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: 03 March 2009
This paper attempts to document and account for cost savings on North American small-grain prairie farms in the early twentieth century. Costs of production are analyzed using the ex post price and yield data abundantly available. Cost conjectures are developed and compared with scattered farm data that itemize inputs and reveal some aspects of farming technique. Total costs per acre, despite year to year fluctuations, appear to have fallen gradually over the entire period consistent with a comprehensive and continuous learning process, rather than only suddenly in the late 1920s when there was a marked increase in the sales of gasoline-powered farm machinery.
1 See Bennett, Merrill K., “Trends of Yield in Major Wheat Regions Since 1885, Part II: Irregular, Stable, and Declining Trends,” Wheat Studies, 14, (03 1938), 249–58Google Scholar, and Dick, Trevor J. O., “Productivity Change and Grain Farm Practice on the Canadian Prairie, 1900–1930,” this JOURNAL, 40 (03 1980), 105–10.Google Scholar
2 For particular innovations, see especially Ankli, Robert E. et al. , “The Adoption of the Gasoline Tractor in Western Canada,” in Canadian Papers in Rural History, Vol. II, Akenson, Donald H., ed. (Gananoque, 1980), pp. 30–33.Google Scholar
3 See Hopkins, E. S. et al. The Costs of Producing Farm Crops in the Prairie Provinces (Ottawa, 1935), p. 34Google Scholar, and Hargreaves, Mary W. M., Dry Farming in the Northern Great Plains 1900–1925 (Cambridge, Massachusetts, 1957), pp. 464–65.Google Scholar
4 This division of costs, defined in this paper to include both fixed and variable costs, is suggested by Parker, William N. and Klein, Judith L. V., “Productivity Growth in Grain Production in the United States, 1840–60 and 1900–10” in Output, Employment and Productivity in the United States After 1860, Studies in Income and Wealth, Vol. 30 (New York, 1966), p. 528.Google Scholar
5 See Silberberg, Eugene, The Structure of Economics (Toronto, 1978), pp. 309–13.Google Scholar
6 Price and yield data are taken for Canada from Strange, H. G. L., A Short History of Prairie Agriculture (Winnipeg, 1954)Google Scholar, Appendices II, III, and IV, and for the U. S. from annual issues of the U. S. Department of Agriculture, Yearbook of Agriculture (Washington, D.C., 1896–1931).Google Scholar
7 The Canadian wage series is based on Canada, Department of Labour, Wages and Hours of Labour (Ottawa, selected years 1920 and later)Google Scholar for 1901–1930, extrapolated backward to 1896 with Olley, Robert E., Construction Wage Rates in Ontario 1864–1903 (Kingston, 1961), pp. 74–75.Google Scholar Average farm wages in the U. S. were obtained from U. S. Department of Agriculture, Yearbook of Agriculture, 1931 (Washington, D.C., 1931), p. 1023.Google Scholar The Canadian farm machinery index relies on a farm implement index before 1913 and an official index of farm machinery prices thereafter. See Urquhart, M. and Buckiey, K., Historical Statistics of Canada (Toronto, 1965)Google Scholar, series L-92 and J-24. The U. S. machinery price index is constructed from retail prices paid by farmers, 1910 and later, extrapolated back to 1896 with a wholesale index of metals and metal products. See U. S. Department of Agriculture, Yearbook of Agriculture, 1931, p. 1021, and U. S. Department of Commerce,Google ScholarHistorical Statistics of the United States, Bicentennial Edition, Part I (Washington, D.C., 1975)Google Scholar, series E-47. A land price index for the Canadian prairies was constructed from the average prices of land sold by railroads and settlement companies. See Mackintosh, W. A., Economic Problems of the Prairie Provinces (New Haven, 1935), p. 285.Google Scholar A land price index for the United States was constructed using census benchmarks for 1900, 1910, and 1920, extrapolated back on the assumption of a linear trend. See Gray, L. C. et al. Farm Ownership and Tenancy (New York, 1976), p. 546Google Scholar, and Hargreaves, Mary, Dry Farming, Chap. 10.Google Scholar
8 See Urquhart and Buckley, Historical Statistics of Canada, series J-34, for the Canadian deflator, and U. S. Department of Commerce, Historical Statistics of the United States, series E23, for the U. S. deflator. For the exchange rate between the Canaiian and the U. S. dollar, see Urquhart and Buckley, series H-627.Google Scholar
9 This view is also held by Grest, E. G. in his An Economic Analysis of Farm Power in Alberta and Saskatchewan (Ottawa, 1935).Google Scholar See also Thomson, John Herd, The Harvests of War: The Prairie West, 1914–1918 (Toronto, 1978), pp. 63–65.Google Scholar
10 While average farm sizes increased only slowly over the period, there is some evidence of the cooperative use of machinery. Threshing, for example, was commonly hired out by independent operators who combined their machines and supervision with local hands as they moved from farm to farm. See MacEwan, Grant, Power for Prairie Plows (Saskatoon, 1971)Google Scholar, Chap. 12. See also Dick, Lyle, “Estimates of Farm Making Costs in Saskatchewan 1882–1914,” unpublished ms. presented to the 11th Conference on Quantitative Methods in Canadian Economic History (Kingston, 1981), pp. 16–17.Google Scholar
11 The factor price coefficients lack robustness for technical reasons that are not central to the issues pursued in the present study. See Varian, Hal R., Microeconomic Analysis (New York, 1978)Google Scholar, Chap. 4. The possibility of inferior factors arises under the hypothesis of cost minimization. See Silberberg, The Structure of Economics, pp. 195–98.Google Scholar Alternatively, the negative coefficient on machinery may be a symptom of aggregation bias, especially when machine costs are small part of total costs, and some elements subsumed under machinery are strongly associated with other factors of production. See Griliches, Zvi, “Specification Bias in Estimates of Production Functions,” Journal of Farm Economics (02 1957), pp. 16–18.Google Scholar
12 The appropriate variance for a given vector X0 is the dispersion of the forecast error measured by 
where 
is the estimated residual variance from the regression, n is the number of observations, and X is the data matruc for the regression. See Maddala, G. S., Econometrics (New York, 1977), pp. 44–65.Google Scholar Strictly speaking, the application of this formula in the context of regressions corrected for serial correlation leads to biased prediction errors. At least part of the bias is upward. See Pindyck, Robert S. and Rubinfeld, Daniel L., Econometric Models and Economic Forecasts, 2nd ed. (New York, 1981), pp. 215–17. I am indebted to Bronwyn Hall for programming assistance and for helpful discussions alerting me to these issues.Google Scholar
13 See Bennett, Merrill K., “Average Pre-War and Post-War Farm Costs of Wheat Production in the North American Spring Wheat Belt,” Wheat Studies, 1 (05 1925), pp. 173–207.Google Scholar
14 See the reporting irregularities noted in, Bennett, “Average Pre-War and Post-War Farm Costs,”Table 1, p. 202.Google Scholar
15 Although Bennett is aware of this influence, he does not apply any criterion of normal yield such as the average yield over a number of years. “Average Pre-War and Post-War Farm costs,” p. 193.Google Scholar
16 See Hopkins, E. S. et al. , The Costs of Producing Farm Crops in the Prairie Provinces (Ottawa, 1935), pp. 32–44.Google Scholar
17 See Bennett, “Average Pre-War and Post-War Farm Costs,” pp. 199–200.Google Scholar
