MetLife (MET)

Business Overview

MetLife earns its money a different way from the P&C insurers on this site, and the tools that work on them do not work here. Group benefits, retirement and life insurance take in policyholder money, invest it, and credit the policyholder a lower rate than the portfolio earns. The gap between the two is the spread, and it does more work than any underwriting ratio. There is no combined ratio on this name, and inventing one would be a category error. Market capitalisation was roughly $57.3 billion as of 10 June 2026.

Adjusted book value per share was $57.07 excluding AOCI at year-end 2025. The multiple prices the 16.0% adjusted return on equity more than the 12.9% GAAP one.

How the Numbers Read

General-account spread is portfolio yield minus credited rate on policyholder liabilities. At 115 bps FY2025, MetLife earned 1.15 percentage points of margin on the general-account block before other expense loads. The life spread guide sets this against P&C economics. There is no loss ratio and no expense ratio to add up, but the same rate exposure runs underneath: the assets reprice on one schedule and the crediting rates on another, and the spread is whatever falls out between them.

Net investment income scale reflects the general-account asset base. FY2025 GAAP net investment income was $22,559 million; adjusted net investment income was $21,432 million. The 4.91% is a yield on the whole portfolio; MetLife's own supplement carries a second, slightly lower yield on a narrower definition. Prudential's comparable figure, 4.32% after investment expenses, is closer to the latter. None of these is an interchangeable "book yield", and a 60 basis point gap between two life insurers is as likely to be the definition as the portfolio.

Once a year the actuaries revisit what they have assumed about how long policyholders live, lapse and claim, and the revision lands in earnings. That annual assumption review cost MetLife $89 million in FY2025, all of it in the third quarter, inside a net $43 million of one-off items for the year. Do not read it as the life-company version of a reserve release. The reserve development guide deals with P&C prior-year development, where Progressive released $1,394 million against earlier accident years; an assumption update is a change of view about the future, not a settled verdict on the past.

MetLife publishes no value of new business, the present value of profits on policies written this year that European insurers report. The nearest read on volume is premiums and fees, up 5% to $49.8 billion once pension risk transfer deals are excluded. Those deals are lumpy by nature, which is why they come out.

The float idea still applies, even though nobody uses the word here: MetLife holds policyholder money for decades before paying it out, and what it earns on that money is most of the return. The float guide works through the mechanism. Spread and business mix do the bulk of the ROE work, with capital management the residual, and the gap between 12.9% GAAP and 16.0% adjusted is what the one-off items did to the headline.

Valuation Framework

Against a 12.9% GAAP ROE, MetLife sits in the typical 12 to 16% ROE band (roughly 1.3 to 1.8× P/BV). On 16.0% adjusted ROE, steady-state P/BV ≈ ROE ÷ 10.0% COE implies about 1.60×. A lower GAAP return maps to a lower multiple, which is what the P/BV vs ROE identity says it should.

The bear case is spread compression, not one year's assumption charge. Start any sensitivity from the two filed numbers: 115 basis points of spread earned on a portfolio yielding 4.91%. Credited rates rising faster than the portfolio reprices is what takes the 115 down, and there is no combined ratio to warn you it is happening.

What to Watch in the Financials

General-account spread in bps. MetLife files 115 bps; track quarter-to-quarter direction against credited-rate moves and portfolio reinvestment.

GAAP versus adjusted ROE. 12.9% GAAP against 16.0% adjusted ex-notables: know which denominator the market is implicitly capitalising.

The annual assumption review. $89 million in FY2025 is small. The line is worth watching because it is lumpy and because it is the one place a life insurer restates what it thinks its liabilities are worth.

Premiums and fees, up 5% to $49.8 billion. This is the volume read in the absence of a new-business value. Pension risk transfer deals can swing the headline without changing spread economics at all, which is why the growth figure excludes them.

Which yield you are quoting. The 4.91% is the whole-portfolio measure from the earnings tables. MetLife's supplement carries a lower one on a narrower definition, so name the basis before comparing it with anything.

Key Risks

Spread compression. If credited rates on legacy blocks reprice up faster than the general-account portfolio yield, 115 bps narrows without any "combined ratio" to flag it.

Assumption-review volatility. These charges are lumpy and they arrive once a year. $89 million was modest, and a heavier one would land on adjusted earnings rather than on the spread.

Rate moves on the asset side. Unrealised gains and losses on the bond book run through AOCI, which can widen the gap between GAAP and adjusted returns even when the spread itself has not moved.

Multiple tied to adjusted returns. The multiple prices mid-teens adjusted ROE. If GAAP ROE near 13% is the market's sustainable view, the multiple needs the adjusted return to be the durable one.