OPM-commissioned study finds subsidized housing correlates with segregation but stops short of proving causation; recommends statewide portal and tax-credit re‑

A state-funded study presented to a Connecticut housing roundtable on Tuesday concluded that subsidized housing is correlated with racial and economic segregation across the state but that the data do not show a proven causal relationship. The Office of Policy and Management contracted New York firm Urbanomics to analyze housing and program data from 1990 through 2020 and to map program locations and participant demographics.

Manisha Shrivastava, an economist and policy coordinator at OPM, said the work flowed from statute: "Public Act 20 one-two required OPM to aggregate data related to state and federal housing programs and then use that data to analyze the impact of such programs on economic and racial segregation in the state." The study team also published data tools and a report appendix cataloguing more than 80 federal and state housing programs.

The report examined nine segregation and inequality indices at multiple geographies and ran spatial and regression analyses comparing the share of subsidized housing to 27 socioeconomic variables. "While there is a correlation between subsidized housing and segregation," Peter, a senior planner at Urbanomics, summarized, "there was no proven causal relationship." Urbanomics said multivariable models showed socioeconomic factors such as population density, tenure and poverty rates had greater predictive power than the share of subsidized units alone.

Urbanomics principal Tina Lund said historical practices remain a central explanatory factor. "Redlining was absolutely the key the basis for segregation," Lund told the roundtable, pointing to maps that overlaid historic red-line boundaries with current demographic patterns and subsidized-housing locations.

Key empirical findings cited by the consultants included multi-decade settlement shifts — the total nonwhite population increased about 20.6 percentage points statewide from 1990 to 2020, Hispanic/Latino share rose roughly 10.8 points and Asian/Pacific Islander share rose about 3.3 points — and that in 65% of Connecticut municipalities a greater share of subsidized-housing heads of household were people of color than in the municipality as a whole.

Based on those findings and outreach to developers, housing authorities and advocates, Urbanomics proposed a package of policies to reduce clustering and improve mobility: a feasibility study and subsequent creation of a single statewide application portal for active federal, state and local housing choice vouchers and site-based public housing opportunities; a fund for security deposits and vacancy-gap payments; mobility counseling and post-move supports; revisions to the Low‑Income Housing Tax Credit qualified allocation plan to encourage greater geographic diversity; and expanded race/ethnicity data collection for subsidized units under existing state law provisions (the presentation referenced "CGS section 8-68d" in discussing data collection).

The recommendations drew immediate questions about feasibility and trade-offs. Committee member Francis said the consultant interviews did not include some local officials and technical experts and warned that alternative septic or package wastewater systems have a poor maintenance record in some places, which could complicate siting recommendations. "It would be a lot less costly for a developer to attach to existing municipal infrastructure than have to rebuild all the infrastructure on-site," Francis said.

An affordable-housing developer who identified himself as David said market and financing structures make small (20–30 unit) projects difficult for tax-credit developers. "The 20 to 30 unit projects... just do not align well with the development model," he said, urging targeted funding to make small projects viable in higher‑income communities.

Nandini, who took part in the Q&A, cautioned that the Low‑Income Housing Tax Credit program and the QAP process involve federal rules and public‑input requirements and that legislating allocation criteria could have unintended consequences for program administration and public engagement.

OPM signaled readiness to help lawmakers evaluate the study recommendations. "We're ready to assist the General Assembly in evaluating the study recommendations in our joint efforts to increase affordable housing as well as workforce housing," Manisha said.

The roundtable did not take formal votes. Lawmakers and stakeholders asked Urbanomics and OPM for additional follow-up on feasibility, environmental review and how to operationalize proposals such as a statewide portal and QAP changes. The presentation, the associated data tools and the full report are available from OPM; committee members indicated the findings will inform forthcoming legislative and administrative discussions.