Spatial Competition With Interacting Agents PDF Download

Abstract: This dissertation contributes to the study of the economics of strategic interactions through the examination of four distinct but thematically related investigations of spatial competition, conflict and cooperation. A spatial accounting, whether conceptualized as local interactions, neighborhood effects, externalities, or other locally defined phenomena, increases the detail used to examine the decisions of economic agents. The first essay examines the theoretical conditions in which spheres of influence form in games of spatial conflict. The term sphere of influence is defined for nation-state competition. Spheres of influence form in non-cooperative games of strategic complements if the contact between the rivals is repeated in both spatial and temporal contexts. The constraints on global cooperation are eased by the ability to form spheres, which allow the nation-states to avoid exhaustive conflict within individual spatial competitions. Employing this sphere of influence model, the second essay develops a theory of cooperative tax strategies to reexamine the problem of local tax competition. Cooperative tax polices assume the form of either uniform or differentiated taxation on local capital. Contrary to previous findings, differentiated taxation of capital may generate higher community welfare, as well as increased resistance to defection from cooperative policy regimes. A spatial econometric examination of state-level local taxation is consistent with these theoretical propositions. The third essay examines another form of spatial competition: defense spending by nation-states. Two theoretical models are developed in this paper: a two-period dyadic rivalry model and an alliance and hegemonic defense provision model. Spatial econometric testing of the model indicates evidence for U.S. hegemonic behavior, but an underlying rivalry in defense expenditure and armed forces for all other nation-states. The final essay also employs the sphere of influence model to study the lack of competitive elections for the U.S. Congress. So-called sweetheart gerrymandering is demonstrated to be the result of bipartisan efforts to allocate voters by political preferences in order to reduce electoral competition within congressional districts. The adoption of these strategies increases the number of party loyalists, reduces bipartisan legislative coalitions, increases the risk associated with the legislative process, but does not change the expected legislative output of Congress.