April 15, 2026
Over the last decade, homelessness has emerged as one of the more pressing social, economic, and public-safety challenges facing urban areas across the United States. While the crisis is national in scope, its impacts are felt most acutely at the local level—where unsheltered encampments, rising overdose deaths, drug-related crime, and visible street disorder increasingly affect residents and businesses.
In Oklahoma, and particularly within the Tulsa Continuum of Care (CoC), these trends are especially pronounced. Tulsa has become a disproportionate center of the state’s homelessness problem, with rapid growth – especially in its unsheltered population – despite the allocation of significant public and private resources. Over the past decade, policy responses have increasingly emphasized Housing First approaches, while shelter capacity, addiction treatment, and other supportive services have not kept pace with rising need. At the same time, concerns about public safety, neighborhood conditions, and economic vitality have intensified – particularly in the parts of Tulsa most affected by unsheltered homelessness.
This report evaluates both the magnitude and composition of resources devoted to addressing homelessness in Tulsa. Despite tens of millions of dollars in annual public and private spending, and a large network of providers and volunteers, the unsheltered population continues to grow. Moreover, relatively few resources are directed toward substance-use treatment and mental health care, raising questions about whether current spending and policy priorities are aligned with the drivers of homelessness in Tulsa.
Figure 1
2024 was the worst year for homelessness across the U.S., with the total homeless population reaching 771,480 according to the U.S. Department of Housing and Urban Development’s (HUD) latest annual Point-In-Time (PIT) counts. For context, the total number of individuals experiencing homelessness nationwide now exceeds the entire population of Wyoming by about 30%. This culminates a long-term trend of rising homelessness going back to the mid-2010’s, and undoing any brief progress made following the pandemic.
Interestingly, the problem is not uniform across U.S. states. While the average state saw its homeless population grow by approximately 35% over the last decade, 22 states and territories saw homeless populations decline between 2014 and 2024. Just four states with some of the largest absolute and relative population growth – California, Oregon, Washington, and Colorado – account for almost half of all U.S. homeless population growth since 2014.
Each of these states varies in geography, population, and climate, but share a common embrace of housing-focused homeless policy.
Figure 2
In Oklahoma, the homeless population increased to 5,467 people (+17.6% year-over-year), and another approximately 1,994 are in “Permanent Supportive Housing” (PSH) and arguably part of an indigent and dependent population – though not literally homeless. Since 2019 alone, homelessness has increased by over 38% in the state of Oklahoma – 21st fastest in the nation – and according to HUD data 2024 was the largest observed dependent and literally homeless population in Oklahoma since at least 2007. More concerning, of the more than 5,000 literally homeless in the state, nearly 40% are living without any shelter (either emergency, transitional, or permanent).
The problem is particularly acute in the Tulsa Continuum of Care (COC), which generally comprises the Tulsa metro area (and precisely mirrors the borders of Tulsa County). Despite accounting for just 17% of the state’s overall population, over 25% of the state’s homeless population resides within the Tulsa COC. Overall, there are 20 people experiencing homelessness for every 10,000 residents living within the Tulsa COC, compared to 13 in the state at large. Moreover, the concentration of homelessness in the Tulsa COC today is a product of a growing trend rather than a reflection of historical norms. In 2007 the Tulsa COC accounted for a roughly even share of both the state’s resident population and homeless population (16.1% vs 15.8%).
Figure 3
The homeless population is, by nature, transient. Individuals experience it differently and over time. Official counts are “point-in-time” – they count those literally without a permanent home on a specific date and time. Within that count, officials differentiate between the “sheltered” – those in a formal setting meant for habitation – and “unsheltered”, or those living on the streets.
Figure 4
People experiencing homelessness relative to the population overall are much more likely to have addiction issues and use drugs and alcohol; more likely to be convicted of a crime, including violent crimes and felonies; more likely to be diagnosed with mental illness; and have significantly higher mortality rates. These issues are more prominent among the unsheltered population specifically – i.e. those not currently receiving services through emergency or transitional shelters.
Since 2007, Oklahoma’s unsheltered homeless population nearly doubled, increasing from 1,132 individuals to 2,216. Growth in the Tulsa COC has been even more dramatic; its unsheltered population has grown by seven fold over the same period. Since 2019 alone both the state and the Tulsa COC saw their unsheltered population increase by 80%.
The increasing unsheltered population has coincided with other negative social outcomes, such as rising crime, addiction issues and open drug use, and overdoses and other mortality issues. Paralleling the rise in the local unsheltered homeless population, the number of drug overdose deaths increased 65% – reflecting the relationship between homelessness and afflictions such as drug addiction.
While the unsheltered population has grown – particularly driven by drug dependency and mental health issues – there is no apparent resource shortage. Federal spending (the bulk of financial support for most state public and non-profit homeless service providers) has ballooned from $2.8 billion in 2010, to nearly $10 billion today. But the Tulsa COC and the state more broadly has increasingly shifted resources towards the creation of permanent housing and housing placement support first (for example, PSH beds versus short-term shelters and crisis intervention services for drug use and mental illness).
Figure 5
As a result, by 2020, the unsheltered population living within the confines of the Tulsa COC had increased by over 272%, while the number of PSH beds increased by over 250%. The number of shelter beds increased by only 21% in this time. Today there are only enough shelter beds for about two-thirds of the regions counted homeless population, when 13 years earlier there were more than enough beds for the entire population captured by PIT counts. We note also that this is likely an underestimate of the shelter gap, as the PIT process likely undercounts the true population experiencing homelessness.
Figure 6
Federal homelessness policy underwent a clear shift over the past two decades – particularly after the passage of the HEARTH Act – and moved away from an emergency and transitional response to homelessness toward permanent supportive housing and Housing First models. Because federal grants make up a large share of the dollars spent by local and non-profit providers, how these programs define what activities are eligible, competitive, and rewarded, has an outsized influence on how providers structure their homelessness programs. Funding priorities, performance metrics, and compliance requirements incentivized providers to align programs and planning around housing-first approaches, making permanent housing solutions the dominant framework for addressing homelessness – and explaining the surge in PSH beds after 2010.
The Housing First framework posits that homelessness itself is the primary barrier to addressing many other concurrent conditions faced by the homeless – like drug use and mental illness – and once housing can be accounted for (hence the “Housing First” terminology), these problems will either resolve themselves or be more readily solvable. Indeed, foundationally, federally-funded “housing first” policies often explicitly require the housing be provided without preconditions related to reducing drug use, taking prescribed medications, etc.
Not only are these policies costly, but they have failed to curtail the rise in homelessness and its associated conditions and behaviors (e.g., mental health issues and drug addiction) and in many instances have created environments for these pathologies to flourish. Studies suggesting there are savings from shifting homeless populations onto permanent housing programs like PSH by way of reduced consumption of homelessness-related services note that those reductions were often offset by the increased cost of the permanent housing intervention, rendering such interventions at best a “break-even proposition” (fiscally speaking).
In general, Housing First programs do not (and often cannot) require participation in services meant to address these concurrent conditions as a prerequisite to either receiving or maintaining their housing assistance, and this permissive policy focus may explain the U.S. failure to address its homelessness crisis despite massive resource investment. According to research, about half of individuals experiencing homelessness have complex behavioral health needs – such as regular drug and alcohol use, hallucinations, or other psychiatric ailments – which cannot be corrected by housing alone. In one study, about 42% of homeless individuals say their regular drug use predated their first bout of homelessness, and in another, 25% of those surveyed self-identified their drug use as the primary reason for their homelessness.
As homelessness has grown in Oklahoma and the Tulsa COC, so too have the social and human costs associated with it. High levels of substance use among the homeless can create additional barriers to recovery and reintegration into society, but often result in more severe social consequences – premature death, severe injury, rising crime and public nuisances, etc. When homelessness per capita rose 38% across Oklahoma between 2015 and 2024, drug related overdose deaths per capita in the state climbed 35%. According to the Oklahoma State Department of Health, Tulsa County had the 5th highest overdose death rate out of all counties in the state between 2020 and 2024.
While data specific to Oklahoma’s homeless population is unavailable, other jurisdictions show that drug related mortality among the homeless is several orders of magnitude higher than among the general population. In LA County between 2020 and 2021, overdose mortality was estimated to be 39 times greater among the region’s homeless population and was the leading cause of death for that population. In Arizona, the Maricopa County Medical Examiner reported that among the 732 deaths among homeless individuals in 2022, 63% were linked to drug-related factors, with most of those linked to methamphetamines or fentanyl.
FIgure 7
Where data exists, it strongly suggests drug abuse – and fatal drug overdose – is far more likely to occur among a region’s homeless individuals than the population at large. We suspect that in Oklahoma too (and the Tulsa COC specifically), the state’s rising drug overdose crisis is correlated with its unmitigated homelessness issues.
Additionally, homeless individuals are more often the victims of violent crime, including homicide, assault, and sexual assault. Studies have shown that the rate of violence victimization among people experiencing homelessness ranges from 14% to 21%, versus an estimated 2.3% for the population overall.
The public safety concerns associated with homelessness extend as well to surrounding neighborhoods and commercial areas. In Tulsa, businesses have reported concerns about the safety of customers and employees following incidents of harassment, vandalism, and disorder near their properties, prompting increased spending on private security and spurring the formation of a community safety coalition in downtown Tulsa. These concerns have become sufficiently prominent that Housing Solutions Tulsa – the lead agency and primary coordinator of the Tulsa Continuum of Care – has published guidance for local businesses on how to respond to encounters with individuals experiencing homelessness on their premises. CSI notes that these incidents appear concentrated around specific types of businesses, including convenience stores and gas stations, fast food restaurants, and other general service establishments.
Law enforcement’s response to the issues surrounding the homelessness crisis was slow, in part due to shifting views on the role of traditional policing in responding to the nations drug use and homelessness issues. Despite growing concerns over the homelessness and drug problem in Tulsa’s streets, it took the City of Tulsa until 2023 before new policies were adopted to address the problems common among the city’s homeless. Since then, attitudes have shifted, with the Governor of Oklahoma deploying state troopers to address homeless encampments across the state – a policy supported by Tulsa business owners.
Although the social consequences of unmitigated public nuisances, like unsheltered homelessness, can be evident, the economic consequences can be opaque. In cities across the country, rising homelessness and a national experiment with tolerance for open drug use and semi-permanent street encampments has coincided with growing business complaints, declining foot traffic, store closures, and declining property values in urban downtowns. These effects are most visible among businesses that depend heavily on walk-in customers and are concentrated in relatively dense areas like Tulsa.
Figure 8
Again, gas stations, convenience stores, and limited-service restaurants such as fast-food establishments are often on the front lines of this impact.
Surveys and business accounts illustrate the scale of the problem. In one Washington area survey, 72 percent of businesses reported receiving customer complaints related to homeless activity nearby, while more than half cited serious safety concerns. Nearly one-third said they had considered leaving the area altogether. High-profile closures underscore these concerns: in 2022, Starbucks shut down multiple locations nationwide, citing safety issues tied to homelessness: loitering and disorder. A Tucson gas station attributed $50,000 in losses over just three months to theft by individuals experiencing homelessness. Similar stories have emerged in large metros such as Houston.
What might these dynamics mean for Tulsa? To explore this question, we look to evidence from other major metro downtown areas between 2019 and 2023 – a period during which the unsheltered homeless population in the Tulsa Continuum of Care grew by nearly 50%, and US homeless counts grew by a third. The Brookings Institute reports that foot traffic in major metro downtown areas declined by between 13% and 24% over this timeframe. While this range is not a precise measure of revenue loss – and conditions vary across cities – it serves as a reasonable proxy for the scale of potential revenue declines associated with rising homelessness and related street disorder. Given the importance of foot traffic for retail sales, especially for businesses with limited online alternatives, reductions of this magnitude are likely to translate into meaningful losses for nearby establishments.
Using Tulsa County’s share of gas station, convenience store, and fast-food employment, we simulated the economic impacts of a 13%, 19%, and 24% reduction in revenue for identical businesses in the Tulsa region using the Regional Economic Models Inc. (REMI) TaxPI+ model for Oklahoma. Inputs were adjusted to reflect revenue impacts to only those businesses within Tulsa County.
The overall, state-wide output reductions at the state level associated with the revenue declines for affected businesses in Tulsa ranged between -0.19% and -0.26% in our low scenario for Gas stations/convenience stores and fast food and other similar restaurants, respectively, to -0.35% and -0.48% in our high scenario.
Figure 9
At our midpoint estimate, reduced foot traffic due to visible homelessness and unmitigated public nuisances will lead to a loss of roughly 1,900 jobs in Tulsa County (-0.56%) and overall economic losses of $151.0M (-0.27%) to GDP, and $86M (-0.16%) to personal income for the area.
Because the problem has grown rapidly in recent years (particularly since 2019), and the states experiments with mitigation remain ongoing, we posit that most of these potential economic costs are prospective and contingent on lack of correction.
Federal agencies made more than $10.3 billion in funding available for homeless assistance in FY 2024 (up from approximately $3 billion a decade ago) – the vast majority of which flows to state and local governments in the form of grants. Based on CSI’s assessment of total homeless spending, Oklahoma’s share of this funding was approximately $70 million. While not all HUD issued grants are for the purpose of addressing homelessness specifically, most, if not all programs, directly relate to homelessness such as affordable housing development and preservation, homeownership, rental assistance, and other supportive housing and services.
Figure 10
However, the full scale of resources devoted to addressing homelessness extends far beyond the flow of federal dollars alone. State and Local governments play a significant role, as do the many nonprofit organizations operating on the ground. Within the Tulsa COC alone, CSI identified at least 24 nonprofit organizations and 13 government agencies responsible for administering direct assistance to individuals experiencing homelessness. These nonprofit organizations alone employed nearly 400 workers and relied on an estimated 7,100 volunteers.
Overall, CSI estimates that total Federal, state, and local (public funding, 32.4%), and private funding (67.6%) to support individuals and families experiencing homelessness in the Tulsa COC reached between $63.6 million and $81.2+ million annually between 2022 and 2024. Based on high and low estimates of both the total and literally homeless populations (see Appendix B), this spending equates to between $15,739 to $41,422 in annual spending per homeless and would-be homeless individual (PIT and PSH populations), and $18,324 to $58,432 for each literally homeless individual (just PIT population). Put another way, the total per-capita spending in Tulsa to alleviate homelessness ranged anywhere from 99% to 366% of the median yearly rent for all property types in the metro area.
Figure 11
Figure 12
Of the 24 non-profits providing direct services, the majority provide food, clothing, and other material support, shelter, or workforce training and other related services. Interestingly, only 4 organizations expressly provided drug addiction and treatment services to the homeless – despite extensive data suggesting the incidence of addiction issues are much higher amongst people experiencing homelessness than other populations. Based on assessments of these organizations 990 filings, impact reports, and other publicly available data, we estimate that only $2.3 million in total went towards addiction treatment, counseling, or other services intended to address drug abuse among the homeless – about 5% of overall non-profit spending. Taking into account all direct spending on homelessness, this percentage falls to about 3%.
The figures presented in this report focus solely on funding and services explicitly targeted to individuals experiencing homelessness. However, given the structure of many public assistance programs, the total level of resources reaching this population is likely substantially higher than what is captured here. For instance, individuals experiencing homelessness may also receive benefits through programs such as SNAP, but because homelessness is not an eligibility criterion for these programs, those resources are excluded from our estimates of direct homelessness spending.
Beyond the direct expenditures documented in this report, considerable resources are directed toward individuals who are not currently homeless but are at risk of becoming so. CSI estimates that this prevention-oriented funding alone accounts for an additional $95 million in resources devoted to addressing homelessness within the Tulsa COC.
Figure 13
FIgure 14
As discussed earlier, the nationwide policy response to homelessness over the last two decades has likely contributed to the growing crisis. While conditions in Tulsa have begun to deteriorate rapidly in recent years, this report acknowledges a new and ongoing effort by state law enforcement to address the issue. In September of 2025 Oklahoma Governor Kevin Stitt launched Operation SAFE (Swift Action for Families Everywhere), which utilizes Oklahoma’s Highway Patrol to clear homeless encampments on underpasses, highways and other areas of state-controlled land.xliv Individuals encountered in these operations are given various options – including possible arrest and jail time, but also shelter and treatment. While the program is still in its infancy, evidence from other jurisdictions suggests these kinds of interventions may have meaningful impacts on, at least, the nature of the problem (e.g., mitigating visible public nuisances, increasing shelter populations, and improving treatment access). More recently, the City of Tulsa – in partnership with local business leaders – has launched Safe Move Tulsa. After a similar shift in sentiments in Arizona and the City of Phoenix, law enforcement began regular raids of large homeless encampments, apprehending drug and other offenders in an attempt to improve safety in the state’s largest urban area.xlv This policy shift culminated in the City of Phoenix clearing “The Zone” – a homeless encampment in downtown Phoenix covering roughly 15 city blocks during its peak – in response to a lawsuit from local businesses establishments affected by the nearby homeless populations.xlvi Although Arizona’s overall homeless population continued to increase after these efforts gained momentum in 2022, the unsheltered population recorded its first meaningful decline in a decade. In Maricopa County, the number of unsheltered individuals fell by 19% between 2022 and 2024, according to annual PIT counts.
Mirroring efforts in Arizona and statewide in Oklahoma, Safe Move Tulsa looks to end tolerance of unmitigated urban camping while providing safer and more humane alternatives. While it is too early to say that Operation SAFE has produced similar results in Oklahoma, evidence from Phoenix and elsewhere suggests sustained coordination among state, county, and city officialsxlvii on efforts of this kind may reduce unsheltered populations, limit public nuisances, and increase access to treatment.
Tulsa’s homelessness crisis continues to worsen despite substantial public and private spending. Unsheltered homelessness has surged—driven largely by untreated addiction and mental illness—while local policy and funding priorities have emphasized permanent housing over emergency shelter, treatment capacity, and public order. The result is a system that spends tens of millions of dollars annually, yet leaves roughly one-third of people experiencing homelessness without shelter and directs only a small share of resources toward addressing the underlying drivers of homelessness.
The economic and social consequences are already visible: rising overdose deaths, reduced foot traffic, business losses, and heightened public-safety concerns. Operation SAFE represents a significant shift in approach and may help change the trajectory, but it is too early for the available data to show measurable effects. Sustained improvement will require a clearer alignment between strategy and outcomes – rebalancing the response toward treatment, adequate shelter capacity, and consistent, lawful enforcement of public-order standards – so Tulsa can reduce unsheltered homelessness and improve safety in a durable way.
Continuum-of-Care – The HUD Continuum-of-Care (COC) Program is “designed to assist sheltered and unsheltered homeless people”. It is a geographic area established by relevant organizations – state and local government and nonprofits – to fulfill specific HUD homeless program responsibilities, including operating and managing federal data collections systems like the Homeless Management Information System (HMIS) and Point-in-Time counts. Relevantly for this paper, Oklahoma’s Tulsa COC encompasses the Tulsa County region generally.
HUD – The Department of Housing and Urban Development (HUD) is the principle federal agency responsible for “addressing America’s housing needs”. As part of the national treatment of homelessness as a housing problem, HUD provides assistance (primarily as grant funding) to state and local governments and nonprofits to provide services for people experiencing homelessness, primarily through its Office of Community Planning and Development (CPD).
Literally Homeless – HUD defines as “literally homeless” (or Homeless Category 1) an “individual who lacks a fixed, regular, and adequate nighttime residence”. This population is differentiated from those who may become imminently homeless (Category 2), do not meet the definition but may be otherwise categorized as homeless under federal law (Category 3), and those who are fleeing domestic violence (Category 4). Both the PIT count and this paper attempt to estimate the true and correct “literally homeless” population.
Permanent Supportive Housing – A form of subsidized housing which is permanent (as opposed to shelter housing) in which cost assistance and (voluntary) supportive services are provided. A part of the “housing first” philosophy, Permanent Supportive Housing (PSH) is intended to address the permanent shelter needs of chronically homeless people; as such, participants in these programs are not included in the PIT counts.
Point-in-Time Count – As a condition of federal funding, all Continuums of Care are required to a conduct a Point-in-Time (PIT) count of people experiencing homelessness within their jurisdiction. The count typically occurs annually. Though intended to be a complete census of the literally homeless population, some stakeholders allege the PIT undercounts the true annual population.
Shelter – A temporary lodging facility for individuals and families experiencing homelessness, domestic abuse, or other harmful situations rendering their current shelter unsafe. This term often includes emergency shelters, transitional housing facilities, and safe haven programs.
The expenditure estimates for the homeless serving ecosystem in Tulsa can be separated into two parts. The first involves the direct expenditures by the various non-profit organizations engaged in homelessness-related services in the region. The other category encompasses all the public funding – consisting primarily of state funding – which is spent either directly, or in a pass-through fashion via grants and other processes, often to private, non-profit organizations.
The annual spending reflects an estimate of the recurring expenditures in the homeless serving ecosystem exclusively within the Tulsa COC. Because of the complexity related to revenue sources flowing into the ecosystem, total expenditures are captured, not revenue, as to avoid any double counting of funds.
CSI undertook a lengthy process to review each expense to remove any dollars that would be re-granted to other organizations, to ensure no significant double counting occurred. We also removed funding not related to homelessness relief efforts. Ultimately, the total expenses related to homelessness captured in the report are less than 12% of total expenses listed for nonprofit agencies included, and we note that given our focus on the Tulsa COC specifically, there are likely organizations who contribute resources to alleviating homelessness that were overlooked in our analysis, either because their spending within the region is not explicit, or because we lack any documentation (impact reports, annual reports, or form 990s) for their expenditures.
Public spending captured in the total reflect estimates of spending from the State of Oklahoma as well as the Tulsa COC and other regional organizations. Upper-bound estimates were calculated using all identified public spending, while the low estimates were calculated using only spending on direct services and administrative expenses for homelessness efforts to avoid the double counting of public monies that may flow to non-profit organizations.
CSI used a range to capture the homeless related expenditures relative to the number of people being served. This report relied on the best practices identified in CSI Colorado’s homelessness report and Arizona’s 2024 homelessness report as it relates to estimation of the actual homeless population. Though there is not a consensus on the best approach, the following three estimates provide a range that covers a daily average, and the total number of people who become homeless, even if just for a brief amount of time. Estimates 1, 2 and 3 show a range in the daily average number of people who are counted as homeless, as well as those in the Homeless Continuum of Care that are housed in permanent supportive housing. They are intended to provide both a lower- and upper-bound on the range of people who may be experiencing homelessness in Oklahoma and the Tulsa COC over time, and to capture two alternative measures (both the traditional or ‘literally homeless’ population and a more expansive population including those in permanent supportive housing).
Each organization’s spending is based on the location of its headquarters. However, if an organization has headquarters in Tulsa but also operates outside of the COC, we only counted estimated expenditures within the region as described below.
Several steps were taken to share out the total expenditures of the non-profit organizations covered in this report to just within the Tulsa COC.
Supporting evidence from financial reports helped to inform us of an assumed share of spending within the Tulsa COC. As an example, unless otherwise specified, if a permanent housing provider has 2/3 of building locations in the Tulsa COC, then 2/3 of spending would be assumed to occur in that area. In cases where there was not supporting evidence, spending was assumed to be distributed across regions based on the share of individuals experiencing homelessness across regions from the PIT count.
The expenditures in this analysis are not comprehensive with regards to capturing every organization that touches the homelessness ecosystem. The figures are collected from a wide range of sources, as there is no reporting requirement among all non-profits and public agencies to isolate expenditures related to homelessness. The primary well-known areas not included in this report’s expenditure analysis include the following:
