BOW - NYSE

Hello, and welcome to Bowhead Specialty's Fourth Quarter 2025 Earnings Call. [Operator Instructions] As a reminder, this conference is being recorded. If you have any questions, please disconnect at this time. With that, I would like to turn the call over to Shirley Yap, Head of Investor Relations. Shirley, you may begin.

Thanks, Marianna. Good morning, and welcome to Bowhead's Fourth Quarter 2025 Earnings Conference Call. I'm Shirley Yap, Bowhead's Chief Accounting Officer and Head of Investor Relations. Joining me today are Stephen Sills, our Chief Executive Officer; Brad Mulcahey, our Chief Financial Officer; and Derek Broaddus, our Head of Casualty. As we introduced last quarter, we'll be inviting an additional member of our management team on our earnings calls to share insights from the area of expertise. Today, we are joined by Derek Broaddus, who heads our casualty team, Bowhead's largest division. Derek will walk us through Bowhead's casualty portfolio and offer his perspective on the casualty markets in which we operate. Turning to our performance. Earlier this morning, we released our financial results for the fourth quarter of 2025. You can find our earnings release in the Investor Relations section of our website. And later this evening, you will also be able to find our Form 10-K on our website. I'd like to remind everyone that this call contains forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. Investors should not place undue reliance on any forward-looking statement. These statements are made only as of the date of this call and are based on management's current expectations and beliefs. Forward-looking statements are subject to risks and uncertainties that could cause actual results to differ materially from those contemplated by these statements. You should review the risks and uncertainties fully described in our SEC filings. We expressly disclaim any duty to update any forward-looking statement, except as required by law. Additionally, we will be referencing certain non-GAAP financial measures on this call. Reconciliations of these non-GAAP financial measures to their respective most directly comparable GAAP measure can be found in the earnings release we issued this morning and in the Investor Relations section of our website. With that, it's my pleasure to turn the call over to Stephen Sills.

Thank you, Shirley. Good morning, everyone, and thank you for taking the time to join us today. I'm very proud of Bowhead's accomplishment in 2025. We delivered disciplined premium growth of 24% for the year, surpassing our original expectation of 20%. We also had a meaningful improvement in our expense ratio coming in below 30% for the year and better than the low 30s range we expected at the start of 2025. Together, these achievements resulted in an over 30% growth in our adjusted net income for the year and adjusted return on equity of 13.6% and diluted adjusted earnings per share of $1.65. I'll begin with gross written premiums. Bowhead's GWP increased 21% in the fourth quarter to $224 million and 24% for the full year to approximately $863 million. We achieved disciplined premium growth from each of our divisions in the quarter and for the full year with casualty driving the increase. Given our emphasis on underwriting discipline and prioritizing profitability over volume, we're pleased to have delivered stronger-than-expected growth in the fourth quarter. In Casualty, GWP increased approximately 26% in the fourth quarter to $133 million and 28% for the full year to $551 million. The growth in both periods was primarily driven by our excess casualty portfolio. Our fourth quarter growth came in stronger than expected, driven by construction project risks that were quoted earlier in the year, but delayed due to macroeconomic factors we discussed in previous earnings calls. The green lighting of these projects added just under 30% to our fourth quarter casualty premiums. While we like the profitability of the construction project business and expect new construction projects to continue, the nonrecurring nature of this business may create lumpiness in our GWP. In our Professional Liability division, GWP increased approximately 4% in the fourth quarter to $48 million and 9% for the full year to $174 million. Our fourth quarter growth was primarily driven by our cyber liability portfolio, where we continue to target small and midsized accounts facilitated by our digital underwriting capabilities. Our full year growth was driven by commercial public D&O and miscellaneous errors and omissions. In our Healthcare Liability division, GWP increased approximately 8% in the fourth quarter to $34 million and 14% for the full year to $116 million. Our growth in both periods was driven by our health care management liability and senior care portfolios. Additionally, our hospitals portfolio, which represents the largest portion of the division's full year premiums at almost 30%, continued to grow while we reduced our total limits deployed. For Baleen, GWP increased 47% from Q3 to over $9.1 million. And we're proud of the fact that for the full year, Baleen generated over $21 million. The momentum we saw in the fourth quarter gives us confidence in Baleen's continued expansion and its anticipated contribution to our broader digital initiative. With a strong year behind us, I'm even more excited about Bowhead's future. As we've said before, Bowhead was built to deliver sustainable and profitable growth across market cycles, and we do that by delivering our products through 2 complementary underwriting models. Our first model is our craft underwriting model, the foundation of our company. It is led by experienced underwriters who specialize in complex nonstandard, high-severity risks and who deliver tailored solutions for our brokers and insurers. Our second model is our digital underwriting model, which represents the technology-enabled low-touch approach to our specialty flow business. This model began with the launch of Baleen in the second half of 2024, focusing on small, harder-to-place risks with restricted coverage. We then expanded this technology to handle the high volume of small and midsized submissions that were within our appetite, but historically, not cost effective for our craft underwriters to get to, a capability we call Express. Express automates the underwriting process that used to be repetitive and time-consuming, allowing our underwriters to make disciplined underwriting decisions within minutes. We first applied Express to our small and middle market cyber liability products in Q2, then broadened it to an E&O product in the second half of 2025. While our craft model delivered over 97% of our GWP in 2025, we've been able to achieve our sub-30 expense ratio even before the digital model is fully scaled. For example, in 2025, head count grew just under 19% from 249 people to 296, while GWP grew 24%. And in the fourth quarter alone, head count increased less than 3%, while GWP grew 21%. Together, Baleen and Express form our digital underwriting model, designed for speed, consistency and disciplined decision-making, all while preserving the underwriting culture that defines Bowhead. We look forward to introducing you to our Head of Digital on a future earnings call so you can hear directly from the team leader driving this effort. Turning to our premium outlook for 2026. We continue to expect profitable premium growth of around 20% for the full year. While we anticipate the growth coming from each of our divisions, we believe the main source of this growth will be driven by our Casualty division, followed by the growth stemming from our digital capabilities. With that, I'll turn the call over to Derek, who's been in the Casualty business over 30 years and won the Insurers 2025 E&S Underwriter of the Year Award. Derek, over to you.

Thank you for the introduction, Stephen, and good morning, everybody. Bowhead wrote its first casualty policy at the end of 2020. So we never wrote large limits for low premiums in the pre-2020 years. We were born in an uprate relatively low limit environment, which still largely exists today. When Bowhead began, the commercial casualty market had just emerged from a 15-year-plus soft market where pricing was suppressed and limits were abundant, all while social inflation was brewing in the background. We think that the payback equation between limit and price in that time was way off. And because of the tail, we still don't think the bill has totally come due for the industry's pre-2020 prior year adverse development. As a 30-year veteran of the industry, I'm happy to say that it has never been a better time to be a casualty underwriter. Our trading partners ask us what differentiates Bowhead's approach to casualty. Well, many of you have heard the insurance business is a people business. The best way to build a successful underwriting organization is to have the best underwriters in the business. Bowhead attracts top talent with our underwriting-first culture focusing on profitability over volume. Underwriters are also attracted to our straightforward distribution model, supporting and partnering with our trading partners. We are overwhelmingly surplus lines. We also are not distracted by fixing a book of business. We are laser-focused on managing and building our current portfolio. It's worth mentioning here that while we remain a predominantly remote organization, we are constantly on video talking about risks with what we call roundtables. Our underwriters are accessible anytime, anywhere. It is an advantage to be able to hire talent no matter where they sit. Another meaningful benefit to this structure is that it is easy to include less senior underwriters from around the country in complicated underwriting meetings that might have been near impossible in a traditional office setting. No one is above being questioned or challenged at Bowhead. In fact, it's encouraged. Additionally, Bowhead Casualty deliberately avoids classes that are well-known hotspots. Two examples are [ for-hire ] Commercial Auto. We don't write risks that are in the business of hauling people or things for others, and we have limited exposure to large national accounts. Our focus remains on profitable classes where we have expertise and experience. We manage limits carefully in today's market. Our average excess limit deployed is just over $5 million rather than the $25 million blocks that were common pre-2020. Large excess towers that once required only a few markets to complete now require many markets at better pricing. We also avoid low price per million, high excess placements that require the deployment of large limits and are more exposed to loss than ever before due to social inflation and nuclear verdicts. Bowhead's Casualty portfolio benefits from today's positive rate environment, lower required limit to participate on towers and the ability to exercise disciplined risk selection. We know that outsized awards and litigation funding are not going away. Social inflation is not a surprise to anyone anymore. Even with improved attachments, risk selection and rate still matter. In terms of our disciplined approach to underwriting, our focus is on deal fundamentals. We believe that walking away from deals that don't make sense is just as important, probably more important than any piece of new or renewal business. Some might say knowing when to walk away is the toughest but most valuable underwriting skill there is. From a market perspective, in excess, limit discipline largely remains. Many excess towers continue to see limit compression from incumbents. This creates new opportunities for underwriters like Bowhead. However, one moderating influence on rate is the movement of admitted markets into the E&S space as was typical in past insurance cycles. Also, nonrisk-bearing MGAs and broker sidecars are bringing more capacity into the U.S. casualty market. We agree with certain industry leaders when they say there is a fundamental misalignment of interest in some nonrisk-bearing underwriting facilities. Overall, we remain confident in our ability to grow profitably. We think the current market is competitive, but there is a relatively healthy balance of rate and limit management. Our brand in casualty continues to grow and strengthen. Submissions are growing faster than we can quote and investments in technology, our digital platform and talent will allow us to capture more opportunities that fit our appetite. With that, I'll pass the call over to Brad to discuss our financial results.

Thanks, Derek. Bowhead had generated adjusted net income of $15.5 million or $0.47 per diluted share and adjusted return on average equity of 14.1% in the fourth quarter of 2025. For the full year, Bowhead's adjusted net income increased 30.2% to $55.6 million or $1.65 per diluted share, and adjusted return on average equity was 13.6%. Our strong results were driven by top and bottom line growth. Gross written premiums increased 21% to $224.1 million for the quarter and 24% for the full year to $862.8 million. Our growth story was consistent throughout the year. We achieved premium growth in each of our divisions, with Casualty continuing to be the largest driver and Baleen generating $21.4 million for the year. Due to the timing of our annual reserve review in Q4 each year, we consider our full year loss ratio a more meaningful metric. For the full year, our 2025 loss ratio of 66.7% increased 2.3 points compared to 64.4% in 2024. The current accident year loss ratio increased 1.8 points due in part to higher expected loss ratios and trends after the annual reserve review as well as mix changes in the portfolio. The prior accident year loss ratio was unchanged as a result of the annual reserve review, but increased 0.5 points due to audit premiums recorded in 2025 that related to prior accident years. As a reminder from previous earnings calls, the audit premium related reserves in the prior accident years is not based on actual losses settling for more than reserved and did not represent an increase in estimated reserves on unresolved claims. We're simply putting loss reserves into the appropriate accident year regardless of when the premiums are billed and earned. And remember, since we've only been in operations for 5 years and write long-tail lines, our actual loss experience is limited. Because of this, our annual reserve review is primarily based on inputs from industry data. Our initial expected loss ratios are derived from a combination of internal pricing data and external benchmarks, while development patterns are mostly based on external benchmark patterns. We attempt to align all industry benchmarks to the nuances of our portfolio, including not writing risks that are in the business of hauling people or things for others and our lack of large national account exposures in casualty. Additionally, the development patterns we use attempt to take into account our excess position in particular lines, which generally results in later development patterns than primary positions. The most recent annual reserve review in Q4 resulted in various adjustments that were smaller compared to our adjustments in Q4 2024. But most importantly, we had no prior accident year development in our aggregate net losses for 2025 as a result of this review. As you will see in our 10-K, we reallocated prior accident year reserves by division to align more closely with the actuarially derived projected loss ratios and development patterns. These reallocations were primarily in professional liability, where we reduced the '21 accident year while increasing the newer accident years and in health care, where we reduced the '23 accident year and increase the '22 and '24 accident years. These were offset by a decrease in casualty for the '22 accident year to align with updated projected loss ratios, all resulting in no prior year development on an aggregate net basis. More specifically, in Professional, the '21 accident year is performing well, resulting in a favorable $3.5 million reduction in IBNR. However, the limited experience in subsequent years, coupled with declining rates, warrants caution. The '22 accident year in particular, where our early experience is deviating from the industry development patterns was increased by $2.8 million at year-end. Similarly, in health care, the '23 accident year is performing well. But in the '22 year, our early experience is also deviating from the industry development patterns. This warranted a $2.2 million increase in the '22 accident at year-end, along with a $3.3 million increase in the '24 accident year out of an abundance of caution. These adjustments to the industry development patterns are another example of conservatism in our reserving. We're reserving as if the industry patterns are correct for now and therefore, reallocating reserves in select areas. Lastly, we increased some of the '25 accident year initial expected loss picks to align with actuarial estimates. In alignment with our conservative approach to reserving, we are carrying loss ratios in the '25 accident year above the industry estimates on a majority of our product groups. Overall, our actual experience of paid claims and reserves continues to be better than we actuarially expected. And at the end of the year, IBNR as a percentage of total reserves was 90%. Turning to our expense ratio. We consider our full year ratio a more meaningful metric to monitor the trending of our expense ratio due to the inherent volatility quarter-to-quarter. For the year, our 2025 expense ratio of 29.8% decreased 1.6 points compared to 31.4% in 2024. The reduction was driven by a 2.3 point decrease in our operating expense ratio, which was partially offset by a 1.1 point increase in our net acquisition ratio. The decrease in our operating expense ratio was due to the continued scaling of our business, scaling that is accelerated by the realization of various technology initiatives to improve efficiencies. The increase in our net acquisition ratio was driven by the increase in broker commissions due to mix changes in our portfolio and to a lesser extent, the increase in the ceding fee we paid to American Family. Overall, the effect of our loss ratio and expense ratio contributed to a combined ratio of 96.5% for the year. As a reminder, we don't write property and we don't write natural catastrophe-exposed risks. Turning to our investment portfolio. Pretax net investment income for the quarter increased approximately 36% to $16.6 million and 44% for the year to $57.8 million. The increase was primarily due to a larger investment portfolio resulting from increased free cash flow. At the end of the year, our investment portfolio had a book yield of 4.6% and a new money rate of 4.5%. The average credit quality of our investment portfolio remained at AA and our duration increased from 2.9 years in Q3 to 3 years at the year-end. Our effective tax rate for the year was 20.1%. As a note, our effective tax rate may vary due to items such as state taxes and stock-based compensation. Total equity was $449 million, giving us a diluted book value per share of $13.45 for the year, an increase of 22% from year-end 2024. Turning to our expectations for 2026. We continue to expect a GWP growth of around 20% for the year. As Stephen mentioned, the growth should come from all divisions but led by continued momentum in our Casualty division and growth driven by our digital underwriting capabilities. From a ceded perspective, although our main quota share and XOL treaties renew in May later this year, we've renewed our cyber quota share treaty effective January 1 of this year at 65%, up from 60% in 2025 and increased our ceding commissions. As a note, at each renewal, we consider various factors when determining our reinsurance coverage. While we may adjust our reinsurance program, including our retention to support capital needs, we expect our reinsurers to maintain a financial strength rating of A or better. Furthermore, we expect our 2026 loss ratio to be in the mid- to high 60s due to product mix and our reliance on industry loss trends. Additionally, we expect our expense ratio to be below 30% for the full year due to the continued scaling of our business, scaling that is accelerated by the realization of various technology initiatives to improve efficiencies. We expect our expense ratio in the first half of the year to be slightly higher than the second half due to payroll taxes. Therefore, we believe our combined ratio will be in the mid- to high 90s for the full year and return on equity to be in the mid-teens. Turning to our investment portfolio. We expect to extend our duration slightly from 3 to 4 years. This change is not because we're predicting interest rates to decrease, but to closer match the duration of our investments to the duration of our liabilities. And lastly, from a capital perspective, in November, we issued $150 million of 7.75% senior unsecured notes that are scheduled to mature on December 1, 2030. We expect the proceeds to be sufficient for our year-end 2026 regulatory capital requirements, but we'll continue to assess throughout the year. With that, we'll turn the call over for questions.

[Operator Instructions] Our first question will come from Meyer Shields with KBW.

Great. Thanks so much. Brad, I appreciate all the detail on the prior year reserve development. Can you walk us through what that implies for price adequacy for 2026 for professional financial lines -- I'm sorry, for professional health care?

Meyer, thanks for the question. Yes, we detailed quite a bit on our prior accident year development, changes around in our IBNR in particular. We think we're priced well. We think pricing is coming in above trend. But we do have a couple of pockets that was just normal changes. I don't want to read too much into it. They were actually pretty small changes. So I don't think there's really a pricing impact from it. It's more just taking a conservative approach to the reserving and adjusting where it was warranted, nipping and tucking around the edges, if you will.

Okay. No, that's helpful. And I guess a question on Baleen. When -- right now, obviously, it's a very, very small percentage. But when -- as it grows, should we think of it as having the same loss ratio characteristics as Casualty? Is it going to be more evenly distributed? That's the wrong way of phrasing it. But when you look at the different segments, you've got different loss ratio profiles there. And I'm wondering how we should think of a mature Baleen in that context?

I think that -- this is Stephen. I think that the Baleen loss ratio will be superior to the general large casualty business. The -- I'm not as certain as when we get into the Express Casualty business, whether that will probably mirror more of what the larger casualty business is. But the Baleen business, based upon the restricted nature of the coverage, we think that, that will have a superior loss ratio.

Okay. Fantastic.

Your next question will come from Rowland Mayor at RBC Capital Markets.

I wanted to quickly ask on how you translate industry data into the loss ratio picks. I assume you're trying to be better than the industry and your business is much more niche, but how do you kind of get granular on that data and use it in your business?

Yes, Rowland, this is Brad. Thanks for the question. We've been doing this for a couple of years now. Obviously, we don't have enough data on our own to set our picks and development patterns. But we do have a third-party actuary who has very detailed proprietary information that they give us. So this is not Schedule P industry data that we're using. This is something that we can slice and dice, as I mentioned, we don't really have a big Fortune 1000 exposure in casualty, for example. That's given everybody a lot of heartburn. So we're able to tailor these industry benchmarks to our portfolio to an extent. There's only so much you can do, obviously. So -- but we think that the proprietary information that we now have helps us. And looking backwards, it has been pretty accurate with foretelling what is happening in the casualty market and the other markets that we participate in. The development patterns are probably the one that we're starting to see our own data, but I don't know if we can say we have a trend in our data for that. So we are definitely using the industry development patterns. And I think that's adding a little bit more conservatism into our reserves as well. Does that help?

Yes. That's super helpful. And then I wanted to talk about the expense ratio target. You're now sub-30. I get there's going to be a step-up in acquisition costs from the deal in May and maybe some first half payroll taxes. But is there a place you're thinking about long term where you can get the expense ratio down to?

Yes. I think we have headwinds with our ceding fee going to American Family, as you point out, but we got a lot of tailwinds in the technology initiatives that we've put in place. We saw halfway through last year, we're starting to see the benefits of those a lot faster than we had thought, surprisingly so. So we're still going to squeeze as much as we can out of this expense ratio. But it is sort of a last year, low 30s. We were happy being in the low 30s, but this is sort of a new paradigm now with some of the tools out there. So hard to say where it will come in, but I think we're comfortable low 30s, and we'll do our best to get it even lower than that.

Your next question will come from Bob Huang with Morgan Stanley.

So my first question revolves around Casualty. I wanted just to follow up with something that you talked about a little earlier. As we -- I think previously, right, you've talked about the undisciplined nature of some of the underwriters, but also the risk of like these eye-watering verdicts from social inflation. As we go into 2026, like is there any sign that pricing environment and excess casualty maybe is beginning to plateau? Is the market significantly offering substantial growth in 2026 and beyond, just given where we are in the underwriting cycle for the casualty side?

Bob, this is Derek. I like directionally the limit discipline that we're seeing in the market. I think that's holding pretty well. I would say that there's a lumpy moderation going on. You're seeing some deals, in particular, that are still dealing with adverse development from prior. And then on other deals, you're seeing 5 years of compounded double-digit rate and great loss experience. So you're going to see a little bit of a mix of response from the market for those 2 different types of risks that are coming in. For the most part, though, as Brad said, I think directionally, we're seeing rate exceed loss trend.

Got it. Okay. No, that's very helpful. My second question is more of AI and automation. So when I look at Baleen on the automated underwriting side, is there a reason to believe that at some point in time, the technology on that side is advanced enough that you can essentially disintermediate brokers as in you're going directly to customers for that line of business or maybe even if that line of business gets bigger, like higher limits, can you skip the brokers and going directly to customers?

In the type of business we do, I don't see that happening anytime soon. I mean, carriers for as long as I've been in the business, have talked about could brokers been disintermediated. At the end of the day, the type of specialty insurance we do is not homogenous. It's not like a family automobile policy or a homeowners policy. There's a lot of complexity to it that I think needs a lot of explaining. And I think the broker brings a lot to the table. And even further than that, the wholesalers play a large role because many of the retailers who are good producers of the business are not experts in the nuances, the ins and outs of some of the specialty insurance. So we don't see that going away anytime soon, number one. Number two, we think the biggest advantage at this time is the speed of being able to get to the business. I think we've mentioned before that close to -- in the casualty space, we don't even have the ability to get to 90% of our submissions that come in the door. It's just -- unless it's a premium that's maybe 50,000 or above, we don't have the resources to handle it. In the next several months, we're going to be getting -- we're going to be able to get our system online in Express where we'll start to be able to handle that business. So it's a matter of being a great underwriter assist in doing the business to help us grow profitably. But the idea of disintermediating is not on our radar.

Your next question will come from Pablo Singzon with JPMorgan.

So first question for Brad. How much did mix contribute to the 1.8% [ attritional ] loss ratio uptick at '25? I think about '26, the reed loss pick should flow through at the same level. So if we use 66.7% in '25 as a base, how would you sort of frame the impact of any mix impacts in '26?

Pablo, thanks for the question. I don't really have an answer for that yet. You're right, we will use our '25 loss picks as sort of the starting point for 2026. But that doesn't mean it's set in stone at that level. We'll review these every quarter. And if we need to make changes, we will based on rate or anything else we're seeing or the industry changing as well. I think there is a -- we're probably reaching like the upper limit of how much mix plays into it as you see the casualty portfolio is such a bigger portion of the overall premium. But even with that -- within Casualty, there's mix. So the primary casualty has a different loss pick from excess, for example. So there's mix within mix, if you will, and we just kind of have to see how that plays out. I wish I could give you a more precise number for next year, but that's the best I have.

Okay. And taking a step back, right, and I appreciate, Brad, you provided a lot of detail on the combined ratio. But I guess as I think about the overall number, it seems to me that all else equal, maybe the loss ratio should go up a bit, right, maybe for mix, acquisition expense will probably go up. And the question is, do you expect to fully offset those with a lower expense ratio? Or will it offset be only partial? And I know that's spitting hairs, but I guess just given where combined ratio is, even a 50 bps movement can be meaningful. So any perspective you can provide there?

Sure. I guess -- and Stephen, feel free to jump in. But I think the way that we approach this is we will try to get as low of an expense ratio as we can regardless of where the loss ratio is going. And we will let the loss ratio do what it does based on how we feel comfortable with our reserves regardless of what the expense ratio is doing. So hopefully, those 2 come together, and there will be some offset if the loss ratio does trend up. We do have the benefit of older accident years that have lower loss picks. As those roll off, each year, you will see that impact the loss ratio. So I think that's why we're saying our target is the mid- to high 60s on the loss ratio. But I wouldn't read too much into that being a huge increase, but that's probably where I would stand on that.

Well said.

Your next question will come from Cave Montazeri with Deutsche Bank.

First question is on Baleen. It looks like growth is picking up nicely after what was arguably a slower-than-expected first half of the year. So I guess my question is, what's been working so well in the second half of 2025? And how should we think about growth in 2026 for Baleen specifically?

Well, part of it has to do with acceptance that there are certain entrenched markets and with relatively small premiums of, say, $5,000, there is not a ready acceptance to market them. So there's a certain amount of hanging around the net, if you will, and continuing getting our message out to brokers of what we're offering, how our policy form compares, how our commission level compares our service level, all those things and ultimately getting the message through until people start to try us, and it becomes more and more accepted. And now as it starts to build, we've started to put more infrastructure behind it in terms of people -- more people going out and speaking to brokers about the business. And then success breeds success. They see that what we've done has -- it's been worth the effort to try us, and they're trying us more. And with adding more marketing people, we've been able to add more distribution points, and that's still building on itself. But also even beyond Baleen, and we tried to make this clear earlier, but even beyond Baleen, building on that technology is enabling us to do that smaller business, not the restricted type business of Baleen, but the smaller business that we call Express, which is going to be another real plus, I think, in '26. Does that help?

Second question -- my second question is for Brad. On the investment portfolio, it's good to hear that you can increase the duration from 3 to 4 years. With your new money yield being below the book yield, would you also consider maybe going up the risk curve? You have a very defensive portfolio right now.

Yes. Thanks for the question, Cave. The answer is no on that one. When we discussed moving our duration up, we explicitly kind of agreed that we're not going to change the risk profile of the portfolio. We like the conservative position in it.

Your last question will come from Cameron Bianchi with Piper Sandler.

This is Cam on for Paul Newsome. Just one question for me. On the lower expense ratio guide for 2026, how much of that improvement would we say is attributable to scale versus mix?

Yes. Good question. I think the -- our previous guidance of low 30s, that was scale. I think the new guidance of being below 30%, that's the impact of the technology. And it's not just technology in the digital platform as well. There's -- we're deploying technology on the craft business as well that's helping with efficiencies. Our claims team is getting more efficient. So we're really seeing that across both. So I'd say the difference between the low 30s and where we actually end up would be that the impact of the non-scaling of the business, if you will.

That concludes the question-and-answer portion of today's call. I will now hand the call back to Stephen Sills, CEO, for closing remarks.

Thank you. Bowhead delivered another strong quarter to end a great year. Before we go, I wanted to say thank you again to our colleagues and brokers for making 2025 such a successful year. Thank you, and we look forward to speaking to you along the way.