Palomar Compounded Earnings 30% While Its Stock Stood Still — The Fight Is Over Which One Is Lying

Every September, Palomar's chairman and CEO, Mac Armstrong, and CFO, Chris Uchida, take the stage at the annual KBW Insurance Conference for a fireside chat selling the same story: a specialty insurer that keeps outgrowing its niche. The story has rarely been easier to retail. In the second quarter of 2026, gross written premium rose 27% year over year to $630.5 million, net earned premium jumped 59.5%, and adjusted earnings per share grew 34%. A specialty insurer compounding that fast is unusual.

And yet the market yawned. The stock has gone essentially nowhere for a year — up about 1% year to date even after nearly matching its 52-week high this summer. Same earnings, same growth, a stagnant price: that compression is the whole duel in miniature. Both camps see the identical numbers. The question is whether the stall is a mispricing the bulls can exploit or a warning the bears have been right about all along.

The shared facts both camps sign

As of Sept. 3, Palomar trades near $136 with a market value of about $3.6 billion. Freeze the record:

• Q2 2026: gross written premium $630.5 million, up 27%; adjusted net income $63.8 million, up 31.4%; adjusted EPS $2.36, up 34%.
• Underwriting: combined ratio 83.3%, up from 78.8% a year earlier; adjusted combined ratio 76.7%, up from 73.1%. The overall loss ratio jumped from 25.7% to 34.5% — driven by attritional losses, not catastrophe claims.
• Mix shift:Casualty is now 31% of premium and grew 37%; earthquake, the product that built the company, is 23% and shrank slightly (down 0.7%). Crop grew 96%; surety and credit grew 236%.
• Capital: an annualized adjusted return on equity of 26.3%, $1.7 billion of cash and invested assets, and $41 million of buybacks in the quarter.
• Guidance: full-year 2026 adjusted net income of $270–$280 million — above the $266–$280 million range set in May alongside expanded earthquake and hurricane reinsurance — a target that embeds only $8–$12 million of catastrophe losses for the entire year.

That last pair of lines is where the fight actually starts.

Round one — Where did the growth come from?

The bull's cleanest argument is diversification. Palomar built its name on earthquake insurance, a volatile niche where one tremor can erase a year's profit. It has spent the past few years deliberately shrinking that exposure: earthquake is now under a quarter of the book and basically flat, while casualty — the classic steady, compounding specialty line — is the largest segment and growing at 37%. Bulls read this as a company exchanging a boom-and-bust catastrophe engine for a machine that compounds predictably. Net earned premium up 59.5% is evidence the market is accepting the new product mix.

The bear reads the exact same migration as the risk, not the cure. Casualty is a long-tail line where the true cost of a policy is not known for years, until the claims stack up and reserving proves whether the pricing was right. Palomar's expertise and data history were built on short-tail catastrophe products; the new growth is concentrated in lines where a 2014-founded specialty insurer simply has not lived through a full cycle. Crop (up 96%) and surety-and-credit (up 236%) add weather and credit correlation on top. The fastest-growing dollars, the bear says, are exactly the dollars with the least seasoning — and reserving is revealed only in arrears.

Round one verdict: the bull wins the description, the bear wins the uncertainty. Diversification is real and directionally de-risking. But "we grew 59%" in earned premium is not proof the new lines will ultimately price out; it is an invoice for risk that won't be paid in full until years of loss development.

Round two — The margin is drifting, and both sides noticed

Here the bear has a fact that genuinely embarrasses the bull. The reported loss ratio jumped from 25.7% to 34.5% in a single year, and the combined ratio climbed from 78.8% to 83.3%. Management says the deterioration is attritional — ordinary claims, not catastrophe events — which is precisely the bad kind. A catastrophe loss is a one-time tail; an attritional loss-ratio creep says the underlying pricing of the growing book is thinning.

The bull's answer is that the new mix is simply carrying more — both richer casualty pricing and the longer realization of earned premium — and that at an adjusted combined ratio of 76.7%, Palomar is still printing roughly 23 cents of underwriting profit on every premium dollar. Add favorable prior-year development, the bull argues, and the trend line is still one of an elite specialty underwriter, not a deteriorating one.

But the bear can point out that a chunk of the current period's income is favorable prior-year development — releases from earlier, over-reserved accident years. Those releases flatter today's results precisely because they are the echo of a book that was priced conservatively years ago. They are not a promise that today's faster-growing, less-seasoned book will produce the same echo. The company's own EPS growth is being helped, in part, by money that is already in the can.

Round two verdict: the bear wins this round, narrowly. The bull's margin is still excellent, but the trend is moving against it at exactly the moment the marginal premium dollar is shifting to long-tail lines. When a growth story's margin compresses while its mix gets less seasoned, the burden of proof moves.

Round three — The catastrophe assumption nobody is paying for

Now the decisive variable. The full-year guidance of $270–$280 million in adjusted net income assumes just $8–$12 million of catastrophe losses all year. For a company that still writes a quarter of its book in earthquake and a crop book with weather exposure, that is a razor-thin assumption — effectively a bet on a benign hurricane season and no significant seismic event.

The bull frames this as discipline: the company raised guidance in May to $266–$280 million and expanded its earthquake and hurricane reinsurance, buying protection so that routine years are not derailed. A $12 million catastrophe allowance is the sign of a business that has hedged away the tail, they argue, and that the low-volatility earnings power is therefore real.

The bear's counter is that reinsurance is not free and does not make the risk disappear — it converts catastrophe risk into a premium line item and a credit-counterparty question. If a large event lands on the remaining cat book, the $8–$12 million assumption is broken in a single quarter, and the "low-volatility compounder" valuation snaps back to a catastrophe-insurer discount. The market's yearlong refusal to re-rate the stock, in this reading, is exactly the correct enforcement: don't pay a stable-compounder multiple until the company has proven it through a loss.

Round three verdict: the bear wins the point, but it is a scenario, not an event. No single loss has occurred to validate it. Yet the bull's guard against it is reinsurance, which costs margin — the same margin that is already drifting.

What the price is demanding

Run both stories through one frame. At $136 and about $3.6 billion of market value, with $270–$280 million of 2026 adjusted net income guided, the market is paying roughly 13 times forward adjusted earnings for a business earning a 26% return on equity and growing adjusted EPS at 30% a year. A 30% grower at 13 times earnings is not expensive by any ordinary growth-valuation arithmetic.

That is the bull's exit line, and it has real force. But the price has done the work already: a stock that compounds earnings 30% while going nowhere for a year is one whose multiple compressed by a third. The market is not ignoring Palomar's growth — it is demandingly discounting the quality and durability of that growth. At 13 times earnings you are paying for the diversification story but keeping only a slim cushion against the margin drift, the unseasoned long-tail mix, and a catastrophe assumption that is essentially zero.

So which camp has the better odds? The bear's strongest claims are real and undeniably present in the numbers — the margin creep, the favorability reliance, the thin cat allowance. But they are risk factors flagged as scenarios, not losses that have actually occurred. The bull can show a 26% ROE, cold cash from operations, self-funded growth, and an improving product mix, all at a multiple that stopped rewarding it a year ago. When the market has already refused to pay up, the asymmetry shifts: the bull no longer has to be right every quarter, just profitable over time.

The ruling

The business case goes to the bull; at $136, so does the stock call — but on a thin margin and with the burden of proof firmly on management. Paying roughly 13 times forward adjusted earnings for a genuinely compounding, de-risking specialty insurer is a reasonable wager, not a steal. The bear is right that the growth is migrating into lines whose claims have yet to season and that the catastrophe allowance is thin; those are the two things that could make this an expensive 13 times.

The ruling flips if the adjusted combined ratio holds above roughly 78% into the back half of 2026, if casualty prior-year development turns adverse, or if a single large earthquake or hurricane loss lands on the remaining catastrophe book. Those are measurable, dateable conditions — and they are exactly what management will be asked about from the KBW stage this September, where a soft question about growth should be the least interesting one asked.