Life insurance KPIs

Everything you need is in six KPIs. Compute each one, and never read any of them alone.

KPI 1. APE (Annualised Premium Equivalent): the volume metric. New-business premium comes in two flavours. Regular premium is paid annually for years, and single premium is paid in one shot. Add them raw and one ₹10 lakh single-premium policy looks like ten ₹1 lakh annual policies, even though the ten annual policies are worth far more over their lives. So the industry uses a convention:

APE = 100% of annualized first-year regular premium + 10% of single premium

Mini-computation. Say an insurer writes ₹2,000 cr of annualised regular new-business premium and ₹3,000 cr of single premium. Then APE = 2,000 + 0.10 × 3,000 = ₹2,300 cr. Track APE growth, where mid-teens is healthy for Indian private insurers in the current decade (check against industry data). But always track it with mix. APE growth bought with ULIPs is worth a fraction of the same growth in protection.

What are VNB and VNB margin?

KPI 2. VNB and VNB margin: the profit on this year's sales. Value of New Business is the present value, at the point of sale, of all future after-tax profits expected from the policies written this year, net of the cost of holding capital against them. Think of it as the life insurer's true "profit on new sales". It's the number the statutory accounts hide.

VNB margin = VNB ÷ APE. For an Indian private life insurer, 25 to 30%+ is strong (the brief's threshold). High-teens is middling, and low-teens tells you the mix is ULIP/par-heavy.

Mini-computation (mix is destiny). Using the ladder's midpoint margins on the ₹2,300 cr APE above:

Now let a bull market tilt the same ₹2,300 cr of APE toward ULIPs. ULIP rises to ₹1,200 cr at the expense of non-par, down to ₹300 cr, and protection, down to ₹250 cr, with par at ₹350 cr and annuity at ₹200 cr. Then VNB = 250×0.65 + 300×0.35 + 350×0.15 + 1,200×0.10 + 200×0.25 = 162.5 + 105 + 52.5 + 120 + 50 = ₹490 cr, margin 21.3%. It's the same "growth", and a fifth less value created. You've just seen the brief's first red flag in numbers, the VNB margin falling on a ULIP tilt. Headline APE is up and the economics are down.

What is embedded value and how is it built?

KPI 3. Embedded Value (EV): what the company is worth if it never sold another policy. It has two parts:

EV = Adjusted Net Worth (ANW) + Value of In-Force (VIF)

• ANW = shareholders' net worth marked to market (free surplus + required capital).
• VIF = present value of future after-tax profits expected from all policies already on the books, minus deductions for the frictional cost of holding required capital, the cost of residual non-hedgeable risks, and the time value of options and guarantees embedded in the policies. Indian insurers report IEV (Indian Embedded Value) under actuarial standards, in the APS 10 framework. An external actuary such as Milliman typically reviews it. Check each company's EV report.

EV is a runoff value: net worth plus the discounted profit stream of the existing book. It deliberately leaves out the value of policies not yet sold, which is the franchise. That exclusion is what makes P/EV meaningful, as you'll see in the valuation section. Whatever the market pays above EV is the price of the future franchise.

Worked simplified EV build (synthetic "Sarayu Life"). ANW = ₹3,000 cr. The actuary projects the in-force book's distributable after-tax profits at ₹800 cr, ₹760, ₹720, ₹680, ₹640 over years 1 to 5. After that they decline 5% a year forever as the book runs off. Use a risk discount rate (RDR) of 12%.

• PV of years 1–5: 800/1.12 + 760/1.12² + 720/1.12³ + 680/1.12⁴ + 640/1.12⁵ = 714 + 606 + 513 + 432 + 363 = ₹2,628 cr.
• Runoff tail from year 6: a declining perpetuity, value at end-year 5 = CF₆ ÷ (RDR − g) = (640 × 0.95) ÷ (0.12 − (−0.05)) = 608 ÷ 0.17 = ₹3,576 cr; PV = 3,576 ÷ 1.12⁵ = ₹2,029 cr.
• Gross VIF = 2,628 + 2,029 = ₹4,657 cr; less cost-of-capital and non-hedgeable-risk deductions of ₹657 → VIF = ₹4,000 cr.
• EV = 3,000 + 4,000 = ₹7,000 cr.

You'll never compute a real EV, because that takes an actuarial model of every policy. But the build shows you what the disclosed number is: a DCF of the existing book. And once you know that, you know what can bend it. A different discount rate bends it. So do different persistency and mortality assumptions, and so do different expense assumptions. An insurer that "improves" EV by softening assumptions shows up in the roll-forward as positive "assumption changes". Read that line every year.

What is RoEV and how is it calculated?

KPI 4. RoEV: the compounding rate of the machine. Each year EV rolls forward:

Closing EV = Opening EV + unwind (expected return) + VNB + operating variances & assumption changes ± economic variances − net capital returned

• Unwind: the VIF is a discounted number, so one year passing un-discounts it by roughly the RDR, plus the expected return on ANW. It's mechanical and takes no skill.
• VNB: new value added by this year's sales, the growth engine.
• Operating variances: reality vs assumptions on persistency, mortality, expenses. Persistent positive operating variance means conservative actuaries, which is a quality marker. Persistent negative variance means reality keeps coming in worse than assumed.
• Economic variances: markets and interest rates moved. That isn't management skill, so leave it out of the operating read.

EV operating profit = unwind + VNB + operating variances/assumption changes and RoEV = EV operating profit ÷ opening EV. For a good Indian private life insurer, a mid-to-high-teens RoEV is healthy. Note that RoEV mechanically ≈ RDR-driven unwind + VNB/EV, so a young fast-grower posts a higher RoEV than a mature book. Compare like with like.

Mini-computation (Sarayu Life, next year). Unwind: 12% × 4,000 (VIF) + 7.5% × 3,000 (ANW) = 480 + 225 = 705. VNB = 700. Operating variance is +45 because persistency beat its assumption, and an assumption change of −50 trues up the expense assumption. EV operating profit = 705 + 700 + 45 − 50 = ₹1,400 cr → RoEV = 1,400 ÷ 7,000 = 20%. Equity markets fell, so the economic variance is −100, and the dividend paid is −150. Closing EV = 7,000 + 1,400 − 100 − 150 = ₹8,150 cr.

How do persistency and the solvency ratio work?

KPI 5. Persistency: the mis-selling detector. Persistency measures whether customers keep paying. For last year's cohort, 13th-month persistency = premium renewed at month 13 ÷ premium due at month 13. It's also tracked at months 25, 37, 49, 61. You'll see it both by premium value and by policy count. Premium-weighted is standard, but a large gap between the two means small customers are lapsing en masse.

Thresholds. A 13th-month figure above 85% is healthy, and the best Indian franchises run 85 to 90%+. At the 61st month, >50% is respectable and ~60% excellent. Why is persistency the mis-selling detector? A customer who understood what they bought keeps paying. A customer who was sold a 20-year product as a 3-year deposit lapses at the first renewal. Falling 13th-month persistency indicts the distribution engine now. A 37th-month cliff marks the cohort mis-sold three years ago, the classic result when agents pushed a "pay only 3 years" framing.

Lapses also destroy the VNB you booked, because VNB assumed those renewals. When persistency comes in worse than assumed, you get a negative operating variance, and RoEV erodes. One straight line runs from the sales incentive to the lapse to the EV write-down.

KPI 6. Solvency ratio: the regulator's tripwire. Solvency ratio = Available Solvency Margin (ASM) ÷ Required Solvency Margin (RSM). Its numerator is assets over liabilities available as a buffer, and its denominator is the capital the formula requires. IRDAI's control level is 150% (1.5×) for life and general insurers alike. That's current as of mid-2026, so verify it at irdai.gov.in. IRDAI has been consulting on a move from this factor-based regime toward risk-based capital, alongside the Ind AS 117 transition. Healthy listed insurers run 180 to 220%+. Mini-computation: ASM ₹9,000 cr, RSM ₹4,800 cr → 187.5%.

It's the buffer that funds growth. Writing guarantee-heavy business consumes RSM, so a fast-growing non-par book with flat net worth drifts down toward 150. That drift is the brief's fourth red flag. An insurer at 155% must retain earnings, raise capital, or slow down. It has lost degrees of freedom.

Why does product mix matter?

KPI 7 (the modifier): product mix. Mix is the lens you hold over all six. Protection and non-par lift margin. ULIP and par dilute it. Annuities add duration and longevity risk. Trace any KPI move to mix before you credit or blame management skill.