CH 2

ESTIMATING HAZARD RISK

EDUCATIONAL OBJECTIVES

  • After learning the content of this assignment, you should be able to:

    • Explain how to analyze and evaluate hazard risk, distinguishing between qualitative and quantitative methodologies.

    • Explain how to estimate expected losses arising from hazard risk, incorporating various adjustment factors.

    • Explain how to apply increased limit factors to hazard loss estimates, particularly for high-severity, low-frequency events.

    • Explain how to estimate the volatility of hazard losses, utilizing probability distributions for better financial planning.

OUTLINE

  • Analyzing and Evaluating Hazard Risk

  • Steps in Estimating Hazard Losses

  • Applying Increased Limit Factors

  • Estimating Hazard Loss Volatility

  • Summary

ANALYZING AND EVALUATING HAZARD RISK

  • Introduction to Hazard Risk Analysis

    • Risk management professionals must determine when to use quantitative versus qualitative methods in analyzing and evaluating an organization's hazard risks for optimal resource allocation. The choice between methods often depends on data availability, the nature of the risk, and the desired level of precision.

  • Types of Risk Analysis

    • Quantitative Analysis: Involves the use of statistical data to predict future losses based on historical data. This method is often preferred when ample, reliable historical data is available for a risk and when financial precision is required.

    • Qualitative Analysis: Utilizes the knowledge and experience of experts to evaluate risks where data is insufficient, unreliable, or non-existent. It is particularly useful for emerging risks, unique exposures, or when rapid preliminary assessments are needed.

    • May include methods such as scenario analysis and failure mode and effects analysis (FMEA).

  • Importance of Risk Evaluation

    • Determines how much of the estimated future losses should be retained or transferred, thereby influencing broader risk treatment decisions such as avoidance, reduction, or transfer through insurance or other mechanisms.

QUANTITATIVE VERSUS QUALITATIVE ANALYSIS

  • Definition of Hazard Risk: A pure risk with only two potential outcomes: a loss or no loss (no chance for gain). Unlike speculative risks (e.g., investing in the stock market), hazard risks (like property damage or liability claims) only present the possibility of adverse outcomes.

    • Focuses solely on potential losses rather than gains.

  • Quantitative Analysis Details

    • Sources of Data:

      • Internal records of past losses from the organization. These can include detailed claims databases, incident reports, and financial loss summaries.

      • Insurance industry data, often aggregated from numerous organizations by actuarial bureaus (e.g., ISO, NCCI) or specialized consulting firms.

    • Internal data is generally more accurate as it entails all types of losses including uninsured ones, providing a comprehensive view specific to the organization's operations.

    • Risks of Relying on Internal Data:

      • Small sample sizes: Insufficient number of past loss events to derive statistically reliable predictions.

      • Limited timeframes: History might not represent future conditions or capture rare, severe events.

      • Lack of operational diversification: Data from a single entity may not reflect broader industry trends or the impacts of varying operational contexts.

  • Qualitative Analysis Techniques

    • Techniques include:

      • Scenario Analysis: Analyzing various future loss scenarios without relying exclusively on past data. This involves identifying potential events, exploring their causes and consequences, and developing 'what-if' stories. It often generates multiple outcomes: optimistic, pessimistic, and most likely scenarios to understand a range of potential impacts.

      • Failure Mode and Effects Analysis (FMEA): A systematic, proactive method for identifying and evaluating potential failure modes in a process or design, assessing their effects, and planning to prevent future losses based on historical precedents and expert judgment. Key steps often include identifying potential failure modes, their effects, causes, current controls, and assigning severity, occurrence, and detection ratings to calculate a Risk Priority Number (RPN).

HAZARD RISK EVALUATION

  • Definition by ISO: Risk level as the magnitude of a risk or combination of risks expressed in terms of consequences (impact or outcome of an event) and likelihood (the chance of something happening).

  • Budgeting Based on Risk Level:

    • Organizations budget for expected losses based on established criteria and the evaluations of risk retention. This involves setting self-insured retentions (SIRs), deductibles, and allocating capital for anticipated but unpredictable loss events.

    • Correlation of hazard risk with operational, financial, and strategic risk types is minimal, providing diversification benefits in a holistic risk portfolio.

  • Factors Influencing Retention Decisions:

    • Cost of insurance premiums: The premium charged versus the cost of internal retention.

    • Financial conservatism of the organization: An organization's willingness to bear financial risk, often reflected in its risk tolerance policy.

    • Other factors include: regulatory requirements, balance sheet strength, access to capital, cash flow stability, tax implications, and the organization's overall risk appetite.

STEPS IN ESTIMATING HAZARD LOSSES

Overview of the Procedure

  1. Collect and Organize Past Data

    • Ideally collect a minimum of five years of past exposure and loss data, striving for 10 or more years for greater statistical credibility. Exposure data refers to underlying variables that correlate with loss (e.g., sales revenue, payroll, number of vehicles, square footage).

    • Key Types of Data: Paid losses (amounts already paid out), reserves (estimated future payments for reported but unsettled claims), and loss adjustment expenses reserves (estimated costs of handling claims, including legal fees, investigation costs, and administrative overhead).

    • Formula for Incurred Losses:

      • Incurred ext{ }losses = Paid ext{ }losses + Loss ext{ }reserves + Loss ext{ }adjustment ext{ }expense ext{ }reserves

  2. Limit Individual Losses

    • Capping individual losses to prevent extreme, infrequent values (catastrophic events) from skewing the forecast for typical losses (e.g., capping losses at 50,000). This practice helps separate predictable, frequent losses from unpredictable, severe events, which are often handled differently in insurance pricing and risk financing.

  3. Apply Trend and Loss Development Factors

    • Trend Factors: Adjusts historical data for changes in economic, social, and legal conditions (e.g., inflation affecting property values or medical costs, changes in legal precedents impacting liability awards, or increased frequency of lawsuits).

    • Loss Development Factors: Used specifically for 'long-tail' liability losses where claims are reported and paid over time. These factors project reported losses to their 'ultimate' value, accounting for claims that emerge later or develop in severity after initial reporting.

      • Example of Ultimate Loss Development Factor Application: A final total of claims is estimated based on past data, extrapolating from immature loss data to what the total cost will be once all claims are closed.

  4. Forecast Future Losses

    • Comparing adjusted total losses and exposure for forecasting future losses based on the organization's projected sales data or other relevant exposure bases. This can involve statistical methods such as regression analysis, actuarial projection techniques, or simple averages, depending on data quality and volume.

APPLYING INCREASED LIMIT FACTORS

  • Purpose: To estimate total anticipated large hazard losses, especially when an organization's internal loss data is insufficient to reliably predict high-severity, low-frequency events. These factors provide a way to estimate the cost of higher limits of coverage.

Overview of the Procedure

  1. Develop increased limit factors from aggregated insurer data, usually provided by actuarial bureaus (like ISO) or specialist consultants, which pool data from many policyholders.

  2. Calculate the factor for specific loss layers (e.g., losses between 50,000 and 100,000) using established industry tables or actuarial methodologies.

  3. Forecast losses across various limit scenarios (e.g., assuming a 100,000 limit versus a 1,000,000 limit) based on established increased limit factors applied to primary layer losses.

Example Application:

  • Tarnton Company’s Forecasting Example:

    • Forecasted Loss for Basic Limit (50,000): 245,000

    • Total Losses Expected (0 to 1,000,000): 510,000 from calculating the differences in forecasts and limits using the increased limit factors.

ESTIMATING HAZARD LOSS VOLATILITY

  • Importance of Volatility Estimates: Provides critical insight into potential variations around the expected loss forecasts, indicating the degree of uncertainty and enabling better financial preparedness, capital allocation, and risk financing decisions.

  • Probability Distributions: Different distributions help understand the total losses, loss frequency, and severity. Common distributions include Poisson or Negative Binomial for frequency, and Lognormal, Gamma, or Weibull for severity. Compound Poisson distributions are often used to model total losses by combining frequency and severity.

    • Expected Value Calculation: Calculates long-term predictions for total losses by combining frequency and severity distributions. The expected total loss is typically the product of the expected frequency and the expected severity (E[Total ext{ }Loss] = E[Frequency] imes E[Severity]).

SUMMARY

  • Organizations utilize risk analysis to estimate future losses through either quantitative (data-driven) or qualitative (expert judgment) methods, chosen based on data availability and risk characteristics.

  • Forecasting loss involves systematic steps including collecting comprehensive past data, limiting individual losses to manage skewing, applying trend and loss development factors for accuracy, and ultimately predicting future losses.

  • Increased limit factors assist in estimating higher risk exposures when original internal data may not suffice, providing industry-standard projections for large claims.

  • Analyzing loss volatility provides organizations an understanding of potential financial impacts due to loss fluctuations, aiding in capital planning and solvency management.

ADDITIONAL METHODOLOGY

  • Probability Intervals: Used to determine likely loss variability ranges (e.g., 95\% confidence intervals, Value at Risk, or Conditional Value at Risk scenarios), assisting with financial preparedness and optimal capital allocation for organizations under various stress events.