Search for a command to run...
It no longer seems controversial to say that governments can sometimes and do sometimes act to carry out the autonomous preferences of officials. Analysts have observed that government leaders at times initiate major policy shifts without regard for the demands of important political and economic groups, that they follow their own economic ideologies, and that they build coalitions to support their preferred policies and ideologies.' In the absence of any developed theory to account for these observations, the ability of officials independently to formulate and execute policy has acquired the label state autonomy. Discussions of the causes of state autonomy have tended to focus on macro-level conditions that might lead officials to form independent preferences or that reduce the ability of societal groups to exert influence. Authors writing from the dependency perspective, for example, link the emergence of an autonomous state to the failure of the national bourgeoisie to achieve hegemony.2 Theda Skocpol and Ellen Kay Trimberger argue that states will act autonomously in order to mobilize national defense against external military threats when traditional rulers have failed to do so.3 The shortcoming of these macro-level explanations is that they describe virtually all developing countries. The hegemonic bourgeoisie has always been a rare and fragile creature; it has not so far found a congenial niche anywhere in the contemporary developing world. Scarcely any part of the earth has escaped the international threats of colonialism, war, and economic imperialism during the nineteenth and twentieth centuries, and few traditional rulers have proved able to mount an effective defense. Thus, if the macro-level speculations about the causes of state autonomy are correct, virtually all states in developing countries should be autonomous. These theories thus fail to explain the vastly differing degrees of autonomy actually observed in developing countries. Nevertheless, they offer a starting point for an explanation. In the face of persistent military and economic threats to sovereignty and well-being, elites committed to change-that is, elites who have preferences at variance with the interests of groups favored by the status quo-do come to power with some frequency in developing countries, as the macrolevel theories would predict. Autonomous actions by these elites, however, occur only infrequently because the efforts of such elites to act on their own preferences often fail. If elites with independent preferences reach power but often fail in their attempts to implement their preferences, explanations of state autonomy need to give some attention to the development of the capacity to carry out these preferences. State autonomy implies that the officials who constitute the state not only have preferences which are more than simple reflections of the preferences of powerful societal groups, but also that they have the capacity, in terms of organizational cohesion, expertise, and extractive and coercive ability, to carry out decisions based on their preferences. Such capacity may depend either on characteristics of the organizations to which officials belong or on attributes of the government itself. In countries governed by Communist parties