No CrossRef data available.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 07 November 2014
The Dominion Bureau of Statistics is currently completing a table of the inter-industry flow of goods and services in Canada for the year 1949. This paper is a preliminary report on the project which tries to answer the following questions: (1) What are the objectives of the Bureau's study of inter-industry relationships? (2) What are the principal characteristics of the 1949 table of inter-industry flows? (3) What are the principal sources of the data, the methods of estimation, and the reliability of the estimates? (4) What are the limitations of the use of the 1949 table for input-output analysis? The task of evaluating the results of the project must be left until the table has been completed and has been carefully studied.
The primary reason for constructing a set of inter-industry accounts for 1949 was to bring together, in an integrated framework, industrial and other economic statistics collected by the Bureau in order to create a tool for statistical management which draws attention to inconsistencies in classification and gaps in the data. The backbone of a good statistical system is standard classification of industries and commodities but, to ensure that the classifications are being implemented uniformly and to ensure that the classifications are analytically useful, statistics based on them should be integrated in some suitable framework. A table of inter-industry flows is ideally suited for this integration since one of its main features, as will be explained later, is that the total value of output of an industry is defined to be identical to the total value of input. If input and output can be estimated independently, the discrepancy between the two estimates provides an indicator of the accuracy of the data. Moreover, as Dr. Milton Gilbert has suggested: “It is only by having such an all-embracing framework that one can establish any order of priority in statistical collection. Otherwise the national ‘program’ of data-gathering tends to be like the program of a variety show; the first act is an acrobat, the second is an Irish tenor, the third is a tap dancer, and so on, with no connection between them. The parts of the program do not make up a whole drama or tell one story.”
This paper was presented at the annual meeting of the Canadian Political Science Association in Toronto, June 3, 1955. All opinions expressed herein are the personal responsibility of the author.
1 The name input-output table may also be used and the two terms are frequently used interchangeably.
2 D. H. Steinthorson has constructed a set of input-output statistics for the Canadian economy for 1939 and 1947 by applying input coefficients for the United States for the year 1939, derived by the U.S. Bureau of Labor Statistics, to control totals obtained from Canadian statistics. See Steinthorson, D. H., “Problems in. Input-Output Analysis of the Canadian Economy,” unpublished Ph.D. dissertation, Harvard University, 1954.Google Scholar
3 Gilbert, Milton, Discussion of “Measurement of National Income,” Econometrica, XVII, Supplement, 07, 1949, 256.Google Scholar
4 The detailed distribution of gross domestic product at factor cost is being used for 1949 weights for the Bureau's index of real output. See Berlinguette, V. R., “Measurement of Real Output,” this Journal, XX, no. 1, 02, 1954, 59–75.Google Scholar
5 Parts of this section have been influenced by Richard Stone and Utting, J. E. G., “The Relationship between Input-Output Analysis and National Accounting” in Netherlands Economic Institute, Input-Output Relations, Proceedings of a Conference on Inter-Industrial Relations held at Dreibergen, Holland (Leiden, 1953), 195–224 Google Scholar; Barna, Tibor, “The Interdependence of the British Economy,” Journal of the Royal Statistical Society, Series A, CXV, 1952, 29–77 CrossRefGoogle Scholar; and Evans, W. D. and Hoffenberg, M., “The Interindustry Relations Study for 1947,” Review of Economics and Statistics, XXXIV, 05, 1952, 97–142.CrossRefGoogle Scholar
6 Government is made to appear in the inter-industry flow system both as a producer of “government service” (which might be more appropriately called “public administration and defence”) and as a final user, “government expenditure on goods and services.”
7 To keep input in balance with output it was necessary to adjust corporation profits to an establishment basis. The industrial distribution of profits in the National Accounts is on a company basis.
8 Dominion Bureau of Statistics, Standard Industrial Classification Manual (Ottawa, 1948).Google Scholar
9 See United Nations, International Standards in Basic Industrial Statistics, Statistical Papers, Series M, no. 17, 1953, 10 Google Scholar; Simmons, W. R., “The Elements of an Industrial Classification Policy,” Journal of the American Statistical Association, XLVIII, 09, 1953, 435–6Google Scholar; and D.B.S., Standard Industrial Classification Manual, 2.Google Scholar
10 See Robinson, E. A. G., The Structure of Competitive Industry (revised ed., Cambridge, 1935), 6–13 Google Scholar; Robinson, Joan, “Imperfect Competition Revisited,” Economic Journal, LXIII, 09, 1953, 579–81CrossRefGoogle Scholar; and D. B. S., Standard Industrial Classification Manual, 1–5.Google Scholar
11 Robinson, E. A. G., Competitive Industry, 13.Google Scholar
12 “A considerable degree of heterogeneity … [in the output of a firm] can be allowed for by supposing the figure which represents the output of the firm to be an index of out-put, … when we are solely concerned with scale, there does not seem to be any harm in … [this simplification].” Hicks, J. R., “The Process of Imperfect Competition,” Oxford Economic Tapers, N.S. VI, 02, 1954, 42.Google Scholar
13 A table for the Italian economy for 1950 has been constructed with 200 rows and 60 columns as a solution to this problem. United States of America, Mutual Security Agency, Special Mission to Italy for Economic Cooperation, The Structure of Growth of the Italian Economy (Rome, 1953).Google Scholar
14 For a fuller discussion see Modlin, C. P. and Rosenbluth, G., “The Treatment of Foreign and Domestic Trade and Transportation Charges in the Leontief Input-Output Table” in Morgenstern, Oskar, ed., Economic Activity Analysis (New York, 1954), 129–99.Google Scholar
15 For an example of the integration of inter-industry flows with National Accounts see United Kingdom, Central Statistical Office, National Income and Expenditure, 1946–1951 (1952), 22–5, 62–4Google Scholar, and for discussions of the interrelation with the Accounts in the Netherlands, Norway, and the United Kingdom see Netherlands Economic Institute, Input-Output Relations, 111–94.Google Scholar For a popular exposition see Edey, Harold C. and Peacock, Alan T., National Income and Social Accounting (London, 1954), 32–5 and 141–51.Google Scholar
16 For an explanation of National Income concepts see Dominion Bureau of Statistics, National Accounts, Income and Expenditure, 1926–1950 (Ottawa, 1952), 83–7Google Scholar; United Nations, A System of National Accounts and Supporting Tables, Studies in Method, Series F, no. 2 (New York, 1953)Google Scholar; and Goldberg, S. A., “The National Accounts—Concepts and Applications,” Canadian Journal of Agricultural Economics, II, spring, 1954, 1–14.Google Scholar
17
In addition to the adjustment for interest and dividends, adjustment should be made for profits, wages, and other factor shares but these are small and tend to cancel out. Sources: D.B.S., National Accounts, 1926–1950, 26–9Google Scholar, and D.B.S., Canadian Balance of International Payments during the Post-War Years, 1946–1952 (Ottawa, 1953), 90.Google Scholar
18 Calculation was made with unrounded figures.
19 For the chemicals and allied products industry, 1949, the estimate was made as follows:
Sources: Figures (1)−(3) from D.B.S., General Review of the Manufacturing Industries of Canada, 1949 (Ottawa, 1952), 23 Google Scholar; small adjustments were made to these figures for the inter-industry flow study. Figure (4) from Table II of this paper. Figure (6) from Department of Trade and Commerce, Private and Public Investment in Canada, Outlook 1951 (Ottawa, 1951), 15.Google Scholar An adjustment was made to eliminate duplication of wages and salaries.
A breakdown of operating expenses of manufacturing firms as percentages of net sales for Canada for the year 1936 is available in MacGregor, Donald C., “Manufacturers' Expenses, Net Production, and Rigid Costs in Canada,” Review of Economic Statistics, XXVIII, 05, 1945, 60–73.CrossRefGoogle Scholar
20 For a description of the commodity flow method see United States Dept. of Commerce, National Income, 1954 Edition: A Supplement to the Survey of Current Business (Washington, 1954), 103–15, 128–33.Google Scholar
21 Lebergott, Stanley, “Measurement for Economic Models,” Journal of the American Statistical Association, XLIX, 06, 1954, 211–13.Google Scholar See also Morgenstem, Oskar, On the Accuracy of Economic Observations (Princeton, N.J., 1950), esp. 37–44.Google Scholar
22 A good exposition and example of input-output analysis is contained in The Structure and Growth of the Italian Economy. For discussions of input-output analysis see Dorfman, Robert, “The Nature and Significance of Input-Output,” Review of Economics and Statistics, XXXVI, 05, 1954, 121–33CrossRefGoogle Scholar; and National Bureau of Economic Research, Conference on Research in Income and Wealth, Input-Output Analysis: An Appraisal, Studies in Income and Wealth, vol. XVIII (Princeton, N.J., 1955).Google Scholar
23 Hicks, J. R., The Social Framework: An Introduction to Economics (2nd ed., Oxford, 1952), 219.Google Scholar
24 Ibid.
25 Ibid.
26 In order to obtain input coefficients which relate more directly to productive processes than to inputs and outputs of establishments, Leontief's group at the Harvard Economic Research Project are attempting to use engineering studies of production functions rather than census of industry data. Leontief, W. W. et at., Studies in the Structure of the American Economy (New York, 1953), part iv.Google Scholar
27 See Leontief, W. W., The Structure of American Economy, 1919–1939: An Empirical Application of Equilibrium Analysis (2nd. ed., New York, 1951), 37–41, 201–2.Google Scholar
28 The implications of various assumptions concerning the relationship between imports and domestic production have been investigated by D. H. Steinthorson, “Problems in Input-Output Analysis.”
29 See Klein, L. R., “On the Interpretation of Professor Leontiefs System,” Review of Economic Studies, XX, 1952–1953, 131–6.CrossRefGoogle Scholar
30 Leontief and his associates have attempted to extend input-output analysis to include “capital coefficients”—the ratio of output in an industry to the capital stock required to produce that output. See Leontief, Studies in the Structure of the American Economy, chaps, iii, vi, vii.
31 Hansen, Alvin H., “On The American Economy in 1960 ,” Review of Economics and Statistics, XXXV, 11, 1953, 263.Google Scholar
32 However, see, for example, Barnett, H. J., “Specific Industry Output Projections” in National Bureau of Economic Research, Conference on Research in Income and Wealth, Long-Range Economic Projection, Studies in Income and Wealth, vol. XVI (Princeton, N.J., 1954), 191–226 Google Scholar, and Leontief, , The Structure of American Economy, 216–18.Google Scholar