¹ Throughout this paper it is assumed that the community's real income generally moves in the same direction as its money income. This, of course, is a “short run” tendency; and even in the short run there are exceptions to the rule.

² Saving is here denned as the accumulation of titles to wealth or future income. Spending on home-produced products thus includes the purchase out of income of both consumable and capital goods. A wage earner buying bread, a farmer purchasing machinery, a corporation erecting new plant out of current profits, all these are spending. Saving is thus, according to our definition, always associated with placing or holding funds at the disposal of the financial institutions or the capital market. Conversely, borrowing and loan-expenditure is conceived as the expenditure of money obtained from financial institutions or through the market; whether that money is devoted to the extension of capital equipment by business men or the purchase by consumers of automobiles and radios on time-payment. While this definition has certain disadvantages they are here outweighed by advantages. When speaking of a country's expenditure upon imports, it is inconvenient and unnecessary from our standpoint to distinguish between capital and consumable goods; a distinction which would be necessary if we were to follow the normal custom of including in “investment” all additions to capital goods. Moreover, from the point of view of the flow of money, which is our present concern, it is largely immaterial whether people are spending upon one sort of home-produced goods or another, a block of iron or a block of ice; as long as they are spending the flow of income is maintained or augmented.

³ Paish, F. W., “Banking Policy and the Balance of International Payments” (Economica, new series, vol. III, 1936, pp. 404–11).CrossRefGoogle Scholar Using alternative terminology, introduced to economists by Messrs Hicks and Allen, we might speak of the “income-elasticity” of imported goods in general.

⁴ The lag here discussed is simply the period between changed incomes and changed expenditures. It must not be confused with the lag which exists between changed incomes and the full effects of changed outlays including all their repercussions. The latter are discussed below.

It must be added that the concept of income is not clear or unambiguous, especially in relation to wealthy individuals, entrepreneurs, and corporations. Arbitrary (or conventional) allowances have to be made for capital appreciation or depreciation, obsolescence, and so forth. Into this whole matter we cannot enter here. A helpful discussion may be found in Hicks, J. R., Value and Capital (Oxford, 1939), chap. xiv.Google Scholar It will be clear, however, that we are referring only to the “final” incomes received. The income of the retailer, for instance, is not the aggregate which he receives from his sales, but only what is left over after paying for his supplies and for his employees and after making such other payments and allowances as his use of land and capital equipment entail.

⁵ Monthly Review of the Bank of Nova Scotia, Dec., 1935. See also Nov., 1935, and May, 1937.

⁶ Ibid., Sept., 1936.

⁷ See Giblin, L. F., Australia, 1930, an Inaugural Lecture (Melbourne University Press, 1930), pp. 10–12.Google Scholar Also Clark, Colin, “Determination of the Multiplier from National Income Statistics” (Economic Journal, vol. XLVIII, 09, 1938, pp. 435 ff).CrossRefGoogle Scholar And especially Clark, Colin and Crawford, J. G., The National Income of Australia (Sydney, 1938), chap. xi.Google Scholar

⁸ Clark, Colin, “Determination of the Multiplier,” p. 438.Google Scholar One would have expected a change in the tariff to have been treated as a change in the propensities to import and to consume rather than as a change in the determinants.

⁹ Mr. Clark was not the first to leave his readers somewhat uncertain whether the multiplier was a figure which related a given additional injection (a) to the total volume of incomes to which the addition gave rise immediately and subsequently, or (b) to the increase in the volume of incomes which existed during the period occupied by the injection. Mr. Keynes, who is chiefly responsible for popularizing the concept, left many of his readers in the same quandary by his exposition in The General Theory of Employment, Interest and Money (New York, 1936). See pages 113–15Google Scholar where the multiplier first appears to be the sum of an infinite series and causal connections are under discussion; then compare this with pages 122-4 and 403 where these causal connections seem to be rejected and the multiplier is moulded to fit whatever events do in fact occur simultaneously. Mr. Keynes has, I think, defended the logic if not the utility of his latter position: for he makes it quite clear that in periods when a new injection is taking place the multiplier which relates this injection to the contemporary increase of total income will, in the absence of adequate activity based upon accurate foresight, be distorted by two factors: first, the (net) additional injection may not really be of the magnitude which appears at first sight (because of the “induced” investment or disinvestment which is described below), and second, the marginal propensity to consume and thus the multiplier may temporarily diverge widely from its normal position. Mr. Clark lumps all the possible “determinants” together without attempting to ascribe causal priority to (say) exports and causal inferiority to (say) budgetary deficits or even to changes in working capital. Formally, therefore, his computations may be correct; but the theory of the multiplier, in elucidating the processes of cause and effect, has lost much of its value. Indeed, as Professor Mackintosh has suggested, if the “multiplier” is simply a mathematical relationship existing between total determinants (autonomous and induced) and total income, it were better described as a “quotient” and thus shorn of its pedagogic pretentions.

¹⁰ The first distribution of increased income may differ very markedly from the average. As the money passes on from the first income recipient to the second and third and so forth, it becomes diffused throughout a variety of industries and its re-distribution, at each step, will more nearly conform to the average. However, if the normal propensity to consume home products is about one-half (as seems probable) each item in the series (1 + ½ + ¼ …) is equal to the sum of all that follow: so that the first distribution is of great importance.

¹¹ Saving, as here defined, does not include accumulation of physical equipment, but only the accumulation of securities, etc.

¹² Of course there were many other factors at work to produce the different behaviour of the economy in the two periods. Part of the construction activity in the earlier period must be attributed to deficiencies caused in the previous decade by the war; also to a rapid trend towards urbanization. See, however, MacGregor, D. C., Memorandum 64 of the London and Cambridge Economic Service, 02, 1937, p. 23.Google Scholar Also “The Role of Mining in Canadian Business” (Monthly Review of the Bank of Nova Scotia, 07, 1936).Google Scholar Also the Annual Report of the Governor of the Bank of Canada, 1939. Also Hasler, W. J., “The Boom and Depression in Canada, 1924-1935” (thesis submitted for the degree of M.Sc. (Econ.), University of London, 05, 1936), chap, viGoogle Scholar; especially the section which explains why the economy reacts far more spectacularly to changes in wheat exports than to pulp and paper exports.

¹³ By borrowing we here mean all methods of obtaining money from a capital market or financial institutions; whether by the issue of bills of exchange, bonds, debentures, mortgages, and other evidences of debt or by the issue of common and preferred shares. Unfortunately the language does not contain a word covering all such forms of money-raising.

¹⁴ Since all this was written the Economic Journal for 06 1939 (vol. XLIX, no. 194)Google Scholar has appeared. Readers may be interested to compare what has been said here with the criticisms of Clark's work advanced by Roy W. Jastram and E. S. Shaw (pp. 358-65); and also with the refinements suggested by D. H. Robertson (pp. 354-6) in regard to the various possible outlays of income, etc.