¹ Haites, Erik F. and Mak, James, “Steamboating on the Mississippi, 1810–1860: A Purely Competitive Industry,” Business History Review, XLV (Spring, 1971)Google Scholar; Mak, James and Walton, Gary M., “Steamboats and the Great Productivity Surge in River Transportation,” Journal of Economic History, XXXII (September, 1972)Google Scholar; and Haites, Erik F. and Mak, James, “The Decline of Steamboating on the Antebellum Western Rivers: Some New Evidence and an Alternative Hypothesis,” Explorations in Economic History, vol. 11 (Fall, 1973)Google Scholar. These in combination with other research comprise the forthcoming book by Haites, Erik F., Mak, James, and Walton, Gary M., Western River Transportation During the Era of Early Internal Improvements 1810–1860 (Baltimore, 1975)Google Scholar.

² Fred Bateman, James Foust, and Thomas Weiss, recipients of National Science Foundation Grant Nos. 2450 and 1456, Collaborative Research in United States Manufacturing, 1850–1870.

³ See Wright, C. D., The History and Growth of the United States Census (Washington, D.C., 1900)Google Scholar.

⁴ Three steamboats, the Logan, Mammoth Cave, and Major Barbour, had the lockage fees reported together with the insurance costs. For the purposes of our calculations, it was necessary to separate these items. This was accomplished by assuming that the insurance costs were equal to 15 per cent of the construction cost. This is the average rate for other tributary boats. The residual becomes the lockage fees.

⁵ The steamboats in the sample constitute about 7 per cent of the number of steamboats and about 10 per cent of the steamboat tonnage in operation in 1850. (See Haites and Mak, “The Decline of Steamboating on the Antebellum Western Rivers,” 35.) The sample, therefore, contains a disproportionately high number of the large trunk river boats.

⁶ Obviously, these data also shed light on other aspects of western transportation, such as vessel operating characteristics and productivity comparisons by route, but for these and other considerations, see the forthcoming study listed in footnote 1.

⁷ It should be pointed out that this designation of routes should not be considered as being fixed. Boats did move from route to route within a single season, as changing economic circumstances dictated. For example, the New Orleans Prices Current recorded 213 steamboat arrivals by sixty-six separate steamboats in New Orleans from Louisville between August 25, 1848 and August 31, 1849; thirty-eight of the vessels made the trip only once while some made as many as fourteen round-trips. In addition, the census data indicate that vessels that operated on trunk river routes had an average annual operating season of 8.9 months while those on the tributary rivers operated one full month longer despite the common knowledge that tributary rivers actually had shorter navigational seasons. It seems clear that the longer operating season witnessed on tributary vessels is explained by their descent to trunk river trades during those periods when tributaries became unnavigable due either to freezing or to low water.

⁸ For simplicity of exposition, assume continuous discounting, then the rate of return (r) earned over the lifetime of an investment (n) is given by the following equation:

where, I is the initial capital investment and II_t and D_t are respectively the profit and depreciation at time t.

Now assume that II_t and D_t are constant with values II and D. And assume that the depreciation allowance is continuously reinvested. In other words, the initial investment I is maintained with an infinite life and the depreciation becomes zero. This last assumption is difficult to justify for an individual project, but is not unreasonable for an ongoing industry.

As a result of the assumptions we have:

The last formulation, i.e., the profits divided by the original investment, is the profit rate calculated in the paper.

See also, Knight, Frank H., “Investment and Its Yield: Quantitative Relations,” reprinted in Readings in the Theory of Income Distribution (Homewood, Illinois, 1951), 409–417Google Scholar.

⁹ We assumed the out-of-pocket costs for one round trip to equal 9 per cent of the annual operating costs of a steamboat. The 9 per cent figure is based on estimates for the Louisville-New Orleans trade and thus undoubtedly overstates the requirements on shorter routes. This bias, however, does not alter our conclusions.

¹⁰ Hunter, Louis C., Steamboats on the Western Rivers (Cambridge, Mass., 1949), 366Google Scholar.

¹¹ Haites and Mak, “Steamboating on the Mississippi, 1810–1860,” Table A2, 72–73.

¹² Embree, Davis, editor of the Western Boatman, suggested (vol. 1, no. 1, January 1849Google Scholar, reprinted U.S. Congress Senate, The Statistics and History of the Steam Marine of the United States, Senate Exec. Doc. 42, 32 Cong., 1 sess., 1852, p. 117) that a steamboat last 24 per cent of its original value during each of its first two years. Thereafter it lost 24 per cent of the remaining value each year.

¹³ The non-linear depreciation functions do not provide full recovery of the original cost. In theory the undepreciated portion of the original cost is the salvage value. In practice this consists of payments from the insurance company and/or revenue from the sale of the vessel for scrap. There is virtually no information on the magnitudes of such payments and revenues nor on their relationship to the theoretical salvage value. We have allowed for full insurance coverage, so it is reasonable to assume that the insurance payments and scrap value would equal the salvage value.

With the above assumption and complete financial records, it is possible to calculate the profit of each individual steamboat as the rate of return over its lifespan. But we have financial records for only one year. So there are two alternatives: 1) assume that all costs and revenues (except depreciation) remain unchanged over the life of the boat and use the rate of return calculation; 2) use a “representative” depreciation charge and calculate the profit rate for the year for which data are available. We have chosen the latter approach. The representative depreciation charge is defined as the original cost of the boat divided by the average lifespan of all boats in the same category. This definition ensures that the original cost will be, on average, fully recovered, hence it incorporates the salvage value.

¹⁴ Lytle, William M., Merchant Steam Vessels of the United States, 1807–1868, publication no. 6 (Mystic, Conn.: The Steamship Historical Society of America, 1952)Google Scholar; and Supplements no. 2 and no. 3 (1954 and 1958).

¹⁵ The Uncle Sam is the only boat in the sample to carry only freight. The Paytona is the only boat in the sample that we know used slave labor. These special characteristics do not appear to have had a significant impact on profitability.

There is a possibility that the Swiftsure No. 4 and the John Simpson are misclassified. In the census they are reported as operating on trunk and tributary routes respectively. The Swiftsure No. 4 is relatively small. At 141 tons it is the smallest trunk boat in the sample (1.6 standard deviations below the mean trunk route size of 381 tons). The size and the fact that it operated twelve months suggest that it may have run primarily on tributary rivers. The John Simpson, at 228 tons, is the largest tributary boat (2.0 standard deviations above the mean tributary river size of 148 tons). It is possible that it operated primarily on trunk routes. If these boats are misclassified, the profit rates no longer look unusual.

¹⁶ With the large number of vessels in the sample that operated on trunk routes, we were able to compare differences in profitability in steamboating between routes. The average rate of return to the 22 vessels operating on the Louisville-New Orleans route was 7.7 per cent, while the remaining nine trunk river vessels averaged 10.4 per cent. We tested the difference between these two means to determine if it was significantly different from zero. Using a two-tailed test, we concluded that the two means were not significantly different at the .05 level of acceptance. The same conclusion was reached on packet vessels vs. all other trunk river vessels. We thus concluded that there were no significant differences in the average rates of return to investment among vessels operating on different trunk river routes.

¹⁷ The average lifespan of steamboats over 200 tons throughout most of the antebellum period, as calculated from the Lytle List, was six years (see Haites and Mak, “Steamboating on the Mississippi 1810–1860,” 54). The average lifespan of all steamboats over most of the antebellum period, calculated from the same source, was five years (see Haites and Mak, “The Decline of Steamboating on the Antebellum Western Rivers,” 27–28, fn. 7).

¹⁸ U.S. Congress, Senate Exec. Doc. 42, 32nd Cong., 1st session, serial set 619 (1852), 114.

¹⁹ The null hypothesis that the mean profit rates for trunk and tributary boats are not significantly different was subjected to a two-tailed test for each set of assumptions. In every case, this hypothesis was rejected at the 1 per cent level of significance. The profit rates of the trunk and tributary boats were significantly different.

See Hunter, Steamboats on the Western Rivers, and also Haites and Mak, “Steamboating on the Mississippi, 1810–1860,” for a discussion of the restrictions on entry.

²⁰ Although insurance premiums covered the loss of the vessel, they obviously did not cover the economic loss suffered by the steamboat operator when the vessel was lost to service.

²¹ After standardizing for differences in the length of the navigation season, the construction cost per ton and the size of the boat, the insurance costs for the tributary boats are significantly higher (at the 1 per cent level of confidence) than for the trunk boats.

²³ See U.S. Congress Senate, The Statistics and History of the Steam Marine of the United States, Senate Exec. Doc. 42, 32 Cong., 1 sess., 1852, (serial set 619), 116–117.

²³ Merrick, George B., Old Times on the Upper Mississippi; The Recollections of a Steamboat Pilot from 1854 to 1863 (Cleveland, 1909), 163Google Scholar, and Haites and Mak, “Steamboating on the Mississippi 1810–1860,” Table A2, 72, 73.

²⁴ Hunter, Steamboats on the Western Rivers, 386.

²⁵ Hall, James, The West: Its Commerce and Navigation (Cincinnati, 1848), 134Google Scholar.

²⁶ Hunter, Steamboats on the Western Rivers, 357.

²⁷ Fogel, Robert W. and Engerman, Stanley L., Time on the Cross (Boston, 1974), 1, 70Google Scholar. This conclusion is based on rate of return calculations rather than profit rates.

²⁸ Non-steamboating profit data are courtesy of Fred Bateman and Thomas Weiss. Estimates for 1860 are forthcoming. See Bateman, Fred, Foust, James, and Weiss, Thomas, “Profitability in Southern Manufacturing: Estimates for 1860,” Explorations in Economic History, Vol. 12 (Summer, 1975CrossRefGoogle Scholar). These 1850 estimates are preliminary estimates of the average profit rate.

²⁹ Ibid.

³⁰ The value of the F–ratio with 30 and 13 degrees of freedom is 6.20. The initial F–value (a = 0.05) for not rejecting the hypothesis that the variances are the same is 2.38.

³¹ See footnote 28 above.

³² Conrad, Alfred H. and Meyer, John R., “The Economicis of Slavery in the Ante Bellum South,” Journal of Political Economy, vol. 66 (April, 1958), 95–130CrossRefGoogle Scholar.

³³ Fishlow, Albert, “The Dynamics of Railroad Extension into the West,” in Fogel, Robert W. and Engerman, Stanley L., eds., The Reinterpretation of American Economic History (New York, 1971), 409Google Scholar.

³⁴ The existence of bonanza profits on the upper Mississippi in the mid-1850s was observed by a contemporary steamboat pilot, George B. Merrick, in Old Times on the Upper Mississippi.