The author would like to acknowledge, with thanks but without implicating them in the results, the helpful suggestions of H. A. J. Green, D. F. Gordon, W. Haque, J. R. Huber, and S. R. Lewis, Jr. Computations were performed at the computing centers of Yale University and the University of the Philippines, with the financial support of the Williams College-AID Research Project on Import Substitution and the University of the Philippines—University of Wisconsin Training Program in Development Economics.

¹ For expositional simplicity, this paper will confine itself to a two-sector model. The full amount of the “tax” does not accrue to the favored sector, of course, since there is some deadweight loss resulting from a less efficient allocation of resources.

² An assessment of the reliability of the data can be found in Bokor, Gustav, Geschichte und Organisation der amtlichen Statistik in Ungarn [History and Organization of the Official Statistics in Hungary] (Budapest: 1896), esp. pp. 69, 137, 183, 186–88. Because of changes in some commodity categories between 1882 and 1883, the latter has been used as the initial year throughout this paper.Google Scholar

³ Kuehne, Roland, Die Geschichte des ungarischen Getreidehandels und die Getreidepreisbildung in Oesterreich-Ungarn [ The History of the Hungarian Grain Trade and Grain Price Formation in Austria-Hungary] (Magyaróvár: Dissertation from the Royal Hungarian Agricultural College, 1911), p. 4.Google Scholar

⁴ See especially Grunzel, Josef, Handelspolitik und Ausgleich in Oesterreich-Ungarn [Commercial Policy and “Compromise” in Austria-Hungary] (Vienna and Leipzig: Duncker and Humblot, 1912)Google Scholar, and Mauekovits, Alexander von, “Die handelspolitischen Interessen Ungarns,” [The Commercial Policy Interests of Hungary] in Beitraege zur neuesten Handelspolitik Oesterreichs [Contributions toward the Newest Commercial Policy of Austria] (Leipzig: Schriften des Vereins fuer Socialpolitik [Papers of the Association for Social Policy] 1901) XVIII.Google Scholar

⁵ Since the Hungarian government was almost completely dominated by the rural magnates and the landed gentry, there was very close correspondence between national policy and the interests of this group. See Jászi, Oscar, The Dissolution of the Habsburg Monarchy (Chicago: University of Chicago Press, 1929), esp. part IV, ch. ii, “Morbus Latifundii.”Google Scholar

⁶ The partners can, however, raise the price of imported goods to their consumers via tariffs or other trade restrictions. They are therefore not price takers in the very strictest sense.

⁷ An index-number problem arises if there is more than one A- or M-good. This problem will be discussed in later paragraphs. It should be noted here also that the loss as calculated represents only a part of the burden of the biased terms of trade: since the prices of goods produced and traded strictly within country A are also affected, there is a larger cost (assuming exports are less than half of total A-goods production) which is unaccounted for by the procedure used.

⁸ Assume balanced trade, so that M = TA. Then

⁹ Nerlove, Marc, The Dynamics of Supply: Estimation of Farmers' Response to Price (Baltimore: The Johns Hopkins University Studies in Historical and Political Science, 1958), series LXXVI, no. 2, esp. pp. 25–26 and 62–65.Google Scholar

¹⁰ See for example Krishna, Raj, “Farm-Supply Response in India-Pakistan: The Case of the Punjab Region,” Economic Journal, LXXIII (Sept., 1963), 477–87CrossRefGoogle Scholar; Falcon, Walter P., “Farmer Response to Price in a Subsistence Economy: The Case of West Pakistan,” American Economic Review, LIV (May, 1964), 580–91Google Scholar; Bateman, Merrill J., “Aggregate and Regional Supply Functions for Ghanaian Cocoa,” Journal of Farm Economics, XLVII (May, 1965), 384–401CrossRefGoogle Scholar; Wharton, Clifton R. Jr, “Malayan Rubber Supply Conditions,” in The Political Economy of Independent Malaya, Silcock, T. H. ed. (Canberra: Australian National University Press, 1963), 131–62Google Scholar; or Eddie, S. M., “Farmer Response to Price in Large-Estate Agriculture,” Economic History Review, 2nd ser., XXIV (Nov., 1971), 571–88.Google Scholar A very useful and lucid exposition of the development and properties of distributed-lag models can be found in Wallis, Kenneth F., “Some Recent Developments in Applied Econometrics: Dynamic Models and Simultaneous Equation Systems,” Journal of Economic Literature, VII (Sept., 1969), 771–96.Google Scholar

¹¹ A problem of autocorrelation of the residuals can arise from the form of the model used. Although there is some hesitation in using ordinary least squares regression technique under such circumstances, we follow here the typical practice of using least squares and presenting its results with reservations. The problems in so doing are summarized in Wallis, 773–75.

¹² Meade, James Edward, A Geometry of International Trade (London: George Allen and Unwin, 1952), pp. 14–16 and Figure II.Google Scholar

¹³ If this were not the case, and we assumed instead that free trade must be balanced trade, it is of course possible that the A-producer would prefer point X'—at less favorable prices but with an import surplus—over the balanced free-trade point, N, which is found by drawing a price line OP_W through the origin and parallel to ST_W. Point X' lies on a higher trade-indifference curve than does point N. The total net import surplus for Hungary during the entire period 1883–1913 amounted to 653,000 crowns (uncorrected for price changes), which represents only about 1.7 percent of total imports (again uncorrected for price changes). Thus the assumptions made about the origin and character of the import surplus do not seem unreasonable; in any case, a surplus so small in relation to total trade is not likely to be very significant.

¹⁴ Since we have assumed country A small in relation to total world trade, the two terms of trade lines also represent the world offer curves of goods to country A (T_W if there is free trade, T_D if tariffs are imposed). Only in the unlikely case that offer curves, O_A and O_M, intersect at a point above the fine T_W would these terms of trade lines not be the effective offer curves of M-goods to country A. Since there was sizable trade with the world in every year of the period, we can rule out this possibility.

¹⁵ Since the analysis is symmetric, this procedure will also reveal any gains, if the domestic terms of trade are actually skewed to favor the agricultural producer.

¹⁶ The terms of trade used in the calculations leading to Table 4 were computed on the assumption that the exporters were specialized in production, but were generalized in consumption. Thus in each case the terms of trade are the ratio of the prices of the group of products being considered to the prices of all imports.

¹⁷ Sándor, P., “Die Agrarkrise am Ende des 19. Jahrhunderts und der Grossgrundbesitz in Ungarn” [The Agrarian Crisis at the End of the 19th Century and the Large Estates in Hungary], in Studien zur Geschichte der Oesterriechisch-ungarischen Monarchie [Studies in the History of the Austro-Hungarian Monarchy], ed. Sándor, V. and Hanak, P. (Budapest: Studia Historica Academiae Scientiarum Hungaricae, 1961), no. 51, pp. 181–82.Google Scholar

¹⁸ Eddie, Scott M., “Agricultural Production and Output per Worker in Hungary, 1870–1913,” The Journal of Economic History, XXVIII (June, 1968), pp. 216–18.Google Scholar

¹⁹ Ibid., 218.