The Brief

The 2024 Los Angeles homeless count brought what appeared to be a rare piece of good news: a 10% drop in unsheltered homelessness inside city limits.

The decline was a bright spot after years of growing homelessness across L.A. County. Since 2018, each homeless count for the region — a requirement for seeking federal funding — brought more bad news. Year after year, even as the region spent more and more money on the homelessness crisis, the number of unhoused people kept going up.

L.A. Mayor Karen Bass and Va Lecia Adams Kellum, the outgoing chief executive of the Los Angeles Homeless Services Authority, have pointed to the drop in 2024 as a sign that major homelessness initiatives are working.

The results of the count surprised some Los Angeles residents and advocates who said the celebrated drop in homelessness doesn’t match what they’ve seen on the streets.

To better understand how LAHSA reached their conclusions, LAist requested the policies and raw data behind the official results announced last June.

We found the 2024 homeless count was complicated by policy changes at LAHSA, shifting guidelines and technical problems — with marked differences in how people were counted in the city of L.A. and in other areas of L.A. County.

How the count works

Volunteers and LAHSA staff canvassed the county across more than 3,000 census tracts covering nearly 4,000 square miles, entering their observations on a phone app and, when the technology didn’t work as intended, on paper forms filled out by hand.

Ultimately, each count is an estimate based on a “point in time,” when homelessness is observed by volunteers throughout the region. The raw data gathered is then reviewed by LAHSA officials for verification. The count’s results typically come out months later.

Marina Flores, left, and Helde Pereira, right, document homeless people seen during LAHSA's annual count earlier this year.

(

Carlin Stiehl

/

LAist

)

LAist’s analysis found LAHSA excluded more observations recorded by volunteers looking for people living outside or signs of homelessness in 2024 than the previous year. LAist also found LAHSA excluded significantly more observations from census tracts within the city of L.A. in 2024 than those in the rest of the county.

Let’s take a closer look at what we found.

People and dwellings

The unsheltered homeless count is based on two categories: People observed to be unhoused and temporary dwellings where people appear to be staying, like tents, encampments, RVs and other vehicles. When all of the observations are validated by LAHSA, a separate annual survey conducted by University of Southern California is used to estimate how many people are living in those temporary dwellings.

• In the 2023 count, LAHSA included 87% of all observation data that volunteers across the region had entered through the phone app. The other 13% was removed from the count and not replaced by data from paper backup forms.
• In the 2024 count, just 81% of app data from volunteers was included — in raw numbers that meant about 2,300 more observations of people and dwellings were dropped than the year before.

Inside and outside city limits

LAist also examined observations by jurisdiction and found:

• 87% of observations made on the app outside the Los Angeles city limits were included in the final 2024 count, similar to the inclusion rates for both the city and the rest of the county in the 2023 count.
• In contrast, 78% of observations inside the city of L.A. made it to the final count in 2024. LAist found about 2,300 more observations were dropped from the city’s count, accounting for almost all of the additional observations LAHSA removed countywide.

What we know about the observations LAHSA removed

LAist found nine areas — all inside Los Angeles city limits — where more than 100 observations of homelessness were removed from LAHSA’s preliminary totals of app and paper data by the time the count was finalized.

LAHSA officials said that, in many areas, the final count was higher than what was observed in preliminary data. When they did exclude data, LAHSA officials said it was due to errors in the entries that, in some cases, made the preliminary data “unambiguously invalid.”

“When you’re working with such a large number of volunteers, human error is part of the process,” LAHSA spokesperson Ahmad Chapman said. “LAHSA staff accounts for this and are thoughtful throughout the process as we look at the data.”

We found the agency made decisions about what data to exclude on a case-by-case basis and lacked documentation explaining some of their decisions.

“LAHSA employs the data reconciliation process to improve the accuracy of the Homeless Count, not to fulfill a narrative,” Chapman said in response to our analysis.

Chapman said the count is meant to be an estimate.

“It’s important to note that Homeless Count data is just one measure of our system. Albeit an important one, it plays its own specific and limited role,” Chapman said.

Zach Seidl, a spokesperson for Bass, responded to LAist’s findings in a statement, saying that the 2024 homeless count results “were certified by the federal government and there is still more work to be done."

Councilmember Monica Rodriguez, who sat on the city’s housing and homelessness committee until the end of 2024, is one local official questioning the results of the annual count.

“ Are we measuring apples to apples?” Rodriguez said in an interview with LAist. “Or are people feeling compelled to manufacture reductions to convey to the public that they're actually garnering some type of progress.”

The data and discrepancies

By the first night of the homeless count, Bass had been in office for a little over a year. Her administration had targeted homelessness with a state of emergency issued on Day One of her term and already spent hundreds of millions tackling the problem.

The 2024 count was the first big test to see if the mayor’s initiatives were working.

Volunteers and LAHSA staff fanned across Los Angeles County starting on Jan. 23, 2024. The count took three days, with most of the work taking place at night.

Volunteers were given a brief training on how to identify signs of homelessness and were told to mark those observations in an app made by Esri, a software company based in Redlands.

The volunteers were instructed to mark observations only in their assigned census tract. GPS from the app would help discard observations made outside the assigned areas to prevent duplication.

Technology problems popped up shortly after volunteers arrived at their designated counting areas and persisted all three nights of the count, according to LAHSA’s reports on the 2024 count and volunteers interviewed by LAist.

Several volunteers told LAist they had trouble logging into the app and had to wait hours to start counting.

Data entry errors were common, according to two volunteers interviewed by LAist.

For example, volunteers in North Hollywood used the app to count more than 50 people living outdoors near a homeless shelter operated by the nonprofit LA Family Housing.

LAHSA officials say the volunteer site coordinator at LA Family Housing checked a box on the app’s review page indicating that no unhoused people were counted. LAHSA deferred to that statement rather than the data gathered by volunteers, even though data was submitted on the app, in this case.

A representative for LA Family Housing confirmed the site coordinator had made an error when they checked the box in the app and recalled computer screen glitches.

Dozens of unhoused people live in the census tract, near the LA Family Housing shelter site.

LAist visited the area and spoke with residents. Timothy Woodhead has been living in a makeshift shelter there for years and is skeptical L.A. is making progress on homelessness.

“Maybe they've gotten a lot more people inside the tiny homes and stuff like that, but there are still plenty of homeless people out here,” Woodhead said. “They kicked me out [of LA Family Housing] like two years ago, and I've just been right here next to them, in the streets. ”

Timothy Woodhead has been living in makeshift shelters in North Hollywood for years. In 2024, LAHSA didn't include Woodhead and his unhoused neighbors in their annual Homeless Count due to an error.

(

Aaron Schrank

/

LAist

)

Two volunteers told LAist there was no easy way to fix app entries made in error, so they relied on paper maps as a backup.

According to count data, LAHSA relied on data from paper backup maps for more than 400 of the 3,249 assigned areas counted in 2024.

But data recorded on the paper wasn’t always included in the final, publicly released count data, even when volunteers intended it to be used in place of app data.

David Hirschman and his wife have regularly volunteered for the homeless count since before LAHSA unveiled its first counting app in 2022.

“ They have made improvements on the app over the years, but my experience was not a great user experience, not a great user interface,” Hirschman said, recalling the 2024 count. “We had to use the paper and highlighters to mark where we went because the app just wasn't good.”

Hirschman was assigned to count a several-block stretch of Chatsworth, where dozens of unhoused Angelenos were living in tents and RVs near a county social services office.

“There's literally blocks and blocks where it's just wall-to-wall homeless folks,” he told LAist. “There’s so many, it’s hard to count accurately.”

Hirschman and other volunteers recorded more than 180 observations of homelessness in the mobile app for this area and 105 on the paper form, but none were recorded in final data from LAHSA.

Chapman, the LAHSA spokesperson, told LAist this was an error caused by the volunteer site coordinator in Chatsworth, who had submitted a note in the app dashboard indicating that the area’s homeless count should come from a paper back-up form instead of the app.

But Chapman said the Chatsworth observations recorded on paper were not included in a separate spreadsheet LAHSA used to record what was written on the paper forms. LAHSA was not able to investigate every instance where contradictory data was provided and relied on the spreadsheet “to complete the job in time as accurately and defensibly as possible,” Chapman told LAist. In this case, that led to none of the area’s observations being included in the final count.

In a 2024 report to the L.A. County Board of Supervisors weeks after the count took place, LAHSA officials said issues with the app “increased doubts and concerns about the technology and reliability of the data.”

Bevin Kuhn, who is responsible for the management of homeless count data as LAHSA’s deputy chief analytics officer, told LAist that the issues they faced in 2024 with the app are now fixed, explaining that they were addressed in a custom app developed with Esri for the 2025 count.

Chapman told LAist in a separate email that all count data was recorded on this app in 2025, which “alleviated the need for a reconciliation process as complex as in 2024.”

What happened after the initial count?

After three days of data entry in January 2024, LAHSA officials moved to the next step: cleaning the data by removing duplicates or mistaken entries.

Chapman told LAist that four to six LAHSA staff have just three weeks to reconcile the data.

Much of the data processing for the 2023 count was automated, using a computer program with a specific set of instructions, according to documents LAHSA provided to LAist after a public records request.

LAHSA didn’t use a similar system in 2024 because the staff member who wrote the program was on leave, according to Bryan Brown, who is associate director of data management at LAHSA and helped lead data processing.

Brown said the remaining staff was “recreating the process to the extent the rest of the team knew” what that program had done.

LAist requested LAHSA’s policies regarding how the 2024 data was processed.

Brown said that while there was no written policy, the process was “understood by the small team that works on this year over year.”

LAHSA invited LAist reporters to their downtown L.A. office to walk through how they collected and cleaned the data of duplicates and errors.

The process was “adjusted” as the LAHSA team worked through the data, Brown said, influenced by unique circumstances in each assigned area.

“It was a nimble process,” Brown said, where they established guidelines that then had to be modified.

The number of observations volunteers submitted on the app could be higher or lower than what LAHSA verified after considering all the data, he told LAist. “It truly varies tract to tract," he said, "situation to situation, times 3,000.”

The agency shared the document below that they created in 2025 in response to LAist questions about how their process in 2024 worked.

LAHSA provided LAist with a flow chart that was made after the count to describe how they processed their data.

(

LAHSA

)

Louis Abramson, an adjunct senior physical scientist at RAND who researches homelessness in L.A., told LAist that transparency, simplicity and consistency from year to year are key when counting such a large number of people. He said he is concerned that changes to how LAHSA conducted the count opened the door to more error.

Apart from his work with RAND, Abramson was a founding board member of the homelessness-focused nonprofit Hollywood 4WRD and volunteered in the LAHSA count in 2024.

Abramson said that the area in Hollywood where he volunteered was recounted by a professional team from LAHSA after Hollywood 4WRD wrote a letter to the L.A. County Board of Supervisors saying that issues with the app compromised their ability to “ensure volunteer safety, data integrity, and to recover physical materials.”

“I think the most damaging thing of 2024 was that confidence was eroded,” Abramson told LAist. “The mechanism that [LAHSA] used to deliver this number became complex enough such that there were lots of ways it could go wrong, and it did go wrong in some of those ways.”

Questions from the community

Two weeks after the count was conducted in January 2024, Jonah Glickman from L.A. County Supervisor Lindsey Horvath’s office reached out to LAHSA with questions, according to emails LAist acquired through a public records request to LAHSA.

Horvath’s district includes the L.A. County’s coast from Venice, where homelessness has been the subject of tension, up to the Ventura County line. Glickman asked about beaches that community members believed had gone uncounted in 2024, and whether LAHSA could provide “a detailed breakdown of the data confirmation process that could be shared with the community.”

Through public records requests, LAist uncovered an internal email conversation between Brown and Sally Malone, LAHSA’s director of government affairs, about Glickman’s question.

“We don’t have a robust ‘official’ documented process,” Brown wrote to Malone. “In part because the back-end and how the system actually operated during the count has changed so significantly year-over-year.” Brown did share partial documentation of the process with Malone, but the agency did not share that information with Horvath’s office.

LAHSA eventually confirmed in an email to Horvath’s office that “Homeless Count volunteers may not have consistently counted some sandy beach areas.”

A spokesperson for Horvath told LAist that LAHSA never responded to the second question from Glickman about their data validation process.

Questions about the accuracy of the count — again an estimate at a point in time — are far from new.

"My staff and I have led the homeless count in my district for years, and it is hard to remember a time where we felt that we could trust the data 100%,” Councilmember Bob Blumenfield said in an emailed statement. Blumenfield said he has raised concerns about the count accuracy to LAHSA before, but the agency’s responses “lacked the urgency I was hoping for.”

A 2017 Economic Roundtable report found L.A.’s homeless count lacked reliable year-to-year comparability, because of inconsistent methods and data collection across different counts.

Pete White, executive director of L.A. CAN, a nonprofit that does housing advocacy work in Skid Row, said the drop in homelessness observed in the 2024 count didn’t match what he was seeing in the streets.

“We have always known that the homeless count methodology was flawed and resulted in undercounts, year in and year out,” White said. “The naked eye could see the inconsistency of reported decreases and increased visible homelessness.”

Nearly half of LAist readers surveyed last summer reported seeing homelessness increasing in their neighborhoods in 2023 and the first half of 2024, while 28% reported decreases.

Are homeless initiatives working?

Some homelessness experts say they do take L.A.'s 2024 homeless count reduction as a sign that the city’s programs such as Inside Safe, which provides city-funded hotel rooms to bring people off the streets, are showing progress.

Mayor Karen Bass discusses the 2025 homeless count.

(

Carlin Stiehl

/

LAist

)

“What we're seeing in other parts of the country, a focus on punitive measures and criminalization, that doesn't work,” said Alex Visotsky, Senior California Policy Fellow at the National Alliance to End Homelessness based in Washington, D.C. “Focusing on getting folks back into housing seems to be bending the curve.”

Others say increased enforcement of anti-camping policies across L.A. County may just be depressing the count.

“Code enforcement, criminalization and sweeps have driven the unhoused deeper into the shadows, and reporting is even more challenging,” said Peter Connery, vice president of Applied Survey Research, a nonprofit research organization that has been contracted to conduct dozens of homeless counts for municipalities in the Western U.S., including L.A.’s first-ever homeless count in 2005.

The homeless count should be considered a “minimum count,” Connery said, because it only reflects what volunteers can see from the street.

Over the past year, LAHSA has been criticized for failing to properly track billions of dollars it spends on services. L.A. County supervisors pulled funding from the agency and a federal judge is currently considering handing control of L.A. city homelessness spending to a court-appointed receiver. The Los Angeles City Council also recently voted to address inconsistent and fragmented data collection in the city’s homeless services system and look for pathways to separate from LAHSA.

Adams Kellum told LAHSA commissioners last month that the L.A. city budget and loss of funding from the county could jeopardize next year’s count.

With this backdrop, LAHSA officials released data from the February 2025 count in March, months earlier than usual. The agency said the data was preliminary and could change, but it once again showed a major drop in unsheltered homelessness that city officials say show their programs are working, despite the mounting criticism of LAHSA from public officials.

In a LAHSA Commission meeting on April 25, Paul Rubenstein, a LAHSA spokesperson, said the reason they released the preliminary estimates was because the agency “felt like this was very important information for stakeholders to have as they were considering significant shifts to the system.”

White, of L.A. CAN, remains skeptical.

LAHSA “released premature data to support City Hall’s narrative of plummeting numbers,” White told LAist. “It has been weaponized for its own political purpose.”

L.A. has been in federal court over allegations the city has not complied with terms of a settlement agreement that laid out milestones to reduce homelessness.

“My view is that they’re in a political battle for their lives right now,” Federal Judge David O. Carter said during a court hearing about the settlement in March. Carter, the judge considering pulling control of homeless spending from the city, said LAHSA’s release of preliminary data could be “political gamesmanship.”