10. The Inter-Bank Market in the Perspective of Fractional Reserve Banking

10

Nikolay Gertchev

---

The Inter-Bank Market in the Perspective of Fractional Reserve Banking

In view of the diversity and depth of the analytical contributions of the Theory of Money and Credit, modern Austrians rightly consider Ludwig von Mises as the monetary theorist of the twentieth century. However, Mises’s monetary doctrine did not reach full completion until the publication of his broader treatise Human Action, which clarifies or even amends, sometimes in a substantial manner, his earlier views on important monetary issues (Gertchev 2004). One question that is further elaborated in Human Action is that of the existence of natural limits to the amount of fiduciary media of exchange that freely competing fractional reserve (FR) banks would be able to issue. In the Theory of Money and Credit, while pointing out that there are no natural limits to the creation of fiduciary media of exchange, Mises adds repeatedly that a single bank could not expand its circulation credit unlimitedly if its competitors decided to act otherwise. In Human Action, Mises strengthens his earlier position by asserting painstakingly that a competitive FR banking system, i.e., one that is free from government intervention and is only subject to private business law, would be very efficient in restricting bank credit expansion, and hence in preventing inflation. Mises’s completed argument, which was adopted and further expounded by Murray Rothbard, has become a building block of the standard Austrian banking theory.

One of the goals of this paper is to revisit the foundation of this building block. We try to show that the self-limitation of bank credit expansion, which Mises considers to be inherent to FR free banking, is not apodictic, but rather dependable on specific actions being undertaken by the banks. This allows us to present the inter-bank loan market as an alternative pattern of bankers’ actions, namely a pattern that results in the removal of these so-called natural limits. The first section presents and critically analyzes the case for the existence of the natural limits. The second section offers an Austrian approach to the inter-bank market.[1] The third and last section contains an empirical illustration of the formal analysis based on recent developments in the euro area.

I. The Case for Natural Limits on Expansion in a Fractional Reserve Free Banking System

In Human Action, Mises goes to great lengths to show how the coexistence of multiple FR banks implies strong restraints on the capacity of the banking system to expand.[2] The argument first establishes the individual constraints for a single institution that would engage in credit expansion alone before drawing a generalization at the system’s level.

Liquidity Outflows from Inter-Bank Settlement

At the level of the individual analysis, Mises makes a crucial distinction between the clients and the non-clients of the expanding bank. The clients are those who express a demand to hold the fiduciary media issued by that specific institution. To the contrary, non-clients are these individuals who do not have a demand to hold the bank’s fiduciary media, and who then present them for redemption in money, should they happen to receive them in exchange of goods and services. As a consequence, an expanding bank will soon be brought to contract by the amount which would be needed to redeem that portion of the additional media of exchange that would have fallen in the hands of non-clients. Otherwise, the bank would not have sufficient money reserves to pay out, and would have to acknowledge its insolvency. In Mises’s own terms:

But now, we assume further, one bank alone embarks upon an additional issue of fiduciary media while the other banks do not follow suit. . . . In order to settle the payments due to nonclients, the clients must first exchange the money-substitutes issued by their own—viz., the expanding bank—against money. The expanding bank must redeem its banknotes and pay out its deposits. Its reserve . . . dwindles. The instant approaches in which the bank will—after the exhaustion of its money reserve—no longer be in a position to redeem the money-substitutes still current. In order to avoid insolvency it must as soon as possible return to a policy of strengthening its money reserve. It must abandon its expansionist methods. (Mises 1998, p. 434)

Mises further notes that, even if everybody was accepting the new fiduciary media without discrimination for their issuer, some of them would have been deposited at competing banks, which in turn would have asked for redemption according to the same principle (ibid., p. 435). The process of inter-bank settlement would then act as a check on each bank’s capacity to inflate alone. In both cases, the fundamental principle at action is rooted in the incapacity of a bank to expand its own clientele, i.e., the demand to hold its fiduciary media, by an expansion of its supply of bank credit:

The concatenation which sets a limit to credit expansion under a system of free banking . . . is brought about by the fact that credit expansion in itself does not expand a bank’s clientele, viz., the number of people who assign to the demand-claims against this bank the character of money-substitutes. Since the over-issuance of fiduciary media on the part of one bank, as has been shown above, increases the amount to be paid by the expanding banks’ clients to other people, it increases concomitantly the demand for the redemption of its money-substitutes. (ibid., p. 441)

Mises generalizes his conclusion to the level of the system on the ground of competing banks’ unwillingness to collude lest they should lose their good reputation. He calls “preposterous” the idea that free banks would form a cartel in order to expand together and thus remove the limits upon individual attempts at expansion. Mises concludes:

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. (ibid., p. 444)

From this, Mises implies that any generalized expansion of credit is due to government intervention in the sector, which alone prevents the operation of the efficient natural check. Such is his confidence in the efficiency of a free FR banking system for averting credit expansion that he declares:

What is needed to prevent any further credit expansion is to place the banking business under the general rules of commercial and civil laws compelling every individual and firm to fulfill all obligations in full compliance with the terms of the contract. . . . Free banking is the only method available for the prevention of the dangers inherent in credit expansion. . . . Only free banking would have rendered the market economy secure against crises and depressions. (ibid., p. 440)

Rothbard reiterates Mises’s argument, while providing one additional point and one clarification. He justifies banks’ unwillingness to collude on profitability grounds: “The banks are competitors, not allies. . . . The longer the Boonville Bank holds off on redemption the more money it loses. Banks therefore have everything to lose and nothing to gain by holding up on redeeming notes or demand deposits from other banks” (Rothbard 2008, p. 118). While Rothbard acknowledges that a bank cartel is logically possible, considerations about tarnishing one’s reputation lead him to conclude that cartelization is highly unlikely in a free banking system. One cannot avoid noticing that Rothbard’s argument is not built upon a priori foundations, but upon contingent considerations about the most plausible course of action being undertaken by free FR bankers. On these fundamentally empirical grounds, he sides with Mises and reaffirms that credit expansion and inflation are not embedded in FR banking per se, but in a FR banking system that is protected by a central bank: “Free banking, then, will inevitably be a regime of hard money and virtually no inflation” (ibid., p. 125).[3]

Rothbard also clarifies that the drain on the expanding bank’s reserves is but a specific case of the more general Hume-Ricardo specie outflow mechanism that explains the equilibration of the balance of payments between nations on the ground of international money transfers: “. . . we should now be able to see that the Ricardian specie flow price process is one and the same mechanism by which one bank is unable to inflate much if at all in a free banking system” (ibid., p. 122). This is indeed a valuable clarification. Even though Mises casts his account of the inter-bank settlement process in terms general enough to cover the case of the inter-nation settlements too, he does not explicitly state that it is but an application of the classical economists’ specie flow price process.

By emphasizing so much the case for the existence of a natural check on credit expansion, Mises and Rothbard revive an old dispute from the beginning of the nineteenth century between advocates of free banking and proponents of the central banking system. Henry Parnell is the first to present the restrictive impact of the inter-bank clearing system (Parnell 1827, pp. 86–87).[4] Rothbard refers to Parnell very favorably as to one of the hard money advocates of the free banking school (Rothbard 1988, p. 237). Parnell’s argument was already challenged at his time, and a critical assessment of Mises’s strong faith in the non-inflationary virtues of a free FR banking system should start from the objections raised by McCulloch and Longfield.

The Critique: How Are Competitors Reacting?

McCulloch pointed out that non-expanding banks are equally impacted by the liquidity drain, because banknote users do not discriminate between banknotes issued by the expanding bank and those issued by others.[5] Non-expanding banks will then have to redeem part of their notes too, and at the end will lose market shares. The only means for them to avoid such an outcome would be to follow suit, lower their own discount rate, and expand their issuance: “The over-issuing bank must necessarily deal with the public on terms more advantageous to them than her rivals; and this circumstance would effectually disable the latter from counteracting her operations, except by following a similar line of conduct” (McCulloch 1831, p. 50; original emphasis). However, McCulloch’s criticism of Parnell misses the point. It refers only to that check on credit expansion which derives from the redemption of that part of banknotes that individuals consider excessive in their total money holdings. As a matter of fact, Parnell’s specific point, namely the redemption demands which come out of the very spending process of the new fiduciary media of exchange, remains unaddressed.[6]

As a matter of fact, Parnell admits that the non-expanding banks will suffer from a liquidity drain, but suggests another final outcome:

The principle of private interest, when the trade of banking is free, provides a complete protection against the interference of weak banks. For if a bank force a larger quantity of its notes into circulation than its capital and fair dealings justify, as the circulation will admit of only a certain amount in the whole, this bank will diminish the quantity of paper in circulation of the other banks, and injure their interest: to this they will not submit, but they will combine together to collect the paper of the offending bank, in order to make a run upon it. (Parnell 1827, p. 89; emphasis added)

This line of defense implies that cartelization of the banking sector is possible, but only among banks of good standing. However, what Parnell omits to clarify is how the collection of the expanding bank’s notes will be funded. The non-expanding banks must acquire the banknotes in exchange of either money reserves or of an additional issue of their own notes. In the latter case, they follow suit and confirm McCulloch’s point. In the former case, they lose an increasing portion of reserves, which makes them objectively weaker.

Longfield builds upon McCulloch’s main observation and further notes the strong negative impact of credit expansion by a single institution on the reserve ratio of the non-expanding banks. The latter’s attempt to come back to the previous higher reserve ratio would impose a contraction upon their banknote issuance, i.e., a further decline in their market shares. If the expanding institution is strong enough and continues its expansion, it could then drive its competitors out of business. Longfield suggests that, acting in self-defense and to avoid this outcome, all banks will follow the first-mover, thereby generalizing the expansionary process and forcing the economy into a boom-bust cycle (Viner 1960, p. 242).[7]

Our short account of this fascinating debate suggests that both camps hold a piece of the truth. On the one hand, Parnell is right in emphasizing that the inter-bank clearing mechanism is draining liquidity out of the expanding bank. On the other hand, McCulloch and Longfield are right in responding that the aggregate impact depends much on the way other banks will react, and they make a compelling case for the choice to follow the expansionary policy. Among the modern Austrians, Jesús Huerta de Soto is an advocate of this middle-way synthesis. He acknowledges that “the first bank to launch an expansionary policy derives the most profit and eventually establishes a position of advantage over its competitors” (Huerta de Soto 2006, p. 634). As a result, free FR banking is “incapable of avoiding credit expansion and the appearance of cycles,” and it “invariably leads to the emergence of a central bank as a lender of last resort” (ibid., p. 666 and p. 638). While recognizing together with Mises and Rothbard the technical aspects of Parnell’s mechanism, Huerta de Soto arrives at a diametrically opposed aggregate conclusion—free FR banking would be naturally inflationary.

This conclusion begs the question whether Rothbard’s overall positive assessment of Parnell as one of the hard money free bankers is not exaggerated. A careful reading of Parnell’s essay reveals that he is a follower of the Ricardo-Smith doctrine according to which paper currency, by substituting itself to gold and silver, increases the productive stock of capital in the economy (Parnell 1828, p. 2). Parnell goes as far as stating that a credit economy requires a transition to paper currency first: “A metallic currency not only deprives a country of one of the best supporters of industry, namely, the assistance of the discounts and loans of bankers, but it occasions a great loss to the public in maintaining it” (ibid., p. 72). As a matter of fact, Parnell is a clear proponent of paper currency, i.e., of convertible fiduciary media of exchange, and an enthusiastic opponent of the monopolization of the issue of fiduciary media by a single bank: “The real evil is not paper money, but the system of banking which has the management of it” (ibid., p. 30). His essay must be seen as a blueprint for monetary reform in England that would be based on freedom from legislative interference, strong capitalization of banks by means of allowing the formation of joint stock companies and accepting the principles of rivalship and competition, thereby “preventing each other from abusing the power of issuing paper money, by forcing too much of it into circulation” (ibid., p. 37). The demonstration that inter-bank clearing will limit credit expansion in free banking is only meant to gain support for the suggested monetary reform by dissipating worries about its possible inflationary impact. The core of the plan, however, is to allow all banks, and not only the Bank of England, to collect the profits from issuing fiduciary media, by stretching their reserve ratios as much as possible: “On the whole, sufficient has been said to show that it is in the interest of every commercial country to introduce a free system of banking, and the most extensive use of paper money, consistent with its convertibility into coin” (ibid., pp. 91–92; our emphasis). Statements like this one would hardly qualify anyone as a hard money free banker.

The criticisms which were addressed to Parnell, and which Huerta de Soto has already integrated in the Austrian model, refer exclusively to the second part of Parnell’s argument, namely the generalization of the individual liquidity outflows at the aggregate level. What we would like to show now is that the presumably unavoidable nature of even the individual liquidity outflows is equally subject to criticism.

II. The Case for an Inter-bank Market in a Fractional Reserve Banking System

An inter-bank market between 100-percent reserve banks would have no special features to analyze. The very fact of keeping an integral reserve implies that any transfer of bank-issued media of exchange, i.e., of monetary certificates in this case, from one bank to another must be accompanied by an immediate inter-bank settlement through an effective exchange of reserves. Otherwise, the 100-percent reserve ratio would be violated. Banks could still lend to each other, but only to the extent that they would engage in the additional, and analytically distinct, activity of financial intermediation, i.e., of borrowing and lending money. If such were the case, an inter-bank market would be limited in size, given the limited scope for a financial sector intermediary to borrow from competitors rather than from capitalist-savers. An inter-bank market related to the specific banking activity is conceivable only in the case of FR banking.

The Inter-Bank Market as an Indispensable Financial Institution

FR banks can operate in a commodity (gold, silver) money regime, without an explicit minimum reserve requirement, and keep their reserves separately from each other. Or they may also function in a fiat paper money system, be subject to a minimum reserve requirement, and pool all their reserves at a single central bank, which is in charge of controlling the quantity of reserve (base) money. While the latter case alone corresponds to present-day reality, there are economic incentives for FR banks to lend to each other under all circumstances.

Let us assume that two FR banks A and B issue fiduciary media of exchange, and that they are fully loaned up to the maximum allowed by their respective targeted reserve ratios. Any payment that a customer of bank A would make to a customer of the bank B would imply a liquidity outflow from A to B. Let us further assume that over the reference time period, after netting out all payments between customers of A and B, bank A has a negative net balance against B. Bank A must then find a solution to the problem of financing this negative balance.

An outright payment, i.e., an immediate settlement of the net balance by a transfer of reserve money, is the only case analyzed by Parnell. The point, however, is that this case is specific to FR banks that keep reserves in excess of their targeted ratio. A fully loaned-up FR bank can fund on the spot, i.e., by an outright money payment, only that fraction r (0 < r < 1) of its net negative balance that corresponds to its reserve ratio. Should the FR bank go beyond this fraction, it would breach its targeted reserve ratio on the remaining fiduciary media of exchange it has issued. Hence, in order to keep the required reserve ratio, bank A must borrow money in order to finance that portion (1—r) of the net negative balance that goes beyond the fraction r covered by reserves. This, in a nutshell, is the fundamental rationale for the existence of an inter-bank market. The periodic net negative balances between banks, related to and implied by their customers’ usage of fiduciary media of exchange, necessarily imply that FR banks have to borrow money.

If bank A borrows from a third party, for instance by seeking re-financing from the bank C, then bank B receives as much reserves as is the net positive balance in its favor, i.e., the increase in the deposits it has issued. It ends up holding reserves above the targeted reserve ratio. It is then in the position to lend out these excess reserves or to expand its bank credit until it is again fully loaned up. Bank B could even consider lending these reserves to A, in which case A would not have to seek re-financing from C. In other words, B could accept delayed payment from A for that portion of A’s negative net balance which physically cannot be financed by A’s corresponding reserves. In addition to the interest gain that B would collect from this operation, there is a longer-term interest in accepting to grant A re-financing. In particular, the net balance of the next reference period could be to the advantage of A, in which case it would be B’s turn to seek re-financing. Having already re-financed A in the first period, B could reasonably expect, and act accordingly, that A will be willing to grant it re-financing in the second period, as well as later on.

The inter-bank market appears as an efficient tool that FR banks can use to deal with the unpredictable effect that customers’ economic transactions exert on banks’ stock of reserves. By granting themselves mutual loans, FR banks can de facto re-finance themselves and remain in control of their targeted reserve ratios. The inter-bank market is the mechanism by which FR banks dissociate their individual holdings of reserves from the liquidity implications of customers’ economic transactions. Hence, banks avoid the necessary swings in their credit granting activity, according to whether they need to replenish their targeted reserves by contracting credit or to use up their excess reserves by expanding credit. In the special case of a commodity money system with decentralized reserves, the inter-bank market also saves the cost of continuously shipping money between banks.

A comparison with the financing of a negative current account balance in international transactions would be highly instructive at this point. The Parnell-Mises-Rothbard model of how a FR banking system operates assumes that any negative balance between banks is financed on the spot through a money transfer. The fact is, however, that this needs not be the case, and that such a negative balance can be financed through a short-term loan, very much like the way in which negative current account balances are not always financed through a transfer of international reserves, but through an inflow of foreign capital by means of short or longer-term loans. Jacob Viner rightly points out that “Such movements of short-term funds in a reverse direction from the actual or incipient movement of specie are helpful to the international mechanism of adjustment in two main ways” (Viner 1965, p. 403). First, they smooth the movements of gold between countries. Second, they avoid a sudden contraction in bank credit, and subsequently in the money supply and in domestic prices. The analogy with the inter-bank market is straightforward.

More importantly, the inter-bank market also makes it possible for all banks to benefit from the expansionary policy of a single credit institution. Let us assume that bank A has noticed a permanent increase in its reserves above its targeted ratio and hence decides to expand its credit. By re-financing the subsequent net negative balance with other banks on the inter-bank market, the expanding bank de facto shares, with its new creditors, its credit expansion-induced profits. Hence, the inter-bank market effectively collectivizes the profits that an over-expanding bank might have tried to collect. This latter conclusion is of particular interest, as it puts into perspective the very notion of an over-expanding FR bank. Thanks to the inter-bank market, every bank can expand its credit by a multiple of its net increase in reserves, as long as it finds refinancing from other banks.[8] The latter have a financial incentive to grant this re-financing, as this is an effective means for sharing into the higher profits derived from the credit expansion. In a sense, it becomes immaterial how new reserves are distributed among banks and which bank starts credit expansion, as the inter-bank market redistributes the associated profits according to each one’s market share in the issuance of fiduciary media of exchange.

This conclusion seriously questions the existence of natural limits on credit expansion in a free FR banking system. As we have shown, it is in the short and long-term interest of any FR bank to take part in the inter-bank market, which de facto prevents the operation of the strict clearing mechanism described by Parnell. By engaging into mutual loan transactions, FR banks can free themselves from the restraints of the immediate outright clearing. Furthermore, we have shown that the very existence of fiduciary media of exchange is preconditioned on banks’ accepting to lend to each other, as the principle of the fractional reserve is incompatible with the capacity to fully fund inter-bank net negative balances with money.

The inter-bank market is indispensable for FR banks in an even more fundamental way. Each FR bank must solve the difficult problem of creating a demand to hold its fiduciary media of exchange. In Mises’s terminology, it must create, maintain, and enlarge its clientele. Now, how plausible is it that the fiduciary media issued by bank A would be acceptable to money holders at large if they are not acceptable to other FR banks? As a matter of fact, another way to look at the inter-bank market is to consider it as the emanation of banks’ mutual recognition of their fiduciary media of exchange. Indeed, a bank B that does not require a money transfer from bank A for spot settlement of their mutual net balance de facto accepts to hold bank A’s fiduciary media of exchange for a more or less determined period of time.[9] The inter-bank acceptability of fiduciary media of exchange is definitely a precondition for their acceptability by banks’ clients. This means that the inter-bank market is an indispensable financial institution of the fractional reserve banking system. But then, contra Parnell, Mises and Rothbard, a free FR banking system is no rampart against inflation and the boom-bust cycle.

Factors that Determine the Extent of the Inter-Bank Market

While FR banks cannot avoid undertaking mutual credit transactions, there are a number of factors that determine the latter’s extent. The size of the inter-bank market depends on the degree of concentration in the banking sector, the level of targeted reserve ratio, the degree of trustworthiness between banks and, most crucially, upon the presence of an unlimited lender of last resort.

The more concentrated the banking sector is, the greater is the scope for netting out inter-bank liquidity outflows, and hence the smaller is the net inter-bank balance that requires financing. On the contrary, the higher the number of institutions, and the smaller any individual market share is, the more complex becomes the multilateral compensation between outflows and inflows of liquidity. It could then be expected that each institution’s net financing needs would aggregate into a larger and more complex nexus of inter-bank credit transactions. A higher degree of rivalry,[10] to borrow Parnell’s quite aptly chosen term, increases then banks’ financing needs and contributes to a more dynamic inter-bank market.

Second, the size of the inter-bank market is inversely related to the targeted reserve ratio. In the extreme case where the reserve ratio is at 100 percent, there is no scope for the emergence of inter-bank credit related to the activity of issuing media of exchange. In the other extreme case of a zero reserve ratio, which is conceivable only in the framework of a fiat paper money, any inter-bank transaction can be but a credit transaction, as no bank has any reserves or aims at holding any of it. Generally speaking, the lower the targeted reserve ratio, the lesser is banks’ capacity to finance their negative balances with other institutions by reserves. The higher must then be the portion of the negative balance financed by a credit transaction. As a result, the inter-bank market is growing in size when the required reserve ratio is declining.

The third key factor, which ultimately determines the size of the inter-bank market, at given composition of the sector and reserve ratio, is the degree of trustworthiness among banks. The stronger the confidence of banks in their respective financial strength and solvency, the more inclined they will be to become creditors to each other. Inversely, when doubts arise about the ultimate capacity of a bank to reimburse the loans it has contracted, its creditors may refuse to keep re-financing it and require payment at the maturity of the loan transaction. To the extent that maturing credit transactions are not rolled-over or renewed, the inter-bank market shrinks in size. Trustworthiness in banks’ financial soundness, in a free FR system, would depend crucially on the quality of debtor banks’ investments as well as on the conservativeness and foresight of their creditors. In particular, inter-bank credit would be reserved for institutions of similar, not lower, standing with respect to financial soundness. This does not imply that free FR banks would all be conservative lenders and very cautious about their financial standing. Rather, this means that the sector will be characterized by a uniform credit policy, from which individual deviations could be quickly sanctioned.

The last factor, which exerts a crucial influence on the size of the inter-bank market, is the presence of an unlimited lender of last resort. Under a fiat paper money regime, which is the only case when a central bank can increase banks’ reserves ad infinitum, any bank can be re-financed by the central bank. Should a bank fail to pay at the maturity of a credit transaction, it could always seek support from the central bank, and there is no technical impediment to such bail-out. In this institutional environment, which describes well present-day reality, liquidity risk practically vanishes away. More importantly, banks’ assessment of their rivals’ trustworthiness becomes strongly biased in favor of granting unconditional loans, as the risk of making losses on them is objectively reduced by the very presence of an unlimited lender of last resort. Banks’ higher willingness to lend to each other translates into an increased size of the inter-bank market and, subsequently, stronger interdependencies between institutions.[11] A fiat paper money regime becomes then exposed to regular banking crises due to a sudden degradation of the trustworthiness among banks. A revaluation of a bank’s risk profile by its creditors would imply requests for repaying the loans at maturity. A bank which has lost the confidence of its rivals is then excluded from the inter-bank market and must seek the assistance of the central bank. While the presence of an unlimited lender of last resort leads to an expansion of the inter-bank market beyond what its size would be under a commodity standard, it also contains the seeds for regular swings in inter-bank loan transactions.

III. The Euro Area Inter-bank Market

The presentation of the euro area inter-bank market provides a useful and timely illustration of the theoretical insights that have been derived in the previous section. This section will draw on statistics gathered and published by the European Central Bank in its Statistical Data Warehouse.[12] To begin with, a few clarifications on terminology would be necessary.

The data on money, banking and financial markets does not refer to banks exclusively, but to the broader category of Monetary Financial Institutions (MFIs). The characteristic mark of MFIs is their ability to increase the money supply in the broader sense. In practice, MFIs include credit institutions (commercial and cooperative banks), money market funds and the Euro-system. The Euro-system, which includes the European Central Bank as well as the 17 National Central Banks of the euro area member countries, is better thought of as the consolidated central bank in the euro area. Hence, for most analytical purposes, one must distinguish between, on the one hand, the Euro-system and, on the other hand, the MFIs, excluding the Euro-system. This latter category corresponds, more or less, to the whole of FR banks in the euro area. Statistics from the ECB report the consolidated balance sheet of the Euro-system as well as the aggregated balance sheet of MFIs, excluding the Euro-system. When put together, these two sources of data provide some concrete figures about the euro area inter-bank market. In statistical terms, the euro-area inter-bank market is represented by the loans granted by MFIs, excluding the Euro-system, to other MFIs, excluding the Euro-system.[13]

The first characteristic of the inter-bank market illustrated by the data is its impressive size. Inter-bank loans amounted to 5.2 trillion of euros in April 2012, which represented more than 15 percent of banks’ total assets and almost 43 percent of banks’ loans to the euro-area economy. Moreover, the significance of the inter-bank market has been even more impressive during the boom years, when for every two euros lent to the economy, banks lent more than 1 euro among themselves (see Chart 1).

Chart 1

The relative size of the euro-area inter-bank market

Source: BCB Statistical Data Warehouse

Second, the inter-bank market has grown in nominal terms, together with the expansion of bank credit to the non-monetary part of the economy. This illustrates our finding that the inter-bank market is the mechanism, by which banks refinance their liquidity needs on a daily basis, without having to borrow substantial amounts from the central bank. The substitutability between borrowing from others banks and refinancing at the central bank has become even more pronounced since the autumn of 2008, when the intensification of the world financial crisis resulted, inter alia, in an increased loss of confidence among European banks (see Chart 2). In order to repay their creditors, who refused to roll-over inter-bank loans, debtor banks had to borrow heavily from the Euro-system. This development has become even more pronounced in late 2011 and at the beginning of 2012. At the same time, the outstanding volumes in the inter-bank market have stabilized, although at levels which are some 15 percent below their peak in October 2008.

Chart 2

The Euro-area inter-bank market (loans to MFIs, excluding the Euro-system) and refinancing at the central bank (borrowings from the Euro-system)

Source: BCB Statistical Data Warehouse

Third, the significance of the inter-bank market for accompanying credit expansion is clearly revealed when comparing the dynamics of loans to the economy with loans among banks (see Chart 3). The stagnation of loans to the economy in the aftermath of the Lehman bankruptcy goes hand in hand with the contraction of inter-bank credit. The modest credit expansion since 2010 has occurred despite a further contraction in inter-bank credit and thanks to the massive liquidity injections by the Euro-system.[14] Interestingly enough, there has been no significant recovery of credit to the economy, as there has been no recovery in the inter-bank lending.

Chart 3

The inter-bank market as a foundation for bank credit expansion

Source: BCB Statistical Data Warehouse

Any of these three features of the euro area inter-bank market is fully consistent with our theoretical conclusions. In addition, they also portray the history of the single currency as a relatively long boom-bust cycle, with the crisis having emerged in late 2008. The shrinking of the inter-bank market since then is to be interpreted as the end of the acceptability of some banks’ fiduciary media of exchange by other peer banks. The refusal to renew inter-bank loans at maturity is tantamount to a request for an outright settlement, i.e., for redemption of the fiduciary media of exchange. Put simply, this is a run on some banks by other banks. And because FR banks have no spare money available, such a bank run may result either in the banks’ bankruptcy or in their refinancing by the central bank. While for now policy makers have taken the latter course of action, the fact that the euro area inter-bank market has not recovered four years after the outburst of the crisis suggests that the run on banks by other banks has not come to an end yet.

Conclusion

The inter-bank market, though impressive by its size and omnipresent in banks’ daily business activity, has not received much attention in economic theory. The purpose of this paper has been to fill this gap from the point of view of Austrian economics. Two main conclusions have been reached.

First, an inter-bank loan market appears indispensable for the existence of fiduciary media of exchange in a decentralized banking system. It is an essential element of the fractional reserve system, in the sense that decentralized fractional reserve banks could not possibly exist without accepting to engage in mutual loan transactions. The inter-bank market is foundational for bank credit expansion.

Second, the reality of the inter-bank market sheds a different light on the expected outcome of the so-called fractional reserve free banking system. In particular, the inter-bank market is incompatible with the view that banking freedom would put natural limits on the issuance of fiduciary media of exchange. This implies that banking freedom is not a guarantee against the inflation of the money supply in the broad sense. Hence, and despite Mises’s and Rothbard’s affirmations to the contrary, the policy of banking freedom could not be an element of a successful strategy for restraining credit expansion by fractional reserve banks. This finding rationalizes the widespread government-controlled policies of prudential regulation and supervision of banks. Whether these policies could be efficient for reining in inflation has been left outside the scope of this paper.

Bibliography

Cook, Timothy, and Robert Laroche, ed. 1993. Instruments of the Money Market. Richmond: Federal Reserve Bank of Richmond.

Gertchev, Nikolay. 2004. “Dehomogenizing Mises’s Monetary Theory.” Journal of Libertarian Studies 18, no. 3: 57–90.

Huerta de Soto, Jesús. [1998] 2006. Money, Bank Credit and Economic Cycles. Auburn, Ala.: Ludwig von Mises Institute.

McCulloch, J.R. 1831. Historical Sketch of the Bank of England: With an Examination of the Question as to the Prolongation of the Exclusive Privileges of that Establishment. London: Longman, Rees, Orme, Brown and Green.

Mises, Ludwig von. [1949] 1998. Human Action, Auburn, Ala.: Ludwig von Mises Institute.

Parnell, Henry. 1828. Paper Money, Banking and Overtrading, Including Those Part of the Evidence Taken Before the Committee of the House of Commons, which Explains the Scotch System of Banking. London: Charles Wood and Son.

Rothbard, Murray N. [1983] 2008. The Mystery of Banking, Auburn, Ala.: Ludwig von Mises Institute.

——. 1988. “The Myth of Free Banking in Scotland. Review of Austrian Economics 2: 229–46.

Smith, Vera. [1936] 1990. The Rationale of Central Banking and the Free Banking Alternative. Indianapolis: Liberty Fund.

Viner, Jacob. [1937] 1960. Studies in the Theory of International Trade. London: Bradford and Dickens.

---

Nikolay Gertchev holds a Ph.D. in economics from the University of Paris II Panthéon-Assas and is currently based in Brussels, Belgium where he works for an international organization.

[1] Throughout this paper, the notion of inter-bank market refers exclusively to loan transactions between banks, independently of their legal characteristics (secured by the pledge of collateral or unsecured). Thus, the inter-bank market is understood to be part of the so-called money market, which typically refers to all short-term financial investment instruments. For a comprehensive review of the money market, see Cook and Laroche (1993).

[2] Chapter XVII, which deals exclusively with money and banking, dedicates to this specific question some 13 pages out of a total of 80 pages.

[3] Consider also the following statements: “Free banking, even where fractional reserve banking is legal and not punished as fraud, will scarcely permit fractional reserve inflation to exist, much less to flourish and proliferate. Free banking, far from leading to inflationary chaos, will insure almost as hard and noninflationary a money as 100 percent reserve banking itself” (Rothbard 2008, p. 117).

[4] Jacob Viner asserts: “That the power to over-issue of a single bank, operating in competition with other banks, was closely limited, had long been known” (Viner 1965, p. 238). He dates Parnell’s mechanism back to an essay by Adam Dickens from 1773, as well as to a book by Lord King from 1804.

[5] After providing a numerical example, McChulloch concludes: “And as all notes are, under the circumstances supposed, equally good in his estimation, he sends those in for payment that comes first to hand. . . . So long as he believes the different notes to be alike good, he will show no preference to one more than to another, but will return them indiscriminately upon their issuers, while he can make a profit by doing so” (McCulloch 1831, p. 48).

[6] Jacob Viner makes this very clear: “McCulloch failed to point out that a single bank which expanded its note issue while other banks remained passive or contracted would suffer a drastic impairment of its reserves” (Viner 1965, p. 241). However, Viner does not notice that Parnell has foreseen McCulloch’s main point and that he diverges in his assessment of the outcome of banks’ interactions.

[7] A lengthy discussion of this argument, together with a numerical annex and additional considerations as to the impact of the loan maturity on the liquidity drain experienced by the expanding bank, is to be found in Vera Smith (1990, pp. 177–85 and 197–200).

[8] This conclusion relativizes the common text-book presentation, also followed by Austrian economists, of the so-called money multiplier. The theory of the money-multiplier states that, at the aggregate level, the banking system can issue fiduciary media of exchange by a multiple of the increase in banks’ money reserves. The multiplier is equal to the inverse of the complementary to one of the percentage of banks’ liquidity outflows. Typically, the aggregate money multiplier is derived from a sequential analysis of individual banks’ credit expansion, whereby each bank expands credit by the exact amount of the increase in its reserves, then loses part of its reserves to other banks because of customers’ transactions, then it is the other banks’ turn to expand credit to the increase in their reserves, etc. The system’s multiple increase in the fiduciary media of exchange is then presented as the limit to the sum of infinite rounds of ever-decreasing issuances by the individual banks. This standard presentation de facto assumes that banks must fund their liquidity outflows by spot transfers of money exclusively, and that accordingly each bank must limit its credit expansion to the amount of liquidity inflow. However, this needs not be the case, as the individual institution might rely on the inter-bank market to finance its un-funded liquidity outflow. When the inter-bank market is taken into consideration, the money multiplier needs not be derived as the sum of an infinite geometrical series. It can be reached in a one-go, as long as the expanding bank can obtain refinancing from its peers.

[9] Even if the inter-bank loan has an initial maturity, nothing precludes it from being extended upon expiration.

[10] Indeed, the notion of competition is to be reserved for the interactions between owners of legitimately acquired private property. There are some doubts as to whether FR banks would fall into this category. On the latter point, see in particular Huerta de Soto (2006).

[11] The presence of an unlimited lender of last resort increases the size of the inter-bank market also through its impact on the required reserve ratio, which is typically very low, given the unlimited bail-out possibilities.

[12] The database is available at this web-address: http://sdw.ecb.europa.eu/

[13] This statistical category is not readily available, however. The aggregated balance sheet of the MFIs, excluding the Euro-system, distinguishes only between loans to MFIs, i.e., including the Euro-system, and loans to non-MFIs, i.e., households, corporates, the government and non-monetary financial institutions, such as pension funds and insurance companies. When looking at the inter-bank market, one needs data for the loans granted from banks to other banks, i.e., for loans from MFIs, excluding the Euro-system, to other MFIs, excluding the Euro-system. This category can only be derived, namely by subtracting MFIs’ deposits at the Euro-system from loans from MFIs, excluding the Euro-system, to MFIs. Indeed, MFIs’ deposits at the Euro-system are classified as part of the broader category of loans to MFIs. These deposits happen to be published as an independent series from the consolidated balance sheet of the Euro-system. That is why one needs to combine information from different balance sheets in order to arrive at data about loans from MFIs, excluding the Euro-system, to other MFIs, excluding the Euro-system.

[14] In addition to the standard collateralized lending to banks, the Euro-system has increased banks’ liquidity by measures that are non-orthodox for the euro area, such as 1) secondary market purchases of government bonds (Securities Market Program), 2) targeted purchases of banks’ bonds (Covered bonds Programs I and II) and 3) dollar loans, after swapping euros for dollars from the FED. Hence, the increase in the amount of borrowings from the Euro-system is only a portion, albeit the largest, of all extra liquidity that has been injected since 2008.