California Needs to Learn from Houston and Dallas about Homelessness

Walk through downtown San Francisco or Los Angeles and you’ll navigate a shifting obstacle course of tents, human waste, and unstable individuals. Business districts that once thrived now see foot traffic evaporate as customers avoid entire blocks. Parents can’t take children to public parks. Elderly residents can’t use their own sidewalks. The social contract that public spaces belong to everyone has collapsed. Between 2015 and 2022, Los Angeles County’s homeless population surged by 56% while Houston’s fell by 32%. Today, California houses 28% of America’s homeless population with just 12% of its residents. San Francisco’s homelessness rate is nearly 20 times higher than Houston’s. Share California has spent over $27 billion on homelessness in recent years. The difference is systems architecture. Houston and Dallas built something that works. California built something that doesn’t, then spent billions pretending otherwise. This matters because instability breeds instability. When encampments persist for years, residents and business owners face constant uncertainty. When sweeps just move people two blocks over, when every neighborhood waits to see if it will host the next relocated camp, the crisis simply shifts location without resolution. Texas Democrats proved there’s an alternative through disciplined systems that create stability by resolving homelessness at scale. They didn’t do this primarily to help homeless people, though that happened. They did it because sprawling street encampments make cities unlivable, and the only way to make them go away permanently is to house the people living in them. What Texas Cities Actually Built The success in Houston and Dallas came from building operational infrastructure to make encampments disappear permanently instead of temporarily. Houston designated the Coalition for the Homeless as a “backbone” organization with real power: control over funding, management of a unified data system, and authority to enforce performance standards across over 100 partner agencies. Dallas did the same with Housing Forward, deliberately restructuring its governing board from one dominated by service providers to one controlled by major sector leaders from philanthropy, corporations, and county government. A board of service providers protects existing programs and funding streams. A board of resource-controllers thinks systemically and imposes strategic alignment. This created what California lacks: a single entity in a city that can tell nonprofits what outcomes they must deliver, find the appropriate metics, and cut funding if they’re aren’t playing ball or refusing to do their job. One organization with the authority to impose order on what was previously chaos. The architecture: centralize strategy, decentralize operations. Texas cities mandated unified intake systems so people enter the housing system the same way regardless of entry point. They required all partners to use the same data system so performance is transparent and comparable. They set clear metrics: housing placements per dollar spent, time from intake to placement, retention rates. But case managers retained autonomy in how they work with clients. Nonprofits could innovate in service delivery. California does the opposite. It allows 44 different Continuums of Care to pursue incompatible approaches with fragmented data systems. But it imposes extensive compliance requirements on how nonprofits must staff programs and structure services. The system can neither coordinate at scale nor adapt locally. Houston has housed over 33,000 people since 2012 with a 90% retention rate. The city’s homelessness rate stands at 52 per 100,000 residents, the lowest of any major U.S. metro. Street encampments don’t migrate from neighborhood to neighborhood because the system resolves them. Only after residents are housed does the encampment close permanently. The tents disappear because the people who lived in them now have addresses. Inspired by Houston’s work, Dallas’s first downtown pilot housed 107 chronically homeless individuals in under 100 days. The site hasn’t returned to an encampment. The operational workflow in Texas: Street outreach teams engage encampment residents. Everyone undergoes the same assessment and gets placed on a unified priority list based on objective vulnerability scores. When housing becomes available, it goes to the highest-priority match. Case managers handle everything: finding apartments, negotiating with landlords, covering move-in costs from flexible funding pools braiding federal grants, local government funds, and private philanthropy. San Francisco’s 200 Days of Displacement Mayor Daniel Lurie was elected on a platform of building more homes and cleaning up the streets. “We are just getting started,” Mayor Lurie said in March 2025, promising to be “relentless” in clearing Mission Street and the side alleys of unpermitted vendors, drug dealing and drug use. Two hundred days later, the results document what enforcement without housing capacity produces. San Francisco’s experience in the Mission District is instructive. For 200 days, the city deployed police, street outreach teams, private ambassadors, and Public Works crews to clear the area around 16th and Mission Streets. The pattern is mechanical. Clear Mission Street between 15th and 16th, drug users and encampments move to Caledonia north of 15th. Clear those alleys, Capp Street fills up. Station an ambassador on Capp to keep it clear, within an hour, “10, 15 people all piled up again,” according to the private security team. When that block finally clears, “around the corner, on the south side 16th street, more than a dozen people were doing drugs out in the open.” San Francisco reorganized its entire street team structure, consolidated workers from seven departments under unified leadership, added private ambassadors working 11:30am to 8pm daily, and maintained consistent police presence. After 200 days, they’ve moved the crisis from block to block. A mother picking up her child from Marshall Elementary found a drug user sitting on the school stairs. She used to call security guards stationed at nearby temporary housing, but that facility closed. Business owners on Capp Street report staff fear for their safety. One resident described “dozens of clothing items and several suitcases scattered throughout the street. Feces was smeared on the sidewalk.” Residents can’t plan because they never know which block will host the next rotation. A restaurant owner found the ambassadors effective at clearing his block while streets behind him got worse. The city achieves temporary cosmetic improvement on prioritized blocks while crisis intensifies elsewhere. Most telling: 200 days into Lurie’s cleanup priority, city officials still cannot answer whether they’re connecting people to permanent shelter or “just pushing people to different places.” When a supervisor demanded data on outcomes in May, the city asked organizations for suggestions on improving data collection. Four months later, it remains “unclear if any changes are underway.” Houston would answer this question in real-time through its mandatory HMIS. Every interaction is logged. Every housing placement is tracked. Retention rates are measured. San Francisco doesn’t know because it lacks the infrastructure to know. The city fills shelter spots quickly each morning, but “it’s unclear if the people in those spots move into permanent housing or if they end up back on the street.” Houston houses people first, then closes encampment sites permanently. San Francisco deploys enforcement first, achieves temporary displacement, and watches areas refill. Houston tracks every person from intake through permanent placement. San Francisco can’t determine if anyone is being permanently housed at all. The Human Cost of Managing Crisis Indefinitely San Francisco’s approach doesn’t just fail to resolve homelessness. It creates dangerous, unaccountable systems that put workers at risk without solving underlying problems. On September 27, 2025, Joey Alexander, a 60-year-old Urban Alchemy employee, approached a man using drugs outside the Main Library near City Hall. Alexander asked him to stop, noting there were families nearby. The man pulled a shotgun from his bag, said “F— Urban Alchemy,” and shot Alexander at close range. Alexander died three days later. Urban Alchemy workers are unarmed. They don’t hold state licenses like private security guards. Their job is to ask unhoused residents to move their belongings off sidewalks, discourage drug users from using in public, and reverse overdoses. Alexander is at least the third Urban Alchemy worker to be shot while on duty. The organization itself embodies California’s dysfunction. Founded in 2018, Urban Alchemy has grown rapidly to tens of millions in contracts across California cities by hiring formerly incarcerated individuals as “ambassadors” to manage street disorder. But the model has serious problems. Two former employees filed lawsuits alleging a supervisor sexually harassed multiple women; Urban Alchemy transferred him to Portland and gave him another supervisory role. When an LA worker was filmed hosing down a sidewalk feet from a homeless person scrambling to save belongings, Urban Alchemy initially fired him, then reinstated him months later. Several ambassadors have been accused or convicted of serious crimes including attempted murder. When Los Angeles’s controller tried to investigate the hosing incident, Urban Alchemy sued to block oversight, and the city attorney sided with the nonprofit against the elected controller. The organization resists transparency about how it spends taxpayer money while collecting millions in public contracts. California outsources street management to contractors with minimal oversight, deploying workers without adequate security backup to enforce behavioral norms where violent individuals may be present. The system regulates workers but not leadership. Violent individuals exist in every city and require law enforcement response, but California compounds the danger by using contractors as the front line while perpetually managing rather than resolving the crisis. Alexander died doing work that required security support he didn’t have, deployed by an organization that resists accountability. Why California’s Architecture Fails Everyone California’s dysfunction stems from getting the architecture of control backwards: decentralizing what should be unified while centralizing what should be flexible. The housing bottleneck. California faces a structural deficit of 3.5 million units and builds fewer than 80,000 homes annually when 180,000 are needed. Median rent stands at $2,200 versus $1,233 in Texas in 2023. Housing someone annually through Houston’s system costs approximately $18,000. Building a single supportive housing unit in Los Angeles averages $600,000, sometimes exceeding $700,000 in 2022. Even if California had Houston’s operational efficiency (which it doesn’t) the underlying economics break the model. Houston has no zoning code, making it cheap and fast to build housing. Dallas maintains far more permissive development policies than California. California’s regulatory gauntlet (restrictive zoning, CEQA litigation weaponized to block development, endless permitting delays) actively prevents housing production. Strategic fragmentation, operational rigidity. California operates nine separate state agencies administering 41 different homelessness programs. Despite spending $10,786 per unhoused person (versus Texas’s $806), the system cannot convert resources into outcomes. California had political advantage. Houston and Dallas are blue cities in red states, forced to build systems without state support. California has unified Democratic control from Sacramento through city councils and used that advantage to avoid hard choices. The nonprofit-industrial complex. California maintains rigorous oversight of most government contractors but homeless services nonprofits operate in a privileged space where political connections matter more than outcomes. The August 2025 San Francisco case: Providence Foundation signed off on $105,000 in falsified invoices for maintenance work that never occurred. Investigators found rust and fungus on walls the nonprofit claimed it had painted. The organization hired family members of executives in violation of anti-nepotism rules. Labor enforcement found wage theft. The consequences: a $1 million settlement, leadership changes, continued city funding. Another provider, HomeRise, operates nearly a third of city-funded homeless housing. An audit found it spent $12,500 on a social event and $200,000 in bonuses while oversight remained minimal. California’s homeless services became a sprawling industry of nonprofits that produce activity metrics (beds provided, meals served, case management hours) while avoiding the outcome metric that matters: permanent housing placements per dollar spent. Texas cities broke this through structural reform. Houston’s Coalition and Dallas’s Housing Forward were given real authority: control over funding decisions, management of unified data, power to enforce performance standards. The unified data system makes every organization’s performance transparent. Organizations that don’t deliver housing placements lose federal grants to competitors. These lead agencies don’t dictate how to do case management. They set clear outcome expectations and require transparent reporting. Nonprofits that perform well get more funding. Those that don’t either improve or exit. California could do this (the state has unified political control and far more resources) but it would require confronting politically embedded organizations. California’s Two Failed Approaches For years before Governor Newsom’s enforcement push, California, especially under Newsom, normalized urban decay. Encampments grew from tent clusters to semi-permanent settlements blocking sidewalks, filling parks, creating public health hazards. The 2016 legislation codifying Housing First was treated as victory rather than starting point, as almost like it was an excuse not to, you know, build the units. Money poured into programs without backbone organizations to coordinate them, unified intake systems, or mandatory data participation. Los Angeles County’s homeless population rose 56% between 2015 and 2022 despite billions in investment. Business districts emptied. Parents couldn’t take children to parks. The frustration grew because the policy imposed costs on working and middle-class residents while delivering no visible progress. Newsom’s pivot to enforcement (executive orders demanding clearances, a model ordinance requiring relocation every three days, threats to withhold funding) represents different dysfunction, same failure. California still hasn’t built infrastructure that would allow enforcement to resolve homelessness rather than relocate it. San Diego conducted 6,400 encampment clearings, stored belongings 36 times, and saw them recovered just 4 times. One woman’s kidney medication was confiscated during a sweep; her body swelled with fluid for days until a doctor could rewrite the prescription. She moved two blocks and rebuilt. Imagine what kind of destabilizing effect it has on the homeless who see this stuff? It’s not exactly a stabilizing experience, and instability does not bring out the best in people. Requiring moves every three days doesn’t make encampments disappear. It makes them proliferate unpredictably, and may even metastasize them like a cancer. For residents, this is worse than static encampments. At least with permanent camps, you knew which blocks to avoid. Now every neighborhood waits to see if it will host the next rotation. California’s permissive era (letting encampments grow) and enforcement era (forced relocations) produce the same outcome. Permanent instability. Both approaches treat homelessness as something to be managed indefinitely rather than systematically solved. The Political Choice California’s Democrats had unified control from Sacramento through city councils and chose to protect organizational interests over building functional systems. Houston Democrats (let alone Dallas Democrats working in a purplish blue government), operating as blue cities in red states without state support, made the opposite choice: confronting nonprofit interests, centralizing strategy, and enforcing performance standards. California spent billions avoiding hard choices. Texas cities with a fraction of those resources built systems that work. Houston’s achievement is particularly instructive given the city’s structural fragmentation. Beyond lacking traditional zoning, Houston must navigate special-purpose districts, restrictive deed covenants, and a ring of heavily zoned suburbs and exurbs that historically resist regional coordination on any issue, and aren’t that fond of the homeless. That the Coalition for the Homeless could unify over 100 partners across this fragmented landscape demonstrates what political will can accomplish. If Houston can build a coordinated system despite these obstacles, California’s failure to do so with far greater structural advantages becomes harder to excuse. What This Means for Other Cities Cities contemplating reform should understand the Texas model requires building specific infrastructure: Designate a single backbone organization per city with real authority over funding, data, and performance standards. This must be an entity focused on system coordination, not service delivery. Half-measures (creating “coordinating committees” without enforcement power) accomplish nothing. Mandate unified intake and data systems. Every person entering the homeless response system goes through the same assessment. Every organization reports into the same database. Performance becomes transparent and comparable. Without it, you’re managing organizational silos, not solving homelessness. Braid flexible funding by combining federal grants with local government commitments and substantial private philanthropy. The flexible private capital removes the small barriers that collapse housing placements. Grant operational autonomy within the performance framework. Don’t micromanage how case managers work or how nonprofits structure programs. Standardize outcomes and data reporting, then let organizations innovate on execution. Deploy encampment resolution only after housing capacity exists. Sweeps without housing placements just relocate the problem. Build the system first, then close sites permanently by housing residents. The political cost is real. Existing nonprofits will resist. Some will need to exit the system. Politicians will face criticism from organizations accustomed to funding without performance expectations. But the alternative is California’s cycle: spending billions to produce worsening outcomes while quality of life deteriorates for everyone. The Fragility of Success Houston (and Dallas’s younger program) have produced measurable results, but their systems remain vulnerable to forces beyond local control. Federal funding dependencies create existential risk. Both Texas cities rely heavily on HUD Continuum of Care grants. Houston receives approximately $45 million annually, Dallas $27 million. These are competitive, performance-based awards. But they’re federal discretionary spending subject to political winds. The Trump administration has signaled plans to fundamentally restructure HUD, potentially cutting housing assistance and tying remaining funds to immigration enforcement priorities. If federal CoC funding gets slashed or conditioned on cooperation with deportation efforts, Houston’s system could lose its financial foundation regardless of operational excellence. The temporary COVID relief funds that accelerated Houston’s recent progress have already expired. The Coalition for the Homeless estimates the system needs $50 million to $70 million in new annual funding just to maintain current service levels. Housing market pressures threaten the model’s core advantage. Houston’s success was enabled by cheap, abundant housing. Between 2015 and 2021, median rent in Harris County increased 29% while wages grew only 23%. Eviction filings now exceed pre-pandemic levels. Rising costs make it harder for clients to achieve self-sufficiency when rental assistance ends. Dallas faces worse constraints. The region has a deficit of nearly 40,000 rental units for households at or below 50% of Area Median Income, projected to nearly double by 2035. Rents grew faster in Dallas than Houston between 2015 and 2024. Even with excellent operational systems, these cities can’t escape basic economics. Local political will can erode. Houston’s model survived three mayoral administrations, demonstrating unusual political durability. But this isn’t guaranteed. The consolidated authority that makes the system effective also makes it easy to dismantle. One budget cycle could defund the Coalition, scatter responsibility back across competing agencies, and return to pre-2012 fragmentation. The comparison remains valid despite vulnerabilities. Houston and Dallas built functional systems during years when federal support was available, local housing markets were manageable, and political leadership was aligned. They converted favorable conditions into measurable progress. California had the same favorable conditions, longer, with more resources and greater political control. The state chose not to build functional systems. Now, as federal support becomes uncertain and housing costs rise everywhere, California has nothing to fall back on. Texas cities at least constructed infrastructure that might survive partial funding cuts. California built nothing that could work even under ideal conditions. The question isn’t whether Houston and Dallas have solved homelessness permanently. They haven’t. The question is whether systematic approaches produce better outcomes than fragmentation. Systems require maintenance, political will, and resources. When those disappear, even functional systems can collapse. California never gave its approach the chance to fail this way because it never built anything coherent enough to succeed in the first place. What Actually Works Houston and Dallas prove systematic approaches can work when conditions align. They built systems that actually house people because sprawling encampments destroy quality of life for everyone. But these systems remain fragile to federal funding cuts, rising housing costs, and political change. The Trump administration’s proposals to restructure HUD funding threaten even high-performing systems dependent on federal grants. California shouldn’t mindlessly import Houston’s specific programs. The lesson is about systems capacity: unified data infrastructure that tracks who you’re serving and what outcomes you’re achieving, consolidated strategic authority that makes and executes decisions, performance measurement that directs resources to what works, operational flexibility that lets frontline workers adapt to local conditions. California had unified political control, massive resources, and years to build this capacity. Instead, it maintained comfortable dysfunction: nine state agencies administering 41 programs, 44 Continuums of Care pursuing incompatible strategies, rigorous audits for most contractors but extraordinary latitude for homeless services nonprofits. The state shifted from permissive tolerance to Newsom’s enforcement without fixing foundational problems. Both produce permanent instability. Texas cities with a fraction of the resources built functional infrastructure while conditions were favorable. After all, how many more years of deteriorating quality of life will residents tolerate before demanding politicians stop doing a bad job and start building systems that achieve (let alone exceed) parity with what Houston and Dallas demonstrate is possible? This is just a couple of Texan cities, globally we seen massive cities who are even *safer* despite being even poorer.