Get Educated
Research Evidence: Why Smaller, Mixed-Income PSH Works Better
Peer-Reviewed Research on PSH Models
Mixed-Income Properties Significantly Outperform Concentrated PSH
Minnesota Housing Finance Agency Statewide Evaluation
Tenants in mixed properties (fewer than 50% PSH units) had 65% higher odds of one-year housing continuity compared to majority-PSH properties
Finding held even after controlling for individual tenant characteristics
Source: Human Services Research Institute - Evaluation of Permanent Supportive Housing in Minnesota
Scattered-Site vs. Single-Site Comparative Studies
Vancouver Randomized Controlled Trial (2017)
Both models achieved similar housing stability (~74% time stably housed)
Scattered-site participants showed better community integration outcomes
Study notes congregate models may provide superior on-site services but scattered-site promotes independence
National Academies of Sciences Review (2018)
Large concentrated PSH developments face "formidable regulatory barriers"
Emphasized community integration over institutional settings per Olmstead decision
Noted scattered-site approaches often end up in higher-crime areas, requiring careful implementation
Source: Permanent Supportive Housing: Evaluating the Evidence - National Academies Press
State and Federal Policy Evolution
Georgia's Strict Limits on Concentrated PSH
Georgia Department of Community Affairs 2025 Policy
Limits entire state to maximum 2 single-site PSH projects per year under 9% tax credit program
Requires extensive underwriting standards including trauma-informed design
Prohibits institutional appearance or SRO units
Illinois Housing Development Authority Standards
2025 PSH Development Requirements
Restricts new PSH developments to 25 units or fewer
Requires mixed-income integration (no 100% PSH)
Based on research showing better tenant outcomes in smaller developments
Federal Housing Guidance
HUD and National Low Income Housing Coalition Position
Recommend developments of 25 units or fewer with mixed-income requirements
Emphasize scattered-site approaches for better community integration
Move away from large institutional models based on outcomes research
Source: National Low Income Housing Coalition Policy Positions
Documented Operational Challenges of Large Facilities
San Francisco's Large-Scale PSH Experience
HomeRise at Mission Bay (141 units)
Generated 1,166 service calls within two years of opening (late 2022-2024)
Included 656 police-related incidents
Located in affluent neighborhood, created significant community tension
Tahanan at 833 Bryant Street (146 units)
Experienced 11 fatal overdoses since opening in 2023
Social worker shot, systematic maintenance failures including mold, broken elevators
44 residents filed lawsuit against operator Mercy Housing
Source: Guns and roaches: Inside a homeless housing project - San Francisco Standard
San Francisco SRO Program (6,000 residents, 70 hotels)
$160 million annual budget with documented "pattern of chaos, crime and death"
Multiple overdose deaths and inadequate oversight despite massive spending
Investigative series documented systematic failures
Source: Broken Homes - San Francisco Chronicle investigation
What Research Shows About Optimal PSH Design
Key Findings for Successful PSH
Development Size
Optimal range: 20-50 units maximum
Smaller developments balance service efficiency with community acceptance
Developments exceeding 50 units face exponentially greater operational challenges
Income Integration
Mixed-income properties (25-50% PSH) consistently outperform 100% PSH
Financial sustainability improved through diverse tenant base
Reduced institutional character and stigma
Service Delivery
Scattered-site requires more coordination but shows better independence outcomes
Single-site allows service economies but requires intensive management
Quality of services matters more than model type
Community Integration
Fair Housing Act and Olmstead decision emphasize integration over segregation
Large concentrated facilities risk creating new forms of institutionalization
Neighborhood compatibility requires careful site selection and design
Bottom Line: Evidence Supports Smaller, Mixed Models
The research is clear: permanent supportive housing works best when developments are smaller (25 units or fewer), include mixed-income integration rather than 100% PSH concentration, and are carefully integrated into communities rather than creating institutional enclaves.
This isn't about opposing PSH - it's about advocating for models that research demonstrates work better for both residents and communities.
All sources accessed and verified September 2025. Links provided for independent verification of research findings.
We’re not sharing these examples to dismiss the idea of housing for those in need—but to ask for smarter planning, better execution, and honest conversation.