Gentrification and the Artistic Dividend: The Role of the Arts in Neighborhood Change

Abstract

Problem, research strategy, and findings: There is a conflict between recent creative placemaking policies intended to promote positive neighborhood development through the arts and the fact that the arts have long been cited as contributing to gentrification and the displacement of lower-income residents. Unfortunately, we do not have data to demonstrate widespread evidence of either outcome. We address the dearth of comprehensive research and inform neighborhood planning efforts by statistically testing how two different groups of arts activities—the fine arts and commercial arts industries—are associated with conditions indicative of revitalization and gentrification in 100 large U.S. metropolitan areas. We find that different arts activities are associated with different types and levels of neighborhood change. Commercial arts industries show the strongest association with gentrification in rapidly changing areas, while the fine arts are associated with stable, slow-growth neighborhoods.

Takeaway for practice: This research can help planners to more effectively incorporate the arts into neighborhood planning efforts and to anticipate the potential for different outcomes in their arts development strategies, including gentrification-related displacement.

Keywords:

• arts
• cultural industries
• gentrification
• neighborhood planning
• revitalization

Acknowledgments

The authors would like to thank Ann Gadwa Nicodemus, Dan Silver, and three anonymous reviewers for their helpful comments on this article.

Research Support

The research was supported by a grant from the National Endowment for the Arts.

Notes

¹ The term ìneighborhoodî has both social and geographic connotations. We use the term in this study simply to mean a small geographic area that is larger than a block and exists within a city or region. Like most neighborhood studies and planning efforts, we define the neighborhood based on the available census data geography. However, the reality is that neighborhoods are extremely difficult to accurately define because individual residents often have different perceptions of what constitutes the defining features and geographic boundaries of their neighborhood.

² Two issues pertain to the sample. First, we excluded the McAllen—Edinburg–Mission (TX) MSA from the study due to missing coordinate data for its CBD. Second, given the available data, it is possible that we do not capture arts activity that could have emerged around newer CBDs in our sample. However, given that gentrification tends to occur in older urban areas, we feel that we capture the vast majority of arts activity.

³ We used two different weights. For absolute numbers (e.g., population), we calculated weights using the formula w = a_int/a₂₀₀₀, where a_int is the land area from the 2000 ZCTA that overlaps with the 2010 ZCTA area and a₂₀₀₀ is the total area of the 2000 ZCTA. For ratios (e.g., the percentage that walked to work), we calculated weights using the formula w = a_int/a₂₀₁₀, where a_int is the same as above and a₂₀₁₀ is the total area of the 2010 ZCTA.

⁴ The 2007–2011 ACS data is a collection of data over a five-year time period and therefore does not capture one point in time. It is, however, the most reliable and best available source of SES data at the micro level and has been used by others in time-series analysis.

⁵ Unfortunately, given data availability, we were not able to study potentially important variables related to land use and property characteristics, although we recognize that they may have an effect on the relationship between arts industries and neighborhood change.

⁶ To identify outliers, we calculate the z score for each variable included in the scatter plot. If a zip code has a z score of 3 or higher (or 3 or lower), we consider it an outlier.

⁷ While we have made an effort to control for a wide range of variables, as in any time series model, changes can occur over the time period that we do not control for that may affect results.