The importance of modeling spatial spillovers in?public?choice analysis

It is frequently assumed that regional observations on local government behavior, voters, regional taxes, etc. can be analyzed using ordinary least-squares (OLS) methods. We discuss spatial regression models in empirical studies of public choice issues using impacts arising from population migration on the provision of county-level government services as an illustration. Spatial regressions allow an examination of the direct and indirect (spatial spillover) effects which taken together determine the total impact (on the dependent variable) arising from a change in the explanatory variables. This decomposition should be quite useful in empirical public choice studies.     

How are we doing? Group-based economic assessments and African American political behavior

Using data from the National Black Election Study, this study tests the importance of group-based economic evaluations in driving African American political behavior.¹ Group-based evaluations powerfully influence presidential approval and vote choice, even controlling for national and personal evaluations and a conception of “linked fate.” More importantly, group-based assessments exert a significant and independent influence on turnout, the central variable in black electoral politics. The results extend and reconsider the implications of group solidarity as a motivator of black political behavior and suggest that a revision of traditional notions of economic voting is in order, at least for African Americans.