How to Analyse a Life Insurance Company (VNB, EV, P/EV)

Look at one line from SBI Life's March-quarter results and you will see why a life insurer is the one financial business you must never read off the profit line. In the three months to March 2026, the company's reported "operating profit" was negative ₹1,046 crore on the back of ₹4,071 crore of revenue, and the income statement showed an operating margin of −26% (Inve data, Q4 FY26). The same company, in the same year, grew the real measure of its franchise — the Value of New Business — 12% to ₹66.7 billion at a healthy 27.5% margin, and its embedded value rose 15% to ₹807.9 billion (SBI Life Q4 FY26 concall). One number screams crisis; the other says the business compounded nicely. The profit-and-loss account is the liar here, and the embedded value is the truth. (Illustration of how to read the numbers, not a view on the stock.)

That gap is not an accounting quirk you can ignore. It is the whole subject. A life insurer sells a promise that pays out over twenty or thirty years, but it books the cost of acquiring that policy — commission, medicals, setup — almost entirely upfront. So a quarter of fast growth can look like a loss, because you have paid to write business whose profit arrives across the next three decades. Read a life insurer on P/E and you will reward the ones writing the least new business and punish the ones writing the most. This is how to read one the way an actuary does: the handful of numbers that actually decide the outcome, where they hide (mostly not in the income statement), and the multiple that makes sense once you accept that ordinary profit doesn't.

A note on the boundary first. You will not re-run an insurer's actuarial model from the outside — embedded value rests on assumptions about mortality, lapses and investment returns that only the appointed actuary fully sees. What you can do is read the direction of the operating metrics and check whether management's account of them survives the next four concalls.

What does a life insurer actually sell, and how does it make money?

Picture a fruit-seller who, instead of selling apples for cash today, signs customers up to pay ₹100 a year for thirty years and hands over a basket of fruit only at the very end. On day one he is poorer — he has paid for the stall, the signboard and the salesman's commission — even though he has just locked in thirty years of payments. Judge him on this month's cash and he looks like he's failing. Judge him on the present value of all those future ₹100 instalments, minus what he'll pay out, and he's building real wealth. A life insurer is that fruit-seller, and embedded value is the honest scoreboard.

The economic engine has three parts, and only one of them is in the headlines. New business — the policies sold this year — is where growth and most of the value creation sits. The in-force book — every policy already on the books — throws off profit slowly as customers keep paying their renewals. And the investment float — the pool of premiums invested until claims fall due — earns a spread the insurer keeps. The first two are what you analyse; the third is why solvency and persistency matter so much. Growth that doesn't persist, or growth bought at a margin so thin it barely covers the upfront cost, is growth that destroys value while flattering the top line.

The metrics that matter — the spine

Forget the standard ratio toolkit. A life insurer is judged on a sector-specific stack of numbers, most of which never appear in the income statement — they live in the investor presentation and the concall. Here is each one, what "good" looks like, and a real number to anchor it.

APE — the right way to count sales

Annualised Premium Equivalent (APE) is the standard sales measure: regular premiums counted in full plus 10% of any single (lump-sum) premium. The 10% rule stops a company from inflating its growth by selling one giant one-time policy that will never recur. Gross premium includes renewals from old policies and so flatters a stagnant insurer; APE isolates the new franchise being built this year.

Where to find it: not the P&L — the investor presentation and the concall's opening remarks. SBI Life's FY26 APE was ₹242.7 billion, up 13% (SBI Life Q4 FY26 concall). ICICI Prudential's, by contrast, grew just 2.2% to ₹106.41 billion in FY26 against a punishing high base (ICICI Pru Q4 FY26 concall). "Good" is double-digit APE growth that holds across the cycle — but APE alone tells you nothing about whether that growth was profitable. For that you need the next number.

VNB and VNB margin — does the growth create value?

This is the single most important pair in the sector. Value of New Business (VNB) is the present value of all future profits from this year's new policies, calculated upfront. VNB margin is that value as a percentage of APE — essentially, how many paise of lifetime profit each rupee of new sales creates.

VNB margin is the insurer's gross margin, and it is set almost entirely by product mix (the next metric). SBI Life ran a 27.5% VNB margin in FY26; ICICI Prudential, 24.7%; HDFC Life, 24.4% in Q3 FY26 (each company's Q4/Q3 FY26 concall). A 3-point margin gap on similar APE is an enormous difference in value created. "Good" is a stable-to-rising margin; a falling margin while APE grows is the tell that the insurer is buying volume with cheaper, lower-margin product — value-dilutive growth dressed up as a good quarter.

Product mix — the hidden lever behind the margin

VNB margin doesn't move on its own; it moves because the product mix moves. Four buckets matter, in roughly ascending order of margin: ULIPs (unit-linked, market-risk passed to the customer — thin margin), par (participating, profits shared), non-par savings (guaranteed-return — fatter margin), and protection (pure term cover — fattest margin of all).

Watch HDFC Life to see the lever in action. As its ULIP mix rose to 37% (from 32% a year earlier) and its non-par savings mix fell from ~30% to 19%, its VNB margin compressed from 25.6% (FY25) to 24.4% (Q3 FY26) (HDFC Life Q3 FY25 and Q3 FY26 concalls). Nothing went wrong operationally — customers simply chose more ULIPs in a buoyant market, and ULIPs carry less margin. ICICI Prudential, the most linked-heavy of the four with ~48% of APE in ULIPs, structurally carries the lowest margin of the group for exactly this reason (ICICI Pru Q4 FY25 concall). When you see margin moving, your first question is always: which way did the mix shift, and was it the customer's choice or the company's push?

Persistency — does the franchise actually stick?

A policy sold and then lapsed is worse than a policy never sold, because the insurer paid the upfront cost and gets none of the renewal stream. Persistency measures how many policies are still being paid for after a given time. The two numbers to watch are 13th-month persistency (did they pay the second-year premium?) and 61st-month persistency (are they still around after five years?). The 13th-month is the leading indicator of sales quality; the 61st-month is the deep test of whether the book is genuinely sticky.

This is where a real warning lives. ICICI Prudential's 13th-month persistency fell from 89.1% (FY25) to 84.5% (FY26) — a near five-point drop (ICICI Pru Q4 FY25 and Q4 FY26 concalls). Management's own stated endeavour was to "hit 85% and above," and Inve's Promise Tracker marks that guidance missed (Inve data). SBI Life, over the same window, held 13th-month persistency at 87.9% (SBI Life Q4 FY26 concall). A falling 13th-month number while sales grow is one of the clearest amber lights in the sector: it usually means the recent growth was sold harder than it was sold well. "Good" is high-80s-and-rising for 13th-month, and a 61st-month grinding upward (HDFC Life's 61st-month improved to 64% in FY26 from the high-50s a couple of years earlier — HDFC Life Q4 FY26 concall).

AUM and solvency — the float and the cushion

Assets under management is the investment float — the premiums sitting invested until claims fall due. SBI Life crossed ₹4.9 trillion in FY26, up 9% (SBI Life Q4 FY26 concall). AUM growth is mostly a function of accumulated premiums and markets; it matters more as scale and brand evidence than as a value driver on its own.

Solvency ratio is the regulatory cushion — available capital over the required minimum. IRDAI's floor is 150%, and a healthy insurer sits comfortably above it: SBI Life at 1.90x (190%), HDFC Life at 177%, ICICI Prudential at a robust 227.3% in FY26 (each company's Q4 FY26 concall). Solvency is a survival metric, not a growth one — but a number drifting toward the floor while the book grows fast signals a capital raise ahead, and the timing of that raise (opportunistic vs forced) tells you about management foresight. HDFC Life's solvency drifted from 194% (FY25) to 177% (FY26) and it announced a ₹1,000 crore preferential capital raise (HDFC Life Q4 FY26 concall) — the cushion being topped up before it got thin.

Embedded value — the scoreboard the P&L hides

Embedded value (EV) is the present value of the in-force book plus the company's net worth — the actuary's estimate of what the franchise is worth today if it stopped writing new business and simply ran off everything already sold. It is the single number that makes a life insurer legible, and it is the foundation of valuation. SBI Life's Indian EV reached ₹807.9 billion in FY26 (+15%); HDFC Life's, ₹62,139 crore (HDFC Life Q4 FY26 concall); ICICI Prudential's, ₹529.89 billion (+10.5%); MFSL's (Axis Max Life's parent), ₹28,110 crore in Q3 FY26 (each company's concall).

The growth in EV, scrubbed of market noise, is the operating return on embedded value (RoEV) — the insurer's true return on its own franchise. SBI Life and HDFC Life run RoEV in the 15–17% range (HDFC Life's operating RoEV was 16.7% in FY25 — HDFC Life Q4 FY25 concall); ICICI Prudential reported FY26 RoEV of 11.9% (ICICI Pru Q3 FY26 concall). That spread — high-teens versus low-double-digits — is the quiet difference between a franchise compounding hard and one merely treading water.

How do you value a life insurer — and why not P/E?

You already have the answer from the first paragraph: the P&L understates a growing insurer, so a P/E built on that profit is meaningless. A fast-growing insurer with a heavy upfront-cost drag can look "expensive" on P/E precisely because it is growing well. The right anchor is the embedded value.

The sector's native multiple is Price to Embedded Value (P/EV) — market cap divided by EV. It asks the only sensible question: how much am I paying for the franchise the actuary says exists today? A P/EV of 1.0x means you're paying for the in-force book and net worth with nothing for future growth; pay above 1.0x and you are buying the value of all the new business the company will write in years to come.

The richer version is the appraisal value: embedded value (what exists) plus a multiple of VNB (the value of future new business, capitalised). This is why VNB margin and APE growth feed directly into the valuation — a company growing APE at 13% with a 27.5% margin and 16% RoEV deserves a higher P/EV than one growing APE at 2% with a 24.7% margin and 12% RoEV, because its future-new-business engine is worth more. So the two numbers move together: a high P/EV is only justified by a high, durable RoEV. Pay 3x EV for a 12% RoEV book and you are paying a growth price for a treading-water franchise — the most common way investors overpay in this sector.

Read EV with one eye open, though. It is assumption-dependent — change the long-term investment-return or mortality assumption and EV moves. Reputable insurers publish the sensitivities; read them, and treat a company that buries or frequently revises its assumptions with suspicion.

A worked case: when "range-bound" wasn't

Take HDFC Life through FY26 — a well-run insurer, which is exactly why it teaches better than a disaster would. At the Q4 FY25 call, management guided that VNB margins "will be more or less range bound" for FY26, with VNB growth tracking topline growth (HDFC Life Q4 FY25 concall). It was a reasonable, confident guide from a credible team.

Watch what the mix did to it. Through FY26, customers leaned into ULIPs (mix up to 37%) and away from higher-margin non-par savings (down from ~30% to 19%) (HDFC Life Q3 FY25 and Q1 FY26 concalls). The VNB margin was not range-bound: it slipped to 24.4% in Q3 FY26 and 24.2% for FY26 including the new GST and surrender-value regulatory drag, against 25.6% the year before (HDFC Life Q3 and Q4 FY26 concalls). In Inve's Promise Tracker, the "range-bound margin" guidance is marked missed, and the headline ambition management had repeated for years — double VNB every 4–4.5 years (a ~16–17% CAGR) — was quietly revised down to a "normalised three-year VNB CAGR of around 9–10%" by Q4 FY26 (Inve data; HDFC Life Q4 FY26 concall: "on a normalized three-year VNB CAGR would be around 9%, 10%.").

Notice what this is and isn't. It is not evidence of bad management — HDFC Life was candid, kept compounding EV at ~16% RoEV, and absorbed a real regulatory shock. It is evidence that the confident margin guide given twelve months earlier deserved less weight than the product-mix trend sitting underneath it. The margin didn't fall because the company stumbled; it fell because mix is the master variable and the customer, not the CFO, sets it. A sequence of guidance — confident margin guide, then a quiet halving of the VNB-doubling ambition — is the tell no single call gives you. (Illustration, not a view on the stock — and a read on how management communicated through one cycle, not a lifetime verdict.)

This is the pattern Inve's Promise Tracker is built to surface: every forward commitment pinned to the quarter it was made, with a verdict as later calls come in.

Red flags specific to life insurers

• APE growing while VNB margin falls. Volume bought with cheap product. The growth is real; the value isn't. Always read the two together.
• 13th-month persistency slipping while sales accelerate. The fastest-sold business is often the worst-quality. A near-5-point persistency drop (as ICICI Prudential saw into FY26) deserves more attention than a strong APE number does.
• Margin "explained away" by mix every quarter. A one-off mix shift is fine; mix moving the same way for six straight quarters while management calls margins "range bound" is the slow leak the EV growth eventually exposes.
• Solvency drifting toward 150% with no capital plan named. The cushion thinning under fast growth means dilution is coming; the question is whether management raises early or is forced to raise late.
• EV assumptions that keep getting friendlier. If investment-return or persistency assumptions are quietly nudged up to flatter EV, the scoreboard itself is being doctored. Read the published sensitivities.
• Single-channel dependence. A bank-promoted insurer doing the bulk of its APE through one bancassurance counter is one distribution-agreement renegotiation away from a growth cliff — watch the channel mix, not just the total.

Frequently asked questions

A repeatable workflow falls out of all this. Anchor on VNB and VNB margin across six to eight quarters, read against the product mix that drives them. Check the persistency trend — 13th-month for sales quality, 61st-month for durability. Confirm the solvency cushion and watch for a looming raise. Then read embedded value and its operating RoEV as the scoreboard, and value the whole thing on P/EV justified by that RoEV, never on P/E. Inve's KPI Screener lines these operating metrics up across insurers with a trend and a data-confidence flag per number, and the concall summaries pull every forward commitment into one guidance table per quarter. If lenders are your other beat, the sibling logic for reading a spread-and-credit business sits in how to analyse an NBFC.

So invert the question you bring to an insurer's results. Don't ask "was this a good quarter for sales?" Ask: if this management were quietly buying growth with cheap, low-margin product that won't persist, what would the numbers look like — and does this book rule that out? Rising APE on a falling VNB margin with slipping persistency does not rule it out; it is the pattern itself.

And the owner's question, the one to sit with before buying a single share of any life insurer: what must I believe about the next twenty years of renewals — not this quarter's APE — for the embedded value I'm paying a premium over to actually be there? If the honest answer leans on a buoyant ULIP market and a confident margin guide rather than on persistency and mix, you have read the headline, not the franchise.

Related reading

• How to Analyse a Bank Stock (NIM, CASA, GNPA, ROA)
• How to Analyse a Building Products Company (Pipes, Tiles)
• How to Analyse a Cables and Wires Stock