¹ Schumpeter, Joseph, Business Cycles (two vols.; New York: McGraw-Hill, 1939).Google Scholar

² Schumpeter, Business Cycles.

³ Jenks, Leland, “Railroads as an Economic Force in American Development,” Journal of Economic History, IV, No. 1 (May 1944), 1–20.CrossRefGoogle Scholar

⁴ Paul Rosenstein-Rodan, “Notes on the Theory of the Rig Push,” Center for International Studies. (Unpublished.)

⁵ Cootner, Paul H., “Transport Innovation and Economic Development: The Case of the U. S. Steam Railroad, 1826–1886.” (Unpublished Ph. D. thesis, M. I. T., 1953.)Google Scholar

⁶ Gayer, Arthur,Rostow, W. W., and Schwartz, A. J., Growth and Fluctuations in the British Economy, 1790–1850 (two vols.; Oxford: The University Press, 1953), p. 1028.Google Scholar

⁷ It is usually profitable to incur additional construction expense to avoid heavy grades in building through hilly country but, unlike the case of canals, it is not mandatory.

⁸ From 1820 to 1830, the United States urban population grew 63 per cent as opposed to only one half that rate in the previous decade, when the major economic forces were spurring agricultural growth.

⁹ Eavenson, Howard, The First Century and a Quarter of the American Coal Industry (Baltimore: The Waverly Press, 1942).Google Scholar

¹⁰ It should be noted that exactly the same kind of coal boom was occurring in England and was the most important force in early British railway development. It is no accident that both Trevithick and Stephenson were deeply involved in the British coal industry. See Kay, Frederick, Pioneers in British Industry (London: Rockliff, 1952).Google Scholar

¹¹ For example, the Delaware and Hudson Canal averaged a rise and fall of nine feet per mile. On the connecting railroad that was constructed from canal head to mines, the average was one hundred feet per mile. See Kay, ibid.

¹² Ringwalt, John L., The Development of Transportation Systems in the United States (Philadelphia: J. Ringwalt, 1888).Google Scholar

¹³ MacGill, Caroline, History of Transportation in the United States to 1860 (Washington: Carnegie Institute of Washineton, 1917), p. 678.Google Scholar

¹⁴ Gray, Lewis C., History of Agriculture in the Southern United States (two vols.; Washington: Carnegie Institutions of Washington, 1933), p. 1016.Google Scholar

¹⁵ ibid.

¹⁶ Smith, Walter B. and Cole, Arthur H., Fluctuations in American Business, 1790–1860 (Cambridge: Harvard University Press, 1935), p. 195.Google Scholar

¹⁷ Gray, History of Agriculture.

¹⁸ Bidwell, Percy W. and Falconer, John I., History of Agriculture in the Northern United States (Washington: The Carnegie Institute of Washington, 1925), p. 512.Google Scholar

¹⁹ ibid.

²⁰ On Lake Erie, wheat prices rose 130 per cent from 1830 to 1837. In New Orleans, the cotton price rise was 80 per cent from trough to peak.

²¹ Schumpeter, Business Cycles (cited in n . l ).

²² Berry, Thomas S., Western Prices Before 1861 (Cambridge: Harvard University Press, 1943), p. 645.Google Scholar

²³ Cole, Arthur H., Wholesale Commodity Prices in the United States, 1700–1861 (2 vols.; Cambridge: Harvard University Press, 1938).CrossRefGoogle Scholar

²⁴ Poor, Henry V., History of the Railroads and Canals of the United States of America (Vol. I [no more published]; New York: J. H. Schultz and Company, 1860), p. 612.Google Scholar

²⁵ The most important transport project in the area was not a railway at all, but the Illinois Canal. See U. S. Congress, 32d Cong., 2d Sess. (1853), H. R. Doc. 136, On the Trade and Commerce of the British North American Colonies and Upon the Trade of the Great Lakes and Rivers, submitted by Israel D. Andrews.

²⁶ About another million bushels were exported to Canada. See Andrews report, ibid.

²⁷ These shifting regional investment incentives are reflected in contemporaneous stock price quotations. See Smith and Cole, Fluctuations, cited in n.16. For example, the New England railroad stock prices were rising to record highs in the 1840's, while the roads which were associated with the development of western expansion plunged in the early forties and recovered only partially in the later prosperity. Then, as the westward movement renewed after 1848, the New England stocks fell while central Atlantic and western roads soared.

When the boom of the fifties collapsed, western railroad stock prices collapsed with it, but eastern railroad stocks climbed steadily. This, too, is not solely a war phenomenon. Just as England and the eastern United States shared a railway boom in the 1840's, we find that British railway mileage completed in the 1860's exceeded that of the 1850's. Here again is evidence against the idea of innovation smoothly wiping out competition as its superiority gradually sinks in.

²⁸ Potter, J., “Atlantic Economy, 1815–60: The U. S. A. and the Industrial Revolution in Britain,” in Studies in the Industrial Revolution, Presnell, L. S., ed. (London: University of London, The Athlone Press, 1960).Google Scholar

²⁹ The need for expansion of grain “capacity” must have been accentuated by three successive wheat crop failures in the Lake Michigan area; see Bidwell and Falconer, History of Agriculture, p. 220 (cited in n.18) and Andrews report, H. R. Doc. 136, pp. 21 ff. (cited in n.25).

³⁰ MacGill, History of Transportation (cited in n.13).

³¹ The Atlanta and LaGrange was chartered in 1847; the Western and Atlantic, and the Southwestern, started building in 1848; and the Mobile and Ohio, and the Alabama and Tennessee roads were chartered in the latter year. See Phillips, Ulrich B., History of Transportation in the Eastern Cotton Belt (New York: Macmillan, 1913), p. 405.Google Scholar

³² The figures given in Table 3 for Chicago are more striking than those for arrivals at Buffalo because of crop failures, but both sets of data show the effect of the same forces. See p. 93 of Andrews report (cited in n.25).

³³ See Smith and Cole, Fluctuations (cited in n.16).

³⁴ In 1848, lumber receipts at Chicago were 60,000,000 board feet. In 1856, they were 450,000,000. See Hoyt, Homer, One Hundred Years of Land Values in Chicago (Chicago: University of Chicago Press, 1933), p. 85.Google Scholar

³⁵ The response of production to settlement was much faster in the 1850's than in the 1830's. This is quite likely due to the lack of tree cover, which reduced the time required for clearing land. It is interesting to note that the first settlers in the Midwest in the 1830's thought that land without trees was infertile and avoided it. See Bidwell and Falconer, History of Agriculture. As evidence of the distortion, note that in 1855, Cincinnati flour prices actually exceeded those in New Orleans and New York. See Berry, Western Prices (cited in n.22).

³⁶ Cotton, in a way, benefited from the Crimean War. United Kingdom cotton consumption fell during that War, holding prices down in the prosperity but raising them to high levels when the War ended as the United States depression was beginning.

³⁷ U. S. Congress (1874), Select Committee on Transportation Routes to Seaboard, Report.

³⁸ Computed from data in the Tenth Census of the United States (1880), Agriculture, Vol. CLI, and from New York State, Auditor of the Canal Dept, Annual Report, 1871 (Albany, 1872), pp. 220–23.Google Scholar It is probably an underestimate, but it is close to the correct figure.

³⁹ Bidwell and Falconer, History of Agriculture.

⁴⁰ Berry, Western Prices.

⁴¹ U. S. Geological Survey, Mineral Resources of the United States, 1882 (Washington: Government Printing Office, 1883).Google Scholar

⁴² Kuhlmann, Charles B., The Development of the Flour-Milling Industry in the United States (Boston: Houghton Mifflin Company, 1929), p. 346.Google Scholar

⁴³ Berry, Western Prices.

⁴⁴ Eavenson, American Coal Industry (cited in n.9).

⁴⁵ U. S. Geological Survey, Mineral Resources (see n.41).

⁴⁶ Of course, what was “demand” for the railroad industry was usually “supply” for its customers; i.e., the need for transportation to new potential sources of raw materials. Also, unless these demands for transportation services had been supplied, this whole discussion would be pointless. The point is that the disequilibrium was created primarily by a shift in the demand curve for transport services, rather than by a shift in the supply curve.

⁴⁷ Colean, Miles M. and Newcomb, Robinson, Stabilizing Construction: Record and Potential (New York: McGraw-Hill, 1952), p. 340.Google Scholar

⁴⁸ To be published in Proceedings of the International Economic Association Conference on “Take Off into Sustained Economic Growth.”

⁴⁹ On the tendency of the overall steam railroad capital-output ratio to decline after 1880, see Ulmer, Melville, Capital in Transportation, Communication and Public Utilities (Princeton: Princeton University Press, 1960)Google Scholar. This tendency is probably at least partially the result of the slowing rate of new investment, so that expanding utilization of older lines was not diluted as much by new extensive roads.

⁵⁰ Rosenstein-Rodan, “Theory of the Big Push” (cited in n.4).

⁵¹ See n.48.

⁵² In 1852, the tonnage of freight carried by rail was estimated at less than 20 per cent of that carried on canals and in the coastal trade. See Taylor, George R., The Transportation Revolution, 1815–1860 (New York: Rinehart and Company, 1951), p. 174.Google Scholar In 1849, railway revenues were only 7 per cent of an estimate of total expenditures on transportation and communication, in Martin, Robert F., National Income in the United States, 1799–1938 (New York: NICB, 1939), p. 146.Google Scholar

⁵³ Ulmer, Capital in Transportation (cited in n.49).

⁵⁴ Kuznets’ results as given in U. S. Department of Commerce, Historical Statistics of the United States, Colonial Times to 1957 (Washington: Government Printing Office, 1960), p. 143.Google Scholar

⁵⁵ Fishlow, Albert, “Technological Change in the Railroad Sector, 1840–1910.” Paper delivered at the National Bureau of Economic Research Conference on Income and Wealth, September 5, 1963. To be published.Google Scholar

⁵⁶ Ulmer estimates 40 per cent labor costs in road investment, based on 1929, 1935, and 1945 data. It was almost surely much more, a decade earlier.

⁵⁷ Swank, James M., History of the Manufacture of Iron in All Ages (Philadelphia: J. M. Swank, 1892), p. 526.Google Scholar

⁵⁸ American Coal Industry (cited in n.9).

⁵⁹ Coal consumption per ton of pig iron was about two tons. Coke consumption was about 1.25 tons per ton, and coal use was about 1.6 tons per ton of coke produced. This implies a figure of 6.5 million tons of coal used in pig-iron production. Assuming another ton of coal for each ton of steel brings the total to about 7.5 million tons. Rail production must have accounted for about 3.3 million tons of coal.

⁶⁰ Hunter, Louis C., “Financial Problems of the Early Pittsburgh Iron Manufacturers,” Journal of Economic and Business History, II (May 1930), 520–44.Google Scholar

⁶¹ Hunter, , “The Influence of the Market on Technique in the Iron Industry,” Journal of Economic and Business History, I (Feb. 1929), 241–81.Google Scholar

⁶² Unpublished thesis as cited in n.5.