In this study, I estimate the spillover effect of increased ethnic heterogeneity on per-pupil spending using a three-year panel of United States school districts. While there is general consensus that higher ethnic heterogeneity decreases public good provision, there is little research on the potential spillover effects of increased ethnic heterogeneity on public good provision. The focus is on two measures of ethnic heterogeneity: the fragmentation in-dex and the polarization index. Both of these indices disaggregate ethnic groups, which is an improvement over studies that use the share of the population that is not White as a proxy for racial mix. Both of these measures better reflect the tension that arises with local public good provision (Alesina, Baqir, and Easterly, 1999; Ajilore and Smith, 2011). In addition to using more appropriate measures of ethnic heterogeneity, this paper makes a contribution through the proper calculation of spillover effects from spatial models. The findings show greater ethnic heterogeneity has negative spillover effects on per-pupil expenditures. The public finance implications of these demographic changes are important since there will be not only increases in ethnic heterogeneity but also the geographic distribution of this heterogeneity.