¹ For a brief summary of post-war business cycles, see Moulton, H. G. and Schlotterbeck, K. T., Collapse or Boom at the End of the War (Brookings Institution, 1942).Google Scholar On the general nature of war and post-war cycles, see Mitchell, W. C., Wartime Prosperity and the Future (Occasional Paper no. 9, National Bureau of Economic Research, New York, 1943).Google Scholar British and American experience after the last war is outlined in Economic Fluctuations in the United States and the United Kingdom, 1918-22 (League of Nations, 1942). A somewhat more detailed account of the American post-war inflation and deflation is presented by Payne, Wilson F., Business Behaviour, 1919-22 (Chicago, 1942).Google Scholar See also Wilson, Thomas, Fluctuations in Income and Employment (London, 1942).Google Scholar The Canadian post-war cycle is analysed by the author in a memorandum presented to the Advisory Committee on Reconstruction under the title, “The War and Post-war Cycle in Canada, 1913-23.”

² The financial effects of Canada's war effort are presented in some detail in my essay, Canada's Financial System at War (National Bureau of Economic Research, 1943).

³ Official figures of war investment were published in the Financial Post, April 24, 1943. Figures of net new investment from 1920 to 1930 are presented in the Report of the Royal Commission on Dominion-Provincial Relations (3 vols., Ottawa, 1940), vol. I, p. 116.Google Scholar

⁴ If the tax system were “neutral” with respect to debt distribution, so that everyone's tax payments were just offset by interest received on government obligations, the last argument against unlimited increases in national debt would disappear. Unfortunately, other considerations may prevent the use of such a tax system.

⁵ Somers, Harold M., “The Impact of Fiscal Policy on National Income” (this Journal, 08, 1942, pp. 364–85).Google Scholar

⁶ Since such prominent “Keynesians” as Professor Hicks and Mr. Kaldor have now confessed their liking for the Robertsonian definitions, it is perhaps no longer necessary to apologize for using them in preference to the Keynesian definitions ( Hicks, J. R., “The Monetary Theory of D. H. Robertson,” Economica, 02., 1942 Google Scholar, and Kaldor, N., “A Model of the Trade Cycle,” Economic Journal, 03, 1940).CrossRefGoogle Scholar However, it might be worth mentioning that in matters of tax policy the “Keynesian” definitions become inconvenient, since the discontinuity of income payments and lags between government outlays and tax collections become particularly important.

Dr. Kalecki has approached the problem of tax policy in depression through “stream” equations of the Keynesian sort (“The General Theory of Commodity, Income, and Capital Taxation,” Economic Journal, Sept., 1937). His general conclusion was that “capital taxation is perhaps the best way to stimulate business and reduce unemployment.” Fundamentally, his argument is that since a capital tax does not enter into prime cost or reduce the return to investment, a tax on all assets including cash balances will not discourage investment, while expenditure of the proceeds will encourage it. However, he throws up a fearsome barricade of assumptions to get his results, and his conclusions are less secure when the barricade is removed. His final conclusion would apply equally to a tax on hoarding. Perhaps most important, his failure to take account of lags between collection and expenditure of government revenue impairs the usefulness of his analysis.

Carsten Welinder uses a period analysis quite similar to the one presented here (“Grundzüge einer dynamische Inzidenztheorie,” Weltwirtschaftliches Archiv, Jan., 1940). There are, however, other limitations in his analysis. First, for some strange reason, he neglects the effects on investment of government expenditure. Second, he fails to show the importance of the form of government expenditure, particularly whether the proceeds enter directly into profits or go first into wages. Third, the possibility that expenditure may precede collection of the tax even with the budget balanced over the fiscal year is not considered. Fourth, the lag between collection and expenditure receives insufficient attention. Fifth, his definition of profits is an accounting rather than an income concept, and it is questionable whether profits in his sense determine entrepreneurial decisions in all cases. His use of “general elasticity of demand” to mean the total effect on consumption of a change in taxes is confusing, and his restatement of principles in terms of this “elasticity” is tautological and redundant.

⁷ This equation illustrates the cumulative nature of inflation. Any uncompensated rise in P ₁ over P ₀ will lead to a rise in profits over “day 1” and so to increased investment in “day 2,” which again raises prices in “day 3,” and so forth.

⁸ Speaking generally, after the last war expenditures were cut before taxes in England, and taxes were cut before expenditures in Canada and in the United States. In the six months after the Armistice, government deficits decreased in Britain and increased in Canada and the United States.

⁹ Since S ₁ = Y ₀ – (C ₁ + T ₁), and C ₁ = C ₀, S ₁, – S ₀ = T ₀ – T ₁ = I ₁, – I ₀. Y ₁ – Y ₀ = T ₀ – T ₁, and not, as might seem at first glance, (T ₀ – T ₁) + (I ₁ – I ₀).

¹⁰ Symmetrical results are obtained for tax increases. A rise in taxation followed by an equivalent rise in expenditures, whether or not anticipated, may lead to an upswing, and is more likely to do so if the increase in expenditures is felt by entrepreneurs in the second day than in the third. When the increase in expenditure is not foreseen and accordingly capitalists do not maintain their outlays in the first day by borrowing or dishoarding, an initial downswing starts, but a reverse tendency sets in when the expenditures are felt.

¹¹ This Journal, 08., 1943, pp. 425–7.Google Scholar

¹² The incidence of payroll taxes is analysed in Dunlop, J. and Higgins, B. H., “Bargaining Power and Market Structures” (Journal of Political Economy, 02., 1942, pp. 23–4Google Scholar).

¹³ Ibid.

¹⁴ Since the tenor of my argument must already have suggested to the reader that in my mind “tax policy” and “expenditure policy” are really inseparable, a word specifically on expenditure policy may be permitted. As we saw in Part I, a subsidy paid to employers equal to marginal wages bill minus marginal productivity of labour for any desired level of employment in any industry would tend to induce employers to hire the desired number of workers in each industry. Selective subsidies of this sort would serve not only to maintain the level of employment and income, but also to allocate labour in an approved manner. There is, therefore, something to be said for such selective subsidies in preference to general unemployment benefits when depression threatens.

¹⁵ Brown, H. G., “The Incidence of a General Output or a General Sales Tax” (Journal of Political Economy, 04, 1939)CrossRefGoogle Scholar and “Correction” (ibid., June, 1939).

¹⁶ “A Diagrammatic Analysis of the Supply of Loan Funds” (Econometrica, July-Oct., 1941).

¹⁷ Black, Duncan, The Incidence of Income Taxes (London, 1939), chap. XI and pp. 220-2, 300–12.Google Scholar

¹⁸ In a frictionless market, the change in relative rates of return would cause marginal shifts throughout the whole range of assets, including some shift from money to other assets which are “safe” when price stability is expected. If there is uncertainty about price trends, so that holding cash is regarded as a “risky” investment, the reduction in income tax may conceivably lead to increased cash holdings. However, in terms of liquidity preference, the result is clear cut, and it seems unlikely that many people would increase their cash holdings because of a rise in “net” rate of return that could be made with a given drop in prices. In the highly institutionalized capital market that we have, such marginal shifts may not take place; but certain types of investor, who make “risky” investments or none at all, will be induced to hold less cash and more assets.

Readers who find this verbal argument unconvincing will find a complete exposition of the diagrammatic analysis upon which it is based in the article referred to above, “A Diagrammatic Analysis of the Supply of Loan Funds.”

¹⁹ See for example Colm, Gerhard and Lehmann, Fritz, Economic Consequences of Recent American Tax Policy, Supplement I of Social Research (New York, 1938), p. 93 Google Scholar; Hicks, Ursula, Finance of British Government, 1920-36 (London, 1938)Google Scholar; Colwyn Committee on National Debt and Taxation, Report (London, 1927), p. 82.Google Scholar Exceptions are Dalton, Hugh (Principles of Public Finance, London, 1936, p. 52)Google Scholar and Jensen, J. P. (Government Finance, New York, 1937, p. 408).Google Scholar

²⁰ Smith, Ricardo, and Bastable all argued that inheritance taxes are paid “out of capital.” Pigou's reply that one might as well say that “a tax on beer is necessarily paid out of beer” does not seem to meet the argument; a tax on beer may very well be paid out of beer ( Public Finance, London, 1928, p. 162 Google Scholar). The Colwyn Committee also denied that the tax leads to actual capital consumption, but both majority and minority agreed that “it is distinctly more damaging to savings than the income tax” (Report, pp. 189-90). Dalton (Principles of Public Finance) and Colm and Lehmann (Economic Consequences of Recent American Tax Policy) are of the reverse opinion. A. de Viti de Marco—a member of Italy's landed aristocracy—insisted that the “instinct” to ensure continuity of the family is the chief reason for saving ( First Principles of Public Finance, trans, by Marget, E. P., New York, 1936, p. 368 Google Scholar). From this premise he concludes that the tax reduces savings, but one could surely reach the opposite conclusion from the same premise.

²¹ Of 1,151 corporations answering the Brookings Institution survey, 668 said they had increased the percentage of earnings paid out in dividends, 296 said they had not, the other answers being indefinite. Of 857 replying, 636 said the dividends were paid at the expense of reserves for bad years, at least in part; of 848 replies, 604 said reserves for expansion were forgone; 558 out of 835 said they sacrificed working capital; only 132 out of 821 sacrificed funds used to reduce deficits, and only 229 out of 805 postponed repayment of obligations ( Kendrick, M. S., The Undistributed Profits Tax, Brookings Institution, 1937 Google Scholar). When allowance is made for the obvious efforts of business men to paint the tax black and their own practices white, it seems that the tax modified dividend policies considerably.

²² See for example, Despres, E., “The Proposal to Tax Hoarding” (American Economic Review, Supplement, 03, 1939).Google Scholar

²³ Dahlberg, Arthur, When Capital Goes on Strike (New York, 1938).Google Scholar

²⁴ Mr. Dahlberg considers 100 per cent reserves necessary to prevent shrinkage in the volume of deposits through bidding away of bank assets by customers anxious to avoid the hoarding tax. The 100 per cent reserve plan is a weighty matter in itself; my own views on it are expressed in “Comments on 100 Per Cent Money” (American Economic Review, March, 1941). By imposing lower rates of tax on money substitutes, the need for 100 per cent reserves is diminished.

²⁵ The failure of dated stamp scrip in Alberta is no proof that a hoarding tax would be unworkable. The Alberta plan was on too small a scale, the coverage was too limited, publicity was confused, there was no official assurance of permanency, the government did not accept the scrip for all taxes, redeemed scrip was not reissued, no provision was made for conversion of scrip into drafts or currency acceptable in other provinces or other countries—indeed, almost every conceivable mistake was made. Cf. Coe, V. F., “Dated Stamp Scrip in Alberta” (this Journal, vol. IV, 02, 1938, pp. 60–91 Google Scholar).

²⁶ See for example, Fisher, Irving, “Income in Theory and Income Taxation in Practice” (Econometrica, 01., 1937)Google Scholar; Friedman, Milton and Poole, Kenyon, on “The Spendings Tax” (American Economic Review, 03, 1943).Google Scholar