Business

Know the Business — MetLife, Inc. (MET)

Bottom line. MetLife is not really a life insurer in the way casual investors picture one — it is a $745B balance sheet with $29B of equity (about 26× leverage) and three economically distinct businesses bolted onto it: a high-margin group-benefits underwriter, a spread-and-scale U.S. retirement / pension-risk-transfer factory, and an Asia + LatAm growth engine with a $736B institutional asset manager (MIM) on the side. The market underprices the diversification (no segment is over ~30% of earnings) and overweights the consolidated 4% net margin and 1.8× book multiple, when the right way to value MET is adjusted ROE on book ex-AOCI, with an option-value tail on MIM if the New Frontier $1T-AUM target lands.

1. How This Business Actually Works

The economic engine: MetLife collects premiums and contract deposits today, parks the cash in a $600B+ general-account portfolio (mostly investment-grade fixed income, structured products, mortgages, private credit), and pockets the spread between portfolio yield and what it owes policyholders — net of claims, expenses, and capital charges. Three different mechanics sit on top of that float, and they behave very differently.

The two mechanics that drive the bulk of profit are investment spread (RIS, Asia, big chunks of LatAm) and underwriting margin (Group Benefits, supplemental health). Spread businesses behave like leveraged bond books; underwriting businesses behave like a P&C-style float-light franchise. Group Benefits reprices annually and absorbs mortality shocks within a year; RIS prices a 30-year liability today and lives with the asset/liability mix for decades. A 200 bp move in the 10-year Treasury hits these two segments in opposite quarters with different magnitudes.

The third leg — MIM — is a different animal entirely. It earns basis-point fees on $736B of Total AUM (post-PineBridge, closed Dec 30, 2025; Q1 2026 disclosure), is capital-light, and reaches profit through asset-management economics. Its $200M of FY2025 adjusted earnings is small but strategically disproportionate: at $1T scale, MIM could reach ~$400-500M of earnings — a stream the market should multiple-rate like an institutional credit manager (15-20×) rather than an insurance subsidiary (8-12×). That re-rating optionality is the cleanest equity story inside MET.

The cost structure is mostly fixed. Claims are pre-priced and slow-moving; the controllable lines are distribution and corporate / IT overhead, where management has guided to a direct expense ratio of 12.1% in 2026 and 11.3% by 2029. Bargaining power is uneven: jumbo employers and pension-plan sponsors squeeze pricing in Group and RIS; retail customers in Asia/LatAm have very little leverage; large distribution intermediaries (Aon, Mercer, big banks) extract real economics.

2. The Playing Field

The peer set splits cleanly into two business models that the market keeps trying to compare on the same scorecard. MET, PRU, LNC, and PFG are leveraged spread-plus-underwriting composites; AFL and UNM are capital-light pure underwriters. Margins, leverage, and book multiples are not directly comparable across the two camps — but ROE is.

The chart shows the iron rule of this industry: the multiple tracks ROE, with two notable dislocations. AFL pays the highest P/B because its 13% ROE comes from low-asset-leverage underwriting that doesn't scare credit-cycle investors. LNC trades below book despite a 12.3% headline ROE because the market doubts the durability of those earnings (legacy VA book, smaller capital cushion). MET sits in the middle: 12.9% GAAP ROE and 1.82× book implies the market is paying a small premium for diversification, but not pricing in management's 15-17% adjusted ROE target band — the gap between GAAP ROE (12.9%) and the Core Adjusted ROE target (15-17%, with FY25 actual 16.0%) is the bull/bear hinge.

MET is the only public peer that meaningfully combines all three economic models (group underwriting, spread/PRT, Asia growth, plus MIM fee income). That diversification has a cost — slower revenue growth than AFL Japan or PRU Japan — but a benefit — earnings stability that no single peer matches. PRU is the apples-to-apples comp on retirement and Asia, AFL is the right comp on group/supplemental, UNM defines what a pure group book looks like, and LNC tells you what happens when capital discipline slips.

3. Is This Business Cyclical?

It is, but not in the GDP-cyclical sense investors usually mean. MET's cycle is a rate cycle and a credit cycle, not an economic-output cycle — and when both go the wrong way at once, the damage is severe and slow to reverse.

Three episodes anchor the cyclical history. 2008-2009: the financial crisis took $4B of net income to a $2.4B loss in a single year as credit spreads widened, equity markets crashed (variable-annuity guarantees became huge liabilities), and net realized losses overwhelmed underwriting. P/B dropped to 0.5x; full GAAP earnings recovery took until 2011. 2016: zero-rate environment compressed spread to the point that GAAP NI fell to $747M and management ultimately spun off the variable-annuity book as Brighthouse in 2017. 2023: GAAP NI fell to $1.4B on actuarial assumption updates and derivative MTM noise — a non-cyclical accounting event but a useful reminder that long-tail liability assumptions are not certain.

The damage pattern is consistent: rates fall and credit widens, the bond book takes mark-to-market hits, new-money yields collapse below portfolio yield, variable-investment income misses plan, and reserves may need strengthening if assumptions on longevity/policyholder behavior prove wrong. MetLife's structural defense is now meaningful: 350%+ RBC, the 2017 Brighthouse spin removed the worst of the variable-annuity guarantees, the rate-sensitive RIS book sits inside an actively hedged ALM framework, and Asia/LatAm/EMEA/MIM together are explicitly described in the 10-K as buffers because they have "limited U.S. interest rate sensitivity."

What the chart will not show, but matters: share count fell from ~1.07B in 2014 to 655M at year-end 2025 — a 39% reduction. Even when book value per share looked flat or compressed, equity per share was still being concentrated. The book-value-ex-AOCI trend, not GAAP NI, is the right cycle scorecard.

4. The Metrics That Actually Matter

These five do almost all the work. Generic measures (revenue growth, gross margin, ROA, EBITDA) mislead in this industry — see the industry primer for why.

Why these and not the obvious ones. Adjusted ROE on book ex-AOCI strips the rate-driven AOCI noise that swings shareholder equity by billions without changing the cash-flow generating power of the business. RBC ratio is the single hardest constraint on capital return — drift below 350% and the cadence of buybacks slows; below 300% the dividend itself is at risk. FCF-to-AE conversion matters because adjusted earnings are non-cash in part (reserve releases, DAC amortization mechanics); the cash that actually reaches the holding company is what funds the $4.4B/year of capital return. Segment adjusted earnings mix quantifies diversification — the more even the mix, the more stable the through-cycle ROE.

5. What Is This Business Worth?

Value here is determined by adjusted earnings power × the multiple the market gives that earnings power, where the multiple is set by ROE durability, capital-return quality, and the mix of fee vs spread vs underwriting income. Pure SOTP doesn't quite fit — the segments share a common general-account, RBC budget, and brand — but a value-driver lens is essential because the segments deserve very different treatment.

The key valuation insight: applying a single P/E or P/B to MET as a whole prices the consolidated business between a spread-only insurer (8-10x) and a fee-heavy asset manager (15-20x). The market currently sits at 16.8x earnings / 1.82x book — pricing it closer to the diversified-financial average than the pure-spread average. What would justify a premium: durable adjusted ROE above 15%, MIM scaling toward $1T with healthy organic flows, sustained Group Benefits margin, and Asia continuing to grow at low double digits. What would justify a discount: a return to the 2008-2009 or 2015-2016 rate environment, RBC drift, a legacy-block reserve correction, or PE-platform competitors compressing PRT spreads.

The right way to underwrite MET as an investor: estimate a normalized adjusted ROE through a full cycle, multiply by book ex-AOCI, then add option value for MIM re-rating. Don't try to forecast quarterly GAAP EPS — too much derivative MTM and AOCI noise — and don't dwell on revenue growth, since most of "revenue" is just the gross flow of premiums and investment income.

6. What I'd Tell a Young Analyst

One. Watch the gap between GAAP ROE (12.9% in FY25) and the New Frontier 15-17% target band (FY25 Core Adjusted ROE was 16.0%). The wedge between the two metrics, not the headline P/E, is the entire equity story.

Two. MIM is the optionality. Small today ($200M of earnings) but on a path to ~$1T AUM. If MIM hits $400-500M of earnings and the market multiples that piece like an institutional asset manager, fair value lifts 10-15% before any insurance segment moves.

Three. Don't trust GAAP NI quarter-to-quarter. FY23's $1.4B GAAP figure looked terrible; adjusted earnings were $5.2B. Read the adjusted earnings reconciliation in the 10-K and the supplement; the buy-side and sell-side both work off the non-GAAP number.

Four. Track three rates at once: 10-year UST (drives RIS spreads), IG corporate spread (drives the bond book mark), and JGB / Japan curve (drives Asia segment). A 50 bp move in any of them shows up in the next earnings supplement before management commentary catches up.

Five. The thesis breaks if RBC drifts below 350% or if the Statement-Based Combined ratio takes a step-down move. That removes the per-share-EPS support that has carried the stock through choppy earnings (share count fell 39% over the last decade).