Comments
iAUDIT! - On Monday, July 14, LAHSA released the 2025 Point In Time (PIT) Count numbers. According to the count, homelessness dropped four percent countywide and a little more than three percent in LA’s City limits. Unsheltered homelessness dropped dramatically. In the LAist and LAT Times articles, some Councilmembers expressed skepticism about the numbers, as well they should. Others, like Councilmember Raman, said the alleged reduction proves City-funded programs are working, especially in her district, which seems to differ dramatically from residents' comments I've seen on social media. Mayor Bass also joined in the cheerleading, saying Inside Safe is working. Before we celebrate, consider a few key points:
Neither the City nor LAHSA really counts people in homelessness programs. They count actions. They count how many times they've sheltered or housed someone, even if that person has been sheltered several times. This has been a common theme in audits and reports for years and has not changed. Therefore, it is quite likely the only increase was the "churn" of people cycling through the system multiple times.
Regardless of whether the minor results claimed are true or not, the City cannot sustain the expenditures needed to get a few hundred more people off the streets from one year to the next. We can't spend more than $430,000 per person through Inside Safe just to get someone housed, and then have no assurance that person remains housed. We can't continue to pay outreach providers for endless "relentless outreach" that puts no limit on the number of times someone must be offered shelter before they accept. We can't afford to pay shelter operators who run their facilities in conditions that would rival medieval Newgate Prison. You can boil a pan of water with a low flame or with a flamethrower; both work but one is a lot more expensive.
Despite Mayor Bass' effusive presentation, we are spending way too much to bring far too few people inside. Success isn't just getting people off the street. It is keeping them off the street and in shelters or housing over the long term, and there's no reason to believe that's happening.
LAHSA’s count has long been criticized for chronically undercounting the unhoused population, and 2025 was no different. As reported by Angela McGregor in the Westside Current, RAND, which released its count July 1, noted serious discrepancies in the consistency of the counts in its sample areas. RAND's report says, if the miscounts are projected across the entire County, the undercount could be anywhere between 25 percent and 60 percent of the real population.
The inherent inaccuracy of LAHSA's PIT count process has been well documented for years, but is now worsened by allegations that Dr. Adams Kellum and her senior staff manipulated the numbers to make her friend Mayor Bass' programs look like they're performing better than they are. There's a reason Judge Carter ordered a monitor to be appointed specifically to verify the numbers in City and LAHSA performance reports.
Much emphasis was placed on a 9.5 percent county-wide decline in unsheltered homelessness. However, what the positive news failed to mention was that unsheltered homelessness still comprises more than 65 percent of all the unhoused in LA County and 62 percent in the City. Compare that percentage to New York City, where unsheltered homelessness is about five percent of the homeless population.
Despite assurances the count is as accurate as possible (including, for the first time, a slide on how data was recorded and reconciled), LAHSA continues to define the count as either a rough estimate or a scientific survey, depending on how it wants to spin the numbers. Slide four of the PIT count presentation says, “The Count is best interpreted as a snapshot of homelessness at the regional level”, and that it is not useful for community or neighborhood level counting. However, many LAHSA contracts are based on the counts within Service Planning Areas (SPA’s). LA County is divided into eight SPA’s. A contract for outreach services may be based on the PIT count’s estimate of unhoused people within a SPA and require a provider to contact a certain percentage of those people, with another requirement to refer a designated number to housing. So, when it suits LAHSA, it claims the count is regional or county-wide, but when it comes to allocating resources, it uses community-level numbers.
Again, despite assurances about improved data accuracy, the count’s presentation still includes a caveat, as it has in years past, that “Data provided courtesy of County Information Officer. It is possible for one person to have multiple permanent housing placements in a year”. In other words, the claim that more people than ever have been housed is nothing more than a report on how many times an unknown number of people have been cycled through the housing system multiple times. Combine the caveat on housing with findings from several reviews and audits about shelter numbers. For example, in 2019, the L.A. City Controller reported a person who enters a shelter in one month, then leaves, but returns to shelter in a different month, will be counted twice. Another City Controller audit in December 2023 stated the shelter bed availability system was so unreliable, outreach staff often called shelters to see if there were available beds. Still another audit from December 2024 found transfers from shelters to permanent housing were far below performance targets, averaging 17 percent between fiscal years 2019 and 2023. The audit also found shelter occupancy never exceeded 78 percent in any year, despite a target of 95 percent. These statistics leave us to wonder how the city and LAHSA can claim all-time shelter and housing rates; where are these people coming from and where do they go if the numbers don’t support the claims?
Finally, reducing homelessness is not a short-term proposition. Between 2017 and 2023, countywide homelessness increased 37 percent, from 55,048 to 75,518. In 2024, it might have declined by 206 people, a number well within the count’s margin or error. 2025’s numbers show a further decline to 72,308, or 3,004 people. So, despite claims of great success for no more than a few thousand fewer people on the street, homelessness remains 31 percent higher than it was in 2017. In 2017, the City spent a little more than $100 million on homelessness and LAHSA’s budget was about $180 million. Even though budgets have increased nearly ten-fold, homelessness has fallen, at most, by a few thousand. Anyone with even a basic understanding of cost-benefit analysis knows the huge expenditures cannot be justified in terms of the infinitesimal reductions in homelessness, if they exist. Bragging about two years of reductions in the hundreds after years of increases in the thousands is intellectually dishonest and morally questionable.
None of this has stopped leaders from bragging about a “trend” in declining homelessness. An online dictionary defines “trend” as “the general movement over time of a statistically detectable change”. Last year’s reduction was called “statistically insignificant” by the USC professor who helped LAHSA with data processing. This year’s reduction was plagued by missed count areas and questionable administration. Whatever one’s definition of “trend” is, it surely is not two years of dubious and minor reductions.
To understand just how shaky the claims of real reductions are, we need to look at two sources of information. The first is LAHSA’s PIT count presentation, where, on slides 3, 4, 12, and 14, it explains how the data was collected and entered. Slide Three notes “100% of [the 2025 count] data was entered digitally, with positive user feedback, enabling simpler and more efficient data reconciliation”. Keep that in mind. The second source is a June 12 LAist article on problems with the 2024 count (LAHSA says count methodology did not change between 2024 and 2025). The article notes only 81 percent of the data from the app was included in the count. Only 78 percent of volunteer counters’ site observations made it to the final count. About 2,300 fewer observations were entered in 2024 than in 2023. Therefore, the initial 2024 reduction was based on fewer numbers being entered into the system.
Now, consider LAHSA’s statement that 100 percent of the data in the 2025 count was entered digitally. Examined in light of the problems with the 2024 count, this statement has no value. It is the equivalent of saying “We entered the data digitally, but forget about the data we ignored”. It says nothing about the data that wasn’t entered--how many observations were excluded for being what LAHSA termed “unambiguously invalid” in 2024. It may very well be hundreds or thousands of people were omitted from the count when the data could not be properly entered via LAHSA’s app.
In an attempt to deflect legitimate questions about the 2024 and 2025 counts, LAHSA released a statement saying, “It is dispiriting for LAHSA staff and the thousands of volunteers who make the count happen to entertain the insinuations that politics helped drive our results”. As much as it may hurt the feelings of LAHSA’s staff to have the accuracy of its data questioned, it is hardly the first time such questions have been asked. In his recent decision appointing an independent monitor to oversee data and performance reporting from the City and LAHSA, federal Judge David O. Carter noted several reports going back almost 20 years questioning the accuracy of LAHSA’s data. As he wrote in the decision. “The creation of LAHSA has, at times, enabled a cycle of blame-shifting. When data inconsistencies or compliance issues arise, the City points to LAHSA. In turn, LAHSA attributes its problems to a lack of information or cooperation from the City. This dynamic has fostered a system in which responsibility is routinely deflected, allowing both entities to evade accountability, with no single party willing to take responsibility”. (p.11). Giving sworn testimony at a hearing in Judge Carter’s court, Emily Vaughn Henry, LAHSA’s former CIO said its homelessness data system was “smoke and mirrors” and she had been instructed “to do whatever we can to make the mayor look good.” The allegations of politically motivated data manipulation are more than “insinuations”; they are multiple accusations from different sources over a long period of time. If there is any “trend” in homelessness numbers, it is this: that the City and LAHSA have persistently failed make substantive improvements in the way they collect and report data and performance.
Almost as soon as the numbers were released, doubts were raised about their accuracy. On Friday, July 18, LAist reported LASHA revised its shelter numbers, saying 437 people previously recorded as being sheltered within City limits were actually located elsewhere. Although LAHSA told reporters the numbers were corrected before last Monday’s release, its staff neglected to inform City Council members, who expressed frustration at the lack of communication and seemingly fluid statistics. As Councilmember Lee told LAist, he’s worked to bring 371 shelter beds online, but data from LAHSA showed only 78 were occupied. He noted LAHSA has not been helpful in resolving discrepancies in the numbers.
The 2025 PIT count may be encouraging, but it must also be considered in the light of billions of dollars expended on programs with no discernable outcomes, based on data that are known to be inaccurate. The City and LAHSA need to practice serious introspection and reassess the way they evaluate their programs before they can legitimately claim progress. Based on last week’s bubbly proclamations of success, we can be reasonable sure no such self-examination will be done anytime soon.
(Tim Campbell is a longtime Westchester resident and veteran public servant who spent his career managing a municipal performance audit program. Drawing on decades of experience in government accountability, he brings a results-driven approach to civic oversight. In his iAUDIT! column for CityWatchLA, Campbell emphasizes outcomes over bureaucratic process, offering readers clear-eyed analyses of how local programs perform—and where they fall short. His work advocates for greater transparency, efficiency, and effectiveness in Los Angeles government.)
