Chapter 6 — Further Reading

Sources are grouped by the book's three citation tiers. Tier 1 is verified canonical material we can stand behind; Tier 2 is real practice and literature whose exact citation should be checked against a current source; Tier 3 is the constructed teaching material used in this chapter. Where a source's specific figures change over time or vary by edition, treat the concept as reliable and verify any number before you quote it.

If you read only one thing

Read the chapters on risk, peril and hazard, and loss frequency vs. loss severity in a standard risk-and-insurance text such as Rejda & McNamara, Principles of Risk Management and Insurance. It is the cleanest available treatment of the exact vocabulary this chapter is built on — pure vs. speculative risk, the families of hazard, and the two dimensions of loss — and it is the foundation the AINS and CPCU curricula assume.

Tier 1 — Verified canonical

  • George E. Rejda & Michael McNamara, Principles of Risk Management and Insurance (Pearson, multiple editions). The standard academic survey; the early chapters define risk, peril, hazard (physical, moral, morale, and increasingly legal), pure vs. speculative risk, and the loss-frequency/loss-severity framework with care. The single best companion to this chapter.
  • The Institutes (American Institute for Chartered Property Casualty Underwriters), AINS and CPCU curricula. The professional body that defines the working vocabulary of risk and underwriting in North America; its foundational courses cover this chapter's terms as tested material. The reference for any reader on the certification path.
  • Hurricane Andrew (1992) — public record. Government, scientific (National Hurricane Center / NOAA), and journalistic documentation of the storm and its insurance aftermath; a Tier-1 reference point for catastrophe severity, the failure of the independence assumption, and the rise of catastrophe modeling (developed in Case Study 1; deepened in Chapters 27, 28, and 30). Use public facts only; the precise insured loss varies by source.
  • Asbestos liability — public record. The extensively documented history of asbestos-related disease, litigation, manufacturer bankruptcies, and the resulting long-tail insurance losses; a Tier-1 reference for latent hazard and the limits of backward-looking loss data (developed in Case Study 2; the occurrence vs. claims-made response is owned by Chapters 21 and 24). Credible cost totals span a wide range and continue to be revised — keep any figure qualitative.

Tier 2 — Attributed, specifics to verify

  • Catastrophe-modeling literature and vendor documentation (e.g., the major cat-model providers). The industry's standard treatment of how hazard, vulnerability, and financial modules turn a peril into a loss distribution, and how the maximum severity (PML) is estimated. The concepts are well established; specific model outputs and methodologies differ by vendor and vintage and should be checked. (Chapter 30 owns this material.)
  • Actuarial and pricing texts on frequency–severity modeling (the loss-cost literature). The standard treatment of why frequency and severity are modeled as separate distributions (count vs. amount) and why severity is heavy-tailed. The structure is canonical; particular distributional assumptions and parameters are context-specific. (Chapters 10 and 32 build on this.)
  • Insurance Information Institute (Triple-I) and similar industry resources on perils, hazards, and the cost of catastrophes. Useful for current context and definitions written for a general audience; verify any specific statistic against the original source, as figures are updated frequently.
  • Risk-management literature on moral and morale hazard and on risk classification. The distinctions are standard; empirical magnitudes (how much insurance changes behavior, how predictive a given classification variable is) are active research areas, so treat quantified behavioral claims with caution.

Tier 3 — Illustrative / constructed (used in this chapter)

  • The Harbor Steel & Fabrication risk inventory — the full exposure → hazard → controls inventory across property, GL, workers' comp, auto, and catastrophe (Figure 6.1 and The Underwriting File). A constructed teaching example built to be realistic; not drawn from any real account.
  • David Okafor, the borderline life applicant — the 45-year-old \$1M term-life applicant used to illustrate whole-person risk classification (§6.5). Constructed; worked in full in Chapter 17.
  • The frequency-vs-severity worked examples — the two \$75,000-expected-loss fleets (Fleet A high-frequency, Fleet B high-severity) and the collision-loss arithmetic. Illustrative round numbers chosen to make the two dimensions legible.
  • The exposure-base table and all sample rates — constructed correspondences and figures used to show how a base tracks the risk; not the actual rates of any insurer.