I'll give you the numbers first, and then I'll discuss them.
In fiscal year 2025 (FY25), life insurers paid out ₹6.3 trillion in benefits, according to the insurance regulator's September 2026 consultation paper on distribution costs. Of this, death claims accounted for ₹0.47 trillion, or a mere 7%. Maturity benefits took 35%, surrender payouts 37%, and annuity and pension payments the remaining 20%.
Read that again. Money paid to customers who gave up their policies far exceeds the money paid to families of customers who died. Surrender payments alone are five times larger than death claims.
The numbers from Irdai's consultation paper deserve far more attention than the new commission rates that everyone, including me, quickly wrote about. The payoff from digging into this 120-page paper is remarkable: it contains unusually revealing data about an industry whose underlying economics are often difficult to see.
These numbers should make an industry—and its regulator—uncomfortable. Life insurance exists, at least in part, to protect families from the financial consequences of a breadwinner's death. Yet death payouts are a small fraction of the industry's total benefits paid. Maturity benefits alone are nearly five times death benefits. Whatever else the industry is doing, insuring against death appears to be a relatively small part of the outcome.
The bigger problem
Then there is the persistency data.
Overall, 61st-month persistency across the industry is just 48%, meaning fewer than half of policies remain in force after five years. By the 10th year, the proportion of policies still in force is as low as 8% for some insurance companies.
That is a remarkable level of attrition. And there is a particularly interesting qualifier in the data.
Online policies, bought without agent involvement, have a 61st-month persistency rate of 71%. A policyholder who bought a policy without the same degree of sales intervention is much more likely to keep it. The gap between online and agent-driven policies raises an obvious question about whether the advice being given is consistently aligned with the customer's long-term interest.
The incentive structure offers part of the explanation.
The paper's commission data show that first-year commissions on savings-linked policies average 25-37% of the premium. For some companies, total first-year distribution costs are as high as 60-80%. Renewal commissions on the same products range from 0-5%.
The incentives are therefore heavily weighted towards acquiring a new customer rather than retaining one. It should not be surprising, then, if the industry's sales machinery is better at generating new policies than ensuring that customers stay with them.
There is another loss hidden in these numbers. According to the data, even customers who surrender at around the five-year point generally do not recover the premiums they have paid. Forget returns; in many cases, they take an actual loss.
The industry's defence against such criticism is that these are savings and investment products, and judging them by the standard of pure protection is unfair because customers buy them for the maturity payout, not the death benefit.
But that defence gets to the heart of the problem. When customers buy products from an insurance company that function primarily as high-commission investment products with a relatively small insurance component, it raises a basic question about whether the product is doing either job particularly well.
If you are holding such a policy, those surrender figures are about you, and the decision to exit is harder than the headline numbers might suggest. The premium you've already lost is gone either way. Continuing to pay for another 15 years simply to avoid acknowledging a bad purchase could turn a five-year loss into a 20-year one.
A more useful comparison is forward-looking: what will the policy pay you from here? Set that against the cost of a term plan plus an index fund for the same annual outgo, and make the decision on those numbers rather than on the money already spent.
None of this necessarily implies criticism of individual sellers, or any cynicism on their part. They are operating within an industry design and incentive structure that they did not create. The more important question is whether that structure produces products and advice that work in the customer's long-term interest.
And the data in Irdai's paper point to a conclusion that is difficult to ignore: for someone whose primary need is life insurance, plain-term insurance remains the clearest form of protection because its purpose is straightforward—pay a benefit if the insured dies during the policy term, rather than combining insurance with a savings or investment proposition.
Dhirendra Kumar is founder and chief executive officer of Value Research, an independent investment advisory firm.