__________________________________________________________________ Nicolás Cachanosky Huerta de Soto ’ s Concerted Expansion : A Prisoner ’ s Dilemma in Free Banking ? Introduction Concerted expansion is a common concern present in discussions about free banking performance . Huerta de Soto ( 2012 [ 1998 ], pp . 664 – 71 ) uses game theory to show that banks have incentives to collude and expand fiduciary media . According to Huerta de Soto , his game illustrates that incentives exist for banks to over-expand to the point that a crisis unfolds . His approach , however , does not accurately describe the problem of concerted expansion in free banking . Van den Hauwe ( 2008 ) critically discusses the general use of the prisoner ’ s dilemma in the free banking literature . The following commentary discusses seven specific limitations to applying the prisoner ’ s dilemma model to concerted expansion in free banking . The critical discussion of these points also helps to focus on relevant challenges for concerted expansion that are not clearly present in Huerta de Soto ’ s treatment . Nicolás Cachanosky ( Ph . D ., Suffolk University , 2012 ) is Assistant Professor of Economics at Metropolitan State University of Denver ( USA ). The author would like to thank Benjamin Powell and Adrián Ravier for helpful comments . He is solely responsible for any errors or omissions . Huerta de Soto ’ s Use of Game Theory in Concerted Expansion In Money , Bank Credit , and Economic Cycles ( 2012 [ 1998 ]), Huerta de Soto defends the 100 percent rule for banking reserves . There are two lines of argument for the convenience of forbidding fractional reserves in banking . The first of these is drawn from what Huerta de Soto refers to as “ traditional legal principles ” that he traces back to the Roman Empire . According to this position , the fractional reserve practice contradicts basic legal principles by ( 1 ) lacking a specific term or maturity date , which according to Huerta de Soto ( 2012 [ 1998 ], pp . 17-18 ) is “ impossible to imagine ” in a monetary loan contract , and ( 2 ) by the absence of safekeeping of the deposit . For Huerta de Soto , the banker performs a fraudulent act if he lends the deposit for his own benefit . Because fractional reserves would be an irregular contract , such practice should be avoided and a 100 percent rule should be enforced . This point of view has been contested by White ( 2007 ) and Yeager ( 2010 ), who argue that such outdated terminology and legal practices are inappropriate for an analysis of contemporary banking and financial practices . The second line of argument , constructed from the first one , pertains to the Laissez-Faire , No . 38-39 ( Marzo-Sept 2013 ): 1-6

__________________________________________________________________ Bank A Does not expand Expands Bank B Does not expand Expands The survival of both ( reduced profits ) The failure of B The survival of A The failure of A The survival of B Large profits for both instability of fractional reserves . An economic crisis may come in two forms : ( a ) a crisis as described by the Austrian business cycle theory , or ( b ) a crisis caused by a bank run that provokes bankruptcy and financial instability . Huerta de Soto ( 2012 [ 1998 ], p . 665 ) acknowledges that any bank that expands its credit faster than its competitors will see its reserves drop quickly . He continues , however , arguing as follows : Not only is fractional-reserve freebanking incapable of avoiding credit expansion and the appearance of cycles , but it actually tempts bankers in general to expand their loans , and the result is a policy in which all bankers , to one extent or another , are carried away by optimism in the granting of loans and in the creation of deposits . It is a well-known fact that whenever property rights are not adequately defined — and this is the case with fractional-reserve banking , which by definition involves the violation of depositors ’ traditional property rights — the “ tragedy of the commons ” effect tends to appear . Thus a banker who expands his loans brings in a handsome , and larger , profit ( if his bank does not fail ), while the cost of his irresponsible act is shared by all other economic agents . It is for this reason that bankers face the almost irresistible temptation to be the first to initiate a policy of expansion , particularly if they expect all other banks to follow suit to one degree or another , which often occurs ( pp . 666-67 ). Huerta de Soto further asserts that issuer banks in free banking are in a similar situation to that of a tragedy of the commons , the difference being that if not all banks expand in concert then the mechanism of adverse clearing comes into play . He illustrates the situation with the table shown above . Following his argument on traditional legal principles , Huerta de Soto compares the situation of free banking to that of a tragedy of the commons where overexpansion of credit substitutes the depredation of resources . Selgin and White ( 1996 , pp . 92-93 ) criticized this analogy in their discussion of a previous article by Huerta de Soto ( 1995 , p . 33 ). They argued that economists “ conventionally distinguish a ‘ pecuniary externality ,’ an effort on someone ’ s wealth transmitted via the price system , from a ‘ technological externality ’,” and that Huerta de Soto “ fails to grasp this distinction when he mischaracterizes the pecuniary externality from fiduciary media as a ‘ tragedy of the commons ,’ a term that properly applies to a particular sort of technological externality .” Huerta de Soto , however , defended the analogy as correct , arguing that “ the issue of fiduciary media stems from the violation of traditional property rights in connection with the monetary bankdeposit contract , and that hence fiduciary media are not a spontaneous phenomenon of a legally based free-market process ” __________________________________________________________________ 2

__________________________________________________________________ ( Huerta de Soto ( 2012 [ 1998 ], p . 667 ). He concludes that “ it is obvious that given the choices above ” the banks will “ initiate a joint policy of credit expansion which will protect both from insolvency and guarantee handsome profits ” ( p . 668 ). Limitations of the Game Applied to Concerted Expansion Prisoner ’ s Dilemma ? The first limitation with Huerta de Soto ’ s treatment is that , although he refers to the game as one “ typically used to illustrate both cooperative games and ‘ prisoner ’ s dilemmas ’ … ” ( p . 668n ), his presentation does not correspond to that of the prisoner ’ s dilemma . 1 The latter has strictly dominant strategies and one Nash equilibrium that happens to be sub-optimal . Huerta de Soto ’ s game equilibrium , on the other hand , is Pareto efficient because the banks collude to seize large profits for both . The situation that the banks face in the presented form thus does not resemble the problem in a conventional prisoner ’ s dilemma . On the contrary , in Huerta de Soto ’ s treatment the players collude in the Pareto efficient equilibrium . The concerted expansion becomes not a dilemma , but the best outcome for the banks given the structure of the presented game . Repeated Game The second limitation is that Huerta de Soto presents the game as a one-move game . The conclusion that a central bank will finally appear is not deduced from 1 Note that in the 2009 Spanish edition this reference to cooperation games and prisoner ’ s dilemmas is absent . the game itself but must be assumed and then added to the end of the game . Although Huerta de Soto argues that colluding banks and central banks are designed to “ orchestrate ” and “ organize ” concerted expansion , the non-repetitive aspect of the game does not describe the concerted expansion problem before the central bank is created . This is a relevant point since the problem of the prisoner ’ s dilemma vanishes in the case of repeated games . Certainly banks are reviewing their decisions frequently , and therefore a repeated game is a more appropriate assumption than a non-repeated game . Barriers to Entry The third limitation has to do with the lack of room in the game for new competitors to join the market when colluded expansion raises income . If collusion were to yield significant returns , as assumed , then it should then attract new competitors into the banking sector . Without such a possibility , the game can hardly be considered an accurate description of the free banking scenario . Although Huerta de Soto ( 2012 [ 1998 ], pp . 669 – 670 ) touches on this point , he concludes that central banks “ generally appear as a result of requests from private bankers ” so that “ the ‘ uncooperative ’ behavior of a significant number of relatively more prudent bankers is prevented from endangering the solvency of the rest .” In other words , he argues that collusion becomes stable because colluding banks promote the ultimate appearance of a central bank to institutionalize the joint credit expansion . But the need and presence of a central bank cannot be an argument that free banking is unstable since the collusion strategy is selfdefeating just as the free-banking literature argues . Colluding banks are unable __________________________________________________________________ 3

__________________________________________________________________ to survive competition absent the protection of a central bank . How is that a valid argument against free banking ? Solvency and Goodwill The fourth limitation is related to solvency and goodwill considerations . This problem was mentioned by Mises ( 1996 [ 1949 ], Chap . 12 ), a reference that Huerta de Soto quotes ( p . 665 ) but a problem that he fails to address . Mises points out that solvency considerations will constrain issuer banks from colluding with less efficient banks ( Mises , 1996 [ 1949 ], p . 441 ). A bank , argues Mises , puts too much at risk by joining a venture with a low goodwill partner . Building goodwill is a difficult and long-term task for banks that can easily be lost setting the stage for bankruptcy . According to Mises : But , some people may ask , what about a cartel of the commercial banks ? Could not the banks collude for the sake of a boundless expansion of their issuance of fiduciary media ? The objection is preposterous . As long as the public is not , by government interference , deprived of the right of withdrawing its deposits , no bank can risk its own good will by collusion with banks whose good will is not as high as its own . One must not forget that every bank issuing fiduciary media is in a rather precarious position . Its most valuable asset is its reputation . It must go bankrupt as soon as doubts arise concerning its perfect trustworthiness and solvency . It would be suicidal for a bank of good standing to link its name with that of other banks with a poorer good will . Under free banking a cartel of the banks would destroy the country ’ s whole banking system . It would not serve the interests of any bank ( Mises , 1996 [ 1949 ], p . 447 , italics added ). Independently of the validity of Mises ’ assessment of solvency and goodwill , the game as applied to free banking does not deal with the problem of differences in solvency and goodwill ; it merely assumes these problems away . Volatility of Reserves The fifth limitation is the impact of the variance of reserves when fiduciary media is expanded on the behavior of banks , an issue that has been pointed out by Selgin ( 1988 , pp . 80 – 82 ; 1994 ). Selgin ’ s point is also not addressed by Huerta de Soto , despite the facts that Selgin ’ s argument was made ten years prior to the first edition of Huerta de Soto ’ s book , and that Huerta de Soto actually quotes Selgin ’ s ( 1994 ) discussion on the effect of the variance of reserves on page 670 ( footnote 98 ). Selgin draws attention to the fact that under concerted expansion , even if the expected reserve value remains unchanged , reserve ’ s variance will increase , which , as a risk measure , will require the banks to increase their precautionary reserve holdings . Selgin recognizes that some concerted expansion may occur if the proportional change in the variance of reserves is smaller than the proportional change of the fiduciary media . According to Selgin ( 1988 , pp . 82 ): Under in-concerted expansion no member of a system of banks expanding in unison ( and in the face of an unchanged demand for money ) will experience any increase in its average net reserve demand ; the change in expected value of its clearing debits . But the growth in total clearings will bring about a growth ( though perhaps less than proportionate ) in the variance of clearing debit and credits , which increases the precautionary reserve needs of every bank . Thus , given the quantity of reserve media , the demand for and turnover of inside money , and the desire of banks to protect themselves against all but a very small risk of default at the __________________________________________________________________ 4