An earlier draft of this article was presented at a Social Science Research Council conference at the University of Chicago. I thank Steve Krasner, David Laitin, John Ruggie, and Mark Zacher for their thoughtful comments. Steve Saideman provided valuable research assistance.

1. Equipment constitutes about $100 billion of this total. About $150 billion of the services market might be open to international competition by 1992. Market estimates are from Aronson, Jonathan D. and Cowhey, Peter F., When Countries Talk: International Trade in Telecommunications Services (Cambridge, Mass.: Ballinger, 1988), p. 7Google Scholar. Note that the study presented in this article excludes the market for television and radio programming.

2. The telecommunications regime reflects a central feature of the modern state: a rough division in the economic arrangements for goods and services. Services were at the core of the state's command over the economic infrastructure, which includes money, communications, transportation, the administration of laws, and the provision of health and education. Governments kept monopolies or enforced extensive controls over entry and pricing for these services while largely permitting competitive private arrangements for goods. Today, restrictions on services are being challenged because they constitute over 60 percent of the gross national product of industrialized nations.

3. Kratochwil, Friedrich and Ruggie, John Gerard, “International Organization: A State of the Art on an Art of the State,” International Organization 40 (Autumn 1986), pp. 753–76CrossRefGoogle Scholar.

4. This discussion modifies Haggard and Simmons' fine review of the literature by conflating two of their categories, functionalism and game-theoretic approaches. Both are examples of how strategic behavior influences the ability to achieve collective goods. See Haggard, Stephan and Simmons, Beth A., “Theories of International Regimes,” International Organization 41 (Summer 1987), pp. 491–517CrossRefGoogle Scholar.

5. See Haggard and Simmons, “Theories of International Regimes”; and Keohane, Robert O., “International Institutions: Two Approaches,” International Studies Quarterly 32 (12 1988), pp. 379–96CrossRefGoogle Scholar.

6. Firms in smaller countries may not share the belief of U.S. firms that successful regulatory initiative at home can trigger changes elsewhere. On firm-specific logic, see Milner, Helen V. and Yoffie, David B., “Between Free Trade and Protectionism: Strategic Trade Policy and a Theory of Corporate Trade Demands,” International Organization 43 (Spring 1989), pp. 239–73CrossRefGoogle Scholar.

7. Cowhey, Peter F., “‘States’ and ‘Politics’ in American Foreign Economic Policy,” in Odell, John and Willett, Thomas, eds., International Trade Policies: Gains from Exchange Between Economics and Political Science (Ann Arbor: University of Michigan Press, forthcoming)Google Scholar.

8. This interpretation follows the tradition of looking at power in particular issue-areas, since the relative explanatory force of the international distribution of power for predicting the regime in any given issue-area remains murky. While it makes sense to examine how particular regimes are “embedded” in broader dynamics of international power, this relationship is hard to define meaningfully. See Keohane, Robert and Nye, Joseph, Power and Interdependence (Boston: Little, Brown, 1977)Google Scholar; and Kratochwil and Ruggie, “International Organization.”

9. According to Gourevitch, a collapse of international market arrangements may also induce a crisis in domestic market arrangements, which in turn sets the stage for a new direction in the regime. See Gourevitch, Peter, Politics in Hard Times (Ithaca, N.Y.: Cornell University Press, 1986)Google Scholar. This was not the cause of change, however, in the case of the telecommunications regime.

10. Haas, Ernst B., “Why Collaborate? Issue-Linkage and International Regimes,” World Politics 32 (04 1980), pp. 357–402CrossRefGoogle Scholar.

11. This is one way to understand Ruggie's intriguing argument that epistemic communities interact with the changing division of labor and tasks assumed by the components of the international system. See Ruggie, John Gerard, “Continuity and Transformation in the World Polity: Toward a Neorealist Synthesis,” in Keohane, Robert O., ed., Neorealism and Its Critics (New York: Columbia University Press, 1986), pp. 131–57Google Scholar.

12. For a good example of how to link the choice of policy ideas in a theory of politicians' preferences, see Hansen, John Mark, “Creating a New Politics: The Evolution of an Agricultural Policy Network in Congress, 1919–1980,” Ph.D. diss., Yale University, 1988Google Scholar.

13. Snidal, Duncan, “Coordination Versus Prisoner's Dilemma: Implications for International Cooperation and Regimes,” American Political Science Review 79 (12 1985), pp. 923–42CrossRefGoogle Scholar.

14. Aronson, and Cowhey, , When Countries Talk, pp. 35–45Google Scholar.

15. The argument presented here is in agreement with Haggard and Simmons' view of the problems in the literature but reformulates their suggested research program set forth in “Theories of International Regimes”.

16. Young, Oran, “International Regimes: Toward a New Theory of Institutions,” World Politics 39 (10 1986), pp. 104–22CrossRefGoogle Scholar.

17. Ruggie, John Gerard, “International Responses to Technology: Concepts and Trends,” International Organization 29 (Summer 1975), pp. 557–84CrossRefGoogle Scholar.

18. The United States was always a bit of a deviant case. While it authorized a dominant monopolistic company, it kept a separate regulatory authority, never merged postal services with telecommunications, and allowed the dominant telephone company to own its equipment supplier. These differences ultimately made reforms easier to launch in the United States.

19. The key problems in the pioneering days of international telegraphic services arose because European countries made international messages subject to inspection and tariffs at each border. This led (literally) to building telegraph stations that spanned both sides of a national boundary. The wing in Germany would receive a telegram for France, write it down, take it through customs in the middle of the building, and hand it over to the French company on the other side of the station. In response, telegraph companies lobbied for international rules that would allow the uninterrupted flow of messages across two or more nations. Hence, the forerunner of the ITU liberalized commerce in communications services. But the bargain had a second dimension. It also reinforced highly profitable national monopolies by excluding competitive entrants. See Ende, Asher H., “International Communications,” Federal Communications Bar Journal, vol. 28, 1975, pp. 147–78Google Scholar.

20. Butler, Richard, “The International Telecommunications Union and Space Communications,” in Pelton, Joseph N. and Howkins, John, eds., Satellites International (New York: Stockton Press, 1987), pp. 33–38CrossRefGoogle Scholar.

21. See Jacobson, Harold K., “ITU: A Potpourri of Bureaucrats and Industrialists,” in Cox, Robert and Jacobson, Harold K., eds., The Anatomy of Influence(New Haven, Conn.: Yale University Press, 1973), pp. 59–101Google Scholar. Although this study did an excellent job of analyzing the processes at the ITU and noted that industrial interests were around to lobby, it showed no systematic understanding of the political economy of the regime. For two other standard works, see Blatherwick, David, The International Politics of Telecommunications(Berkeley, Calif.: Institute for International Studies, 1987)Google Scholar; and Codding, George A. Jr, and Rutkowski, Anthony M., The International Telecommunications Union in a Changing World (Dedham, Mass.: Artech House, 1982)Google Scholar.

22. For a typical view, see Weil, Gordon L., Communicating by Satellite: An International Discussion (New York: Carnegie Endowment for International Peace, 1969)Google Scholar.

23. See, for example, Ruggie, “International Responses to Technology.”

24. The United States did not withdraw its formal reservations from the international telephone conventions until 1973, when the ITU changed these rules into voluntary CCITT recommendations. See White, Rita Lauria and White, Harold M. Jr, The Law and Regulation of International Space Communication (Dedham, Mass.: Artech House, 1988), pp. 37–38Google Scholar. Nonetheless, the United States actively worked within the system. For example, according to AT&T's internal documents and interviews with the company's officials, AT&T abided by special legal reservations imposed by the Federal Communications Commission, but it largely viewed international services according to CCITT rules. AT&T did not view the ITU as a cartel, since the ITU did not set prices.

25. This principle reflected the broader view of the proper way to organize utilities, as discussed earlier in this article.

26. Krasner, Stephen D., “International Regimes for Global Communications: Technological Dynamism and Sovereign Control,” paper presented at the Conference on Utility Regulation, Stanford University, Stanford, Calif., 04 1988Google Scholar.

27. For example, in Chiron and Rehberg's astute analysis, the authors properly note that international communications services require the mutual consent of two or more nations and that the rules are therefore not set unilaterally. They mistakenly conclude that this makes telecommunications services unique. The same problem applies to obtaining mutual consent by two or more nations to any form of international trade. See Chiron, Stuart Z. and Rehberg, Lise A., “Fostering Competition in International Telecommunications,” Federal Communications Law Journal 38 (03 1986), pp. 1–58Google Scholar.

28. Aronson, and Cowhey, , When Countries Talk, pp. 8–9Google Scholar.

29. The CCITT had four rules implementing the norm. First, telephone rates for international calls would be set bilaterally between the carrier of the country of origin and that of the country of ultimate destination. Carriers from the two national monopolies could establish a single “accounting rate” for calls between their countries and then split the revenues roughly “fifty-fifty” (with a fee for any third countries through which the calls passed). Second, no matter which route a call actually followed between the two countries, the accounting rate would remain the same. Thus, unlike an airline traveler, a telecommunications carrier could not take a cheap route from New York to Brussels and then take a standard short hop to Frankfurt in order to save money over the expensive direct route from New York to Frankfurt. Third, PTTs should keep prices as low as possible and should relate them to costs. Although PTTs could cross-subsidize some services for social purposes, they were obligated to avoid “harmful” competition among services. PTTs cited this latter obligation when refusing discounts for largervolume users. Fourth, there could be no “resale” of capacity. This meant th at if a telephone circuit was leased to one party, that party could not resell its unused capacity to other parties.This prevented something akin to the airline charter business for telecommunications and hindered competition in enhanced services. See Aronson and Cowhey, When Countries Talk.

30. The literature on bargaining suggests that a country may be better off conforming to the model of domestic monopoly if other countries follow this model. Providers of monopolistic services may be better able to bear costs longer in a negotiation and to internalize costs, both of which are potential advantages in bilateral bargaining. The U.S. government corrected the problem by retaining the right of review if its competitive carriers were being whipsawed by a monopolistic PTT. See Chiron and Rehberg, “Fostering Competition”.

31. The creation of a separate Intelsat system, instead of minority foreign ownership in a U.S. venture or simple joint ownership of satellites between AT&T and foreign PTTs, was a function of U.S. domestic politics. See Cowhey, “‘States’ and ‘Politics’”; and Ende, “International Communications.”

32. For critical commentaries on the traditional theories of economies of scale and scope, see Leonard Waverman, “Competition and Regulation in the U.S. Interexchange Telecommunications Market,” and Crandall, Robert, “Fragmentation of the Telephone Network,” both in Crandall, R. and Flamm, K., eds., Changing the Rules: Technological Change, International Competition, and Regulation in Communications (Washington, D.C.: Brookings Institution, 1988), pp. 62–113 and 114–46Google Scholar.

33. Even within the EC, divergent and incompatible designs for selected new services were appearing in the 1970s.

34. It is commonplace to accept ISDN as a goal, but there are still bitter fights over what ISDN means for the design and competitiveness of telecommunications services. See Drake, William J. and McKnight, Lee, “Telecommunications Standards in the Global Information Economy,,” Project Promothee Perspectives, no. 5, 03 1988, pp. 21–26Google Scholar.

35. The ITU also reserves parts of the spectrum for different functional uses. Winning an allocation for signals of international communications satellites took four years and two ITU conferences.

36. The industrialized nations advocated regional multilateral meetings on an ad hoc basis to discuss “burden sharing” as a substitute for sole reliance on bilateral consultation. In 1973, they also agreed to a new article (Article 33) of the ITU Convention, which acknowledged a duty to discuss possible burdens on developing nations, but they worked to show how more simultaneous uses could be squeezed out of the spectrum. See Wijkman, Magnus, “Managing the Global Commons,” International Organization 36 (Summer 1982), pp. 511–36CrossRefGoogle Scholar.

37. Originally, the ITU allowed colonial powers to represent their possessions up to a limit of six votes.

38. The recommendations hold a lower legal status than do the regulations that supplement the ITU Convention, but the ITU Secretary General wrote that they were “almost universally followed”. See Butler, , “The International Telecommunications Union and Space Communications,” p. 35Google Scholar.

39. The ITU has undertaken several initiatives on development assistance, as its charter requires. Many are useful, but they are not crucial for the overall workings of the regime.

40. The United States has always insisted that it could only cooperate with an organization that allowed for RPOAs in order to accommodate the fact that AT&T was a private firm and that the United States had multiple private telex and telegram carriers.

41. Putnam, Robert D., “Diplomacy and Domestic Politics: The Logic of Two-Level Games,” International Organization 42 (Summer 1988), pp. 430–33CrossRefGoogle Scholar.

42. This is an extension of the argument that functionalist analysis is insufficiently political. A recent example of the problem is Blatherwick's The International Politics of Telecommunications.

43. For a discussion of this error, see March, James and Olson, Johan, “The New Institutionalism: Organizational Factors in Political Life,” American Political Science Review 78 (09 1984), pp. 734–50CrossRefGoogle Scholar.

44. Jacobson, , “ITU,” p. 84Google Scholar.

45. Ibid. The ITU Convention does in fact call for efficient service and low rates. See White, and White, , The Law and Regulation of International Space Communication, p. 66Google Scholar.

46. Companies have earned as much as $700,000 in profits from leasing a single international circuit that cost only $53,000 to provide, even with cost structures that are wildly inflated. See Stapley, Barry, “Managing Communications: The Value of Choice,” Telecommunications Policy 5 (06 1981), pp. 149–51CrossRefGoogle Scholar. This figure is not an aberration. In my examination of internal planning documents of major telecommunications companies, I found annual rates of return on international services for most PTTs to be at least 35–40 percent.

47. “Natural monopoly” as a dominant metaphor in telecommunications persisted without major controversy in a manner similar to the persistence of “naturalist realism” as a dominant motif in painting. As Bryson's brilliant analysis of “naturalist realism” notes, realism depends on the clever creation of a series of “signs” and “techniques” that are self-camouflaging so as to avoid pointing out the artificial construction of the “realist” vision. Painters have a rich agenda for paradigmatic (Bryson uses the word schematic) development of signs and techniques available to keep them and their viewers happy. But, just as vitally, successful naturalist painting depends on viewers who expect to see that type of construction of reality. In other words, “natural orders” depend on techniques that are self-effacing yet richly laden with developmental opportunities for engaged practitioners and “consumers” who have no incentive to challenge this manner of reconstructing the world. Much the same held true for the telecommunications regime. See Bryson, Norman, Vision and Painting (New Haven, Conn.: Yale University Press, 1983)Google Scholar.

48. Economists have long criticized telephone regulation as inefficient. Recent reexamination of the historical record shows that competition was possible far earlier than previously recognized. In the case of AT&T, for example, records show that the firm used its initial control of long distance technology to stifle competition at the local level, the exact opposite of how engineers traditionally ranked the opportunities for competition. See Garnett, Robert W., The Telephone Enterprise (Baltimore, Md.: Johns Hopkins University Press, 1985)Google Scholar. For a compelling theory of the evolution of networks, see Noam, Eli M., “The Public Telecommunications Network: A Concept in Transition,” Journal of Communication 37 (Winter 1987), pp. 30–48CrossRefGoogle Scholar.

49. Noam, “The Public Telecommunications Network.”

50. While global factors (such as the diffusion of new technology) may set the stage for widespread adoption of a broadly similar approach to public policies, it is a mistake to impute national change to the international system. In Politics in Hard Times, Gourevitch suggests that the intersection of global changes with local politics should produce broad policy movements globally, but the reason for adopting one particular response to the global change is largely domestic. Sometimes, as in telecommunications, a transnational coalition may reinforce the policy movement. See Keohane, Robert O. and Nye, Joseph S. Jr, “Power and Interdependence Revisited,” International Organization 41 (Autumn 1987), p. 748CrossRefGoogle Scholar; and Ruggie, John Gerard, “International Regimes, Transactions, and Change: Embedded Liberalism in the Postwar Economic Order,” International Organization 36 (Spring 1982), pp. 379–416CrossRefGoogle Scholar.

51. In “International Regimes for Global Communications,” Krasner noted the Marconi case as proof for the durability of “sovereignty norms,” but he ignored the question of when and in which form such claims of sovereignty are politically sustainable. The confusion about sovereignty in the literature on regimes probably has its roots in the powerful influence of neofunctionalism. In The Uniting of Europe (Stanford, Calif.: Stanford University Press, 1958)Google Scholar, Ernst Haas explored the problem of when legitimacy and authority would move from the state to international institutions (the question of transference of sovereignty). But regime analysts require no such transfer of sovereignty to look for variations in the powers delegated to regime institutions. Moreover, if sovereignty is a constant, as regime analysts assume, then invoking sovereignty as a norm is hardly an explanation for state behavior.

52. In the case of the Marconi Company, conferences in 1903 and 1906 passed ineffectual regulations to indirectly curb its monopolistic practices. After several well-publicized tragedies at sea (including the loss of the Titanic) were made worse by radio communications problems and after an insider stock trading scandal involving the Marconi Company and Lloyd George occurred, Britain capitulated on the interconnection issue at an international conference in 1912. See Archer, Gleason L., History of Radio to 1926 (New York: American Historical Society, 1938)Google Scholar.

53. The 1912 international conference did not end the issue of monopolies. It was only resolved by the cartel agreement of 1922. The equipment cartel was supplemented by a cartel agreement on radiotelegraphy service territories. See Tomlinson, John, The International Control of Radio-Communications (1938Google Scholar; reprint, New York: Kraus Reprint Company, 1972); and Jolly, W. P., Marconi (New York: Stein & Day, 1972)Google Scholar. Ronald Coase made the classic case for a frequency auction in “The Federal Communications Commission,” Journal of Law and Economics 2 (10 1959), pp. 1–40CrossRefGoogle Scholar.

54. Oslun, Robert John, “Communications Satellite Policy: Revolutionary or Evolutionary?” Ph.D. diss., American University, Washington, D.C., 1975Google Scholar.

55. The estimates are based on interview data and include indirect savings from more efficient configuration of future computer systems.

56. Computer networks often demanded capabilities that telephone companies could not provide in the 1970s. Recommendation D-6 of the ITU essentially allowed for independent suppliers of the network services most closely related to computer operations, but the PTTs tried to sharply curtail the scope of the exemption.

57. After liberalization, companies such as Sony and Kyocera became significant investors in new telephone firms in Japan.

58. Many new U.S. firms expect to become significant exporters in entirely new equipment lines opened up by competition. U.S. firms are the emerging world leaders in wide area networks and transmission multiplexes, which soon will have annual markets of $1 billion, according to interviews with company officials in 1989.

59. Political leadership has always worried about the impact of competition on the price of local service because it is the most visible cost for the local voter. Moreover, immediate price increases are memorable, while subtle subsequent gains on pricing get lost. Consider the grumbling in the United States. Local services have climbed by roughly 50 percent. Nevertheless, when the lower cost of telephone equipment, the reduction in long distance service costs, and general inflation are taken into account, the average household's real cost for telephone services is probably slightly lower now than it was before divestiture. The upper middle class household, which votes in a far larger percentage, has definitely gained. Data on service and equipment costs are from the internal documents of the New York State Public Service Commission. Estimates of the share of GNP commanded by the industry are from “Rewiring the World,” The Economist, 17 October 1987, p. 4.

60. According to data from the New York State Public Service Commission and various interviews, the cost of the network equipment has dropped significantly owing to curtailed cross-subsidies in both the United States and Japan since the introduction of competition. Productivity estimates are from the Communications Workers of America, Information Industry Report, Washington, D.C., 12 1987Google Scholar.

61. According to interviews, this is the conclusion of internal corporate studies of several telecommunications carriers.

62. Even more acute is the impact of geographic voting clusters on policy. The most expensive subsidies are for rural areas, areas which are overrepresented in the apportionment of legislative representation in many countries. The cross-subsidies reduce the overall efficiency of pricing, and because regulators seek revenues for cross-subsidies from services that are least pertinent to voters, they often tap international services.

63. The European Commission has also directed that this be the general rule for the EC by 1992.

64. The EC rules on local content intentionally resemble the “Buy America” policy of the United States, according to interviews with EC officials in Brussels in July 1989. The proposal indeed resembles the logic of strategic trade philosophy that Krasner has correctly endorsed for this market, although Krasner does not sufficiently note that it would also constitute a major liberalization of the market. See Krasner, Stephen D., Asymmetries in Japanese-American Trade: The Case for Specific Reciprocity (Berkeley, Calif.: Institute for International Studies, 1987)Google Scholar.

65. All three countries still face problems on this score because they introduced elaborate methods to achieve some special support for new entrants in the basic services market while also retaining some cross-subsidies for households.

66. Information is based on interviews with EC officials in Brussels in 1989.

67. Even the U.S. companies that were most critical of the outcome of the ITU World Administrative Telephone and Telegraph Conference saw no significant threat to their competitive strategies, according to interviews with officials of several data communications companies in May and June 1989.

68. Fora history of bilateral talks, see Aronson, Jonathan D. and Cowhey, Peter F., “Bilateral Telecommunications Negotiations,” in Bressand, Albert and Nicolaidis, Kalypso, eds., International Services Yearbook (Cambridge, Mass.: Ballinger, 1989)Google Scholar.

69. For a discussion of grievance processes as a substitute for direct legislative decision, see Cowhey, “‘States’ and ‘Politics’”. “Market access” implies that all barriers to effective global access to agreed upon service markets are subject to trade review. See Cowhey, Peter F., “Telecommunications and Foreign Economic Policy,” in Newburg, Paula, ed., New Directions in Communications Policy, vol. 2 (Durham, N.C.: Duke University Press, 1989)Google Scholar.

70. In the short run, for example, it appears that the United States lost significantly on CPE issues as a result of increased imports. But in the long run, the computer industry appears to have been strengthened by making the U.S. approach to networking minicomputers and work stations with low cost and flexible communications systems into the new global computing standard.

71. Huber, Peter W., The Geodesic Network: 1987 Report on Competition in the Telephone Industry (Washington, D.C.: Antitrust Division of the Department of Justice, 1987), p. 1.29Google Scholar.

72. See Farrel, Joseph and Saloner, Garth, “Standardization, Compatibility, and Innovation,” Rand Journal of Economics 16 (Spring 1985), pp. 70–83CrossRefGoogle Scholar; Farrel, Joseph and Saloner, Garth, “Installed Base and Compatibility: Innovation, Product Preannouncements, and Predation,” American Economic Review 76 (12 1986), pp. 940–55Google Scholar; and Stanley Besen and Garth Saloner, “Compatibility Standards and the Market for Telecommunications Standards,” in Crandall, and Flamm, , Changing the Rules, pp. 177–220Google Scholar.

73. For a similar conclusion about a more extreme version of this game in regard to currency blocks during the Great Depression, see Oye, Kenneth A., “The Sterling-Dollar-Franc Triangle: Monetary Diplomacy, 1929–1937,” in Oye, Kenneth A., ed., Cooperation Under Anarchy (Princeton, N.J.: Princeton University Press, 1985), pp. 173–99Google Scholar.