The relation between the supply of money and the general level of prices has been the subject of some conflict of opinion, and it will be well to examine somewhat closely the nature of the problem involved. Each operation of purchase or sale is not merely the acquisition, or the offer, of goods for money, but also the offer, or acquisition, of money in exchange for goods. If prices generally are on a high level the money obtained by the sellers of goods is correspondingly great in quantity, while if they are on a low level it will be in corresponding degree small in quantity. It is, therefore, obviously true that, in exchanges involving the actual passage of money from the buyer to the seller of goods, the quantity of money used will vary with the general level of prices. Now we have seen that in large groups of transactions in the modern business world it is not necessary that actual money should pass from buyer to seller of commodities, while, so far as concerns transactions of purchase where actual money does not pass, the level of prices appears, at any rate at first sight, to have no relation with the quantity of money in use. Though actual money does not pass in such transactions, a document conferring a claim to money does, in general, pass, and the form which such documents take is of importance in relation to the problem before us. They are of two main varieties, cheques and bills of exchange, which may be distinguished although the cheque is itself technically a bill of exchange. What we know as a cheque is an instruction to a banker by one of his clients to pay a specified sum of money to a named person, who may, if he wishes, collect the amount through 169an agent. The right to draw cheques implies the possession of a credit with the bankers upon whom the cheques are drawn. That credit may be based on the actual entrusting by the customer to the banker of money or immediate claims to money. It may, however, take its origin in an arrangement in virtue of which the customer acquires the right to draw upon the banker, within agreed limits, without the actual deposit of funds beforehand. The extent of the cheques which can be drawn is related to, and limited by, the extent of the credit accounts between customers and bankers. Though such accounts can be increased, as already remarked, without the actual deposit of money, the banker, in determining the aggregate amount of the advances which he accords to his customers, must have regard to his capacity to meet the resultant demands without default. His capacity to do so depends primarily upon the amount of actual money which he holds and the amount which, in the natural course of business, will accrue to his credit from day to day. The extent of his advances to customers is, accordingly, conditioned by the state of his reserves, and, taking the commercial community as a whole, it is true that the extent of the credit accommodation which bankers can extend to their customers as a whole is limited by the reserves in the possession of the banking community as a whole. It may also be affected by the distribution of these reserves between different bankers. As the aggregate of cheques which can be drawn is limited by the credits with bankers, and these credits are not indefinitely expansible without relation to the money resources of bankers, it follows that purchases which can be effected by the use of cheques without the actual handling of money in the narrower sense are limited in the last resort by the money supply. If this proposition be difficult to realise on account of the involved relations which have to be taken into account, it is possible to consider the problem from another point of view, namely, that the transactions of purchase and sale may be divided into two classes which are, in the main, clearly separable from one another. On the one hand, there are sales and purchases in connection with which actual money passes 170at each transaction; and, on the other hand, there are sales and purchases in connection with which documents representing claims to money suffice to effect the exchange. Into the former class fall the larger part of retail sales and purchases, and the latter class is mainly made up of wholesale transactions. Unless a change in the level of prices, or a change in the general money-supply, involves the transfer of transactions from the class in which money serves as the actual intermediary in the exchange to the other, the relations between money-supply and price-level can be considered in connection with the former class only of transactions. This course is justified by the further consideration that, though we may treat these two classes of transactions as separate from one another, the nominal price-levels expressing the rates of exchange of commodities in these two classes are not independent. It is not, in practice, except for comparatively short periods of time, possible for important changes to occur in the price-levels of one of the groups without the occurrence of corresponding changes in the other. For an examination of the problem before us we do not need to determine whether such changes take their origin in the one group or the other. It is sufficient that, in whichever group they arise, they are propagated into the other group. If this can be granted it will follow that the possibility of making provision for indefinite expansion of prices in the group of exchanges in which actual money does not pass will not justify the assertion that the movements of money-supply and prices are independent unless that independence can be established for the other group. As it has already been pointed out that in the group of transactions in which actual money serves as intermediary in exchanges a rising level of price implies a larger supply of money, and vice versa, unless the volume of transactions undergoes compensatory changes, a decreasing volume of transactions with an increasing price-level may be consistent with an unchanged money-supply, and an increased volume of exchanges with a lowered price-level may also be effected without change in the money-supply, while, as observed earlier, changes in the practice of the community covered 171by the designation “rapidity of circulation” or “efficiency” of money may be equivalent to corresponding changes in money-supply.