Clearing Encampments or Resolving Homelessness?

Sweeps are a routine experience for unsheltered Angelenos, but they rarely provide a pathway indoors and are becoming more punitive. On average, one in four PATHS respondents reported experiencing a sweep in any given month. Between 2023 and 2026 the share of sweeps involving a shelter or housing offer fell by nearly half, while the share resulting in a citation rose by almost 60%.

The constraint is the short supply of meaningful offers, not an unwillingness to accept help. Two thirds of respondents who received a shelter or housing offer during a sweep accepted it, reaching 78% for motel or hotel placements and 80% for bridge or transitional housing. Even congregate shelter, which very few identified as a preferred option, was accepted by nearly half of those offered it. These findings challenge the common assumption that people choose to stay outside.

The consequences of sweeps extend well beyond being told to move. Sweep exposure was associated with increases in forced displacement, police contact, property loss, food insecurity, emergency department use, harassment, physical or sexual violence, poorer self rated physical health, and greater psychological distress.

ERPs perform substantially better than conventional sweeps in shelter yield and retention as well as service access, but their reach is limited. From December 2023 through July 2026, 22% of the cohort reported an ERP invitation, while 60% experienced a sweep over the same period. On a per person reached basis, ERPs produced roughly twelve times the shelter retention rate of sweeps, and were associated with a 10 percentage point increase in receiving help applying for services.

Better than a sweep, however, is not the same as resolving homelessness. Among people invited to an ERP and followed for six months, half never joined, and a further 23% joined but never entered interim housing. Of the roughly one quarter who did get indoors, 10% had already exited by six months. That leaves 10% still in interim housing and 7% in permanent housing, about one in six of those invited, and fewer than 2% of the full PATHS cohort.

This report weaves together two complementary sources of evidence. The first is quantitative data from the PATHS longitudinal survey. The second is narratives written by currently or recently houseless people, or poverty scholars, who have been exposed to repeated sweeps and ERP programs such as Inside Safe.

These houseless Herstories and Histories were produced through a Street Writing Workshop conducted by Aetna Street Solidarity and POOR Magazine in the San Fernando Valley, using a community led process called WeSearch that reassigns authorship and control to the people whose lives are being described. Major findings were then reviewed with the HPRI Lived Expert Advisory Board. Future PATHS reports will build on this partnership by involving WeSearchers in questionnaire design, analysis, and interpretation.

PATHS can answer this at a monthly level and cumulatively across all months of follow up. From April 2023 through July 2026, an average of one in four respondents reported experiencing a sweep in any given month. These experiences add up across multiple months: almost two thirds of returning respondents experienced a sweep in their first year in the study, and over half experienced at least two sweeps.

Rates of police contact are similarly high, driven both by sweeps and by day to day exposure to law enforcement. Six in ten respondents reported police contact in their first year. A quarter received a citation in that first year and more than one in ten were arrested. By comparison, the annual arrest rate of US adults is about 5 in 1,000, making PATHS respondents more than 20 times more likely to report an arrest. These estimates reflect only the average nine months per year in which respondents completed a survey, and arrests may lead to loss of contact with the study surveys, so these figures likely underestimate the true level of police contact.

Sweeps in Los Angeles rarely include an offer of interim housing or shelter, and such offers are becoming rarer. Just one in nine sweeps, about 11%, was associated with a shelter or housing offer, an even lower rate than the 13% reported in 2023.

The very nature of sweeps appears to have changed over time. The rate at which sweeps involved a shelter or housing offer fell by half, from about 17% in 2023 to about 9% in 2026. Over the same period the share of sweeps leading to a citation rose from about 9% to about 14%. Whereas the 2023 report noted that a sweep was more likely to lead to a citation than a shelter move, in 2026 a sweep is far more likely to lead to a citation than a shelter offer. In short, the enforcement component of sweeps has grown while the housing component has declined.

Citations carry repercussions beyond the fine itself: unpaid or unresolved citations can trigger a bench warrant for arrest and a cascade of downstream negative effects, including on housing access and employment.

These trends also mirror changes in the composition of sweep teams. Respondents most commonly reported being notified by police or law enforcement, about 63% of sweeps, followed by sanitation workers at about 38% and homeless outreach workers at about 17%. From 2023 to 2026 the share notified by outreach workers declined from 18% to 15%, while the share involving police rose from 64% to 68%.

While popular media and public sentiment often portray unhoused people as resistant to shelter, around two thirds of respondents who received an offer, 227 of 350, accepted it. Acceptance was highest for settings that met basic expectations of privacy and stability.

Across every option, the rates at which people actually accepted offers were considerably higher than the hypothetical interest they had expressed at enrollment. Among those offered a motel or hotel like setting, 78% accepted, compared with 53% who had indicated a general interest in such a stay. More strikingly, while just 10% of respondents had expressed general interest in congregate shelter, nearly half, 46%, accepted it when offered. Supportive housing offers were accepted at the same rate as motel and hotel offers.

Declining an offer, in other words, usually reflects a mismatch between what was offered and what people need, such as privacy and safety. On the whole, respondents are considerably more willing to accept whatever option is offered.

Because so few sweeps lead to shelter or housing, the impact of a sweep is felt almost entirely by people who remain on the street afterward.

Within the same person's timeline, several consequences were substantially and consistently more common in the month of a sweep than in that person's previous month without one. The largest and most consistent differences were in police contact and forced displacement, both approximately 16 percentage points higher, which correspond to the mechanics of the operation itself: a police involved event that requires someone to move. Also notable were increases in property loss, food insecurity, emergency department visits, verbal harassment, and sexual or physical violence.

These increases are measured against baselines that are already extraordinarily high. Even in months without a sweep, roughly three in four respondents experienced food insecurity, about half were displaced, one in three lost their belongings, and nearly one in four experienced physical or sexual violence. With baseline levels this high, even an increase of a few percentage points represents a substantial added burden on people already living under extreme hardship.

ERPs are pitched as a solution to unsheltered homelessness as a whole, but in practice they specifically target large encampments, typically defined as those with five or more dwellings. The high intensity of outreach and the limited availability of hotel rooms have limited the total number of operations to 136 in the City of LA's Inside Safe program and 77 in LA County's Pathway Home since they launched in December 2022 and August 2023 respectively.

Inside Safe invitations reached approximately 6% of the cohort, 83 of 1,361, while Pathway Home reached 13%, 179 of 1,361. Another 12%, 168 respondents, indicated they were unsure which program they were invited to, which points to real difficulty in identifying which agency is reaching out to them.

Invitation and sweep exposure also overlap substantially at the person level: among everyone ever invited to an ERP, 305 people, 76% had also been swept at some point, and 53% experienced a sweep in the very same survey month as an invitation.

If the goal was to design an encampment operation that fares better than a conventional sweep, ERPs appear to accomplish it. They are more likely to bring people indoors, more likely to connect them with services, and less likely to expose them to enforcement, displacement, and property loss.

But ERPs are costly and resource intensive operations that reach a relatively small share of the unsheltered population, and their apparent success depends heavily on which outcomes are measured and which people remain in the denominator. A retention rate of approximately 60% among those who enter interim housing may appear encouraging, but it excludes people who were never invited, those who did not enroll, those who enrolled but never received a placement, and those who returned outside. These groups constitute most of the people encountered through the intervention.

An operation may successfully clear a location while leaving most of the people who lived there homeless and increasingly scattered, from the relative safety of large encampments in accessible locations into rough sleeping or smaller encampments in isolated places where they become even harder to reach. If the objective is resolving homelessness rather than resolving the visible presence of an encampment, the current results remain inadequate.

Current definitions of encampment resolution can further obscure this distinction by treating site clearance as an outcome alongside placement into durable housing. An operation may be classified as resolved even when only 26% of those invited enter interim housing and 7.5% reach permanent housing within six months. Site clearance, interim placement, continued residence indoors, and entry into permanent housing are different outcomes and should be reported as such.

With approximately 21 months remaining before the Games, improving ERPs will be important, but it will not be enough. The findings point to three priorities.

1. Expand access to permanent housing within the remaining time. Los Angeles cannot build all the affordable housing it needs before the Games, but it can examine whether additional housing can be secured through rental subsidies, flexible financial assistance, master leasing, property acquisition or conversion, and stronger incentives for landlords accepting tenants with vouchers. Without more housing at the end of the pathway, interim placements will remain bottlenecks instead of bridges to permanent homes.

2. Strengthen the pathway from outreach to permanent housing. ERPs need multiple placement options, interim settings that preserve greater autonomy, and continued engagement when an initial offer does not result in placement. People may experience repeated sweeps, but ERP invitations are much less likely to recur: Los Angeles currently appears better equipped to return and move someone again than to return with a different offer of assistance. People with lived experience should be paid a living wage, provided sufficient hours to build careers from their expertise, and given meaningful roles in program design, implementation, evaluation, and governance.

3. Develop safe options for people who cannot be placed indoors. Most people living outside are never reached by ERPs or do not get placed in interim housing. If they cannot remain where they are and no appropriate indoor option exists, they still need somewhere safe and lawful to go. Repeatedly directing people to another sidewalk merely transfers the problem and exposes them to another cycle of displacement. Los Angeles should consider a broader range of alternatives, including safe parking and sanctioned camping programs that provide sanitation, water, storage, health care, outreach, transportation, and continuing pathways toward housing. Such programs are not substitutes for permanent housing, but the absence of any viable alternative leaves repeated displacement as the default.