¹ See particularly the discussion by Seybert, , Statistical Annals of the United States (Philadelphia, 1818), 281 Google Scholar, and by Seaman, , Progress of Nations (New York, 1846), 266–72.Google Scholar

² Giffen, Robert, “The Use of Import and Export Statistics,” Journal of the Royal Statistical Society, XIV, 1882 Google Scholar, reprinted in his Essays in Finance (London, 1880)Google Scholar; and Hobson, C. K., The Export of Capital (London, 1914).Google Scholar

³ Imlah, Albert, “British Balance of Payments and the Export of Capital, 1816–1913,” Economic History Review, 2nd series, II, 1952, 208–39Google Scholar, and Cairncross, Alec, Home and foreign Investment, 1870–1913 (Cambridge, 1953).Google Scholar

⁴ “The Beginning of American Financial Independence,” Journal of Political Economy, VI, Dec., 1897, 187–208 Google Scholar, and “Foreign Trade of the United States from 1820 to 1840,” ibid., VIII, Dec, 1899, 34–58.

⁵ “The Balance of Trade of the United States,” Review of Economic Statistics, Preliminary, I, 07, 1919, 213–68.Google Scholar

⁶ Canada's Bahnce of International Indebtedness, 1900–1913 (Cambridge, Mass., 1924).Google Scholar

⁷ French International Accounts, 1880–1913 (Cambridge, Mass., 1933).Google Scholar

⁸ This study is confined to shipping earnings although, of course, the transport account includes land transport as well as insurance and brokerage fees.

⁹ For a discussion of some current problems of estimating freight earnings see Moneta, Carmellah, “The Estimation of Transportation Costs in International Trade” Journal of Political Economy, LXVII, Feb., 1959, 42–3.Google Scholar

¹⁰ For the United States there are several estimates of freight factors during the nineteenth century, e.g.: Bates, William, American Navigation (Boston, 1902)Google Scholar, Appendix; and American State Papers, Commerce and Navigation, II, 395.Google Scholar The most reliable of the estimates is that of Wells, David, Report of Special Commissioner of the Revenue, 1869, 31.Google Scholar It is also possible to derive freight factors from R. Giffen's estimates of United Kingdom earnings.

¹¹ Sources for the rates and freight factors presented here are contained in the “Study of the United States Balance of Payments, 1790–1860” by Douglass North presented at the 24th Conference on Income and Wealth. Proceedings to be published.

¹² See North, Douglass, “Ocean Freight Rates and Economic Development, 1750–1913,” Journal of Economic History, Dec., 1958, 540.Google Scholar

¹³ North, , “Ocean Freight Rates,” 542.Google Scholar

¹⁴ For example, Baltic rates fell more rapidly in the first half of the nineteenth century and less rapidly in the second half of the nineteenth century than rates on the North Atlantic.

¹⁵ The immigrant trade was lucrative and was an important influence in lowering rates on the North Atlantic import trade into Europe, but it is the only major exception.

¹⁶ Export of Capital, 174–87. The classic study of freight earnings as an important invisible item in a country's balance of payments was that of Sir Robert Giffen. Giffen's work will not be discussed here because his estimates relate only to selected years. However, as has already been pointed out, the influence of his work on subsequent studies of freight earnings has been very significant. For a discussion of his work see Hobson, , Export of Capital, 172–4Google Scholar, and Imlah, , “British Balance of Payments,” 213–15.Google Scholar

¹⁷ For a more detailed discussion on the Board of Trade's indices see Isserlis, L., “Tramp Shipping Cargoes and Freights,” Journal of the Royal Statistical Society, Part I, 1938, 55–8.Google Scholar

¹⁸ Cairncross, , Home and Foreign Investment, 175.Google Scholar

¹⁹ Isserlis, , “Tramp Shipping Cargoes,” 67.Google Scholar Data for 1931 from a sample of British shipping companies cited by Isserlis indicated that over 10 per cent of gross British earnings was from trade between foreign ports, and that this was probably low, because of depressed trade.

²⁰ Home and Foreign Investment, 174–5.

²¹ Isserlis, , “Tramp Shipping Cargoes,” 74, 122.Google Scholar His index is from 1869 to 1936.

²² Imlah, , “British Balance of Payments,” 214–18.Google Scholar Mr. Imlah has since revised part of his estimation procedure in his recent book, Economic Elements in the Pax Britannica (Cambridge, Mass., 1958), 48–55.Google Scholar Since the former study provides the best description of the procedure, it will be cited except when there are differences between the two works.

²³ This index was calculated for another study by Imlah, : “Terms of Trade of the United Kingdom, 1798 to 1913,” Journal of Economic History, Nov., 1950, 171–2.Google Scholar

²⁴ Ibid., 180–2, and Cairncross, , Home and Foreign Investment, 176.Google Scholar The latter's inward index is from 1870 to 1912, while the combined index is from 1870 to 1913, so that these were the years chosen for comparison with Imlah's index, which covers a longer period.

²⁵ Canada's Balance of International Indebtedness, 14–78.

²⁶ Moneta, , “Estimation of Transportation Costs,” 55–6.Google Scholar

²⁷ Harry D. White in calculating French earnings from 1880 to 1913 derives his results from Hobson's study by using the proportion of French tonnage to British tonnage. However, for calculating the credit due to France for internal transportation of goods between two other countries, White is able to make fairly direct calculations since rail rates are available for the whole period. French International Accounts, 56–66.

²⁸ Sterns' study occurs in the two articles cited in n. 4 above; for the study by Bullock et al. see n. 5 above.

²⁹ Actually, there are two available registered tonnage series for the United Kingdom. One series is simply net registered tonnage, and it includes tonnage in the coasting trade in addition to that in foreign trade. The other series is tonnage not engaged in the home trade, and this series does not include tonnage engaged in the “near seas trade” between ports in England and continental ports between the River Elbe and Brest. Hobson, and Imlah in his first estimates, used the former series, while Cairncross believes less error is involved in using the latter series.

³⁰ Imlah's method may be briefly stated. Robert Giifen estimated that the annual earnings net of port costs abroad per registered ton of sail was £5 and of steam £15 in 1880 (“The Use of Import and Export Statistics,” 259–70). Giffen's estimates were based on a sample of earnings of sail and steam vessels for several large shipping firms, and have generally been regarded as reliable estimates for that year.

In both of his studies, Imlah used these estimates as his base. As an index of how freight rates and hence earnings per ton changed over time, he used an import price index as was discussed above. Up to this point Imlah's method is the same as that of Hobson and Cairncross. There is a nominal difference in that Hobson and Cairncross do not calculate sail and steam earnings separately. However, they do convert sail into steam tonnage in the ratio of 3 to 1, which is the same ratio of earnings which Imlah uses in his base year; so, in effect, the two methods are equivalent. The methods would only differ if Imlah had separate freight rate indices of sail and steam.

Imlah improves on the method of Hobson and Cairncross by adding an index of the activity of the fleet. In “British Balance of Payments,” Imlah took total tonnage entered in cargo and in ballast divided by total registered tonnage in each year, setting the value in the base year, 1880, equal to 100. This is equivalent to estimating the earnings per ton of ships entering in cargo and in ballast. In Economic Elements, the activity index is modified, ana the resulting measure which Imlah uses is earnings per ton for the average of entrances and clearances in cargo only. One other difference is that in his first calculations Imlah allowed .3 per cent a year as an allowance for increasing efficiency of shipping.

³¹ This procedure was used in an early study of Anglo-American trade by Chapman, Sidney, History of Trade between the United Kingdom and the United States (New York, 1899), 14–18.Google Scholar

³² Export of Capital, 182.

³³ Clearance and entries of a given ship are similar to total sales, while the registered tonnage of the ship represents in some sense the fixed capital of a ship (firm). Earnings per ton entered may fall when there are increases in efficiency of ships, while earnings per registered ton rise. If clearances and entries are an accurate gauge of the actual quantities of goods shipped, and the freight index is reliable, then Imlah's method should produce a good indirect estimate of the effects of faster service on the earnings of vessels. Imlah does not explicitly recognize this point in his writings.

³⁴ The understatement would arise because freight rates are likely to fall, ceteris paribus, owing to the increased carrying capacity, but clearances and entries need not rise to the increased carrying capacity, even though the quantity shipped does rise. What happens to total earnings will depend on the elasticity of demand, but since Imlah's method does not allow for the increase in quantity, total earnings must necessarily be less than otherwise.

³⁵ White, , French International Accounts, 60–1Google Scholar; Cairncross, , Home and Foreign Investment, 172–3Google Scholar; Imlah, , Economic Elements, 54.Google Scholar

³⁶ North, “United States Balance of Payments,” App. I.