The Insurance Act 1938 is the statute under which every insurer in India holds its licence, holds its capital and answers for its conduct, and it contains one provision that decides more contested life insurance claims than any other sentence in Indian financial law. Section 45 provides that no policy of life insurance shall be called in question on any ground whatsoever after three years have run from the last of four dates to occur. A family that understands that sentence, and knows how the three years are counted, understands the strongest protection the statute gives it. A business reader needs a different part of the same Act: how a company gets registered, how much capital it must bring, how much of it foreign investors may own, and where the limits on its spending and investing sit.

Three boundary notes will keep the cluster honest. The Act’s text is followed in its amended form, with the pre amendment position stated only where the comparison does legal work, as it does for Section 45. Figures that sit in regulations rather than in the Act, the solvency ratio, the expense ceilings, the surveyor threshold, are described as regulation made, because parliamentary text and subordinate law deserve different weight in a reader’s confidence. And judgments are used for what they decide, not decorated with citations: the series cites Mithoolal Nayak, Asha Goel and Satwant Kaur Sandhu because each of them governs how a claim is actually fought.

This article works through the Act in that order of importance. It begins with what stood before 1938 and why it failed, then takes the architecture of the statute apart: the gate at the door, the capital rules, the prudential spine of investments and solvency, the rule that no risk begins before the premium is paid, and the provisions that treat a policy as property that can be nominated and assigned. The centre of the article is Section 45, stated exactly, with the table that shows what an insurer may do on each side of the three year line. The second half follows the sector the Act has governed: nationalisation in 1956 and 1972, the reopening of 1999, the large amendment of 2015, and the foreign investment argument that has run from a ceiling of 26 per cent to a ceiling of 100 per cent. One caution belongs at the start: the regulator and the insurance ombudsman are owned by the companion article on the IRDAI Act 1999, and this article links there rather than retelling that story.

Before 1938: insurance sold faster than it was supervised

Insurance reached India early and was regulated late. The first life office on Indian soil was the Oriental Life Insurance Company, set up in Calcutta in 1818, and the first Indian-owned life insurer, the Bombay Mutual, followed in 1870. General insurance has its own founding dates: the Triton Insurance Company, established in Calcutta in 1850, is generally counted the first general insurer in the country, and the Indian Mercantile Insurance Company of 1907 the first Indian owned general insurer. Between those dates and 1938 the market filled with agency houses and joint stock offices of every quality, Indian enterprise expanding in the swadeshi decades while the largest share of premium, particularly in general lines, continued to flow to foreign offices. For the rest of the nineteenth century the business grew on the strength of agency networks and public trust, with no statutory requirement that an insurer demonstrate it could pay the claims it was collecting premiums against. The predictable result was a long tail of weak offices whose promises were worth more than their funds.

Parliament responded in pieces. The Indian Life Assurance Companies Act, 1912 was the first statute aimed at life business, and the Provident Insurance Societies Act of the same year addressed the small premium provident societies that sold cover to poorer households. An amending Act of 1928 widened the legislative grip to general business. The patchwork had two structural defects. It supervised life and general insurance under different logics although the same weaknesses, thin capital, related party lending, premium income spent as it arrived, appeared in both. And it required deposits and returns without giving the supervising officer a coherent code against which to judge an insurer’s whole position.

The market the patchwork governed also had a shape that reformers disliked. European and other foreign offices held much of the life business of the cities, while Indian offices, many founded in the nationalist impulse of the early twentieth century, competed for the remainder, often on price and on commission to agents rather than on reserves. Inspection powers under the 1912 scheme were thin, publication of accounts was not standardised into comparability, and there was no mechanism by which a policyholder, or a rival, could tell a sound office from an insolvent one until the cheques stopped being honoured. The Act also built the industry’s own institutions into the statutory scheme, a feature easy to miss. It recognised an Insurance Association of India with a Life Insurance Council and a General Insurance Council, bodies of the insurers themselves with functions in the framing of standard practices and the discipline of the market, whose executive committees the Act armed with defined powers. Self regulation under statute was the 1938 compromise between a free market and a department: the industry would organise itself, in the open, under a code and a Controller. The councils survive in modern form as industry bodies alongside the Authority, their statutory functions largely absorbed by the regulator, their existence a reminder that the Act never imagined supervision as the state alone.

The consolidating impulse of 1938 was therefore not tidiness for its own sake. It was the judgment, reached after two decades of incremental statutes, that insurance needed one code, one registrar with real powers, and one set of financial disciplines applied to life and general business alike. The Act that followed borrowed heavily from the British assurance legislation of the period, adapted to a market in which the state itself would, within twenty years, become the largest insurer.

The Act of 1938 replaced the patchwork with a single consolidating statute. Its long title states the ambition plainly: an Act to consolidate and amend the law relating to the business of insurance. It was enacted as Act 4 of 1938 and brought into force on 1 July 1939. The statute it created has been amended many times, most heavily in 1968, in 1999 and in 2015, but its skeleton is the skeleton of 1938, which is why a statute drafted for a market of agency driven offices still frames a market of listed insurers and foreign joint ventures.

The first decades: the code at work

The Act’s first seventeen years, from its commencement on 1 July 1939 to nationalisation in 1956, are the least written about and the most instructive, because they show the code governing the market it was drafted for. The register filled with Indian and foreign offices for life business and a separate population for general lines. Deposits were made, returns were filed, the Controller’s office acquired the habits of a supervisor, and the statutory forms for accounts and actuarial valuation began to make one insurer’s position comparable with another’s, which is the precondition of every later discipline in this article. The war years interrupted growth and normalised state direction of the economy; the years after independence brought expansion, new Indian offices, and a rising official impatience with the industry’s governance.

That impatience had particular targets. Insurance companies had become, in a number of prominent instances, appendages of industrial houses, their funds invested in group concerns and their boards arranged for the convenience of promoters rather than policyholders. Weak offices still failed. The security argument for nationalisation, that a family’s savings in a life policy should not depend on the solvency of a promoter controlled company, drew its force from this record and not from doctrine alone. The amendment of 1950, tightening investments and curbing the managing agents, was Parliament’s attempt to discipline the private market into adequacy; the decision of 1956 was the judgment that the attempt had not gone far enough. Readers of the later debates should hold the sequence in view, because the same two step, regulation attempted, ownership changed, frames the argument over foreign control that the article reaches at its end.

How the Insurance Act 1938 is put together

The Act is best read as four machines sharing one housing. The first machine controls entry: who may carry on insurance business at all, in what corporate form, and with how much capital. The second machine is prudential: it governs what an insurer does with the money it collects, through investment rules, separate accounts, returns, audit and a solvency floor. The third machine governs the policy itself: nomination, assignment, the premium before risk rule, and the limits Section 45 places on an insurer that wants to escape a policy it has issued. The fourth machine is supervisory and penal: inspection, directions, adjudication and the penalties that give the first three machines their teeth.

Two features of the drafting explain much of what follows. The Act deals with life and general insurance inside one statute, dividing them by class of business rather than by separate codes, so a provision such as the investment rule states a pattern for life funds and a different pattern for general funds side by side. And the Act was written for supervision by a government officer, the Controller of Insurance, whose office carried the regime for six decades; when a statutory Authority replaced the Controller at the end of the twentieth century, the amendment technique was largely to substitute the Authority into the 1938 text rather than to rewrite the text. Readers who wonder why a modern regulator works through a statute of 1938 have their answer there: the Act is the operating system, and the Authority is the operator the system was re-pointed at.

The Act in its amended form runs to well over a hundred sections, besides its definitions, its schedules of forms and the body of regulations beneath it, and no honest article can pretend to walk all of it. This article walks the load bearing parts and says so.

The definition section does quiet but decisive work. An insurer means an Indian insurance company, a statutory body established by an Act of Parliament to carry on insurance business, an insurance co-operative society, or a foreign company engaged in reinsurance business through a branch established in India. An Indian insurance company is a company formed and registered under the companies legislation whose sole purpose is a class of insurance business and whose foreign shareholding stays within the limit the Act prescribes. Those two definitions, read with the prohibition in Section 2C, draw the entire perimeter of lawful insurance in India.

A reader meeting the Act for the first time should also be warned about its numbering. The sections inserted by later amendments carry letters and compound tags, Section 42D, Section 64VB, Section 64VA, Section 105C, because Parliament has amended the code into itself for eighty years rather than renumber it. The Part on general insurance business runs through the sixties of the numbering, which is how the premium before risk rule acquired the label 64VB, sitting far from the policy provisions it protects. Nothing in the labels indicates importance. Section 45, a plain number in the original drafting, is worth more litigation than the whole of Part V.

The amendments that rebuilt the Act

The statute of 1938 was amended within two years of the Republic and has been amended in every decade since, but four revisions did structural work. The first great amendment came in 1950. It tightened the investment provisions into something close to the Section 27 pattern the Act carries in substance, and it began the statutory curbing of the managing agency system, under which managing agents controlled insurance companies on slender capital of their own while drawing commission on the whole premium income. The managing agent, an inheritance from the agency house era of Indian commerce, is the recurring villain of the pre nationalisation record in official accounts: control without capital, remuneration without risk. The curbing begun in 1950 was completed by nationalisation itself, and the modern Act’s insistence that voting rights follow equity capital in Section 6A is the distant descendant of that quarrel.

The amendment of 1968, carried in the era when public ownership was extending across the financial sector, tightened the Controller’s hand: it inserted Section 64VB, the premium before risk rule, as a discipline on the cover on credit habit, and it sharpened the management expense and solvency provisions that nationalisation would soon make the government’s own problem. The general insurance nationalisation of 1972 worked through its own Act, but its operating code remained the Act of 1938.

The amendment of 1999 was the second structural event. Rather than re enact the code, the IRDA Act amended the Insurance Act, the LIC Act and the nationalisation Act in one motion, substituting the Authority for the Controller and the government across the 1938 text and removing the statutory monopolies that had made most of the Act dormant. Three years later the amendment of 2002 admitted insurance co-operative societies to the perimeter, a change that matters to the co-operative strongholds of the sector and explains why Section 2C speaks of co-operative societies as first class citizens of the gate. The amendments of 2015, 2021 and 2025, each of which gets its own treatment later in this article, are the later structural events: they rewrote the policy provisions, repriced the penalties and raised the ownership ceiling in steps to the whole of the capital.

Running beneath the amendments is a law reform thread. The Law Commission of India examined the Act in its 190th Report, on the revision of the Insurance Act, and recommended a modernising overhaul, including a fixed outer limit on the insurer’s power to call a policy in question. The three year bar of 2015 grew from that recommendation line, which is worth remembering when the bar is criticised as an accident of drafting. It was the product of a decade of official reconsideration, carried into law by an amendment bill that had itself been pending, in one form or another, since 2008.

The Controller of Insurance: the Act’s first operator

For the Act’s first six decades its operator was an officer of the Government of India. The Act vested administration in a Controller of Insurance, an office inside the finance administration of the state, and the Controller registered insurers, received their returns, licensed their agents and answered for the sector to the government of the day. The arrangement had the strengths and weaknesses of departmental supervision everywhere. It was inexpensive and continuous, and it kept insurance inside the general financial administration alongside banking and the public debt. It was also, structurally, a supervisor without an institution: no independent board, no dedicated fund of supervisory talent insulated from transfer, and, after 1956 and 1972, the awkward position of a government officer supervising insurers the government itself owned.

The rule making architecture of the Act still bears that history. Under Section 114 the Central Government makes rules under the Act; under Section 114A, inserted when the Authority was created, the Authority makes regulations consistent with the Act and the rules. Much of the lived law of insurance, solvency computation, expense ceilings, surveyor thresholds, intermediary conduct, sits in those regulations, which is why this article repeatedly distinguishes what the Act itself fixes from what it empowers the Authority to fix. The 1999 substitution changed the operator without changing the instrument: the Authority stepped into the Controller’s statutory shoes, and the divide between government rules and the Authority’s regulations was drawn at the same moment. Any account of Indian insurance that treats the Act of 1938 and the Authority of 1999 as rivals misreads the engineering. The Act supplies the powers; the Authority is the hand Parliament fitted to them after sixty years of its own hand proving too slow.

The gate at the door: Section 2C and registration

Section 2C states the perimeter as a prohibition. No person may begin or carry on any class of insurance business in India unless that person falls within one of the permitted forms: in substance, an Indian insurance company, an insurance co-operative society, a statutory body created for the purpose, or a foreign reinsurer’s branch of the kind the Act admits. Carrying on insurance business outside the perimeter is not a licensing irregularity; it is an offence, and it exposes the persons running the business to prosecution as well as to the Act’s penalties.

Within the perimeter, Section 3 requires registration. An insurer must obtain a certificate of registration from the Authority for the class of business it proposes to write, and the registration provisions give the Authority the corresponding powers of refusal, suspension and cancellation. The grounds on which registration may be taken away track the Act’s own disciplines: failure to satisfy the capital requirements, failure to maintain the required solvency, failure to comply with the Act and the regulations under it, and conduct that the Authority finds prejudicial to policyholders. A cancelled insurer does not vanish with its liabilities; it stops writing new business while its existing book is run off, transferred or wound up under supervision, because the policyholders of a failed insurer are the people the whole apparatus exists to protect.

Registration under the Act is class specific, and the classes matter commercially. Life insurance, general insurance, health insurance carried on exclusively, and reinsurance are distinct registrations with distinct capital attached. A composite licence that lets one company write everything does not exist; groups operate through separate companies for separate classes, which is a direct survival of the 1938 habit of separating accounts and funds by class. The life and general separation is the sharpest form of the rule: a company registered for life insurance may not write general business, and the two great nationalisations of 1956 and 1972 built separate state structures on the same line, so the Indian market’s institutional geography, life companies here, general companies there, health specialists beside them, is the Act’s class map drawn in buildings and brands. Proposals to permit composite licences, one company across life and general lines, have been consulted on by the government in amendment exercises over the years and have not reached the statute book; the 1938 separation still governs.

The co-operative form deserves its sentence, because it is the perimeter’s oldest Indian shape. Co-operative insurance societies, owned by the members they insured, were a feature of the market long before 1938, and the original Act recognised co-operative societies among the forms that could carry on insurance business; the amendment of 2002, passed in the licensing era, regularised their admission to registration under the modern scheme. Co-operative insurers remain a small part of the market by premium, but the form matters to the Act’s completeness: the perimeter is defined by function and solvency, not by corporate fashion, and a member owned society that meets Section 6 and submits to Section 64VA is as lawful an insurer as a listed company.

The registration gate also explains the Act’s treatment of the two state owned giants. The Life Insurance Corporation is not registered under the Insurance Act at all; it is a statutory body created by its own Act of 1956, which applies many provisions of the Insurance Act to it with modifications. The state owned general insurers, by contrast, are companies within the Act’s scheme. The perimeter, in other words, has always had a parliamentary side door for insurers Parliament creates directly.

In practice the gate is passed in stages, because registration is the point at which the state decides who may be trusted with the public’s savings. The Authority examines the promoters and their fitness, the capital they commit and the schedule on which they will add to it, the business plan and the classes it proposes to write, and the governance it offers, before a certificate is granted; the detail of that scrutiny lives in the registration regulations, but its statutory root is the discretion Section 3 gives the Authority to refuse an applicant that cannot satisfy it. The staged design has a purpose beyond caution. Insurance failures are discovered late, when the policies are already sold and the claims are maturing, so the system spends its scepticism at the door rather than at the funeral.

Capital and ownership: what Section 6 demands

Capital is the second half of the entry gate, and Section 6 fixes it in figures that have been stable since the sector reopened. No insurer may be registered unless it has a paid up equity capital of one hundred crore rupees where it carries on life insurance or general insurance, one hundred crore rupees where it carries on health insurance exclusively, and two hundred crore rupees where it carries on reinsurance exclusively. A foreign company writing reinsurance through an Indian branch is measured differently: it must have net owned funds of not less than five thousand crore rupees. Section 6A adds the structural rules around the number: the capital of an Indian insurance company consists of equity shares of a single face value and such other forms of capital as the regulations specify, voting rights attach to equity shares alone, and the paid up amount must be uniform across shares outside a short window for calls. The section also polices movements in the register: a transfer of shares that crosses the thresholds the Act specifies requires the Authority’s prior approval, so that control of an insurer cannot change hands as a purely private transaction between shareholders. Read with the foreign investment ceiling in the definition, the scheme is complete: capital fixes how much loss the owners stand behind, Section 6A fixes how the ownership itself may be constituted and traded, and the definition fixes how much of it may be foreign.

The size of these figures is itself a policy choice worth noticing. One hundred crore rupees is a serious but not forbidding sum: high enough to exclude the thinly capitalised offices whose failures littered the pre 1938 record, low enough that Indian promoters without a foreign partner can still enter. The reinsurance figure is higher because a reinsurer’s book is concentrated risk written on other insurers’ promises. The branch figure is higher still, because a branch has no separate Indian capital and the home balance sheet is the only fund an Indian claimant can look to.

What the capital is for should be stated plainly, because it explains why the figure is a condition of registration rather than a fee. Paid up capital is the loss absorbing layer that stands between an insurer’s mistakes and its policyholders: underwriting losses, investment losses and expense overruns are charged against the owners’ money before they can touch the funds held against claims. The requirement that the capital be paid up, in money, before registration, and that shares be of one class with voting rights confined to equity, prevents the two classic evasions, capital promised but never contributed, and control exercised through paper that carries no risk. It is also worth recording that the figures have stood still while the market grew around them. One hundred crore rupees was a demanding threshold when the sector reopened in 2000; in a market whose premium income has multiplied since, a static figure is a barrier that lowers itself a little each year, and the real entry discipline has migrated to the Authority’s scrutiny of business plans and to the solvency margin an insurer must fund as it grows.

Ownership sits in the definition rather than in Section 6. The definition of an Indian insurance company caps the aggregate holding of foreign investors, including portfolio investors, at the percentage the Act states, and attaches conditions to that holding. The percentage has been the single most amended number in the statute: 26 per cent when the sector reopened, 49 per cent under the amendment of 2015, 74 per cent under the amendment of 2021, and 100 per cent under the amendment of 2025. The full sequence, and the argument about it, is taken up later in this article. What belongs here is the mechanism: the ceiling is a condition of being an Indian insurance company at all, so an insurer whose foreign holding crosses the ceiling does not merely breach a rule, it falls outside the definition on which its registration rests.

Reinsurance and the national reinsurer

Reinsurance, the insurance of insurers, sits inside the same Act with two features of its own. The first is capital: a reinsurer needs two hundred crore rupees of paid up equity capital under Section 6, double the direct insurer’s figure, and a foreign reinsurer operating through a branch must show net owned funds of five thousand crore rupees, because the branch has no Indian capital of its own and its Indian cedants are, in substance, lending against the home office balance sheet. The second feature is Section 101A, inserted when the sector reopened, under which every insurer must re insure such percentage of its business with the Indian reinsurer as the Authority specifies after consulting the Reinsurance Advisory Committee. The Authority designated the General Insurance Corporation of India in that role when it was reconstituted as a pure reinsurer at the reopening, and the obligatory cession has been lowered in steps over the years as the market deepened.

The branch route deserves a paragraph because it reverses the Act’s oldest instinct. From 1938 until 2015, a foreign insurer could not operate in India except through an Indian company: the 1938 perimeter, like the settlement of 1999, forced foreign risk capital to take Indian corporate form, with Indian capital locked inside it. The amendment of 2015 admitted the foreign reinsurer’s branch as a fourth form inside the perimeter, on the theory that reinsurance is a wholesale market between sophisticated parties, in which the branch’s five thousand crore funding requirement and the Authority’s supervision give cedants enough protection without a separate incorporation. Lloyd’s of London was among the first foreign reinsurance presences to take the branch route. Direct insurance remains behind the incorporation wall. The distinction is a deliberate piece of regulatory engineering: households buy from Indian companies, insurers may buy capacity from branches. The cession side is ordered as well: the reinsurance regulations lay down an order of preference that requires Indian insurers to offer their cessions first within India, to the Indian reinsurer and to the Indian branches and insurers entitled to them, before placing business abroad, so that reinsurance premium, like the savings it reinsures, is retained in the domestic market to the extent the market can hold the risk. Retention of premium was a nationalisation era instinct; in the branch era it survives as a rule of cession, administered by regulation, beneath a section of the Act of 1938.

The prudential spine: deposits, investments, accounts and solvency

The 1938 draftsman’s answer to insurer failure was a deposit. As enacted, Section 7 required every insurer to deposit cash or approved securities with the Reserve Bank of India, scaled to the size of its business, so that a fund outside the insurer’s own control stood behind its promises. The deposit regime survived for three quarters of a century and was then abolished: Sections 7 to 9, the deposit sections, were omitted by the amendment of 2015. The deposit was not repealed because security had stopped mattering. It was repealed because a modern solvency regime does the same job with better information, measuring the whole balance sheet instead of freezing one slice of it.

Why did the deposit with the Reserve Bank disappear from the Act?

The deposit was tied to business volume, not to risk, and it froze securities that earned the policyholder nothing while the insurer carried a real cost. The solvency system asks a better question: whether assets exceed liabilities by a margin proportioned to the risks actually taken on. Parliament made that substitution in 2015.

Two insurers with identical premium income and very different risk books deposited the same amount under the old rule, which is the definition of a blunt instrument. With the deposit sections omitted, the investment pattern of Section 27 and the solvency margin of Section 64VA carry the prudential weight together, and the paragraphs that follow take them in that order.

The investment pattern is Section 27. For life insurance business, the section requires the relevant fund to be held, first, as to 25 per cent in Government securities, second, as to a further sum of not less than 25 per cent in Government securities or other approved securities, and as to the balance in approved investments of the kinds the regulations specify. For general insurance the pattern is lighter: 20 per cent in Government securities, a further sum of not less than 10 per cent in Government securities or other approved securities, and the balance in accordance with the Authority’s regulations. A parallel rule directs insurers to invest and keep invested a stated share of their funds in infrastructure and social sector assets, so the pattern is not only about safety but about where insurance money is put to work.

The pattern carries a cost that should be reckoned honestly, because policyholders pay it. Compelling half a life fund into government securities and approved paper guarantees the sovereign a deep, captive buyer for its debt, and captive buyers do not bargain hard on yield; the return forgone is borne by the fund, which is to say by the bonuses and maturity values of the policyholders whose money it is. The defence of the pattern is equally concrete: life insurance liabilities are long, guaranteed in large part, and intolerant of loss, and a statutory floor of sovereign paper matches the promise the product makes. The 25 and 25 of Section 27 are the Legislature’s own balance between those two truths, struck in an era of thinner markets and kept, in substance, through every amendment since; the investment regulations have modernised what counts as approved and how exposure to a single group is capped, but the architecture of compulsion itself has not been seriously disturbed since 2000. A reader who wants the shortest possible statement of Indian insurance’s social contract could do worse: private capital may gather the savings of households, and in exchange the state directs where a fixed share of those savings must sleep.

How much of an insurer’s assets must remain in government securities?

Under Section 27 a life insurer must hold half the fund in Government securities or, as to the second quarter, in Government securities or other approved securities. A general insurer must hold 30 per cent across the two categories on the same pattern. The balance must go into approved investments within the regulatory limits.

Those limits cap exposure to any single investee or group and steer a share of the fund towards infrastructure and the social sector, so the freedom the balance appears to give is a supervised freedom.

Around the investment pattern sits the accounting machinery. Section 10 requires separate accounts for each class of business and separate funds, so that life policyholders’ money cannot quietly subsidise a general book. The life fund carries an additional protection that goes to the nature of the participating contract. A with profits policy promises the policyholder a share of the surplus his fund earns, and the statutory scheme confines the life fund to the purposes of the life business and restricts what may be taken out of it for shareholders, so that the surplus belongs, in the proportions the applicable law and the policy conditions fix, to the policyholders whose premiums created it. The Life Insurance Corporation’s own statute states the famous version of the rule for its fund, ninety five per cent of the surplus to policyholders; the private market works under the corresponding discipline of the Act and the Authority’s regulations. The principle predates both formulations and is one of the oldest in the code: the life fund is a trust like reservoir, not a profit pool, and the shareholder is paid from its edge and not from its body. Sections on returns, audit and actuarial investigation oblige the insurer to lay its position before the Authority in audited numbers and to have its liabilities valued by an actuary; the appointed actuary, a statutory figure inside every insurer, certifies the reserving on which the balance sheet stands. These provisions read as routine and are in fact the load bearing wall of the regime, because every supervisory decision, on solvency, on expenses, on a disputed reserving practice, is taken on numbers produced under them.

The actuarial leg of that wall is Section 13, which requires the financial condition of a life insurer, including a valuation of its liabilities, to be investigated by an actuary at regular intervals and reported in the prescribed form. A life insurer’s liabilities are not invoices; they are estimates of payments falling due across half a century, sensitive to assumptions about mortality, interest and expenses that only actuarial method can discipline. The appointed actuary system, built on the Act’s requirement of actuarial investigation and elaborated in the regulations, gives the supervisor a professionally accountable name inside each company whose certificate converts those estimates into numbers the Authority can act on. When insurers and the regulator argue about reserving, they are arguing inside the room Section 13 built.

The solvency floor is Section 64VA. Every insurer and reinsurer must at all times maintain an excess of the value of its assets over the amount of its liabilities of not less than 50 per cent of the minimum capital the Act requires of it. Above that statutory floor, the section empowers the Authority to specify a control level of solvency by regulation, and the operating standard in the regulations requires an available solvency margin of one and a half times the required solvency margin. An insurer that cannot maintain the Section 64VA floor is deemed insolvent for the purposes of winding up on the Authority’s application, and an insurer that slips below the control level must file a financial plan, within a period the section caps at six months, showing how the margin will be restored. The two tier structure is best understood with illustrative figures. Suppose an insurer’s liabilities and risk profile produce a required solvency margin of one thousand crore rupees under the regulations. The control level then obliges it to hold an available margin of one thousand five hundred crore; so long as it does, it trades freely. If losses or fast growth pull the available margin to one thousand three hundred crore, the insurer is solvent, its assets still exceed its liabilities by more than the required margin, but it is in breach of the control level and owes the Authority a financial plan. If the available margin fell below the statutory floor that Section 64VA itself fixes, fifty per cent of minimum capital over liabilities, the deeming provision and the winding up machinery are engaged. Policyholders reading an annual report should look for the ratio, not the rupee figure: the distance between the ratio an insurer reports and one and a half is the truest single number in the document. The journey from a deposit in a vault to a ratio watched quarter by quarter is the clearest single illustration of how the 1938 chassis carries twenty first century supervision.

The prudential machine has an endgame, and the Act provides for it in advance, because an insurer that cannot pay must not be allowed to fail the way an ordinary company fails. A book of life policies is a stack of promises running decades into the future; its value in a transfer is the reason the Act’s amalgamation and transfer provisions require the Authority’s sanction for any scheme that moves policies from one insurer to another, with the scheme examined for what it does to the policyholders being moved. Winding up stands behind transfer as the last resort, and Section 64VA’s deeming provision, an insurer below the statutory floor is deemed insolvent for winding up purposes, exists so that the end can be orderly and early rather than chaotic and late. The design preference runs through the whole part: correction first, transfer second, liquidation last, and policyholders ranked as the interest every stage must answer to.

Expenses are policed on the same principle. Section 40 restricts payments by way of commission or remuneration for soliciting business to agents and intermediaries entitled to receive them, and Sections 40B and 40C place the expenses of management of life and general business within limits the Authority prescribes. The detail sits in expense of management regulations, but the statutory peg is what matters: Parliament long ago decided that the share of the premium rupee consumed by the machinery of selling and administering insurance is a policyholder protection question, not a private bargain between insurer and shareholder.

Is there a ceiling on what an insurer may spend on commission and management?

Yes in structure, though the figures live below the statute. The Act confines commission to persons lawfully entitled to receive it and requires expenses of management to stay within limits the Authority prescribes by regulation, framed separately for life and for general business.

An insurer that exceeds the limits faces regulatory consequences, including restrictions on how the excess is borne; in broad terms, costs above the ceiling cannot be loaded onto the policyholders’ fund and must fall on shareholders, which aligns the penalty with the party that chose the spending.

No premium, no risk: Section 64VB

One short provision reverses the instinct of every credit based market. Section 64VB provides that no insurer shall assume any risk in India in respect of any insurance business on which premium is ordinarily payable until the premium payable is received by the insurer or is guaranteed to be paid in the prescribed manner, or a deposit of the prescribed amount is made in advance. The provision was inserted in 1968, and its target was the practice, familiar in the tariff era, of cover notes issued on trust and premiums collected late or never. The rule is strict but not mindless, and its exceptions mark the boundary of the mischief it aimed at. Premium may be taken as constructively paid where it is guaranteed in the manner the rules prescribe, and government business has its own accommodation within the scheme, because the vice the section attacks is unsecured private credit extended by habit and favour, not the settled payment machinery of the state or a bank’s guarantee standing in the premium’s place. What the section will not tolerate is the informal understanding, the agent’s assurance that the money will follow, the cover note issued as an act of salesmanship. In a market built on the pooling of other people’s money, Parliament decided that an insurer’s generosity with risk it has not been paid for is a form of spending the pool’s funds without authority, and Section 64VB is the minute of that decision.

The rule does three kinds of work. It protects the insurer’s pool: a claim cannot arise under a risk the pool was never paid to carry, which matters to every other policyholder whose premiums fund that claim. It protects the policyholder against a subtler loss, the discovery after a fire or a death that the cover the family believed in was never perfected because an intermediary held the premium. And it disciplines intermediaries, who must pass premiums on rather than finance their business with clients’ money. The section is applied with equal rigour in reverse: where the insurer or its agent has in fact received the premium, the insurer cannot hide behind an internal failure to record receipt. The general position stated here is information about the statute, not advice on an individual case; in a disputed claim the dates of payment and receipt are exactly what decide the matter, and they deserve professional reading.

The litigated edge of the section is payment by cheque. Where a proposer hands a cheque to the insurer or its agent and cover is issued, and the loss occurs before the cheque is presented, the courts have treated acceptance of the cheque as receipt for the purposes of the risk, with the bargain unwinding if the cheque is dishonoured; where no cover was issued and the cheque was never the insurer’s, the section protects the insurer. The thread running through the decisions is the section’s own logic: risk follows money the insurer has taken into its hands, in whatever instrument, and does not follow money still in the customer’s hands. Premium paid to an agent authorised to collect it is treated as paid to the insurer, which is why the identity of the person who took the money, agent, broker or stranger, features in nearly every Section 64VB dispute.

The policy as property: assignment and nomination

A life policy is a chose in action, money owed by the insurer on a future event, and the Act treats it as property that can be dealt with. Section 38 governs assignment and transfer. A policyholder may assign a policy, wholly or in part, with or without consideration, by an endorsement on the policy itself or by a separate instrument, signed and attested; notice in writing must be given to the insurer, which records the assignment, and the assignee then holds the rights the policyholder had, subject to the equities between the parties. The amendment of 2015 modernised the section in two ways: it required the insurer to accept and act on assignments rather than sit on them, and it confined refusal of registration to stated grounds, principally that the assignment is not bona fide, is against public interest, or is for trading in policies. A bank that lends against a policy, an employer holding a keyman policy, a parent settling a policy on a child: each transaction runs through Section 38.

A neighbouring device should be distinguished before the nomination discussion, because families confuse them. Group insurance, the master policy under which an employer or an association covers its members, is not assignment and not nomination: the group policyholder holds the contract and the member holds a certificate and a beneficial interest defined by the scheme’s rules, with nomination for the member’s benefit operating inside the scheme. The master policy model lets the Act’s classes do welfare work at scale, employer cover, government scheme cover routed through insurers, without individual underwriting of each life, and the group mechanism’s terms sit in the filed scheme and the regulations rather than in Sections 38 and 39. The distinction matters on death: a group member’s family claims under the scheme’s nomination and the master policy, not under a personal policy the member never held.

Nomination under Section 39 is the mirror device, and the distinction is the one policyholders most often blur. An assignment transfers the policy during the policyholder’s lifetime; a nomination merely names the person to whom the insurer should pay when the policyholder’s death makes the money payable, and it can be changed or cancelled at any time before maturity. The insurer that pays the nominee in accordance with the nomination obtains a valid discharge under Section 39, which is the practical point of the machinery: the insurer pays the right person without adjudicating the family’s succession dispute.

For most of the Act’s life, the nominee’s own right was small. In Sarbati Devi v. Usha Devi the Supreme Court held that a nomination under the unamended section did not make the nominee the owner of the proceeds: the nominee received the money as a trustee or collector for those entitled under the law of succession, and the Act was not a third mode of succession alongside a will and intestacy. The Court’s reasoning rested on the section as it then stood, which gave the nominee a right to receive and the insurer a discharge, and said nothing about beneficial ownership.

The amendment of 2015 changed the text on which that reasoning rested. Section 39(7), as substituted, provides that where the policyholder nominates his parents, or his spouse, or his children, or his spouse and children, or any of them, the nominees are beneficially entitled to the amount payable, unless it is proved that the policyholder, having regard to the nature of his title to the policy, could not have conferred a beneficial title. Section 39(8) carries the entitlement forward: if such a nominee dies after the insured but before the money is paid, the nominee’s heirs or legal representatives take the beneficial interest as part of the nominee’s estate. Two safeguards frame the gift. Creditors’ rights are expressly preserved, so a beneficial nomination does not put the proceeds beyond the reach of a policyholder’s creditors. And the beneficial provisions apply to policies maturing for payment after the commencement of the 2015 amendment, leaving the older law to govern older maturities. Policies under a trust of the Married Women’s Property Act type are carved out of the nomination scheme by Section 39(12), since the trust, and not the nomination, governs them. The contrast is worth drawing, because it shows two centuries of Indian law solving the same anxiety by different machinery. A policy taken under Section 6 of the Married Women’s Property Act, 1874 creates a statutory trust for the wife and children at the moment of the policy: the money never forms part of the husband’s estate and is beyond his creditors, a stronger protection than any nomination gives, purchased at the price of rigidity, since the policyholder cannot later redirect it. Nomination under Section 39 is the flexible device, changeable at will, its beneficial form confined to the family circle, its protection against creditors expressly withheld. Advisers who treat the two as interchangeable misunderstand both, and the 1938 draftsmen, by carving the trust policies out of the nomination code, showed they understood the difference.

The result is a two track law of nomination. A nomination of a stranger or a non family holder still operates on the Sarbati Devi principle: the nominee collects for the estate. A nomination within the stated family circle transfers beneficial ownership itself, unless the policyholder’s own title to the policy was so limited that he could not give that ownership away. Families drafting nominations since 2015 have been making succession decisions whether they knew it or not, which is why the section deserves its place among the Act’s most consequential provisions.

Two practical details complete the picture. A nomination is made by incorporating it in the policy or by later endorsement, is communicated to and recorded by the insurer, and may name more than one nominee in stated shares; where a nominee is a minor, the policyholder appoints a person to receive the money on the minor’s behalf if death occurs during the minority. And the beneficial nominee rule has an unsettled edge that families should know exists: the Supreme Court has not pronounced on the amended section, and how a beneficial nomination interacts with a later will, or with the claims of heirs who say the policy was bought with joint family funds, is ground on which the older trustee doctrine is still argued. The proceeds are also not beyond the law’s ordinary reach in the meantime: Section 39(9) preserves the rights of creditors, so a beneficial nomination is a succession device, not a shelter from debt.

One interaction between the two devices deserves a paragraph of its own, because it decides real claims. As a rule, an assignment cancels the nomination it overtakes: the assignee takes the policy as owner, and the nominee’s expectancy falls with the policyholder’s title. The Act makes one humane exception for the commonest security transaction of all: where the policy is assigned to the insurer itself as security for a loan on the policy, the nomination is not extinguished, and on repayment of the loan and reassignment of the policy the nominee’s position revives. The exception recognises that a policy loan is an incident of the policy’s own life, not a change of its destination. Its absence would have meant that a loan taken and repaid in middle life could silently disinherit the nominee of the borrower’s old age, a trap the draftsmen of the amendment chose to disarm in terms.

Section 45 and the three year line

Every proposal for life insurance is a negotiation conducted on unequal information. The proposer knows his own body, habits and history; the insurer knows the statistics of thousands of lives. Insurance law absorbs that asymmetry through the duty of utmost good faith, which requires the proposer to disclose every fact material to the risk. In Satwant Kaur Sandhu v. New India Assurance Company the Supreme Court described a material fact as any fact which would influence the judgment of a prudent insurer in fixing the premium or determining whether to take the risk. A proposer who suppresses such a fact gives the insurer ground to repudiate. Against that stands the family’s position: premiums are paid for decades, the proposer is dead when the claim is tested, and an insurer hunting a file for an error after the event can defeat the very security the policy was bought to create.

Section 45 is the statute’s resolution of that tension, and it works by time rather than by balancing. As substituted in 2015, its first sub-section states the bar in unqualified terms: no policy of life insurance shall be called in question on any ground whatsoever after the expiry of three years from the date of the policy, that is, from the date of issuance of the policy or the date of commencement of risk or the date of revival of the policy or the date of the rider to the policy, whichever is later. The phrase on any ground whatsoever does the decisive work. Fraud, the ground insurance law usually treats as dissolving everything, is inside the bar: a policy that has crossed the line cannot be questioned even by an insurer who can prove deliberate deception in the proposal.

From which date does the three year period in Section 45 run?

The period runs from the last of four dates to occur: the date the policy was issued, the date the risk under it commenced, the date the policy was revived after a lapse, and the date a rider was added to it. Count from whichever falls last, not from the first premium.

A revival or a later rider starts a fresh three years for the matters it brings into the contract, because the insurer has underwritten again on a fresh declaration; the third illustrative case below turns entirely on that sentence.

The choice of the last occurring date was not decorative. Life insurers have long used backdating, setting the commencement of risk a few months before the policy is actually issued so that the life assured falls within a younger age band and a lower premium. Without the words whichever is later, backdating would quietly shorten the insurer’s own contestable window while appearing to benefit the customer, and an insurer could find its three years partly spent before the proposal was even signed. The counting rule forecloses the game in both directions: whichever date falls last governs, issuer’s convenience and customer’s alike. The same rule governs the revival trap in the third illustrative case below, and it is the first thing to verify in any repudiation, because an insurer that counts from the wrong date may be outside its window without knowing it.

Within the three years, the section permits challenge on exactly two grounds and surrounds each with conditions. The first ground is fraud. An insurer invoking it must communicate in writing to the insured or to the legal representatives or nominees or assignees the grounds and materials on which the decision rests; a repudiation announced without that statement does not satisfy the section. The section defines fraud in terms drawn from the general law of contract, acts committed with intent to deceive, while providing that mere silence about a fact is not fraud unless the circumstances imposed a duty to speak or the silence was itself equivalent to speech. The second ground is misstatement: that a statement made in the proposal or in documents required for the policy was inaccurate or false in respect of a fact material to the expectancy of the life of the insured. Materiality under this limb is demanding: a fact counts as material only if it has a direct bearing on the risk undertaken, and the onus lies on the insurer to show that, had it known the fact, no policy would have been issued on any terms.

The insured’s side of the ledger is stated in the section as well. No repudiation lies on the fraud ground if the insured proves that the misstatement or suppression was true to the best of his knowledge and belief, or that there was no deliberate intention to suppress the fact, or that the fact was within the insurer’s own knowledge, as where its agent had been told. One asymmetry is deliberate and should be seen plainly: where the policyholder is not alive, the proviso to the fraud provisions places the onus of disproving the fraud on the beneficiaries, which is a hard burden for a widow or child to carry and the aspect of the 2015 settlement most open to criticism. Where a policy is repudiated on the misstatement ground alone, the section is generous in the other direction: the insurer shall pay back the premiums collected under the policy within ninety days. And the section closes one familiar door: the insurer may still require proof of age at any time and adjust the terms of the policy on the basis of the correct age, and doing so is not a calling in question.

The table below states the whole design at a glance, and it is the findable centre of this article.

The three year line: what an insurer may and may not do under Section 45

Question	Within three years of the last start date to occur	After three years have run
May the insurer call the policy in question at all?	Yes, on the two grounds the section allows	No, on any ground whatsoever
What grounds are available?	Fraud, and misstatement or suppression of a fact material to the expectancy of the life	None; proved fraud is inside the bar
What must the insurer communicate?	The grounds and the materials behind the decision, in writing, to the insured or the representatives, nominees or assignees	Nothing can cure the lateness of the challenge
Who carries the burden?	The insurer must prove fraud, or that the fact was so material no policy would have been issued; if the policyholder is dead, beneficiaries must disprove fraud	The question does not arise
What happens to the money on a successful challenge?	The policy is repudiated; on the misstatement ground alone, premiums collected must be paid back within ninety days	The claim must be paid; the insurer may still call for proof of age and adjust terms to the correct age
When is the date counted from?	The last of issuance, commencement of risk, revival or rider to occur	The same counting rule, completed

One question of reach must be answered before the comparison with the old law, because it decides live claims. Policies issued before the amendment of 2015 are governed by the substituted section where the claim arises after the amendment commenced: the right the section restricts is the right to call a policy in question, and that right is exercised, if at all, when the claim is repudiated, so the Authority clarified and the courts have proceeded on the footing that the new three year bar governs repudiations after 26 December 2014 whatever the vintage of the policy. The transition mattered enormously in practice, since the portfolio of 2015 was written entirely under the old law; overnight, every policy more than three years from its last starting date to occur acquired the absolute protection, and the stock of contestable old claims shrank to those within three years of issue, revival or rider.

The bar of 2015 replaced a weaker protection and the comparison shows what Parliament intended. The original Section 45 ran for two years, and after that period a policy could still be called in question if the insurer showed that a statement was on a material matter, was fraudulently made, and was known by the policyholder to be false. Those three requirements, restated in Mithoolal Nayak v. Life Insurance Corporation and applied in LIC v. Asha Goel, made late repudiation difficult but not impossible, and the insurer’s burden was the litigation battleground for half a century. The 2015 substitution lengthened the period to three years and then made the post period position absolute. Fraud stopped being an answer to the passage of time.

What must an insurer put in writing before it repudiates inside the three years?

The insurer must state, in writing, both the grounds on which it repudiates and the materials on which the decision is based, and must send that statement to the insured or to the legal representatives, nominees or assignees of the insured.

A repudiation letter that announces a conclusion without disclosing the grounds and the material relied on fails the condition the section attaches to its own power, which is why claimants should always demand the full statement before answering it.

The old law at work: two years and three conditions

To see what the 2015 substitution changed in litigation practice, the old section has to be understood on its own terms. Before 2015 the shield period was two years, and it was a shield with three gates in it. After two years a policy could still be called in question if the insurer showed that a statement in the proposal or in the medical and other documents was on a material matter or that facts material to the risk had been suppressed, that the statement or suppression was fraudulently made, and that the policyholder knew at the time that the statement was false or that the facts were material. The three conditions were cumulative, and the burden of each lay on the insurer.

The Supreme Court structured that burden in Mithoolal Nayak v. Life Insurance Corporation of India, the foundational reading of the old section, insisting that an insurer seeking to escape a policy after the period had to prove all three elements strictly, since the consequence of success was to strip a dead man’s family of the claim. LIC v. Asha Goel applied the same discipline half a century later, with the Court pointedly observing that repudiation is not a mechanical exercise and that the insurer’s satisfaction about materiality and fraud must be real. In between and around those poles, the case law built the practical law of the early claim: insurers investigated deaths occurring soon after issue as a matter of routine, repudiations rested most often on undisclosed illness and undisclosed earlier policies, and the courts tested each against the three conditions. Reliance Life Insurance Company v. Rekhaben Nareshbhai Rathod, decided on the old law even in 2019 because the policy predated the substitution, confirmed that the duty of disclosure reaches the proposer’s other policies, whose existence a prudent insurer plainly weighs. P. C. Chacko v. Chairman, Life Insurance Corporation upheld a repudiation where earlier treatment had been suppressed. The litigation pattern was consistent: the insurer’s case was available but never easy, and everything depended on proof of what the proposer knew. The standard of that proof shaped outcomes as surely as the conditions themselves. Fraud in insurance litigation is proved on the civil standard, the balance of probabilities, but courts demand evidence of a cogency proportioned to the gravity of the allegation, and an insurer that pleaded fraud on suspicion, a thin file and a convenient inference lost cases it might have won on the misstatement limb, where the question is the narrower one of what a prudent insurer would have done with the truth. Skilled claims departments learned to plead the two limbs in the alternative and to build the materiality case on underwriting evidence rather than assertion: the manual that would have rated the risk up, the guideline that would have declined it. The 2015 text preserves that craft inside the three years, in the insurer’s duty to state its grounds and materials, and abolishes its use outside them.

The 2015 substitution must be read against that record. Parliament did not conclude that the three conditions were wrongly decided; it concluded that the contest should have a closing date at all, and that the closing date should bind even a meritorious fraud case. Where the old law asked whether the insurer could prove enough, the amended law asks a single question of the calendar.

Two claims, one line

Three illustrative cases, stripped to their dates, show the line doing the work. In the first, a policy is issued in March 2017 on a proposal that conceals continuing treatment for heart disease; the policyholder dies in February 2019. The claim falls within three years of issuance, and the insurer may call the policy in question on the fraud ground if it states its grounds and materials in writing and proves deliberate concealment of a fact it would have weighed; the family’s answers are the three defences of Section 45(3), and if the policyholder’s concealment was deliberate, the repudiation will ordinarily stand.

In the second case the proposal, the concealment and the man are identical, but the death occurs in June 2021, more than three years after the risk commenced. The insurer that discovers the treatment records in its claim investigation has, in law, discovered nothing it can use: the policy shall not be called in question on any ground whatsoever, and the claim must be paid. The distance between the two outcomes is not a difference in desert. It is the statute’s judgment that at some point the insurer’s time to test the bargain has run out.

The third case is the one families most often misjudge. A policy issued in 2016 lapses in 2019 when premiums stop, is revived in January 2022 on a fresh declaration of health, and the policyholder dies in September 2023. The revival, not the original issue, is the last of the section’s four dates to occur, and the claim is within three years of it; the insurer may call the policy in question on the strength of the revival declaration, including a suppression in that declaration itself. A revival form is therefore not a formality. It is a fresh act of underwriting, and the three year line runs from it.

The case for the absolute bar is certainty where certainty matters most. A life policy is bought precisely against the event that leaves the family least able to fight: the earners’ death. If every policy remained defeasible forever, the product would carry a permanent asterisk, and the failure would fall on widows and children rather than on the proposer who made the error. The government that carried the 2015 amendment presented the three year bar, in its own summary of the Act, as one of the amendment’s principal consumer welfare measures, a point after which the insurer’s remedy is to investigate promptly rather than to investigate eventually. There is a commercial logic as well: a bright line is cheaper to administer than a fraud inquiry conducted decades after the proposal, when the agent is untraceable and the medical evidence is thin.

The case against the bar is that it rewards a lie that survives three years. Insurers and their associations argued through the long years in which the amendment bill was pending, from its first introduction in 2008 to the ordinance route of December 2014, that an incontestability rule without a fraud exception lets a proposer conceal a terminal diagnosis or a dangerous habit, pay three years of premiums, and bind the pool to a risk it never priced. Honest policyholders fund the resulting claims, so the bar transfers money from the truthful to the dishonest. The drafting answer available to Parliament was a fraud proviso, and Parliament declined it: the words on any ground whatsoever were a considered choice, made against the industry argument, in favour of finality. Readers should hold both parts of that sentence together. The bar is strong because it was meant to be, and the fraud it shelters is the price the statute consciously pays for it.

Holding a claim file together across those three years is unglamorous and decisive. The claimant’s best evidence is the proposal form copy, the premium receipts, the medical reports the insurer itself ordered, and every letter the insurer sent; where the file is held in office document and presentation formats, it can be reviewed without the original software through this online PowerPoint viewer, which a family assembling papers for a grievance will find easier than reinstalling a suite of programs. What the insurer knew, and when it knew it, is what the fraud and materiality tests turn on, and the file is where knowledge is proved.

The proposal form: utmost good faith in daily practice

Section 45 is the courtroom face of a duty that does its work earlier, at the proposal form. A life policy begins as a set of answers: age, occupation, income, habits, illnesses, treatments, other policies. The insurer prices the risk on those answers and on its own medical examination where one is held, and the contract that results incorporates the proposal as its basis. The duty of utmost good faith, which the common law of insurance imposes on both parties and with special weight on the proposer, is what makes those answers legally radioactive: an answer that misstates a fact material to the risk, or the silence that suppresses one, gives the insurer its Section 45 grounds if death comes inside the contestable period.

Two features of the machinery moderate that severity in practice, and both come from the statute rather than from judicial kindness. The first is the agent’s position. The explanation appended to Section 45 treats an insurance agent, for the formation of the contract, as the insurer’s agent, however the insurer’s internal paperwork describes him. The consequence is far reaching: a fact told to the agent is, for these purposes, told to the insurer, and an answer the agent filled in wrongly is a poor foundation for repudiating the widow. Claim litigation returns to this point constantly, because the agent who sold the policy to an illiterate or hurried proposer, and completed the form himself, is the rule rather than the exception in much of the market. The second moderating feature is the insurer’s own medical process. Where the insurer examined the proposer and passed him, its plea that an undisclosed condition was material faces the obvious question why its own examination, conducted for exactly this purpose, is worth less than its after the event reading of the file. Neither feature abolishes the duty of disclosure. Together they explain why Section 45 litigation is fact litigation, fought over who knew what and who wrote what, rather than over abstract principle.

Financial underwriting belongs to the same picture. An insurer may decline or limit cover that is out of proportion to the proposer’s income and obligations, not from prudery but because insurance that greatly exceeds the economic loss it replaces distorts the incentives the contract depends on. The Act does not spell out this underwriting discipline; it presupposes it, in the registration and solvency architecture, and leaves its content to the insurer’s filed underwriting policy under the Authority’s product regulations. The pillar’s silence here is typical: the Act fixes the points at which law will intervene, the incontestability line, the premium rule, the nomination and assignment code, and leaves the commercial craft of risk selection to the market it licenses.

Selling insurance lawfully: rebating, agents, intermediaries and surveyors

A statute that regulates promises must also regulate the selling of them, because the characteristic insurance scandal is a sale made by misdescription. Section 41 is the Act’s oldest conduct rule: it prohibits rebating. No person shall allow or offer to allow, directly or indirectly, as an inducement to take out or renew or continue a policy, any rebate of the whole or part of the commission payable or any rebate of the premium shown on the policy; nor shall the person taking the policy accept a rebate outside what the insurer’s published prospectuses or tables allow. The rule looks technical and aims at something simple: competition on the price printed in the policy rather than on secret discounts returned by the agent, which destroy the premium rate’s integrity and are the usual companion of a sale the buyer did not understand. The penalty for default under the section may extend to ten lakh rupees.

Section 41 does not stand alone in the economics of distribution. Section 40 confines payments of commission or remuneration for soliciting or procuring business to the agents and intermediaries the Act recognises, so that the premium rupee may not leak to unlicensed hands, and the ceilings on what distribution may cost have migrated over the Act’s life: detailed commission limits that once stood in the statute’s own text have been withdrawn into the Authority’s expense of management regulations, where they sit with the wider ceilings on what an insurer may spend to acquire and administer business. The direction of the migration matches the pattern this article has traced in solvency and in surveyor thresholds: Parliament fixed the prohibition and the entitlement, and left the price of distribution to the regulator who can watch it quarterly. What has not changed since 1938 is the premise that distribution cost is a policyholder protection subject, because every rupee of commission is a rupee of premium that will not be available to pay a claim.

Who may sell is fixed by Sections 42 and 42D. An insurance agent, the individual tied to one insurer’s products, must hold a licence under Section 42; an intermediary, a category the Act defines to include brokers, corporate agents, third party administrators and surveyors among others, is licensed or registered under the corresponding provisions; and an insurer that appoints or pays a person outside the permitted channel answers for it with a penalty that may reach one crore rupees, quite apart from the consequences for the unlicensed seller. The amendment of 2015 added a prohibition directed at a different abuse: insurance may not be promoted through multilevel marketing schemes, the pyramid style networks in which a policy becomes the token of a money circulation plan.

Surveyors occupy their own corner of the Act because they stand between the insurer and the loss. Section 64UM provides that surveyors and loss assessors are licensed by the Authority, that claims above a floor fixed by the regulations must be supported by a report from a licensed surveyor before the insurer settles, that the Authority may call for an additional independent report where it doubts the first, and that surveyors work under a code of conduct the regulations prescribe. The statutory floor was twenty thousand rupees in the old text; the amendment of 2015 replaced the figure with an amount the regulations specify, moving the threshold into subordinate legislation where it can be revised without amending the Act. The surveyor’s report is not the settlement: an insurer remains free to settle for more or less than the surveyor assesses, but it must be able to explain the difference, and in motor and fire claims the surveyor’s paper trail is usually the spine of any later dispute. The institution carries a built in tension that the Act manages rather than resolves. The surveyor is appointed and paid by the insurer whose liability he measures, and claimants have always suspected the arrangement; the licensing and disciplinary hand of the Authority, its power to call for a second and independent report, and the insured’s right under the policyholder protection regulations to receive the report, are the system’s answers. A surveyor who serves insurers carelessly loses claimants’ trust; one who serves claimants carelessly loses his licence. The equilibrium is imperfect and durable, like much of the Act.

The architecture of distribution the Act built has proved elastic enough to absorb channels its draftsmen never imagined. The tied agent of Section 42, licensed for one insurer, remains the largest channel in life insurance. Beside the agent stands the broker, an intermediary who represents the buyer rather than any insurer and who must answer for advice across the market; the corporate agent, a company or bank distributing for insurers from its own counters, which is the statutory form behind bancassurance; and the third party administrator, which services health claims between insurer and hospital under the intermediary provisions. Web aggregators and insurance marketing firms arrived by regulation within the Act’s definition of intermediary rather than by amendment of the definition itself. The pattern is the Act’s characteristic one: a spare statutory category, filled in over time by subordinate law, so that a channel invented in the internet era holds its licence under a statute passed when insurance was sold from brass plate offices.

One statute, four markets: the general classes

The pillar is usually read for its life provisions, but the Act governs four markets, and its definition section separates them with care. Fire insurance business, marine insurance business and miscellaneous insurance business are each defined classes, and the miscellaneous class carries the long tail of modern general insurance: motor, liability, engineering, aviation, crop and health among them. Crop insurance shows the class at its most public: the national crop scheme is delivered not by a government department but by empanelled insurers writing the miscellaneous class under the Act, so that a statute of 1938 supplies the legal chassis for a welfare programme its draftsmen could not have pictured, the state buying capacity in the market it licenses rather than building an insurer of its own. Registration, capital, investment patterns and expense ceilings are stated class by class, which is why the same statute can impose a 50 per cent government securities pattern on life funds and a 30 per cent pattern on general funds without contradiction: the promises differ in length and in kind.

The newest class illustrates the Act’s method of absorbing change. Health insurance was written for decades inside the miscellaneous class by general insurers, although its economics, indemnity against hospital bills, annual repricing, a claims relationship with hospitals rather than with the insured alone, resembled life insurance in none of those respects. The system first recognised the difference in subordinate legislation and in the registration of standalone health insurers, and the amendment of 2015 wrote the recognition into the Act itself, making health insurance business a distinct class with its own capital figure in Section 6. The statute of 1938 thus acquired, by amendment, a class of business its draftsmen had filed under miscellaneous, without any disturbance to its architecture. A code that can admit a new market by inserting a definition and a figure is a code built to last, and the health class is the cleanest recent proof of that property.

The miscellaneous class holds the Act’s most socialised obligation, and it deserves notice before the monopoly narrative resumes. Motor third party insurance is compulsory under the Motor Vehicles legislation, which means the general insurers are conscripted as the carriers of a welfare promise: every vehicle must be covered for the death and injury it may cause, whether or not the risk is attractive. The Insurance Act answers the conscription with the minimum underwriting obligation in Section 32D and, in the regulatory machinery beneath it, has at times answered it with pooling arrangements for risks no insurer would take voluntarily, including a declined risk pool for commercial vehicles whose record made them uninsurable at tariff prices. The pairing is characteristic of the statute: a parliamentary command in the Motor Vehicles Act, an Insurance Act obligation to keep the command supplied with capacity, and regulation to share the resulting losses across the market. General insurance in India is a private industry carrying statutory duties of a public kind, and the third party line is where that bargain is struck hardest.

The monopoly decades: 1956, 1972 and what they did to the Act

The Act of 1938 was written for a competitive market and spent its middle decades governing a monopoly. In 1956 the Government of India nationalised life insurance. The decision, announced by the government of the day as a step to spread insurance widely, in particular to rural areas, to give the policyholder absolute security and to mobilise public savings more effectively, was executed through the Life Insurance Corporation Act, 1956: 245 Indian and foreign insurers were merged into the Life Insurance Corporation of India, which began business on 1 September 1956 as the sole lawful seller of life insurance. The Insurance Act did not disappear; the LIC Act applied many of its provisions to the Corporation with modifications, Section 45 among them, so the discipline of the private market statute crossed into the monopoly.

General insurance followed in 1972 under the General Insurance Business (Nationalisation) Act. The private general insurers were taken into public ownership and grouped in time into four public sector companies under a holding structure, and the Insurance Act continued as their operating code, administered by the Controller of Insurance and overlaid with tariffs fixed through the Tariff Advisory Committee. For a generation, the Act’s competitive provisions, its entry gate, its commission rules, operated in a market with no entrants to gate and little price competition to police.

One institution of the monopoly era survives inside the Act and is worth naming, because it shows the statute carrying a whole pricing regime. The Tariff Advisory Committee, constituted under Section 64U, was the body that fixed tariff rates for general insurance business: fire, engineering and much of motor insurance were sold at prices the Committee set, so that insurers competed, to the extent they competed at all, on service rather than price. Tariffing was defended as protection against the ruinous rate cutting that had destroyed insurers before 1938 and attacked as a cartel with statutory force. The Authority progressively detariffed the market from 2007, leaving insurers to file their own prices under regulatory oversight, and the Committee’s rates survive in binding form mainly where Parliament and the regulator still treat price as a public question, above all motor third party cover. The section that housed the tariff sits quietly inside a market statute, which is the usual fate of monopoly machinery in a code that outlives the monopoly.

The structures nationalisation built are worth one paragraph of detail, because the Act was administered through them for a generation. The Life Insurance Corporation worked as a single national corporation with zonal and divisional offices, its monopoly protected by the LIC Act and its discipline imported from the Insurance Act by the application device of Section 43 of that Act. In general insurance, the nationalised companies were grouped under the General Insurance Corporation as a holding company with four operating subsidiaries, National Insurance, New India Assurance, Oriental Insurance and United India Insurance, whose names became the market’s geography for thirty years. Tariffs fixed the prices, the Controller supervised the forms, and the Insurance Act’s provisions on commission, expenses and investments applied to the public insurers as they had to the private market, a point worth making against the lazy assumption that nationalisation suspended the statute. It did not. It changed the shareholders.

The monopoly decades are the necessary background to every later argument in this article. Nationalisation was carried on real grievances: private insurers had failed, funds had been misused, and cover barely reached beyond the cities. Its cost was the cost of all monopolies: service standards with no rival to shame them, products that changed slowly, and penetration of insurance in national income that stayed low for decades. When reformers of the nineteen nineties argued for reopening the sector, they were arguing against the lived record of the monopoly the 1938 Act had been made to serve; when defenders of the public insurers resisted foreign capital later, they were arguing from the same record in the opposite direction. The statute sat impartially beneath both positions, which is a comment on statutes rather than on either side.

Reopening the sector: Malhotra, the 1999 Act and the 26 per cent settlement

Reform began with a committee. The Committee on Reforms in the Insurance Sector, appointed in April 1993 under R. N. Malhotra, a former Governor of the Reserve Bank of India, reported in January 1994 and recommended, in essence, that the sector be reopened to private companies, that foreign insurers enter alongside Indian partners rather than alone, that the LIC and the general insurance holding structure be restructured, and that an autonomous statutory authority replace the Controller. The report supplied the blueprint, and the wider story of the era, the unmaking of the licence permit system across the Indian economy, is traced in the companion article on the passage from the licence raj to liberalisation.

The committee’s wider recommendations explain the settlement that eventually emerged. It proposed that the Life Insurance Corporation be restructured and the government’s holding in the public insurers be brought down in time towards 50 per cent; that the General Insurance Corporation be converted into a reinsurer with its subsidiaries made independent companies; that brokers be admitted as a channel representing the buyer; that foreign insurers enter through joint ventures with Indian partners rather than as branches; and that the business be supervised by an autonomous authority with statutory powers over licensing, investments and tariffs. Nearly every element of that list became law or practice within a decade, which is why the report is cited in this article at headline level but with respect: it is one of the few committee reports in Indian economic legislation whose table of contents the statute book eventually reproduced. An interim regulator was set up by executive action in 1996 while the legislative argument ran its course, so the Authority that the Act of 1999 created had a working predecessor and a settled agenda.

Legislation took the rest of the decade. The Insurance Regulatory and Development Authority Act, 1999, Act 41 of 1999, created the Authority, ended the statutory monopolies by amending the LIC Act, the nationalisation Act and the Insurance Act itself, and substituted the Authority for the government and the Controller across the 1938 text. The Authority was incorporated as a statutory body in April 2000 and the first private registrations followed. Its constitution, powers and accountability are the subject of the companion article on the IRDAI Act 1999 explained, which also owns the insurance ombudsman; here it is enough to record that from 2000 the 1938 Act is administered by a specialist regulator rather than by a department of government.

The ownership price of reopening was set at 26 per cent. Foreign investment in an Indian insurance company was capped at 26 per cent of its capital, a figure chosen in the politics of the time as high enough to attract foreign insurers’ capital and expertise and low enough to keep control unmistakably Indian. Joint ventures between Indian houses and foreign insurers formed across life, general and health lines, and the market the Act governed changed character within a few years: private insurers took share from the incumbents, distribution professionalised, and the Act’s dormant provisions on agents, capital and solvency began to do daily work. But the 26 per cent figure also embedded the argument that has never since closed, because every subsequent increase had to be legislated against the charge that the original settlement had promised control would stay Indian.

The register the gate admitted changed the market faster than the monopolies’ defenders expected and more unevenly than the reformers promised. Private life insurers took registration through the first decade, then general insurers, then the standalone health companies the class system made possible. Distribution was rebuilt around the licensed agent and the corporate agent bank; products changed character, from the endowment plans of the monopoly years towards unit linked plans in the market’s boom phase and, after the correction of 2010 and the regulatory tightening that followed, towards protection products whose term premiums fell sharply as mortality was priced by competing underwriters. The Authority’s annual publication of claim settlement figures, insurer by insurer, introduced a discipline the Act’s framers would have recognised as their own idea in modern dress: comparability as supervision. The LIC, competing for the first time in four decades, retained the largest single share of the life market, which the defenders of nationalisation read as proof of public trust and its critics read as the residue of incumbency; the Act, characteristically, records only that the same sections came to govern all of them.

The amendment of 2015: the modern hinge

The Insurance Laws (Amendment) Act, 2015, Act 5 of 2015, is the largest single revision the statute has received. A bill to amend the insurance laws had been introduced in 2008 and had not completed its passage by 2014; the government took the ordinance route on 26 December 2014, and the Act, assented to in March 2015, operates from that December date. Its contents rearrange several parts of this article at once.

On ownership, it raised the foreign investment ceiling in an Indian insurance company from 26 per cent to 49 per cent, coupling the increase with the condition that the company remain Indian owned and controlled. On policyholder rights, it substituted Section 45 with the absolute three year bar and substituted Section 39 with the beneficial nominee rule, and it reformed Section 38 on assignment. On the prudential side, it omitted the deposit sections, recognised health insurance as a distinct class with its own capital, admitted foreign reinsurers to operate through Indian branches, and made registration continuing rather than a matter of periodic renewal. On enforcement, it raised the Act’s penalties into a band running from one crore rupees to twenty five crore rupees for the graver defaults, created an adjudication mechanism under Section 105C before an officer of senior rank, and sent appeals from the Authority’s orders to the Securities Appellate Tribunal. It widened the definition of insurance intermediary to gather the new distribution channels, brokers, corporate agents, web aggregators and their successors, into the licensing net under a single statutory description, an apparently technical change that mattered because conduct regulation can only reach what the definition catches. It also wrote fresh social obligations into the Act, Sections 32B to 32D, requiring insurers to write stated proportions of business in rural areas and in the social sector, and requiring every general insurer other than those the Authority exempts to underwrite a minimum percentage of motor third party insurance, the compulsory cover that the Motor Vehicles Act regime depends on. Section 105B backs those three obligations with a penalty that may extend to twenty five crore rupees.

Two of its quieter contents deserve record. The amendment reached beyond the Insurance Act to the nationalisation statute, allowing the General Insurance Corporation and the public sector general insurers to raise capital from the market with the government’s holding kept at not less than 51 per cent, the same staged logic of dilution under retained control that the amendment applied to foreign capital. And it legislated against the ordinance criticism it had itself attracted: taking the ordinance route for a measure pending since 2008 drew objection in Parliament and in public commentary that a bill of this weight deserved the full legislative process, an objection the government answered by pointing to the years the bill had already spent unpassed. The episode belongs in this article because it explains the Act’s texture: the 2015 text is a negotiated settlement between a decade of committee scrutiny and a single session of decision, and its compromises, control conditions on foreign capital among them, carry the marks of both.

Read as a whole, the 2015 Act is the moment the 1938 statute stopped being a consolidating code with amendments and became the deliberate chassis of a mixed market: foreign capital admitted further under Indian control, policyholder finality strengthened against insurers, penalties repriced for companies of modern scale, and social obligations imposed as the price of the licence. Every debate since, including the foreign investment debate that resumed within six years, has been conducted inside the frame the 2015 Act built.

Foreign investment: the numbers and the argument

The sequence is short enough to state in one paragraph and has occupied Parliament for a quarter of a century. The 1999 settlement capped foreign holding at 26 per cent. The Act of 2015 raised the ceiling to 49 per cent and attached the Indian owned and controlled condition. The Insurance (Amendment) Act, 2021, Act 6 of 2021, in force from 1 April 2021, raised the ceiling to 74 per cent, substituted the relevant limb of the definition of an Indian insurance company, removed the owned and controlled condition, and substituted safeguards of a different design. The Sabka Bima Sabki Raksha (Amendment of Insurance Laws) Act, 2025, passed by Parliament in December 2025 with most of its provisions in force from 5 February 2026, raised the ceiling to 100 per cent, completing the sequence. One exception runs through the whole story: foreign investment in the Life Insurance Corporation has its own lower ceiling of 20 per cent under the foreign investment policy, untouched by the general figures.

The exception is principled rather than accidental, and its reason sits in the LIC Act rather than in investment policy. The sums assured by the Corporation’s policies are guaranteed by the Central Government under Section 37 of that Act: the Indian state stands behind every LIC promise, a sovereign guarantee no private insurer carries. An institution whose liabilities the taxpayer guarantees is not an ordinary candidate for majority private, let alone foreign, ownership, and the Corporation’s path has accordingly been one of small dilutions under retained state control, including its public listing, while the general ceiling moved to the whole of the capital. The juxtaposition is the cleanest statement of the settlement the sequence reaches: ownership may go where it likes in the market the state regulates, and stays where it is in the institution the state guarantees.

The middle step of that sequence took seventeen years, and the delay is itself evidence in the argument. A rise to 49 per cent was announced as government policy as early as the Budget of 2004, and an amendment bill carrying it was introduced in 2008; the measure could not complete its passage for a decade, blocked by the same control argument this article examines below, pressed inside Parliament by parties of the left and by members protective of the public insurers. When the increase finally came in 2015, it came with the owned and controlled condition as the drafted price of the votes that carried it. The six years between the 49 per cent ceiling and the 74 per cent ceiling, and the four years from 74 per cent to the whole of the capital, show the argument’s centre of gravity moving: each later step was debated as a question of conditions, in boards and reserves, rather than as the threshold question of whether foreign control was acceptable at all.

The safeguards at the 74 per cent stage show how the case against higher foreign holding was answered in drafting rather than in rhetoric. Under the conditions prescribed when the 2021 increase took effect, an insurer with foreign investment was to have a majority of its directors and a majority of its key management persons as resident Indian citizens, with at least one of the chairperson, the managing director and the chief executive officer a resident Indian citizen. Where foreign investment exceeded 49 per cent, two further conditions bit: at least half the board was to consist of independent directors, relaxed to one third where the chairperson was independent, and if the insurer declared a dividend while its solvency stood below one and one fifth times the control level, at least 50 per cent of its net profit was to be retained in a general reserve. The 2025 Act relaxed the governance conditions again, leaving the requirement that at least one among the chairperson, the managing director and the chief executive officer be a resident Indian citizen. The direction of travel is plain: as the ceiling rose, the safeguards thinned from ownership, to ownership plus governance, to governance, to a single residency peg.

What the ceilings did on the ground is visible in the joint ventures’ own history. Under the 26 per cent ceiling the foreign partner was structurally a minority technology and brand supplier, and the long refusal of Parliament to move to 49 per cent produced visible frustration: New York Life sold out of its Indian life joint venture in 2012 and ING left its life venture the following year, exits explained publicly in the language of global strategy and read in the market as verdicts on the ceiling. When the ceiling rose to 49 per cent in 2015, several partners bought up to it; when it rose to 74 per cent in 2021, the pattern repeated, with Ageas raising its holding in its Indian life venture towards the new ceiling and Nippon Life moving to raise its stake in its Indian life joint venture likewise. Capital followed the ceiling at each step, which is the supporters’ best evidence; the ventures also became, at each step, less Indian in their share registers, which is the critics’ best evidence. Both readings are descriptions of the same transactions.

The case for the increases, made by the government at each stage and supported by the industry, is a case about capital and penetration. Insurance is a business that consumes capital as it grows, since every new policy must be reserved against from the first day; the Finance Minister’s budget statements for the increases of 2021 and 2025 argued in substance that Indian promoters alone could not supply the capital the sector’s growth required, and that foreign capital at majority levels would deepen a market whose penetration of national income has persistently lagged comparable economies. Insurers publicly welcomed the 2021 increase on exactly that reasoning, that a 74 per cent ceiling would unlock investment the 49 per cent ceiling had kept waiting. On this view the ceiling was never a symbol; it was a tap, and each turn of it was calibrated to the capital the sector could absorb.

The case against, pressed by the organised employees of the public sector insurers, prominently the All India Insurance Employees’ Association, and by members across the opposition benches at each stage, is a case about what insurance money is. Premiums are not ordinary investment funds: they are savings entrusted against death, illness and accident, accumulated over decades, and held in the meantime by whoever controls the insurer. Majority foreign control, on this argument, puts long term Indian household savings under boards answerable elsewhere, with profits, and eventually strategic choices about which risks to write and which to leave, flowing out of the domestic economy. The argument does not deny that capital is needed; it denies that control is a fair price for it, and it points to the public insurers’ record of carrying social sector and rural business as evidence of what ownership decides.

A fact both sides must accommodate is that insurance penetration in India, premiums as a share of national income, has remained in low single digits across the whole period of the argument, under the monopoly, under the 26 per cent ceiling and under 49 per cent alike. Supporters of each increase read the figure as proof the previous ceiling was binding: capital was the constraint, and each rise was followed by investment. Critics read the same figure as proof the ceiling was never the binding constraint at all, that penetration follows incomes, distribution reach and trust in claims payment, none of which a share register supplies. The figures cannot settle the dispute because the counterfactual, the market at the old ceiling for the same years, does not exist. What the record does show is the direction of the state’s own revealed preference across four decades and governments of different colours: confronted with the choice, every administration since 1999 has raised the ceiling or held it, and none has lowered it.

The honest verdict on the sequence is that both cases were right about the mechanism and the statute kept choosing between the values. The capital did come at each stage, and penetration arguments were not invented; the control concern was equally real, and each Act answered it with conditions rather than dismissal, the reserve retention rule of 2021 being the most inventive example, a safeguard aimed at the precise fear, that a foreign controlled insurer might distribute profits while its margin thinned. With the ceiling at 100 per cent, the argument moves to ground the Act has always owned: whether capital adequacy, investment rules and supervision can do the whole work of protecting policyholders regardless of who owns the insurer. That is, in the end, a question about the strength of Sections 6, 27 and 64VA, and the statute of 1938 is where the answer will be written.

Supervision, penalties and the consumer’s parallel routes

The fourth machine, supervision, runs on Sections 33 and 34 and their neighbours. The Authority may investigate the affairs of an insurer, calling for its books and examining its officers, and may give directions where the insurer’s conduct prejudices policyholders or the public interest; it may call for returns, appoint auditors and actuaries to special inquiries, and act against insurers whose solvency or conduct fails the Act’s standards. The penalty code was repriced in 2015 and states the modern scale: a general penalty of one crore rupees attaches to defaults under several central provisions, the social and third party obligations carry up to twenty five crore rupees under Section 105B, and adjudication of the scheduled defaults is conducted by an adjudicating officer under Section 105C after notice, with appeal lying to the Securities Appellate Tribunal, the tribunal built for securities markets and borrowed for insurance by the 2015 amendment because appellate capacity had to be found somewhere credible. The presence of a securities tribunal in an insurance statute is a small emblem of Indian financial law’s habit of sharing institutions across sectors.

These are not paper figures. Adjudication orders imposing the one crore penalty under the Act’s general penalty provision for claims handling failures and regulatory defaults have been passed against insurers and tested on appeal before the Securities Appellate Tribunal, which has upheld penalties of that order where the default was established. The penalty code’s credibility, on which the pricing of deterrence depends, rests on that record of use.

In ordinary times the fourth machine does quieter work. Supervision runs on the returns and actuarial reports the prudential sections compel, tested by on site inspection against the insurer’s own records. Section 33 gives the investigation its teeth: the Authority may investigate the affairs of an insurer through its own officers or through auditors and actuaries it appoints, may examine officers and records on the matters the section specifies, and the cost of the exercise falls on the insurer investigated, a detail that concentrates minds in boardrooms the way a fine rarely does. Its characteristic product is not a penalty but a direction: to strengthen a reserve, to stop writing a line of business that is consuming capital, to replace an officer, to correct a practice in claims settlement. The direction power matters more than the penalty code in the life of the Act precisely because it is exercised early, while correction is cheap. The penalty figures make the headlines; the directions do the supervision.

The policyholder does not depend on the supervisor alone. A rejected claim may be pursued as a consumer dispute, and the architecture of that route, commissions, pecuniary limits and the choice between forums, is set out in the companion article on the consumer law of India. The insurance ombudsman, the low cost forum built for insurance grievances specifically, is treated in the companion article on the IRDAI Act, which owns the regulator’s institutions. What belongs to this article is the interaction: Section 45 decides what the insurer is allowed to argue wherever the claim is fought, so the three year line follows the family into the ombudsman’s office and the consumer commission alike. Two supporting disciplines complete the claimant’s statutory environment. The policyholder protection regulations made under the Act require insurers to decide claims within stated timelines and to pay interest for delay beyond them, converting the file bound procrastination that once exhausted claimants into a cost the insurer books. And a repudiation, to be worth the paper it is written on in any of these forums, must still show the grounds and materials Section 45 demands, which is why the first document a claimant’s adviser calls for is never the policy. It is the letter.

Reading the Act with its neighbours

No financial statute works alone, and the Insurance Act’s position in the architecture repays a short map. Its closest neighbour is the LIC Act of 1956, bound to it by the application device of Section 43 of that Act, under which named provisions of the Insurance Act, Section 45 among them, apply to the Corporation with notified modifications. The device is unusual and effective: Parliament legislated policyholder protection once, in the market statute, and imported it into the monopoly statute, so that the two regimes could not drift apart on the questions that matter to claimants. The nationalisation Act of 1972 performs a similar office for general insurance. Around those two stands company law: an Indian insurance company is a company, formed and governed as such, but on every question the Insurance Act answers, capital for registration, investment of funds, the transfer of a book, the special law prevails over the general, and the companies legislation applies in the space the Insurance Act leaves.

The regulator map explains the Act’s borrowed tribunal. Indian finance is supervised in silos, the Reserve Bank for banks and payment systems, the securities regulator for markets, the Authority for insurance, the pension regulator for retirement funds, coordinated loosely at the level of the Financial Stability and Development Council. A financial group may answer to three of them at once, and an insurer that lists its shares adds securities disclosure to insurance supervision. When the amendment of 2015 needed an appellate forum for the Authority’s adjudication orders, it borrowed the Securities Appellate Tribunal rather than building a new bench, and insurance appeals have since been heard by the tribunal that also hears securities appeals. The arrangement is untidy and has worked, which is a fair description of much of the architecture this article has described.

The silo system’s most famous stress test belongs in this map. In 2010 the securities regulator and the Authority issued competing directions over unit linked insurance plans, investment products wearing an insurance wrapper, each regulator claiming the product for its own statute. The quarrel was closed by an ordinance that settled the classification in the Authority’s favour and created a standing mechanism for resolving such inter regulatory disputes. The episode is remembered because it exposed the cost of the silo design, two statutory regulators with plausible texts and no arbiter; its insurance lesson is quieter. The Insurance Act’s definitions, wide enough in 1938 to govern products not yet invented, had drawn a security like contract into the insurance perimeter without anyone planning it, and Parliament, asked to choose, chose the Act.

Assessment: an old chassis under modern loads

The Act of 1938 has governed, in succession, three different insurance markets: a competitive market of mixed Indian and foreign offices, a public monopoly, and a liberalised market of joint ventures and listed insurers moving towards full foreign ownership. That it has done so without wholesale replacement says something real about the durability of functional drafting: entry, capital, investments, policy rights and supervision are the permanent questions of insurance law, and the Act asks all five. Its weaknesses are the weaknesses of its age. Much of its operative detail has migrated into regulations, so the statute alone no longer tells a reader the solvency ratio, the expense ceiling or the surveyor threshold that governs a live dispute. Its policyholder provisions, though transformed in 2015, still place on bereaved beneficiaries the burden of disproving fraud inside three years, a burden the text imposes plainly and which any honest account must record as the settlement’s hardest edge.

The migration of detail into regulations cuts both ways and should be judged as a design choice rather than a decay. Solvency ratios, expense ceilings and surveyor thresholds must move faster than Parliament can, and a code that froze them would date as badly as the deposit sections dated. The price is that the operative law of insurance is no longer legible from the statute alone, and that the forum in which the real figures are set is a regulatory consultation rather than a parliamentary debate. The amendments of 2015, 2021 and 2025 suggest the system has found its rhythm: Parliament settles the questions of value, the ownership ceiling, the incontestability line, the penalty scale, and the Authority settles the questions of calibration. Trouble begins only if the two exchange roles. The chain of examples runs through this article: the surveyor’s statutory figure of twenty thousand rupees stood in the Act until 2015 made it a regulation figure; the commission ceilings stood in the Act until the expense regulations took them; the solvency control level sits in regulation under a floor the Act itself still states in Section 64VA. In each case the Act kept the question of principle and surrendered the question of number, which is the correct division when numbers must move with inflation and principle must not.

The operation of the 2015 settlement over its first decade supports one measured conclusion. A hard contestability window does not abolish claim investigation; it relocates it. Insurers responded to the absolute bar by moving scrutiny to the two places the statute left open, the proposal stage, where underwriting of new policies grew more searching, and the three years themselves, where early claims are investigated with an intensity the old law spread over a longer and less certain contest. Whether that relocation is a social gain depends on the reader’s seat: the family of the fourth year is better protected than any reform could have made it under the old law, while the family of the second year faces a better resourced opponent. The settlement’s defenders and critics each hold half of that truth, and the Act, having chosen finality, leaves the argument there.

Two items of unfinished business stand out for the next amendment exercise. The first is the asymmetry this article has recorded twice: the LIC answers to the Insurance Act through the filter of its own statute, and every other insurer answers to it directly, an arrangement whose justification weakened as the Corporation moved towards public listing and direct competition. The second is the burden the fraud provisions place on beneficiaries when the policyholder is dead. It was a deliberate feature of the 2015 settlement, cohering with the insurer’s need for some answer to the claim it cannot test against the deceased, but it sits oddly beside the same amendment’s consumer welfare purpose, and it is the provision a future Parliament is most likely to revisit. Neither item is an argument against the chassis. Both are arguments that the chassis is due for the kind of overhaul it has received roughly once a decade since 1999.

The verdict of this article is the one its structure has argued. The Insurance Act 1938 survives because its centre holds: the perimeter in Section 2C, the capital in Section 6, the investment pattern in Section 27, the premium rule in Section 64VB, and the three year line in Section 45. Reforms have replaced the operator, repriced the penalties and rewritten the ownership ceiling three times in a generation, but each reform has been drafted as an amendment to the 1938 text rather than as a new code, including the repeal and replacement of the nuclear liability and atomic energy statutes’ neighbour provisions in the energy field. A reader who has followed the argument to this point holds the essential map: the Act decides who may promise, on what capital, with what money, and for how long the promise may be questioned. The regulator’s own story, and the ombudsman who closes so many of the disputes this article has described, belong to the next article in the cluster.

One sentence of perspective closes the assessment. Statutes are usually replaced when their categories fail; the Insurance Act 1938 has instead been amended, in 1950, 1968, 1999, 2002, 2015, 2021 and 2025, each time by a Parliament that examined the categories and kept them. Entry, capital, investments, policy rights, supervision: the five questions were the right five in 1938, when the market was offices and agents, and they remained the right five when the market became joint ventures, bancassurance and foreign branches. That is the deepest reason the pillar stands on a statute from the age of the Controller of Insurance, and why reformers with a free hand have repeatedly chosen to rebuild inside it rather than begin again.

For the reader who came to this article with a claim rather than a syllabus, the whole argument reduces to the discipline of the three dates and the paper that proves them: the date the policy was issued, the date its risk began, and, where the policy lapsed and was brought back, the date of revival, with any rider’s date beside them. Fix those dates from the policy document itself, count three years from the last of them to occur, and the family’s position under Section 45 is known before any letter is written or any forum approached. The statute rewards the reader who counts. It has been rewarding such readers, and punishing those who guess, since the first of July 1939.

Frequently Asked Questions

Q: What does the Insurance Act 1938 regulate?

The Act regulates the business of insurance in India from end to end. It fixes who may carry on insurance business and in what form under Section 2C, the registration of insurers by class under Section 3, the minimum capital under Section 6, the investment of funds under Section 27 and the solvency margin under Section 64VA. It governs the policy itself through nomination under Section 39, assignment under Section 38, the rule in Section 64VB that risk follows premium, and the limits Section 45 sets on questioning a life policy. It also polices conduct, through the rebating ban in Section 41 and the licensing of agents, intermediaries and surveyors, and it arms the supervisor with inspection, direction and penalty powers. It does not create the regulator, which the Act of 1999 does, and it does not decide individual claims.

Q: What capital and registration does an insurer need?

An insurer must first fall within a permitted form under Section 2C, in substance an Indian insurance company, an insurance co-operative society, a statutory body or a foreign reinsurer’s branch, and must then hold a certificate of registration from the Authority under Section 3 for each class of business it writes. The capital condition in Section 6 is a paid up equity capital of one hundred crore rupees for life insurance, for general insurance and for health insurance carried on exclusively, and two hundred crore rupees for reinsurance carried on exclusively. A foreign company writing reinsurance through an Indian branch must instead show net owned funds of not less than five thousand crore rupees. Registration is continuing, and it carries continuing obligations, including the solvency margin, rather than a one time permission.

Q: How much foreign investment is allowed in insurance?

The ceiling has moved in steps. It was 26 per cent when the sector reopened under the settlement of 1999. The amendment of 2015 raised it to 49 per cent with the condition that the company remain Indian owned and controlled. The Insurance (Amendment) Act, 2021, Act 6 of 2021, in force from 1 April 2021, raised it to 74 per cent, dropped the owned and controlled condition and substituted governance safeguards, including resident Indian directors and key managers and a rule retaining at least half of net profit in a general reserve where dividend was paid at weak solvency. The amendment of 2025 raised the ceiling to 100 per cent. The Life Insurance Corporation is different: foreign investment in it is capped at 20 per cent under the foreign investment policy.

Q: Can an insurer reject a claim after three years?

No. Section 45(1) provides that no policy of life insurance shall be called in question on any ground whatsoever after three years from the date of the policy, counted from the last of the date of issuance, the date of commencement of risk, the date of revival and the date of any rider to occur. The words on any ground whatsoever include fraud: an insurer who discovers after the period that the proposal contained a deliberate falsehood still cannot question the policy, and the claim must be paid. Two qualifications are built into the section rather than imposed on it. A revival or a rider starts a fresh three years for what it introduces, and the insurer may at any time call for proof of age and adjust the policy’s terms to the corrected age, which the section says is not a calling in question.

Q: What is the rule on nomination and assignment of a policy?

Nomination under Section 39 names the person to whom the insurer should pay when the policyholder’s death makes the money payable; it can be made at purchase, changed or cancelled later, and payment to the nominee gives the insurer a valid discharge. Since the amendment of 2015, a nominee who is the policyholder’s parent, spouse or child, or a combination of them, is beneficially entitled to the money, not a mere collector for the estate, unless the policyholder’s own title to the policy was too limited to confer that benefit. Assignment under Section 38 is different in kind: it transfers the policy itself during the policyholder’s lifetime, by endorsement or separate instrument with written notice to the insurer, and the assignee then holds the policyholder’s rights. While an assignment subsists it displaces an earlier nomination.

Q: Why must premium be paid before risk is assumed?

Section 64VB provides that an insurer shall not assume any risk on which premium is ordinarily payable until the premium is received, or guaranteed to be paid in the prescribed manner, or deposited in advance. The rule protects the common pool: an insurer that carries risk it was never paid to carry makes every other policyholder subsidise the unpaid claim. It protects the customer as well, against the discovery after a loss that cover was never perfected because an intermediary held the premium and never passed it on. And it protects the insurer’s solvency, which is built on premium actually collected. Inserted in 1968 against the habit of issuing cover on credit, the section makes payment, not paperwork, the moment the risk begins.

Q: What did the 2015 insurance amendment change?

The Insurance Laws (Amendment) Act, 2015, Act 5 of 2015, operating from 26 December 2014, is the statute’s largest revision. It raised the foreign investment ceiling from 26 per cent to 49 per cent with an Indian owned and controlled condition. It substituted Section 45 with the absolute three year bar on questioning a life policy, made family nominees beneficially entitled under Section 39 and reformed assignment under Section 38. It recognised health insurance as a distinct class, admitted foreign reinsurers through Indian branches, omitted the old deposit sections and made registration continuing. It raised penalties into a band of one crore to twenty five crore rupees, created adjudication under Section 105C with appeals to the Securities Appellate Tribunal, imposed rural, social sector and motor third party obligations under Sections 32B to 32D, and barred the promotion of insurance through multilevel marketing schemes.

Q: How are surveyors and intermediaries regulated?

Surveyors and loss assessors, the professionals who measure a loss for the insurer, are licensed by the Authority under Section 64UM. A claim above a threshold the regulations fix must be supported by a licensed surveyor’s report before settlement, the Authority may call for an additional independent report, and surveyors work under a prescribed code of conduct; the report informs the settlement without dictating it. Intermediaries are regulated in parallel. An insurance agent must be licensed under Section 42, and intermediaries such as brokers, corporate agents and third party administrators are licensed or registered under Section 42D and the regulations under it. An insurer that pays commission to a person outside the permitted channel faces a penalty that may reach one crore rupees, and acting as an agent without a licence is itself penalised.

Q: What is the difference between the Insurance Act 1938 and the IRDAI Act 1999?

The two statutes divide the work between substance and institution. The Insurance Act 1938 is the substantive law of the business: entry, capital, investment of funds, the rights of policyholders under Sections 38, 39 and 45, conduct rules and penalties. The Insurance Regulatory and Development Authority Act, 1999 is the institutional law: it created the Authority as the sector’s regulator, ended the statutory monopolies of the Life Insurance Corporation and the nationalised general insurers, and substituted the Authority for the government and the Controller of Insurance throughout the 1938 text. The later Act therefore amends the earlier one rather than replacing it, and a reader who wants to know what an insurer must do reads the Act of 1938, while a reader who wants to know who watches the insurer reads the Act of 1999.

Q: How can an insurer’s registration be suspended or cancelled?

Registration under Section 3 is continuing but conditional, and the Authority may suspend or cancel it on grounds stated in the Act, which include failure to satisfy the capital requirements of Section 6, failure to maintain the solvency margin required by Section 64VA, failure to furnish the returns and information the Act demands, failure to comply with the Act, the regulations or the Authority’s directions, and conduct prejudicial to the interests of policyholders. Cancellation does not erase the insurer’s promises. An insurer whose registration is cancelled may not write new business, and its existing policies must be serviced, run off or transferred under supervision, with the Authority’s orders in the matter appealable to the Securities Appellate Tribunal. The sanction is designed to stop the rot at the door while keeping faith with the people already inside.

Q: Can a policyholder assign a life policy to a bank as security for a loan?

Yes, and it is one of the commonest uses of Section 38. The policyholder assigns the policy to the bank by endorsement on the policy or by a separate instrument, signed and attested, and gives written notice to the insurer, which records the assignment. The bank as assignee then holds the policyholder’s rights to the extent of the assignment: if the policyholder dies while the loan is outstanding, the bank may recover its dues from the proceeds. The policy itself continues in force so long as premiums are paid. When the loan is repaid, the bank reassigns the policy to the policyholder and the original position is restored. While the assignment subsists it displaces an earlier nomination, so a family that borrows against a policy should nominate afresh after the reassignment.

Q: What happens to a claim when the policyholder dies within three years of the policy starting?

The claim falls inside the window in which the insurer may test the policy, and it will usually be investigated. The insurer may call the policy in question only on the two grounds Section 45 allows: fraud, or misstatement or suppression of a fact material to the expectancy of the life. It must communicate the grounds and the materials behind its decision in writing to the legal representatives, nominees or assignees. The insurer carries the burden of proving fraud or materiality, with one hard exception: where the policyholder is dead, the beneficiaries bear the onus of disproving fraud. If the repudiation rests on misstatement alone and succeeds, the premiums collected must be paid back within ninety days. A family refused payment may pursue the grievance machinery, the insurance ombudsman and the consumer commissions.

Q: Does Section 45 apply to health insurance and other general insurance policies?

No. By its terms Section 45 governs policies of life insurance, and its three year bar is a protection built for life claims, where the person who answered the proposal questions is dead when the claim is examined. Health and other general insurance policies are annual contracts of a different structure, and an insurer’s power to resist a claim under them comes from the policy contract and from the Authority’s policyholder protection regulations, which impose their own disciplines on repudiation, including requirements of reasoned orders and regulatory limits on contesting long standing health cover. A claimant under a health policy therefore cannot borrow Section 45, but is not unprotected: the protection sits in subordinate legislation and in the general law of contract rather than in the section.

Q: Who is a beneficial nominee under Section 39 of the Insurance Act?

A beneficial nominee is a nominee who owns the policy money rather than collecting it for the estate. Section 39(7), inserted by the amendment of 2015, confers that status where the policyholder nominates his parents, or his spouse, or his children, or his spouse and children, or any of them. Such nominees are beneficially entitled to the amount payable by the insurer unless it is proved that the policyholder, having regard to the nature of his title to the policy, could not have conferred a beneficial title. If the nominee dies after the insured but before the money is paid, Section 39(8) passes the beneficial interest to the nominee’s heirs or legal representatives as part of the nominee’s estate. The provisions apply to policies maturing for payment after the 2015 amendment commenced, and creditors’ rights against the policyholder are expressly preserved.

Q: What must an insurer prove to call a policy in question on the ground of fraud?

The insurer must act within three years of the last of issuance, commencement of risk, revival or rider to occur, and must prove fraud as the section defines it: a statement or suppression in the proposal or the documents supporting it that falls within the section’s explanation of fraud, acts done with intent to deceive, such as asserting what is not true, actively concealing a known fact, or making a promise without intending to keep it. It must communicate its grounds and materials in writing. Section 45(3) then gives the insured three answers: that the statement was true to the best of his knowledge and belief, that there was no deliberate intention to suppress, or that the fact was within the insurer’s knowledge. Where the policyholder is dead, the proviso places the onus of disproving the fraud on the beneficiaries.

Q: What happens to a life policy when premiums stop being paid?

The answer depends on how long the policy has run. A young policy lapses after the days of grace allowed by its terms: it carries no cover while unpaid, and a death in that state gives the family no claim. A policy that has acquired a surrender value is treated differently, because real money has accumulated under it; under the Act’s surrender value provisions such a policy does not simply die for non payment but is kept alive as a paid up policy for a reduced sum, and the policyholder may instead surrender it for its value, whose computation the regulations govern. A lapsed policy may also be revived within the window the policy and the regulations allow, on payment of arrears and fresh evidence of health. A revival matters beyond the cover itself, since it starts a new three year count under Section 45 for the matters it introduces.

Q: Why was life insurance nationalised in 1956 and general insurance in 1972?

The government that nationalised life insurance in 1956 stated its case as one of reach, security and savings: to spread insurance much more widely, in particular to the rural areas, to ensure absolute security to the policyholder, and to mobilise public savings more effectively for national purposes. The context was a private market with failures and misuse of policyholders’ funds in its record, and 245 Indian and foreign insurers were merged into the Life Insurance Corporation. General insurance followed in 1972 on parallel reasoning, as part of the period’s wider policy of public ownership of commanding heights of the economy. The later criticism, recorded by the Malhotra Committee, was that the monopolies so created delivered neither the service standards nor the penetration the nationalisation case had promised, which set the terms of the reopening debate.

Q: Does the Insurance Act 1938 apply to the Life Insurance Corporation?

Not in the ordinary way. The Corporation is not an insurer registered under the Insurance Act; it is a statutory body created by the Life Insurance Corporation Act, 1956, Act 31 of 1956. Section 43 of that Act applies specified provisions of the Insurance Act to the Corporation, with any modifications the Central Government notifies, and Section 45 is among the provisions so applied, which is why a Corporation policy carries the same three year protection as a private insurer’s. Since the monopolies ended in 1999 the Corporation has competed with private insurers for the same customers, and in most operational respects, products, policyholder protection and conduct, the Authority regulates it alongside them. Its ownership is separately treated: foreign investment in the Corporation is capped at 20 per cent under the investment policy.

Q: What penalties follow a breach of the Insurance Act 1938?

The penalty scale was repriced in 2015 for companies of modern size. Rebating in breach of Section 41 carries a penalty that may extend to ten lakh rupees. A general penalty of one crore rupees attaches to defaults under several central provisions, and an insurer that pays commission to a person not entitled to receive it faces a penalty of up to one crore rupees. The heaviest financial penalty in the Act, up to twenty five crore rupees under Section 105B, backs the obligations to write rural and social sector business and minimum motor third party business under Sections 32B to 32D. Carrying on insurance business without registration is a criminal offence punishable with imprisonment and fine. Penalties are imposed after adjudication under Section 105C, with appeal to the Securities Appellate Tribunal.

Q: What can the Authority do when an insurer falls below the control level of solvency?

Section 64VA requires every insurer to maintain an excess of assets over liabilities of at least 50 per cent of its minimum capital, and empowers the Authority to set a control level by regulation, operated as an available solvency margin of one and a half times the required margin. An insurer that cannot maintain that level must submit a financial plan showing how it will restore the margin, within a period the section caps at six months, and the Authority may direct the steps the situation requires in the interests of policyholders. If the insurer falls below the statutory floor itself, rather than the control level, it is deemed insolvent for the purposes of winding up, which the Authority may seek. The sequence matters: the control level is an early warning trigger for correction, long before policyholders’ money is actually short.