¹ It grew out of a larger study of the Canadian economy and its prospects sponsored by the Canadian Pacific Railway Company. All opinions expressed are those of the author. J. A. Sawyer and D. H. Steinthorsen of the Dominion Bureau of Statistics and J. A. Pincus of Harvard University have suggested many improvements.

² Dominion Bureau of Statistics, Reference Paper No. 72, The Inter-Industry Flow of Goods and Services, Canada, 1949 (Ottawa, 1956)Google Scholar, hereafter cited as The Inter-Industry Flow. J. A. Sawyer has developed some parts of this document more fully in “Some Aspects of the Structure of the Canadian Economy: An Application of the 1949 Input-Output Table,” presented at the Economics Seminar, Dominion Bureau of Statistics, Feb. 6, 1957 (mimeo.).

³ Goldberg, S. A., “Some Recent Developments in the Dominion Bureau of Statistics,” this Journal, XXI, no. 1, 02, 1955, 61 (italics in original).Google Scholar

⁴ Ibid.

⁵ For a clear introduction to inter-industry analysis, see Leontief, W. W. et al., Studies in the Structure of the American Economy (New York, 1954), chap. ii.Google Scholar The type of analysis performed in this paper has many precedents, but cf. especially Leontief, W. W., The Structure of the American Economy, 1919-1939 (2nd ed., New York, 1951), Part iv.Google Scholar Descriptions of many such uses of inter-industry analysis appear in Cameron, B., “The Future of Inter-Industry Analysis,” Economic Record, XXXI, no. 61, 11, 1955 Google Scholar, and W. D. Evans and M. Hoffenberg, “The Nature and Uses of Inter-Industry Relations Data and Methods” in National Bureau of Economic Research, Conference on Research in Income and Wealth, Studies in Income and Wealth, XVIII, Input-Output Analysis: An Appraisal (Princeton, N.J., 1955), 90–115.Google Scholar

⁶ In the following list, the number outside the parentheses refers to one of the sixteen sectors listed in Table I, and the numbers in parentheses indicate which sectors of the forty-two-sector table were included in that particular grouping: 1 (1); 2 (2); 3 (3); 4 (5); 5 (4, 6); 6 (7–17); 7 (19–22, 30); 8 (23, 24); 9 (26–29, 31); 10 (32); 11 (33–35); 12 (18, 25, 36); 13 (37); 14 (38); 15 (39–41); 16 (42).

⁷ “A Review of Input-Output Analysis” in N.B.E.R., Input-Output Analysis, 159–68.Google Scholar

⁸ Because of this possibility, it is very important when using an inter-industry table to know whether the base year contained any major peculiarities, such as relative price disturbances, recessions, capacity bottlenecks, import restrictions, etc. The last-mentioned did exist in Canada in 1949; see The Inter-Industry Flow, 23.

⁹ Leontief has, however, made progress at computing capital requirements per unit of output for American industries. See “Domestic Production and Foreign Trade; The American Capital Position Re-examined,” Proceedings of the American Philosophical Society, XCVII, no. 4, 09, 1953, Table 2.Google Scholar

¹⁰ “Post-war Economic Development and Policy in Canada,” this Journal, XX, no. 4, 11, 1954, 454.Google Scholar

¹¹ In The Inter-Industry Flow, inventory accumulation is kept separate from the rest of gross domestic investment. Whenever, in the following pages, a change in gross domestic investment is assumed, this means a change only in its non-inventory components. In short, when exogenous changes are traced through the system, it is always on the assumption that all flows are fully adjusted to the change, while all stocks remain constant.

¹² These elasticities were computed from data on family budgets given in D.B.S., Reference Paper No. 64, City Family Expenditure, 1953 (Ottawa, 1956)Google Scholar, App. B. Some guess-work was necessary in applying these, since the classes of expenditure used in this study are not at all identical with those used in the inter-industry table. Moreover, for two sectors—electrical apparatus and supplies, and imports (largely foreign travel)—no remotely applicable elasticities fell out of these computations. For these two sectors, elasticities were borrowed from Dewhurst, J. F. et al., America's Needs and Resources (New York, 1955), 1005, 1008.Google Scholar This eclectic procedure is not justifiable except on the ground that the changes in final demand are all largely illustrative approximations.

¹³ In 1949, governments purchased their own “production” of $1,279.3 million of “government service” and $282.0 million of community service from sector 16 of our condensed table, $8.0 million of communication from sector 15, and $0.5 million of transportation and $2.0 million of trade from sector 14 (figures furnished by Mr. Sawyer of D.B.S.).

¹⁴ An alternative approach to that taken here would be to trace all government expenditures through the inter-industry table to the point where they pass to some private agent, either a “primary factor of production” or a private business firm. This could not be done here without lengthy explanation and elaborate assumptions.

¹⁵ Gibson, J. D., “The Changing Influence of the United States on the Canadian Economy,” this Journal, XXII, no. 4, 11, 1956, 423.Google Scholar

¹⁶ It should be noted here that the primary sector includes non-ferrous metal smelting and refining, and the manufacturing sectors include the tertiary industries which repair and service their respective products. See The Inter-Industry Flow, 30, 31.

¹⁷ For an earlier exercise in this sort of analysis for the United States, see Leontief, W. W., “Exports, Imports, Domestic Output, and Employment,” Quarterly Journal of Economics, LX, no. 2, 02, 1946 Google Scholar, reprinted in Leontief, The Structure of the American Economy; also United States Mutual Security Agency, The Structure of the Italian Economy (Rome, 1953), chap. iv.Google Scholar

¹⁸ Gibson, , “Post-war Economic Development and Policy in Canada,” 440–3.Google Scholar

¹⁹ Eastman, H. C., “Recent Canadian Economic Policy: Some Alternatives,” this Journal, XVIII, no. 2, 05, 1952, 140.Google Scholar

²⁰ Safarian, A. E., “Foreign Trade and the Level of Economic Activity in Canada in the 1930's,” this Journal, XVIII, no. 3, 08, 1952, 336–40.Google Scholar

²¹ Looked at broadly, this attitude to “vulnerability” is simply a recognition that variables subject to direct influence only from within the economy are “policy variables” subject to public stabilization, while variables subject to external influence must fend for themselves. See Tinbergen, J., “The Relation between Internal Inflation and the Balance of Payments,” del Lavoro, Banca Nazionale, Quarterly Review, no. 23, 10-Dec, 1952, 4–8 Google Scholar; for a general discussion of these matters, see Myrdal, G., An International Economy (New York, 1956), chap. iv.Google Scholar

²² This assumption is discussed infra.

²³ See, for example, Haberler, G., “The Market for Foreign Exchange and the Stability of the Balance of Payments,” Kyklos, III, no. 3, 1949, 213.Google Scholar

²⁴ This figure is not particularly large, although one would expect it to be higher for Canada than for the United States. This is in fact the case. Leontief has shown that in the United States in 1951, a general change in exports would only have altered imports by 3.2 per cent; see “Factor Proportions and the Structure of American Trade,” Review of Economics and Statistics, XXXVIII, no. 4, 11, 1956 Google Scholar, Table 1, n. 3, and also 394–5. No completely comparable figure seems to exist for a European country, although one of Billewicz's methods of deriving the import content of British exports seems roughly similar to inter-industry analysis. His figures of 16.4 per cent for the United Kingdom in 1938 and 14.0 per cent in 1946 are probably a little low. See Billewicz, W. Z., “The Import Content of British Exports,” Economica, XX, no. 78, 05, 1953, 167 and passim.Google Scholar

²⁵ The standard treatise on this point is Machlup, F., International Trade and the National Income Multiplier (Philadelphia, 1943).Google Scholar

²⁶ Polak, J. J., An International Economic System (Chicago, 1953), 123–6Google Scholar, gives 0.68 as his value for Canada's “reflection ratio,” which is equal to the product of the foreign trade multiplier (2.13) and the marginal propensity to import (0.32). These figures can be taken only as suggestive because, first, the data consist of short time series of annual observations during the most abnormal years of the 1930's, and, second, Polak's least-squares method of estimation may result in biased estimates of parameters for those countries whose imports are large enough significantly to influence the total level of world trade.

³⁷ D.B.S., Review of Foreign Trade, Calendar Year, 1953 (Ottawa, 1954), 13, 43–4.Google Scholar

²⁸ “Post-war Economic Development in Canada,” 444.

²⁹ Watts, G. S., “Some Longer-Term Factors in the Canadian Balance of International Payments,” this Journal, XVI, no. 1, 02, 1950, 15.Google Scholar

³⁰ See, e.g., the summary of O. J. Firestone's researches in this field, cited by Walton, E. A., “The Vulnerability of the Canadian Economy,” this Journal, XX, no. 1, 02, 1954, 15.Google Scholar

³¹ Bryce, R. B., “The Effects on Canada of Industrial Fluctuations in the United States,” this Journal, V, no. 3, 08, 1939, 373–86.Google Scholar

³² Beckerman, W., “The World Trade Multiplier and the Stability of World Trade 1938 to 1953,” Econometrica, XXIV, no. 3, 07, 1956, 239–52.CrossRefGoogle Scholar

³³ Bishop, R. A., “Input-Output Work as a Basis for Development Planning,” Monthly Bulletin of Agricultural Economics and Statistics, V, no. 5, 05, 1956, 1–10 Google Scholar; A. Gosfield, “Input-Output Analysis of the Puerto Rican Economy” in N.B.E.R;, Input-Output Analysis, 321–62, but cf. “Comments” by D. Creamer and W: Isard, 362–7.

³⁴ Evans, W. D., “Marketing Uses of Input-Output Data,”, Journal of Marketing, XVII, no. 1, 07, 1952, 11–21 CrossRefGoogle Scholar; Leontief, W. W., “A New Approach to the Problem of Market Analysis” in American Management Association, Marketing Series, no. 59, 1945.Google Scholar