Skip to main content
Original research · Elevyr · California RCFE roster

California gained senior-care beds. Seventeen counties lost them.

An analysis of every residential care facility for the elderly on California’s licensing roster, comparing beds open on 31 December 2020 against beds open on 31 December 2024. The statewide number is flat and close to useless. Underneath it, almost a third of the state’s capacity closed and was rebuilt somewhere else.

+2.8%

Statewide change in licensed beds, 2020 to 2024

29.9%

Of the 2020 bed stock closed during the window

17 of 36

Measurable counties that ended with fewer beds

2

Counties that ended with no licensed beds at all

Download the dataset (CSV, 50 counties)No email, no form, no signup.

California licensed 201,720 residential care beds for older adults as of 31 December 2020. Four years later it licensed 207,439. The state added 5,719 beds, a gain of 2.8 percent.

That number is true and it is close to useless.

Underneath it, 3,175 facilities closed. They took 60,390 beds with them, which is 29.9 percent of everything the state had open at the start of the window. California did not hold steady for four years. It tore down nearly a third of its capacity and built back slightly more than it lost, and it built back in different places.

Seventeen of the 36 counties large enough to measure ended the period with fewer beds than they started with. Two counties ended it with none.

The counties that lost beds

Counties with at least 250 licensed beds in 2020. Thirty-six of the state’s 50 counties with licensed capacity clear that floor. The full 50-county table, including the small ones, is in the CSV.

CountyBeds 2020Beds 2024NetChangeClosures in window
Imperial557247-310-55.7%7 facilities, 525 beds
Tuolumne307175-132-43.0%4 facilities, 357 beds
Sutter500416-84-16.8%2 facilities, 84 beds
Yolo1,4521,248-204-14.0%12 facilities, 395 beds
El Dorado947859-88-9.3%12 facilities, 179 beds
Humboldt606550-56-9.2%7 facilities, 110 beds
Tulare1,1211,020-101-9.0%17 facilities, 315 beds
Butte1,1161,038-78-7.0%21 facilities, 318 beds
Napa1,2491,166-83-6.6%22 facilities, 409 beds
Nevada541517-24-4.4%3 facilities, 87 beds
Santa Cruz1,1131,064-49-4.4%10 facilities, 211 beds
Alameda9,3939,072-321-3.4%100 facilities, 2,042 beds
Santa Barbara3,8993,772-127-3.3%40 facilities, 943 beds
San Mateo5,4225,254-168-3.1%66 facilities, 1,304 beds
Monterey1,9081,865-43-2.3%20 facilities, 791 beds
Sacramento10,90510,759-146-1.3%289 facilities, 4,784 beds
Stanislaus2,8242,807-17-0.6%43 facilities, 1,118 beds

Imperial County lost more than half its licensed elder-care capacity in four years. Tuolumne lost 43 percent. Neither county is large, and that is the point. When a county has 557 beds and seven buildings close, there is no second operator down the road.

Two counties now have none

Inyo and Plumas both had licensed residential care beds at the end of 2020 and had none at the end of 2024. Each lost a single facility and had nothing else. Inyo’s was licensed for 76 beds and closed in July 2021. Plumas’s was licensed for four and closed in December 2021.

Both counties sit below the 250-bed floor used in the tables above, which is exactly why a county-size threshold is a reporting convenience and not a judgment about who matters. A resident of Inyo County has no licensed residential care option inside the county line at all.

The counties that gained

CountyBeds 2020Beds 2024NetChange
Placer3,5774,159+582+16.3%
Shasta1,0851,207+122+11.2%
Orange23,03225,085+2,053+8.9%
Ventura5,7456,252+507+8.8%
Santa Clara9,31610,028+712+7.6%

Orange County alone added 2,053 beds. That is more than the entire 2024 licensed capacity of Sutter, Tuolumne, Imperial and Nevada counties combined.

Closing is normal. Replacement is not.

The more interesting column is not net change. It is churn, meaning how much of a county’s 2020 capacity went dark at some point in the four years.

CountyBeds closed, share of 2020 capacityFacilities closedNet change
Tuolumne116.3%4-43.0%
Imperial94.3%7-55.7%
Sacramento43.9%289-1.3%
Monterey41.5%20-2.3%
Stanislaus39.6%43-0.6%
Riverside39.3%225+1.6%

Tuolumne is above 100 percent because facilities licensed after 2020 also closed before 2024. The county cycled through more beds than it ever held at once.

Sacramento and Riverside are the comparison that matters. Sacramento lost 289 facilities holding 4,784 beds and finished the period down 1.3 percent. Riverside lost 225 facilities holding a comparable share of its base and finished up 1.6 percent. Both counties replaced almost everything they lost.

Imperial and Tuolumne lost a similar share of their capacity and replaced none of it.

So closure is not the variable that decides whether a county keeps its beds. Every county loses buildings. Some counties get new ones.

What this data does not say

This is licensed capacity, not occupancy. California’s public roster reports the number of beds a facility is licensed to operate. It does not report how many are filled, so nothing here measures demand, waitlists or whether a county has a shortage. A county can gain licensed beds and still have families driving two hours to visit.

It also does not explain why any facility closed. The roster records a closure date and nothing else. Retirement, sale, enforcement action and bankruptcy all look identical in this file.

And it covers residential care facilities for the elderly only. Skilled nursing facilities are licensed under a separate state program and do not appear in this file.

Methodology

Source

California’s most recent published licensing extract for Residential Care Facilities for the Elderly, dated 25 May 2025. It contains 12,522 facility records: 7,939 licensed, 3,821 closed, 739 pending and 23 on probation. Two facility types appear and both are included, 12,389 records of standard residential care for the elderly and 132 continuing care retirement communities. One record carries no facility type and is excluded below.

The extract is 16 months old

25 May 2025 is the newest file the state publishes as of 22 September 2026. It was re-pulled on 21 September 2026 and is byte for byte identical to the copy analyzed here, md5 89d372790d0505196667f3ba17e824f0. Every figure on this page therefore describes the state’s own most recent published accounting, not a live count.

What was measured

Beds open on 31 December 2020 against beds open on 31 December 2024. A facility counts as open on a given date when its status is not pending, its first license date is on or before that date or is blank, and its closure date is after that date or is blank. Both endpoints measure the same quantity, which makes the comparison immune to facilities that opened and closed inside the window.

An earlier version of this analysis measured closures against the 2020 stock and reported a percentage loss for each county. That method was discarded before publication because it lets a churn-heavy county report a loss larger than the capacity it ever held. The churn table above reports that ratio as churn, which is what it actually is, and never as a loss.

Closure coverage

Closure dates in this file begin 27 May 2020 and run through 19 May 2025. A facility that closed before May 2020 carries no closure date, so the opening balance counts only facilities the file shows as having existed into the window.

Exclusions

739 pending records that were never licensed, one of which also carries no county. Eleven records carrying a closure date while currently showing an active status, where the active status was taken as correct and the closure ignored. Fourteen counties holding 1,793 beds between them fall below the 250-bed floor and are excluded from the county tables above, but they are included in the statewide totals and in the downloadable CSV.

Parsing

The file was read with a standards-compliant CSV parser, not by splitting on commas. Facility addresses in this roster contain commas, and a naive split silently shifts every column after the address field. Two runs on the same file produce identical output.

Corrections

If you find an error in this analysis, write to ed@elevyr.com. Corrections get made on this page with a dated note, not quietly.

Disclosure

Elevyr sells an AI voice phone agent to senior living communities, including operators in the counties named here. Nothing in this analysis is a claim about any individual facility, and no operator was contacted before publication. The counts come from the state’s file and can be checked against it by anyone.

Cite this page

These figures may be reproduced with attribution and a link. No permission form, no email required.

Plain text

Elevyr, "California Senior-Care Bed Loss by County, 2020 to 2024," September 2026, https://www.elevyr.com/data/california-rcfe

HTML

<a href="https://www.elevyr.com/data/california-rcfe">Elevyr, "California Senior-Care Bed Loss by County, 2020 to 2024," September 2026</a>

Press and data requests go to ed@elevyr.com.

Sloane, our AI Admissions Agent for senior living, exists because of what happens after a family finds a community that still has a bed. Whether that inquiry reaches a person is a different question from whether the bed exists.