Marsh & McLennan shares fall on AI fears, but the threat may be overblown
Marsh & McLennan and other major insurance brokers fell 8% to 11% in February after OpenAI released an insurance-focused app designed by Insurify. The market read the move as a warning sign that ChatGPT and similar tools could disrupt the brokerage business. But the decline has created a buying opportunity for investors who understand what Marsh actually does.
Marsh's stock has tumbled roughly 28% since peaking in the first half of 2025. Much of that drop reflects the insurance market's shift from a profitable "hard" cycle-where prices rose between 2017 and early 2024-to a softer one where rates have flattened and fallen. The company's valuation has compressed accordingly. Shares now trade at 16.8 times forward earnings, down from 22 times a year ago, according to UBS.
What Marsh actually does
Marsh Risk, the company's largest division, generates $14.4 billion in annual revenue by assembling insurance pools for complex risks. Large projects-say, a clean hydrogen facility-often require multiple insurers and reinsurers to spread exposure. Marsh brokers negotiate these deals and manage claims afterward, a time-consuming process most insurers won't handle in-house.
This brokerage work requires human judgment. An AI tool can streamline paperwork, but navigating competing interests among multiple stakeholders and structuring deals demands the kind of relationship management and problem-solving that remains difficult to automate.
The insurance arm operated at a 32% adjusted margin last year. That efficiency translates to strong cash generation: the division produced $5.3 billion in operating cash flow from $7.3 billion in total adjusted operating income.
The consulting risk is real-and being managed
Marsh's consulting arm-which includes Oliver Wyman and Mercer-faces genuine pressure from AI for Insurance and other automation tools. But management is not ignoring the threat.
Oliver Wyman's AI Quotient service has advised on more than $50 billion in AI deployments. The unit grew revenue 6% in the first quarter, outpacing the group's 4% top-line growth. Internally, Marsh has targeted $400 million in cost savings over three years through AI-powered document processing and other tools, which have already saved an estimated one million staff hours.
UBS estimates these efficiency gains could push Marsh's return on invested capital to near 30% by 2030, up from 25% today.
Cash flow makes the case
The stock's valuation looks reasonable when measured against cash generation rather than earnings alone. Marsh trades at a forward free cash-flow yield of 6.2% for 2026, 6.7% for 2027, and 8.1% for 2030.
Management has authorized a $6 billion share buyback and deployed $750 million in the first quarter, buying stock while the market sold. The company converts most operating cash into free cash flow because capital requirements are low-its main assets are people and technology, both of which it funds through the income statement.
For insurance professionals evaluating the sector, Marsh's valuation now reflects genuine cycle headwinds rather than structural obsolescence. The company's core brokerage business remains difficult to disrupt, and management is actively deploying AI to strengthen its position rather than defend against it.
Your membership also unlocks:
