Encampments: why sweeps don't work, and what does

Tent encampments are the most politically charged feature of contemporary American and Canadian homelessness. They are visible. They generate complaints. They appear, are cleared, and reappear. They have driven cities to spend billions of dollars on enforcement that does not, by any rigorous measure, reduce the underlying homelessness rate.

This article is about what works and what doesn't, based on the research.

What an encampment sweep is

A sweep is a coordinated effort by police, sanitation, or parks workers to remove people and their belongings from a public space. Sweeps typically:

• Give 24-72 hours notice (sometimes less; sometimes none).
• Disperse residents to other public spaces.
• Confiscate or destroy belongings deemed "abandoned."
• Cite or arrest residents who refuse to leave.

A small share of sweeps include housing offers — usually a shelter bed, sometimes a hotel voucher. Most do not include any housing offer. Even when offers are made, the offered shelter is frequently unsuitable for the people being displaced (no pets, no partners, sobriety requirements, curfews that conflict with work, gendered intake that excludes trans people, etc.), and acceptance rates are often under 30%.

What the evidence shows about sweeps

Multiple peer-reviewed studies and government reports have evaluated sweep policies. The findings converge:

• Sweeps do not reduce the number of unhoused people in a city. They relocate them. Population studies before and after sweep waves consistently find unchanged or higher counts.
• Sweeps increase mortality. A 2023 University of Colorado study estimated that aggressive enforcement of camping bans in 23 cities was associated with significantly elevated death rates among unsheltered residents, primarily through loss of medications, disruption of medical care, and exposure.
• Sweeps make future housing harder. Confiscated IDs, lost medications, and new misdemeanor or felony records on a person's history all create barriers to entering shelter, getting benefits, and being approved for an apartment.
• Sweeps are expensive. A 2020 Seattle audit estimated the per-cleanup cost at $15,000-$30,000 when accounting for police, sanitation, social workers, and storage. A 2022 LA City Controller audit put per-camp removal costs above $50,000. These figures are routinely higher than what it would cost to provide voluntary outreach, storage, and a housing voucher to the displaced residents.
• Sweeps damage trust in outreach. Outreach workers report that after sweep waves, residents become much harder to find and much less willing to engage with caseworkers who they reasonably suspect of being affiliated with enforcement.

What actually clears encampments

Cities that have substantially reduced their encampment populations — Houston, parts of Salt Lake City, Helsinki, Anchorage during a focused period — all share a common pattern:

1. A real housing offer per resident

Not a shelter offer. Not a "we have a bed somewhere." A specific, viable, voluntary offer of permanent or transitional housing tailored to the resident's situation — for example, a unit that allows their partner and pet, that doesn't require sobriety, that is reachable from where they work or get healthcare.

This is the binding constraint. Without housing on the receiving end of the sweep, there is no sweep, only a relocation.

2. Sustained, voluntary outreach before any clearance

Outreach workers build trust over weeks or months, learning who lives at the camp, what each person's needs are, and what housing offer would actually work. This is the slow expensive part. Cities that try to skip it get failed sweeps.

3. Coordinated entry that produces real placements

The encampment connects to the local Continuum of Care system. Each resident gets a coordinated-entry assessment. The CoC assigns them to a housing program. The program has actual vacancies.

If any of these three steps is missing — particularly the third — the model breaks down. A lot of cities have the first two steps and skip the third because their housing pipeline is too small.

4. Time and trust

Houston's reduction in unsheltered homelessness took years of consistent execution. There is no two-week version. Cities that promise rapid results almost always end up doing sweeps and calling them success.

A specific example: Houston

Houston has cut unsheltered homelessness by roughly 60% since 2011. They did this not by criminalizing camping, but by building a coordinated system that places about 4,000 people per year into permanent supportive housing or rapid rehousing. They use outreach teams to identify camp residents, do coordinated-entry assessments on site, and present each person with a tailored housing offer. The encampments are then cleared with the resident's voluntary participation — and stay cleared because the residents are housed.

This took twelve years. It cost less per person than enforcement-based approaches in comparable cities. It is the model that the federal government (HUD) holds up as the gold standard.

A specific example: Helsinki

Finland eliminated long-term street homelessness almost entirely between 2008 and 2020. The approach was straightforward: build or buy units, hand them out via Housing First, and provide voluntary supportive services. There was no enforcement campaign. There were no sweeps. The population was rehoused by being given housing.

Helsinki's chronic homelessness is now below 50 people for a metro area of 1.5 million.

What individuals can do

If your city is considering encampment sweeps:

• Ask elected officials specifically: "Is there a housing offer per resident, before clearance?" If not, the sweep will fail by every metric except optics.
• Support the Housing First pipeline. The binding constraint is housing supply at the low end and rental vouchers.
• Oppose ordinances that criminalize sleeping outside without offering an alternative. The 2024 US Supreme Court case Grants Pass v. Johnson held that such ordinances are constitutional, but constitutionality is not effectiveness — cities are free to choose otherwise.

If you encounter an encampment in your neighborhood:

• Don't call police as a first response. Call the local outreach team (211 will route you).
• Do call about specific safety issues (open fires, weapons, medical emergencies). The right response there is not enforcement, it's emergency services.
• Support — politically and with money — the local providers doing outreach and housing placement.

Bottom line

The choice cities face is not "sweep the camp" or "do nothing." It is "sweep the camp" or "build the housing pipeline that makes the camp empty itself." The first option is faster, more politically appealing, and reliably ineffective. The second is slower, less appealing, and reliably works.

Two decades of research are unusually consistent on this. The cities that are reducing unsheltered homelessness are doing the second thing.

Read more: What actually works: Housing First · Common myths about homelessness.