Accidents, Occurrences, and Claims

Why These Distinctions Matter

A single harmful incident can injure several people, while one person may make more than one demand for relief. Conversely, an insurer’s contract may group a sequence of related harmful events into one policy-defined occurrence. These are different ways of describing a loss: an accident describes what happened, an occurrence applies the contract’s grouping language, and a claim describes a demand for relief.

The counts therefore need not match. A candidate who counts injured people as accidents, or assumes that each claim must be a separate occurrence, can misclassify a scenario before even applying the policy wording. The reliable approach is to identify the events, read the stated definition of occurrence, and then count the demands separately.

Definition: An accident is an unexpected, unintended event that causes injury or damage, as understood in the context of the coverage. A policy’s use of the word may be subject to its own wording. An occurrence is the unit or group of events treated as one occurrence under the applicable policy definition. A claim is a demand or assertion seeking compensation, benefits, or other relief. The exact contract and scenario determine how these terms are applied.

These definitions are related but not interchangeable. One accident can produce several injuries and several claims. A policy may also define an occurrence to include a series of exposures to the same harmful condition, so that more than one harmful event is grouped as one occurrence. There is no universal rule that every accident is one occurrence or that every occurrence produces one claim.

Count Events, Occurrences, and Claims Separately

A practical classification begins by making a separate list for each count. First identify the accident or accidents described in the facts. Next, apply the specific occurrence wording; do not substitute an informal idea of what seems like “one incident.” Finally, count the demands that qualify as claims under the assumptions given.

Identify what happened.
Describe each unexpected event or harmful exposure. Do not count each injury as a separate accident unless the facts or wording support that classification.

Apply the occurrence definition.
Use the hypothetical policy wording provided. It may treat a single event as one occurrence or group related events under a stated rule.

Count demands.
Count claims according to the stated meaning of claim and the demands made. A claimant is a person or entity making a demand; the number of claimants does not automatically determine the number of claims.

Check the relationship among the counts.
Explain whether the facts make each claim attributable to one occurrence, and whether any occurrence has no claim or more than one claim.

The policy definition matters because “occurrence” is not merely a synonym for “accident.” For example, a hypothetical definition might state that an occurrence includes continuous or repeated exposure to substantially the same harmful conditions. Under such wording, several exposures may be grouped together. Different wording may treat separate releases or incidents separately. The answer must follow the definition actually supplied; a reader should not invent a grouping rule that the policy does not state.

Why the Counts Can Differ

Suppose a scenario has a set of occurrences and each claim is assigned to exactly one of them. Let \(m_j\) be the number of claims assigned to occurrence \(j\). Then the total number of claims is the sum of those occurrence-level claim counts:

N_C=\sum_{j=1}^{N_O}m_j

Here \(N_C\) is the number of claims and \(N_O\) is the number of occurrences. The formula is a counting identity: partition the claims by their assigned occurrence, count the claims in each group, and add the group sizes. It does not say that claim count must equal occurrence count. Also, the identity requires the stated assignment rule. If a claim can relate to multiple occurrences, or if a claim has not been assigned to any occurrence, the simple one-group-per-claim count needs to be revised to reflect those facts.

If every occurrence has at least one claim, then \(m_j\geq 1\) for each \(j\). In that case, \(N_C=N_O\) if and only if every \(m_j=1\): the sum of \(N_O\) positive integers equals \(N_O\) only when each integer is one. Without the assumption that every occurrence has at least one claim, equality of the totals does not imply exactly one claim per occurrence. For example, two occurrences could have \(m_1=0\) and \(m_2=2\), giving two claims in total.

Worked Example: Several Injuries From One Sudden Event

A hypothetical liability policy defines an occurrence as one discrete accident. A delivery van backs into a market stall, causing the stall to collapse. Three visitors are injured. Each visitor independently sends a written demand for compensation, and the scenario states that each demand is one claim. Count the accidents, occurrences, claimants, and claims.

Method. Classify the physical event first, then apply the stated occurrence definition, and count each stated demand. This avoids mistaking the number of injured people for the number of accidents.

The van’s impact and the resulting collapse describe one sudden harmful event. On the facts given, that is one accident. The policy defines an occurrence as one discrete accident, so the event is one occurrence.

There are three injured visitors, each of whom makes a demand. Thus there are three claimants and, because each written demand is stated to be one claim, three claims.

The count is therefore 1 accident, 1 occurrence, 3 claimants, and 3 claims. The check is that all three demands arise from the same stated event, but the occurrence definition groups the event, not the number of demands.

Common mistake avoided: Counting three injured visitors as three accidents. The people are separate claimants; their number does not multiply the one event into several accidents.

Worked Example: Repeated Exposure and Demands at Different Times

A hypothetical policy defines one occurrence to include continuous or repeated exposure to substantially the same harmful condition. A faulty ventilation system releases the same irritating vapor into a shared workshop on four consecutive mornings before it is repaired. Five workers are affected. Each worker submits one written demand, but the demands arrive over a period of two weeks. The scenario states that each demand is one claim. How many accidents, occurrences, claimants, and claims are there under the stated assumptions?

Method. Count the harmful events described, then apply the policy’s express grouping language. Count claims from the demands, not from the days on which those demands arrive.

The facts describe four separate morning releases. Treating each release as a separate event for the accident count gives four accidents. The wording then groups continuous or repeated exposure to substantially the same harmful condition into one occurrence. All four releases have the same stated source and condition, so they form one occurrence under this hypothetical definition.

Five distinct workers make one demand each. Therefore there are five claimants and five claims. The claims arrive over two weeks, but the scenario does not define a new occurrence each time a demand is submitted. Demand timing alone does not change the count of harmful events or the policy-defined grouping.

The result is 4 accidents, 1 occurrence, 5 claimants, and 5 claims. The check is that the grouping rule applies to the four exposures, while each separately stated demand remains a separate claim.

Common mistake avoided: Treating each arrival of a demand as a new occurrence. The occurrence count follows the policy definition and the harmful conditions, not the administrative dates when demands are received.

Worked Example: One Claimant With Demands From Separate Events

A hypothetical policy defines each separate accidental discharge from a piece of equipment as one occurrence. A greenhouse’s irrigation unit discharges hot water accidentally on Monday and again on Friday after a separate failure. On Monday, gardener R and gardener S are injured; on Friday, gardener R and gardener T are injured. Each injured person submits one demand for each incident in which that person was injured, and each such demand is one claim. Count the accidents, occurrences, distinct claimants, and claims.

Method. Treat each discharge under the given occurrence definition, then count demand-injury pairs because the facts specify one claim for each such pair. Count distinct people separately from the total number of claims.

There are two separate accidental discharges, one on Monday and one on Friday. Each discharge is a separate accident and, under the policy wording, a separate occurrence. Thus there are two accidents and two occurrences.

The distinct people who make demands are R, S, and T, so there are three distinct claimants. Monday produces demands from R and S, or 2 claims. Friday produces demands from R and T, or 2 claims. The total is \(2+2=4\) claims.

The result is 2 accidents, 2 occurrences, 3 distinct claimants, and 4 claims. The check is that R appears as one distinct claimant but makes two demands, one for each separate incident. The number of claims is therefore not the number of distinct claimants.

Common mistake avoided: Counting R only once in the claim total because R is one person. The scenario defines a separate claim for each incident-related demand, so R contributes two claims.

A Counting Trap: Equal Totals Do Not Prove One-to-One Pairing

The claim and occurrence totals can happen to be equal even when their relationship is not one claim per occurrence. The following small count illustrates why the assumptions behind a conclusion must be stated.

Worked Example: Equal Totals With an Unclaimed Occurrence

A hypothetical scenario identifies two occurrences. Each claim is assigned to one occurrence. The first occurrence has no demand, while two separate claimants make demands arising from the second occurrence. Each demand is one claim. Determine the number of claims and whether each occurrence has exactly one claim.

Method. Count claims within each occurrence, then add them. Do not infer a one-to-one relationship from equal overall totals.

For the first occurrence, \(m_1=0\). For the second occurrence, \(m_2=2\). Therefore:

$$ N_C=m_1+m_2=0+2=2 $$

There are \(N_O=2\) occurrences and \(N_C=2\) claims, so the totals are equal. However, the first occurrence has zero claims and the second has two; it is false that each occurrence has exactly one claim.

The result is 2 occurrences and 2 claims, with claim counts per occurrence of 0 and 2. The check is \(0+2=2\), which matches the total number of claims without producing a one-to-one pairing.

Common mistake shown and corrected: The incorrect conclusion is, “There are two claims and two occurrences, so each occurrence has one claim.” The corrected conclusion is that equal totals alone do not establish how claims are distributed. If every occurrence is known to have at least one claim, equality does imply one claim per occurrence; without that condition, it does not.

Exam-Style Classification Checks

When a problem gives a policy definition, quote or paraphrase the part that controls the grouping. When it does not give enough information to determine an occurrence count, state what is missing rather than assuming a standard definition. This discipline is aligned with the SOA Exam FAM (Fundamentals of Actuarial Mathematics) syllabus.

Do not equate injuries with accidents. Several people can be injured in one accident.
Do not equate claims with claimants. One person may make more than one claim, while a demand may also involve more than one claimant depending on the stated facts.
Do not assume every occurrence produces a claim. An event may cause no demand, or a demand may not yet have been made.
Do not let the demand date redefine the event. A later demand is not, by itself, a new accident or occurrence.
Do not impose a grouping rule that was not supplied. Occurrence classification depends on the applicable definition and the facts.

Key takeaway: Keep three questions separate: What harmful event or events occurred? How does the policy define and group occurrences? How many demands qualify as claims? State any assumptions about how claims are assigned to occurrences before comparing their totals.

Check Your Understanding

For each problem, distinguish the event count from the policy-defined occurrence count and the claim count. Use only the definitions and assumptions stated in the problem.

A hypothetical policy defines an occurrence as one discrete accidental event. A shelving unit falls once in a store, injuring four customers. Each customer submits one demand, and each demand is one claim. Give the number of accidents, occurrences, distinct claimants, and claims. (Answer: 1 accident, 1 occurrence, 4 distinct claimants, 4 claims.)
A hypothetical policy groups repeated exposure to the same harmful condition into one occurrence. A machine emits the same dust on three consecutive days before repair. Two workers each make one demand for injury from the exposure, with each demand constituting one claim. Give the counts, treating each daily emission as a separate event. (Answer: 3 accidents, 1 occurrence, 2 claimants, 2 claims.)
A hypothetical policy treats each separate accidental spill as a separate occurrence. A tank spills once on Tuesday and once on Saturday. Worker J is injured in both spills and submits one separate demand for each; worker K is injured only on Tuesday and submits one demand. Each demand is one claim. Give the counts. (Answer: 2 accidents, 2 occurrences, 2 distinct claimants, 3 claims.)
There are three policy-defined occurrences, with claim counts \(m_1=1\), \(m_2=0\), and \(m_3=2\). Each claim is assigned to exactly one occurrence. Find the total claim count and explain whether equal claim and occurrence totals establish one claim per occurrence. (Answer: 3 claims; no, the totals are 3 claims and 3 occurrences, but the occurrence-level counts are 1, 0, and 2.)
A hypothetical policy groups continuous exposure to one source into a single occurrence. A leaking pipe exposes three tenants to the same condition over several days. Each tenant sends two separate demands, and the scenario states that each demand is a separate claim. No further event or grouping rule is stated. Give the number of claims and, treating each tenant’s demands as separate claims, the number of occurrences. (Answer: 6 claims and 1 occurrence; the total is 3 tenants multiplied by 2 demands each, while the stated repeated-exposure wording groups the exposure into one occurrence.)

What You'll Learn