Session 03: Insurance Business Model
Learning Objectives
- Deconstruct how insurers generate profit through underwriting income and investment income, and explain why investment income on the "float" is central to the insurance business model
- Compute gross premiums by decomposing them into pure premium, expense loading, and profit/contingency loading
- Analyze the underwriting process from application to policy issuance and the claims process from FNOL to settlement
- Calculate and interpret the seven key insurance financial metrics: loss ratio, expense ratio, combined ratio, claim settlement ratio, persistency ratio, IBNR, and solvency ratio
- Build a simplified insurance Profit & Loss statement in Excel and conduct sensitivity analysis on claims frequency and severity
1. How Insurers Make Money
Most businesses make money by selling a product for more than it costs to produce. An insurer does this too — but it has a second, equally important revenue stream that most non-insurance businesses do not have. Understanding both revenue streams, and how they interact, is essential to understanding why insurance is a unique business.
1.1 The Two Engines of Insurance Profitability
Insurers generate profit from two distinct sources:
Engine 1 — Underwriting Profit: This is the profit from the core insurance operation. The insurer collects premiums and pays claims and operating expenses. If premiums exceed claims plus expenses, the insurer earns an underwriting profit. If not, it incurs an underwriting loss.
Underwriting Profit = Earned Premium − Incurred Claims − Operating ExpensesEngine 2 — Investment Income: Insurers collect premiums upfront but pay claims later — sometimes months later (motor, health), sometimes years later (life insurance), sometimes decades later (liability, workers' compensation). During this gap, they invest the premiums and earn returns. These premiums held pending claim payment are called the "float."
Investment Income = Invested Assets (Float + Capital) × Investment Yield (%)The total profitability equation is therefore:
Profit Before Tax = Underwriting Profit + Investment Income
Or equivalently:
Profit Before Tax = (Earned Premium − Claims − Expenses) + (Float × Investment Yield)1.2 When Underwriting Loses Money
It is entirely possible — and common — for insurers to run underwriting losses while remaining profitable overall. An insurer with a combined ratio of 105 (spending ₹105 on claims and expenses for every ₹100 in premiums) is losing ₹5 on underwriting. But if it holds ₹10,000 crore in float earning 7% annually (₹700 crore in investment income) against ₹5,000 crore in earned premiums, the ₹250 crore underwriting loss is more than covered by ₹700 crore in investment income. The company is profitable — but dependent on investment markets.
This is both a strength and a vulnerability. When interest rates are high, insurers with large floats thrive. When interest rates are low (as they were globally from 2009–2021), investment income shrinks, and underwriting discipline becomes critical. The insurer that relied on investment income to cover sloppy underwriting gets exposed when rates fall.
🤔 Exercise 1.1 — Is This Insurer Profitable?
Below are the financials of three insurers. For each, calculate: (a) Underwriting Profit/Loss, (b) Investment Income, (c) Profit Before Tax. Then answer the reflection question.
| Item | Insurer A | Insurer B | Insurer C |
|---|---|---|---|
| Earned Premium | ₹200 Cr | ₹200 Cr | ₹200 Cr |
| Incurred Claims | ₹150 Cr | ₹155 Cr | ₹160 Cr |
| Operating Expenses | ₹45 Cr | ₹50 Cr | ₹50 Cr |
| Float (investable assets) | ₹300 Cr | ₹250 Cr | ₹500 Cr |
| Investment Yield | 6% | 3% | 2% |
| Underwriting Result | |||
| Investment Income | |||
| Profit Before Tax |
Reflection question: Which insurer is most at risk if interest rates fall further? Explain your reasoning.
View Solution — Calculations and Risk Ranking
Calculations
| Item | Insurer A | Insurer B | Insurer C |
|---|---|---|---|
| Underwriting Result | 200 − 150 − 45 = +₹5 Cr | 200 − 155 − 50 = −₹5 Cr | 200 − 160 − 50 = −₹10 Cr |
| Investment Income | 300 × 6% = ₹18 Cr | 250 × 3% = ₹7.5 Cr | 500 × 2% = ₹10 Cr |
| Profit Before Tax | ₹23 Cr | ₹2.5 Cr | ₹0 Cr (break-even) |
Risk Ranking (from most to least at risk if rates fall)
- Insurer C (most at risk): It is already break-even — its underwriting loss (−₹10 Cr) exactly cancels its investment income. Its investment yield is only 2%. If rates fall to 1.5%, investment income drops to ₹7.5 Cr and it becomes loss-making. It also has the largest float (₹500 Cr), meaning its total profitability swings most with rates. C is entirely dependent on investment markets to survive its poor underwriting.
- Insurer B: Has a small underwriting loss (−₹5 Cr) but a thin ₹2.5 Cr profit cushion. A 0.5% drop in yields would roughly halve its profit to ₹1.25 Cr. Still marginally dependent on investment income.
- Insurer A (least at risk): Has an underwriting PROFIT of ₹5 Cr — profitable on the core business even without investment income. Its ₹18 Cr investment income is a bonus, not a lifeline. Even if yields fall to 2%, it stays profitable.
Key insight: Insurer A is the only one with a sustainable business model — it can survive low interest rates. B and C are dependent on investment markets, which is dangerous. When the CFA/board of an insurer reviews results, the first question should always be: "Is the profit coming from underwriting (sustainable) or investment income (dependent on markets)?"
2. Premium Calculation
The premium is the price of insurance. Setting it correctly is the single most important decision an insurer makes. Price too high, and customers go to competitors. Price too low, and claims plus expenses exceed premiums — the insurer loses money on every policy it sells. Getting the premium right requires understanding its components.
2.1 The Building Blocks of Premium
| Component | What It Covers | Typical Share | Example: Motor OD (₹10,000 Premium) |
|---|---|---|---|
| Pure Premium | Expected claims cost per policy. Calculated as: (Claim Frequency × Average Claim Severity) or (Total Expected Claims ÷ Number of Policies). | 55–70% | ₹6,200 (62%) — the amount the actuary expects will be paid in claims for this risk pool |
| Expense Loading | Acquisition costs (commission to agents/brokers/aggregators), policy administration, technology, salaries, rent, and other operating expenses. | 20–30% | ₹2,500 (25%) — commissions, admin, IT, office, salaries |
| Profit Loading | The insurer's target underwriting profit margin. This is what shareholders expect as a return on the capital they have committed to the insurance company. | 3–8% | ₹500 (5%) — target underwriting profit |
| Contingency Loading | A buffer for uncertainty. Claims experience may be worse than expected (higher frequency, higher severity, or both). This loading provides a margin of safety. | 3–8% | ₹800 (8%) — safety margin for adverse deviation |
Gross Premium = Pure Premium + Expense Loading + Profit Loading + Contingency Loading
Example: Motor Own-Damage Insurance
Pure Premium = ₹6,200 (expected annual claims per policy)
Expense Loading = ₹2,500 (commission + admin + overhead allocation)
Profit Loading = ₹500 (5% underwriting margin)
Contingency Loading = ₹800 (8% safety margin)
─────────────────────────────────────
Gross Premium = ₹10,0002.2 Pure Premium Calculation — A Deeper Look
The pure premium is the actuarial estimate of expected claims. It is calculated from historical data and adjusted for expected future conditions:
Pure Premium = (Total Expected Annual Claims) ÷ (Number of Exposure Units)
Where: Total Expected Claims = Claim Frequency × Number of Policies × Average Claim Severity
Example: Motor OD Portfolio
Number of Policies = 100,000
Claim Frequency = 12% (12 claims per 100 policies per year)
Average Claim Severity = ₹52,000
Total Expected Claims = 100,000 × 0.12 × ₹52,000 = ₹62,40,00,000 (₹62.4 crore)
Pure Premium per Policy = ₹62,40,00,000 ÷ 100,000 = ₹6,240The pure premium is then loaded with expenses, profit, and contingency to arrive at the gross premium that the customer actually pays.
🧮 Exercise 2.1 — Build a Premium from Scratch
You are the pricing actuary for a new motor insurance product. Use the data below to calculate the premium step by step.
Given data: 50,000 policies | Claim frequency: 8% | Average claim severity: ₹45,000 | Expense loading: 25% of pure premium | Profit loading: 8% of pure premium
Calculate
- Total expected claims: Number of policies × claim frequency × average severity.
- Pure premium per policy: Total expected claims ÷ number of policies.
- Expense loading per policy: Pure premium × 25%.
- Profit loading per policy: Pure premium × 8%.
- Gross premium: Pure premium + expense loading + profit loading.
Challenge
A competitor sells the same product at a premium of ₹4,500. Using the loss ratio and expense ratio reasoning from Section 5.1, explain what this implies about the competitor's pricing. Is it sustainable?
View Solution — Full Calculation
Step-by-step calculation
Step 1: Total expected claims
= 50,000 × 8% × ₹45,000
= 50,000 × 0.08 × ₹45,000
= 4,000 claims × ₹45,000
= ₹18,00,00,000 (₹18 crore)
Step 2: Pure premium per policy
= ₹18,00,00,000 ÷ 50,000
= ₹3,600
Step 3: Expense loading
= ₹3,600 × 25% = ₹900
Step 4: Profit loading
= ₹3,600 × 8% = ₹288
Step 5: Gross premium
= ₹3,600 + ₹900 + ₹288
= ₹4,788Challenge answer
The competitor's premium of ₹4,500 is below your gross premium of ₹4,788 — but above the pure premium of ₹3,600. Let's decompose it:
- If the competitor's loss ratio is the same as yours (₹3,600/₹4,500 = 80%), they are paying 80% of every premium rupee in claims — a high but not impossible loss ratio.
- The remaining ₹900 (20%) must cover BOTH operating expenses AND profit. That is a very thin margin. If the competitor's expense ratio is 18%, their profit margin is only 2% — barely break-even.
- Two possibilities: (a) The competitor has genuinely lower costs (digital distribution, no commissions) — sustainable advantage. (b) The competitor is under-pricing risk — charging below what the claims will cost once actual claims experience is known. If claim frequency rises above 8% or severity above ₹45,000, the competitor's loss ratio will spike and they will lose money on every policy.
The takeaway from the Pro Tip above: You cannot tell from the price alone whether ₹4,500 is a sustainable competitor advantage or a time bomb. You would need to know their actual cost structure. But the math shows: at ₹4,500, there is very little room for expense + profit. If they are a digital-first carrier with an 18% expense ratio, they are barely break-even. If they are a traditional carrier with a 28% expense ratio, they are already losing money on underwriting — and the losses will catch up.
3. Underwriting: The Gatekeeper
Underwriting is the process of deciding which risks to accept, at what price, and on what terms. It is the insurer's primary defense against adverse selection — the tendency of high-risk individuals to seek insurance more actively than low-risk individuals. A good underwriter is worth their weight in gold because they prevent bad risks from entering the pool in the first place.
3.1 The Underwriting Process
- Application: The prospective policyholder submits an application (proposal form) with information about the risk to be insured — age, health status, vehicle details, property location, business operations, claims history, etc. The accuracy and completeness of this information is critical; material misrepresentation can void the policy.
- Risk Assessment: The underwriter evaluates the risk. This involves: reviewing the application, checking internal databases (prior claims, previous policies), accessing external data (credit scores, motor vehicle records, medical databases), and applying underwriting guidelines that specify acceptable risk parameters.
- Decision: The underwriter makes one of four decisions: Accept at standard terms, Accept with modification (higher premium, higher deductible, restricted coverage, specific exclusions), Refer to a senior underwriter (the risk is outside standard guidelines), or Decline (the risk is unacceptably high or falls outside the insurer's risk appetite).
- Pricing: If accepted, the final premium is set. This may be the standard rate (for risks that fit the actuarial model), a debited rate (higher premium for higher risk), or a credited rate (lower premium for lower risk — e.g., a No Claim Bonus discount).
- Policy Issuance: The policy document is generated, terms and conditions are finalized, and the policy is issued to the customer. The insurer is now on risk.
3.2 Risk Classification
Underwriters classify risks into groups that share similar expected loss characteristics. The purpose is to ensure that each policyholder pays a premium that reflects their risk — preventing cross-subsidization where low-risk customers overpay to subsidize high-risk customers (which would drive the low-risk customers to competitors).
Common rating factors in Indian insurance:
- Motor: Vehicle make/model, cubic capacity, age of vehicle, geographical zone (Zone A/B/C), Insured Declared Value (IDV), claim history (No Claim Bonus), driver age, vehicle usage (private/commercial)
- Health: Age, gender (contentious — banned as a rating factor in the EU since 2012), pre-existing conditions, BMI, smoking status, occupation, geographical location, sum insured, room rent limit
- Property: Construction type (RCC, load-bearing, temporary), occupancy (owner-occupied, tenant, commercial), location (flood zone, earthquake zone, crime rate), sum insured, fire protection systems, age of building
- Life: Age, gender, smoking status, occupation, income, medical history, family medical history, lifestyle (hazardous activities), policy term, sum assured
💼 Exercise 3.1 — You Are the Underwriter
Four insurance applications land on your desk. For each, decide: Accept at standard rate / Accept with modification / Refer to senior underwriter / Decline. State the specific modification or the reason for referral/decline.
| # | Application | Your Decision — Write Directly Here |
|---|---|---|
| 1 | Term Life — Rohan, 28: Non-smoker, healthy, income ₹12 L/year, no claims history, wants ₹1 Cr cover. | |
| 2 | Health — Priya, 45: Disclosed pre-existing type-2 diabetes (well-controlled with medication). Wants ₹5 L individual health cover. | |
| 3 | Motor — Arjun, 38: 3-year-old luxury SUV, wants own-damage cover. Has filed 4 claims in the past 2 years (3 of them small bumper repairs). | |
| 4 | Property — Meera, 42: Commercial building in a severe flood zone. Her previous insurer declined to renew. Wants ₹2 Cr fire + flood cover. |
Tip: You can type your answer directly here. Nothing is saved — your text stays only on this screen. Compare your answer with the solution below.
View Solution — Underwriting Decisions
| # | Decision | Rationale |
|---|---|---|
| 1 | Accept at standard rate | Rohan is a textbook low-risk applicant: young, healthy, non-smoking, adequate income for the sum assured, clean history. He fits the actuarial model perfectly. No modification needed. Issue immediately. |
| 2 | Accept with modification | Diabetes is a well-understood pre-existing condition. Since it is disclosed and controlled, the risk is manageable. Standard modifications: (a) impose a waiting period for diabetes-related claims (typically 12–36 months in Indian health insurance), (b) possibly a premium loading of 10–30%, (c) exclusion or limitation on certain diabetic complications. Never decline a controlled, disclosed condition outright — it is a huge and growing customer segment. |
| 3 | Refer to senior underwriter | Four claims in 2 years is a significant red flag — especially if 3 were small bumper repairs. Patterns like this can indicate: (a) a genuinely careless driver, (b) possible fraud (claiming for damage that never happened), or (c) repair-garage collusion. The risk does not clearly fit standard guidelines — a senior underwriter should investigate the claim history in detail (what were the claims for? were they legitimate?) before deciding. Possible outcomes: accept with a heavy loading, accept with a higher deductible, or non-renew after the current term. |
| 4 | Refer — or Decline | A commercial building in a severe flood zone that the previous insurer declined to renew is a high-risk application. Standard underwriting would: (a) require a surveyor report assessing flood exposure, construction quality, and mitigation measures (elevation, flood barriers), (b) likely apply a very high premium loading or a flood exclusion, (c) possibly require the building to have flood protection measures before accepting. If the surveyor report is unfavourable, decline. If the applicant installs flood barriers, accept with a flood sub-limit. The key principle: never accept a high-risk property without a physical survey. |
4. Claims: The Moment of Truth
Claims is where the insurer's promise becomes real. A policyholder who has paid premiums for years and then suffers a loss will judge the insurer entirely by the claims experience. A well-handled claim builds loyalty; a poorly handled claim destroys it — and the policyholder tells everyone they know.
4.1 The Claims Lifecycle
1. FNOL — First Notice of Loss
The policyholder reports the loss to the insurer. This is the insurer's first opportunity to set the tone: empathy, clarity, speed. Digital FNOL (via app, WhatsApp, website) is replacing phone-based FNOL for simple claims, but complex or emotional claims (death, serious injury) still warrant human contact.
2. Triage
The claim is categorized by type, complexity, and estimated severity. Simple claims (minor motor damage) may be routed to straight-through processing. Complex claims (liability disputes, large property losses) are assigned to experienced adjusters. Potential fraud indicators trigger investigation routing.
3. Investigation
The insurer verifies: Did the loss occur? Is it covered by the policy? Are there any exclusions that apply? Is the claimed amount reasonable? This may involve: reviewing documents, interviewing the policyholder and witnesses, engaging a surveyor/loss assessor, reviewing police/FIR reports, checking for fraud indicators.
4. Assessment (Quantum)
Once coverage is confirmed, the insurer determines the amount payable. For property/motor: surveyor assesses repair/replacement cost, applies depreciation. For health: TPA verifies medical bills against policy terms and reasonable & customary charges. For liability: legal assessment of damages.
5. Settlement or Rejection
The insurer communicates the decision. If approved: payment is made (NEFT to bank account is now standard in India). If partially approved: clear explanation of what was covered and what was not. If rejected: detailed written explanation citing the specific policy clause, with information on how to appeal.
6. Recovery (Subrogation)
If the loss was caused by a third party, the insurer may pursue recovery from that party after compensating the policyholder. This is called subrogation — the insurer "steps into the shoes" of the policyholder to recover from the responsible party. In motor insurance, insurers actively pursue recoveries from at-fault third parties.
4.2 Claims Leakage
Claims leakage is the difference between what an insurer actually pays in claims and what it should have paid under optimal claims management. Leakage happens through:
- Overpayment: Paying more than the loss warrants due to inadequate investigation, inflated repair estimates, or fraud
- Underpayment: Paying less than owed (inadvertently or deliberately), leading to disputes, complaints, regulatory intervention, and reputational damage
- Process inefficiency: Excessive claims handling costs due to manual processes, rework, unnecessary surveyor visits, or slow decision-making
Industry estimates suggest claims leakage of 5–10% of total claims spend — meaning a large Indian general insurer paying ₹3,000 crore in annual claims could be leaking ₹150–300 crore. This is the financial prize for AI-based claims automation.
🔍 Exercise 4.1 — Spot the Claims Leakage
Each scenario below involves claims leakage — the gap between what an insurer should pay and what it actually pays. For each: (a) identify whether leakage occurred, (b) classify it as overpayment / underpayment / process inefficiency, and (c) suggest a fix.
| # | Scenario | Leakage Type |
|---|---|---|
| 1 | A health claim includes a room charge of ₹40,000/day, but the policy's room rent limit is ₹20,000/day. The claims processor did not catch it and paid the full amount. | |
| 2 | A motor claim of ₹60,000 takes 45 days to settle because the surveyor's report was lost twice and the file had to be re-processed. The customer is furious. | |
| 3 | An auto-approval system pays every claim under ₹15,000 automatically — without checking whether the policy was active on the accident date. | |
| 4 | A valid claim is settled for ₹75,000, but the policyholder was actually owed ₹1,00,000 — the adjuster applied the wrong depreciation rate to the damaged asset. |
View Solution — Leakage Analysis
| # | Type | Analysis and Fix |
|---|---|---|
| 1 | Overpayment | The insurer paid ₹20,000 more per day than the policy allows. This is pure leakage — the excess room charge should never have been paid. Fix: Implement an automated claims audit that checks billed amounts against policy limits BEFORE payment — exactly what the insurer in the Real World example above did, saving ₹18 Cr/year. The check is simple: policy room limit vs. actual room charge. |
| 2 | Process inefficiency | The claim was correct but the PROCESS was broken — a lost surveyor report caused 45 days of delay. The financial leakage here is indirect: customer dissatisfaction, potential complaints to IRDAI, and the cost of re-processing. Fix: Digitise surveyor report submission (report uploaded directly to the claims system via app/web), track report delivery with automated reminders, and set SLA alerts if a surveyor report is overdue by more than 5 days. |
| 3 | Overpayment + fraud risk | The auto-approval rule was designed for speed but has a dangerous gap: it does not verify policy coverage. A claim on an expired or void policy would be paid. This is both leakage (paying what is not owed) and a fraud vulnerability. Fix: Add a mandatory coverage check to the auto-approval workflow — verify the policy is active and the event is covered BEFORE any automatic payment. This is a classic "control gap" in an otherwise efficient process. |
| 4 | Underpayment | The insurer paid ₹25,000 less than owed. While this is the opposite of leakage in cash terms (the insurer saves money), it is a regulatory and reputational liability: the policyholder will complain, IRDAI tracks claim settlement disputes, and the insurer may face penalties plus the cost of correcting the payment. Underpaying genuine claims also destroys customer trust. Fix: Standardise depreciation rate tables in the assessment system (remove adjuster discretion), and build a post-payment audit that samples settled claims to verify accuracy — both overpayments AND underpayments. |
Key insight: Leakage is not only about paying too MUCH — it is also about paying too LITTLE (which creates regulatory and reputational risk) and paying slowly (which destroys trust). A well-managed claims operation controls all three.
5. Key Insurance Metrics
Insurance has its own vocabulary of financial metrics. Mastering these seven metrics will allow you to read an insurer's financial statements, evaluate an InsurTech's performance, and understand any insurance conversation.
5.1 The Seven Essential Metrics
| # | Metric | Formula | What It Tells You | Benchmark |
|---|---|---|---|---|
| 1 | Loss Ratio | Incurred Claims ÷ Earned Premium × 100 | What percentage of every premium rupee is paid out in claims? The core indicator of pricing adequacy. | 55–75% (motor, health); 40–60% (property); 30–50% (life protection) |
| 2 | Expense Ratio | (Operating Expenses + Commissions) ÷ Gross Written Premium × 100 | How efficiently does the insurer operate? What does it cost to acquire and service each rupee of premium? Note the base: Indian insurers report expenses against written premium while the loss ratio uses earned premium. Some analysts put both on earned premium instead — always state which base you are using, since the two can differ by several points. | 20–35% (traditional); 15–25% (digital-first); LIC agency costs push this higher |
| 3 | Combined Ratio | Loss Ratio + Expense Ratio | The single most important metric in general insurance. Below 100 = underwriting profit. Above 100 = underwriting loss. Investment income must cover the gap. | 95–105% is typical; below 95 is excellent; above 110 is concerning |
| 4 | Claim Settlement Ratio | Claims Paid ÷ (Claims Paid + Claims Rejected) × 100 | What percentage of claims received are paid? A measure of claims fairness and customer-centricity. High is generally good, but 100% could indicate inadequate fraud controls. | Health: 85–95%; Life (by number of policies): 97–99% (term insurance has lower settlement ratios than endowment) |
| 5 | Persistency Ratio | Policies Renewed ÷ Policies Up for Renewal × 100 | What percentage of customers renew their policies? The insurance equivalent of customer retention. Critical because acquisition costs are recovered over multiple renewal years. | Life (13th month): 75–85% (private), 65–75% (LIC); General (annual): 70–85% |
| 6 | IBNR | Incurred But Not Reported — an actuarial estimate of claims that have occurred but not yet been reported to the insurer | How adequate are the insurer's reserves for claims that have already happened but haven't been filed yet? Under-reserving IBNR is the most common cause of insurer insolvency. | Varies by line — higher for long-tail lines (liability, 30–50% of total reserves), lower for short-tail (motor, 5–15%) |
| 7 | Solvency Ratio | Available Solvency Margin ÷ Required Solvency Margin | Does the insurer have enough capital to survive a severe but plausible loss scenario? The ultimate measure of insurer financial strength. | IRDAI minimum: 1.5; Strong insurers: 2.0+; Warning zone: 1.5–1.75; Action zone: below 1.5 |
5.2 The Combined Ratio — Why It Matters Most
The combined ratio deserves special attention because it is the single number that tells you whether an insurer's core business is profitable. It combines the two largest costs — claims and expenses — and compares them to premiums earned.
Combined Ratio Breakdown — Example:
Earned Premium: ₹100
Loss Ratio: Incurred Claims: ₹72 (Loss Ratio = 72%)
Expense Ratio: Op. Expenses: ₹28 (Expense Ratio = 28%)
─────────────────────────
Combined Ratio: ₹100 (Combined Ratio = 100%)
Underwriting Profit: ₹0 (Break-even on underwriting)Every point of combined ratio represents 1% of earned premium. For an insurer with ₹5,000 crore in earned premium, reducing the combined ratio from 105 to 102 means an additional ₹150 crore in underwriting profit — or a ₹150 crore reduction in underwriting loss. This is why insurers obsess over combined ratio improvement.
🧮 Exercise 5.1 — Calculate All Seven Metrics
Use the financial data below to calculate all 7 insurance metrics for "StableGuard Insurance." Show your working for each.
| Input Data | Value |
|---|---|
| Earned Premium | ₹300 Cr |
| Incurred Claims | ₹210 Cr |
| Operating Expenses + Commissions | ₹75 Cr |
| Claims Paid | ₹195 Cr |
| Claims Rejected | ₹15 Cr |
| Policies Renewed | 72,000 |
| Policies Due for Renewal | 90,000 |
| Outstanding Claims + IBNR | ₹40 Cr |
| Available Solvency Margin (ASM) | ₹260 Cr |
| Required Solvency Margin (RSM) | ₹130 Cr |
Calculate: Loss Ratio, Expense Ratio, Combined Ratio, Claim Settlement Ratio, Persistency Ratio, IBNR (as % of incurred claims), Solvency Ratio.
Interpretation: Which single metric is the most concerning for this insurer, and why?
View Solution — All 7 Metrics
Calculations
| # | Metric | Calculation | Result | Benchmark |
|---|---|---|---|---|
| 1 | Loss Ratio | 210 ÷ 300 × 100 | 70.0% | 55–75% → healthy |
| 2 | Expense Ratio | 75 ÷ 300 × 100 — earned premium used as the base here because written premium is not given; see the base note in 5.1 | 25.0% | 20–35% → normal |
| 3 | Combined Ratio | 70.0 + 25.0 | 95.0% | Below 100 → underwriting profit ✓ |
| 4 | Claim Settlement Ratio | 195 ÷ (195 + 15) × 100 | 92.9% | Health 85–95% → acceptable |
| 5 | Persistency Ratio | 72,000 ÷ 90,000 × 100 | 80.0% | 70–85% → good |
| 6 | IBNR as % of Incurred Claims | 40 ÷ 210 × 100 | 19.0% | Depends on line — for short-tail lines this is high |
| 7 | Solvency Ratio | 260 ÷ 130 | 2.00 | Min 1.5; Strong ≥ 2.0 → comfortable |
Interpretation — Most Concerning Metric
Most concerning: IBNR at 19% of incurred claims. Here's why:
- Most of the other metrics are healthy — a 95% combined ratio means the insurer is profitable on underwriting, and a solvency ratio of 2.0 is comfortable.
- But IBNR of 19% of incurred claims is high. For short-tail lines (motor, health), IBNR should be 5–15%. At 19%, the insurer is holding a large reserve for claims that have occurred but not yet been reported.
- Two interpretations: (a) the insurer is being conservative with reserves (safe, but ties up capital and depresses reported profit), or (b) the insurer is discovering that claims are emerging higher than expected — which would suggest the loss ratio is actually worse than the reported 70%.
- The other metrics would look different if the IBNR were inadequate. An insurer that under-reserves IBNR appears to have a good combined ratio and strong solvency — but is actually insolvent in waiting. This is why IBNR adequacy is the most dangerous blind spot.
Key takeaway: A healthy set of headline metrics can hide a reserving problem. Always probe the IBNR and reserve adequacy — it is the metric most likely to be "optimistically wrong."
6. The Insurance P&L in Excel
Building an insurance P&L model in Excel is the best way to internalize how the pieces fit together. Once you have built the model and can adjust assumptions to see the impact flow through, you will understand the insurance business model at a level that reading alone cannot provide.
6.1 The Simplified Insurance P&L Structure
INSURANCE PROFIT & LOSS STATEMENT
A. PREMIUM
Gross Written Premium (GWP) xxx
Less: Reinsurance Ceded (xx)
Net Written Premium xxx
Change in Unearned Premium Reserve (xx)
Net Earned Premium (NEP) xxx ← The top line for underwriting
B. CLAIMS
Claims Paid (during the period) xxx
Change in Outstanding Claims Reserve xx
Change in IBNR Reserve xx
Incurred Claims xxx ← The largest cost
C. OPERATING EXPENSES
Commission / Acquisition Costs xx
Salaries & Benefits xx
Technology & Administration xx
Rent, Utilities, Other xx
Total Operating Expenses xxx
D. UNDERWRITING RESULT
Net Earned Premium xxx
Less: Incurred Claims (xx)
Less: Operating Expenses (xx)
Underwriting Profit / (Loss) xx ← Positive = underwriting profit
E. INVESTMENT INCOME
Investment Income on Float & Capital xx
F. PROFIT BEFORE TAX
Underwriting Result + Investment Income xx6.2 Building the Model in Excel — Step by Step
Open Excel and create the following structure:
' Cell B1: Label — "ASSUMPTIONS"
' Cell B2: Number of Policies = 100000
' Cell B3: Average Premium = ₹10,000
' Cell B4: Claim Frequency = 12%
' Cell B5: Average Claim Severity = ₹52,000
' Cell B6: Commission Rate = 15%
' Cell B7: Operating Expense per Policy = ₹900
' Cell B8: Reinsurance Cession Rate = 10%
' Cell B9: Investment Yield = 7%
' Cell B10a: UPR Change Rate = 5% (used by B14)
' Cell B10b: Reserve Strengthening = 8% (used by B18)
' Cell B10c: Float as % of GWP = 40% (used by B27)
'
' Labels sit in column A; values and formulas sit in column B.
' The last three rates are inputs, not constants — keep them in cells so that
' no formula below contains a hardcoded number.
' ---- PREMIUM CALCULATION ----
' Cell B11: Gross Written Premium =B2*B3 → ₹100,00,00,000 (₹100 Cr)
' Cell B12: Reinsurance Ceded =B11*B8 → ₹10,00,00,000
' Cell B13: Net Written Premium =B11-B12 → ₹90,00,00,000
' Cell B14: Change in UPR =B13*B10a → ₹4,50,00,000
' Cell B15: Net Earned Premium =B13-B14 → ₹85,50,00,000
' ---- CLAIMS CALCULATION ----
' Cell B16: Number of Claims =B2*B4 → 12,000
' Cell B17: Total Claims Paid =B16*B5 → ₹62,40,00,000
' Cell B18: Change in Reserves (OS+IBNR) =B17*B10b → ₹4,99,20,000
' Cell B19: Incurred Claims =B17+B18 → ₹67,39,20,000
' ---- OPERATING EXPENSES ----
' Cell B20: Commission =B11*B6 → ₹15,00,00,000
' Cell B21: Operating Expenses =B2*B7 → ₹9,00,00,000
' Cell B22: Total Operating Expenses =B20+B21 → ₹24,00,00,000
' ---- RESULTS ----
' Cell B23: Underwriting Result =B15-B19-B22 → ₹(5,89,20,000) — Underwriting Loss
' Cell B24: Loss Ratio =B19/B15 → 78.8%
' Cell B25: Expense Ratio =B22/B11 → 24.0% (base = GWP, per Section 5.1)
' Cell B26: Combined Ratio =B24+B25 → 102.8%
'
' Note on bases: the loss ratio is measured on EARNED premium while the expense
' ratio is measured on WRITTEN premium. That is the Indian reporting convention
' and it is what Section 5.1 defines. If you instead put both on earned premium,
' the expense ratio becomes 24.00/85.50 = 28.1% and the combined ratio 106.9%.
' Neither is wrong — but always say which base you are using before comparing
' two insurers, because the gap is worth roughly 4 points here.
' ---- INVESTMENT INCOME ----
' Cell B27: Average Float =B11*B10c → ₹40,00,00,000
' Cell B28: Investment Income =B27*B9 → ₹2,80,00,000
' ---- BOTTOM LINE ----
' Cell B29: Profit Before Tax =B23+B28 → ₹(3,09,20,000) — Overall Loss6.3 Sensitivity Analysis
The real power of the Excel model comes from sensitivity analysis — changing one assumption and observing the impact flow through to the bottom line:
| Scenario | Change | New Combined Ratio | New PBT (₹ Cr) | Impact |
|---|---|---|---|---|
| Base Case | — | 102.8% | (3.09) | Loss of ₹3.09 Cr |
| Better Underwriting | Claim frequency drops from 12% to 10% | 89.7% | 8.14 | ₹11.23 Cr improvement |
| Worse Claims | Claim severity rises from ₹52K to ₹60K | 115.0% | (13.46) | ₹10.37 Cr deterioration |
| Expense Efficiency | Expense per policy drops from ₹900 to ₹700 | 100.8% | (1.09) | ₹2.00 Cr improvement |
| Higher Investment Yield | Yield rises from 7% to 9% | 102.8% | (2.29) | ₹0.80 Cr improvement |
| Premium Increase | Average premium rises 5% to ₹10,500 | 98.6% | 0.57 | ₹3.66 Cr improvement |
Notice that better underwriting (reducing claim frequency by 2 percentage points) has a far larger impact on profitability than higher investment yield (increasing yield by 2 percentage points). This is not always true — it depends on the size of the float relative to premiums — but for general insurers with combined ratios above 100, underwriting improvement almost always dominates investment yield improvement.
Two details in that table repay attention. First, a severity shock is unforgiving: claim severity is multiplied by claim count and then grossed up by the reserve strengthening factor, so a 15% rise in severity costs ₹10.37 crore, not the ₹9.60 crore you get from multiplying claim count by the increase alone. Second, the premium increase is less powerful than it looks. Raising the average premium by 5% lifts earned premium by ₹4.28 crore — but commission is 15% of gross written premium, so it rises too, giving back ₹0.75 crore. Raising prices also does nothing to fix the underlying loss ratio; it only spreads the same claims over a larger premium base, and it assumes every customer renews at the higher price.
📈 Exercise 6.1 — Which Lever Matters Most?
Recall the base case from the sensitivity table: 100,000 policies, ₹10,000 average premium, 12% claim frequency, ₹52,000 average severity, ₹900 expense per policy, 15% commission rate, 10% reinsurance cession. The result was a combined ratio of 102.8% and a PBT loss of ₹3.09 Cr.
Your task: decide which lever to pull — and prove it with numbers.
- Lever A — Claims frequency: Improve underwriting selection so claim frequency drops from 12% to 11%. How much does the combined ratio improve? (Hint: total claims = 100K × 11% × ₹52,000.)
- Lever B — Expense ratio: Cut operating expenses per policy from ₹900 to ₹700. How much does the combined ratio improve?
- Compare: Which lever — a 1-percentage-point improvement in claims frequency OR a ₹200 cut in per-policy expenses — has the bigger impact on the combined ratio? Why is that the case?
- "Make it profitable" challenge: Starting from the base case (PBT = −₹3.09 Cr), propose ONE realistic change that turns the insurer profitable. Show the numbers. (You may adjust claims, expenses, premium, or any combination of ONE primary driver.)
View Solution — Lever Analysis
Step 1 — Lever A: Claims frequency 12% → 11%
Claims PAID at 12%: 100,000 × 0.12 × ₹52,000 = ₹62.40 Cr
Claims PAID at 11%: 100,000 × 0.11 × ₹52,000 = ₹57.20 Cr
Careful — the loss ratio uses INCURRED claims, not claims paid.
Incurred = paid × 1.08 (the 8% reserve strengthening from the model):
Incurred at 12%: ₹62.40 Cr × 1.08 = ₹67.39 Cr
Incurred at 11%: ₹57.20 Cr × 1.08 = ₹61.78 Cr
Saving: ₹5.62 Cr
Loss ratio improves from 78.8% to 72.3% (saving ₹5.62 Cr on ₹85.5 Cr earned premium)
Combined ratio improves by 6.57 points (from 102.8% to 96.3%)
PBT swings from −₹3.09 Cr to +₹2.52 Cr — profitable on this lever aloneStep 2 — Lever B: Expense per policy ₹900 → ₹700
Expense savings: 100,000 × ₹200 = ₹2.00 Cr
Expense ratio improves by 2.00 points (from 24.0% to 22.0%)
— note the expense ratio base is GWP (₹100 Cr), so ₹2 Cr = exactly 2 points
Combined ratio improves by 2.00 points (from 102.8% to 100.8%)
PBT improves from −₹3.09 Cr to −₹1.09 Cr — better, but still a lossStep 3 — Comparison
Lever A (claims frequency) is 3.3× more powerful than Lever B (expenses) — 6.57 combined-ratio points against 2.00.
Why? Because incurred claims are ~79% of premium earned, while expenses are 24% of written premium. A 1-percentage-point improvement applied to a LARGER base (claims) produces a bigger absolute saving. This is the fundamental insight: in insurance, the loss ratio is almost always the bigger lever than the expense ratio — because claims dominate the cost structure.
Step 4 — "Make it profitable" challenge (sample answer)
Several single-lever solutions work. The most realistic one:
Option: Improve claims frequency from 12% to 10.7% (better underwriting selection)
Claims PAID at 10.7%: 100,000 × 0.107 × ₹52,000 = ₹55.64 Cr
Incurred at 10.7%: ₹55.64 Cr × 1.08 = ₹60.09 Cr
Saving vs base (₹67.39 Cr): ₹7.30 Cr
New PBT: −₹3.09 Cr + ₹7.30 Cr = +₹4.21 Cr PROFIT ✓
New Combined Ratio: 94.3%
For reference, exact break-even is a frequency of 11.45%. Anything below that
turns the insurer profitable, so 10.7% clears the bar with room to spare.
This is achievable through: better risk selection (decline or re-price the top
5% of high-risk applicants), a predictive underwriting model, and tightening
claims validation to catch inflated claims.Alternatively, a 7% premium increase (from ₹10,000 to ₹10,700) would achieve a similar result — but this risks losing price-sensitive customers. The underwriting-quality route improves the combined ratio without sacrificing volume. This is why insurers with strong data capabilities consistently outperform on combined ratio.
7. Indian Insurance Financials — A Comparative View
How do Indian insurers actually perform on these metrics? Here is a comparative snapshot based on publicly available data.
7.1 General Insurance — Combined Ratio Comparison
| Insurer Type | Loss Ratio | Expense Ratio | Combined Ratio | Underwriting Status |
|---|---|---|---|---|
| Public Sector General Insurers (avg.) | ~82% | ~28% | ~110% | Loss-making — high claims from motor TP and government schemes |
| Private Sector General Insurers (avg.) | ~72% | ~30% | ~102% | Near break-even — better risk selection, higher commission costs |
| Standalone Health Insurers (avg.) | ~65% | ~28% | ~93% | Profitable — but medical inflation is compressing margins |
| Digital-First General Insurers | ~68% | ~22% | ~90% | Profitable — lower expense ratio is the digital advantage; early years may show higher loss ratios as they scale |
7.2 Life Insurance — Metrics That Matter
Life insurance uses different metrics because the product structure is different (long-term contracts, savings components, mortality risk):
- New Business Premium (NBP): The premium from policies sold in the current year. A growth metric. LIC still dominates with ~59% of first-year premium, but private players are gaining share every year.
- Assets Under Management (AUM): The total invested assets backing policyholder liabilities. LIC's AUM exceeds ₹50 lakh crore — making it India's largest institutional investor.
- Value of New Business (VNB): The present value of future profits from new policies written in the current year. The life insurance equivalent of "how much value did this year's sales create?" VNB margin (VNB ÷ New Business Premium) is a key metric — private players target 25–30% margins.
- Persistency (13th and 61st month): What percentage of policies are still in force 13 months and 61 months after sale? These are the most watched metrics in life insurance because acquisition costs are recovered over the first 3–5 years of premium.
📊 Exercise 7.1 — Interpret the Numbers
Read the comparative tables in this section (combined ratio by insurer type, life insurance metrics) and answer the questions below with reasoning. There is no single "right" answer — the quality of your reasoning is what matters.
- Public-sector general insurers have a combined ratio of ~110% — yet they continue to operate. How is this possible? Explain the mechanism that keeps them in business despite the underwriting loss.
- Digital-first insurers have expense ratios around 22% vs ~30% for traditional private insurers — but their loss ratios are similar. Why does a lower expense ratio NOT automatically translate into a lower combined ratio? What does this imply about digital insurers' risk selection?
- A life insurer's 13th-month persistency drops from 80% to 74% in 6 months. What is the likely financial impact? What would you investigate first to understand the cause?
View Solution — Model Interpretations
1. How can a 110% combined ratio be sustainable?
Three mechanisms keep a loss-making insurer operating:
- Investment income covers the underwriting loss: The insurer earns 5–8% on its investment portfolio (the float + capital). If the float is large relative to premiums, investment income can exceed the 10% underwriting loss. This is the same mechanism explored in Exercise 1.1 — but it means the insurer is dependent on investment markets.
- Capital reserves absorb the loss: The insurer draws down its solvency margin. This is unsustainable long-term — if the solvency ratio falls below 1.5, IRDAI intervenes. The insurer must eventually either fix underwriting or raise capital.
- Cross-subsidisation: Profitable lines (e.g., health, travel) subsidise the loss-making motor TP line. The combined ratio of 110% is the portfolio average — some lines are below 100%, others above 120%.
2. Why doesn't a lower expense ratio guarantee a lower combined ratio?
Because the combined ratio = loss ratio + expense ratio. A digital insurer with a 22% expense ratio and a 68% loss ratio has a combined ratio of 90% — excellent. But the key phrase in the question is "loss ratios are similar." A digital insurer with a 22% expense ratio but a 78% loss ratio has a combined ratio of 100% — no better than a traditional insurer at 30% expense ratio + 70% loss ratio.
The implication: digital insurers' lower distribution costs do not automatically mean better profitability. If their automated underwriting is less selective than a human underwriter (accepting more high-risk applicants in pursuit of volume), the loss ratio rises to offset the expense advantage. The sustainable competitive advantage of digital insurers is not the expense ratio alone — it is the ability to maintain a LOWER loss ratio through better data-driven risk selection AND a lower expense ratio simultaneously. When a digital insurer's loss ratio is similar to a traditional insurer's, the expense advantage is real but modest — and could be erased by any claims experience deterioration.
3. Persistency drop from 80% to 74% — impact and investigation
Financial impact: A 6-point drop in persistency is significant. Life insurance acquisition costs are recovered over 3–5 years. A 74% persistency means 26% of policies lapse before their acquisition cost is fully recovered — the insurer is effectively paying acquisition costs for customers who do not stay long enough to repay them. A 6-point persistency drop can reduce new-business value (NBV) by 10–20% and increase the "strain" on the insurer's capital.
First investigations:
- Segment the drop: Is the decline across ALL products or concentrated in one product/channel/agent? A drop in term insurance persistency is different from a drop in endowment persistency.
- Premium changes: Did any product have a premium increase in the last 12 months that could explain policyholders leaving?
- Claims experience: Did any cohort experience a claim denial or poor claims experience?
- Channel quality: Are agents/brokers selling to customers who are not a good fit (leading to early lapse)? Is there a "lapse risk" in a specific distribution channel?
- Economic factors: Is the policyholder base under financial stress (income decline, job losses) that would cause lapsation?
Key principle: Persistency is a leading indicator — by the time it drops, the damage is partially done. The investigation should aim to identify the cause before the next renewal cycle, so corrective action (better customer selection, retention programmes, premium adjustments) can be taken.
Hands-On Project: Build an Insurance P&L Model in Excel
You are a financial analyst at a mid-size Indian general insurer. Build a simplified P&L model in Excel and use sensitivity analysis to answer key business questions. This model will give you intuition for how insurance economics work that no amount of reading can replace.
Steps
- Set up the assumptions section with clear labels and input cells. Use the structure from Section 6.2. Make all calculations reference the assumption cells — no hardcoded numbers in formulas.
- Build the P&L with the following sections: Premium, Claims, Operating Expenses, Underwriting Result, Investment Income, and Profit Before Tax.
- Calculate the combined ratio and format it as a percentage. Add conditional formatting: green if below 100, red if above 100.
- Build a sensitivity table showing the combined ratio and PBT for different claim frequency assumptions (8%, 10%, 12%, 14%, 16%) and different average premium assumptions (₹9,000, ₹9,500, ₹10,000, ₹10,500, ₹11,000).
- Answer these questions by adjusting assumptions and observing the results: (a) What premium increase is needed to achieve a combined ratio of 95? (b) What claim frequency reduction achieves the same 95 combined ratio? (c) Which is more impactful — a 10% improvement in expense ratio or a 10% improvement in loss ratio? Why?
- Write a 200-word management brief summarizing your findings and recommending one operational priority for the coming year.
View Solution / Walkthrough
Key Answers
(a) Premium increase for 95 Combined Ratio: Starting from base case (102.8 combined ratio), the insurer needs to reduce the combined ratio by 7.8 points. Since every 1% premium increase reduces the combined ratio by approximately (Claims+Expenses)/Premium² — roughly 0.98 points at base — the required premium increase is approximately 8% (from ₹10,000 to ₹10,800). This assumes claims and expenses stay constant in absolute terms. In reality, premium increases may reduce demand, so the model should also consider volume elasticity.
(b) Claim frequency reduction for 95 Combined Ratio: A 7.8-point combined ratio improvement from claims reduction requires reducing incurred claims by approximately ₹6.66 crore. At an average severity of ₹52,000, this means approximately 1,280 fewer claims — reducing frequency from 12% to 10.72%. This is achievable through better underwriting selection, but would require rejecting or re-pricing some currently accepted risks.
(c) 10% improvement — expense ratio vs. loss ratio: In the base case: Loss Ratio = 78.8%, Expense Ratio = 24.0%. A 10% improvement in expense ratio (from 24.0 to 21.6) saves ₹2.40 crore. A 10% improvement in loss ratio (from 78.8 to 70.9) saves ₹6.74 crore. The loss ratio improvement is 2.8× more impactful because claims are a much larger portion of costs than expenses. For insurers with combined ratios above 100, the loss ratio is almost always the bigger lever — but it is also the harder one to pull without sacrificing growth (raising prices) or increasing risk (loosening underwriting).
Management Brief (Sample)
Our P&L analysis reveals the company is operating at a combined ratio of 102.8%, resulting in a pre-tax loss of approximately ₹3.1 crore despite positive investment income. While investment returns partially offset underwriting losses, this is not a sustainable strategy — investment yields are volatile, and the core insurance operation must achieve standalone profitability. Sensitivity analysis indicates that reducing claim frequency by 1.3 percentage points (from 12.0% to 10.7%) through tighter underwriting selection would restore underwriting profitability (combined ratio 95%) and deliver a pre-tax profit of approximately ₹5 crore — superior to the premium increase alternative, which risks volume loss in a competitive market. We recommend prioritizing underwriting discipline: implement stricter risk selection guidelines for motor and health lines (which account for 70% of our portfolio), invest in a predictive underwriting model to identify high-risk applicants before policy issuance, and set a 12-month target of reducing the combined ratio to 98 — with quarterly monitoring of the loss ratio by line of business.
3-2-1 Reflection — Before You Move On
Insurance economics is a numbers game. Write down what you can now calculate from memory — retrieval practice makes the formulas stick.
3 Things I Learned Today
2 Ratios I Can Now Calculate From Memory
1 Question I Still Have About Insurance Economics
Key Takeaways
Insurers make money two ways — underwriting profit and investment income on the float. The most powerful insurance business model generates profits from both simultaneously.
The combined ratio (loss ratio + expense ratio) is the single most important metric in general insurance. Below 100 = underwriting profit. Above 100 = underwriting loss that investment income must cover.
Claims leakage of 5–10% represents a massive financial opportunity — and the primary business case for AI-based claims automation in insurance.
A 1-point improvement in combined ratio on a ₹5,000 crore premium base is worth ₹50 crore to the bottom line. Insurers obsess over combined ratio for good reason.
Building the Excel P&L model with sensitivity analysis teaches you more about insurance economics than any textbook. The model reveals that loss ratio improvement is typically 2–3× more powerful than expense ratio improvement — but much harder to achieve.
Test Your Understanding
1. An insurer's combined ratio is 104. What does this mean?
2. What is the insurance "float"?
3. The pure premium component of an insurance premium covers:
4. What does IBNR stand for and why is it significant?
5. In the underwriting process, after receiving an application, the underwriter's first substantive step is to: