Experience Rating for New or Short-History Risks

Why New and Short-History Risks Need a Defined Treatment

An experience rating plan uses a risk’s own past experience in determining a prospective premium or premium component. That approach requires relevant, sufficiently complete data. A new risk may have no history at all; another may have a few months of exposure but not enough completed experience to meet the plan’s rules. In either case, the plan needs a clear treatment that does not mistake missing information for favorable experience.

The design question is not simply whether a risk is “large enough.” A plan can specify conditions for entering the plan and separate conditions for using a particular period of its experience. It also needs a stated basis for rating a risk while its own data are not yet eligible. These decisions should be made in advance and applied consistently to risks meeting the same conditions. This topic is aligned with the CAS Exam 8 (Advanced Ratemaking) syllabus.

Definition: A new or short-history risk is a risk that lacks enough relevant, eligible experience to meet the experience plan’s requirements for using its own past data. Entry eligibility determines whether the risk may participate in the plan. Experience-data eligibility determines whether specified historical exposure and loss data may be used in a prospective rating.

Entry Rules and Data-Eligibility Rules Are Different

An entry rule is about the risk’s participation in the plan. For example, a plan might apply only to specified coverage types or to risks meeting a stated premium threshold. A risk that fails an entry rule is rated outside that plan, subject to the applicable rating approach. Meeting an entry rule does not necessarily make the risk’s experience usable.

A data-eligibility rule specifies what history can be included. It may require a minimum number of completed experience periods, a minimum amount of earned exposure, a particular coverage basis, and losses valued as of a specified date. A risk can therefore enter the plan but initially receive no adjustment based on its own experience. Conversely, data should not be made eligible merely because they are available: the data must satisfy the plan’s requirements.

The exact criteria depend on the plan. For example, the plan could require two complete annual periods and at least 100 exposure units in total. Another plan could use different criteria. The important design feature is that the criteria are stated clearly enough to determine whether the risk qualifies without discretionary judgment at renewal.

Key distinction: Entry asks, “Is this risk included in the experience rating plan?” Data eligibility asks, “Does this particular history meet the plan’s rules for use?” A risk may satisfy the first test and fail the second.

Specify the Temporary Treatment

If a risk enters the plan before its own experience is eligible, the plan needs a temporary rating basis. The temporary method should be defined in advance and should use information permitted under the applicable rating rules. Common approaches include:

Manual or class rating: use the applicable prospective rate and exposure basis without an adjustment from the risk’s own ineligible history.
A defined pool or common benchmark: use a specified pooled experience or benchmark when the plan is designed to rate a group of risks together.
A staged transition: use the stated temporary basis until the risk meets the experience requirements, then begin using eligible data in a specified prospective period.
A permitted external history source: if the plan allows it, use documented prior experience that meets the plan’s requirements for relevance, completeness, coverage, and valuation.

These are alternatives, not interchangeable labels for the same method. A manual rate does not become the risk’s own experience. A pool represents a defined group, not the individual risk’s observed results. Prior records should not be imported simply because the current risk has little history; the plan must establish that the records represent comparable coverage and exposure and satisfy its data rules.

The plan should also state the transition point. Ordinarily, newly eligible experience is used in a prospective rating after the required data are available, checked, and valued on the prescribed basis. A period that has not finished, or losses that have not reached the required valuation date, should not be treated as though the plan’s requirements have been met.

How to Test a Short History

A reliable decision follows the plan’s tests in order. First determine whether the risk is within the plan’s scope. Then check whether the experience periods are complete, whether the exposure minimum is met, and whether the losses meet the plan’s valuation and coverage rules. If any required condition fails, apply the stated temporary method rather than using a partial result as if it were eligible.

Check plan entry.
Confirm that the risk and coverage meet the plan’s stated participation rules.

Check the history requirements.
Count only the periods and exposure that meet the plan’s definition of eligible experience.

Check data quality and valuation.
Confirm that losses are on the required coverage basis and have been valued by the stated date.

Assign the correct treatment.
If all conditions are met, the eligible history may be used for the applicable prospective period. Otherwise, use the designated temporary basis and identify when the test will be made again.

Why Minimum History Requirements Matter

A minimum-history requirement prevents a small or incomplete set of observations from being treated as a stable description of a risk’s experience. A simple calculation illustrates the issue. Suppose, only for illustration, that each exposure unit has an independent loss observation with variance $\sigma^2$, and the observations have the same expected loss. For $n$ units, the average loss is $\bar X$. Independence gives:

\operatorname{Var}(\bar X) =\operatorname{Var}\left(\frac{1}{n}\sum_{j=1}^{n}X_j\right) =\frac{1}{n^2}\sum_{j=1}^{n}\operatorname{Var}(X_j) =\frac{\sigma^2}{n}.

Thus, under those simplifying assumptions, increasing the number of observations reduces the variance of the average. This does not prescribe a universal minimum, and actual insurance losses may differ in frequency, severity, dependence, and exposure. It does explain why a plan may set a minimum history requirement and why a small amount of favorable experience is not, by itself, a sound reason to use that history.

Worked Example: Enters the Plan but Fails the Exposure Test

A hypothetical commercial property plan includes every risk with the covered policy form, so a small business is within the plan. The plan permits use of a risk’s own experience only after two complete annual periods and at least 100 earned location-years across those periods. The risk has two complete periods with 45 and 38 earned location-years. Its manual rate is $42 per location-year, and its prospective exposure is 80 location-years. Decide whether its own experience may be used and calculate the temporary manual premium.

Method. Test entry and experience eligibility separately, then calculate the premium on the stated temporary manual basis.

Entry test. The covered policy form is included, so the risk enters the plan.

History tests. The two-period requirement is met. Total eligible exposure is $45+38=83$ location-years. Since $83<100$, the exposure requirement is not met. The risk’s own experience is therefore not eligible for use yet.

Temporary premium. Apply the manual rate to the prospective exposure:

80 \times \$42 = \$3{,}360.00.

The temporary manual premium is $3,360.00. Check: $80 \times 40=\$3{,}200$ and $80 \times 2=\$160$, giving $\$3{,}360$. The common mistake is to treat two completed periods as sufficient by themselves; the plan requires both periods and the minimum exposure.

Worked Example: Enough Exposure but No Second Complete Period

A hypothetical liability plan allows an account’s own experience after two complete annual periods and at least 100 earned units in total. The first period has 70 earned units. At the proposed renewal, the second period is still in progress and has accumulated 50 units over ten months. The manual rate is $35 per unit, and the prospective exposure is 120 units. Determine the status and the temporary premium.

Method. Apply the completion condition before totaling exposure for eligibility. Then use the prescribed manual basis if the condition fails.

The observed exposure so far totals $70+50=120$ units, which is at least 100. However, only the first annual period is complete; ten months of the second period are not a second complete annual period. The two-period condition therefore fails, so the account’s own history is not yet eligible.

The manual premium is:

120 \times \$35 = \$4{,}200.00.

The temporary premium is $4,200.00. Check: $12 \times 35=420$, and multiplying by ten gives $4{,}200$. The common mistake is to see that the total exposure exceeds 100 and overlook the separate requirement for two complete periods.

Worked Example: Identifying the First Eligible Prospective Period

A hypothetical fleet plan requires two complete annual periods, at least 100 earned vehicle-years in total, and losses valued at least six months after the end of the second period. A fleet has 58 vehicle-years in its first completed period and 52 in its second. The second period ends December 31, and the losses are valued June 30. The next policy period begins July 1. Determine whether the history meets the stated requirements and when it can first be used.

Method. Check the number of complete periods, aggregate exposure, and valuation date separately.

There are two complete annual periods. Their total exposure is $58+52=110$ vehicle-years, and $110\geq100$. Six months after December 31 is June 30, so the losses are valued on the required date. All three stated conditions are met.

The history may therefore be used in the prospective rating beginning July 1, subject to the plan’s other applicable procedures. The eligible experience is not retroactively applied to the completed periods. The common mistake is to count a valuation made before June 30 as satisfying the six-month condition or to use qualifying history before it is available for the prospective rating.

Common Mistakes in New-Risk Treatment

Combining entry and data eligibility. A risk’s participation in the plan does not automatically make its experience usable.
Ignoring one of several conditions. If a plan requires complete periods and a minimum exposure, both tests must pass.
Using partial periods as complete periods. Accumulated exposure may count toward a specified exposure test, but it does not turn an unfinished period into a completed one.
Treating missing history as favorable experience. No reported losses in a very short period do not establish that the risk has no expected losses.
Failing to name the transition date. State the first prospective period for which eligible experience can be used, rather than implying that the data alter a period already underway or completed.

A sound plan makes the result reproducible: the same entry rule, data tests, temporary basis, and transition timing should lead to the same treatment for risks with the same relevant circumstances. The calculations above illustrate how to apply stated rules; they do not prescribe universal thresholds or a single temporary method for all plans.

Check Your Understanding

For each hypothetical plan, apply the stated rules exactly and identify the correct treatment.

A plan requires one completed annual period and 60 earned units before using individual experience. A risk has one complete period with 54 units and an applicable manual rate of $28 per unit for 75 prospective units. Is its own experience eligible, and what is the temporary manual premium? (Answer: Not eligible; $75 \times \$28=\$2,100.00$.)
A plan requires two completed annual periods and 90 total earned units. A risk has two completed periods with 44 and 47 units. Does it meet the stated experience requirements? (Answer: Yes; $44+47=91$ units and two complete periods meet both requirements.)
A plan requires two completed periods, 120 total earned units, and losses valued six months after the final period. A risk has 61 and 63 units in two complete periods ending September 30; losses are valued March 31. What is the result? (Answer: Eligible for the next prospective rating, since $61+63=124$, both periods are complete, and March 31 is six months after September 30.)
A plan admits all covered risks but requires three complete annual periods before an individual history can be used. A new risk has 160 total units over two complete years and a third period that is eight months complete. Can its history be used now, even though total exposure is substantial? (Answer: No; the third period is incomplete, so only two complete periods are available.)
A plan requires two complete periods and at least 80 earned units. A risk has 41 units in each of two completed periods, but its losses have not yet reached the required valuation date. What should the plan do at this renewal? (Answer: Do not use the history yet; apply the plan’s stated temporary rating basis until the valuation requirement is met and the experience is available for a prospective period.)

What You'll Learn