Chapter 11 — Key Takeaways

A one-page card. Pricing and rating in the space of a desk reference.

The core claim

A premium is built, not guessed — three blocks, then modified for the individual risk — and the discipline that makes the whole thing honest is rate adequacy: charge enough for the risk accepted, especially when the soft market is pushing the other way. The punishment for getting it wrong is delayed by two or three years, which is exactly why it is so hard.

The premium build-up (memorize this)

  pure premium (expected loss = frequency × severity)      ← Chapter 10
+ expense load   (commissions, ops, taxes, overhead ~30%)
+ profit & contingencies load (margin + cushion ~5%)       ← the THIN block
─────────────────────────────────────────────────────────
= indicated MANUAL (class) rate
  then ADJUST for THIS risk:
    × experience rating   (its own losses, weighted by CREDIBILITY)
    × schedule rating     (documented credits/debits for what the numbers miss)
= the modified, charged rate

Rules of thumb

• Loss cost ≠ rate. A loss cost (ISO/NCCI) is the expected-loss component only. Multiply it by the carrier's loss cost multiplier (LCM) to get a rate. Quoting the loss cost as the rate sells the business with zero expense and zero profit.
• Permissible loss ratio = 1 − (expense % + profit %). If expenses are 30% and profit 5%, the rate is set so the pure premium fills 65% and no more. rate = pure premium ÷ permissible loss ratio.
• A relativity is a multiplier on expected loss, not a probability. 1.30 = 30% more expected loss than baseline; 0.90 = 10% less. They chain by multiplication.
• Experience rating is bounded by credibility. A thin loss history is mostly noise — weight it lightly; the class stays the better predictor. One shock loss is not a frequency trend.
• Schedule rating must be documented. Every credit/debit needs a risk-based reason tied to a filed category and must stay within the filed maximum. "We want the account" is not a category.
• A debit experience mod + a credit schedule mod is not a contradiction. Past losses justify the debit; new controls justify the credit. (This is Harbor Steel.)
• Retrospective rating hands risk back to the insured. Final premium tracks actual losses within a min/max band — partial self-insurance for large, strong accounts.
• Fast growth is a red flag for inadequacy. In a competitive market, the cheapest carrier is usually cheap because its rate is too low, not because it underwrites better.

The statutory standard (Chapter 4)

A filed rate must not be: inadequate (protects solvency) · excessive (protects consumers) · unfairly discriminatory (price differences must reflect risk, not protected class).

Key terms

manual/class rate · base rate · rating factor (relativity) · experience rating · schedule rating · expense loading · profit & contingencies loading · retrospective rating · rate adequacy · (plus: loss cost, loss cost multiplier, permissible loss ratio, minimum premium, composite rating)

Themes advanced

• Pricing follows risk — the premium must be adequate for the risk accepted, built from nameable, defensible parts.
• The combined ratio tells the truth — above 100% the underwriting loses money; soft-market underpricing shows up there, on a delay.

What you could defend to your manager

"I built Harbor Steel's property rate from the class — base rate times the construction, protection, and metal-fabrication-occupancy relativities — then modified it for the risk. Experience rating moved it only modestly: two fires in five years is low-credibility evidence on frequency, though it flags a controllable hazard. Schedule rating nets to a debit — the end-of-life roof and the housekeeping and hot-work history outweigh the genuine credits for improving management and maintained sprinklers, and the hot-work- program credit is held contingent on verification. The indicated price is debit-rated and adequate. The terms that make it work — the wind deductible, the ACV-roof endorsement, the BI period — are the next chapter. The price is built; the deal is not yet done."