A record compensatory verdict closed every news cycle, then the same jury came back a week later with punitive damages no reserve model prices. For an auto book, the lesson is the timing, not the dollars.
On June 3, 2026, a Van Nuys jury handed Brian Panish a $176 million verdict against Rebecca Grossman and Scott Erickson. The number was already a record. Then the trial did not end.
The case was Iskander v. Grossman. Two boys, Mark and Jacob Iskander, ages 11 and 8, were killed in a marked crosswalk on Triunfo Canyon Road in Westlake Village in 2020. Two Mercedes SUVs. One of them driven by Grossman, a former Hidden Hills socialite now serving 15 years to life for their deaths. The jury found both drivers liable for $176 million in compensatory damages to the parents. Panish had asked for $439 million. He did not get it. He got the next phase.
A week later, on June 10, the same jury came back with punitive damages. $21 million against Grossman. $1.17 million against Erickson. The total climbed to roughly $198 million. The headline number that closed every news cycle the week before was now wrong by twenty-two million dollars.
This is the part that should hold a reserve actuary's attention. The verdict did not arrive whole. It arrived in two pieces, a week apart, and the second piece was a punitive award that no compensatory model prices.
Reserves get set the moment a claim looks serious. A carrier or a self-insured fleet looks at the facts, the venue, the plaintiff firm, and books a number. That number is a guess about where the case lands. When a verdict comes in, the guess resolves. The book moves. The reinsurance treaty either attaches or it does not, depending on whether the loss pierces the layer.
A single-day verdict is clean. You read the figure, you true up the reserve, you cede what belongs above the retention, and you close the file. A verdict that grows after the headline is not clean. The compensatory number resolves one set of reserves. Then a second number lands on top of it, days later, after the file already looked settled in the ledger. The reinsurance attachment point that was comfortably clear of the $176 million figure may sit right inside the $198 million figure. The same loss now pierces a layer it did not pierce on June 3.
Trial drama, nuclear verdicts, and the plaintiff-firm tactics behind them. Court-reporter prose, no consultant filler. Read by litigation leaders at F500 legal departments and national carriers. Free.
Punitive damages are the tail that compensatory models do not carry. Compensatory loss is bounded by something real. Medical bills, lost earnings, the value a jury puts on grief. It is large here, but it is anchored. Punitive damages are anchored to nothing but the jury's appetite to punish. Panish called the punitive phase "the most important part of the trial." He told the jury to punish and deter. That is a number with no ceiling and no actuarial input. It is the developing tail on an auto file, and it shows up after the case already read as resolved.
Act one is the record compensatory verdict that makes the news. Act two is the punitive phase, days later, where the number grows again.
For an auto book, the structural lesson is the timing, not the dollars. The dollars are someone else's problem in this particular case, because Grossman is a convicted criminal whose conduct likely strips coverage clean off the loss. But the mechanism is general. The plaintiff bar in Los Angeles now runs a two-act trial. Act one is the record compensatory verdict that makes the news. Act two is the punitive phase, days later, where the number that was already a record grows again. Panish won a $488 million verdict for a single pedestrian last year. He runs this play at scale.
A reserve set on the compensatory verdict is a reserve set before the trial is over. The file is not done when the big number prints. It is done when the punitive jury comes back. Anyone who trued up reserves on June 3 trued them up to a number that was about to move.
The reinsurance question is sharper still. Treaties attach at a point. A loss that crosses the point in two installments, on two dates, with the second installment driven by a punitive standard no model forecasts, is a loss that tests whether the attachment math was built for a verdict that arrives once or a verdict that keeps arriving.
Panish has built a machine that makes verdicts grow after the headline. The reserve and the treaty were built for verdicts that stop. One of those two things is wrong. Right now it is the books.
The Executive Briefing is six questions. It shows you exactly where the gaps are.
Take the Executive Briefing →Trial drama, nuclear verdicts, and the plaintiff-firm tactics behind them. Court-reporter prose, no consultant filler. Read by litigation leaders at F500 legal departments and national carriers. Free.