Commercial motives may be regarded as sufficient to establish the general principle that, under ordinary circumstances, business men will use those forms of currency which are most convenient, economical, and safe, but some exceptions must be noted. Custom has a good deal of influence in this connection and may for a long time maintain in circulation forms of money which are in reality less convenient and economical than others which might be employed. In England, France, and Germany, for example, gold coins are used to a much greater extent than in the United States. This is due chiefly to the fact that bank-notes of low denominations are not permitted in those countries, but in part to the acquisition of the habit of using coin instead of paper in ordinary retail trade. Legislation may favour one form of currency and forbid another, as in the cases of banknotes of low denominations in the countries just mentioned and of the silver certificates and treasury notes of the United States. The influences of custom and law, however, are most likely to show themselves in the retardation of the development of banking institutions rather than in the limitation of the circulation of the forms of currency they are permitted to issue.

The only real danger connected with the normal expansion of bank credit comes from commercial crises. These calamities can rarely be foreseen, though they occur with a considerable degree of regularity. They are characterized by a general shattering of credit and a tendency to demand payments in cash and to hoard gold. The result is a rapid diminution of cash deposits on the one hand and of the use of bank currency on the other, the failure of business men to meet their notes when they fall due, and an increase in the demands made upon the banks for legal-tender money. Provision* against such exigencies can, however, be made, which, while not adequate to entirely remove the danger, may reduce it to a minimum.

Up to this point we have been attempting to show that a normal limit to the expansion of bank credit is set by the amount of mercantile securities which the ordinary processes of commerce under the forms of the credit system bring into existence. We must now explain how. it may be stretched far beyond this limit by dishonest or unwise practices.

A considerable demand for loans comes from men who desire to undertake enterprises of uncertain utility and success and sometimes of a purely fraudulent nature; and if banks accept the paper of such men, they are apt to come to ruin, or at least to suffer great loss. The failure of such enterprises usually renders impossible the payment of the discounted notes at maturity, and leaves the banks liable for the redemption of the currency they have issued without the usual means of accomplishing it Such events represent only half-completed exchanges, the first transfer having been made and the credit granted without the completion of the exchange by the counter-transfer. So far as the banks are concerned, the results are similar to those which follow commercial crises, the difference consisting in the nature of the event which caused the interruption of the commercial process. It should be observed that enterprises of this character are very apt to be financed by borrowed capital, the promoters themselves being equipped with enthusiasm, optimism, plausibility, and shrewdness rather than money. As the chief dispensers of loanable capital, therefore, the officials of banks are in a position either to render great service to the community by refusing to discount the paper of such people, or to do great harm by furnishing them the funds they desire. So far as they yield to the temptation of investments of this character, they are expanding their credit beyond legitimate and safe limits. Many varieties of perfectly sound securities are not suitable means for the expansion of bank credit. To this category belong bonds, stocks, and mortgages, and indeed most securities which represent fixed rather than circulating capital. Such credit instruments usually mature only after very long periods of time and depend for their value upon the profits of business enterprises rather than upon the successful completion of ordinary commercial processes. They do not, as it were, automatically turn themselves into cash every two or three months, and thus furnish to the bank the means of redeeming or retiring the currency issued at the time of their purchase. Take as an illustration the stock certificates of a manufacturing establishment, which represent the capital invested in the enterprise. They give to their holders the right to participate in the profits of the concern, and have a market price equal to the capitalized value of the actual or estimated dividends. Their life is as long as the factory, and they are transmutable into cash only by the dissolution of the business or by sale upon the stock-markets. In either case the process of transmutation is forced and not automatic, and does not take place with sufficient frequency or certainty of results to meet the needs of banks. As we shall show in the following chapter, banks usually possess certain funds which may, perhaps without great danger, be invested in this class of securities, and bank-notes are frequently issued against them exclusively, but there are few, if any, exceptions to the general principle that large investments of this sort involve an overstraining of bank credit.

* See Chapter IX (Bank Currency, Its Regulation And Safety), sec. 7.

In this connection a distinction between the real object of a bank's investment and the security which it demands and receives therefor must not be overlooked. Large quantities of the class of securities we are now discussing are often deposited with banks as security for the payment of discounted notes. They are not in this case the property of the banks, and are available for use only in case of the non-payment of the discounted notes at maturity. Their presence in the bank's possession under these circumstances is an evidence of careful management and in no way constitutes an exception to the above mentioned principle.

The expansibility of bank credit and the impossibility of setting exact limits to it, together with the quantity theory of money and certain facts whose explanation is not apparent to the superficial observer, are responsible for a theory of bank currency an account of which may throw some light upon the subject we are discussing.