The Evolution of a Free Banking System
In a relatively primitive society in which individuals are just beginning to trade with each other, 'coincidence of wants' problems would arise frequently if market participants were restricted to barter. Some goods would be more in demand than others, however, and at some stage individuals would realize that they had a better chance of getting the goods they wanted if they first accepted some popular intermediate good and then swapped it for the good they wanted to consume. This resort to 'indirect exchange', which employs a certain class (or classes) of intermediate goods, would allow individuals to avoid the 'coincidence of wants' problem, but their transactions costs remain high. In particular, they would still need to measure the quantity of the goods they were offered and assess their quality. They would therefore prefer intermediate goods whose quantity was easily measured and whose quality was relatively uniform. To minimize transport and storage costs, market participants would also want goods that were sufficiently scarce for small amounts to have a high exchange value. Historically, people have tended to converge on the precious metals as desirable intermediate goods and to abandon alternatives as the advantages of precious metals became more apparent.
The use of precious metals as intermediate goods would still leave individuals with the inconvenience of weighing lumps of metal and assessing their purity. This would create an opportunity for some individuals to act as intermediaries and make their living assessing the purity of the metal brought to them and recasting it into pieces of more convenient size. As such practices spread, the fineness and sizes of metal pieces would gradually become standardized, and the private intermediaries would mark the pieces to show their weight and quality. The profits made by the earliest of these intermediaries would attract others, and they would compete with each other for business. It would not take long for them to realize that they could attract more business by using distinctive marks on the metal pieces they issued. The intermediaries would thus become private mints and their metal pieces privately issued coins.
Each of these private mints would exist primarily to maximize its own profits, which could be generated in several ways. One would be by offering competitive minting fees. Another would be by developing a reputation for probity to reassure prospective customers that they would not be cheated. A third would be by innocation: mints would experiment with coins of new denominations, alternative metals, and so on. Any successful innovations would be imitated by the other mints and would become widely adopted. It bears stressing that these mints would have no incentive to cheat by overstating the weight of their coins, because such deception would be easy to detect, and this would harm the mint's reputation and hence its business. Furthermore, the law would classify such activity as fraud.
As an aside, it is exactly at this stage that the state historically has intervened in the monetary system. Governments realized they could use their coercive powers to create a legal monopoly that would make the minting business very profitable. Even if the government's service was inferior to that of the private mints, the public could be forced to accept it, as the state would prohibit its subjects from using the coins of other mints. The government could then impose high minting charges or misrepresent the weight of the coins it issued. Note that it is only the state's monopoly over the means of legal coercion that enables the state mint to stay in business. A private mint could not provide an inferior service and survive because it would have no way of compelling people to use its services.
The use of coins would still involve considerable costs, particularly the cost of storage (including the costs of ensuring that one's coins were safe), and the cost of moving coins around. To avoid storage costs, some people would be prepared to pay others who already had the facilities to keep gold safe - those with strongboxes - to store their gold for them. In practice, this would mean that metalsmiths ('goldsmiths') or merchants who regularly kept large amounts of gold or silver would be asked to look after other people's gold for a fee, and they would probably do so, because the marginal cost would be quite low. Depositors would obtain receipts from those holding their gold or silver attesting to the value of each deposit.
As the practice spread it would increasingly happen that, when two people agreed on an exchange, one would go and withdraw his coins and hand them over to the other, who would deposit them again. Provided that the party accepting payment was satisfied that the goldsmith was likely to honour his commitment, it would be more convenient for him simply to accept the goldsmith's receipt and save both parties the bother of visiting their goldsmiths. Goldsmiths's receipts would this begin to circulate as a medium of exchange in their own right. At the same time, the goldsmiths would begin to notice that only a small proportion of their deposits of gold would be demanded in redemption over any given period, and they would realize that they could lend out some of the gold deposited with them and face little danger of being unable to meet their liabilities. Thus lending activity would give them an opportunity to earn an additional profit.
The goldsmiths would thus become bankers and begin to compete with each other for deposits. One way of doing this would be to offer interest on deposits, replacing the earlier fees charged depositors for the safe keeping of their money. More important, the goldsmith-bankers would also compete for deposits by offering guaruntees to prospective depositors that the receipt notes issued by them would retain their value. Perhaps the most persuasive guarantee they could offer would be to make their notes 'convertible', that is, to promise to conver their notes back into specie. The goldsmith-bankers would in fact have offered such promises right from the start, of course, because no one would have placed deposits with them unless they were assured that they could withdraw them. These guaruntees would have the status of legally binding contracts, and the violation of such contract would therefore expose the banker to the legal penalty for default, which we will assume to be sufficiently high to make a banker careful to avoid it.
This committment to convertibility is one of the most important features of a free banking system, and it has several major implications. First, it would help ensure that banknotes remained relatively stable in value. The value of convertible notes would be tied to the value of gold. It follows that the exchange value of notes against goods in general would fluctuate only with changes in the relative price of gold - that is, the exchange rate between gold and other goods - and we would not normally expect this to be particularly volatile. We might therefore expect the price level to be reasonable stable.
Second, the commitment to the convertibility would provide an effective discipline against goldsmith-bankers who issued an excess of notes. When banks issued convertible notes, their circulation would be limited by the demand to hold them. That demand would depend on such factors as the precise features of the convertibility contract (for example, whether the depositor had to give notice when he wanted to withdraw his deposit), the bank's reputation, the familiarity of its notes, the number of branches it maintained, and so on. Any notes issued beyond the demand to hold them would be returned for redemption. A bank would not deliberately choose to issue an excess of notes, because they would not remain in circulation long enough to justify the expense of putting them into circulation and then taking them back again. If a bank sought to increase its note issue, it might attempt to improve its reputation, advertise its notes more, or open more branches. But it could not increase its note issue simply by putting more notes into circulation. It is one thing to put more notes into circulation, but quite another to keep them there.
The next stage in the evolution of the banking system would be the development of a note clearing system that would arise out of bankers' attempts to raise their profits by increasing the demand for their notes. In the beginning, no banker would accept the notes of other banks when such notes were submitted by the public because to do so would make rival's notes more acceptable and raise his competitors' profits. But any two banks could make themselves jointly better-off by agreeing to accept each other's notes. Each bank would benefit, because the public would more readily accept the notes of either of the two banks, given the knowledge that the other bank would accept the notes at par as well. The notes of those two banks would thus become slightly more attractive that alternative media of exchange such as gold or the notes of other banks. Thus additional bank pairs would be formed, and it would become increasingly apparent that the easiest way to organize the note exchange system would be to meet regularly at a central clearing session where the banks would hand back each other's notes and settle the differences. In this way a central clearing system would evolve out of the bank's own private self-interest.
The clearing system is important because it would provide a further restrainte on the ability of any one bank to overissue its notes. Without a clearing system a bank that overissued would face a reserve drain only from the general public's returning its notes for specie, and it might take some time for this to force the bank to restrain its issues. Once the clearing system was in operation, however, a bank issuing more notes than the public wanted would also face reserve losses at the central clearing sessions. These losses would occur as the public deposited the extra notes at other banks and those banks returned them to the issuing bank. A bank that over-issued notes would thus lose reserves through two channels - through direct redemption by the public, and through indirect redemption via the clearing system - but the latter channel would be likely to operate more quickly.
We have seen that the banker's self-interest would lead to note convertibility and to a central clearing system, and both of these would discipline any bank that overissued its notes. They would therefore contribute significantly to the stability of the monetary system. However, if a bank was committed to redeeming all its note liabilities on demand, it would still face a problem of potential illiquidity, given its ability to redeem only a fraction of its liabilities at any given time. With sufficient advance notice a sound bank would be able to meet demands for redemption by liquidating assets, but a problem could arise if it failed to receive the notice it needed. It is this lack of notice that gives rise to the possibility that an otherwise sound bank might become illiquid and unable to honour its obligations.
Two institutions would develop to deal with the problem. The first would be the growth of a market in short-term liquidity. This would arise because bankers' holdings of liquidity would be subject to random short-term fluctuations that were difficult to predict accurately. At any given time some banks would find themselves with more liquidity than they had anticipated and others with less. Those with 'excess' reserves would be willing to lend them out on a short-term basis, while those that were short of reserves would be willing to borrow them, making both groups better-off. Experience would teach the lending banks what kind of collateral policy to adopt, what information they needed from prospective borrowers, and so on.
However, a bank could borrow only if others were ready to lend to it. This is an important qualification, because it means that the banking system as a whole might not be able to obtain the reserves needed from the liquidity market, even though any individual bank could. This constraint could pose a problem if an unexpectedly high demand for cash caused the short-term liquidity market to dry up temporarily as everyone demanded more reserves and no one was willing to supply them. In principle, this could cause the banking system to collapse. Since the danger is caused by the banks' committment to redeem their notes on demand without notice, the banks might try to avoid it by modifying the convertibility contract on their notes. Instead of guarunteeing to redeem their notes for specie on demand, the banks could reserve the right to defer redemption for some pre-specified period on the condition that noteholders would be paid pre-specified compensation when the notes were finally redeemed. In other words, the bankers could insert clauses into the convertibility contract that would give them the option of deferring redemption.
These 'option clauses' would need to be carefully designed. To remain in business, a bank that introduced option clauses would need to reassure the public that its notes were still safe. It would want to make a credible promise that it would use the option only in exceptional circumstances and that noteholders would suffer no losses even then. To be convincing the bank would need to set the compensation paid to noteholders at a level high enough for it never to be in the bank's interest to exercise the option except in an emergency. The bank might also stress the advantages of the option clause to the risk-averse noteholder. Those noteholders slow to react to a run on specie would lose little or nothing by their failure to be first in line, and indeed would gain from the compensation the bank would have to pay for suspending convertibility. And even if the bank should turn out to be insolvent as well as illiquid, then losses would be shared on a pro rata basis among noteholders and other creditors rather than falling disproportionately on those who were not quick enough to demand redemption before the bank suspended. It would thus be clear to the public that the bank would have recourse to the option only as a last resort, and that the option, if anything, would probably make individuals better off.
Trial and error in the market place would determine the period over which redemption could be deferred and the interest to be paid on notes whose redemption had been suspended. The exact form of option clause is therefore difficult to predict beforehand. The compensatory interest rate would presumably be linked to the interest rate of the short-term liquidity market. A plausible formula would be 'x points above the average rate prevailing in the short-term market over the past y months'. With such a formula, the option would never be exercised in 'normal' times because a bank could always obtain liquidity more cheaply in the short-term market. If a liquidity crisis were to develop, however, the short-term interest rate would rise sharply, and once it rose beyond a certain threshold level it would be cheaper for the bank to obtain - or, strictly speaking, to retain - liquidity by invoking its option of deferring payment. For simplicity we assume that all banks face the same threshold level, leading all banks to invoke their options simultaneously.
This would set in motion a chain of events that would break the crisis, send interest rates downward again and alleviate the shortage of liquidity. In the period immediately after the banks started to exercise their options, market interest rates would remain above the penalty rates the banks were paying through the use of their options, making it worthwhile for banks to borrow by invoking their options and to lend on the short-term liquidity market. The banks would thereby channel liquidity to where the demand for it was greatest, thus beginning to alleviate the shortage of liquidity and causing market interest rates to begin to fall. The banks would continue these arbitrage activities until the market interest rate had come down to the penalty level. Once it reached that level the banks would no longer derive any benefit from exercising their option to defer redemption, and they would be ready to resume redemption on demand. By the time interest rates had fallen to that level, the public's panic over liquidity would have abated. As the demand for liquidity continued to fall interest rates would return to pre-panic levels. The introduction of option clauses would thus protect the liquidity of the banking system and break the panic. The knowledge of this would itself make the banking system considerably more stable by eliminating the possibility of a bank run starting because of the public's self-fulfilling expectations of a run.
This completes our discussion of the evolution of a laissez-faire banking system. Note, in particular, how institutions like convertibility, a clearing system, a market for short-term liquidity and option clauses would develop and protect the banking system against shocks. The sole driving force behind these stabilizing mechanisms would be individuals' self-interested attempts to protect themselves against adverse conditions. A free market monetary system would thus be highly stable.
Back to the Free Banking book page.