An earlier version of this article was presented at the 8th Purdue Conference on the Application of Economic Theory and Quantitative Techniques to Problems of Economic History, February 2, 1968. We are indebted to Lawrence A. Harper for comments and William Moss for research assistance on an earlier draft of this article. Any shortcomings, of course, remain the responsibility of the authors. We would like to acknowledge the provision of financial assistance from the National Science Foundation and the Purdue Research Foundation.

¹ See, for examples, Nettels, Curtis P., The Roots of American Civilization (New York: Appleton-Century-Crofts, 1938), p. 264;Google Scholar and North, Douglass C., Growth and Welfare in the American Past (Englewood Cliffs, N. J.: Prentice-Hall, Inc., 1966), p. 40.Google Scholar

² We refer to the trade discussed in this paper as the “overseas trade” of the colonies. The term “foreign trade” would not be proper because Great Britain and British West Indies were not foreign areas.

³ These records are in the Public Record Office in London and catalogued there as Customs 16/1. They cover the period of January 5, 1768, to January 5, 1773. For a more complete discussion of these estimates of exports, imports, and the other estimated debits due to the immigration of slaves and indentured servants, see the sources cited in Table 1. The term “commodity trade” will be used to include all debits and credits due to these items.

⁴ A word of caution should be given at this point regarding the regional breakdown of the estimated balances of commodity trade. The balances represent the trade of each colonial region with each overseas trading area. There was undoubtedly some modification of these balances brought about by the trade that took place among the colonies themselves (as suggested in the statement of Benjamin Franklin quoted below). No attempt has been made to account for the effects of trade between the colonies in these balances.

⁵ Quoted from Faulkner, Harold U., American Economic History (8th ed.; New York: Harper & Brothers, Publishers, 1960), p. 81.Google Scholar

⁶ Nettels, The Roots, p. 256; and Callender, Guy S., “The Early Transportation and Banking Enterprises of the States in Relation to the Growth of Corporations,” The Quarterly Journal of Economics, XVII (Nov. 1902), 137.Google Scholar

⁷ The expenditures made by the British government in the colonies were for defense, for costs of civil administration and justice that were assumed by Great Britain, and for costs of maintaining the customs administration and enforcing customs regulations. Colonial revenues (mainly customs duties) that accrued to the British Government must be deducted from these British expenditures. See section III for a discussion of the probable magnitude of these expenditures.

⁸ See North, Douglass and Heston, Alan, “The Estimation of Shipping Earnings in Historical Studies of the Balance of Payments,” The Canadian Journal of Economics and Political Science, XXVI (May 1960), 266–67.Google Scholar

⁹ See, for example, Faulkner, Economic History, pp. 80–81.

¹⁰ See Gary M. Walton, “Trade Routes, Ownership Proportions, and American Colonial Shipping Characteristics,” presented to the International Commission of Maritime History, University of Seville, Seville, Spain, September 28, 1967 (forthcoming, The Reports of the Congress, 1967), and “New Evidence on Colonial Commerce,” The journal of Economic History, XXVIII (Sept. 1968), 363–89.Google Scholar

¹¹ Between 51 and 52 percent of the total tonnage entering colonial ports was colonial-owned. This would indicate that the amount of colonial-owned tonnage entering foreign ports was approximately equal to the amount of noncolonial-owned tonnage entering colonial ports.

¹² See the statement by Harper, Lawrence A. in U.S. Bureau of the Census, Historical Statistics of the United States, Colonial Times to 1957 (Washington, D.C.: U.S. Government Printing Office, 1960), p. 745;Google Scholar and Davis, Ralph, The Rise of the English Shipping Industry in the Seventeenth and Eighteenth Centuries (London: Macmillan and Co., Ltd., 1962), p. 7n.Google Scholar For a discussion of the different definitions of tonnage, see Lane, Frederic C., “Tonnages, Medieval and Modern,” The Economic History Review, 2d ser., Vol. XVII (Dec. 1964), 213–33.CrossRefGoogle Scholar

¹³ The tonnage figures in Table 2 include all vessels legally entering and clearing colonial ports. The Navigation Acts excluded foreign ships from colonial ports, but ships owned in the colonies had the same privileges as those owned by residents of Great Britain. For the present purposes, then, we must know what proportion of the tonnage entering and clearing colonial ports was owned by residents of the American colonies as opposed to the tonnage owned by residents of other parts of the Empire. The clearances, of course, are an ex ante figure, and not all vessels successfully completed their voyages. Insurance rates have been used to approximate losses en route.

¹⁴ See William Walton's Insurance Book, New York Chamber of Commerce Library, New York; Ezekiel Price's Policy Books, Boston Athenaeum, Boston, Mass.; John Hancock Collection, Volume 28, Manuscripts Division, Baker Library, Harvard University; and rough journals of the London Assurance Company, London, England.

¹⁵ More formally, if no vessels were lost en route, shipping earnings for any year that accrued to a particular colonial region “d” in its trade on a particular route between colonial region “r” and overseas area “a” would be:

where is shipping earnings of colonial region “d” in its trade between colonial region “r” and overseas area “a.”

and are total shipping tonnages that cleared colonial region “r” for overseas area “a”, and entered colonial region “r” from overseas “a,” respectively, in tons;

and are the percentages of colonial-owned vessels from region “d” which cleared and entered, respectively, in the trade route between “r” and “a”;

and are the average freight rates charged by ships traveling from region “r” to area “a,” and from area “a” to region “r,” respectively, in pounds sterling per ton; and

and are the percentages of shipping capacity (tonnage) utilized in the trade route between “r” and “a,” and “a” and “r,” respectively. Total shipping earnings (of components a and b) of the colonies for any year would be:

If freight rates were available for every commodity on every route, it would be a simple task directly to estimate shipping earnings by the price charged times the quantity shipped. The lack of rates for every commodity, however, requires that the quantities of commodities be converted to a standard unit of shipping tonnage, the quantity for which we have been able to construct average freight rates.

¹⁶ See Walton, Gary M., “A Measure of Productivity Change in American Colonial Shipping,” The Economic History Review, 2d ser., Vol. XXI (Aug. 1968), 268–82.Google Scholar Some of the rates used by Walton have previously appeared in published form, for example, Hemphill, John M. II, “Freight Rates in the Maryland Tobacco Trade, 1705–62,” Maryland Historical Magazine, LIV (Mar., June 1959), 36–58 and 153–87;Google Scholar Davis, Rise of English Shipping, p. 283; Bell, Herbert C., “The West India Trade Before the American Revolution,” The American Historical Review, XXII (Jan. 1917), 274;Google ScholarHarper, Lawrence A., The English Navigation Laws (New York: Columbia University Press, 1939), p. 268;Google ScholarWeeden, William B., Economic and Social History of New England, 1620–1789 (New York: Hillary House Publishers, 1963), I, 126, 261, and 369.Google Scholar

¹⁷ Factor prices of the major inputs into shipping (mainly seamen's wages and shipbuilding costs) were relatively stable from the seventeenth century to the American Revolution, and there were few reasons for these costs to vary for routes of similar distance. The existence of economies of scale might be one reason for rates in similar routes to differ and for the long-run supply curve not to be horizontal. But shipping was probably of sufficient volume in most of the trans-Atlantic routes so that whatever economies of scale existed were captured on all routes. See Walton's, discussion of shipping costs in “Sources of Productivity Change in American Colonial Shipping, 1675–1775,” The Economic History Review, 2d ser., Vol. XX (Apr. 1967), 67–78.CrossRefGoogle Scholar One might raise a question about whether or not freight rates fluctuated substantially so that cost conditions are irrelevant in the short run. The series on tobacco rates suggests a significant degree of short-run stability, remaining at: about £7 to £8 per “ton” (of four hogsheads) during peacetime. This stability of tobacco rates supports the use of the same rate per route for each year in this period (as given in Table 4). It should be noted, however, that Davis (Rise of English Shipping, p. 282) would probably disagree with this contention, at least for die West Indian sugar trade.

¹⁸ See North and Heston, “Estimation of Earnings,” p. 269; Davis, Rise of English Shipping, pp. 185–88; North, Douglass C., “Ocean Freight Rates and Economic Development, 1750–1913,” The Journal of Economic History, XVIII(Dec. 1958), 539;Google Scholar and Bell, “The West India Trade,” p. 278. The African trade was an exception, of course, to this pattern of utilization.

¹⁹ See Gray, Lewis C., History of Agriculture in the Southern United States to 1860 (Gloucester, Mass.: Peter Smith, 1958; originally published by the Carnegie Institution of Washington, 1933), I, 220–21; Davis, Rise of English Shipping, p. 282; and Walton, “A Measure of Productivity Change,” p. 271.Google Scholar

²⁰ See Walton, “New Evidence.” We do not mean to imply that ships engaged in trade between the colonies and the West Indies typically may not have gone to several islands, or in the southern European trade, to several southern European ports. We do mean that routes involving two or more overseas areas were not typical, such as, from the colonies to the West Indies to Great Britain and back to the colonies.

²¹ The sources are C. O. 133/17 and C. O. 142/18, Public Record Office, London. We are indebted to Professor Lawrence A. Harper, who kindly permitted us to use his copies of the Barbados naval office lists for 1773.

²² This was roughly the proportion of exports going to Great Britain from Barbados and Jamaica relative to total exports from all the West Indies. See Davis, Rise of English Shipping, p. 298.

²³ Davis. Rise of English Shipping, p. 283; and Walton, “A Measure of Productivity Change.”

²⁴ See Walton, “A Measure of Productivity Change;” Davis, Rise of English Shipping, p. 239; and Lane, “Tonnages,” p. 220. Also see the discussion about the meaning of “tons burden” in the Appendix available from the authors.

²⁵ Nettels, Curtis P., The Money Sypply of the American Colonies before 1720 (Madison: University of Wisconsin Press, 1934), p. 70.Google Scholar

²⁶ White, Philip L., The Beekmans of New York in Politics and Commerce, 1647–1877 (New York: The New-York Historical Society, 1956), p. 348.Google Scholar

²⁷ See Richard B. Sheridan, “The British Credit Crisis of 1772 and the American Colonies,” The Journal of Economic History, XX (June 1960), 166–67. Callender (“Early Transportation,” p. 173) has said the debt was $28 million, which would have meant about 6 million pounds sterling. Thomas Jefferson said that Virginians “certainly owed two millions sterling to Great Britain at the conclusion of the war” (Boyd, Julian P., et al., eds., The Papers of Thomas Jefferson [Princeton: Princeton University Press, 1954], X, 27).Google Scholar A statement of the debts submitted to the British Government in February, 1791, by British merchants claimed a total of 4,984,655 pounds sterling was owed them by Americans (London, Public Record Office (Chatham Papers) 30/8/343/167, “List of Debts due by Citizens of the United States of America to the Merchants and Traders of Great Britain contracted previous to the year 1776 with interest on the same to January 1, 1790”). Since these claims (column 1) included 14 years' interest (which the document states that “as near as can at present be computed amounts to two millions and upwards”) some adjustment must be made if they are to approximate the amounts alleged by the British merchants to be the indebtedness at the beginning of 1776. Assuming that this interest was calculated at a rate of 5 percent (according to Sheridan, “British Credit Crisis,” pp. 166–67, this was the rate used) compounded annually, the 1776 debts would have been the amounts shown in column 2 (rounded to the nearest thousand pounds sterling): See tabulation on p. 253.

Interest on 2,500,000 pounds sterling would have been about 125,000 to 200,000 pounds sterling annually if the average rate of interest charged was from 5 to 8 percent. (Gipson, Lawrence H., “Virginia Planter Debts before the American Revolution,” The Virginia Magazine of History and Biography, LXIX (July 1961), 260,Google Scholar quotes one source which said Virginia planters were paying 8 percent on their debts. Curtis P. Nettels says planters “were obliged to pay an interest charge of 6 per cent or more …,” The Roots, p. 256). There are two reasons, however, why interest paid to British merchants was probably much less than this. First, it is very likely that the amounts claimed by British merchants were exaggerated (see Land, Aubrey C., “Economic Behavior in a Planting Society: The Eighteenth-Century Chesapeake,” The Journal of Southern History, XXXIII (Nov. 1967), 482–83).Google Scholar Second, explicit interest payments would not have been made on part—probably the largest part— of this debt since it was outstanding short-term trade credit granted by British merchants. The point is that the prices charged included normal interest costs under the usual credit terms (as stated above), and the values of imports estimated in Table 1 have been based upon the prices charged.

²⁸ White, The Beekmans, p. 254.

²⁹ We are indebted to Ronald Hansen for this point.

³⁰ See Bell, “West India Trade,” 283–85, for mention of indebtedness of West Indian merchants to North American colonists. We have no indication, however, of the yearly principal outstanding, so it is not possible to estimate interest earnings of colonial merchants in the West Indies.

³¹ Pares, Richard, Yankees and Creoles (Cambridge: Harvard University Press, 1956), pp. 81–82.Google Scholar

³² Bell, “West India Trade,” pp. 283 and 285.

³³ Donnan, Elizabeth; ed., Documents Illustrative of the History of the Slave Trade to America (Washington, D.C.: Carnegie institution of Washington, 1932), III, 215–16.Google Scholar

³⁴ James F. Shepherd, “Commodity Exports from the British North American Colonies to Overseas Areas, 1768–1772: Magnitudes and Patterns of Trade,” Explorations in Entrepreneurial History (forthcoming).

³⁵ Robertson, Ross M., The History of the American Economy (2d ed.; New York: Harcourt, Brace and World, Inc., 1964).Google Scholar

³⁶ According to Macpherson, David, Annals of Commerce (London, 1805), III, 570, annual tonnages (registered tons) built in the colonies were roughly 20.000 in 1769, 21,000 in 1770, and 24,000 in 1771, of which about two-thirds were built in New England. The value of this tonnage based upon building costs of 3 to 4 pounds sterling (see Walton, “A Measure of Productivity Change”) was approximately £70,000 per year, far below the yearly gross sales of shipping services. Value-added measures dp not exist for purposes of comparison.Google Scholar

³⁷ P. R. O. Customs 16/1; There are no estimates of the costs of the customs administration to Britain, but some have said the costs were greater than the customs revenues by sizable amounts. For example, Miller, John C., in Origins of the American Revolution (Stanford: Stanford University Press, 1959), p. 83,Google Scholar states: “To collect two thousand pounds in customs duties in the colonies cost the British government eight thousand pounds.” There were also other costs of administering the colonies, some of which were borne by the British Army. The army not only garrisoned the Indian country and enforced the boundary line, but it “bore the expenses of the Indian Department arid the cost of the civil governments of such ‘infant’ colonies as Nova Scotia, Georgia; and the Flbridas” (Sosin, Jack M.; Whitehall and the Wildernèss: The Middle West in British Colonial Pdlicy 1760–1775 [Lincoln: University of Nebraska Press; 1961], p; 79).Google Scholar Amounts for subsidizing these “infant” colonial governments and costs of “general surveys of his Majesty's dominions in North America” were £19,068 in 1768 and £18,797 in 1769 (Cobbett, William, The Parliamentary History of England [London, 1813;Google Scholar reprinted by the Johnson Reprint Corporation, 1966], XVI, pp. 413–418 and 621–626). (This source does not list the subsidies after 1769.) Salaries in the Northern Indian Department were said by General Gage in 1768 to be £3299 (Garter, Clarence E, ed.; the Correspondence of General Thomas Gage, 1763–1775 [New Haven: Yale University Press, 1933], II, 385–86).Google Scholar Earlier in the same year Gage had said that £3000 should be provided for occasional presents to the Indians and for various other expenses (Carter; Correspondence of Gage, p. 63). If the Southern Indian Department's costs were roughly of this magnitude, then expenditures for the Indian ‘departments would have been at least £12,000 per annum: According to Sosih (Whitehall and the Wilderness, p. 132); they were £20,000. Given only this fragmentary evidence; it is difficult to place any precise estimate on such expenditures by the British Government in the colonies; but it is probable that they were in the neighborhood of £30;000 to £40,000 or more, per annum for this period.

³⁸ The number of regiments and the estimates below of regimental strength and pay and subsistence of a regiment are taken from Burg, Maclyn P., “An Estimation of the Cost of Defending and Administering the Colonies of British North America; 1763:1775” (unpublished paper Department of Economics, University of Washington, 1966).Google Scholar Her information was based upon statements of regiment size, number; location, and pay scales in Curtis, Edward E., The Organization of the British Army in the American Revolution (New Haven: Yale University Press, 1926); pp. 3–5, 158;Google Scholar Carter, Correspondence of Gage; pp; 689–90; and Fortescue, JohnCorrespondence of George the Third (London: Macmillan and Co., 1928), III; 319;Google Scholar It is suggested that this estimated pay and subsistence of the army is only a lower bound to what British military expenditures in the colonies actually were. Note, for example, the statement by Sosin (Whitehall and the Wilderness, p. 132): “… on February 18 [1767], … the Secretary-at-War brought in the extraordinary expenditures of the American army for the previous year. They came to over £300,000. Thus, the annual cost of the American defense establishment to Great Britain was over £700,000 [in 1767]. Only a relatively small portion of this sum, £20,000, went to the Indian departments. The difficulty in supplying the remote forts in the interior accounted for most of the added expense.” The expenses of the military establishment probably did not remain this high over the period 1768–1772, however. “Since the ministers were acutely aware of the need for a reduction of the financial burden on Great Britain, they carried out a gradual withdrawal of the forces from the interior in the years from 1768 to 1774. Indeed, the mounting tensions which increased the need for troops on the turbulent seacoast accelerated the process” (Sosin, Whitehall and the Wilderness, p. 220).

³⁹ In 1763, there were 27 British naval vessels assigned to North American stations (London Magazine, XXXII [Dec. 1763], 674). In December 1774, there were 24 ships manned by 2835 men stationed in North America, and 30 ships and 3435 men in June 1775 (Barnes, G. R. and Owen, J. H, eds., The Private Papers of John, Earl of Sandwich, 1771–1782 (Greenwich, England: The Naval Records Society, 1932), I, 42.Google Scholar Also see Clark, William B., ed., Naval Documents of the American Revolution (Washington, D.C.: U.S. Government Printing Office, 1964), I, 1351.Google Scholar This estimate of the number of ships in the five years, 1768–1772, assumes that the number between 1763 and 1775 was approximately constant. The. estimated expenditures are taken from Burg, “Cost of Defending the Colonies,” and are based on evidence about the type of vessels and naval manning characteristics and pay published in Allen French, The First Year of the American Revolution (Boston: Houghton Mifflin Co., 1934), p. 345;Google ScholarBullocke, J. G., ed., The Tomlinson Papers (Greenwich, England: Naval Records Society, 1935), pp. 150–51;Google Scholar and A, Michael LewisSocial History of the Navy, 1793–1815 (London: George Allen and Unwin, Ltd., 1960), pp. 294–95, 300, and 304.Google Scholar

⁴⁰ The persistent complaints of colonists that there was never much specie circulating in the colonies because it was immediately shipped to England to pay debts there seem to point to the existence of balance-of-payments deficits with England. This, of course, is true; but it says nothing about the size of such deficits. A “specie drain” would exist even with a zero balance on the overall current account. All specie coming into the colonies from surplus areas would be shipped to pay debts in the deficit areas (assuming no domestic source of specie and none to begin with). These complaints of a “specie drain” miss the point. The complaints should have been voiced against the lack of a stable domestic money supply, which was the real problem (as was recognized by some colonists).

⁴¹ See the discussion of the relation between trade and development in Shepherd, “Commodity Exports.”

⁴² Historical Statistics, p. 757.

⁴³ The evidence presented by Nettels suggests relatively large naval expenditures in the colonies for the period, 1708–1711; see Nettels; The Momy Supply, p. 195.

⁴⁴ Historical Statistics p. 757.