Kinsale’s stock is down, but that does not make the short thesis right. A closer look at the Bear Cave allegations, customer complaints, retention, pricing, exclusions, and why the business may be more cyclical than broken.
It would be interesting to check with the states in, say Kinsale's top 5 for underwriting and see what the complaint rates are, e.g. complaints per policy or complaints per $1M premium (states vary a bit in what they track data on, so you may have to be flexible). If Kinsale's complaint rates are similar to other competitors, then that short argument would appear to be less valid.
Good job, Alexander, as usual. Just want to make two comments:
1. Handling all the underwriting and claim functions internally is a strategic decision. Closely scrutinizing every claim is probably quite natural, versus the MGA model.
2. Although I agree with your statement that the price going down by 35% does not prove the short case was right, as with Brad Safalow of PAA Research's July 2025 original publication. However, there were some valid points in Brad's arguments, such as the extreme valuation and macro headwinds. It's a masterful short in July 2025, and I think the PAA folks are sharp. Whether they are right or not does not really matter as long as they got the 35% decline right.
I advised caution several times in two quality communities and on Twitter/X since middle-2025. However, when the Bear Cave short article dropped, I felt it was a bit silly and a little "Johnny-Come-Lately". As I commented by quoting Buffett, "What the wise do in the beginning, fools do in the end."
I don’t know Palomar well enough to have a strong view, but at first glance, they seem to be growing faster right now, partly because they are more concentrated in fewer products than Kinsale.
That said, the profitability gap is hard to ignore. Both Kinsale and Palomar wrote roughly $2 billion of GWP last year, but net income was very different: around $197 million for Palomar versus $503 million for Kinsale.
What I find especially interesting is the composition of the combined ratio. Both companies are currently around a record-low ~76% combined ratio, but Palomar gets there in a very different way: a loss ratio of only ~28% versus ~57% for Kinsale, and an expense ratio of ~48% versus ~21% for Kinsale.
Part of the explanation may simply be product mix. Palomar has meaningful exposure to catastrophe and low-frequency/high-severity lines like earthquake. In quiet years, those lines can produce very low loss ratios because the big losses simply do not occur. But that does not necessarily mean the normalized loss ratio is structurally 28%. It may just look exceptionally good in benign years.
To me, the expense ratio is the more important number when thinking about scalability. Kinsale’s model clearly shows operating leverage. Palomar’s much higher expense ratio makes me less comfortable, even though the headline combined ratio looks great.
So while Palomar may currently look stronger on growth, I feel much better with Kinsale because the scalability of the model is much clearer. Palomar’s low loss ratio may be real for now, but I would be careful extrapolating it too far into the future.
IMHO, the beauty of KNSL’s business is the niche that it serves. William and the other NKOTB managers may be excited that float is akin to negative interest rates which may now cause some crowding creating a softer market. Michael’s shown discipline so we wait for the weigh in and ignore today’s vote. At the point that too many TikTok streams become annoying for KNSL, Michael can ask Prem for Kasowitz’s number; wonders happened for http://ffh.TO afterwards. Someone should then tell Ari Emanuel to hold off booking Dylan.
It would be interesting to check with the states in, say Kinsale's top 5 for underwriting and see what the complaint rates are, e.g. complaints per policy or complaints per $1M premium (states vary a bit in what they track data on, so you may have to be flexible). If Kinsale's complaint rates are similar to other competitors, then that short argument would appear to be less valid.
Good job, Alexander, as usual. Just want to make two comments:
1. Handling all the underwriting and claim functions internally is a strategic decision. Closely scrutinizing every claim is probably quite natural, versus the MGA model.
2. Although I agree with your statement that the price going down by 35% does not prove the short case was right, as with Brad Safalow of PAA Research's July 2025 original publication. However, there were some valid points in Brad's arguments, such as the extreme valuation and macro headwinds. It's a masterful short in July 2025, and I think the PAA folks are sharp. Whether they are right or not does not really matter as long as they got the 35% decline right.
I advised caution several times in two quality communities and on Twitter/X since middle-2025. However, when the Bear Cave short article dropped, I felt it was a bit silly and a little "Johnny-Come-Lately". As I commented by quoting Buffett, "What the wise do in the beginning, fools do in the end."
What, if anything, makes Kinsale a better buy at these levels than Palomar?
I don’t know Palomar well enough to have a strong view, but at first glance, they seem to be growing faster right now, partly because they are more concentrated in fewer products than Kinsale.
That said, the profitability gap is hard to ignore. Both Kinsale and Palomar wrote roughly $2 billion of GWP last year, but net income was very different: around $197 million for Palomar versus $503 million for Kinsale.
What I find especially interesting is the composition of the combined ratio. Both companies are currently around a record-low ~76% combined ratio, but Palomar gets there in a very different way: a loss ratio of only ~28% versus ~57% for Kinsale, and an expense ratio of ~48% versus ~21% for Kinsale.
Part of the explanation may simply be product mix. Palomar has meaningful exposure to catastrophe and low-frequency/high-severity lines like earthquake. In quiet years, those lines can produce very low loss ratios because the big losses simply do not occur. But that does not necessarily mean the normalized loss ratio is structurally 28%. It may just look exceptionally good in benign years.
To me, the expense ratio is the more important number when thinking about scalability. Kinsale’s model clearly shows operating leverage. Palomar’s much higher expense ratio makes me less comfortable, even though the headline combined ratio looks great.
So while Palomar may currently look stronger on growth, I feel much better with Kinsale because the scalability of the model is much clearer. Palomar’s low loss ratio may be real for now, but I would be careful extrapolating it too far into the future.
Again, great writing. Always look forward to your interpretations
IMHO, the beauty of KNSL’s business is the niche that it serves. William and the other NKOTB managers may be excited that float is akin to negative interest rates which may now cause some crowding creating a softer market. Michael’s shown discipline so we wait for the weigh in and ignore today’s vote. At the point that too many TikTok streams become annoying for KNSL, Michael can ask Prem for Kasowitz’s number; wonders happened for http://ffh.TO afterwards. Someone should then tell Ari Emanuel to hold off booking Dylan.