¹ It is convenient to adopt the solecism of Mr. Keynes and to use the term “classical” to denote the writings of the followers of Ricardo, as well as those of his forerunners.

² In this paper the logic of the classical method is nowhere in dispute.

³ In discussing “applied” economics, Pareto (and Barone) encountered difficulty in the use of the general equilibrium method and were compelled frequently to neglect the interdependence of economic phenomena and to resort to the particularist method.

⁴ See ProfessorPigou's, insistence on the small industry (Economic Journal, 06, 1927).Google Scholar

⁵ Thus in Marshallian language, the change in output in the large industry will affect the producers' surpluses in the small industries; and the total effect requires that a balance be struck between these. And, of course, the total effect depends on how the funds so raised are used by the government: if the expenditure is of the sort that increases demand for output in the small industries, the change in the set-up becomes even more marked, i.e., the contraction of the large and the expansion of the small. And this effect will be still more evident if the government action happens in a depression, because as the large and small industries contract output in the downswing of the business cycle, the relatively greater release of factors by the large industry is likely to make the demands of the small industries more elastic, and the tax on the output of the large industry will reinforce this effect of the depression on the degree of monopsony.

⁶ The third determinant implies that the repercussion is not merely a matter of the elasticity of demand in the small industries and the elasticity of the supply in the large industry. These concepts were used by ProfessorPigou, (Industrial Fluctuations (New York, 1927), ch. v)Google Scholar to measure repercussions. But the question is concerned more with the changes in elasticity. For that type of investigation, account has to be taken of the amount of “slack” available in the interconnections between industries, e.g., the stocks held by the different industries, the amount of unused resources, etc., at the moment of tax impact. These conditions will greatly affect the short-period repercussions, and their velocity, and this in turn may have considerable effect on the path to the new equilibrium, following the tax. The question is deeply concerned, therefore, with the cyclical conditions at the moment of tax impact. In the upswing of the cycle, as the large industry increases output, its increase in demand for factors is proportionately greater than in the small industries (after the “slack” has been taken up), and the demand of the latter becomes more inelastic. A tax imposed on the large industry in those conditions would partly offset the increased monopsonist position of the large industry, to a greater or less degree according to the size of the tax, and the size of the large industry as compared with the small. In a downswing the opposite holds, and a tax on the large industry would reinforce the conditions imposed by the cycle (see footnote 5). From this point of view the repercussion of a tax would go further through the system in the downswing, and the sector for study would be correspondingly enlarged.

In a system which combines a few trustified industries with many small ones, the study of the tax repercussions through industries merits extensive inquiry, and it appears that at least two sets of cases have to be studied, given the size of the tax and the size of the industries: (1) the case where there is a “slack” available; (2) the case where there is a condition approaching equilibrium but the economic system is about to move up to boom or down to depression.

⁷ Equilibrium theory does study the effect of changes in data, e.g., the weapon of quasi-rents, as well as the new equilibrium necessitated by any such change. But the change in data themselves cannot be analysed by equilibrium theory, which presumes that these changes are induced by non-economic circumstances, and that they do not grow, therefore, from within the economic system. Changes in data are, therefore, given from the point of view of equilibrium theory.

⁸ McQueen, R., “Economic Aspects of Federalism” (Canadian Journal of Economics and Political Science, vol. I, 08, 1935, p. 352).CrossRefGoogle Scholar

⁹ Report of the Committee on National Debt and Taxation (H. M. Stationery Office, 1927, Cmd. 2800).

¹⁰ I.e., questions concerning theory of incidence and effects of income tax, of the incidence and effects of different kinds of taxes in the disequilibrium of perfect competition, and in the equilibrium and disequilibrium of imperfect competition, questions of incidence and effects of given taxes used in certain ways, questions of discontinuous adaptations through time.

¹¹ This greater variability in the agricultural regions is partly due to the greater inelasticity in the supply of labour in these industries, partly due to the inelasticity of supply in agricultural commodities (i.e., difficulties of closing-down, finding alternative employment for resources, etc.), and partly due to the fact that in the production of raw materials, competition is more perfect than in manufacturing production, generally speaking.

¹² A careful study of these conditions is necessary when analysing, for example, the taxable capacity of an economy. In an economy containing rigidities, and therefore sensitive to impacts, the actual taxable capacity probably varies considerably as between the upswing and the downswing of the usual business cycles. Also the incidence and the effects of taxes are likely to vary in the different phases of the cycles, not merely in degree but also in kind.

¹³ Eléments d'économie politique pure (Paris, 1926), pp. 65, 214 ff., 259 ff.Google Scholar

¹⁴ Mathematical Psychics (London, 1932), p. 17 Google Scholar, app. 4. Also Papers Relating to Political Economy (New York, 1925), vol. II, p. 311.Google Scholar

¹⁵ E.g., by de Marco, De Viti in First Principles of Public Finance (London, 1936).Google Scholar

¹⁶ The quantity of state services, e.g., defence, health, etc., is more or less fixed for each individual, although, of course, different citizens may acquire greater or less utility from the same service.

¹⁷ It is true that an individual may consume less whiskey in order to reduce his tax payments, but this sort of action is limited in extent. Anyway such action is regarded as perversive by the state, unless it happens to have fixed the tax rate with moral ends in view.

¹⁸ The term “limitational” was introduced by Ragnar Frisch—“Einige Punkte einer Preistheorie mit Boden und Arbeit als Productionsfaktoren” ( Zeitschrift für Nationalökonomie, Bd. III, 19331, p. 64).Google Scholar The above use of the term is not the same as in Professor Frisch's paper. He uses the term “limitational” to refer to factors of production which have to be combined in certain fixed proportions in order to produce certain quantities, and where no substitution of increments is therefore possible. Factors of production which are uniquely determined by technical considerations are “limitational” in his sense. In this paper, however, “limitational” is used to describe conditions which are determined by political considerations. The parallel between the two conditions is clear: in each instance the household or firm is confronted with a datum which limits free choice, and which prevents substitution of a little more of one thing for a little less of another.

¹⁹ This is not a necessary result of a socialist economy. Freedom of choice in consumption is compatible with such a system. But if the consumption is directed by the state, the limitational nature of the case is obvious.

²⁰ The connection between the growth of the state as employer, and the question of wage rates in the sheltered industries, especially in England, is not remote.

It may be noted that the growth of state unemployment insurance, health insurance, etc., tends to alter the functions of trade unions. They become increasingly interested in the wage rates and wages bills, and if they can and do fix a wage rate which is too high to give full employment to their members, the cost of relief lies on the taxpayer, and not on the funds of the union.

²¹ Of course other conditions assisted this process. The possession of a skilled Civil Service, and the growth of a certain level of citizenship, are necessary for the successful operation of direct differential taxation. These were the means that made the change readily possible. By the time of the Asquith Budget of 1908, it was realized that “retrenchment was over and done with”, the tendency towards social services in both state and municipal finance being unmistakable.

²² Thus for Canada, given the institutional or legal definition of function between the Dominion and provincial governments—in which the provinces have jurisdiction over property, over education, welfare, and local works—it could be deduced theoretically that in the twentieth century there would be a relatively greater increase in provincial than in Dominion expenditures. This expectation was supported by fact, despite the effect of the War on the Dominion expenditures.

If the above thesis holds, provincial expenditures are likely to continue to increase, relatively to Dominion, other things remaining constant. This raises two issues in the discussion of Dominion-provincial relations. (1) The social services are national, not local questions. The needs are national, and probably some efficiency is to be gained by having a national policy in their fulfilment. Of course the administration need not be national. (2) The funds to meet these are essentially a question of national revenues. Thus the trend of the tax system has to be considered, and the probable future emphasis on direct differential taxation in the newer countries has to be weighed.

The long trend in the content of expenditures and taxes, and the variability of relative expenditures and relative taxes in the course of business fluctuations, are the economic factors that determine any delineation of Dominion-provincial relations.

²³ Economists are particularly indebted to Mr. Keynes for his general theory of employment and interest, especially in its emphasis that the money rate of interest is not governed by the marginal efficiency of capital in the way the classical school maintains. This idea was common among Mercantilists, and was not unknown to Adam Smith. They, living in an age in which government expenditures on war and other unproductive enterprises were highly significant, were impressed by the burden of internal debt and the consequential transfers from those interested in development, to creditors. Smith, as if replying to Ricardo and Professor Pigou, says explicitly that interest on internal debt is not a mere payment from the right hand to the left: he regards the debt as a factor which raises the money rate of interest and which is likely, therefore, to hinder the inducement to invest. This idea was lost in the nineteenth century when the dynamics of the capitalist system and the inducements to invest made internal debt and the whole sector of public finance relatively unimportant. But in the post-war period, the burden of the debt and the enlargement of public finance create again conditions analogous to those of the eighteenth century. Mr. Keynes's discussions of mercantilism are particularly interesting from this point of view.