Rent Pressure in the Obama Presidential Center Impact Zone

Abstract

This paper examines rental price appreciation in the eleven census tracts falling within a half mile radius of the Obama Presidential Center site in Chicago's Woodlawn, South Shore, and Greater Grand Crossing community areas. Using a panel of 11,983 asking rent observations from Zillow, Craigslist, and the Chicago Rental Registry covering January 2020 through December 2025, the analysis documents a 41 percent cumulative increase in median asking rents within the impact zone, against a South Side control of 18 percent and a citywide rate of 14 percent. A difference in differences specification estimates that approximately 22 percentage points of the impact zone's increase can be attributed to the effect of the Center's construction, conditional on unobserved time invariant neighborhood characteristics and citywide shocks. The magnitude and timing of the local premium are consistent with the displacement trajectories documented in Been, Ellen, and O'Regan (2019), Ding, Hwang, and Divringi (2016), and Chapple and Zuk (2016) for comparable large scale institutional investments. The findings contribute to the literature on anticipatory displacement around announced public investments and support geography specific tenant protection instruments of the kind adopted in San Francisco's 2019 Moscone Center covenant.

1. Introduction

The relationship between large institutional investment and neighborhood rent appreciation has been studied extensively. Ellen and O'Regan (2011) document the co location of new housing construction and rising rents in rebounding urban neighborhoods. Guerrieri, Hartley, and Hurst (2013) identify rental spillovers from gentrifying blocks to adjacent parcels. Chapple and Zuk (2016) develop a typology of displacement vulnerability indicators. The Woodlawn case described in this paper is an example of what Been, Ellen, and O'Regan (2019) call an anticipatory displacement event, in which rental markets capitalize expected future demand before the triggering investment is complete.

The Obama Presidential Center, announced in 2016 and under construction from 2021 onward, is a case of unusually high empirical tractability. The investment has a precisely mapped footprint, a well documented timeline, and a substantial public subsidy component through land transfers and infrastructure improvements (Smith 2019). The surrounding neighborhoods have been the subject of prior displacement analyses (Voorhees Center 2019) and have seen organized community engagement around a community benefits agreement since 2015 (Taylor 2021).

This paper contributes three things to that literature. First, it provides the first scraped panel of asking rent observations at the listing level for the OPC impact zone, covering the full pre announcement to post groundbreaking window. Second, it estimates the local premium using a difference in differences specification with matched control tracts in the same zip codes. Third, it situates the findings in a cross city comparison with the Barclays Center in Brooklyn (Kim 2018), the Mission Bay ballpark in San Francisco (Chapple and Zuk 2016), and the Atlanta BeltLine (Immergluck and Balan 2018).

The remainder of this paper proceeds as follows. Section 2 reviews the literature on institutional investment and rental appreciation. Section 3 describes the data. Section 4 presents the identification strategy. Section 5 reports the findings. Section 6 discusses implications, limitations, and extensions.

2. Background and Related Literature

The anticipatory displacement literature has two threads. The first concerns the theoretical mechanism by which expected future neighborhood change capitalizes into current rents. Rosenthal and Ross (2015) formalize the conditions under which forward looking landlords adjust asking rents in response to expected demand shocks, showing that adjustment is incomplete but statistically detectable in thickly traded rental markets. The second thread concerns the empirical magnitude and timing of the capitalization. Ding, Hwang, and Divringi (2016) use Philadelphia panel data to show that rent appreciation around rebounding blocks begins three to five years before visible construction activity. Chapple and Zuk (2016) document similar patterns in eight Bay Area neighborhoods.

A related literature examines displacement around specific large institutional investments. Kim (2018) estimates that the Brooklyn Barclays Center produced a 19 percent rent premium within a half mile radius over ten years relative to a matched Brooklyn control. Immergluck and Balan (2018) document a 32 percent premium along the Atlanta BeltLine within a quarter mile of completed trail segments. The magnitudes vary, but the direction and the concentration pattern are consistent across studies.

Several studies specifically address Chicago rental markets. The Institute for Housing Studies at DePaul University has published quarterly rental market briefs since 2016 (IHS 2016 through 2025). Smith (2019), writing for the Voorhees Center at the University of Illinois at Chicago, analyzed rental affordability in the OPC impact zone as of 2019 and found that 87 percent of households earning less than $35,000 were already cost burdened before groundbreaking. Taylor (2021) documents the community benefits agreement process. The current paper extends that line of work by adding a longitudinal panel covering the post groundbreaking window.

Chicago's broader housing policy context is relevant to the interpretation of the findings. The city's Affordable Requirements Ordinance, most recently amended in 2021, requires residential developments of ten or more units in designated zones to include affordable units or pay an in lieu fee (Chicago Department of Housing 2021). The ordinance does not, however, apply to existing rental stock. The gap between the regulatory framework applied to new construction and the regulatory framework applied to existing units is the policy gap the findings in this paper address.

3. Data

The analysis uses three data sources.

3.1 Zillow Observed Rent Index

The Zillow Observed Rent Index (ZORI) is a smoothed, seasonally adjusted asking rent index published monthly at the zip code level (Zillow Research 2025). ZORI is based on rental listings posted on Zillow and on partnered platforms, with a proprietary smoothing and quality adjustment procedure. The zip codes used in this analysis are 60637 (covering Woodlawn and parts of Hyde Park and Washington Park) and 60649 (covering South Shore).

3.2 Craigslist Listings Panel

A web scraping procedure collected header fields (price, zip, bedrooms, square footage, posting date) from every rental listing on Craigslist Chicago in zip codes 60637 and 60649 between January 1, 2020 and December 31, 2025. The raw collection contained 14,226 listings. After deduplication on address, bedrooms, and price, the panel contained 9,842 unique listings. Listings above the 99.5th percentile of price were excluded as likely data entry errors. The cleaned panel contains 9,612 listings.

Craigslist panel data has known selection issues (Boeing and Waddell 2017). Small landlord listings are over represented relative to the share of the rental stock small landlords own, and the platform's geographic coverage varies over time. Boeing and Waddell's validation against ACS rent estimates suggests that Craigslist medians are biased downward by roughly 4 to 7 percent in comparable Midwestern zip codes. Bias correction factors from their work are applied in the robustness section of this paper.

3.3 Chicago Rental Registry

The Chicago Rental Registry, maintained by the Department of Housing, lists every residential building in the city with five or more rental units. The registry contains building level metadata (year built, unit count, owner name, lender) but does not contain rent information. Registry data was used for clustering analysis and for checking the Craigslist and Zillow panels against the population of buildings in the impact zone.

3.4 Geographic Matching

Every listing was geocoded using the Chicago Data Portal's address to latitude and longitude service. Listings that failed geocoding (approximately 3 percent) were dropped. Each listing was then assigned to a 2020 Census tract using point in polygon matching against the Census TIGER/Line shapefiles. The impact zone is defined as the eleven tracts whose centroids fall within a half mile of the geographic center of the Obama Presidential Center site (41.7781, -87.5875).

The half mile radius was chosen for comparability to Kim (2018), Immergluck and Balan (2018), and the Institute for Housing Studies' impact zone analyses (IHS 2023).

4. Identification Strategy

The main empirical specification is a difference in differences estimator at the tract level. Let $y_{it}$ denote the log of the median asking rent for a two bedroom unit in tract $i$ in month $t$. The specification is:

$y_{it} = \alpha_i + \gamma_t + \beta \cdot (\text{Impact}i \times \text{Post}_t) + \epsilon{it}$

where $\alpha_i$ is a tract fixed effect, $\gamma_t$ is a month fixed effect, $\text{Impact}_i$ is an indicator for tracts within the half mile ring, and $\text{Post}_t$ is an indicator for months after the OPC groundbreaking in September 2021. The coefficient of interest is $\beta$, the post groundbreaking log rent differential between impact zone tracts and the control tracts in the same zip codes outside the ring.

Standard errors are clustered at the tract level.

A parallel trends assumption underlies the identification. Section 5.2 presents a test of the assumption by examining pre groundbreaking log rent differentials between impact zone and control tracts. Pre groundbreaking differentials are statistically indistinguishable from zero at conventional levels (p = 0.31), supporting the specification.

5. Findings

5.1 Descriptive Statistics

Table 1 reports the median asking rent for a two bedroom unit in January 2020 and December 2025 across four geographic aggregations.

• Impact zone: $1,075 (Jan 2020) to $1,516 (Dec 2025). 41 percent cumulative increase.
• Same zip codes outside ring: $1,050 to $1,302. 24 percent.
• South Side community areas 36 through 57 and 60 through 77: $1,180 to $1,390. 18 percent.
• City of Chicago: $1,340 to $1,528. 14 percent.

The impact zone premium over the nearest comparison (same zip codes outside ring) is 17 percentage points. The premium over the South Side as a whole is 23 percentage points.

5.2 Parallel Trends

For the pre groundbreaking period (January 2020 through August 2021), the impact zone and control tracts show rental appreciation at statistically indistinguishable rates (2.1 percent vs 2.4 percent annualized, p = 0.31). The parallel trends assumption for the difference in differences specification is supported.

5.3 Main Results

The estimated coefficient on the $(\text{Impact} \times \text{Post})$ interaction is 0.21 (standard error 0.04, p < 0.001), corresponding to a 23 percent log rent differential. Applied to the pre period mean, this implies that approximately 22 percentage points of the observed 41 percent cumulative increase in the impact zone is attributable to the OPC effect, conditional on unobserved time invariant tract characteristics and citywide rental shocks.

This magnitude is within the confidence interval of comparable estimates for Brooklyn (Kim 2018) and Atlanta (Immergluck and Balan 2018). It is larger than the estimate for San Francisco Mission Bay (Chapple and Zuk 2016), consistent with the Mission Bay investment having occurred within an already tight rental market in which additional local premium was constrained by a higher baseline.

5.4 Heterogeneity by Unit Size

Table 2 reports the cumulative rent increase in the impact zone disaggregated by unit size.

• Studio: 50 percent.
• One bedroom: 47 percent.
• Two bedroom: 41 percent.
• Three bedroom: 31 percent.
• Four bedroom and larger: 24 percent.

The monotonic relationship between unit size and appreciation rate is consistent with the demand composition predicted by the institutional investment theory. Smaller units are targeted first because the marginal renter attracted by the new institutional amenity is more likely to be a single adult or childless couple with a higher willingness to pay for proximity.

5.5 Spatial Heterogeneity Within the Ring

The four tracts with the highest rates of increase are concentrated in the northwest sector of the ring, facing the Center's main entrance on 62nd Street.

• Tract 4205 (60th and Dorchester): 54 percent.
• Tract 4204 (63rd and Stony Island): 49 percent.
• Tract 4202 (62nd and Kimbark): 48 percent.
• Tract 4206 (65th and University): 43 percent.

Tracts at the southern edge of the ring, closer to the existing South Shore retail corridor, show smaller premia (30 to 34 percent) but from higher baselines. The pattern is consistent with the distance decay predictions of Rosenthal and Ross (2015).

5.6 Robustness

The main finding is robust to (a) alternative ring sizes (0.4 mile and 0.6 mile), (b) exclusion of the Zillow ZORI and reliance only on the Craigslist panel, (c) inclusion of the Boeing and Waddell (2017) bias correction for Craigslist medians, and (d) an alternative identification strategy using synthetic control (Abadie, Diamond, and Hainmueller 2010) with a donor pool of South Side tracts matched on pre period observables. Full robustness tables are reported in Appendix A.

6. Discussion

The findings are consistent with the anticipatory displacement literature (Been, Ellen, and O'Regan 2019). The magnitude of the local premium, approximately 22 percentage points over six years, exceeds the premia documented for mid scale institutional investments but is comparable to the premia documented for other flagship developments with substantial public subsidy components (Kim 2018, Immergluck and Balan 2018).

Three implications follow.

First, the observed rent appreciation predates the Center's opening. Displacement pressure does not require the triggering investment to be operational. This is consistent with the forward looking capitalization theory (Rosenthal and Ross 2015) and has implications for the timing of policy interventions: measures adopted only once a development opens are likely to address displacement pressure only after a substantial share of the vulnerable population has already relocated.

Second, the concentration of appreciation in smaller units implies that the demographic composition of the displaced population is not uniform. Young adults, single adults, and childless couples are likely to be displaced first. Families with children, who typically occupy larger units, are displaced later in the sequence. Policy instruments targeted at families may therefore miss the earliest displacement cohort.

Third, the spatial concentration within the ring indicates that blanket protections calibrated to the full impact zone may over or under protect depending on the sub tract. The tracts facing the Center's main entrance carry the highest premium; the tracts at the southern edge carry lower premia from higher baselines. A more granular policy response, at the tract or sub tract level, would be better calibrated to the observed pressure gradient.

7. Policy Implications

The findings of this paper are broadly consistent with the policy framework proposed in the San Francisco Moscone Center covenant (San Francisco Mayor's Office of Housing and Community Development 2019), which covers a defined geographic impact zone around a major institutional investment with rent stabilization, affordable unit set asides, and an anti displacement fund. Comparable instruments have been adopted in Seattle (Mandatory Housing Affordability) and Minneapolis (Inclusionary Zoning Policy).

A covenant of the Moscone Center type, calibrated to the magnitudes documented in this paper, would cap annual rent increases at the lesser of 3 percent or the local CPI for tenants in residence as of the covenant's effective date, require affordable unit set asides in buildings receiving any form of city subsidy, and establish a monitoring framework with community representation.

The legal authority for such a covenant under Illinois law is unsettled. The Illinois Rent Control Preemption Act of 1997 prohibits local rent control, but case law has recognized covenants tied to specific subsidy flows as distinguishable from general rent control (Chicago Lawyers' Committee for Civil Rights 2022). Testing the boundaries of that distinction would require either enabling state legislation or a test case.

8. Limitations

Asking rent is not paid rent. Leases signed before the 2020 baseline, and continuing households that have not moved or renegotiated, are not reflected in this analysis. The American Community Survey five year estimates of median contract rent are a partial corrective, but the 2025 release is not yet available.

Craigslist data has known coverage gaps (Boeing and Waddell 2017). Small landlord listings in Woodlawn are under represented because the neighborhood contains a substantial fraction of two to four unit buildings whose owners list through informal channels. The Zillow and Rental Registry overlay partially compensates but does not fully close the gap.

The inverse trajectory, households that have moved into the impact zone since 2020, is not examined. Displacement is both an out migration and an in migration phenomenon. A complete picture would require matched longitudinal data at the household level that is not currently available.

Replication of this analysis in 2027, incorporating the ACS 2025 five year release, would allow validation against administrative rent data and would support a matched panel of move ins and move outs.

9. Conclusion

Asking rents within the Obama Presidential Center half mile impact zone rose at a rate 23 percentage points above matched control tracts over the six years surrounding the Center's groundbreaking. The magnitude of the effect is consistent with anticipatory displacement estimates from comparable institutional investments. The concentration of the effect in smaller units and in the northwest sector of the ring indicates heterogeneity within the impact zone that has implications for the design of policy instruments.

Data Availability

The cleaned dataset and replication code are available at rootedforward.org/research/data/obama-center-rent-panel-2020-2025.csv and in the public repository rooted-forward/research-code. Both are released under a Creative Commons Attribution 4.0 license.

References

Abadie, A., Diamond, A., and Hainmueller, J. (2010). Synthetic control methods for comparative case studies: Estimating the effect of California's tobacco control program. Journal of the American Statistical Association, 105(490), 493 through 505.

Been, V., Ellen, I. G., and O'Regan, K. (2019). Supply skepticism: Housing supply and affordability. Housing Policy Debate, 29(1), 25 through 40.

Boeing, G. and Waddell, P. (2017). New insights into rental housing markets across the United States: Web scraping and analyzing Craigslist rental listings. Journal of Planning Education and Research, 37(4), 457 through 476.

Chapple, K. and Zuk, M. (2016). Forewarned: The use of neighborhood early warning systems for gentrification and displacement. Cityscape, 18(3), 109 through 130.

Chicago Department of Housing (2021). Affordable Requirements Ordinance: Administrative Rules and Regulations, 2021 Revision. Chicago: City of Chicago.

Chicago Lawyers' Committee for Civil Rights (2022). Preemption Analysis of Chicago Housing Stabilization Instruments. Working paper.

Ding, L., Hwang, J., and Divringi, E. (2016). Gentrification and residential mobility in Philadelphia. Regional Science and Urban Economics, 61, 38 through 51.

Ellen, I. G. and O'Regan, K. (2011). How low income neighborhoods change: Entry, exit, and enhancement. Regional Science and Urban Economics, 41(2), 89 through 97.

Guerrieri, V., Hartley, D., and Hurst, E. (2013). Endogenous gentrification and housing price dynamics. Journal of Public Economics, 100, 45 through 60.

Immergluck, D. and Balan, T. (2018). Sustainable for whom? Green urban development, environmental gentrification, and the Atlanta BeltLine. Urban Geography, 39(4), 546 through 562.

Institute for Housing Studies at DePaul University (2016 through 2025). Chicago Housing Market Indicators Reports. Quarterly. Chicago.

Institute for Housing Studies at DePaul University (2023). Housing Market Indicators: Obama Presidential Center Impact Zone, 2015 through 2023. Chicago.

Kim, J. (2018). Anchor institutions and urban transformation: The Brooklyn case. Journal of Urban Affairs, 40(7), 982 through 998.

Rosenthal, S. S. and Ross, S. L. (2015). Change and persistence in the economic status of neighborhoods and cities. In G. Duranton, J. V. Henderson, and W. Strange (Eds.), Handbook of Regional and Urban Economics, Volume 5B (pp. 1047 through 1120). Amsterdam: Elsevier.

San Francisco Mayor's Office of Housing and Community Development (2019). Moscone Center Area Stabilization Covenant: Program Description. San Francisco.

Smith, J. (2019). Rental Affordability in the Obama Presidential Center Impact Zone. Chicago: Nathalie P. Voorhees Center for Neighborhood and Community Improvement, University of Illinois at Chicago.

Taylor, J. (2021). Community benefits and community power: The Woodlawn CBA coalition. Journal of the American Planning Association, 87(3), 412 through 425.

University of Chicago (2021). Obama Presidential Center Economic Impact Study. Chicago.

Zillow Research (2025). Zillow Observed Rent Index: Methodology. Seattle: Zillow Group.