Constraints on distributional equity in the location of public services

Studies of the distributional impacts of public service locations have uncovered little bias against low-income groups in geographical access to salutary public services. This paper questions the meaning and interpretation of such Endings because the mathematical procedures used for measuring service impacts ignore two important constraint factors-urban spatial structure and differences in the geographical mobility of income groups. Simulation methods are used to analyze the effect of each factor on the range and frequency distribution of distributional impact scores associated with possible service location patterns in a sample of US cities. The results show that typical urban structures tend to favor location patterns that give greater access to low- than high-income neighborhoods. Differences between cities are a function of the relative locations and densities of income groups. However, if geographical mobility varies by income group, the majority of alternative location patterns are found to benefit high-income groups above others. Methods for incorporating the two constraint factors in measures of distributional impacts are discussed.