¹ Johnson, Harry G., The Canadian Quandary: Economic Problems and Policy (Toronto, McGraw-Hill, 1963), pp. xx, 352, $5.50.Google Scholar

² Part 3, “Monetary Policy.”

³ According to Johnson this “expectation of forthcoming reserves” is “derived partly from reserves actually provided in the current month and partly from an accurate prediction of reserves actually provided in the subsequent month, the adjustment occurring with a distributed lag” (p. 179).

⁴ This might seem a quibble if changes in cash-adjustment assets were immediately reflected in deposits; but they may not be, and certainly not in some “desired level” such as that defined by Johnson. Deposits are subject to many short-run influences outside the immediate control of the banks. Within any short period of time, deposits may be changing rapidly, up or down, as a result of net loan extensions or repayments which may occur in a branch banking system regardless of the immediate response of a bank's head office to its changing cash position. Dependence on deposit data may create spurious lags, or distort true lags, which do not, therefore, reflect at all accurately the response of the banks to changes in their cash. For a brief but clear account of the cash management problem of the chartered banks see Canadian Bankers' Association, Submission to the Royal Commission on Banking and Finance (Toronto: the Association), 07, 1962 Google Scholar (reproduced in Supplement to the Canadian Banker, Spring, 1963), paragraphs 41–6.Google Scholar A full theoretical analysis of monetary control in terms of changes in excess cash and corresponding changes in cash-adjustment assets is presented by Galbraith, J. A., The Economics of Banking Operations: A Canadian Study (Montreal, 1963), 427–38.Google Scholar A brief summary of recent literature relating changes in bank cash (or “reserves”) to changes in cash-adjustment assets appears in Johnson, Harry G., “Monetary Theory and Policy,” American Economic Review, LII, 06, 1962, 358–60, 364-5.Google Scholar

⁵ Perhaps this should not be called a lag at all; the initial response is immediate, especially if the amount of excess cash (or the deficiency) is too large to be worked out in the money market at short notice on some future occasion. Nevertheless, the necessary spreading out of the operation may appear statistically as a lag in response to the large initial change in cash. Indeed the answer to apparent slippages due to averaging would seem to be not to blame existing reserve-formula arrangements, but to cause larger cash changes and thus to overcome any tendency to use the averaging provisions to delay response. However, the Bank of Canada seems to take a different view (cf. Bank of Canada, ubmissions by the Bank of Canada to the Royal Commission on Banking and Finance (Ottawa: the Bank, 05, 1962), no. III, p. 34, paragraph 46.Google Scholar)

⁶ Uncertainty about monetary policy and perverse expectations regarding interest rates could presumably be overcome by an appropriate signal of central bank intentions. This is not discussed by Johnson, but it was emphasized in the Submission by the Canadian Bankers' Association (paragraphs 478–93), and was apparently a matter of great interest to the Commission.

⁷ It must of course be remembered that, under the peculiar Canadian system of reserve requirements (which Johnson recognizes and describes, p. 179), the cash ratio computed on a statutory basis cannot tell us anything about the response of the banks to a change in cash. The statutory deposits used in computing the ratio are defined as “the average of such deposit liabilities at the close of business on Wednesdays in each of the four consecutive weeks ending with the last Wednesday but one of the preceding month.” (Bank Act, sec. 71(2)). Since statutory till money is similarly defined, the only variable in the ratio is the level of chartered bank deposits with the Bank of Canada during the current month. This means that excess cash for all banks together cannot be increased or diminished without action by the central bank. The response of individual banks may be immediate, but the system of chartered banks as a whole is powerless to create or destroy excess cash (“Law of the Conservation of Excess Cash”?) and so cannot affect the over-all cash ratio in any way. It follows, therefore, that when the central bank initiates changes in over-all excess cash, what may seem to be an “outside lag” may in fact be an “inside lag”; viz., a lagged response by the monetary authorities to the existence of an excess (or deficiency) of cash in the banking system.

⁸ It would be especially unfortunate if Johnson's lag of “half a month” should be hastily interpreted to mean that, under our special system of reserve requirements, the lag could be eliminated by shortening the reserve period from one month to two weeks. This would have no effect on the true lags in the system and would only increase the daily average cushion of cash (not, under the circumstances, excess cash) which the banks have to maintain over requirements in order to meet unexpected swings at the end of the reserve period. In the absence of a “federal funds” market, a loss of cash cannot be offset on the same day that cash-adjustment assets are reduced. Banks can get cash on the same day by borrowing from the Bank of Canada; but this is objectionable for more reasons than Johnson's observation that there is a “strong convention” against it. It means not only that a bank may appear to be running its cash in a rather sloppy manner, but, far more important, it means that the bank must cover a temporary shortage (perhaps of one day only) by borrowing for a minimum period of one week at a penalty rate (Bank rate). Now, under any averaging system, the size of this cash cushion for the averaging period as a whole will be a constant; hence the shorter the averaging period the larger the daily average value of this “cash cushion” must be. At the limit, a one-day “averaging” period (the Canadian system up to 1954), Canadian banks actually carried twice the legal reserve requirement; i.e., 10 per cent instead of the pre-1954 requirement of 5 per cent. This was not excess cash but a cushion against daily swings in cash which could not be corrected by averaging within the reserve period.

⁹ It is possible that Johnson's lag is a measure of the time it takes for the “dynamic” (or period-analysis) deposit multiplier to work itself out (to a “reasonable” proportion of its value at infinity). This is a possible interpretation of his description of what he is trying to measure (quoted above p. 259). If so, it is not a measure of bank response to changes in cash, but may nevertheless be a valuable contribution to banking theory and to a general “applied” theory of the multiplier. This, too, deserves further exploration.

¹⁰ Mainly small business, i.e., borrowers in the loan category up to, say $100,000. In practice, banks try to soften the impact of tight money on this type of borrower.

¹¹ The use of the term “expansion multiplier” may be misleading: there is only a multiplier which may lead equally well to deposit expansion or contraction, depending on the sign of the multiplicand.

¹² I am ignoring in the text any constitutional problems involved in shifting the power to initiate taxation from the Minister of Finance to the Bank of Canada.

¹³ There is a special case in which Johnson's “plausible” assumption is true; viz., if the legal requirement is below the actual cash requirements of the banks, and both requirements are computed on an average basis. In this case the higher the legal ratio (up to the point where it coincides with actual cash requirements) the closer the banks will work to it. But this of course no longer applies once the legal ratio exceeds the actual cash requirements of the banks; in other words, the banks work as closely as possible to the higher of (1) their actual cash requirements or (2) the legal reserve ratio. Once the legal ratio equals or exceeds actual cash requirements a higher reserve ratio will in no way induce a closer approach to the legal ratio and in no case (not even in the special case above) will a higher legal ratio in any way improve the central bank's control over chartered bank deposits.

¹⁴ See Marsh, D. B., “Keeping the Dollar at Fixed Parity May Be Costly”, Financial Post, 06 24, 1961, 35.Google Scholar

¹⁵ This has implications for Johnson's alleged outside lag within the chartered banking system (see above pp. 259–60). The way to minimize lags between changes in cash and corresponding changes in cash adjustment assets, and hence in turn lags between changes in cash and changes in chartered bank deposits, is to eliminate legal reserve requirements altogether. Banks would still have a minimum requirement for clearing cash and till money to which they would work as closely as possible, and bank cash would still be under the complete control of the Bank of Canada. (See also: Canadian Bankers' Association, Submission, paragraph 468 Google Scholar; McLaughlin, W. Earle, Submission to the Royal Commission on Banking and Finance (Montreal: The Royal Bank of Canada, 07, 1962), paragraphs 23–4Google Scholar; Gurley, J. G. and Shaw, E. S., Money in a Theory of Finance (Washington, DC, 1960), 262–4, 271 Google Scholar; and Galbraith, , Economics of Banking Operations, 433, n.136.)Google Scholar In the absence of legal reserve requirements there would be no “averaging” lags, since there would be no “reserve period” over which averaging could take place. The only lags left would be due to the unprofitability of short turn-arounds in cash-adjustment assets, and possible delays in adjusting excess cash, due to uncertainty or to market imperfections (see above p. 259). Since banks would be working closely towards their own minimum cash requirements, any cash deficiency would have to be made up at once to guarantee sufficient cash to meet their clearings. It is hard, therefore, to understand the reasoning behind the position taken by the Bank of Canada (Submissions, no. III p. 34, paragraph 45) that responses would be unduly delayed if there were no legal cash-reserve requirements. This might apply to a non-profit institution such as a central bank, but not surely to profit-maximizing institutions such as the chartered banks.