# The underwriting-to-claims cycle: the engine that runs an insurer
A trucking company with 40 delivery vans applies for coverage on a Monday. By Friday, one of those vans rear-ends a sedan at a red light. Whether the insurer makes or loses money on this account comes down to a chain of decisions that started before the accident ever happened.
That chain is the underwriting-to-claims cycle. It is the core operating engine of every insurer, and once you understand it, the industry's financials stop looking mysterious.
Let's follow one commercial auto policy through the whole loop.
Underwriting is the process of deciding whether to accept a risk, and on what terms. The underwriter is the person (or increasingly, the model) making that call.
Our applicant is a regional courier with 40 vans. The underwriter pulls information to answer one question: how likely is this fleet to cause losses, and how big?
They look at:
The underwriter can accept, decline, or accept with conditions (for example, requiring the company to install dashcams). This is risk selection: the insurer is deliberately choosing which risks to take on.
Once the risk is accepted, it needs a price. Rating is the process of calculating the premium.
The starting point is the loss cost: the expected cost of claims for a given exposure, based on historical data. Actuaries build these estimates from large pools of similar risks.
The premium is roughly:
> Expected losses + expenses + profit margin = premium
For our courier, suppose the actuarial tables suggest an expected loss cost of $4,000 per van per year. Add loss adjustment expenses (the cost of investigating and settling claims), general operating costs, commissions, and a target profit. The final premium might land around $6,500 per van, so about $260,000 for the fleet.
Rating factors adjust that base. A clean five-year record earns a credit. Operating in a high-litigation state adds a surcharge. Rates and rating factors are also filed with and regulated by state insurance departments in the US, which review whether rates are adequate, not excessive, and not unfairly discriminatory. The National Association of Insurance Commissioners is a useful public starting point for how this regulation works.
The customer accepts the quote. The insurer issues the policy: this is binding, the moment coverage legally begins.
The policy runs for a defined policy period, typically 12 months. During that window, the insurer is on the hook for covered losses up to the policy limits.
At this point the insurer has a promise outstanding and $260,000 in premium coming in, but it has not actually "earned" that money yet. That distinction matters more than it sounds.
Here is a concept that trips up many newcomers: written premium versus earned premium.
Insurance earns premium ratably across the policy period. After one month of a 12-month policy, the insurer has earned about one twelfth, roughly $21,700. The rest sits as unearned premium, a liability, because if the customer cancels, that portion must be refunded.
Why does this matter? Because you cannot judge profitability by cash collected. You judge it by earned premium matched against the losses that occurred during that same period.
🎬 [VIDEO: "Earned vs Written Premium Explained" — youtube.com — a short walkthrough of how insurers recognize premium over a policy term]
Friday afternoon, van number 22 rear-ends a sedan. This triggers the claims side of the cycle.
A claim is a formal request for payment under the policy. The process:
1. First notice of loss (FNOL). The driver reports the accident. The clock starts.
2. Assignment. A claims adjuster takes the file. The adjuster investigates, determines coverage, and estimates the cost.
3. Reserving. This is critical. The insurer sets a loss reserve: an estimate of what this claim will ultimately cost, booked immediately even though payment may take months or years.
In a rear-end collision, the rear driver is usually at fault. Our insurer expects to pay for the sedan's repairs, plus possible bodily injury claims (whiplash, medical bills), plus its own insured van's damage if the policy includes collision coverage.
The adjuster might reserve $18,000: $6,000 for property damage and $12,000 for potential injury. That reserve counts as a loss the moment it is set, not when the check clears.
Reserving is estimation, and estimation can be wrong. If injury claims escalate (a lawyer gets involved, the injury is worse than it looked), the actual payout could hit $40,000. The insurer must then strengthen the reserve, and that increase hits earnings.
Chronic under-reserving is one of the most common ways insurers get into trouble. It makes today's results look good and pushes pain into future years.
Now we can measure. Two ratios do the heavy lifting.
The loss ratio measures claims cost against earned premium:
> Loss ratio = (incurred losses + loss adjustment expenses) / earned premium
"Incurred losses" means paid claims plus changes in reserves, so it captures the full expected cost, not just checks written.
If our commercial auto book earns $10 million in premium and incurs $6.5 million in losses, the loss ratio is 65 percent.
The loss ratio ignores the cost of running the business. The combined ratio fixes that by adding expenses:
> Combined ratio = loss ratio + expense ratio
The expense ratio covers commissions, underwriting costs, and overhead as a share of premium.
If losses are 65 percent and expenses are 30 percent, the combined ratio is 95 percent.
The rule of thumb:
A combined ratio of 95 percent means the insurer kept 5 cents of underwriting profit per premium dollar. At 108 percent, it lost 8 cents on the dollar from underwriting.
Here is the twist. Insurers collect premium upfront and pay claims later. In between, they invest that money (the float). Many insurers run combined ratios slightly above 100 percent and still turn a total profit because investment income covers the gap.
But a business that relies on investment income to survive is fragile. Disciplined insurers aim for an underwriting profit on its own.
Vérification des acquis
1. What is the fundamental purpose of underwriting in an insurer's operating cycle?
2. An underwriter agrees to cover a fleet only if the company installs dashcams in all its vehicles. This best illustrates which concept?
3. In the pricing formula 'expected losses + expenses + profit margin = premium', what does the loss cost represent?
4. Select ALL correct answers. Which of the following are factors an underwriter would legitimately examine to assess a commercial auto fleet's risk?
Sélectionnez toutes les réponses correctes.
5. Select ALL correct answers. Which statements correctly distinguish risk selection from rating?
Sélectionnez toutes les réponses correctes.
The cycle is not a straight line. It is a loop.
When our courier's claim closes at, say, $22,000, that outcome feeds back:
This feedback is why insurance is often called a business of continuous repricing. Every claim is data. Every renewal is a chance to correct.
Commercial auto specifically has been a challenging line in recent years, with the industry frequently reporting combined ratios above 100 percent, driven partly by rising litigation costs and larger jury awards (sometimes called "social inflation"). These are widely discussed industry trends rather than a precise single figure.