¹ Hirschman, , Exit, Voice, and Loyalty (Cambridge: Harvard University Press, 1970).Google Scholar

² Pizzomo, Alessandro, “Political Exchange and Collective Identity in Industrial Conflict,” in Crouch, Colin and Pizzorno, Alessandro, eds., The Resurgence of Class Conflict in Western Europe since 1968 (New York: Holmes and Meier, 1978)Google Scholar; Collier, Ruth Berins and Collier, David, Shaping the Political Arena (Princeton: Princeton University Press, 1991).Google Scholar

³ Esping-Andersen, Gosta, Politics against Markets (Princeton: Princeton University Press, 1985)Google Scholar; Przeworski, Adam and Sprague, John, Paper Stones (Chicago: University of Chicago Press, 1986).Google Scholar

⁴ Krasner, , “Regimes and the Limits of Realism,” in Krasner, Stephen D., ed., International Regimes (Ithaca, N.Y.: Cornell University Press, 1983), 2.Google Scholar

⁵ Although rules and procedures are also important, I argue that they can be subsumed under norms.

⁶ See, e.g., Grodzins, Morton, The Loyal and the Disloyal (Chicago: University of Chicago Press, 1956), 5Google Scholar; Guetzkow, Harold, Multiple Loyalties: Theoretical Approach to a Problem in International Organization (Princeton: Center for Research on World Political Institutions, Princeton University, 1955), 8Google Scholar; Shain, Yossi, The Frontier of Loyalty (Middletown, Conn.: Wesleyan Unversity Press, 1989), 22Google Scholar; and Withey, Michael J. and Cooper, William H., “Predicting Exit, Voice, Loyalty, and Neglect,” Administrative Science Quarterly 34, no. 4 (1989), 522.CrossRefGoogle Scholar

⁷ Dasgupta, Partha, “Trust as a Commodity,” in Gambetta, Diego, ed., Trust: Making and Breaking Cooperative Relations (Oxford: Basil Blackwell, 1988), 56.Google Scholar For a related and intriguing discussion of how trust mediates the perceptions and responses of employees regarding the breach of psychological contracts by their employers, see Robinson, Sandra L., “Trust and Breach of the Psychological Contract,” Administrative Science Quarterly 41, no. 4 (1996).CrossRefGoogle Scholar

⁸ Schaar, , “Loyalty,” in International Encyclopedia of the Social Sciences, vol. 9 (New York: Macmillan Company and the Free Press, 1968), 485.Google Scholar

⁹ Pérez-Díaz, , “Unions' Uncertainties and Workers' Ambivalence,” International Journal of Political Economy 17 (Fall 1987), 120.Google Scholar

¹⁰ Regini, Mario, “Industrial Relations in the Phase of ‘Flexibility,’” International Journal of Political Economy 17 (Fall 1987), 92.Google Scholar

¹¹ Hirschman (fn. 1), 30.

¹² Ibid., 40.

¹³ Stinchcombe, Arthur L., Constructing Social Theories (New York: Harcourt, Brace and World, 1968), 123.Google Scholar

¹⁴ Although the line between norm-based voice and norm-breaking voice depends on the norms that have emerged in each party-union alliance, a common example is the distinction between issuing threats to hold a general strike (norm-based voice) and actually mobilizing a general strike (normbreaking voice) when the party is in government.

¹⁵ For a detailed discussion of the response of affiliated unions to the party's economic reforms in each case, see Burgess, Katrina, “Alliances under Stress: Economic Reform and Party-Union Relations in Mexico, Spain, and Venezuela” (Ph.D. diss., Princeton University, 1998).Google Scholar

¹⁶ An eloquent example of this return to norm-based voice was the decision by AD's labor delegation to vote with the party to reject a motion of censure against Pérez's cabinet in March 1992, despite the CTV's strong criticism of the cabinet and its policies. ElNacional, March 27, 1992.

¹⁷ Although the catalyst for the 1988 general strike was a Youth Employment Plan to allow temporary contracts for workers under twenty-five years of age, the UGT demanded a fundamental reorientation in the governmentés economic program. For a thorough discussion of the UGTés demands, see the articles in Claridad 29–30 (1989).

¹⁸ Exit did not mean that UGT leaders stopped negotiating with the PSOE, however. Since workers and unions could not avoid the consequences of the party's policies even after the collapse of the alliance, they continued to exercise “voice after exit.” See Hirschman (fn. 1), 101,104.

¹⁹ Hirschman (fn. 1), 78. Note that loyalty is not equivalent to silence. A decision to remain with a product or organization does not necessarily lead to the exercise of voice. Instead, an actor may go along “with a deteriorating status quo without engaging in pressure tactics to reverse it. Barry, Brian, “Review Article: Exit, Voice, and Loyalty,” British Journal of Political Science 4, no. 1 (1974), 95.CrossRefGoogle Scholar But rather than expanding the menu of options to include “silent non-exit,” many scholars mistakenly use loyalty in its stead. This move creates a dichotomy between loyalty and voice that is inconsistent with Hirschman's model.

²⁰ Barry (fn. 19), 95.

²¹ Ibid., 96.

²² Hirschman (fn. 1), 79, 91. Barry recognizes this solution when he writes that “[w]hat I think [Hirschman] wants to say here is that the experience of having had influence in a collectivity and the expectation of having influence in the future tend to increase a person's commitment to that collectivity, in other words loyalty to it in the ordinary meaning of ‘loyalty.’” Barry (fn. 19), 99.

²¹ Hirschman, Albert O., Rival Views of Market Society and Other Recent Essays (New York: Viking, 1986), 81.Google Scholar

²⁴ As suggested above, disaffected labor leaders may continue to adhere to alliance norms even after their reserves of loyalty toward the party have been depleted, especially if the party has the capacity to punish them for disloyalty. At this point, however, loyal behavior is no longer motivated by loyal attitude and thus becomes highly sensitive to short-term calculations of interest and power.

²⁵ For a similar discussion of labor leaders who are pulled in strategically contradictory directions by the policies of labor-backed parties in government, see Gourevitch, Peter et al., Unions and Economic Crisis: Britain, West Germany, and Sweden (London: George Allen and Unwin, 1984), 11.Google Scholar

²⁶ I recognize that these loyalty dilemmas may not set in immediately when a labor-backed government adopts painful reforms. Like labor leaders, workers are likely to have reserves of loyalty that allow the unions to cooperate with the reforms without being viewed as disloyal. Worker patience will wear thin, however, if the governing party appears to be betraying its principles and/or the union appears to be selling out to the reformers. Moreover, as discussed above, reserves of worker loyalty may already be quite shallow as a result of the secular decline in class consciousness and collective identities.

²⁷ For an excellent discussion of the impact of union competition on labor's responses to market reform, see Murillo, Maria Victoria, “From Populism to Neoliberalism: Labor Unions and Market-Oriented Reforms in Argentina, Mexico, and Venezuela” (Ph.D. diss., Harvard University, 1997).Google Scholar

²⁸ Factors such as economic conditions, international pressures, ideological proclivities, internal party politics, and electoral dynamics are likely to shape the propensity of autonomous party leaders to challenge their own executive.

²⁹ Middlebrook, Kevin, The Paradox of Revolution (Baltimore: Johns Hopkins University Press, 1995), 28.Google Scholar

³⁰ Mexican presidents (and all other elected officials) are constitutionally prohibited from running for reelection.

³¹ Casanova, Pablo González, Elestadoy los partidos políticos en México (Mexico, D.F.: Ediciones Era, 1981), 122–29Google Scholar; Pacheco, Guadalupe, “La XIV Asamblea Nacional del PRI,” Estudios Poltticos 3a (1991), 77–78.Google Scholar

³² Middlebrook (fn. 29), 66–67.

³³ It should be recognized, however, that the CTM's crisis of legitimacy in the 1940s was ameliorated in the 1950s and 1960s when the PRl's strategy of “stabilizing development” enabled the CTM to deliver positive gains in real wages, nonwage benefits, and employment to its members. See Middlebrook (fn. 29), 120–22.

³⁴ For a discussion of these tactics, see García, Javier Aguilar and Arrieta, Lorenzo, “En la fase mas aguda de la crisis y en el inicio de la reestructuración o modemización,” in García, Javier Aguilar, ed., Historia de la CTM, 1936–1990, vol. 2 (México, D.F.: Institute de Investigaciones Sociales, UNAM, 1990)Google Scholar; and , María Xelhuantzi López, “Reforma del Estado Mexicano y sindicalismo” (M.A. thesis, Universidad Nacional Autónoma de México, 1992).Google Scholar

³⁵ In addition, several of these benefits proved short-lived or illusory. Many of the CTM-owned businesses established in the mid-1980s were later dismantled as part of the PRIés privatization program. Meanwhile, the economic cabinet set wage levels and and then delivered them to the CTM for ratification. Although real wages made a partial recovery in the early 1990s, the real minimum wage continued its precipitous decline, and real wages failed to keep pace with productivity gains and in 1992 were still at only 61 percent of their 1980 value. Villanueva, Carlos García and Stoll, Alfred, “Las relaciones laborales en México y su contexto ecónomico y social,” in Wannöffel, Manfred, ed., Ruptura en las relaciones laborales (México, D.F.: Fundación Friedrich Ebert, 1995), 221–24.Google Scholar

³⁶ Quoted in El Financiero, May 7,1993. Another CTM official drew an explicit parallel between the reaccommodation in the 1990s and the shift by President Luis Echeverria in the 1970s from support for independent unionism to renewal of the alliance with the CTM. Confidential author interview, Mexico City, November 23,1994.

³⁷ PSOE statutes required members of the socialist group in parliament to adhere to party discipline regarding their legislative action and votes. PSOE, Anexo del acta del 29 Congreso (1981), 52. The party leadership also deducted stiff fines from the paychecks of deputies who missed parliamentary sessions and committee meetings, thereby discouraging abstention as a form of silent protest against the government. Share, Donald, Dilemmas of Social Democracy (New York: Greenwood Press, 1989), 125.Google Scholar

³⁸ Donaghy, Peter J. and Newton, Michael T., Spain: A Guide to Political and Economic Institutions (Cambridge: Cambridge University Press, 1987), 66.Google Scholar

³⁹ Bar, Antonio, “Spain,” in Blondel, Jean and Müller-Rommel, Ferdinand, eds., Cabinets in Western Europe (London: Macmillan, 1988), 112.Google Scholar

⁴⁰ Share (fn. 37), 56.

⁴¹ Gillespie, Richard, The Spanish Socialist Party (Oxford: Clarendon Press, 1989), 346–47.Google Scholar

⁴² Escobar, Modesto, Works or Union Councils? The Representative System in Medium and Large Sized Spanish Firms, Estudio/Working Paper no. 43 (Madrid: Instituto Juan March, 1993), 37.Google Scholar

⁴³ Although only 10—15 percent of salaried workers were union members in the late 1980s, about SO percent voted in factory council elections, a figure that increased to 80 percent in workplaces where elections were held. Escobar (fn. 42), 6; Lobo, Faustino Migueléz, “Las organizaciones sindicales,” in Miguélez, Faustino and Prieto, Carlos, eds., Las relaciones laborales en España (Madrid: siglo veintiuno editores, 1991), 220.Google Scholar

⁴⁴ Paramio, , “Los sindicatos y la política en España, 1982—1992,” in Guerra, Alfonso and Tezanos, José Felix, eds., La decada delcambio (Madrid: Editorial Sistema, 1992), 531–32.Google Scholar

⁴⁵ UGT leaders were particularly sensitive about being too closely associated with the PSOE's reforms after performing poorly in key firms and sectors in the 1986 factory council elections. Following the elections, Redondo complained that “each time a minister speaks … we lose votes; each time the 27 liberalizing measures are mentioned, we lose votes. … There is a problem of identification. Workers identify us with the government or with the actions of public managers.” Quoted in El Patýs, December 18,1986.

⁴⁶ Author interview with Luis Martinez Toval, Madrid, June 29,1995.

⁴⁷ Although it is possible that UGT leaders moved more quickly toward exit than the rank and file would have demanded, the timing of their movement suggests they were responding, at least in part, to signals from workers. Moreover, several opinion polls suggest that workers sympathized with the UGT's criticism of the government. See, e.g., Bouza, Fermýn et al., Perfil, actitudesy demandas deldelegado y afiliado a UGT (Madrid: Fundacion Largo Caballero, 1989), 252–53Google Scholar; and REIS 38 (April-June 1987), 358–59.Google Scholar For a good discussion of the radicalism of Spanish labor leaders relative to the rank and file, see Fishman, Robert, Working-Class Organization and the Return to Democracy in Spain (Ithaca, N.Y.: Cornell University Press, 1990), 39–45.Google Scholar

⁴⁸ A year earlier, the PSOE had refused to defend a budget compromise negotiated by Guerra but then rejected by González. This outcome prompted one UGT leader to complain that “the party has been suppressed in the government-party relationship.” Quoted in Cambio 16 (October 26, 1987).Google Scholar

⁴⁹ Although members of a critical tendency in the PSOE, the socialist left, sympathized with the UGT's position, they lacked any real power within the party because of electoral rules that placed control primarily in the hands of the central party apparatus.

⁵⁰ As in Mexico, the Venezuelan president is constitutionally prohibited from succeeding himself, although he can run for reelection after being out of office for at least ten years.

⁵¹ Corrales, Javier, “El presidente y su gente,” Nueva Sociedad 152 (1997), 97.Google Scholar

⁵² Arismendi, León and Iturraspe, Francisco, “Sistemas electorates sindicales,” in Magallanes, Manuel Vicente, ed., Mandate Político, Evolutión Electoral, Comunkaciony Sociedad (Caracas: Publicaciones del Consejo Supremo Electoral, 1990), 257.Google Scholar

⁵³ UGT members constituted a sizable share of the delegates to PSOE congresses, but the UGT did not act as an organized faction within the party.

⁵⁴ Coppedge, Michael, Strong Parties and Lame Ducks (Stanford, Calif: Stanford University Press, 1994), 109.Google Scholar

⁵⁵ For a detailed discussion of the struggle for control of SUTISS, see Ellner, Steve, Organized Labor in Venezuela, 1958–1991 (Wilmington, Del.: Scholarly Resources, 1993), 153–62.Google Scholar

⁵⁶ Quoted in El National, April 6, 1989.

⁵⁷ Delpino resigned in frustration from his post as CTV president, and the CTV began cooperating with Perez's privatization policies in return for selective benefits for laid-off workers.

⁵⁸ For a more detailed analysis of these trends, see Burgess, Katrina, “Interest Representation at a Critical Juncture? Unions and Parties in a Liberalizing World” (Prepared for delivery at the annual meeting of the American Political Science Association, Atlanta, Ga., September 2—5, 1999).Google Scholar

⁵⁹ For informative discussions of the changing nature of interest representation, see Dalton, Russel J.Beck, Paul Allen, and Flanagan, Scott C., “Electoral Change in Advanced Industrial Democracies,” in Dalton, Beck, , and Flanagan, , eds., Electoral Change in Advanced Industrial Economies: Realignment or Dealignment? (Princeton: Princeton University Press, 1984)Google Scholar; Chalmers, Douglas A.Martin, Scott B., and Peister, Kerianne, “Associative Networks: New Structures of Representation for the Popular Sectors?” in Chalmers, Douglas A. et al., eds., The New Politics of Inequality in Latin America (Oxford: Oxford University Press, 1997)CrossRefGoogle Scholar; and Hagopian, Frances, “Democracy and Political Representation in Latin America in the 1990s: Pause, Reorganization, or Decline?” in Agüero, Felipe and Stark, Jeffrey, eds., Fault Lines of Democracy in Post-Transition Latin America (Miami, Fla.: North-South Center Press, 1998).Google Scholar