IRDAI Whitepaper

IRDAI's 23 September 2026 draft on the economics of insurance distribution, played through the chain.

₹51,841 Cr a year stops flowing to the people who sell insurance. Buyers get ₹19,006 Cr back as lower prices and insurers keep ₹19,006 Cr. The other ₹13,829 Cr simply disappears, because ₹47,087 Cr of premium is never written.

Section 02 · Reported terms · model read-out

The rule, and what it means

What the draft changes, how far each payout falls against its cap, and the dates that decide the final rule.

Section 03 · Δ annual profit after tax · ₹ Cr

Who wins, who loses

Fourteen listed names ranked by change in annual profit, with the year-by-year path and the arithmetic behind each figure.

Rupee losses concentrate in PB Fintech (Policybazaar) and Bajaj Finance; relative pain is worst at PB (−172% of profit). Across the 14 names the net change is +1,843 Cr a year.

Life insurers are valued on the value of new business (VNB), not accounting profit. Lost new-business premium × an assumed 25% VNB margin gives a rough annual value at risk.

Section 04 · Deep dive · most exposed listed name

PB Fintech: valuation and upside

Policybazaar and Paisabazaar, the name most exposed to the reset: how a revenue cut becomes a profit cut, what the share is worth, and which inputs matter most.

At current settings PB is worth about ₹664 a share (−45% vs ₹1,210). The price sits closest to A · Softened final rule (₹1,398). The biggest swing factor is fy31 net margin (±₹218), so the valuation debate is about that input as much as the draft.

Revenue falls 45.2%, but costs don't fall with it: only 50% of lost revenue comes back through lower ad spend and partner payouts. The rest comes straight out of profit, and PB's core pre-tax margin is only about 5.6%. The cost offset (Model Lab) changes profit, not value per share: valuation runs on the revenue cut.

Value per share as each input moves across its plausible range, everything else held at current settings.

Solid line: value at current settings. Dashed: share price. Growth, margin and multiple follow scenario B. Also tested: Enforcement (±₹95), Discount rate (±₹57), Sales recaptured online / direct (±₹54), Growth FY27–28 (±₹44), Volume sensitivity (±₹23).

Section 05 · ₹ Cr a year

Where the commission pool goes

Every rupee of lost distribution payout ends up in one of three places: buyers, insurers, or premium that is never written.

Of every ₹100 of lost payouts, ₹37 goes back to buyers, ₹37 stays with insurers and ₹27 disappears because less insurance is sold. The bigger that last slice, the more the reform costs growth rather than just moving money.

Section 06 · Scenarios side by side · 1,000 simulated end-states

How sure can we be?

Every scenario side by side, then 1,000 random combinations of the uncertain inputs to show how wide the range really is.

Across the five rule presets, PB's profit change runs from −172% to −19%. Across 1,000 random draws of the unknown inputs, 80% of outcomes fall between −119% and −41%. The choice of rule moves the answer more than the unknown inputs do.

All scenarios

Section 07 · Move any lever · every section above updates

Model lab

All seven levers and every product cap in one place. Changes here flow through every section above.