Worked examples

Kern is auctioning 988 parcels in September. The median one opens above what the county says it's worth.

Last verified 2026-09-12 · primary sources

Kern Worked examples
On this page

Kern County is putting 988 tax-defaulted parcels up on September 14–16, 2026, on GovEase. It is the biggest California list of the year that anyone outside a county office can actually read, and it is the one where the cheap-looking numbers are most likely to cost someone money.

Here is our read, published five weeks before bidding opens. This is a free pre-sale desk analysis: county records checked against county records. We did not inspect anything, pull title, walk a boundary, or call county planning. It is not advice and not a recommendation to bid on or avoid any parcel. The full sortable table is here, with a CSV.

Update, 12 September: redemption closed, and the last to leave were mostly buildings

Redemption closed at 5 p.m. on Friday 11 September, so the list is final: 988 parcels. That is 250 of the Board’s 1,238 gone since May — 20% of the sale.

The last five days are the sharpest version of the pattern this post has tracked all along. Twenty-two parcels left between 7 and 12 September, and twelve of them carried a structure — out of a list that was only 6.6% structures. As rates: 17.9% of the 67 parcels with a building on them redeemed in those five days, against 1.1% of the 943 bare ones.

Their assessed values say it louder. The median parcel that left was assessed at $39,418; the median one that stayed, at $2,788.

Twenty-two is a small number and the rates above carry the caution that comes with it. But it points the same way as the two windows below, and the same way Fresno did: what reaches Monday’s auction is overwhelmingly bare desert, because the parcels with something on them were worth rescuing.

Update, 7 September: another 25 gone, and the same parcels are leaving

Three weeks on, the county’s list is down to 1,010. Twenty-five more owners paid off their tax debt between 15 August and 7 September, which brings the total to 228 of the Board’s 1,238 — 18% of the sale — since May.

Measured on its own, this second window says the same thing the first one did:

  • 10 of the 78 parcels carrying a structure redeemed. 14 of the 957 bare ones did. That is 12.8% against 1.5% — a parcel with a building on it was nearly nine times more likely to be rescued in this window.
  • Redemption climbs with assessed value. Below $5,000 assessed, 1.0% came off. Between $5,000 and $25,000, 4.2%. Above $100,000, 23.5%.
  • The parcels that left were assessed at a median of $13,498 against $2,831 for the ones that stayed — nearly five times the value — while their opening bids were close, $5,400 against $4,300.

That last line is the mechanism in one sentence. The debt is roughly the same; what differs is how much the owner loses by ignoring it. Redemption is not a judgement about the price. It is a judgement about the asset.

Redemption runs to 5:00 p.m. on Friday, 11 September, so the list will be shorter again by the sale.

Update, 15 August: the county’s own list landed

We published this in May off the Board’s approving resolution, months before the county posted anything, and said then that redemptions would thin it. The list is now live on GovEase and we can say by how much.

1,035 of the 1,238 parcels survived. 203 — one in six — came off between May and August. Two things about that are worth more than the headline:

  • Every one of the 1,035 appears in the resolution, and not a single minimum bid changed. The read published in May was the sale, three months early.
  • What redeemed is not what you would guess. On price the 203 are indistinguishable from the survivors: same median opening bid, same median assessed value, the same median ratio. The one real difference is what is standing on them. 22% of the redeemed parcels carried a structure, against 7.5% of the parcels still on offer. Owners find the money for a building and walk away from bare ground.

That last point is the practical one. On a desert list the parcels with something built on them are the ones most likely to disappear before the gavel, so a plan built around one is the most fragile kind of plan.

The table now shows the live list and GovEase’s item numbers, which is what the auction actually runs on. Everything below was written from the resolution and is left as it stood, because the method is the point.

The list was hiding in a scanned PDF

The county’s tax-sale page says the list will appear on GovEase about thirty days before the sale. That is true, and it is also not where the list first became public. Kern’s Board of Supervisors adopted Resolution 2026-186 on May 12, and Exhibit A to that resolution is the whole sale: 1,238 parcels with item numbers, tax rate areas, default dates and minimum bids.

It is published as a scanned image with no text layer, so there is nothing to copy and paste. We read it with OCR, which is exactly the kind of step that quietly introduces errors into a spreadsheet nobody checks. So we checked it, against the county’s own data rather than against itself:

  • 1,238 of 1,238 parcel numbers resolve in the Kern County Assessor’s parcel service.
  • 1,238 of 1,238 tax rate areas we read off the scan match the assessor’s own value for the same parcel — a second, independent field agreeing on every row.
  • Every minimum bid is a whole multiple of $100, which is the increment Kern bids in. A misread digit would almost always break that.

That is why we are willing to publish it. Three checks that a bad read would fail.

What the join says

Every parcel is matched to the assessor’s record for land value, improvement value and acreage. 1,237 of 1,238 returned a value. Then one division: opening bid ÷ assessed value.

The median parcel opens at 1.29 times the county’s own assessed value. Not 0.3. Not 0.8. The middle of this list asks for about thirty per cent more than the county thinks the parcel is worth.

706 of the 1,237 open above assessed value. That is 57% of the sale.

(Both figures are the May read of the Board’s 1,238-parcel resolution, and both count parcels whose assessment is nominal — under $1,000, a Proposition 13 leftover rather than a valuation. We no longer count those as opening above assessed value, because dividing a real bid by $400 says nothing. On the final 988-parcel list, with those parcels banded separately, the figure is 434.)

This is not a Kern peculiarity. San Bernardino’s July list had 858 of 1,250 parcels in the same position, and we said so at the time. It is what a big desert list looks like once you have something to measure the price against, and it is the single most useful thing to know before you bid in one.

Why a cheap lot is often the expensive one

The mechanism is the same one we described in what the opening bid actually measures. The opening bid is not a valuation. It is the amount required to redeem the parcel — five or more years of unpaid taxes, penalties, and the county’s costs of sale. Nobody sets it by looking at the land.

So on a parcel worth very little, the debt keeps compounding against an asset that is not growing. Give it long enough and the debt passes the value. A $4,600 opening bid on a two-acre lot sounds like almost nothing until you notice the assessor carries that lot at $1,900. You would be paying 2.4 times the county’s own figure for the privilege of owning something whose previous owner stopped paying $500-a-year taxes on it, and who was probably right to.

The reverse case is in the same list. 131 parcels open at a quarter of assessed value or less, the cheapest at 3.4% of it. Those are the ones where the debt is small relative to something genuinely worth having — a parcel that fell into default over a modest bill rather than a parcel nobody wanted.

The cheapest opening bids on this list are not the best value on it. Sort the table by bid ÷ assessed and the order changes completely. That is the whole reason we publish the ratio.

What the list is made of

Only 122 of the 1,238 carry any improvement value at all. This is a land sale, not a house sale: 90% of it is bare ground.

Only 286 have a street address on file, and Rosamond alone accounts for 119 of those. The other 952 parcels are land the county can locate on a map and has never had reason to give an address to. That matters for what you can find out before September: with no situs address, the ordinary tools — a listing site, a street view, a comparable sale — have nothing to key on. The assessor’s map and the parcel boundary are what you have.

Twenty-five parcels are under a tenth of an acre. We flag those, because under most zoning they cannot be built on, and a lot that cannot be built on is not cheap land, it is a recurring tax bill with no exit.

What this screen cannot see

The ratio is a screen, not a verdict, and assessed value is a county figure rather than an appraisal. Under Proposition 13 it is anchored to the last change of ownership, so a parcel held since the 1980s can be assessed far under what it would fetch — which cuts in the bidder’s favour and is exactly why the low-ratio end of the list is worth reading carefully rather than trusting.

Everything else still applies and none of it is in our table: legal and physical access, whether a parcel is buildable, water and septic feasibility, fire hazard severity and what it does to insurance, county zoning, and any liens that survive the sale. In Kern specifically, bonds and assessments levied by agencies other than the Tax Collector can remain outstanding after the sale, and the IRS has 120 days to redeem a parcel carrying a federal lien.

Before you bid

  • September 14–16, 2026, online at GovEase. A $5,000 deposit is required and bidding moves in $100 increments.
  • The right of redemption is forfeited at 5:00 p.m. on Friday, September 11, the last business day before the sale. Parcels will drop off this list between now and then — the resolution list dates from May.
  • Check the county’s own published list when it lands on GovEase around mid-August, and compare it to ours. The difference is the parcels somebody paid for in the meantime.

Verify all of this against the county before you act on it — the Kern County Treasurer-Tax Collector’s tax sale page is the authority, not us.

Sources

  • Kern County Board of Supervisors Resolution No. 2026-186, adopted May 12, 2026, Exhibit A — item numbers, parcel numbers, tax rate areas, default dates and minimum bids for all 1,238 parcels, read from the county’s published scan.
  • Kern County Assessor’s public parcel service (Assessor_Public) — land value, improvement value, acreage and situs for every parcel, matched by assessor tax number.
  • Kern County Treasurer-Tax Collector, general tax sale information — sale dates, deposit, bid increments, redemption cutoff and the terms of sale.

Want the next California sale like this one — the full read, before bidding opens?

Or see the free Riverside brief · browse the rest of the research.

Worked examples

What Los Angeles tax-sale parcels sold for in April: 121 sales against their opening bids

Los Angeles posts the price of every parcel sold at its tax auctions, so that former owners can claim the surplus. We matched the 121 sales at April's auction to their opening bids. 30 went at exactly the opening bid. 61 went for at least twice it, and the median sale was 2.03 times its opening bid. Only 6 of the 121 had a building, and those 6 brought in $2.2 million of the $4.9 million paid.

Los Angeles Worked examples

Worked examples

We published a screen on 105 Butte parcels, then watched the auction. Here is where it failed.

Before Butte County's August re-offer we published all 105 parcels with a screen on each one. The auction has closed, so we scored it. Among the parcels the screen could rate it ordered them correctly — 33% of the cheapest-looking sold against 11% of the dearest. The problem is the 35 it could not rate. Those sold best of all, and were bid up hardest.

Butte Worked examples

Worked examples

The opening bid is a clock: what 1,252 Los Angeles parcels show about why tax-sale prices stop making sense

Los Angeles publishes something almost no other county does — the year each parcel first went tax-defaulted. Join it to the assessor's values across all 1,252 parcels in the October sale and the pattern is almost mechanical. Among parcels the county assesses above $10,000, fewer than one in a hundred of those defaulted under ten years opens above assessed value. Of those defaulted thirty years or more, 98.7% do. The property is not what moved.

Los Angeles Worked examples