Social inflation in 2027: why rate increases no longer protect your loss ratio

Social inflation in 2027 will keep breaking the link between rate and profit in commercial casualty, because carriers have spent a decade pushing rate to outrun rising claim severity and in 2026 that strategy is running out of room. The casualty lines still losing money are rarely underpriced by just a few points, and more often they were selected and reserved on data that was never accurate enough to catch the severity now sitting in the book. The defense heading into 2027 is therefore accurate loss-history and exposure data at the point of risk selection, which a carrier controls account by account, while rate remains a blunt instrument shared with the whole market. Our earlier look at social inflation and the 2025 results covers what the last cycle did to profitability, and this piece looks ahead.

Why don't rate increases fix social inflation anymore?

Rate increases no longer fix social inflation because severity is rising faster than rate can follow, and commercial auto is the clearest case. According to AM Best, the line posted a $4.9B underwriting loss in 2024, its 14th straight year of losses, with claim severity climbing about 8% a year, more than double the roughly 3% economic inflation rate.

Rate did not sit still during those years. As per Conning, commercial auto has taken 55 straight quarters of rate increases and still runs above a 100 combined ratio, while liability costs have risen more than 60% since 2015. When severity compounds at 8%, a 10% rate increase buys roughly one year of cover before it falls behind again, a pattern we traced in 14 years of commercial auto losses despite rate hikes.

What will social inflation do to your 2027 loss ratio?

Social inflation widens the gap between the premium you book and the losses you later pay, and the driver is the tail far more than the average claim. According to Marathon Strategies, US corporate nuclear verdicts hit a record 135 in 2024, and since 2020 the count of large awards is up 309% while the dollars awarded are up 273%.

That tail is funded. Third-party litigation funding reached $16.1B in 2024, according to Westfleet Advisors, and funded plaintiffs tend to settle less and try more cases, so the result lands in reserves. AM Best estimates commercial auto is under-reserved by $4B to $5B industry-wide, and a reserve miss is a loss ratio you booked as a profit.

The current market makes this harder. As per the Council of Insurance Agents and Brokers, overall commercial premiums fell 2% in the second quarter of 2026, the first broad decline in years, with commercial property down 6.3% and workers compensation down 3.2%, while only the casualty lines held firm, with commercial auto up 4.5% and umbrella up 5.3%. Carriers are pushing casualty rate into a softening market and still not closing the gap.

Why is social inflation a risk-selection problem more than a pricing one?

Social inflation is mainly a selection problem because it punishes specific accounts, and rate spreads the cost of severity across every policyholder without being able to tell a clean fleet from one carrying a hidden loss. Selection can make that distinction, as long as the data behind it is right.

Most of the severity that hurts a book was visible at intake, sitting in a prior loss run, a driver schedule or a claims history that was keyed by hand and read too fast. A missed prior claim understates the exposure, an understated exposure prices the risk below its real cost, and under 8% severity growth that one error compounds every year the policy renews. This is the mechanism behind the real cost of inaccurate underwriting data, and it explains why a soft market can still produce hard losses.

How do you underwrite against social inflation in 2027?

You underwrite against social inflation in 2027 by making the data under each decision accurate and defensible, and three levers move the loss ratio here without any rate filing:

• Extract loss history and exposure data with contractual accuracy. Pibit AI reads loss runs, driver schedules and claims histories through extraction and research and guarantees 99.9% field-level accuracy with our human-in-the-loop, since one dropped claim is a severity signal you will pay for later.
• Score risk in a way you can defend. The risk score is deterministic and factor-level, so a reserving actuary or a reinsurer can see why a risk was accepted, which is the point of field-level provenance, and a score you can decompose is one you can defend when a nuclear verdict lands.
• Do it at volume without adding headcount. Pibit AI customers see 2x underwriter capacity, so accurate selection can scale while the market softens around the book, and the same discipline shows up in commercial auto, where data accuracy moves the loss ratio.

Expert validation is what keeps the accuracy steady on the messy documents where social inflation tends to hide.

Social inflation will keep setting the loss-cost ceiling on casualty through 2027. The market caps how far rate can go, while the accuracy of the data you select and reserve on stays in your own hands, which is where the next point of loss ratio will come from. If you would like to see this on your own casualty book, you can request a demo.