Much of the public finance literature on neighborhood disparities and service provision focuses on the availability and quality of public goods and amenities (Cutler, Glaeser and Vigdor 1999; Jargowsky 2003; Massey & Denton 1993; Wilson 1987).  However, there is little research on how private services affect the quality and desirability of particular neighborhoods (for residents and businesses).  The access to and variety of nearby retail services, like a supermarket, pharmacy or Laundromat, can be part of a household location decision, just as public safety, sanitation and transportation are.  Retail services are not only relevant for understanding the location decisions of households and the viability of neighborhoods; there are significant public finance implications as well.  The proximity and nature of local retail services can impact local property and land values and, in turn, the tax revenue that is generated off of them. Theoretical predictions are ambiguous: retail services can be both positively capitalized (the “amenity effect”) and negatively capitalized (the “nuisance effect”).  This project is, to our knowledge, the first empirical exploration of this question and has implications for polices related to neighborhood development, local business formation and performance and overall fiscal health.      

We rely on parcel-level sales and assessment data as well as establishment-level retail data, both of which span over two decades.  Therefore, we can identify changes in property values against changes in local retail services, over a very extended time period.  We can also compare these changes across neighborhoods of varying economic and demographic make-ups. We rely on a standard hedonic framework, where the nearby retail activity variable(s) operate as locational characteristics that contribute to the valuation of a particular residential property. 

However, we recognize that identifying the direction of effect is not straightforward; while we may observe a positive relationship between changes in retail activity and prices, it may be the case that rising prices attract retail (and not vice versa). In order to better identify the impact of retail change on changes in home prices, we will employ two strategies to identify meaningful transitions in the local retail landscape in order to better identify significant shifts in property values.  First, as part of our data collection, we have information on every rezoning in New York City between 2003 and 2011.  We use this information to identify swaths of the city that were rezoned to commercial or mixed use (i.e. from manufacturing or exclusively residential) and implement this dramatic shift in land use as a shock to identify meaningful changes in retail activity.  Second, we instrument for change in retail activity at the tract level with establishment shift share measures calculated relative to a larger, sub-municipal geography (such as ZIP code or public use microdata area), which should be directly correlated with changes in neighborhood retail activity but not with changes in within-neighborhood prices.  We also test for different price outcomes across neighborhoods with different economic trajectories and across various types of retail that may impose different amenity or nuisance effects (such as necessity versus luxury goods/services).