7,900 Missing from LAHSA’s Homeless Count

RAND has different homeless numbers than LAHSA.

LAist released an October 15 story “LA’s Official Homeless Tally Increasingly Undercounts People on the Streets, RAND Study Finds,” noting that RAND researchers also found the Los Angeles Homeless Services Authority’s (LAHSA) annual count decreased in accuracy over the past two years.

Tim Campbell a Westchester resident who spent a career in the public service and managed a municipal performance audit program wrote to CTN that “RAND’s count, which is conducted by professionals in targeted areas, found significant discrepancies from LAHSA’s February 2025 PIT count.

“That’s important because RAND counts three areas with high homeless populations: Venice, Hollywood, and Skid Row,” Campbell said. “Undercounts in these areas have an outsized effect on the overall count.”

The LAist story click here notes that “LAHSA’s official count in those Los Angeles neighborhoods previously had aligned with RAND’s own estimates, but a 26% undercount emerged in 2024 and widened to 32% this year, RAND found.

“‘In 2022 and 2023, at least in the areas where we count, the [Point-in-Time count] was, as far as I can tell, excellent,” said Louis Abramson, lead author of the RAND report. “I don’t know what has happened afterwards.”

Hollywood’s official tally captured 81% of what RAND found, Venice was at 76% and Skid Row’s was just 61% accurate.

If the pattern RAND holds, it means up to 7,900 people and dwellings may be missing from the city’s most recent count.

Campbell writes, “Why is the 7,900 number so important? First, it’s more than the combined 2024 and 2025 decreases in the City’s homeless population trumpeted by LAHSA and Mayor Bass.  At best, homelessness has stayed the same.

“Second, major funding decisions use the count to set priorities. As LAHSA’s local jurisdiction count web page says, the count will be used to determine areas where new Measure A funding will be targeted,” Campbell said and added “It’s bad enough the agencies that will receive Measure A money sit on the board that makes funding recommendations, but now they’ll make those recommendations using bad data.”