Every California county that runs a tax sale ends up with parcels nobody bid on. The law expects this: under the Revenue and Taxation Code the tax collector has to keep trying to sell, so the leftovers come back — sometimes weeks later as a re-offer, sometimes next year, sometimes at a minimum bid the board has voted to reduce.
A price that went down looks like an opportunity. It is worth being precise about what it actually tells you.
Three lists, all of them leftovers
We published three California sales this week that let you see the effect directly.
Butte County’s August 7–10 auction is a re-offer in its entirety — 105 parcels, every one of which failed to sell at the county’s June sale. That makes it an unusually clean natural experiment: a list containing nothing but the parcels a room full of buyers already declined.
Placer County brings 14 of its 43 October parcels back from 2025, thirteen of which drew no bid at all.
El Dorado brings back 11 of its 37 November real-property items from its November 2025 auction.
What the leftovers look like
Take the ratio we screen every sale by: the opening bid divided by the county’s own assessed value. Lower means the county is asking less relative to what it thinks the parcel is worth.
| Sale | Median bid ÷ assessed | Opens above assessed |
|---|---|---|
| Fresno, September (first offer only) | 0.13× | 11 of 127 |
| El Dorado, November (mixed) | 0.16× | 7 of 37 |
| Placer, October (mixed) | 0.23× | 13 of 41 |
| Butte, August (all re-offer) | 0.64× | 44 of 105 |
| Fresno, September (re-offers only) | 5.96× | 25 of 34 |
Butte’s median parcel opens at five times the ratio of Fresno’s first-offer parcels, and 42% of its list opens above the county’s own assessed value — against 9% of Fresno’s first offers.
The sharpest version of this is inside a single sale. Fresno’s September auction contains both kinds, priced off the same assessor roll on the same day, so nothing about the county, the market or the valuation basis differs between them. The 127 parcels being offered for the first time open at a median of 0.13× assessed value. The 34 coming back open at 5.96× — and 25 of the 34 open above assessed value, against 11 of 127.
That is the finding. The re-offer list is systematically worse value than the first-offer list, and it is worse in the specific way that matters: a much larger share of it costs more than the county says it is worth.
Narrowed since publication. Los Angeles publishes the year each parcel went tax-defaulted, which let us test whether re-offer status is doing the work here. It mostly is not. In LA the re-offered parcels look better on this ratio, because they are younger debt — a median of 6 years in default against 19 for the first offers — and once you split both groups by default age the gap nearly closes. The variable that survives is time in default; a re-offer marker is a proxy for it, and in a county that re-offers quickly the proxy points the wrong way. The counties measured above still show what they show, but read it as a symptom of accumulated years, not of the re-offer itself. The full working is in The opening bid is a clock.
Why, mechanically
This follows from what an opening bid is. It is the amount needed to redeem the parcel — years of unpaid taxes, penalties and the county’s costs — not an estimate of value. Two things happen at a sale:
- Parcels where the debt is small relative to the land get bought. They are the ones with headroom.
- Parcels where the debt has outgrown the land do not get bought, because there is no headroom to buy.
Sale by sale, the first group leaves and the second group stays. A re-offer list is what is left after everyone who could read a ratio has already bought. It is not a rack of things people missed; it is a rack of things people looked at and priced.
The price cut is information, not a discount
Placer makes this visible. Thirteen parcels drew no bid in 2025, so the county cut their minimums — the Truckee block by about 42% across the board. One went from $3,089,200 to $1,803,100.
That is a genuine, substantial reduction, and it is still not a bargain: after the cut, that parcel opens at 49 times the county’s own assessed value of $6,824. Several of its neighbours are in the same position. The market saw them at the old number and passed; the new number is lower and is still multiples of what the assessor thinks the land is worth.
A price cut on an unsold parcel tells you the county wants it gone. It does not tell you the parcel is worth having. Those are different facts and only one of them is about the land.
Where the leftovers are genuinely interesting
None of this makes re-offers useless. Two cases are worth the attention:
A parcel that failed to sell for a reason that has changed. A sale held four days after a list dropped is a sale most people could not do the work for. Butte’s June deposit deadline, Fresno’s two-day window, Mariposa’s same-morning registration — access, not value, decides some of these. A parcel that went unsold because nobody had time is a different animal from one that went unsold because everyone had time.
A parcel that is cheap on the ratio and is coming back anyway. They exist. Butte has 15 parcels opening at a quarter of assessed value or less, and El Dorado’s re-offers include five whose bid the county has cut. The point is not to avoid re-offers, it is to stop treating “reduced” as a synonym for “cheap”.
How we flag it
Every parcel we can identify as previously offered carries an Offered before, didn’t sell flag in the table. It is derived from the county’s own documents — its prior sale list, its published results, or its own re-offer marking — never inferred.
Butte’s whole list carries it. Placer’s thirteen carry it. El Dorado’s eleven carry it. Fresno marks 34 of its parcels as previously offered, and the county has authorised a reduced minimum on 19 of those.
The tables:
- Butte County, August 7–10, 2026 — 105 parcels, all re-offers
- Placer County, October 21, 2026 — 43 parcels, 13 re-offers
- El Dorado County, November 6, 2026 — 37 real-property parcels, 11 re-offers
- Fresno County, September 10–11, 2026 — 90 parcels, 25 previously offered
All free, sortable, and downloadable. As always: a screen, not a verdict, and not a recommendation to bid on or avoid anything.
Sources
- Butte County Treasurer-Tax Collector re-offer list for the August 7–10, 2026 auction, and the county’s June 2026 sale.
- Placer County Tax Land Sale list for October 21, 2026, and the county’s published 2025 sale results, which uniquely report minimum bid, winning bid and assessed value together.
- El Dorado County tax sale auction list for November 6, 2026, and the county’s published November 5, 2025 notice and results.
- Fresno County Board of Supervisors Resolution 26-245, Schedules A and B.
- Assessed values throughout are from each county’s own assessor parcel service. Fresno’s come from its live parcel layer (FC_PARCEL_SELECT), read on August 7, 2026; an earlier version of this piece used the county’s 2022 roll, which put the first-offer median at 0.17× rather than 0.13×.