Article contents
Residential Capital Formation in Canada, 1871–1921*
Published online by Cambridge University Press: 07 November 2014
Extract
Partly because there is at present so much concern with problems of economic growth and partly because much has already been done to account for the years before the First World War, interest is now quickening in the process and magnitude of capital formation in Canada between Confederation and the end of the nineteenth century. An important part of Canadian capital formation has been in housing, and a historical series on housing investment, covering the years 1926–1941, was presented in an official publication prepared by Firestone and Urquhart in 1945. This series was superseded by revised estimates (which still form the basis for the historical figures in the National Accounts) made by Firestone in 1951; Buckley, partly on the basis of Firestone's later estimates, produced a series going back to 1896; and Firestone has recently returned to the problem. A consistent series from Confederation to the turn of the century is, however, still lacking.
The main purpose of this paper is to derive and present annual estimates of residential capital formation for the years 1871–1921. It is in three parts: the first presents annual estimates of the number of dwellings completed; the second considers the problem of securing annual average values and presents the final estimates of residential capital formation; and the third compares the new estimates with those published earlier, discusses the estimates and their limitations, and makes suggestions for their improvement. An effort has been made to present the material in such a way that the reader could go a long way towards reproducing the results. Inadequate data invite endless ingenuity: but unbridled wit can sometimes be dangerous and an effort has been made to keep the paper within the bounds of the plausible.
- Type
- Research Article
- Information
- Canadian Journal of Economics and Political Science/Revue canadienne de economiques et science politique , Volume 29 , Issue 1 , February 1963 , pp. 40 - 58
- Copyright
- Copyright © Canadian Political Science Association 1963
Footnotes
This paper owes a great deal to the kindness of a number of people. Some initial work was done while I was assisting with the Fixed Capital Stock Project at DBS in the summer of 1960. Mr. Ower, the DBS Librarian, was most helpful in providing source material, and Messrs. Rymes and Jones of the DBS willingly discussed early problems with me. Mr. Rymes has followed the work closely throughout and I am particularly grateful to him for his useful criticism and comment. An earlier draft of the completed paper was written at the University of Glasgow and usefully commented upon by Professor A. K. Cairncross, Mr. Gordon Fisher, Mr. Peter Hart, and Professor J. R. Parkinson then all of that University and by Dr. S. A. Goldberg of the DBS. The present version was written while I was a member of the Institute for Economic Research, Queen's University, where Mr. Alain Duncan provided valuable research assistance. I am grateful to the University of Glasgow and the Houblon-Norman Fund for travel grants in support of my research. Except where I have used official data and specified it as such, I am alone responsible for the figures presented in this paper and for the assumptions and calculations on which they are based.
References
1 Firestone, O. J. and Urquhart, M. C., Public Investment and Capital Formation (Ottawa, 1945).Google Scholar
2 Residential Real Estate in Canada (Toronto, 1951).Google Scholar
3 Buckley, K. A. H., Capital Formation in Canada, 1896–1930 (Toronto, 1955).Google Scholar
4 Canada's Economic Development, 1867–1953 (London, 1958).Google Scholar
5 The 1911 census contains additional information on the house-building industry; but the information is limited in extent and not in useful form.
6 For fuller discussion of the difficulties, see Census of Canada, 1931, I, chap. v. The main problem is that the distinction between rural and urban turns on incorporation.
7 Compare, for example, the discussion (“The Third Volume of the 1881 Census and Its Critics”) in Census, 1881, IV, xii–xiv, and the definition in Census, 1921, III, vii. The accuracy of house counts probably increases over time, but no estimate has been made of the likely error arising from inaccurate counts.
8 For a working definition of these concepts in American usage see Blank, D. M., The Volume of Residential Construction, 1889–1950, NBER Technical Paper no. 9 (New York, 1954), chap. 1.Google Scholar
9 The theoretical reason for the exclusion is the fact that most structures in this category would be financed from public funds and included, therefore, in public investment. The explanation of the relative insignificance of these buildings in fact is the census definition which normally requires a separate entrance to the street or dwelling. The same circumstance probably reduces the concern that need be felt about the lack of information on hotels.
10 The estimates of completion times seem plausible in light of information for later years. See Firestone, , Residential Real Estate, 412–13.Google Scholar Any error deriving from this procedure should be small as a proportion of the decade completions.
11 See Firestone, , Canada's Economic Development, Table 90, p. 299.Google Scholar
12 The proportion of one-fifth was based on the assumption that urban vacancies would be higher than rural. This tends to be confirmed by data for later years. See Firestone, , Residential Real Estate, 382–3.Google Scholar The decision to exclude temporary dwellings in the prairie provinces is justified by the proportion of these dwellings known to be rural.
13 See ibid. Firestone uses a rate of 4 per cent in Canada's Economic Development, 301. For the 1890's, a rate of 6 per cent was used for the non-prairie farms.
14 See Buckley, , Capital Formation in Canada, 121 Google Scholar, and Firestone, , Residential Real Estate, 384–7.Google Scholar
15 Capital Formation in Canada, p. 121.
16 See Firestone, , Canada's Economic Development, 191–200.Google Scholar
17 See Firestone, , Residential Real Estate, 385.Google Scholar
18 This seasonal pattern differs significantly from that which now prevails. There is, however, no reason why this shoula not be so. At a time when a greater proportion of imports came by sea and the climate probably had a more restrictive effect on economic activity than now, significant differences are to be expected. For a discussion of the more recent pattern and a consideration of the difficulties in its derivation, see DBS, Review of Foreign Trade, First Half-Year, 1958 (Ottawa, 1958), chap. IV.Google Scholar
19
The formulae used were: (a) For years ending March 31: calendar year N = fiscal year N +
fiscal year (N + 1) and (fo) for years ending June 30: calendar year N = ⅗ fiscal year N + ⅖ fiscal year (N + 1). The arbitrary element in these calculations will clearly be the less, the less violent annual fluctuations in the original figures. The effect of the formulae is, of course, to postulate a three-month lag between import and consumption.
20 The revised figures were made available to me by Mr. T. K. Rymes of DBS.
21 Residential Real Estate, 33
22 Ibid., 83.
23 An equivalent of one in sixteen completions and $3 in $141 in value.
24 Capital Formation in Canada, 120.
25 See, for example, the 1931 Census Monograph, Housing in Canada, Table 2, p. 549.
26 As adjusted by CMHC.
27 Keyfitz, N., The Growth of the Canadian Population, Population Studies IV (1950), 47–63.Google Scholar
28 Published in Residential Real Estate, Table 72.
29 The difference arises from the fact that Buckley obtains total investment by adding estimates for prairie farm, other farm, and non-farm houses and by adjusting the prairie farm value to allow for changes in size. See Capital Formation in Canada, 121, and Table N, p. 139.
30 It would be desirable to test the effect of some of the arbitrary assumptions made in this paper. The difficulty is in choosing among the alternative and equally arbitrary assumptions in order to obtain a standard of comparison. The assumptions behind the series of window glass consumption may be tested to some extent by comparing the results given above with those that would have been attained if an index of window glass imports taken directly from the Trade Reports had been used to obtain the annual volume series. There is a suggestion in a graph of the two alternative methods of obtaining the series of housing completions that, apart from the 1870's, the general pattern is very similar. A finer calculation of the annual ratio of window glass consumed to houses built would have made some difference to the standing of some successive years. This is again particularly so of the 1870's.
31 See Weber, , “A New Index of Residential Construction,” Scottish Journal of Political Economy, II, no. 2, 06, 1955, 104–32.Google Scholar
32 The present study began with the compilation of numerous tables on population, housing (occupied, vacant, and in construction) and imported window glass from census and trade reports. These tables proved too voluminous for inclusion in a Journal article, but the tables may be obtained in mimeographed form from the author.
- 1
- Cited by