency or notes, which for the time get “stuck” in circulation instead of returning to the banks. The crisis has arrived (Wicksell, Reference Wicksell1908c, p. 49).
The crisis (and the upper-turning point of the price level) is associated with a real-balance effect caused by an increase in the demand for money balances, which “appear much too small in relation to the new level of prices. The demand for goods contracts, their supply increases, and prices fall ...” (Wicksell [1898] Reference Wicksell1936, p. 62; see also Boianovsky Reference Boianovsky1995, p. 398). Under these circumstances, banks may also increase their own preference for liquidity, and try to strengthen their cash position by rejecting credit requests (or imposing conditions so severe that they are impossible to meet), refusing to renew old loans etc. However, this is a “dangerous” policy, since ensuing bankruptcy and reduction in the value of collateral of businesspeople will affect the savers’ confidence in the banks’ solvency. The right policy, according to Wicksell (Reference Wicksell1908c, p. 49) is quite the opposite: “full flexibility, free and unimpeded credit to the last cent, although charging an interest rate adjusted to the circumstances and higher than usual, but without unnecessary difficulties.”Footnote 5 He referred to Bagehot’s Lombard Street as the classic source of those ideas, which he had described elsewhere as “conventional wisdom” (see Boianovsky and Trautwein Reference Boianovsky and Trautwein2001, pp. 347–348, 359).Footnote 6 Of course, the crisis itself could be prevented if banks raised their interest rates at the start of the upward price movement (ibid). Moreover, in order to attract deposits and be able to increase their credit activities, banks should raise not just the interest charged on loans, but also the interest paid on deposits (Wicksell Reference Wicksell1908c, pp. 49–50).
Even if the banks’ lending policy is the right one, there remains the matter of the higher demand for “ready money” during crises. Wicksell ([1907] Reference Wicksell1953, pp. 58–59) had discussed before how the “general lack of confidence, concern over one’s personal security etc” paralyzes the velocity of circulation of money.
But this aspect of the matter is also the one in respect of which economists ... have made most headway. The remedy ... lies in having a sufficient reserve of legal tender—whether of metal or of paper money is a matter of indifference—which is put to use only in times of economic recession and of threatening panic, in order, for the time being, to meet the urgent shortage of credit. This is an economic discovery by means of which, as we may hope, crises will for all time have lost some of their worst impact.
However, against Wicksell’s expectations, a few months after he delivered his Oslo talk, the American panic of October 1907 indicated that the lesson had not been learned just yet. As pointed out by Wicksell (Reference Wicksell1908c, p. 50), banks should keep an “available reserve” of legal means of payment, which was often mixed up with the banks’ cash reserve for redemption of their notes. In Sweden, France, and most other European countries, the available reserve consisted of the private banks’ stock of central banknotes plus the central bank’s “unused right” to issue notes. In the case of the centralized British banking system, it corresponded essentially to the stock of gold and notes of the Banking Department of the Bank of England. The second lesson from the 1907 panic was the absence of any regulations in the American banking system about such available reserves, which were even discouraged by the bank legislation. The strict reserve requirements meant the opposite of available or true reserves, observed Wicksell (Reference Wicksell1908c, p. 51), referring to a remark he attributed to Bagehot: “a reserve is to be used, otherwise it is better that it does not exist.”Footnote 7 Although Wicksell’s view—that the American bank legislation effectively discouraged keeping available reserves—was not supported by contemporary commentators (e.g., Sprague Reference Sprague1908, Reference Sprague1910) and economic historians (e.g., Wicker Reference Wicker2000), there was a consensus that it did pose a legal obstacle in the path of reserve availability in periods of financial emergencies (Wicker, p. 126). Sprague (Reference Sprague1908, p. 366) referred to the fact that the arithmetical reserve ratio in the USA had become a “sort of fetish to which every maxim of sound banking policy is blindly sacrificed.”
After some detailed comments on plans for reform of the American banking system on the wave of the 1907 panic—mostly based on reports from The Economist—Wicksell (Reference Wicksell1908c, p. 54) mentioned an emerging consensus about the centralization of the system according to the European pattern, which would be the only way—with a relatively small available reserve—to supply the market during periods of money scarcity. Indeed, as an outcome of the 1907 crisis, the Federal Reserve System was established in the United States a few years later. As pointed out by Charles Rist ([1938] 1940, p. 421), for the first time the concept of a central bank was linked up, not with the issue of notes, but with the centralization of a country’s gold reserve.Footnote 8
In contrast with Rist, Wicksell did not support the gold standard as the best monetary system. It is well known (see Boianovsky Reference Boianovsky1998) that he believed the financial systems of his day were approaching what he called the “pure-credit” or “ideal banking system,” which should be complemented by a monetary reform replacing gold by banknotes as the standard of value. The steady progress in the direction of an increase in the (virtual) velocity of circulation of money should be distinguished from periodic fluctuations arising out of the “exaggerated use of credit and followed by a reaction known as a credit or money crisis, which arises from a lack of confidence between individuals” (Wicksell ([1906, 1915, 1929] Reference Wicksell1935, p. 67). Such lack of confidence means that the demand for a medium of payment turns more towards hard cash and banknotes in times of crisis (ibid, p. 90). The relevant question was whether reserves of gold should be kept to be used in periods of instability. Wicksell’s (ibid, p. 122) answer was that such reserves would be entirely unnecessary even as a precaution for unforeseen circumstances. He explained:
The distrust of banknotes, even those of central banks, which so often appeared in former days in times of unrest, has now completely vanished.... What the business world is afraid of nowadays in times of crisis is that the banks’ credit facilities will be exhausted and credit thereby become stringent, and not that their instruments of credit will lose their value and their purchasing power. Nowadays we never hear of a “run” on gold by the public, but frequently of a run by businessmen and bill brokers to get their bills discounted at the central Bank, in case the bank reserves or the unused portion of the statutory note issue falls unusually low and the private banks begin to restrict credit in consequence.
The American crisis of 1907 was a partial exception to that, but it did not warrant the view that a pure-credit system was not feasible. In a passage added to the second 1915 edition of Volume II of the Lectures, Wicksell observed, “The famous panic in the U.S.A. in 1907 was clearly connected with the peculiar banking conditions in that country, and no repetition of the occurrence is likely since the American banking system has been reorganized on a more rational basis” (ibid). Of course, Wicksell could not anticipate that a hundred years later, the proliferation of unregulated financial innovations would bring about another wave of intense instability in the American economy.
IV. THE AMERICAN CRISIS AND THE APPROACH TO ECONOMIC FLUCTUATIONS
At the outset of his 1908 essay, Wicksell (Reference Wicksell1908c, p. 41) suggested that the fundamental cause of the American crisis of 1907, like most other crises, was probably something that we “could call, in a more comfortable than precisely clear way, overspeculation or overcapitalisation, not to mention the more popular but even less precise concept of overproduction.” This pointed to capital scarcity as a main factor behind the 1907 upper turning point, but the connection is not straightforward. Indeed, in his 1914 lecture on economic crises, Wicksell would tell his students that at the beginning of the twentieth century, the American economy enjoyed a boom, mainly caused by huge investments in railroads. “The capital used exceeded the country’s own savings, so that a great deal of European capital had to be brought in. In 1907, with the collapse of the copper syndicate and the fall of the Knickerbocker trust, a serious financial crisis set in.... At the same time production declined swiftly.” Recovery was also “very rapid, so that by 1909 a new boom had already set in” (modern accounts confirm Wicksell’s description; see, e.g., Bruner and Carr Reference Bruner and Carr2009, pp. 7–8, 143–144).
As claimed by Wicksell (Reference Wicksell1908c, p. 46) against the underconsumption approach, the 1907 crisis was not preceded by overproduction of goods (see also Sprague Reference Sprague1908, p. 355, for a similar opinion). On the contrary, “industries found themselves with large orders, but did not dare to deliver the goods out of fear that they would not get paid. At the same time, they could not get credit to pay their workers.... Most firms (e.g., the Westinghouse big steel factory) had tied up their own available assets in new plants during the boom.” Hence, the crisis was associated with widespread illiquidity brought about by the collapse of credit in the financial market.
Wicksell’s sense of “overspeculation” may be inferred from his remark—written a year after the 1908 essay—that its origins “are to be found in the monetary sphere, in the wrong choice of monetary and banking policy,” which is part of the cumulative process of price change caused by a difference between the natural and bank rates of interest. The business cycle, on the other hand, results from changes in the natural rate of interest itself as caused by irregular technical progress (Wicksell [1909] Reference Wicksell and Sandelin1999, p. 44, n. 2). As explained in the article about the American crisis, once the cumulative process of price change starts to affect the price expectations of agents, one may observe “the phenomenon which almost always precedes a crisis: the chances of gain generated by the price rise bring about an almost feverish atmosphere of excitement.” New ventures are started up, which, “because of their high construction costs, have in reality smaller prospects of profit than ever before, or which completion may perhaps exceed the existing capital resources of the country” (Wicksell Reference Wicksell1908c, p. 48). It is implicit in this argument that inflationary expectations will make longer (capital-intensive) production methods more profitable, with the continuation of the cumulative process until it is brought to an end by a crisis (see also Boianovsky Reference Boianovsky1995, pp. 394–396, for Wicksell’s analytical distinction between business cycles, caused by oscillations of the natural rate of interest, and economic crises, provoked by miscalculation about the value of the natural rate induced by price-level movements).
Interestingly enough, Dennis Robertson ([1934] 1940, pp. 87–88) would argue that the ‘capital scarcity’ hypothesis provided “an almost completely satisfactory explanation of the sharp short set-back of 1907.” In Robertson’s framework, the upper turning point is caused by a shift leftwards of the curve of supply of loanable funds back to its original position—after the initial income transfer from wages to profits is reversed—and the ensuing rise in the market rate of interest above its “quasi-natural” value. However, as pointed out by Robertson, a crisis may also be brought about by saturation with existing capital goods, which was his sense of “overproduction.” Instead of capital scarcity, the saturation hypothesis was Robertson’s preferred explanation for the 1929 crisis, as opposed to the 1907 one (see Boianovsky and Presley Reference Boianovsky and Presley2009). The assumption of a temporary saturation of investment opportunities, when no positive market rate of interest may equilibrate saving and investment at full employment, played an important role in Wicksell’s ([1907] Reference Wicksell1953) interpretation of the cyclical downturn (see Boianovsky and Trautwein 2006, pp. 178–179, on Wicksell’s notion of a negative natural rate of interest in the depression).
The 1907 crisis also attracted the attention of Irving Fisher, who wrote an anonymous short note about it that same year in the November issue of the Yale Review (Fisher’s authorship was established in Fisher Reference Fisher1961, p. 23). According to Fisher (Reference Fisher1907, p. 229), in order to understand the crisis, one should not be distracted by factors such as the “organization and promotion of trusts with their ‘undigested securities’ and their arbitrary effect on prices, and the particular characteristics of the individuals and firms who have failed,” which were merely “precipitating causes.” The general explanation should be found in the “progressive rise in prices due undoubtedly to the increasing supplies of gold.” Fisher then developed his well-known argument about the imperfect adjustment of the nominal rate of interest to the anticipation of price increases, which causes the real rate of interest to decline and brings about the upswing. The upper turning point comes along when the nominal rate of interest adjusts itself completely to the rate of price increase, and borrowers are unable to repay their loans.
The fundamental factor in the situation is that the business world has not forecast the general rise in prices.... Many who have plunged into speculative ventures find that their expenses, especially for interest, have unexpectedly risen and have begun to withdraw from their ill-fated enterprises.... We are entering upon a commercial crisis, and if our analysis is correct the effect will not be confined to New York City or this country (Fisher Reference Fisher1907, p. 229).
Although there are some common points between Wicksell’s discussion of overspeculation and Fisher’s notion that the real rate of interest comes down during the boom, it is worth noting that their respective theoretical frameworks are distinct. Wicksell’s was not a monetary theory of the business cycle, in contrast with Fisher’s. The peculiar characteristics of the American financial system, which Fisher acknowledged but considered of secondary importance, played a crucial role in Wicksell’s interpretation of how the strive for liquidity pushed the bank rate of interest upwards, and eventually overshot the natural or equilibrium rate. Illiquidity explained the turning of the cumulative process of rising prices—started by a divergence between the natural and markets rate of interest, and accelerated by upward price expectations—into a downward one accompanied by output contraction.Footnote 9
V. FINAL REMARKS
The two lessons drawn by Wicksell from the 1907 American crisis—to recapitulate, the attributes of an efficient banking system, and the pivotal role of available reserves—may be found in part in previous contributions made by other monetary economists (especially Henry Thornton and Walter Bagehot). However, Wicksell articulated those elements in a new way, which reflected his concept of a pure-credit economy and his model of the cumulative process. In particular, although the notions (if not the words) of a bank’s liquidity and solvency had been often deployed in discussions about bank runs in the eighteenth and nineteenth centuries (see de Boyer Reference Gårdlund2009), Wicksell’s clear distinction between them opened new ground—or so it would, had his 1908 essay become available in one of the major European languages.
The terms “solvency” and “liquid” could be found in contemporary narratives of the 1907 crisis (see, e.g., Sprague Reference Sprague1908, pp. 857, 360), but they were not systematically related to one another. Indeed, the issue of the interdependence between insolvency and illiquidity of financial institutions has resurfaced as part of the analytical effort to understand the 2007–09 crisis (see, e.g., Morris and Shin Reference Morris and Shin2009). Wicksell’s argument about the interaction between flexibility, liquidity, and solvency, which points to the microeconomic effects (on banking firms) of macroeconomic instability, also broke new ground. His notion of a bank’s flexibility was related to Bagehot’s and Thornton’s views about the role of the lender of last resort, but added new elements to it, such as the stabilization of the price level as a goal of monetary policy. Moreover, Wicksell’s discussion of the 1907 crisis makes it clear that the concepts of solvency and liquidity apply also to a pure-credit economy, where financial markets are not complete and other riskless nominal assets are not perfect substitutes for the liabilities of the central bank (Boianovsky and Trautwein Reference Boianovsky and Trautwein2006, section 2). Wicksell’s interpretation of the historical episode of 1907 is consistent with the later-developed hypothesis (see, e.g., Gorton Reference Gorton1988) that panics result from reassessments of the riskiness of the banking system when a new piece of information becomes available (such as the failure of a particular corporation with links to the banking system), and agents, who cannot tell the value of a bank’s assets, are not able to distinguish the degree of risk of one bank from another.
