Reading time: 9 minutes · Listening time: 6 minutes. Prefer to listen? Flip reads this week's Brief here.
Nobody in your renewal meeting will say the word "hardening." They'll just stop offering decreases.
I sat in two rooms this week with senior underwriters from two of the largest financial lines carriers in the market. Two different companies, nearly the same conversation. Both raised the NERA and Cornerstone midyear reports before I did. Both cited those reports as the thing justifying their pricing outlook for H2 right now.
I'm trying to get both to sponsor these Briefs, so you can use your imagination on which carrier partners!
Here is what I took from both convos though... The crafty underwriters read both of these instruments and weight them very differently. Settlements are the scoreboard. Filings are the weather report. Almost everything that confuses buyers about this market comes from reading those two as if they were the same thing.
Here's what made the cut:
- The ordinary securities case got more expensive this year, and the four headline settlements hide it.
- Four desks sit between a research report published in July and the number on your renewal quote. The third is the one nobody explains.
- Run an Insurtech, MGA or mutual and you'll likely never be a direct securities defendant. The trend reaches your renewal anyway.
For the last 20 years, these reports have been getting baked into the narrative. This week we break down why at a 7th grade level!
Frequency starts the conversation. Severity ends it.
Summary
Settlements are the scoreboard.
All of 2025 produced 73 securities settlements worth $3.1 billion. The first half of 2026 produced 39 worth $2.2 billion. Just over half the count, nearly three quarters of the dollars. Four cleared $100 million, and the mega settlement is the number people reach for to explain a total. It tells you nothing about the other 35.
The median tells you about the other 35.
In Cornerstone's dataset it came in at $20 million, up from $17.6 million for full-year 2025, itself the highest in at least nine years. Cornerstone put the average at $56.4 million, up from $42.2 million. NERA counts a different set of cases and reported its own average at $54 million, up 32%, with its median unchanged at $18 million.
Here is what moved it, and it is not what most people assume.
The cases being settled got bigger. Median plaintiff-style damages, meaning the loss investors claimed measured the same way across cases, ran $660 million this year, up from $290 million. And 36% of settlements involved claimed damages above $1.5 billion, against 14% across 2017 to 2025. Yet the median settlement came in at 5.4% of those damages, down from 6.3%.
Defendants paid a smaller share of a much larger number. Your excess layers sit on the number, not the percentage.
(source: Cornerstone Research, Securities Class Action Settlements: 2026 Midyear Assessment, July 2026; NERA Economic Consulting, Recent Trends in Securities Class Action Litigation: H1 2026 Update, July 21, 2026)
So what?
A settlement is a loss that has crystallized.
That is why this report outranks any filing count in a pricing model. A filing is an allegation the defendant may beat. A settlement is an agreed number, and it lands in the severity assumptions behind next year's rate. When the median moves, the ordinary case got more expensive. That lands hardest above the primary layer, because excess towers, the layers stacked above your primary policy and each held by a different carrier, are priced off severity.
None of which has touched your premium yet. Investment managers are still seeing D&O decreases. So are plenty of banks. If your renewal came in flat this spring, that was the lag, not the verdict.
The four desks between a July report and your renewal quote
Summary
Carriers do not keep separate money for public companies.
One portfolio backs the public D&O, the private D&O, the mutual's D&O, and the managing general agent's errors and omissions cover. One reinsurance treaty sits underneath it. One rate-need calculation runs across the whole book.
It moves the portfolio, and the portfolio prices you.
Both underwriters walked me through the same sequence.
The first desk is the research house. Cornerstone and NERA publish frequency and severity for the whole market, independent of any single carrier. No one insurer sees enough claims to know what is happening, so they all read the same two reports to find out.
The second is the actuary. Industry data gets weighed against the carrier's own claims. Where the two disagree, the carrier trusts itself. Where they agree, the line gets repriced. Not your account. The line.
The third is where it reaches you, and it is the one nobody explains. Your D&O and a listed company's D&O frequently sit under the same reinsurance treaty, the contract your carrier uses to lay off part of its own risk. Some carriers segment those treaties, public in one section and private in another. Even then, the same reinsurers price both sections off the same market-wide data. It is all a reinsurer has.
Once the treaty reprices, the correction stops being a view your underwriter holds and becomes a cost they carry.
The fourth desk is the underwriter across from you. They never get a memo instructing them to harden. They get less authority. The decrease they could sign last year now needs a referral that mostly is not clearing.
That is why the renewal meeting sounds like sympathy rather than strategy.
Run a mutual and you are very unlikely to ever face a securities class action as a direct defendant. The correction still reaches your renewal, through a treaty you are not party to and will not see. The same four desks sit under your errors and omissions cover and your crime cover, though those may sit with a different carrier under a different treaty.
The LION Lens
What happened — Among monoline D&O writers, the carriers who write this as their main line, direct premium peaked near $15 billion in 2021 and fell four years running, to just over $10 billion in 2025. In 2025 the direct loss ratio rose about five points, to 54.5. Reserves for accident years 2023 and 2024 both developed adverse (source: AM Best, US Director & Officers' Liability Remains Profitable, But Warning Signs Are Evident, June 2026).
Why it matters — That is the second desk's problem in numbers. Roughly a third of the premium base is gone and the claims got more expensive. Two recent accident years are already short, and that is next year's problem more than this year's. A reserve gap reaches your quote a renewal cycle after it reaches the balance sheet.
Practical implications — In the second quarter, US financial and professional lines turned positive at 1%. D&O came in at 1% on its own, reversing a 3% decline the quarter before. Outside casualty, it is the only major US line not falling (source: published industry rate indices, Q2 2026).
So what?
This one is not built like the hard markets you have priced through before.
Capital is not leaving. Global reinsurance capital hit a record $790 billion at the end of March, and property catastrophe treaty renewed down 15% to 25% at midyear (source: published reinsurance market estimates, midyear 2026). US property rates fell 13% last quarter. The global composite fell 6%, its eighth straight decline (source: published industry rate indices, Q2 2026). All of that sits on the property side. Whether the same capital reaches financial lines treaties on the same terms is a separate question, and in both rooms this week the answer was no.
That gap is the story.
Nobody pulled back because they ran out of money. One line stopped earning its cost of capital while everything around it got cheaper. Call it micro-firming.
It also explains why the headline looks so small. A carrier that needs margin and cannot get twenty points of rate takes it somewhere else. That is what the quarter recorded. Participations trimmed. Insurers off the top of towers. Fresh scrutiny of large-cap appetite. And Side A capacity moving down the structure, the piece that covers directors directly when the company cannot indemnify them (source: published industry rate indices, Q2 2026).
The composite also hides who is paying it.
| Sector |
D&O |
E&O |
Cyber |
Fiduciary |
| Regional and mutual insurers |
−2% to +2% |
+1% to +5% |
−4% to +1% |
+2% to +5% |
| MGAs |
−3% to +1% |
0% to +4% |
−4% to +1% |
+2% to +5% |
| Insurtechs |
+2% to +7% |
+4% to +9% |
0% to +6% |
+3% to +6% |
| Banks |
−4% to 0% |
0% to +4% |
−7% to −2% |
0% to +3% |
| Investment managers |
−7% to −3% |
−5% to −1% |
−8% to −3% |
0% to +3% |
Ranges are indicative and reflect general market direction, not binding quotes. These are LION book observations from renewals quoted May through July 2026, not a published index (source: LION Specialty, H2 2026).
Read the top row against the bottom one. A regional insurer is looking at flat D&O with errors and omissions up as much as five points. An investment manager is still being offered reductions on both. Same quarter, opposite directions.
Errors and omissions is firming harder than D&O for every buyer here. That happens when the loss pressure sits in professional liability, not the boardroom. Our D&O column reads softer than the national number because of book mix. The published composite carries large-cap public business. Our book sits in the middle market.
If you write on delegated authority, your errors and omissions renewal is not priced off a table. It is priced off your file. Underwriters want the carrier audit results, the bordereaux history and the complaint log. The managing general agents who bring those unprompted get a different conversation.
Cyber runs against the whole table. Capacity is still abundant and well-controlled accounts are still getting decreases. Across all commercial buyers, cyber came off 3.5% in the first quarter, one of the steepest declines of any line (source: The Council of Insurance Agents & Brokers, Q1 2026 P/C Market Survey). In our book it is the smaller buyers and the insurtechs seeing tighter terms. Fiduciary runs on a track of its own, and nothing in these two reports drives it. That is how a flat renewal in total hides a firming D&O line underneath.
The one-point composite is real. It is also the middle of a spread running from minus eight to plus nine.
The LION POV
How we are advising clients on this:
-
Read the panel, not the rate. Your panel is the carriers sharing your program. Track who cut their line, who moved up the tower, and who quoted but declined to lead. That is the fourth desk showing you its authority problem.
-
Rebuild the top of your tower first, or test your single limit against the median. High excess is where capacity is thinnest and where a shifting severity distribution bites first. If you buy one limit rather than a tower, ask what it was sized against, and when. Your carrier is reserving to a $20 million median.
-
Bring your governance record, don't wait to be asked. A parallel derivative action, a suit brought on the company's behalf against its own directors, rode along with 61% of these settlements, up from 51%. Better your underwriter prices yours on evidence than on the average.
The carriers repricing this book are behaving rationally. Four years of shrinking premium against a rising loss ratio is not a position anyone holds indefinitely.
Want us to run your program through this same sequence before your next renewal? Reach out for a confidential conversation.
Your expiring premium was priced against a claim mix that no longer exists
Summary
Filings are the weather report.
They tell you what is forming, not what it will cost, and underwriters discount them hard for good reason. Of the 105 cases resolved in the first half, 65 were dismissed. The count matters less than the composition, and the composition turned over completely.
Three years ago the new money in plaintiff-side securities work was crypto.
This year crypto is down to three filings, on pace for the fewest since 2019. No COVID-19 cases. What replaced them looks like this. A company describes an AI capability on an earnings call. The stock moves when someone questions whether it exists as described. A complaint follows. Cornerstone counted 15 in the first half. NERA, on a slightly different definition, counted 18, already past the 17 filed in all of last year.
The counts undersell them.
Those cases drove $385 billion in Disclosure Dollar Loss, the market value investors say evaporated on the days the alleged truth came out. That is 73% of the half-year total, from roughly one filing in eight.
(source: Cornerstone Research and the Stanford Law School Securities Class Action Clearinghouse, Securities Class Action Filings: 2026 Midyear Assessment, July 29, 2026; NERA Economic Consulting, Recent Trends in Securities Class Action Litigation: H1 2026 Update, July 21, 2026)
So what?
Cornerstone counted 121 core federal filings in the first half, excluding merger-objection suits, against 93 in the back half of 2025. NERA, counting a wider universe, put it at 118. Same direction, different rulers.
If the current pace holds, any given NYSE or Nasdaq company faces a 4.4% chance of being sued this year. That is the highest since 2019, above the 3.8% average since 2012. Your expiring policy was rated on a book where most of that had not happened yet.
One number deserves a second look. Cornerstone recorded no cybersecurity-related securities class action filings in the first half. That reflects plaintiff economics, not a drop in the underlying exposure. Plaintiffs found a target with bigger stock drops attached to it.
Read the scope of these reports before you take comfort from any zero in them. Both count securities class actions and nothing else. Regulatory enforcement runs on its own track, and the SEC has its own view of AI disclosure. A quiet year in these numbers is not a quiet year in your exposure.
One item worth putting in front of your general counsel. They may want to review, with counsel and under privilege, what your institution has said publicly about AI against what you can evidence today. That review needs a current model inventory behind it. A disclosure you cannot document is the one that gets tested.
If you are a mutual, an MGA or a private institution, the analog is not the earnings call. It is your marketing materials, your policyholder communications and your regulatory filings, wherever AI capability has been described.
The Bottom Line
The scoreboard is what already happened, and it is reliable. The weather report is what might happen, and it is not. Carriers price at the speed of the scoreboard, which is why this turn arrives as one point and a thinner panel instead of a headline.
Watch for the two to converge. If a carrier reserves a nine-figure AI case, the weather report becomes the scoreboard.
Both reports are free. Your underwriter read them in July. The lag between that reading and your renewal is the only advantage on offer, and it closes on your date, not theirs.
Three questions for your next risk committee
- When was the top of our tower last sized against a median settlement? Not against the limit we have always bought.
- Which carriers cut their line or moved up our tower at the last renewal, on D&O and on E&O? Do we know why?
- Has our disclosure committee reviewed every public AI statement we have made? The test is what we can evidence today.
LION has published a structured review of the five most common D&O program gaps: the D&O Contract Vigilance Blueprint, a five-day email course available to clients and subscribers preparing for renewal.
- The policy mistakes we see most often when reviewing D&O programs
- Where your personal assets sit when the company can't indemnify you
Want it? Just reply to this email with the word "blueprint" and I'll sign you up.
Thank you for reading today's edition.
Stay Covered Everybody,
-FLIP
P.S. Want to share this edition? Copy the link below:
https://lionspecialty.kit.com/posts/the-micro-firming-phenomenon-brewing-in-h2-margin-not-capital-profitability-and-the-economics-of-underwriting-financial-lines
And if this was forwarded to you, subscribe here: https://lionspecialty.kit.com/.
P.P.S. Nothing in this briefing constitutes legal, coverage, or compliance advice. This is market intelligence designed to help you ask sharper questions of your advisors and make better decisions at renewal.
You're receiving this because you subscribed to the LION Specialty Boardroom Brief.
Unsubscribe · Update your preferences