Community Sorting: Evidence from Northern Ireland.

The aim is to make use of information about residential area choices and the timing of moves to uncover the preferences for housing, segregation and other dynamically-evolving area attributes in Northern Ireland (NI). The 2001 Census indicates that roughly 42% of Catholics and 39% of Protestants live in areas (1 km grid squares) where over 90% of the people are of the same community background as them (Shuttleworth and Lloyd, 2009). Many papers have considered the degree of segregation and what the consequences are but few try to establish what are the causes and dynamics that perpetuate it and what households’ preferences are actually over.

For NI specifically it is possible that violence (or the fear of violence) goes a long way to explaining segregation and therefore looking at the effects of crime and violence on location decisions will be important. The study will seek to establish how large the role of education and income are likely to play as more mixed areas tend to consist of better educated and wealthier households.

This study is rooted in both the Tiebout sorting literature and the more general topic of segregation. Preferences will be quantified by estimating the marginal willingness to pay for area attributes by applying a dynamic discrete choice model for housing and neighborhood attributes.