We have written before that a tax-sale opening bid is not a price — it is the amount needed to redeem the parcel, so it tracks debt rather than value. That was inferred from a regression on one county’s list. Los Angeles lets us test it directly, because its Board of Supervisors publishes something almost no other California county does: the year each parcel first went tax-defaulted.
Put that beside the assessor’s current values for all 1,252 parcels in the October 17–20 sale and the story stops being an inference.
The finding
Take only the parcels Los Angeles assesses at more than $10,000, so nobody can argue the denominator is a rounding artifact, and keep the 697 of those for which the county also records a default year. Sort them by how long they have been in default.
| Years in default | Parcels | Median bid ÷ assessed | Opens above assessed |
|---|---|---|---|
| Under 5 | 93 | 0.11× | 0% |
| 5–10 | 183 | 0.15× | 1.1% |
| 10–20 | 214 | 0.70× | 14% |
| 20–30 | 55 | 1.24× | 78% |
| 30 or more | 152 | 1.75× | 98.7% |
Not one of the 93 parcels defaulted less than five years opens above the county’s own valuation. Nearly every parcel defaulted more than thirty years does. The land did not change. The clock ran.
This is what “the opening bid is the redemption amount” means when you can actually see the time axis. Every year a parcel sits in default, another tax bill, another 18% redemption penalty and another set of county costs land on the floor under it. The floor only goes up. Assessed value, meanwhile, is pinned by Proposition 13 to a base year and creeps at 2% — and for these parcels the base year is often the 1970s or 80s, because a parcel in default has not changed hands. One number compounds and the other is frozen. They cross, and after they cross the county is asking more than it says the parcel is worth.
The crossing point in this sale is somewhere between ten and twenty years. That is a useful thing to carry to any California list, whether or not the county prints the default year.
What this list actually is
It is worth being concrete about what “the largest tax-sale list in California this year” contains, because it is not what the name suggests.
764 of the 1,252 parcels — 61% — are in the high desert, in Hi Vista, Black Butte, Butte Valley, Roosevelt, Redman, Alpine Butte and Antelope Acres. They are the residue of mid-century paper subdivisions: land platted into lots during a boom that never arrived, sold to buyers who mostly never visited, and quietly abandoned to the tax roll. Their median parcel has been in default 19 years and 69% of them open above assessed value. Across the other 488 parcels in the county the median default is 8 years and the figure is 34%.
The oldest default on the list is 1978. That parcel — 2.34 acres in Antelope Acres, assessed at $6,715 — opens at $21,274. It has been offered and passed over for the better part of five decades, and it has grown more expensive every year it failed to sell.
The re-offer trap, corrected
We published a finding last week that re-offer lists are systematically worse value than first-offer lists. Los Angeles lets us check whether that is the real mechanism, and it partly is not.
314 of these 1,252 parcels were already offered at LA’s April auction and came back. 184 of them came back more expensive — the statutory re-offer at a “reduced minimum price” still collected another cycle of taxes on the way. So far, so consistent.
But on the headline ratio the re-offered parcels look better than the rest of this list: a median of 0.16× assessed against 1.55× for the first offers. That inverts what Butte and Fresno showed. The reason is the clock again. The re-offered parcels have a median of 6 years in default; the first-offer parcels have 19. They are not a better class of property. They are younger debt. Split both groups by default age and the gap between re-offer and first offer nearly vanishes inside every band.
So the honest version of the earlier finding is narrower than we wrote it: re-offer status is not itself the signal. It is a proxy for time in default, and in a county that re-offers quickly it points the opposite way. Time in default is the variable that survives. Where a county does not publish it, a re-offer marker is a weak stand-in — and it can have the wrong sign.
Where the ratio stops working
Screening by bid ÷ assessed has a failure mode this list shows clearly. 101 of the 1,252 parcels are assessed at under $1,000 — 32 of them under $100, and 14 at exactly $10. Those are not valuations. They are Prop-13 base-year leftovers from decades-old nominal transfers that were never re-based.
Divide a real opening bid by $10 and you get a ratio of 1,875×, which looks like the most outrageous parcel in California and means almost nothing. Two of them sit side by side in Vernon: 2.59 acres of industrial land assessed at $10, opening at $2,064, and its neighbour — 1.73 acres, assessed at $455 — opening at $78,578.
We now flag these as Nominal assessment on every sale table rather than let the ratio speak. It changes the top-line number honestly: across all 1,251 parcels with a ratio, 55% open above assessed value; restricted to the 725 assessed above $10,000, it is 31%. Both are true. The second is the one that means what people think it means.
Two things a bidder should check before October
The list is a May snapshot, and it has already moved. The resolution was adopted on May 5. When we read the assessor’s record in August, 62 of the 1,252 parcels no longer carried the five-year default that put them on the list — 30 show taxes fully current with no default year at all, and another 32 have had the old default cleared and only re-entered default in 2026. Redemption runs until 5:00 p.m. on October 16, so more will go. Check the parcel you care about against the assessor before you spend a weekend on it.
Four rows share a legal description with a neighbour. The county flagged them on its own list: items 3462 and 3463 in Vernon both read TR=2455*POR OF LOT 1, and items 3506 and 3507 in Carson both read C F 145 FOR DESC SEE ASSESSOR'S MAPS. In each pair the two parcels are different sizes with very different opening bids — $2,064 against $78,578 in the Vernon case. Match those by AIN on the assessor’s map, never by the description.
What we cannot see
The screen does not know whether a parcel has legal access, whether the desert lots are reachable by a maintained road, what a decades-old improvement is actually standing in, or what liens survive the sale. A ratio is a screen, not a verdict, and assessed value is a county figure rather than an appraisal. We do not make recommendations. The full list, with every rule printed beside it, is at Los Angeles County, October 17–20, 2026.
Sources
- Los Angeles County Board of Supervisors letter 26-2339, adopted May 5, 2026 — the resolution approving the 2026B online auction, with the full 1,252-parcel table. Auction dates, the $5,000 deposit and the October 16 redemption deadline are quoted from it directly.
- Los Angeles County Assessor public parcel record, read for all 1,252 AINs in August 2026: land value, improvement value, lot size, use type, tax status and tax-defaulted year. Roll effective July 31, 2026.
- Los Angeles County’s 2026A auction resolution (April 18–21, 2026), used to identify the 314 parcels being offered again and to compare their April and October minimum bids.