This paper argues that the sign of external effects of coalition formation provides a useful organizing principle in examining economic coalitions. In many interesting economic games, coalition formation creates either negative externalities or positive externalities on nonmembers. Examples of negative externalities are research coalitions and customs unions. Examples of positive externalities include output cartels and public goods coalitions. I characterize and compare stable coalition structures under the following three rules of coalition formation: the Open Membership game of Yi and Shin (1995), the Coalition Unanimity game of Bloch (1996), and the Equilibrium Binding Agreements of Ray and Vohra (1994).