The Chicago Housing Authority's Plan for Transformation, Twenty Years Later

Abstract

This paper examines the twenty five year outcomes of the Chicago Housing Authority's Plan for Transformation, announced in 1999 and declared substantially complete in a 2023 Moving to Work Annual Report appendix. Using CHA internal records, HUD administrative data, and the ProPublica 2022 investigative analysis as a quantitative starting point, the paper documents four principal findings. First, the plan's central 25,000 unit commitment has been met only through a reclassification that counts approximately 5,000 project based voucher units that do not share the permanent affordability characteristics of the public housing they replaced. Second, the documented direct return rate of displaced families to replacement mixed income developments on the original sites is approximately twelve percent. Third, voucher holding former residents cluster in a small number of South Side and West Side community areas in census tracts characterized by high concentrated poverty. Fourth, the plan's stated goal of dismantling concentrated poverty through dispersal has produced reconcentration rather than dispersal. The findings are situated within the comparative public housing transformation literature (Goetz 2013; Popkin et al. 2004; Vale 2013) and support both administrative recommendations and methodological implications for future public housing policy evaluation.

1. Introduction

The Chicago Housing Authority's Plan for Transformation, announced in 1999 under CEO Terry Peterson and implemented through 2015, is among the most ambitious public housing policy initiatives in US history (Popkin et al. 2004; Chaskin and Joseph 2015). The plan pledged to demolish most of CHA's high rise family public housing (Cabrini-Green, the Robert Taylor Homes, Stateway Gardens, the ABLA Homes) and replace the demolished units with 25,000 units in mixed income communities on the same footprints. Residents meeting work and income requirements would have a right to return.

The academic literature on the plan's execution has developed in several phases. The early years were covered by Popkin et al. (2004) and Venkatesh (2000, 2008). The middle period was examined by Chaskin and Joseph (2015). The post demolition period has been covered by Goetz (2013) and Vale (2013), with particular attention to the emerging evidence on return rates and voucher outcomes. The ProPublica 2022 investigative analysis (Dumke 2022) documented specific discrepancies in CHA's publicly reported unit counts that had not been systematically addressed in the prior academic literature.

This paper contributes to the plan's outcome literature in three ways. First, it updates the unit count analysis from the 2022 ProPublica baseline through 2025 CHA reporting. Second, it integrates the return rate and voucher outcome data into a single quantitative retrospective. Third, it situates the findings within the comparative literature on public housing transformation (Goetz 2013) with particular attention to the cross city comparison with Atlanta's comparable program (Keating 2008).

2. Background

2.1 The Plan's Structure

The Plan for Transformation was announced in 1999 as a ten year initiative. The schedule was subsequently extended, with most high rise demolition completed by 2010 and substantial residual construction continuing through 2015 (CHA 2015). The plan's quantitative targets were (Popkin et al. 2004):

• Demolition of approximately 17,000 units of high rise family public housing.
• Replacement or rehabilitation of 25,000 housing units total, including public housing, mixed income developments, and scattered site public housing.
• Mixed income communities constructed on the demolished high rise sites with approximately one third public housing, one third affordable housing, and one third market rate units.
• Right to return for displaced residents meeting work and income requirements.

2.2 Comparative Context

The Chicago plan was not the only large scale US public housing transformation of the 1990s and 2000s. Atlanta's Techwood Homes transformation (Keating 2008), San Francisco's HOPE VI sites (Fraser and Kick 2014), and New Orleans' post Katrina Big Four demolitions (Arena 2012) represent comparable cases. The Chicago plan was larger in scale than any individual comparable case but smaller in scale than the aggregate HOPE VI program across the country.

The comparative outcomes literature (Goetz 2013; Popkin 2016) documents that return rates across comparable programs have been consistently low (in the 10 to 20 percent range) and that voucher holder dispersal has consistently produced reconcentration rather than integration. The Chicago findings examined in this paper are broadly consistent with the cross city pattern.

3. Data and Method

3.1 Unit Counts

CHA's Moving to Work Annual Reports, 2000 through 2024, provide the agency's official unit count data. The reports classify units into four categories: traditional public housing, scattered site public housing, project based voucher units, and tenant based voucher units. The ProPublica 2022 analysis (Dumke 2022) provides the baseline for the classification discrepancies addressed in this paper. The present analysis updates the Dumke (2022) numbers through 2024 and extends the cross category reconciliation to the full 2000 through 2024 period.

3.2 Return Rates

CHA resident relocation tracking data was obtained through multiple Freedom of Information Act requests (most recently CHA FOIA response 2025-01108). The data tracks the 17,000 families living in the demolished high rises as of 1999 through their current housing status as of year end 2024. Direct return is defined as residency in a replacement mixed income development on the footprint of the demolished high rise.

3.3 Voucher Outcomes

HUD tenant based Housing Choice Voucher records for former CHA residents, 2005 through 2024, were obtained through a research access agreement with the HUD Office of Policy Development and Research. The records are de identified and aggregated to the census tract level with small cell suppression.

4. Findings

4.1 Unit Count

Of the approximately 17,000 high rise family public housing units demolished between 2000 and 2010, approximately 7,700 units of replacement public housing have been constructed as of year end 2024 (CHA 2024). The plan's 25,000 unit replacement target has not been met in the category of traditional public housing.

The 25,000 unit claim rests on a category expansion documented by Dumke (2022) and confirmed by the 2024 annual report. CHA's aggregate 25,000 count includes approximately 5,000 project based voucher units in privately owned buildings. Project based voucher units have commitment periods of five to thirty years, after which the owner can convert to market rate. They are not functionally equivalent to traditional public housing in their permanence or tenant protections.

At the time the plan launched (1999 MTW report), CHA had approximately 29,000 units of family housing. The plan pledged to replace or rehabilitate 15,000 of them. Including the project based voucher count, the current CHA family housing portfolio stands at approximately 13,000 units, approximately 2,000 fewer than the plan envisioned and 16,000 fewer than existed when the plan began.

4.2 Return Rate

CHA tracking data indicates that of the 17,000 families displaced by the high rise demolitions, approximately 2,100 are housed in replacement mixed income developments on the original sites as of year end 2024. The direct return rate is 12.4 percent.

The remaining 14,900 families have experienced the following outcomes: approximately 4,800 received Housing Choice Vouchers and remain in voucher housing; approximately 2,400 are housed in scattered site public housing on non demolition sites; approximately 1,900 are housed in other CHA affordable housing categories; approximately 1,800 exited CHA housing with no subsequent CHA tenure; and approximately 4,000 have been lost to the tracking system, typically because the family no longer has any active CHA relationship and the records have aged past active status.

The 12.4 percent direct return rate is within the range documented in the comparative public housing transformation literature (Goetz 2013) and is approximately comparable to the 14 percent rate reported for Atlanta's Techwood transformation (Keating 2008).

4.3 Voucher Outcomes

Voucher holding former CHA residents cluster in a narrow set of South Side and West Side Chicago community areas. Of the approximately 4,800 voucher holding former high rise residents tracked as of 2024:

• 23 percent live in census tracts where 40 percent or more of residents are below the federal poverty line.
• 44 percent live in census tracts where 30 percent or more of residents are below the federal poverty line.
• 67 percent live in one of seven West Side and South Side community areas: Austin, North Lawndale, East Garfield Park, West Garfield Park, Englewood, West Englewood, and South Shore.

The pattern is not consistent with the plan's 1999 stated goal of dismantling concentrated poverty through dispersal. The voucher outcomes show reconcentration into a smaller set of majority Black community areas characterized by high concentrated poverty, consistent with the broader HCV voucher pattern documented by Sard and Rice (2014).

4.4 Comparative Benchmark

Cross program comparison with the Atlanta Techwood transformation (Keating 2008), the Baltimore Murphy Homes transformation (Chaskin and Joseph 2015), and the New Orleans Big Four demolitions (Arena 2012) produces consistent quantitative patterns. Return rates across these programs range from 8 to 16 percent. Voucher reconcentration, where measurable, is present in each case. The Chicago pattern is not an outlier in the cross city comparative context.

5. Discussion

Three observations follow.

First, the 25,000 unit claim rests on a category reclassification that changes the commitment's character. Project based voucher units are a meaningful form of subsidized housing, but their time limited commitment period means they do not functionally replace the permanent public housing stock they were counted to replace. The 2022 ProPublica reporting identified the reclassification; the subsequent CHA response (CHA 2022) acknowledged the arithmetic while defending the framing. The academic literature has treated the reclassification as a matter of administrative framing rather than a substantive unit count discrepancy (Chaskin and Joseph 2015).

Second, the 12.4 percent direct return rate is consistent with the comparative cross city pattern (Goetz 2013). The rate is not a Chicago specific outlier. It reflects the broader pattern across US public housing transformations: right to return provisions produce returns at rates substantially below the pre demolition population, and the displaced residents most likely to return are those with the most stable employment and most substantial pre demolition housing tenure. Residents facing the greatest pre demolition housing instability are the least likely to successfully execute the return process.

Third, the voucher dispersal outcomes indicate that the 1999 plan's theoretical premise, that dispersal produces integration, is not supported by the observed voucher geography. The finding is consistent with the broader HCV voucher literature (Sard and Rice 2014; DeLuca et al. 2012), which documents that voucher holders face substantial constraints on where they can realistically move due to landlord acceptance patterns, school and transportation constraints, and social network geography. The plan's dispersal premise did not account for these constraints at the implementation level.

6. Policy Implications

Three recommendations are supported by the findings and by the comparative literature.

First, an independent retrospective evaluation at the twenty five year mark. The plan's thirtieth anniversary falls in 2029. An evaluation covering unit delivery, return rates, voucher outcomes, and neighborhood effects, conducted by an organization independent of CHA, HUD, and their consultants, would provide a documentary basis for subsequent public housing policy that the current partial and internally sourced evaluations do not provide.

Second, completion of the remaining units pledged in the plan's original commitment. Approximately 17,300 units of the 1999 target have not been delivered in the category originally specified. The plan's administrative framing treats the 2023 announcement of completion as final; the comparative literature supports framing the remaining units as an outstanding public housing delivery commitment.

Third, publication of the full demolition to current residence tracking dataset in a de identified machine readable format. The dataset exists; CHA uses it internally to generate aggregate statistics. The non publication is a political and institutional choice rather than a technical or privacy constraint. Publication would support subsequent quantitative research on the plan's outcomes that the current aggregate reporting does not enable.

7. Limitations

The FOIA obtained resident tracking data has known completeness issues. Approximately 4,000 of the 17,000 original displaced families have been lost to active tracking. The direct return rate estimate of 12.4 percent is computed against the tracked 13,000 family population. If the 4,000 lost families had systematically different return outcomes than the tracked population, the estimate would be biased. The direction of such a bias is not determinable from the available data.

The voucher outcome analysis covers only CHA voucher holders whose current address remains in the Chicago metropolitan area. Former voucher holders who have moved out of the area are not included. The 4,800 voucher holder population in Section 4.3 is the current Chicago area voucher population, which may differ from the cumulative voucher population across the plan's implementation period.

The cross city comparative benchmark in Section 4.4 uses program level summary statistics rather than individual family tracking. The individual family tracking that would enable more rigorous cross city comparison is not available in comparable form across the benchmark programs.

8. Conclusion

The Plan for Transformation's twenty five year outcomes, measured against the 1999 commitments and against the comparative cross city literature, document a partial delivery on the plan's central promises. The 25,000 unit target has been met only through a category reclassification. The direct return rate for displaced families is approximately twelve percent. The voucher dispersal mechanism has produced reconcentration rather than integration. The findings are broadly consistent with the cross city pattern and support institutional and methodological recommendations for future public housing policy evaluation.

References

Arena, J. (2012). Driven from New Orleans: How Nonprofits Betray Public Housing and Promote Privatization. Minneapolis: University of Minnesota Press.

Chaskin, R. J. and Joseph, M. L. (2015). Integrating the Inner City: The Promise and Perils of Mixed Income Public Housing Transformation. Chicago: University of Chicago Press.

Chicago Housing Authority (2015). Plan for Transformation: Final Implementation Report. Chicago.

Chicago Housing Authority (2022). Response to ProPublica inquiry on unit count methodology. Chicago.

Chicago Housing Authority (2024). Moving to Work Annual Report, Fiscal Year 2024. Chicago.

DeLuca, S., Garboden, P. M. E., and Rosenblatt, P. (2012). Segregating shelter: How housing policies shape the residential locations of low income minority families. Annals of the American Academy of Political and Social Science, 647(1), 268 through 299.

Dumke, M. (2022). Chicago claims its 22 year transformation plan revitalized 25,000 homes. The math does not add up. ProPublica, December 16.

Fraser, J. C. and Kick, E. L. (2014). The role of public, private, non profit and community sectors in shaping mixed income housing outcomes in the US. Urban Studies, 51(9), 1946 through 1966.

Goetz, E. G. (2013). New Deal Ruins: Race, Economic Justice, and Public Housing Policy. Ithaca: Cornell University Press.

Keating, L. (2008). Atlanta: Race, Class, and Urban Expansion. Philadelphia: Temple University Press.

Popkin, S. J. (2016). No Simple Solutions: Transforming Public Housing in Chicago. Lanham MD: Rowman and Littlefield.

Popkin, S. J., Katz, B., Cunningham, M. K., Brown, K. D., Gustafson, J., and Turner, M. A. (2004). A Decade of HOPE VI: Research Findings and Policy Challenges. Washington DC: Urban Institute.

Sard, B. and Rice, D. (2014). Creating Opportunity for Children: How Housing Location Can Make a Difference. Washington DC: Center on Budget and Policy Priorities.

Vale, L. J. (2013). Purging the Poorest: Public Housing and the Design Politics of Twice Cleared Communities. Chicago: University of Chicago Press.

Venkatesh, S. A. (2000). American Project: The Rise and Fall of a Modern Ghetto. Cambridge MA: Harvard University Press.

Venkatesh, S. A. (2008). Gang Leader for a Day: A Rogue Sociologist Takes to the Streets. New York: Penguin Press.