Risk, Pooling and the Indian Insurance Market
Insurance is the business of turning uncertain individual losses into predictable shared costs. This chapter builds the vocabulary of risk, shows with arithmetic why pooling works, explains which risks insurers will accept, and then maps the Indian industry from the Insurance Act 1938 to the Sabka Bima Sabki Raksha Act, 2025. It also teaches how penetration and density are computed, because both appear regularly in awareness papers.
KarmSakha preparation topics covered
- Scope of insurance and banking awareness tests and how to read a notification
- Risk, pure and speculative risk, peril, hazard and loss
- Physical, moral and morale hazard
- Avoidance, reduction, retention and transfer of risk
- Pooling, expected loss and the law of large numbers
- Six conditions of an insurable risk
- Life, general and health business and reinsurance
- Timeline of Indian insurance from 1938 to 2021
- Ownership and capital reforms of the Sabka Bima Sabki Raksha Act, 2025
- Insurance penetration and density, with 2024-25 figures
What This Book Covers and How Insurance-Sector Tests Use It
This book is a banking and insurance awareness text. It is built for three kinds of reader: a candidate preparing for an insurance agent certification of the IC38 type, a candidate facing the insurance and financial or banking awareness section of an insurance-company recruitment paper, and anyone who needs a clean grounding in how insurance and banks work. Eleven chapters run from risk and pooling, through the principles of the insurance contract, underwriting, life, general and health products, regulation and grievance redressal, and then to money, the Reserve Bank, deposit insurance and the macroeconomy. Insurance occupies roughly the first eight chapters and banking and macro the last three.
Rules in this field move faster than printed study material. Several figures in the 2023 edition of the Insurance Institute of India's IC38 material have since been replaced by IRDAI regulations of 2024 and by legislation passed in December 2025. Throughout the book the current rule is taught and the older figure is labelled as historical, because an examiner who sets a question after a change expects the new rule. Every time-sensitive statement carries a year or a source phrase such as 'as per the IRDAI Regulations, 2024'. Treat those phrases as part of the fact; a number without its date is only half a fact.
Exam patterns differ between bodies and change from cycle to cycle: the number of questions, the sections, the timing, the passing standard and any negative marking are fixed by each notification and can differ between a certification body and a recruiting insurer. This book therefore states no pattern as a rule. Before you begin serious revision, read the latest official notification or syllabus of the test you are taking and note the section names. Use the book to master concepts and to practise original questions; use the notification to decide where to spend your hours. The practice questions here test understanding and short calculations, which is the style most awareness sections reward.
Risk, Hazard, Peril and Loss
Risk, in insurance usage, is uncertainty about a financial loss. Two ideas are needed straight away. A pure risk offers only the possibility of loss or no loss: a house may burn or may not, a driver may cause an accident or may not, but nothing can be gained. A speculative risk offers loss, no loss or gain, as when a person buys shares or starts a new business. Insurers ordinarily accept pure risks only, because a loss from a pure risk is outside the insured's deliberate control, whereas a speculative loss is the price of an attempt to profit. Risks are also split into fundamental risks, which hit large groups at once (inflation, war, widespread unemployment), and particular risks, which fall on an individual or a small set (theft, a fire in one shop). Private insurance is built chiefly around particular pure risks.
Three terms are confused again and again, so fix them precisely. A peril is the cause of a loss: fire, flood, theft, collision, death. A hazard is a condition that raises the chance that a peril will occur or the size of the loss if it does. The loss is the reduction in the value of the asset or the financial harm that follows. Take a shop that stores firecrackers beside its electrical panel: fire is the peril, the stored firecrackers and bad wiring are hazards, and the burnt stock is the loss. An asset exposed to such a peril, such as stock, a vehicle, or a person's earning capacity, is the subject of the cover. The usual exam trap is to call a hazard a peril, or the reverse.
Hazards are of three kinds. A physical hazard is a tangible feature of the risk, for example wooden construction, a building located in a flood plain, or poorly maintained brakes. A moral hazard is dishonesty or a defect of character that influences the frequency or severity of loss, such as setting fire to unsold stock to claim insurance, exaggerating a claim, or hiding a known illness at proposal. A morale hazard is not dishonest; it is carelessness or indifference that arises because the loss is insured: leaving a car unlocked because 'it is insured anyway'. The distinction matters in underwriting. Physical hazards can often be measured or reduced by inspection, whereas moral hazard is controlled by careful selection, disclosure duties and policy conditions, and morale hazard by deductibles, which keep some loss with the insured.
| Term | Meaning | Example |
|---|---|---|
| Peril | Cause of loss | Fire, theft, flood, collision, death |
| Physical hazard | Tangible condition raising chance or severity | Thatched roof, chemicals stored near a flame |
| Moral hazard | Dishonesty or character defect | Deliberately setting fire to stock, inflating a claim |
| Morale hazard | Carelessness because insured | Not locking a car, ignoring maintenance |
| Pure risk | Loss or no loss only | House fire, accident |
| Speculative risk | Gain, loss or no loss | Stock trading, a new venture |
Risk Management Techniques
A person or business facing a pure risk has four basic responses, and insurance is only one of them. Avoidance means not undertaking the activity: a firm declines to manufacture a hazardous chemical, or a family decides never to build on a river bank. It removes the risk completely but also removes the benefit of the activity, so it is practical only for risks not worth carrying. Reduction, also called loss control, lowers either the probability of loss (loss prevention, such as safety training or lightning conductors) or the size of the loss once it happens (loss minimisation, such as sprinklers, fire doors or airbags). Reduction does not eliminate risk, but it improves the risk an insurer will later be asked to take, which is why good safety practice can earn better terms.
Retention means bearing the loss oneself. It may be unplanned, when a person is simply unaware of a risk, or planned, when a decision is made to carry the loss from current income or from a set-aside fund. Planned retention is sensible for small, frequent and affordable losses, and it is built into insurance itself through deductibles and excesses. Transfer shifts the financial consequence of the loss to someone else. Insurance is the best-known transfer, but a lease clause that makes the owner liable for repairs, a hedge, or a contract that makes a supplier carry the transit risk are transfers too. A sound programme normally retains small losses, reduces what can be reduced, and insures the large, rare losses that would be ruinous if retained.
The techniques are usually matched to a risk by two questions: how likely is the loss and how severe would it be? High frequency with low severity (minor scratches on a delivery fleet) points to retention and loss control, since insuring such losses mostly returns the insured's own money minus expenses. Low frequency with high severity (death of the sole earner, destruction of a factory) is the natural field of insurance. High frequency with high severity should be avoided or redesigned, and low frequency with low severity can simply be ignored. This simple grid explains why a small phone-repair expense is rarely insured while a hospital stay or a building fire is. Notice also that insurance does not stop a loss from happening; it only compensates the financial effect.
| Frequency and severity | Sensible response | Illustration |
|---|---|---|
| High frequency, low severity | Retain and control | Small dents on a delivery fleet |
| Low frequency, high severity | Transfer by insurance | Fire destroying a factory, death of sole earner |
| High frequency, high severity | Avoid or redesign the activity | Operating in a persistently flooded location |
| Low frequency, low severity | Accept and ignore | Loss of a cheap accessory |
Pooling and the Law of Large Numbers
Insurance works because many people with similar exposures contribute to a common fund from which the few who suffer a loss are paid. A single owner cannot know whether his house will burn this year, but a group of owners can predict, fairly closely, how many of them will. The standard illustration used in the Insurance Institute of India's material runs as follows. Suppose 100,000 houses, each worth Rs 50,000, and a probability of 0.002 that any one is destroyed in a year. The expected number of losses is 100,000 x 0.002 = 200 houses, and the expected loss is 200 x Rs 50,000 = Rs 1,00,00,000, which is Rs 1 crore. Divided over 100,000 owners, each must contribute Rs 1,00,00,000 / 1,00,000 = Rs 100. Each owner swaps an uncertain loss of Rs 50,000 for a certain cost of Rs 100.
This is not just a trick of division. The justification is the law of large numbers: as the number of independent, similar exposures grows, the actual average loss per exposure moves closer to the expected loss. The absolute spread of total losses does grow with the pool, but the spread relative to the expected total shrinks roughly in proportion to the square root of the number of exposures. That is why a large insurer can charge a premium close to expected loss with confidence, while a pool of ten members could be badly wrong in any given year. Two conditions are needed: exposures must be reasonably similar, so that one probability fits all, and losses must be independent, so that one event does not hit the whole pool together.
A second original illustration shows the same logic with different numbers. A market association has 20,000 shops, each with a 0.005 yearly probability of a fire that destroys stock averaging Rs 4,00,000. Expected losses are 20,000 x 0.005 = 100 shops, so the expected total is 100 x Rs 4,00,000 = Rs 4,00,00,000, or Rs 4 crore. The pure share per shop is Rs 4,00,00,000 / 20,000 = Rs 2,000, which also equals 0.005 x Rs 4,00,000. Real premiums are higher than this figure because the insurer must also meet expenses, commission, reserves and profit, a topic developed in the chapter on underwriting and rating. What pooling gives is the foundation, the pure cost of the risk, from which the rest is built.
What Makes a Risk Insurable
Not every pure risk is acceptable to an insurer. Commonly taught as six economic conditions (the IC38 study text lists a slightly different six: a large number of homogeneous exposures, a definite and measurable loss, a fortuitous loss, only a small percentage of the group suffering loss at a time, economic feasibility, and a risk not contrary to public policy; the two lists overlap heavily), an insurable risk needs: a large number of similar exposure units, so that the law of large numbers can operate; a loss that is accidental, meaning outside the control of the insured and not intended; a loss that is definite and measurable in time, place and amount, so that a claim can be settled; a loss that is not catastrophic for the insurer, meaning that a single event should not strike most of the pool at once; a probability of loss that can be calculated, at least approximately, from past data; and a premium that is economically feasible, so that people are willing to pay it. The condition of insurable interest, treated in the next chapter, is a legal companion to these economic tests.
Applying the conditions explains many practical exclusions. A business loss from deliberate speculation fails the accident test and is a speculative risk anyway. A loss that the insured causes on purpose is not accidental. Losses such as slow wear and tear are certain rather than uncertain, so there is no pooling to be done. War and nuclear risks are traditionally excluded from standard covers because one event can hit a very large number of insureds together and the loss cannot be estimated. Floods and earthquakes sit in an uneasy middle: they can strike many policyholders at once, yet property policies often include them, usually with higher rates, sub-limits or zoning, and the insurer protects itself by reinsurance, which spreads the heavy loss across other insurers.
The conditions are guidelines rather than laws, and markets stretch them. Where data is thin, as with a new technology risk, insurers may still write cover at a cautious price. Where loss is not easily measured, such as a life, the contract fixes a sum assured in advance instead of measuring loss, which is why life cover is described as a benefit contract rather than an indemnity. A risk can fail one test and still be insured if the other conditions are strong or if the state steps in, as with government-supported crop and health schemes where premium affordability is a policy concern. When a question asks why a risk is uninsurable, find the single condition most clearly broken and name it.
