Skip to main content

Oligopoly Agreement and the Timing of American Railroad Construction

  • C. Knick Harley (a1)

The railroads in the American West were constructed in a few concentrated building booms. This timing of construction resulted from the alternate creation and collapse of imperfect property rights to rights of way in partially settled areas. These “property rights” arose from strategic behavior within the railroad oligopoly. When enforcement costs of cooperative action were low, the railroads were able to create rents by avoiding construction ahead of demand. When enforcement became difficult, however, construction was the only way to capture rents on unbuilt lines so a construction boom ensued.

Hide All

1 See Harley C. Knick, “Transportation, the World Wheat Trade, and the Kuznets Cycle, 1850–1913,” Explorations in Economic History, 17 (07 1980), 218–50.

2 Fishlow Albert, American Railroads and the Transformation of the AnteBellum Economy (Cambridge, Massachusetts, 1965), Chap. 4.

3 The data from the Kansas Board of Agriculture are available in machine-readable form made available by the Inter-University Consortium for Political and Social Research. The data for “Adjustments to Resource Depletion—The Case of American Agriculture—Kansas 1874—1936” were originally collected by William N. Parker. Stephen J. De Canio, and Joseph Trojanowski. Neither the original collectors of the data nor the Consortium bears any responsibility for the analysis of interpretations presented here.

For a candid assessment of the quality of these statistics see Malm James C., Winter Wheat in the Golden Belt of Kansas (Lawrence, Kansas, 1944).

4 These figures are illustrative, but fit the Kansas data fairly well. They are used for illustrative purposes throughout the paper. Fortunately the qualitative conclusions are not enormously sensitive to this exact specification of interest rates, final earnings, and growth rates within a relevant range of values. Obviously more complex examples could be constructed but it seems unlikely they would significantly alter the results.

5 These net earnings will, of course, be just paying the rate of interest (6%) on the total capital invested. This total capital is the sum of the initial investments and the foregone compounded interest on this capital.

6 Marglin Stephen, Approaches to Dynamic Investment Planning (Amsterdam, 1963).

7 Seventeen rather than the 12 years of building “ahead of demand” mentioned above because settlement would grow more slowly in the absence of the railroad.

8 This proposition is a very simple extension of the classical Austrian Capital Theory problem of the optimal date to cut a tree. If the value of the timber increases as the tree grows but at a decreasing rate then the optimal time to harvest is when the rate of increase in value equals the rate of interest. The present value of the timber at time zero, given optimal harvest may be illustrated diagrammatically, by extending a line of slope (l–r) from a tangency to the growth curve to the yaxis. This is a rental value of the timber land. The present value of the growing timber becomes zero when a line with slope (l–r) passing through the origin intersects the growth curve.

Although this proposition is generally known it is often not fully appreciated. See in particular Fogel Robert, “Railroads as an Analogy to the Space Effort: Some Economic Aspects,” Economic Journal, 76 (03 1966), where he accepts the zero net present value criterion. This was brought to my attention by Barzel Yoram, “Investment, Scale and Growth,” Journal of Political Economy, 79 (03/04 1971), 216, n. 2.

9 The efficiency, from a social point of view, of delay depends on an implicit assertion that railroad freight charges equalled long-run marginal cost. If monopoly pricing of rail services existed profit maximizing and efficient timing would, of course, diverge.

10 See Harley C. Knick, “Western Settlement and the Price of Wheat, 1872–1913,” this JOURNAL, 38 (12 1978) and Harley, “Transportation.”

11 These data are from: Chicago price: Harley, “Transportation,” pp. 246–47; Kansas prices:Parker et al. (see footnote 3 above); Railroad rates: Hyde John and Newcomb H. T., “Changes in the Rates of Charge for Railway and Other Transportation Services,” United States Department of Agriculture Statistical Bureau, Bulletin, 15 (1898), pp. 2028 and 48.

12 A Chicago price of 97¢ implies a local price of about 67¢ a bushel or a bit less at the 99th meridian. The United States Department of Agnculture Yearbook 1893, pp. 515 and 517, estimate the cost of production of wheat, except rent, at between $7.00 and $7.50 per acre. Yield per acre in average years was between 13 and 17 bushels per acre in the mid-1880s and the general price level was some 20% higher than in 1893. These imply costs, excluding rent, of between 50¢ and 70¢ per bushel. Local price would be below 70¢ at the 99th meridian and around 50¢ at the west end of the state.

13 Grodinsky Julius, The Iowa Pool: A Study in Railroad Competition, 1870–84 (Chicago, 1950);Grodinsky , Jay Gould: His Business Career, 1867–1892 (Philadelphia, 1957); and Grodinsky , Transcontinental Railway Strategy, 1869–1893 (Philadelphia, 1962).

14 Transcontinental Railway Strategy, p. 105.

15 Ibid., p. 114.

16 Iowa Pool, p. 92.

17 Ibid., pp. 97–100.

18 Transcontinential Railway Strategy, Chap. 8.

19 Gould Jay, pp. 240–44 and Transcontinental Railway Strategy, Chap. 9.

20 Transcontinental Railway Strategy, pp. 96–100; 162–78.

21 For a discussion of the details of this collapse see Grodinsky, Transcontinental Railway Strategy, Chaps. 15 and 16.

22 North Douglass C., “International Capital Flows and the Development of the American West,” this JOURNAL, 16 (12 1956), 493505 and The Economic Growth of the United States, 1790–1860 (New York, 1961), pp. 6674.

23 The data used in the regression are described in Harley, “Transportation.” The Kuznets deflator has been extended to the pre-Civil War period on the basis of Gallman's implicit deflator and the Warren and Pearson index. The regressions were also run using a real interest variable which was constructed by subtracting the average price change in the five previous years from the nominal rate. The interest variable remained insignificant. Inclusion of the price of rails yielded a positive regression coefficient.

24 This is the same data set used in Harley, “Western Settlement.”

25 See, for example, Fishlow's Albert discussion in Davis Lance E., Easterlin Richard A. and Parker William N., et al. , American Economic Growth: An Economist's History of the United States (New York, 1972), pp. 500–05.

26 The same conclusion holds if a “real” interest rate is used.

27 This conclusion is even stronger when allowance is made for expectations of price declines. Prices declined very rapidly shortly after the Civil War but rather more slowly by the mid-1870s.

28 Thus the New York Times (03 16, 1876) stated: “The fact that the money markets of Europe are now closed against new railroad enterprises emanating from this country is too palpable to be denied. The growing disposition to invest in our railroad securitites, which has for some months been one of the characteristics of the London market, manifests itself only within very narrow limits. Investors confine themselves to well known and prosperous companies, whose capital affords ample security for money borrowed and the soundness of whose management has been attested by the uninterrupted payment of dividends during the hardest of hard times.”

The Burlington and the Rock Island were two such “well known and prosperous companies.”

Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

The Journal of Economic History
  • ISSN: 0022-0507
  • EISSN: 1471-6372
  • URL: /core/journals/journal-of-economic-history
Please enter your name
Please enter a valid email address
Who would you like to send this to? *


Full text views

Total number of HTML views: 0
Total number of PDF views: 3 *
Loading metrics...

Abstract views

Total abstract views: 117 *
Loading metrics...

* Views captured on Cambridge Core between September 2016 - 19th November 2017. This data will be updated every 24 hours.