Chapter 33 Quiz

Twenty questions to check your grasp of fraud, misrepresentation, rescission, the SIU, and fraud analytics — fifteen multiple-choice and five short-answer. Answers are in the collapsed block at the bottom; try the whole set before opening it. Keep the chapter's bright line in mind: a red flag is a prompt to investigate, never a finding of fraud.

Multiple choice

1. In the long run, who principally bears the cost of insurance fraud?

  • A. The insurer's shareholders, permanently
  • B. The honest policyholders in the same pool, through the premium
  • C. The state fraud bureau
  • D. Reinsurers, through catastrophe treaties

2. Which of the following is the clearest example of hard (premeditated) fraud?

  • A. Rounding payroll from \$11.3M down to "about \$11M"
  • B. Adding a few never-stolen items to a real burglary claim
  • C. Staging a collision with paid "victims" to collect on a policy
  • D. Forgetting a small claim from several years ago

3. A material misrepresentation requires all of the following EXCEPT:

  • A. A false statement of fact
  • B. Materiality (it would have changed the underwriting decision)
  • C. The insurer's reliance on the statement
  • D. A criminal conviction of the applicant

4. The classic test for whether a misstatement is material is whether:

  • A. The applicant intended to deceive
  • B. A reasonable, prudent underwriter knowing the truth would have declined, charged more, or changed terms
  • C. The misstatement was made in writing
  • D. The loss exceeded the policy limit

5. Concealment differs from misrepresentation chiefly in that concealment:

  • A. Can never support rescission
  • B. Is an active false statement rather than a silence
  • C. Is a material silence, and additionally requires that the applicant knew the fact and its materiality
  • D. Applies only to life insurance

6. Rescission of a policy means:

  • A. Ending coverage at the next renewal
  • B. Denying one disputed claim under a valid policy
  • C. Voiding the policy from its inception, returning premium and treating it as if it never existed
  • D. Increasing the premium retroactively

7. Which is a recognized limit on the insurer's power to rescind?

  • A. Incontestability periods (notably in life insurance)
  • B. The doctrine of innocent misrepresentation
  • C. Waiver or estoppel if the insurer knew and issued anyway
  • D. All of the above

8. Post-claim underwriting refers to the disfavored practice of:

  • A. Underwriting a renewal after the prior term's claims are known
  • B. Doing the verification at claim time and using a trivial discrepancy as a pretext to rescind and avoid a valid claim
  • C. Auditing payroll after the policy period ends
  • D. Referring every claim to the SIU

9. The cardinal rule about a single red-flag indicator is that it is:

  • A. Conclusive proof of fraud
  • B. A prompt to investigate, never a finding of fraud
  • C. Grounds for immediate rescission
  • D. Irrelevant unless the claim is large

10. Which combination most warrants an SIU referral (as opposed to a routine question)?

  • A. A single rounded payroll figure
  • B. One small forgotten claim, otherwise clean
  • C. A cluster across families — over-insurance, urgency to bind, a concealed loss, and shifting addresses
  • D. An applicant who is privacy-minded about their home address

11. The relationship between the underwriter and the SIU is best described as:

  • A. The underwriter investigates; the SIU sells policies
  • B. The underwriter spots and refers; the SIU investigates and substantiates
  • C. The SIU sets the price; the underwriter handles claims
  • D. They are the same function under different names

12. "Most SIU referrals come back cleared." The chapter treats this as:

  • A. Evidence the SIU is failing
  • B. A feature — a referral is a question for specialists, and clearing the innocent protects the honest customer
  • C. A reason underwriters should stop referring
  • D. Proof that fraud is rare

13. Link analysis is the fraud-analytics technique most lethal to:

  • A. Honest size-outlier businesses
  • B. Soft fraud committed by individuals
  • C. Organized fraud rings, because it refuses to view connected policies in isolation
  • D. Misstated payroll

14. A high fraud score from a predictive model should be treated as:

  • A. A finding sufficient to rescind the policy
  • B. A reason to deny the claim immediately
  • C. A strong prompt to investigate, which the SIU must substantiate or clear with evidence a human can defend
  • D. Conclusive proof of intent

15. A fraud model trained on a decade of past investigations risks learning:

  • A. Who actually committed fraud, perfectly
  • B. Who got investigated — perpetuating any historical skew, a proxy-discrimination danger Chapter 35 owns
  • C. The correct price level for the class
  • D. Nothing, because models cannot be biased

Short answer

16. In one or two sentences, explain why fraud is described as "adverse selection in its most deliberate form," and name the two ordinary underwriting tools that double as fraud defenses. (§33.1, §33.3)

17. A loss-run order proves an application answer was false. State precisely what this does establish and what it does not establish, and name the element that separates a clarifiable gap from a rescindable fraud. (§33.3)

18. Give the three-part response the chapter prescribes when a single red flag surfaces (ask before you assume; ask in writing and document; route by severity), and explain in one sentence why "ask in writing" is not just bureaucracy. (§33.5)

19. Explain why the innocent-explanation column of the red-flag table is a genuine part of the detection method. What statistical error does an underwriter make who ignores it? (§33.5)

20. In the Harbor Steel file, the application shaded the cause of the 2023 fire. Explain in three to four sentences why this is a red flag and a clarification but not fraud and not a rescission issue — referencing materiality, intent, and the absence of a cluster. (The Underwriting File, §33.2–§33.4)

Answer key (try the whole quiz first) **1. B** — Fraud losses are passed into the price; the honest insureds in the pool ultimately pay, along with the friction cost of the controls built to catch fraud. (§33.1) **2. C** — Hard fraud is the deliberate *manufacture* of a loss; a staged collision is the textbook case. A and B are soft fraud (shading/padding a real situation); D is honest error. (§33.2) **3. D** — Materiality, falsity, and reliance are the elements; a criminal conviction is *not* required to establish a material misrepresentation for civil/contract purposes such as rescission. (§33.3) **4. B** — The prudent-underwriter test: would the truth have changed the decision, the terms, or the price? Intent is a separate question (and required only in some states). (§33.3) **5. C** — Concealment is a material *silence* and additionally requires that the applicant knew the fact and knew, or should have known, it was material — which is why it is harder to prove. (§33.3) **6. C** — Rescission unwinds the contract from inception; it is far heavier than non-renewal (forward only) or a single coverage denial under a valid policy. (§33.4) **7. D** — All three are genuine limits; rescission is a hedged remedy, not a free undo button. (§33.4) **8. B** — Post-claim underwriting substitutes claim-time verification for underwriting-time verification and uses a pretextual discrepancy to escape a valid claim; courts and regulators disfavor it. (§33.4) **9. B** — A red flag is a prompt to look, never a conclusion; each flag usually has an innocent explanation. (§33.5) **10. C** — A *cluster* across different families is what warrants a referral; any single flag (A, B, D) is at most a routine question. (§33.5) **11. B** — The underwriter is the front line of detection (spots and refers); the SIU is the specialist of investigation (investigates and substantiates). (§33.6) **12. B** — Clearing the innocent diligently is how the system protects honest customers and avoids bad-faith exposure; a referral is a question, not a verdict. (§33.6) **13. C** — A ring's defense is that each policy looks innocent alone; link analysis exposes the shared connections precisely because it refuses to view them in isolation. (§33.7) **14. C** — A score is a likelihood, not a fact; it decides where to *look*, and the SIU establishes the truth with defensible evidence — the fraud-side analogue of the documented override. (§33.7, Ch.32) **15. B** — The model can learn *who got investigated* rather than *who committed fraud*, perpetuating historical skew — the proxy-discrimination/algorithmic-bias problem Chapter 35 treats in full. (§33.7, Ch.35 preview) **16.** Fraud is adverse selection with *intent* — the applicant or claimant does not merely expect a loss but plans to manufacture, exaggerate, or conceal one and lies to make the policy pay. Two ordinary tools that double as fraud defenses: the application (its material questions) and the loss run / CLUE order (which exposes concealed history). Inspections and aligned-incentive terms (deductibles) also qualify. (§33.1, §33.3) **17.** The loss-run order *establishes the discrepancy* — it proves the answer was false. It does *not* establish that the applicant *knew* and *intended to deceive*. That intent element is what separates a clarifiable gap (honest oversight) from a rescindable fraud, and data cannot supply it — only investigation and the applicant's own account can. (§33.3) **18.** (1) *Ask before you assume* — give the applicant a fair chance to explain, because the explanation is usually innocent. (2) *Ask in writing and document the answer* — both because the answer may itself become evidence (a second false explanation is far more damning than the first omission) and because the file must show responsible handling. (3) *Route by severity* — clear a trivial flag yourself; refer a serious flag or a cluster to the SIU rather than playing investigator. "Ask in writing" is not bureaucracy because the written record protects the honest applicant and builds the defensible file if it is not. (§33.5) **19.** For every red-flag family there is an innocent explanation that is *usually the true one* (most distressed businesses never commit fraud; most urgency is a real deadline). Ignoring that column makes the underwriter commit a base-rate error — treating a weak, common indicator as strong evidence of a rare event — which produces false accusations, soured relationships, and bad-faith exposure. (§33.5) **20.** The shaded cause is *material* (the cause of a fire bears on the hazard and the controls, so it would have affected the terms), which makes it a red flag requiring a response. But materiality is not fraud: fraud needs *intent*, and innocent explanations (a coarse application "cause" category; the broker's summarization; an honest characterization of a welding fire as "accidental") are at least as likely. There is no *cluster* — the fire's existence and size were honestly disclosed, there is no over-insurance, urgency, or distress, and a strong broker has attached corrective controls. So it is a documented *clarification*, not an SIU referral, and nothing approaches the material-fraud-procured-the-policy standard rescission requires. (The Underwriting File, §33.2–§33.4)