The last lever: portfolio underwriting when price has a ceiling and risk doesn’t

For most of the last decade, carriers could correct a drifting book with rate or reserves. Both levers are now running out at once. Q2 2026 marked the eighth straight quarter of falling global commercial rates, with property down 12% and cyber falling for 12 quarters running, while US casualty rates rose 7% yet still lost ground to severity, according to Sandeep Haridas of IntellectAI.

Price has a ceiling; risk doesn’t. What’s left is selection - what a carrier writes, how much, and whether the bound book matches its intent. Too often the answer arrives 18 months late, in a loss ratio.

More than 40 industry experts and over 225 senior underwriters will gather to discuss these challenges at Underwriting Innovation USA, hosted by Intelligent Insurer at The Westin Times Square, New York, on 17 November 2026. In his morning keynote, Haridas will explain why rate and reserving can no longer correct a portfolio, how 2025’s decade-best industry result concealed an other-liability combined ratio of 114%, and how concentration, capacity and marginal contribution can become growth instruments rather than governance ones.

Haridas is EVP and insurance business head at IntellectAI, where he delivers advanced underwriting tools and strategic digital roadmaps for insurance carriers, MGAs and wholesalers. Previously, at Liberty Mutual Insurance, he drove digital transformation and process optimisation for Global Risk Solutions North America. He has nearly two decades of experience spanning technology, underwriting strategy and P&C insurance.

IntellectAI is the badge and lanyard sponsor of Underwriting Innovation USA 2026.

Q: Why are rate and reserves no longer enough to fix a drifting book?

Rate and reserves matter, but neither is a substitute for sound risk selection. I was brought up in adversity. I spent years in underwriting units where combined ratios touched 126% and the president’s chair was a revolving door. That business has since healed, but it taught me a lot.

Rate is no guarantee. In admitted markets, rates are regulated by the state, and elected insurance commissioners won’t approve a 25% or 30% increase just like that. Increases can linger for years, and competition, capacity and what the customer is willing to pay, expressed through the broker, limit you further, even in surplus lines. Meanwhile your cost of risk keeps rising: salaries, inflation, operating costs, AI investment.

Reserves deal with business you’ve already written. Favourable development can support reported results, but it can’t improve the book. If you have garbage in your book, reserves won’t clean it up.

That’s why our focus at IntellectAI is selection: what you write, on what terms, how much you retain and how each account changes your portfolio. A book rarely drifts because someone abandons the strategy. It drifts through thousands of small decisions made under understandable pressures. The executive challenge is to see their collective effect early enough to respond.

Q: How can concentration and capacity limits drive growth rather than restrict it?

We usually ask about concentration defensively: are we overexposed, and what should we stop writing? Ironically, as a schools underwriter and later product and underwriting director, I worked in a unit that wrote a majority of the public school districts in a state.  We never asked whether we were over-concentrated, because the answer was obviously yes.

The same analysis rarely shows where you’re under-exposed. Look closely and you may be near your limit in one segment while leaving capacity unused in more attractive ones. For us, that might have been charter or religious schools. That intelligence won’t come from brokers, who want you to write everything they bring you.

But unused capacity isn’t a reason to deploy it. Just because you have room in primary liability for nuclear power plants doesn’t mean you should fill it. You need to know your next account’s contribution to expected return. Does it diversify the book, or add to exposures you’re already on the hook for? Is your capital best placed today, and are you set up for tomorrow?

Used this way, concentration and capacity limits turn portfolio management into a growth conversation. They redirect underwriting attention, which is as scarce as capital, towards the business you want, rather than letting the order in which submissions arrive decide what gets written.

Q: What’s the earliest sign a book is drifting from intent, helping you design your Portfolio Navigator?

The earliest sign is a change in underwriting behaviour. Track your business mix weekly or monthly, by line and by state, and watch for shifts. If you wait for claims to reveal a problem, you’ve already lost.

You might see more quotes falling outside appetite, or more referrals on rate adequacy, with underwriters pushing for flat renewals when the organisation wants at least 10%. You might see exposures building in a segment where you should be holding steady. No single indicator tells you. It’s a jigsaw, and you need to know whether you’re assembling a tiger or a peacock. One can kill you. The other is good for Instagram.

That need for earlier visibility and practical intervention is the foundation of Portfolio Navigator, which we’re releasing in November and which already has an order before launch. It takes live policies and quotes, applies the carrier’s chosen measures and shows where each sits in the portfolio, so leaders can investigate patterns without exporting data elsewhere. Interventions are built in too, from referring part of the book to your best underwriters to automatic non-renewals, so you can act with a scalpel or a jackhammer. The goal is to connect seeing the drift, understanding it, responding and checking the response was actioned.

Q: What’s the one thing a CUO should focus on? 

Truly seeing the gap between portfolio intent and what’s actually being bound. If your strategy is E&O for nurses’ associations in hospitals with fewer than 200 beds, stick to it, and make sure what you bind every day fits. The chief underwriting officer should be able to ask: where are we deviating, why, is it deliberate, and who is addressing it? That matters all the more if the book is already trending 10 or 15 points above expected loss ratios.

It’s not only about the negatives. Where should you be growing? One client writes D&O for accounts with under $100 million in revenue, but there’s a big opportunity in the $100 million to $200 million band. The rate is adequate, the risk is worth taking and the capacity is there. All we had to do was turn the dial, and we could only see it because we’d captured data on every declined submission.

IntellectAI connects that executive question to everyday underwriting decisions through Xponent and Portfolio Navigator. Want to push a non-negotiable 3% rate increase across the book? The extra income is easy to calculate; the business you’ll lose in price-sensitive segments isn’t. Portfolio Navigator lets you preview those “if this, then what” scenarios, with referrals, non-renewals and a full decision ledger in one workflow. It works with Xponent or any carrier’s ecosystem as an independent judge of a book’s character, and its agentic AI investigates, guides and monitors without making decisions. The goal is to give underwriting leaders the time and evidence to steer the book while decisions are still being made.

Q: How do you spot the outliers in a book that looks healthy on average?

Average is never a measure in this industry. Averages are for tracking how tall your kids are getting. They tell you how the book adds up, not which decisions produce the result.

One toxic account, or five, can produce a multi-million-dollar full-limit loss inside a book that looks fine. A big broker may push you to take the garbage for a few gems, and you take it rather than lose their favour. And with renewals making up 80% or more of a book, exposures can change faster than you expect. One account going from $1 million to $7 million in payroll is an anomaly. A whole sub-segment growing that fast throws your projections out of the window.

To find these pockets, draw lines. Anything over $25 million in segment premium, or wherever payroll, TIV or square footage is accumulating, gets a closer look. Benchmark against similar risks and industry data such as AM Best, then follow the signals down to individual accounts and relationships. Portfolio Navigator’s arrows show where an account was last year and where it is now, so you see movement, not a snapshot. It’s like watching how fast the universe is expanding.

And watch behaviour. I was an underwriter, and I was very good at making up stories. Is the reason for an exception sound, or is it November and someone has a quota to meet? A flat first renewal is fine. A fourth isn’t.

Did you get value from this story?  Sign up to our free daily newsletters and get stories like this sent straight to your inbox.