Crime Pays. No Amnesia Required.
School drop-off, newly public insurance documents, and a ledger that comes out to about 700 to 1.
Summer time may cause temporary amnesia.
Every August, during the first days of school drop-off, some parents appear to forget and rush to see if merging from two lanes to one is still the way it goes.
This year did not disappoint.
I rolled the wheel around the last turn and through the mess as Liam leaned toward his brother.
“Anthony, is my classroom the same as yours?”
Anthony, almost nine and attentive to more-and-more by the day, didn’t hear as the door swung open to Ms. Hallie and the first day fanfare.
Adalyn pushed past the boys and through the door telling Liam “I’m coming with you to figure this out.”
I reached out to wave goodbye and felt time rushing across my hand like starwars hyperspeed.
Somewhere in my head, Joy from Inside Out decided this was a core memory.
Then my phone went off.
Meeting in fifteen minutes.
Of course.
Some days this past year felt like I had a meeting every fifteen minutes. When I wasn’t in one, I was preparing for one, leaving one, or discovering that someone had placed another one directly on top of lunch.
We acquired another company. If current projections hold, Brelly could grow roughly fivefold this year. The workdays stretched to sixteen and seventeen hours. I passed on opportunities I would have chased a year earlier because there was no room left to chase them.
I got some gray hair out of the deal.
Meanwhile, artificial intelligence made starting things easier than it has ever been. New tools appeared every week. My parents could play spades in the afternoon and decide by dinner that they should launch an app.
I love that about this moment.
I am also tired.
After the meeting, I got back to my car and saw another notification. This one was from Alfred, my AI chief of staff.
Secret documents released.
Oklahoma Watch had published a new installment in its investigation of insurance claim practices. Lawyers had released internal documents and court exhibits describing how insurers studied claim payments, hired consultants, changed roof-claim procedures, and measured the money they could keep.
I did not have time to read hundreds of pages in a parking lot. So I asked AI to help me work through the documents and show me what repeated.
One phrase kept coming back.
“Economic opportunity.”
That was the language in the Allstate and McKinsey material from the 1990s. Claim dollars that could remain unpaid were called “opportunity dollars.” One internal analysis priced the opportunity at $484 million in a year. A roof-process slide said managers needed performance measures with “the teeth to drive behavior.” Another measured the dollars left behind when adjusters failed to follow the process.
Some slides celebrated higher rates of claims closed without payment with handwritten notes: “AWESOME!!” and “WOW!!”
Those are real exclamation points. I checked.
The current State Farm material is less open. Much of it remains sealed or redacted, so some of the largest claims come from plaintiffs’ lawyers describing documents the public cannot yet read. Those lawyers allege that State Farm built a program to cut wind and hail payments by 50 percent, saved more than $1 billion in its first year, and estimated roughly $15,000 saved for every homeowner fully denied.
State Farm disputes the allegations and denies wrongdoing.
The evidence is not identical. The companies are not identical. The legal cases are not verdicts proving every allegation.
But the resemblance is difficult to ignore.
Hire consultants. Rename unpaid claims as savings. Take authority from the adjuster who saw the roof. Track who approves too much. Reward lower payments. Keep the paperwork out of public view.
A generation passed between the programs.
No amnesia required.
Then I found the ledger.
An attorney who spent years fighting for the Allstate documents estimated that its claims program saved the company $15 billion over roughly a decade. That is an estimate, not an audited admission. In a rough ledger assembled from contempt penalties, fines, and regulatory settlements, I could identify about $21 million in direct consequences.
It is important to do the math.
$15,000,000,000 ÷ $21,000,000 = 714
Round down. Call it 700 to 1.
Fine. Maybe the lawyers cost another $79 million. I have no evidence that they did. I am being absurdly generous because lawyers rarely work for free and round numbers are nice.
Call the total cost $100 million.
$15,000,000,000 ÷ $100,000,000 = 150
Not a bad day at the casino.
Crime pays.
State Farm has a separate ledger, and it produced an even stranger lesson.
In State Farm v. Campbell, a Utah jury awarded $145 million in punitive damages after finding bad-faith claim handling. The United States Supreme Court ruled that award excessive. On remand, the punitive award landed at about $9 million.
Again, it is important to do the math.
($145,000,000 - $9,000,000) ÷ $145,000,000 = 93.8%
The Supreme Court gave the bad apples a 93.8 percent discount on the jury’s lesson.
That is the part I cannot shake.
A homeowner pays an insurance premium because some risks are too large and too random to carry alone. A storm does not care whether the mortgage is due or whether the family has enough cash to replace a roof.
The policyholder thinks he bought someone else to carry that uncontrollable risk.
Then the claim arrives, and he learns he may have bought permission to ask for access to a pool of money somebody else controls.
The company holds the gate. It decides what counts as damage, who gets to inspect it, which estimate survives, and how long the policyholder can afford to argue.
From the sound of these documents, some companies also hired consulting firms to improve the equation for paying less.
All of the pressure meets at the policyholder. The storm, the mortgage, the contractor, the insurer, and eventually the lawyer all push toward the same person. The company moves numbers on a report. The homeowner still needs a roof.
This is why I care about transparency in insurance claims.
A market cannot correct behavior it cannot see. A policyholder cannot make a real choice when every company’s promises sound the same and the evidence of what happens after a loss remains scattered across sealed court files.
I do not want to wait for the next regulator, lawsuit, election, or leaked document to make this understandable.
I want the information in the policyholder’s hands. Let people see how claims are handled. Let them compare what was promised with what was paid. Let them decide which companies deserve their premiums.
That is the work I went back to after the notification.
Later that evening, I closed the computer for a break and sat down with the family for dinner.
I asked Liam if he had solved the classroom mystery.
“Yep,” he said. “Mine’s more cool.”
Everyone approved.
He went back to eating as if the matter had been thoroughly investigated and peer reviewed.
I looked around the table at all three of them.
The time was going to keep moving across my hand whether I approved of that arrangement or not. Some of it would pass while I was working. I could not pretend otherwise, and I could not get those hours back.
But I was glad this was the work.
It was work I could explain to my kids. I hoped it would be work they could be proud I chose.
Dinner ended.
I went back to it.