The effect of neighbourhood diversity on volunteering: Evidence from New Zealand

We make two contributions to the emerging empirical literature that identifies a negative relationship between neighbourhood heterogeneity by factors such as race, ethnicity, income etc., and individuals’ likelihood of contributing to public goods or trusting their neighbors. First, we show that studies that attempt to estimate the effect of a concave neighbourhood characteristic like heterogeneity on outcomes of interest may obtain biased results if they use small or large neighbourhood boundaries alone. Such approaches omit the effect of heterogeneity between small neighbourhoods, and can result in biased estimates of heterogeneity’s effects even when this “between heterogeneity” has no economic effect. Second, with this problem in view, we use two levels of neighbourhood cross section and panel data from the 1996, 2001 and 2006 censuses in New Zealand to test whether heterogeneity by race/ethnicity, birthplace, income or language negatively affect New Zealander’s probability of volunteering. We find that addressing neighbourhood size matters. We then find robust evidence that ethnic/racial neighbourhood heterogeneity is associated with lower volunteering rates. We also find some evidence that language, birthplace and household income heterogeneity lower volunteering rates, but the evidence is less robust, particularly for language and income.