Insurance and Risk Management

The Concept and Anatomy of Risk

Risk is defined in insurance economics as uncertainty concerning the occurrence of a loss:

• Peril: The specific direct cause of a loss (e.g., fire, flood, collision, earthquake, theft).
• Hazard: A condition that creates or increases the frequency or severity of a loss from a peril:
  • Physical Hazard: A physical condition that increases the chance of loss (e.g., defective electrical wiring, icy mountain highways).
  • Moral Hazard: Dishonesty or character defects in an individual that increase the frequency or severity of a loss (e.g., fraudulent arson to collect insurance proceeds).
  • Morale (Attitudinal) Hazard: Carelessness or indifference to loss because an individual possesses insurance (e.g., leaving doors unlocked because property is insured).
  • Legal / Regulatory Hazard: Characteristics of the legal system or regulatory environment that boost the frequency or severity of claims (e.g., adverse court litigation trends).

---

Classification of Risk

Taxonomy of Risk
  - Pure Risk vs Speculative Risk
      - Pure Risk: Only possibilities of loss or no loss (insurable)
      - Speculative Risk: Possibility of loss, no loss, or gain (uninsurable)
  - Fundamental Risk vs Particular Risk
        - Fundamental (Systemic) Risk: Affects entire economy or large population (flood, war)
        - Particular (Idiosyncratic) Risk: Affects individuals or discrete firms (house fire)

---

The Economic Mechanism of Insurance and the Law of Large Numbers

Insurance is a financial arrangement that redistributes the costs of unexpected losses by pooling a large number of independent, homogeneous exposure units:

• The Law of Large Numbers: As the number of independent exposure units ($N$) increases, actual loss experience approaches the expected loss probability with increasing precision.
• Risk Reduction through Pooling: For an individual exposure unit with loss standard deviation $\sigma$, when $N$ independent identical exposures are pooled together:

  Swipe to see the full formula↔

  $$\text{Expected Loss per Unit} = E(L) = \mu$$

  Swipe to see the full formula↔

  $$\text{Standard Deviation of Average Loss per Unit} = \sigma_{\bar{L}} = \frac{\sigma}{\sqrt{N}}$$
  As $N \to \infty$, the risk (standard deviation of average loss) approaches zero, transforming individual uncertainty into predictable group certainty.

---

Requirements of an Ideally Insurable Risk

For private commercial insurers to underwrite a risk economically:

1. Large number of homogeneous exposure units (enables accurate loss forecasting via the Law of Large Numbers).
2. Accidental and unintentional loss (prevents moral hazard).
3. Determinable and measurable loss (definite in time, place, and monetary amount).
4. Non-catastrophic loss to the pool (exposures must be geographically dispersed so that a single event does not bankrupt the insurer).
5. Calculable chance of loss (enables actuaries to establish economically viable premiums).
6. Economically feasible premium (substantially less than the face value of the policy).

---

Core Concepts and Terminology: Introduction to Risk and Insurance

• Peril vs Hazard: A peril is the specific event causing a loss (e.g., earthquake, fire). A hazard is an underlying condition that increases the probability or severity of that loss occurring (e.g., storing flammable chemicals near a furnace).
• Law of Large Numbers in Insurance: A mathematical principle stating that as the number of similar, independent exposure units observed expands, the actual loss experience will converge closely toward the expected theoretical probability.
• Pure Risk: A risk category where the only possible outcomes are financial loss or no loss (with zero possibility of gain), representing the core domain of commercial insurance.

---

Numerical Illustration 1: Risk Reduction through Pooling and the Law of Large Numbers

Problem Statement: An industrial logistics firm operates a fleet of cargo trucks in the Birgunj-Kathmandu transit corridor. Actuarial historical data indicates that each truck has a 10.0% probability of experiencing an accident during any given year ($P = 0.10$). If an accident occurs, the property damage loss is exactly Rs. 500,000. If no accident occurs, the loss is Rs. 0.

Required:

1. For a single independent truck ($N = 1$):
   • Calculate the Expected Loss ($E(L)$).
   • Calculate the Variance ($\sigma^2$) and Standard Deviation ($\sigma$) of loss.
   • Compute the Coefficient of Variation (CV).
2. If the firm forms a mutual risk pool combining 100 independent trucks ($N = 100$):
   • Calculate the Expected Loss per truck.
   • Compute the Standard Deviation of the average loss per truck ($\sigma_{\bar{L}}$).
   • Compute the Coefficient of Variation of the pooled portfolio.
3. If the pool expands to 10,000 independent trucks ($N = 10,000$), calculate the new standard deviation of average loss and CV.
4. Explain how this demonstrates the mathematical engine of commercial insurance.

Step-by-Step Solution:

1. Single Truck Analysis ($N = 1$):

• States of Loss:
  • Accident: Loss $L_1 = \text{Rs. } 500,000$ (Probability $P_1 = 0.10$).
  • No Accident: Loss $L_2 = \text{Rs. } 0$ (Probability $P_2 = 0.90$).

    Swipe to see the full formula↔

    $$\text{Expected Loss } E(L) = (0.10 \times 500,000) + (0.90 \times 0) = \mathbf{\text{Rs. } 50,000}$$

    Swipe to see the full formula↔

    $$\sigma^2 = 0.10(500,000 - 50,000)^2 + 0.90(0 - 50,000)^2$$

    Swipe to see the full formula↔

    $$\sigma^2 = 0.10(450,000)^2 + 0.90(-50,000)^2 = 0.10(202,500,000,000) + 0.90(2,500,000,000)$$

    Swipe to see the full formula↔

    $$\sigma^2 = 20,250,000,000 + 2,250,000,000 = \mathbf{22,500,000,000 \text{ Rs.}^2}$$

    Swipe to see the full formula↔

    $$\sigma = \sqrt{22,500,000,000} = \mathbf{\text{Rs. } 150,000}$$

    Swipe to see the full formula↔

    $$\text{Coefficient of Variation (CV)} = \frac{\sigma}{E(L)} = \frac{150,000}{50,000} = \mathbf{3.00}$$
    An individual truck owner faces extreme risk: standard deviation is 300% of expected loss.

2. Pooled Portfolio of 100 Independent Trucks ($N = 100$):

By pooling 100 trucks, risk per truck plunges by 90% (from CV = 3.00 down to 0.30).

3. Pooled Portfolio of 10,000 Independent Trucks ($N = 10,000$):

4. Mathematical Synthesis: By expanding the pool to 10,000 vehicles, relative risk falls to an almost negligible 0.03 (a 99% risk reduction). The insurance company can comfortably charge each owner a premium of Rs. 50,000 plus an administrative loading fee (e.g., Rs. 5,000), knowing with 99.7% statistical certainty that the average loss will not exceed Rs. 54,500 ($50,000 + 3\sigma$).

---

Numerical Illustration 2: Insurable Risk Premium Calculation with Loading Factors

Problem Statement: An insurance company underwrites fire insurance for 2,000 small retail shops in Pokhara. The property value of each shop is Rs. 2,000,000. Actuarial statistics show:

• Probability of total destruction by fire: 0.20% per year ($P = 0.002$).
• The insurer incurs administrative, underwriting, and policy issuance expenses equal to 25% of the pure premium ($Loading = 0.25$).
• The insurer mandates a profit and contingency buffer of 10% of the gross premium ($Profit = 0.10$).
• The government of Nepal levies an insurance regulatory fee of 1.0% on gross premiums and a flat stamp duty of Rs. 20 per policy.

Required:

1. Calculate the Pure (Net) Premium per shop.
2. Calculate the Gross Commercial Premium per shop.
3. Compute the Total Annual Gross Premium collected from all 2,000 shops.
4. Calculate the Total Statutory Revenue remitted to the government (Regulatory Fee + Stamp Duty).

Step-by-Step Solution:

1. Pure (Net) Premium per Shop:

2. Gross Commercial Premium ($GP$): Let $GP$ = Gross Premium.

Gross Premium formula with profit margin $m = 10\%$:

3. Total Gross Premium Collected (2,000 Shops):

4. Statutory Remittances to Government:

• Regulatory Fee ($1.0\% \times 11,111,120$): Rs. 111,111.20
• Stamp Duty ($2,000 \text{ policies} \times \text{Rs. } 20$): Rs. 40,000.00

  Swipe to see the full formula↔

  $$\text{Total Remitted to Government} = 111,111.20 + 40,000.00 = \mathbf{\text{Rs. } 151,111.20}$$

Insurance & Actuarial Examination Focus: In TU examination evaluations, students often confuse perils (the specific cause of loss, like fire) with hazards (conditions that increase the probability of loss, like faulty wiring).

---

Paper Khoj independent study notes — unreviewed draft, not affiliated with Tribhuvan University.