2a policy and state downsizing, and the Bank created a new branch devoted to reform of developing country public sectors. These changes constituted an open admission that politics was a critical component of development, and that the state was not simply an obstacle to growth but often a necessary underpinning for it. Increasingly, donor agencies have seen the promotion of democratic accountability as one tool in the fight against corruption. This modest degree of convergence should not, however, obscure the continuing degree of compartmentalization that exists in the field of development. While paying lip service to the importance of institutions, most economists and field practitioners still see politics as at best an obstacle to the real work of development, which is improvement in incomes, health, education and the like, and not as an independent objective of development strategy. (Amartya Sen is an important exception to this generalization.) The democracy promotion agencies, for their part, spend relatively little time worrying about economic growth, social policy or public health, which in their view are goods often used by authoritarian regimes to buy off populations and prevent democratization. The intellectual confusion surrounding development has led to severely Balkanized policies both in the United States and in the international community that often work at cross purposes from one another. For example, the authoritarian or semi-authoritarian regimes of Meles Zenawi in Ethiopia, Paul Kagame in Rwanda and Yoweri Museveni in Uganda have been aid darlings over the past decade because of their track records in promoting economic, health and social goals. At the same time, democracy promotion groups have been highly critical of them and have supported opposition groups and civil society organizations seeking accountability and limits on executive power. To be sure, aid agencies don’t object to greater government accountability on the part of these regimes, w