| In the debate on how globalized manufacturing transforms domestic institutions, little consensus has emerged on whether national or international factors play a greater role in shaping institutional outcomes. This study compares patterns of change in the institutions regulating business-government relations. With similar economic, cultural and historical backgrounds, Morocco and Tunisia are well matched for comparing the local effects of integration in world markets. Focusing on the same industrial sectors---textiles and ready-to-wear apparel---further controls for economic differences across the cases. Given these similarities, standard trade theories would predict the construction of parallel domestic political coalitions in the two countries. In fact, Moroccan and Tunisian industrialists mobilized in distinct ways and, as a result, post-reform institutional arrangements took on varied guises in each country. In Morocco, struggles between export and locally oriented manufacturers undercut the clientelist system, producing more formal, institutionalized patterns of business-government relations. In Tunisia, where the expected trade-based cleavage not politicized, the state corporatist system of interest representation remained intact.;I argue that different patterns of interest group mobilization are the key to understanding these distinct institutional outcomes. While the policy preferences of Moroccan and Tunisian businesspeople underwent identical shifts in the face of economic change, varied capital structures, which were shaped by the size of the domestic market, fueled different patterns of collective action. Sustained by a sufficiently large local market, the existence of a well-connected domestic elite in Morocco promoted an oppositional identity among new exporters, galvanizing them to lobby vigorously for their interests. In Tunisia, the small local market induced policy-makers to create an offshore economy well before the adoption of comprehensive trade liberalization in the 1980s. Big, protectionist capital did not occupy a preponderant role in the state's social base and tensions within the private sector did not erupt. The explanation is founded on a multi-level model of institutional change, which stresses the dynamic interaction of firm-level preferences, the role of national and sectoral-level institutions in shaping initial producer behavior, and the role of global incentives in reinforcing or creating new classes. The findings defy prevailing conceptualizations of the relationship between international economic change and domestic politics, which either predict uniform reactions to world price shifts or maintain that national institution remain fixed in the face of exogenous economic change, and underscore the importance of integrating global economic variables in models of institutional change. |