07.2 · Insurers, AMCs, exchanges & brokers¶

Why this matters: India's listed financial sector is much more than banks. Life insurers sell forty-year promises and report profits that bear little relation to the value they create in a year; asset managers earn fees on money they do not own; exchanges and depositories are regulated near-monopolies whose profits swing with retail activity and with the regulator's mood. Each has its own metrics and its own valuation logic, and each is mis-analysed when forced into a P/E or a DCF template.

Learning objectives — after this lesson you can:

Read a life insurer through embedded value, value of new business, VNB margin, persistency and product mix, and value it on P/EV.
Read a general insurer through combined ratio, float and reserving.
Analyse an asset manager through AUM mix, yields, flows, TER regulation and operating leverage.
Analyse exchanges, depositories, brokers and RTAs through volumes, take rates and regulatory risk, with the 2024–25 SEBI derivatives measures as the live example.
Choose the right valuation approach for each.

Prerequisites: 07.1 Banks & NBFCs, 06.7 Other valuation methods · Time: ~100 min

1. Life insurance¶

1.1 Why accounting profit is the wrong lens¶

A life insurer selling a 20-year policy incurs most of its costs (agent commission, underwriting, acquisition) in year one and collects premiums and earns investment income for two decades. Under Indian accounting the first-year strain depresses profit when the business is growing fastest; a shrinking insurer can show rising profit as old policies mature. Statutory profit therefore tells you almost nothing about value creation. The industry's answer is embedded value accounting, disclosed by all listed Indian life insurers (verify the current status of Ind AS 117 / IFRS 17 adoption for insurers, which IRDAI has been phasing in — as of 2026 embedded value remains the primary valuation disclosure).

1.2 The metrics¶

Metric	Definition	What it tells you
Embedded value (EV)	Adjusted net worth + value of in-force business (PV of future profits from policies already sold, after cost of capital)	The economic book value; what the existing book is worth if no new policy were ever sold
Value of new business (VNB)	PV of expected future profits from policies sold this year	The value created by a year's selling — the insurer's "earnings"
VNB margin	VNB ÷ annualised premium equivalent (APE = 100% of regular premium + 10% of single premium)	Profitability of new sales; driven by product mix and expense efficiency (FY26 VNB margins: SBI Life 27.5%, ICICI Prudential 24.7%, HDFC Life 24.2%, LIC 21.2% — Business Standard, Jun-2026; verify current disclosures)
APE growth	Growth in new-business volume	The volume driver
Persistency (13th, 25th, 37th, 61st month)	Share of policies still in force after 1, 2, 3, 5 years	Quality of sales; lapses destroy VNB that was already booked
Product mix	ULIP (unit-linked; low margin, market-sensitive) / participating / non-participating savings (high margin, interest-rate risk) / protection (highest margin) / annuities	Margin and risk profile
Operating RoEV	(EV growth from operations) ÷ opening EV	The economic return on the franchise, usually mid-to-high teens for good insurers
Solvency ratio	Available solvency margin ÷ required; IRDAI minimum 150%	Capital adequacy
Channel mix	Bancassurance / agency / direct / online	Distribution cost and dependence (bank-owned insurers rely on the parent bank)

1.3 Valuation: P/EV and appraisal value¶

\[\text{Appraisal value} = \text{EV} + \text{Structural value} = \text{EV} + \text{VNB} \times \text{multiple}\]

The multiple on VNB is a growing-perpetuity of future years' new business: $\text{VNB}_1 / (k_e − g)$ with a fade. In practice analysts quote P/EV (market cap ÷ embedded value): historically 2–4x for Indian private insurers in growth phases, lower for LIC and for slower growers (verify current levels). The justified P/EV is

\[\frac{P}{EV} = 1 + \frac{\text{VNB}_1 / \text{EV}}{k_e − g} \;(\text{with fade})\]

Worked example — Vindhya Life (fictional). EV ₹20,000 Cr; VNB ₹1,600 Cr (VNB/EV 8%); operating RoEV 17%; $k_e$ 12%; VNB growth 12% for ten years fading to 5%.

ev, vnb, ke = 20000.0, 1600.0, 0.12
growth = [0.12] * 5 + [0.10, 0.08, 0.07, 0.06, 0.05]
pv, v = 0.0, vnb
for t, g in enumerate(growth, 1):
    v *= (1 + g); pv += v / (1 + ke) ** t
tv = v * 1.05 / (ke - 0.05); pv += tv / (1 + ke) ** 10
appraisal = ev + pv
print(round(pv), round(appraisal), round(appraisal / ev, 2))
# structural value ≈ 34,500; appraisal ≈ 54,500; P/EV ≈ 2.7x

A P/EV of 2.7x is what 8% VNB/EV and a decade of 12%→5% growth justify at a 12% cost of equity. At 3.5x the market is assuming faster or longer growth, higher margins, or a lower $k_e$ — the reverse-DCF logic of 06.6 applies directly. Sensitivities that matter: persistency (a 5-pp fall in 13th-month persistency can cut VNB by a tenth), interest rates (non-par guarantees), and equity markets (ULIP AUM and fees).

2. General insurance¶

Non-life insurers (motor, health, fire, crop) write one-year contracts, so the accounting is more honest, and the economics are Buffett's: float (premiums collected before claims are paid) invested for the insurer's benefit, plus an underwriting result.

Metric	Definition	Reading
Loss ratio	Claims incurred ÷ net earned premium	Underwriting discipline; health and motor third-party run high
Expense ratio	Operating expenses + commissions ÷ net written premium	Distribution cost
Combined ratio	Loss ratio + expense ratio	< 100% = underwriting profit; Indian general insurers have often run above 100%, earning their profit on float investment income
Float	Reserves for unpaid claims + unearned premium	Interest-free capital if combined ratio ≤ 100%; check its cost when above
Reserving adequacy	Prior-year claim development (reserve releases or strengthening)	Persistent releases = conservative reserving (good); strengthening = past under-reserving (bad)
Solvency ratio	As for life; minimum 150%
Growth by segment	Motor OD/TP, health retail vs group, crop, commercial	Mix drives loss ratios; crop is volatile and government-priced

Valuation: P/E and P/B work (profits are annual), but adjust for float value and reserve quality. An insurer at 100% combined ratio with float of 2x equity invested at 7% earns 14% on equity from float alone; that is the franchise.

3. Asset management companies (AMCs)¶

An AMC earns a percentage of assets under management (the total expense ratio, TER) for managing mutual funds. The economics are the purest operating leverage in finance: revenue is AUM × yield; costs are mostly fixed (people, technology) plus distributor commissions that scale with AUM.

Metric	Definition	Reading
AUM and mix	Equity / hybrid / debt / liquid / ETF-passive	Equity AUM yields ~4–8x what liquid/ETF AUM yields; mix drives revenue more than total AUM does
Net flows vs mark-to-market	Decompose AUM growth	Flows are the franchise; MTM is the market
SIP book	Monthly systematic inflows	Sticky, retail, the industry's structural tailwind (industry SIP flows ran above ₹25,000 Cr/month through 2025–26 — verify current AMFI data)
Yield (revenue/avg AUM)	Blended TER retained by the AMC after distributor share	Falls as AUM grows (SEBI's slab-based TER caps step down with scheme size) and as passive share rises
Cost-to-income; operating margin		Best-in-class AMCs run 70%+ operating margins on core revenue
Market share of equity AUM; flows		Consolidation toward the top 5–7 players
Other income	Treasury on the AMC's own surplus	Often 20–30% of PBT; value at cash-like multiples

Regulation is the swing factor. SEBI sets TER caps by AUM slab and scheme type, has cut them several times (2018 was the big reset), and periodically consults on further reductions, on distributor commission structures and on performance-linked fees (verify the status of SEBI's 2025–26 TER and expense-rationalisation proposals). A 10-bp TER cut on equity AUM flows straight to the AMC's revenue and, with fixed costs, disproportionately to profit.

Valuation: P/E (30–45x has been common for listed Indian AMCs in strong markets — verify), or better, market cap ÷ AUM (roughly 5–12% of equity AUM for the leaders, less for debt-heavy books) cross-checked with a DCF of fee income under explicit TER-cut scenarios. The trader's instinct is right here: an AMC is a long-only leveraged bet on the equity market with a regulator holding a call on the fee rate.

4. Exchanges, depositories, RTAs, brokers¶

4.1 The plumbing and its economics¶

Business	Revenue driver	Take rate	Cost structure	Moat	Regulatory exposure
Stock exchanges (NSE — IPO of ₹22,562 Cr in September 2026, listing expected 24-Sep-2026; BSE; MCX for commodities)	Traded volumes × transaction fee; listing fees; data; co-location; clearing	Basis points of turnover; per-lot in options	Mostly fixed (technology); huge operating leverage	Liquidity network effect + licence	Transaction-charge caps; product rules (weekly expiries, lot sizes); interoperability
Depositories (CDSL, NSDL)	Number of demat accounts (annual issuer charges, account maintenance) + transaction charges	Small fixed fees per account/transaction	Fixed	Duopoly licence + network	Fee caps; account-opening cycles
Clearing corporations	Clearing fees; margin float income			Licence	Margin rules
Registrars (RTAs) (CAMS, KFin)	AUM of mutual funds serviced × bps; folio counts; IPO/corporate registry	Small bps	Fixed	Duopoly with switching costs	AMC fee pressure flows through
Brokers (discount and full-service; listed: e.g. Angel One, Zerodha unlisted)	Orders × brokerage; interest on client float and margin funding; distribution	Per order or bps	Semi-variable (customer acquisition)	Weak — price competition; some scale in technology	Direct: brokerage caps, float rules (SEBI's "ASBA-like" client-fund handling), F&O curbs

4.2 The 2024–25 SEBI derivatives measures — a live case¶

Index options had become the world's largest derivatives market by contract count, driven by Indian retail trading weekly expiries. SEBI's study found most retail F&O traders lost money, and from November 2024 it phased in measures: one weekly-expiry index per exchange, larger minimum contract sizes (Nifty lot from 25 to 75, later adjusted; Bank Nifty weekly expiries withdrawn), upfront collection of option premiums, intraday position-limit monitoring and removal of calendar-spread margin benefit on expiry day (Zerodha summary; NSE lot-size circulars; lot sizes in 2026). Volumes in the affected segments fell sharply; exchange and broker revenues that depended on retail options activity fell with them.

The analytical lessons:

Regulatory take-rate risk is not hypothetical for businesses whose revenue is a slice of a flow the regulator considers socially harmful. Model it as a scenario with a probability, not as a footnote.
Volume is the state variable. Build the exchange model as volumes × fee, and stress volumes (−30%, −50%) with costs fixed: the operating leverage that made the stock a multi-bagger works in reverse.
Mix matters: cash equities, index options, stock futures, commodities and currency have different fee structures and different regulatory sensitivities.
Second-order effects: fewer weekly expiries concentrated volume on the surviving ones (BSE's Sensex weekly gained share); brokers' float income rose with margin requirements; discount brokers' order counts fell. Every change in plumbing rules reallocates revenue across the chain.

Trader's lens

You have seen this from the other side. An exchange is short a put on retail participation and long the regulator's forbearance; a broker is the same with less moat. When you look at MCX's or BSE's P/E, decompose the earnings by product and ask which products SEBI, the ministry or a court could switch off, and what fraction of profit sits there. That is a jump-risk analysis, and it should be priced as one.

4.3 Valuation¶

Exchanges and depositories: DCF of fee income with explicit volume and take-rate scenarios (the regulator is a scenario), cross-checked against global exchange multiples (typically 20–30x earnings for scaled, diversified exchanges — verify) and against the value of the non-volume revenue (listing, data, technology) which deserves a higher multiple than the volume revenue. Brokers: P/E with a cyclical haircut, and a close look at how much of profit is float interest (rate-sensitive) versus brokerage (volume-sensitive). RTAs: DCF of servicing fees keyed to mutual-fund AUM growth with AMC-fee-pressure pass-through.

5. Choosing the method — summary¶

Business	Primary metric	Valuation	Key risk to scenario
Life insurer	EV, VNB, VNB margin, persistency	P/EV; appraisal value (EV + VNB multiple)	Persistency; product regulation (e.g., surrender-value rules of 2024); rates
General insurer	Combined ratio, float, reserving	P/E and P/B with float-adjusted RoE	Claims inflation (health), crop, pricing cycles
AMC	Equity AUM, flows, yield, margin	P/E; mcap/AUM; DCF under TER scenarios	TER cuts; passive shift; market drawdown
Exchange / depository	Volumes, take rate, product mix	DCF with regulatory scenarios; global exchange multiples	Regulatory product/fee changes; competing venue
Broker	Active clients, orders, float	P/E with cyclical haircut	Regulatory curbs; price competition; rates
RTA	Serviced AUM, folios	DCF; P/E	AMC fee pressure

India notes

IRDAI's 2024 changes to surrender values (higher guaranteed surrender values earlier in a policy's life) reduced margins on non-par savings products and forced repricing — an example of product regulation moving VNB margins across the industry within a quarter (verify the specifics and subsequent margin trends).
Bank-owned insurers and AMCs (SBI, HDFC, ICICI groups) depend on the parent bank's distribution; check the bancassurance share and any regulatory push for "open architecture".
NSE's long-delayed IPO finally happened in September 2026 (₹22,562 Cr, subscribed 5.7x, allotment 22-Sep, listing expected 24-Sep-2026 — Kotak Neo), creating by far the largest listed exchange; its first annual report will be the reference document for exchange economics in India. Until then, listed exchange analysis meant BSE and MCX — smaller, more product-concentrated and hence more regulatorily sensitive.

Common mistakes

Valuing a growing life insurer on statutory P/E (it looks absurdly expensive) or a shrinking one as cheap.
Treating VNB margin as fixed — it moves with product mix, and mix moves with regulation and rates.
Ignoring persistency: a VNB booked on a policy that lapses in year two was never real.
Valuing an AMC on total AUM without the equity/debt mix.
Extrapolating exchange volumes from a retail boom year.
Forgetting that other income (treasury) is a large share of AMC and exchange profits and deserves a lower multiple.

Key terms¶

Term	Meaning
Embedded value (EV)	Net worth + PV of future profits on in-force life policies
Value of new business (VNB)	PV of expected profits from policies sold in a period
VNB margin	VNB ÷ annualised premium equivalent (APE)
APE	Annualised premium equivalent: regular premium + 10% of single premium
Persistency	Share of policies still in force after a given period (13th, 25th, 37th, 61st month)
Operating RoEV	Operating growth in embedded value ÷ opening EV
P/EV	Market capitalisation ÷ embedded value
Appraisal value	EV + value of future new business
Combined ratio	Claims ratio + expense ratio for a general insurer; < 100% = underwriting profit
Float	Premiums held before claims are paid; invested for the insurer's account
TER	Total expense ratio charged to a mutual-fund scheme; capped by SEBI in slabs
SIP	Systematic investment plan; recurring retail mutual-fund inflows
Take rate	Fee as a share of transaction value or volume
Weekly expiry	Index option contracts expiring weekly; the focus of SEBI's 2024–25 curbs
RTA	Registrar and transfer agent; services mutual funds' investor records

Check your understanding¶

An insurer's statutory profit rises 20% while APE falls 10% and VNB falls 15%. Is the business improving?

Answer

No. Statutory profit is rising because fewer new policies mean less first-year strain, while the in-force book releases profit; VNB — the value created by selling — is shrinking. The franchise is worth less; the accounting says the opposite.

Recompute Vindhya Life's appraisal value if $k_e$ is 13% and VNB growth fades to 5% after five years instead of ten (5 years at 12%, then 5%).

Answer

Run the loop with growth = [0.12]×5 then terminal at 5% from year 5: structural value falls to ~₹27,900 Cr and appraisal to ~₹47,900 Cr, i.e. P/EV ≈ 2.4x — a 100-bp higher $k_e$ and a shorter growth runway remove a third of the structural value; P/EV is very sensitive to both.

An AMC has ₹3,00,000 Cr AUM, 55% equity at a 60-bp net yield and 45% debt/liquid at 12 bp; costs ₹700 Cr. What is operating profit, and what does a 10-bp cut in the equity TER do to it?

Answer

Revenue = 1,65,000 × 0.0060 + 1,35,000 × 0.0012 = 990 + 162 = ₹1,152 Cr; operating profit = 1,152 − 700 = ₹452 Cr. A 10-bp equity cut removes 1,65,000 × 0.0010 = ₹165 Cr of revenue → profit ₹287 Cr, −37% from a 14% revenue cut. That is operating leverage in reverse.

Why did SEBI's 2024 measures hurt some plumbing businesses more than others?

Answer

Revenue concentration: exchanges and discount brokers whose growth had come from retail index options lost the most; depositories (account-based fees) and RTAs (AUM-based) were largely unaffected; BSE gained relative share as the surviving weekly expiry venue for Sensex. The lesson is to map revenue by product to regulatory sensitivity.

Why does a general insurer with a 102% combined ratio still earn a return for shareholders?

Answer

It pays 2% of premium for the use of its float, which it invests. With float of 2x equity earning 7%, investment income adds ~14% to RoE, more than covering the 2% underwriting loss on premium. The business is an asset manager funded by policyholders — attractive while the combined ratio stays close to 100%.

Go deeper¶

Listed Indian life insurers' annual embedded-value reports (with independent actuary certification) — read one cover to cover; they are the clearest financial documents in India.
IRDAI annual report and AMFI monthly data — industry volumes, mix, SIP flows.
SEBI's consultation papers and circulars on F&O (2024–25) and on mutual-fund expenses — primary sources for the regulatory scenarios.
Warren Buffett's Berkshire letters on insurance float (esp. 1995–1999 and the "float" appendices).
Sources used: Zerodha on SEBI's index-derivative rules; Jainam summary of SEBI F&O rules; NSE circular on lot sizes.

← Previous: 07.1 Banks & NBFCs · Module index · Next: 07.3 Cyclicals & commodities →