The Disparate Neighborhood Impacts of the Great Recession: Evidence from Chicago

Abstract

We advance scholarship about how macroeconomic forces differentially manifest themselves across local spaces by developing a holistic conceptual framework and empirical analyses involving multilevel change modeling. Unlike prior work, we examine differential rates of change in neighborhood indicators. We illustrate our approach with Chicago data measuring the crime, housing, and economic domains of neighborhood quality of life over the 2000–2009 period. We find that the local dynamic manifestations of macroeconomic cycles were far more nuanced than have been previously observed. Neighborhood indicators moved along distinct trajectories, sometimes but not necessarily tracking each other or the overall business cycle, and they changed with varied intensities. The Great Recession of 2006–2009 had disparate negative impacts on lower-income and minority-occupied neighborhoods' local job opportunities, home prices, and home foreclosures, though this was not true for credit or crime indicators. Credit indicators performed geographically much differently than in the prior Chicago recession.

Key words:

• neighborhoods
• multilevel change models
• disparate impact
• business cycles

Acknowledgements

This research was supported from a grant from the MacArthur Foundation to MDRC related to the evaluation of the New Communities Program. The opinions expressed in this article are of the authors and do not necessarilyrepresent those of MDRC, the Foundation, or their respective Boards of Trustees.

Notes

1 ²The Census tract designations used for these transformations are the definitions created after the 2000 decennial Census; data collected or assembled using earlier designations was transformed to the 2000-era designations using the relational matrices published by the US Census Bureau. For most of the neighborhoods, the definitional boundaries align with tract boundaries such that the neighborhood-level measure is the aggregation of the tract-level measure. In cases where this is not true, the tract values were apportioned between multiple neighborhoods based on the distribution of the tract's population between the multiple neighborhoods.

2 ³With multiple incidents, the report is classified in the UCR category of the most serious crime (generally, the crime with the highest potential penalty). Note that these are police reports and do not reflect later adjudication of the incident (e.g., an assault recorded on the initial report as a criminal act later adjudicated as justifiable self-defense is still included).

3 ⁴HMDA data are used instead of sales price data due to the relatively long time series available; HMDA data was available starting in 1992 while the sales price data was only available as two incongruent time series that covered a more limited period. The HMDA and sales price data exhibited the same trends for the times in which we had overlapping coverage. Galster et al. (2005) demonstrated the value of HMDA data as a source for constructing neighborhood indicators.

4 ⁵Both filed and completions data exclude ownership transfers that occur as the result of financial distress (short sales or deed-in-lieu-of-foreclosure transactions).

5 ⁶Details of the cluster analysis procedures, indicator variables employed, and allocations of particular neighborhood to cluster groups are available from the authors.

6 ⁷This is a more refined way of dealing with “abnormal” spikes/troughs in longitudinal trajectories, which examination of our data indicates is focused on specific neighborhoods rather than across the entire sample.

7 ⁸The multilevel model random effects are different from the random effects commonly used in econometric time series models. In the multilevel formulation, the random effects are equivalent to main effects, with the descriptor “random” referring to the nature of the neighborhoods (i.e., they are theoretically drawn at random from some larger population). In contrast, the econometric random effects describes the effect of individual differences (i.e., random disturbance), which is necessary to account for to generate unbiased estimates for the main effects.

8 ⁹The trend and trajectory seen for Chicago and its neighborhoods is consistent with data reported by the FFIEC regarding national trends in these loans. An additional explanation for the results reported here is the cutoff point for reporting; that is, due to inflation and other trends in business capitalization, one million dollars may no longer reflect the high point for the types of loans the data collection program was intended was track.

9 ¹⁰There are several potential explanations for this long-term decline, including reductions in the pool of teenagers, stabilization of local drug markets, mandatory prison sentencing requirements, and new policing strategies.

10 ¹¹This was the conclusion of Galster et al. (2007) and Lim and Galster (2009) in their analyses of neighborhood crime dynamics. Note that this argument cannot be made in the aforementioned case of foreclosures because in general there was a positive correlation between starting levels and rates of change: 0.80 for Chicago overall.

11 ¹²Correlations are available upon request.