We read 174 DFPI escrow shutdowns. Here is what actually closed offices.

Every escrow conference has a session on wire fraud, and every session has a number. The number changes depending on who is presenting. Sometimes it is a national loss figure from the FBI. Sometimes it is "offices like yours are closing every month." I have sat through enough of these to notice that nobody ever cites the actual enforcement record for our own state.

So we read it. All of it.

The Department of Financial Protection and Innovation publishes its enforcement actions against licensed escrow companies. We went through every shutdown from 2002 through 2026 — 174 of them — pulled the underlying documents, and classified each one by cause. This is what the record actually says.

How we did it

Nothing clever. For each enforcement action, we fetched the DFPI's page and downloaded the underlying documents: the Notice and Summary of Findings, the Demand for and Order Taking Possession, or whatever the department filed. We extracted the text and searched it for the terms that distinguish causes — hacked, wire, embezzled, misappropriated, unauthorized, shortage.

Each case got a classification and a confidence level. Where the documents named the cause plainly, we marked it high confidence. Where we had to match a case against a description elsewhere — for example, a DFPI Bulletin describing an unnamed conservatorship — we marked it medium. Cases with only an Order Revoking or an Order to Discontinue, and no trust shortage indicators at all, we classified as procedural.

We are publishing the classification so anyone can check the work: download the full classification (CSV) — all 174 cases with entity, year, cause, evidence quote, source URL, and confidence level. If you think we miscategorized a case, call me.

What the record shows

Of the 174 shutdowns:

Cause	Offices	Share
Procedural (failed audits, missed reports, net-worth and bond deficiencies, surrender issues)	134	~77%
Internal embezzlement	31	~18%
Outside wire fraud	6	~3.5%
Trust shortage, cause unspecified in the record	3	~2%

The six outside wire fraud cases, with what the record shows:

Entity	Year	Amount
Escrow Technologies, Inc.	2025	$4,281,686 wired out; $3,804,686 estimated shortage
Fountain Valley Escrow, Inc.	2024	$1.9M wired; $480K shortage
Integrity Escrow, Inc.	2024	$500K
Redwood Escrow Services	2023	~$2M (medium confidence)
Ridgegate Escrow, Inc.	2019	$107K
Efficient Services Escrow Group	2014	$1,558,339

The largest is recent. In November 2025, a caller impersonating City National Bank's fraud department talked employees at Escrow Technologies into authorizing 40 wire transfers to 11 financial institutions. Thirty-five processed. Five were rejected. The DFPI's Notice and Summary of Findings puts the trust shortage at $3,804,686 — the largest confirmed outside wire fraud loss for a California escrow company. The source document is on the DFPI's site: Notice and Summary of Findings, Escrow Technologies, Inc.

One more pattern worth stating plainly: every confirmed outside wire fraud case in the record involved an independent escrow company, not an underwriter-owned entity. Independents carry this exposure alone.

The honest interpretation

If you came here expecting the record to say wire fraud is the thing closing California escrow offices, it doesn't. Outside wire fraud is real, the losses are large, and the trend is recent — three of the six cases are from 2023 onward. But it is a minority cause. About 3.5% of shutdowns.

Internal embezzlement closed roughly five times as many offices. And the overwhelming majority of shutdowns — about 77% — are procedural: failed audits, late or missing reports, net-worth and bond deficiencies, surrender problems. Offices that lost their license without anyone stealing anything.

Here is the pattern that cuts across all three categories. Read enough of these findings documents and the same sentence structure keeps appearing: the department could not determine, the records did not show, the file did not reflect. Whether the trigger was a fraudulent wire, an employee moving money between files, or an examiner asking routine questions, the office's position depended on what its files could demonstrate — and the files usually couldn't.

The checking may well have happened. Escrow officers are careful people; it is the job. But a review that lives in someone's memory, or in a phone call nobody logged, does not exist as far as the examiner, the adjuster, or the lawyer is concerned. The problem was never the checking. It's that the file doesn't show who checked.

What a careful office can take from this

A few things follow from the record, none of which require buying anything:

Weight your worry by the numbers. Procedural failures close more offices than fraud does. The unglamorous work — reconciliations on time, reports filed, bond and net-worth thresholds maintained — is the bulk of the survival problem.

Treat internal controls as seriously as external ones. The record says the person moving money without authorization is five times more likely to be inside the office than outside it. Dual controls on disbursements and file-to-file transfers address the larger documented risk.

Write the review down when it happens. Every category of case turns on the same question after the fact: what did the office check, against what source, and who checked it? A callback logged with a note number, a payoff demand compared with a date, an open item marked open — these are cheap to record in the moment and nearly impossible to reconstruct honestly later.

Know what your file would show today. Pull a recent file where a disbursement changed and ask whether a stranger could reconstruct the review from what's in it. If the answer is no, that is the gap — not the absence of another tool, but the absence of a record.

Why we did this work

Veto exists because of the pattern in this data. It records the wire review an escrow office already runs — what changed, what was checked, what stayed open, who reviewed — and files one page before the money moves. It does not check anything itself and it does not prevent anything; the record is the product. The office decides. Veto records the review.

— Sebastian Heyneman