Neighborhood Diversity, Economic Health, and The Role of the Arts

ABSTRACT

Policymakers pursue a range of strategies aimed at diversifying neighborhoods despite research indicating the complicated and potentially damaging results of these efforts. One increasingly common approach is to incorporate the arts into planning efforts in the hope of enhancing diversity and catalyzing positive neighborhood change. Using data from the Cultural Data Project, the authors determine where newly established New York City arts organizations locate in terms of neighborhood racial, income, and industry diversity. They then analyze how diverse contexts interact with an arts presence to impact neighborhood economic health over time. They find that neighborhoods with high levels of racial diversity and low levels of income and industry diversity benefit most from an arts presence. However, the arts are attracted predominantly to neighborhoods with moderate levels of racial diversity and high levels of income and industry diversity. This complicates the use of the arts as a tool in urban revitalization policy.

Notes

We remove seven Neighborhood Tabulation Areas from the analysis, which are designated as parks or cemeteries or do not have census data associated with the area.

The CDP only includes information on nonprofit arts organizations that choose to participate and is thus less comprehensive than the National Center of Charitable Statistics (NCCS). The CDP estimates that it has data on approximately 80% of all New York arts organizations. Further, it collects information that allows for a more nuanced analysis of arts organizational types. Thus, although not 100% comprehensive, the database provides a breadth of data that is unmatched elsewhere.

For fiscal year 2010, 1,228 organizations reported data. Of these organizations 299 were founded between 2000 and 2010. However, 40 organizations were removed from the final sample. These organizations are classified as “non‐arts” or they are located in areas that do not have complete demographic and industry data relevant to the study. Two organizations reported data twice for the 2010 fiscal year. A full list of removed organizations is available from the authors on request.

As the presence of already established arts organizations may likely impact neighborhood outcomes, we attempted to incorporate a three‐way interaction variable including pre‐existing arts organizations, the presence of new arts, and our diversity dummy variables. However, the three‐way interactions were not significant. As such, we include two‐way variable interactions (Diversity quartiles × New art) and added a dummy variable indicating whether the neighborhood is already home to a high concentration of arts organizations.

Local and community focused organizations are theorized to build social capital and stabilize diverse neighborhoods. Although this distinction does not directly test the relationship between arts organizations and their impact on social capital, it does provide some differentiation between types of arts organizations.

Although an estimate taken over a 5‐year period, we use these data to approximate the year 2010. In the text we discuss disadvantage in terms of change between 2000 and 2010.

In our final models, we drop the percentage of residents in managerial occupations and multifamily housing due to high collinearity with the percentage of residents with advanced degrees and rental‐occupied housing, respectively. We also conduct variance inflation factor (VIF) tests to address potential collinearity concerns. Robust standard errors are used to address slight heteroskedasticity. It is possible that the presence of newly established arts organizations and the presumed accompanying change in neighborhood disadvantage may not reflect a causal relationship but rather both signal the presence of other influential variables. To address this issue, we employ Hausman's test for endogeneity. We first predict the presence of new arts organizations using the model's independent variables and controls and incorporate the regression residuals into the original equation predicting change in neighborhood disadvantage. We also attempted to include an endogenous dummy treatment effect where we defined the treatment as the establishment of new arts organizations predicted by the presence of pre‐existing arts and diversity measures. However, the results were insignificant for both methods, suggesting that the presence of new arts organizations can be considered exogenous.

However, Stern and Seifert () construct their income diversity measure using a threshold of 40% for each income bracket to characterize neighborhoods dominated by single income groups versus those that are more diversified. Like Stern and Seifert's approach, our income entropy measure identifies neighborhoods with significant representations of income brackets on opposite ends of the spectrum as highly diversified. However, it is possible that using an income polarization variable may produce different results.

Additional information

Notes on contributors

Nicole Foster

Nicole Foster is a PhD student at the University of Texas, Arlington. Her research focuses on tactical urbanism and arts‐driven community development.

Carl Grodach

Carl Grodach is a Senior Lecturer at the Queensland University of Technology. His research focuses on the urban development impacts of arts organizations, cultural industries, and cultural policy.

James Murdoch

James Murdoch III is a PhD student at the University of Texas, Arlington. His research focuses on regional economic development and cultural sustainability.