Chapter 8 Quiz

DataField.Dev

Chapter 8 Quiz

Twenty questions to check your grasp of information gathering: the sources, what each can and cannot prove, and the rules that govern their use. Answers are in the collapsed key at the bottom — try the whole quiz before opening it.

Multiple choice

1. The single most important reason an underwriter cannot simply accept whatever a broker submits is that:

A. brokers are usually dishonest
B. the submission is curated by parties with an interest in the underwriter's "yes," so it reflects what survived their judgment about what helps the case
C. applications are illegal to rely on
D. the underwriter is required by law to re-type every field

2. On a commercial application, a key underwriting question is left blank. The disciplined underwriter treats this as:

A. equivalent to a "no" / zero
B. a clerical issue to ignore
C. a missing answer and a red flag — go back and get it filled before quoting
D. grounds to automatically decline

3. The most valuable document in a commercial file, read for cause and trajectory rather than total dollars, is the:

A. declarations page
B. ACORD application
C. loss run
D. certificate of insurance

4. A loss run showing many small claims and no large one is primarily a signal about:

A. a single catastrophic hazard
B. management and safety culture (a frequency signal)
C. bad luck
D. the building's roof

5. The MVR is best described as:

A. a shared cross-carrier claims database
B. a credit-derived score predicting insurance loss
C. the official state driving record (license status, violations, at-fault accidents)
D. a physical survey of a vehicle

6. CLUE answers which specific adverse-selection move?

A. an applicant who pays premiums late
B. an applicant who switches carriers precisely to escape a claims history the prior carrier's loss run would show
C. an applicant who buys too much coverage
D. an applicant who lives in a flood zone

7. The industry's actual claim about the credit-based insurance score is that it:

A. proves a specific applicant is careless
B. shows that credit causes accidents and fires
C. is statistically correlated with future loss across large populations, even after controlling for other factors
D. measures creditworthiness for a loan

8. Which statement about the credit-based insurance score is true?

A. it is permitted identically in every state
B. it can be statistically valid and ethically contested at the same time, and several states restrict or ban it
C. it is banned everywhere in the United States
D. it has no documented correlation with loss

9. The primary value of an inspection over the application is that it:

A. is cheaper and faster than reading documents
B. independently verifies physical facts, surfaces hazards the applicant no longer notices, and produces loss-control recommendations
C. eliminates the need for loss runs
D. is performed by the applicant

10. An inspection's most important limitation is that it:

A. cannot photograph anything
B. is a snapshot of a single day and depends heavily on the inspector's skill
C. is illegal in most states
D. always contradicts the application

11. An underwriter cares about an applicant's financial condition mainly because:

A. it is fun to read balance sheets
B. financial distress is a leading indicator of loss — deferred maintenance, cut training, and at the extreme, moral hazard
C. insurers lend money to applicants
D. it is required for every personal-auto policy

12. Under the FCRA, an adverse action that triggers a consumer notice includes:

A. only an outright denial of coverage
B. denying, cancelling, non-renewing, or charging a higher premium based in whole or part on a consumer report
C. only cancellation
D. any underwriting decision whatsoever

13. The FCRA obligation underwriters most often get wrong is:

A. sending a notice when they decline
B. forgetting that charging more because of a consumer report also requires the adverse-action notice
C. pulling reports for applicants
D. keeping the file

14. Proxy discrimination refers to:

A. hiring a proxy to underwrite for you
B. using a permitted, ostensibly neutral factor that stands in for a protected characteristic, producing a discriminatory result
C. declining every applicant in a ZIP code
D. a type of reinsurance

15. "It's just data, and the data predicts loss" is not a complete defense for using a factor because:

A. data never predicts anything
B. a factor can be predictive and an illegal proxy at the same time — predictive because it correlates with a protected class
C. underwriters are not allowed to use data
D. prediction is irrelevant to underwriting

Short answer

16. A loss run shows a property fire history that is bad in years 1–2 and clean in years 3–5. Explain why this trajectory might be reassuring — and the one thing you would confirm before relying on it. (§8.2)

17. State precisely what kind of tool a credit-based insurance score is ("a correlation-based class factor, not a fact about the individual"), and why that distinction matters for fairness. (§8.3)

18. You raise a personal-home applicant's premium partly because of information in a CLUE report. List the three things the FCRA adverse-action notice must do. (§8.6)

19. Explain the chapter's claim that "the data tells you what is there; the underwriter notices what is not," using one concrete example of a missing item in a submission. (§8.7)

20. Why is the non-renewal of Harbor Steel by its prior carrier a reason to order independent loss runs rather than rely on the applicant's loss summary? Tie your answer to adverse selection. (§8.2, §8.3, The Underwriting File)

Answer key (try the whole quiz first)

**Multiple choice** 1. **B** — the submission reflects what the applicant and broker chose to include; both are honest but interested, so the underwriter must order independent reports and notice what is missing. 2. **C** — a blank on a loss-predicting question is a missing answer to be filled, never inferred as a favorable "no." Adverse selection lives in exactly these gaps. 3. **C** — the loss run is the most valuable commercial document; read it for cause and trajectory before the total dollars. 4. **B** — high frequency (many small claims) is usually a management/safety-culture signal that eventually produces a large loss. 5. **C** — the MVR is the official state driving record. 6. **B** — CLUE shows claims filed across *all* carriers, catching the applicant who switches to escape a history the single prior carrier's run would reveal. 7. **C** — the claim is purely statistical correlation across populations, not causation and not a judgment about an individual. 8. **B** — it is both statistically valid and genuinely contested; state treatment varies, with some restricting or banning it. 9. **B** — the inspection verifies physical facts, surfaces unnoticed hazards, and yields loss-control recommendations (your future conditions). 10. **B** — it is a single-day snapshot whose quality depends on the inspector. 11. **B** — financial distress precedes loss through deferred maintenance, cut training, and, at the extreme, moral hazard. 12. **B** — adverse action includes *charging a higher premium*, not only denial. 13. **B** — forgetting that charging more triggers the notice is the classic, scalable error. 14. **B** — a neutral-looking permitted factor standing in for a protected characteristic. 15. **B** — predictiveness can coexist with being an illegal proxy; predictiveness is the start of the analysis, not the end. **Short answer** 16. The trajectory suggests the risk may have *fixed* a problem — a hazard addressed, management improved, losses trailing off — which is far better than a clean-early/bad-late pattern. Before relying on it, confirm *what changed*: the specific corrective action behind the improvement (a new manager, a control installed, a process changed) and that it is durable, not just a lucky stretch. A loss run shows the losses, not the response; you have to verify the response. 17. A credit-based insurance score is a *correlation-based class factor*: it tells you that applicants who look like this person, *as a group*, have run higher loss ratios. It is **not** a fact about whether this particular individual is careless. The distinction matters because pricing a person on a group-level statistical signal — especially one shaped by income, historical discrimination, and life events unrelated to risk behavior — is exactly where the line between actuarial validity and social fairness becomes contested, and where regulation (state by state) intervenes. 18. The notice must: (1) tell the consumer that an adverse action was taken; (2) identify the consumer- reporting agency that supplied the report so the consumer can obtain a copy; and (3) inform the consumer of their right to dispute the accuracy of the information and to obtain a free copy of their report. (For credit-based insurance scores, the key factors that adversely affected the score must also be disclosed.) 19. A model scores the fields it is given but has no concept of the field that *should be there and isn't*. Example: a fabrication plant's submission that never mentions a hot-work permit program — when hot work is the obvious hazard and caused one of its fires — is telling you something by omission. The underwriter notices the absence and goes back to the broker to fill it; the model cannot ask a question it does not know it needs to ask. 20. The prior carrier walked away citing catastrophe exposure *and* loss history, so its own view of the account is unfavorable — which is information, not noise. Relying on the applicant's self-prepared loss summary would let exactly the unfavorable history that drove the non-renewal go unseen; that is the adverse-selection trap, where the account most likely to have hidden losses is the one being shopped to a new carrier. Independent, currently-valued loss runs (and a cross-carrier claims history) show what actually happened rather than what the applicant chose to summarize.