Statutory P/E is broken, as the accounting section showed, so the lens is price to embedded value:
P/EV = market capitalization ÷ embedded value
How do you calculate and read P/EV?
Read it from first principles. EV already contains net worth plus the runoff value of the existing book. Pay 1.0× EV and you get the in-force machine at actuarial fair value, with the future franchise thrown in free. Pay more, and the premium is the price of future new business, the profitable sales not yet made. So decompose it:
Implied new-business multiple = (Market cap − EV) ÷ VNB, or "how many years of current-scale VNB am I pre-paying?" Growth means the multiple isn't literally years. Still, it's the cleanest cross-company gauge of how much franchise optimism is in the price.
Worked example (HDFC Life-style; illustrative, modelled on ~FY24 scale; pull the actual EV report and current price). EV ≈ ₹47,000 cr. VNB ≈ ₹3,500 cr on APE ≈ ₹13,000 cr, a margin of ~27%. Market cap ≈ ₹1,40,000 cr.
P/EV = 140,000 ÷ 47,000 ≈ 3.0×Implied NB multiple = (140,000 − 47,000) ÷ 3,500 ≈ 26.6×
The market is paying EV plus ~27 years of today's VNB. Is that demanding? Sanity-check it the Phase 3 way, by asking what must be true. VNB has to compound in the mid-teens for a decade-plus with stable margins. RoEV has to hold in the high teens. And there can't be a mis-selling or regulatory shock.
Advertisement
What explains a P/EV of 3.0× against 0.85×?
Now look at LIC (illustrative, ~FY24 scale). EV ≈ ₹7.2 lakh cr and market cap ≈ ₹6.1 lakh cr, so P/EV ≈ 0.85×, a discount to runoff value. Investors are pointing at a low VNB margin (~mid-teens), a par-heavy mix, single-digit RoEV on a giant EV, and the overhang of state ownership. They doubt the EV's assumptions and assign the franchise no value, or a negative one.
Between 3.0× and 0.85× lies the whole grammar of the sector. P/EV is to RoEV and VNB growth what P/B is to ROE. It's the same excess-return logic you used on banks, one storey up. Here's a quick coherence test you can always run. If RoEV ≈ cost of equity and new business adds nothing, fair P/EV ≈ 1×. Every tenth of a turn above that has to be paid for, either by RoEV above the cost of equity or by a valuable VNB franchise.
How do you value life insurers outside India?
US/global note. EV reporting is a European and Asian tradition. US life insurers (MetLife, Prudential Financial, Lincoln) rarely publish EV, so US analysts value them on price to book value ex-AOCI vs ROE, plus adjusted operating EPS. Ex-AOCI strips the bond mark-to-market noise out of equity. A MetLife-style profile of ~1.2–1.4× book ex-AOCI against a ~13% ROE (illustrative, mid-2020s scale) is the same excess-return read in GAAP clothing. Asian listcos under IFRS 17, such as AIA (historically ~1.2–1.8× EV), increasingly point analysts to the CSM balance and new-business CSM. Those behave like VIF and VNB respectively. Economically they're identical, and only the wrapper changes.
So when you compare an Indian insurer at 2.5× EV with a US peer at 1.2× book, remember what you're comparing. Their growth runways are different. Their metrics are different too, because EV already capitalises future in-force profit and book value does not. And their guarantee books are different. Never read the gap as automatic mispricing.