B.5 Development

Question 1

Reason to develop losses in ratemaking

Answer 1

So that rates in the future policy period will be adequate to cover the ultimate costs coming from policies written using those rates.

Question 2

Four steps in estimating ultimates

Answer 2

1. Exploratory analysis of the data: Identify key
   characteristics and anomalies, and balance to verified
   sources.
2. Apply appropriate techniques to estimate ultimates
3. Evaluate the conflicting results of the different
   techniques: Reconcile and explain the different outcomes.
4. Monitor projections of actual versus expected
   development: Update or correct projections with new
   information.

Question 3

Areas in estimating ultimates for which actuarial

judgment is needed

Answer 3

 -Determining the optimal combinations of claims to use (i.e., the granularity)
 -Assessing the effect of an insurer’s operational changes on the data
 -Adjusting data for known and quantifiable events
 -Evaluating the strengths and weaknesses of different
estimation methods
 -Making the final selection of the estimated ultimate

Question 4

What changes a closed claim counts as a % of

reported claim counts triangle can show

Answer 4

Whether there has been a speedup or slowdown in closing claims over time.

Question 5

What changes an average paid on closed claims

triangle can show

Answer 5

This can reflect both severity trends and speedups or
slowdowns in the closing of small claims relative to large
claims.

Question 6

What changes an average case reserve triangle can

show

Answer 6

This will reflect severity trends, speedups or slowdowns
in the closing of small claims relative to large claims, and
changes in case reserve adequacy.

Question 7

What changes a paid loss to reported loss triangle can

show

Answer 7

This will reflect speedups or slowdowns in the closing of

claims and changes in case reserve adequacy

Question 8

Key characteristics to use in deciding triangle

granularity

Answer 8

 -Similarity in coverage (i.e., similar laws, policy terms)
 -Volume of claim counts (i.e., credibility)
 -Reliability of case reserves
 -Report lag
 -Settlement lag
 -Likelihood of claims reopening
 -Claim severity

Question 9

Reasons that losses can develop downwards

Answer 9

Reasons for negative loss development include case reserve decreases, deductible recoveries, subrogation, and salvage.

Question 10

Possible selections for age-to-age factors

Answer 10

- Straight average: Straight average of the LDFs.
 -Weighted average: Weight the LDFs by the earlier age
losses. It is faster though to sum the relevant values from
the 2 columns of the original loss triangle and then divide.
 -Geometric average: Multiply N LDFs then take the Nth
root.
 -Medial average: Throw out the highest and lowest LDF and straight average the rest.
 -Judgment
 -Latest year
 -Industry benchmark factors

Question 11

Characteristics actuaries look at when selecting

age-to-age factors

Answer 11

 -Smooth progression of age-to-age factors across columns: Ideally, these factors should steadily decrease with age.
 -Stability of age-to-age factors for the same column: The
more stability, the more consistent the development pattern. Usually the first ages have less stability since the data is immature.
 -Credibility of experience: If you have limited or unstable
data, you can use industry benchmark factors.
 -Changes in patterns: Systematic patterns can identify
changes in operations or environment.
 -Applicability of historical experience: Do you expect the
future development to be like the past development?

Question 12

Impact on future estimates based on the reported
chain ladder method with a recent decrease in case
reserve adequacy

Answer 12

If case reserve adequacy has recently decreased, historical age-to-age factors would be based on higher case reserve adequacy. Using those age-to-age factors to project recent data (that has lower case adequacy) would cause you to underproject ultimate losses.

Question 13

Impact on future estimates based on the paid chain
ladder method with a recent increase in claim
settlement rates

Answer 13

If claim settlement rates have recently increased, historical age-to-age factors would be based on slower settlement rates. Using those age-to-age factors to project recent data would cause you to overproject ultimate losses.

Question 14

Common methods for selecting a tail factor

Answer 14

 -Doing a special study that contains more years of data.
 -Using an industry benchmark tail factor.
 -Fitting a curve (e.g., exponential decay) to the LDFs and
extrapolating the tail factor.
 -Use reported-to-paid ratios at the latest paid development period.
 -Judgment (e.g., arbitrarily picking a value like 1.05).

Question 15

Formulas for ultimate, IBNR, and unpaid claims

estimates

Answer 15

Ultimate = Latest value x age-to-ultimate factor
IBNR = Ultimate - Latest reported loss
Unpaid = Ultimate - latest paid loss

Question 16

Impact of a large loss in a cumulative triangle on

age-to-age factors

Answer 16

Spike when the loss appears, followed by a drop.

Question 17

Do triangles need to be adjusted for one-time

changes?

Answer 17

Usually not, as changes can be isolated to appear on separate rows if needed.

However, triangles do need to be adjusted for law changes that impact open claims or in-force policies. This can be done by on-leveling, or considering the impact of the changes when selecting LDFs.