AIG (AIG)
The P&C turnaround case: an FY2025 General Insurance combined ratio of 90.1% and core operating ROE of 11.1% versus 7.5% GAAP.
Educational analysis for professional use. This profile is not investment advice or a recommendation to buy or sell any security, and it is not personalised.
Snapshot
Business Overview
AIG has spent five years shrinking back to a property-and-casualty insurer, and in May 2026 it sold the last of its stake in Corebridge Financial, the life and retirement business it began separating in 2021. What is left is General Insurance, run through North America Commercial, International Commercial and Global Personal, plus a small Other Operations. The underwriting is closing on the peer pack; the market's open question is which return that improvement settles at. Market capitalisation was roughly $40.5 billion as of 10 June 2026, the smallest of the four P&C names on this site.
FY2025 General Insurance combined ratio was 90.1%, still underwriting profit (below 100%) and below the preliminary US industry prints of 92.2% (AM Best) and 92.9% (Verisk), but above Chubb at 85.7%, Progressive at 87.4%, and Travelers at 89.9%. GAAP ROE was 7.5%; core operating ROE was 11.1% in the same release. Adjusted book value per share excluding AOCI was $78.02 at year-end 2025.
Which return you believe decides what the stock is worth. Insurers earning 8 to 12% on book tend to trade around 0.9 to 1.3× it; AIG's 7.5% GAAP return sits under that band, its 11.1% core operating return inside it and above a 10% cost of equity. Which of the two the market capitalises is the whole question, not whether this year's underwriting justifies book.
How the Numbers Read
The combined ratio to use is General Insurance, which is where the underwriting sits; Other Operations carries corporate costs that no underwriting ratio should absorb. At 90.1%, General Insurance earned 9.9 cents of underwriting profit per premium dollar before investment income. That is the weakest of the four P&C names covered here, and the closest to the industry average, while still clearing the below-95% strong-underwriting mark.
Prior-year reserve development, or PYD, was favourable $548 million net of reinsurance in FY2025, equal to 2.1 points favourable on the GI combined ratio. That is real help to the reported year, and it says nothing about how 2025's own business was priced: see accident year versus calendar year. The reserve development guide notes industry adverse PYD of $7.8 billion on US liability lines in calendar 2024 after a long favourable streak; AIG's release does not prove current-year adequacy.
Investment income on the GI portfolio added FY2025 net investment income of $4,215 million on a GAAP basis ($3,778 million on an APTI basis). Book yield on the General Insurance portfolio was 4.43% for FY2025, and 4.59% annualised in the fourth quarter. Read the small print before ranking that against anyone: AIG's figure covers its bonds and loans, not every asset it owns, which is a narrower base than Travelers' 3.1% after-tax whole-portfolio yield and a different one again from Chubb's 5.1% on fixed income.
GAAP ROE of 7.5% against core operating ROE of 11.1% means one-off charges, investment marks and the running-down corporate operations still sit between the underwriting result and the reported return. The P/BV vs ROE guide shows how far the multiple moves with the sustainable return; the open question here is whether 11.1% core operating is AIG's real number.
Valuation Framework
A GAAP ROE below 8% fits the typical screening band for sub-8% sustainable returns. Core operating ROE of 11.1% would map to roughly 0.9 to 1.3× P/BV in the 8 to 12% band if the market believed it was durable. The stock prices the GAAP return, not the core operating one; how much of that gap closes depends on how far the restructuring still has to run.
Steady-state P/BV ≈ ROE ÷ 10.0% COE implies about 0.75× on 7.5% GAAP ROE and about 1.11× on 11.1% core operating ROE. Until GAAP ROE moves toward core operating, the multiple should stay closer to the lower implied line than the upper one.
What to Watch in the Financials
Core operating ROE versus GAAP. Sustained convergence toward 11%+ core without GAAP catching up would keep the reported multiple depressed even as operating earnings improve.
GI combined ratio excluding PYD. 2.1 points of favourable development flattered FY2025. Accident-year combined ratio near 92% would read differently from a reported 90.1%.
What the 4.43% covers. It is the yield on bonds and loans held in General Insurance, so it is not comparable with a life insurer's whole-portfolio figure without adjustment.
Capital return versus book build. A buyback struck below book value accretes per-share adjusted BVPS if underwriting holds; it does not fix a combined ratio drifting toward industry average.
Key Risks
Restructuring stumbles. Legacy reserves, reinsurance programmes, or expense saves that slip would hit core operating ROE before GAAP ROE, narrowing the bull case on the turnaround.
Industry loss inflation. GI combined ratio at 90.1% has less cushion than Chubb or Progressive if commercial and personal lines reprice slower than severity trends.
GAAP ROE drag persists. If notables and non-operating items keep GAAP ROE in single digits while peers report mid-teens returns, the multiple may stay depressed even with improving GI metrics.
Multiple trap. A rerating requires the market to accept 11.1% core operating as sustainable. If the GAAP-to-core gap does not close, improving GI metrics may not lift the multiple.
AIG's price rests on which return the market believes, last year's or the turnaround's. The primer discounts a turnaround to intrinsic value per share.
The Excel model is the primer's three residual-income valuations live across 13 sheets: change the combined ratio, the spread or the cost of equity and the justified P/BV moves.