I acknowledge Larry Neal, Jeremy Atack, Tom Ulen, Hugh Aitken, Michael Bernstein, David Wheelock, the editors of this JOURNAL, and the participants of the University of Massachusetts Economic History and Economic Development workshop for their helpful comments on earlier versions of this article. Also, I thank the firms of Krupp, Gutehoffnungshütte, Thyssen, Marine et Homécourt, and Cockerill, and all the people in each who made my work so enjoyable. Finally, I gratefully acknowledge the Social Science Research Council jointly with the American Council of Learned Societies, the Ford Foundation, and the William and Flora Hewlett Foundation, for funding my research.Google Scholar

¹ The Versailles Treaty forced German firms to sell off 74.5 percent of their iron ore holdings, 26.6 percent of their blast furnaces, and 19.2 percent of their rolling mills. The treaty also disbanded the German Customs Union by removing Luxemburg, the Lorraine, and the Saar from German cartels, and further weakened Germany's position by eliminating tariffs on products exported to Germany from these areas.Google Scholar

² Svennilson, lngvar, Growth and Stagnation in the European Economy (Geneva, 1954), p. 124.Google Scholar

³ Ibid., p. 130.

⁴ Hexner, Emil, The International Steel Cartel (Chapel Hill, 1943), p. 174.Google Scholar

⁵ For an analysis of the classic Marshallian rules, see Hicks, J. R., The Theory of Wages (London, 1935), p. 242.Google Scholar

⁶ This is not the first international agreement between these countries. Already in 1907 there had been a short-lived cartel for the export trade of structural shapes. Germany had a 73.45 percent quota, Belgium and Luxemburg had 15.05 percent, and France had the remaining 11.5 percent (Archives de l'sÉtat de Liège, Cockerill 128).Google Scholar

⁷ The quotas were originally for aggregate annual production of 25,278,000 tons: Germany 40.45 percent, France 31.89 percent, Belgium 12.57 percent, Luxemburg 8.55 percent, and the Saar 6.54 percent. If the annual agreed-upon production increased, then the German relative share would rise, while the other members's quotas would fall. The quota increases are based on million ton increases up to a total tonnage of 29,278,000; at this value the German quota reaches its peak at 43.18 percent. See article 4 of the Cartel Agreement, published in Hexner, The International Steel Cartel, appendix IIIA.Google Scholar

⁸ Notes from the Executive Committee Meeting of the ISC. June 8 and 9, 1927 (Krupp Historisches Archiv, WA77/V 1027); and Report from the 38th Executive Committee Meeting of the Rohstahlgemeinschaft, May 24, 1928 (Krupp Historisches Archiv, WA77/V960).Google Scholar

⁹ For an overquota production between 7.5 and 28 percent they paid $1; between 28 and 30.8 percent $2; and beyond 30.8 percent they had to pay the full $4.Google Scholar

¹⁰ In Jan. 1927 Austria, Hungary, and Czechoslovakia formed the Central European Group, which had a combined quota of 7.272 percent of the ISC members's production, over two-thirds of which belonged to Czechoslovakia. Poland was not a member of the group, but did have a penetration agreement with the ISC, which limited the trade from ISC countries into Poland. See Hexner, The International Steel Cartel, p. 73. The basic sliding-scale penalties were: for production up to 7.5 percent over quota the penalty was only $1; for overquota production from 7.5 to 10 percent the penalty was $2, and beyond 10 percent it went to the original $4. Appendix to a letter from Meyer, M. to Laurent, T. on Aug. 4, 1928 (Archives Nationales, 139AQ102).Google Scholar

¹¹ Extract taken from the notes of the second meeting of the “Small” Commission on July 26, 1929 (Krupp Hisrorisches Archiv, WA77/V962).Google Scholar

¹² The domestic cartel would have even benefited from the import quota if a tariff was also charged. For a detailed analysis of this, see Bhagwati, Jagdish, “On the equivalence of tariffs and quotas,” in Baldwin, Robert E., ed., Trade, Growth, and the Balance of Payments (Chicago, 1965), pp. 53–68.Google Scholar For a historical perspective of the interrelation between tariffs and iron and steel cartels in Germany, see Webb, Steven B., “Tariffs, Cartels, Technology, and Growth in the German Steel Industry, 1879 to 1914,” this Journal, 40 (06 1980), pp. 309–30.Google Scholar

¹³ Minutes of the Meeting of the Executive Committee of the International Cartels, Aug. 6, 1930 (Gutehoffnungshütte Historisches Archly, 40000090/8).Google Scholar

¹⁴ Desgranges, Paul, Le Comptoir Sidérurgique de la France (Paris, 1976), p. 74.Google Scholar

¹⁵ Svennilson, Growth and Stagnation, p. 129.Google Scholar

¹⁶ Convention concernant le contingent d'sacier Lorraine et Luxemburg à importer en Allemagne, June 11, 1934 (Archives Nationales, 139AQ135[2]).Google Scholar

¹⁷ The breakdown of the French export limits in tonnage were 80,000 tons of semi-finished goods;101,000 tons of merchant bars; 11,200 tons of thick plates; and 51,000 tons of structural shapes. These totals were distributed among the Lorraine firms in fixed percentages (Archives Nauionales, 139AQ135[2]).Google Scholar

¹⁸ Ulen, Thomas S., “A Critique of Cartel Theory” (College of Commerce and Business Faculty Working Paper No. 599, University of Illinois, Champaign).Google Scholar

¹⁹ Hexner, The International Steel Cartel, p. 24.Google Scholar

²⁰ The German steel industry had a long history of cooperation and even governmental support. Even during the Weimar Republic, the steel industry, while constrained by new labor laws, was very powerful.Google Scholar

²¹ For an analysis of this, see Barbezat, Daniel, “The AVI and the German Steel Cartels 1925–1939” (paper presented at the 28th Annual Cliometric Conference, May 1988).Google Scholar

²² These figures are from the Comité des Forges, quoted in “Historique des accords Sarrois et du contingent Lorraine-Luxembourgeois,” June 11, 1934 (Archives Narionales, 139AQ135[2]). Figures commonly quoted for France in 1913, for example, in Svennilson, are the sum of the French production and the Lorraine production.Google Scholar

²³ Taken from the Yearly Report of the Vereinigte Stahiwerke (Duisburg, 1932), p. 203.Google Scholar

²⁴ Although neither France nor Germany had a strong antitrust movement, powered by an upwardly mobile lower middle class, the German steel industry was better able to unite and even work with the government on steel trade issues. For an excellent review of the French frustrations in organizing over the 1920s, see Maier, Charles S., Recasting Bourgeois Europe (Princeton, 1975), pp. 516–45.Google Scholar

²⁵ See Caron, François and Bouvier, Jean, “Structure des firmes, emprise de l'É,” in Braudel, Fernand and Labrousse, Ernest, eds., Histoire économique et sociale de la France (Paris, 1980), vol. 4, part 2, esp. pp. 789–93.Google Scholar

²⁶ Ibid., pp. 789–91.

²⁷ Lévy-Leboyer, Maurice, “Innovations and Business Strategies in Nineteenth and Twentieth Century France,” in Carter, Edward C. II, Forster, Robert, and Moody, Joseph N., eds., Enterprise and Entrepreneurs in 19th and 20th Century France (Baltimore, 1976), p. 122.Google Scholar I do not mean to suggest, however, that the French were not committed to making rational organizations. In stating that the French cartels were no match for the organization in German cartels, one must not, as William Parker has pointed out, “attribute to a difference either in rapacity or intelligence what is in fact simply a difference between land animals and underwater creatures.” Pounds, Norman J. G. and Parker, William N., Coal and Steel in Western Europe (Bloomington, 1957), p. 336.Google Scholar

²⁸ Report on the Economic Situation of Belgium 1926, p. 61.Google Scholar

²⁹ The large fall in German market share in the fourth quarter of 1928 is the effect of the lockout in the iron and steel industry in Nov. 1928.Google Scholar

³⁰ Report on the Meeting of the A-Product Cartel, Jan. 31, 1929 (Gutehoffnungshütte Historisches Archly, 40000010/4b).Google Scholar

³¹ The pig iron arrangement even fixed prices and was renewed in 1929 without undue problems. Economic and Trade Conditions in Belgium 1929, report prepared by N.S. Reyntiens for the Department of Overseas Trade.Google Scholar

³² Griffin, James M., “OPEC Behavior: A Test of Alternative Hypotheses,” American Economic Review, 75 (12. 1985) pp. 954–63.Google Scholar

³³ For convenience I have dropped the subscript “t” from each of the variables. Also, the regression was run with Nov. 1928 dropped, the month of the large German general strike.Google Scholar

³⁴ The data are the differences of logs and so represent percentage changes.Google Scholar

³⁵ This is done by nothing that P _x = β_x (T – _P) and that P_∞ = β_∞ (T – P_f), then rewriting the second equation and solving for T, we get: T = [1 + β_∞/β∞_P∞ Now we can substitute this equation into the first equation for P _x and solve for the relationship between the Germany and France, as presented in the text.Google Scholar