William Riker ((1964) Federalism: Origin, Operation, Significance. Boston: Little Brown) stressed the problem of the contested nature of federal institutions and argued that federations existed amidst the ongoing challenge to their rules, that federal institutions were being continuously endogenously produced in the interaction of political parties rather than serving as self-enforceable constraints on the political process. As parties changed, so did federalism, and eventually the balance was bound to shift to either one or the other extreme as far as the degree of centralization was concerned. An alternative approach to essentially the same problem of federal instability was to conceptualize the underlying game differently, as a game of coordination, so that institutions would be accepted as constraints and would therefore be self-enforceable because they allow the players to avoid the chaos and successfully converge to an outcome with payoffs exceeding their reservation values (Hardin, 1989, Ordeshook, 1992). The third proposed solution, consociationalism, emphasizes the elite effort to overcome the conflictual nature of the institutional choice (Lijphart, 1977). Here, as in the coordination argument, the hope is that one could create incentives for politicians to view the existing rules as advantageous and to avoid redistribution by means of the institutional revision. Yet, just like the coordination argument, it is based on an implicit assumption that politicians are more easily motivated to act “cooperatively” than are their constituencies. The missing step in the literature is the mechanism by which this more or less “cooperative” behavior of elected politicians could be sustainable in the environment of popular accountability. An essential component in building the theory of institutional design is to show the possibility in a democracy of elected politicians cooperating on institutional matters even when each of their constituencies would prefer to adjust the constitutional terms to its own advantage. Elite “cooperativeness” must be sustainable even in the presence of outside challengers promising to stay closer to the constituent preferences. Here, I present a model of mass-elite equilibrium of constitutional legitimacy, which demonstrates the possibility to motivate the incumbents to sustain the institutional stability while at the same time protecting them from electoral defeat. I also discuss the difficulties and limitations that such a solution faces, in particular, in plural societies. Copyright Springer Science+Business Media, Inc. 2005
