Basic Track II 2001 CLRS September 2001 New

Basic Track II 2001 CLRS September 2001 New Orleans, Louisiana CLRS Basic Track II 1

Introduction l Topics Covered » » Comparison of Results from Paid and Incurred LDMs Reasonableness Checks – – – » Current Year Sensitivity Analysis – – – » Ultimate Loss Ratios Frequency/Severity Pure Premium Rate Level Adequacy Claim Severity Claim Closure Rate Adequacy of Case Reserves Frequency/Severity Projection Importance of Tail Factors CLRS Basic Track II 2

Comparison of LDM Projections CLRS Basic Track II 3

Comparison of Loss Development Methods Underlying Assumptions PLDM: No changes in the payment pattern Pro PLDM: “Hard” data; no estimates involved ILDM: No changes in case reserve adequacy ILDM: Uses all the available information PLDM: May generate large, volatile loss development factors & take longer to develop to ultimate Con CLRS Basic Track II ILDM: Uses case reserves, which are estimates, to develop estimates of ultimate losses 4

Key Assumptions & Potential Problems Assumptions Sample Problems Claims settlement patterns unchanging Increasing delays in claim closing rates Case reserving practices & philosophies unchanging Conscious effort to improve case reserve adequacy; Introduction of new case reserving procedures No claim processing changes Change in data processing; Revised claim payment recording procedures Policy limits have no impact on loss development Increasing frequency of full policy limits claims; Changing policy limits CLRS Basic Track II 5

Key Assumptions & Potential Problems Assumptions Sample Problems Loss development unaffected by changing loss cost trends Surges in inflation; Increased litigation; Diminished policy defenses No change in mix of business Changes in reinsurance coverages; Increased long-tail exposures; Introduction of new or revised coverages No cyclical loss development Claims settlement or reserving impacted by business underwriting cycles No data anomalies Catastrophic or unusual losses reflected in loss experience; Unusual claim settlement/reporting delays CLRS Basic Track II 6

Comparison of Estimated Reserves CLRS Basic Track II 7

Formulas to Derive IBNR Reserves l Once an estimate of ultimate loss has been obtained, the arithmetic of IBNR is simple. Ultimate Losses Unpaid Losses Minus Paid Losses Reported Losses Case Reserves Minus Case Reserves CLRS Basic Track II 8

Other Reserving Methods Tested l Discussed in subsequent CLRS sessions » » l Expected Loss Technique Bornhuetter-Ferguson Method Severity/Frequency Method Many, many others Note that development method may also be applied to claim counts. CLRS Basic Track II 9

Development Method on Claim Counts CLRS Basic Track II 10

Reported Counts DM: LDF Selected CLRS Basic Track II 11

Reported Counts DM: Projection CLRS Basic Track II 12

Reasonableness l Ultimate losses should be measured for reasonableness against relevant indicators: » premium – » exposures or number of policies – » frequency, pure premium claim counts – l loss ratios severity Assumptions & methods should be documented and subjected to sensitivity analysis. CLRS Basic Track II 13

Reasonableness Checks: Ultimate Loss Ratios CLRS Basic Track II 14

Reasonableness Checks: Ultimate Loss Ratios CLRS Basic Track II 15

Reasonableness Checks: Frequency & Severity CLRS Basic Track II 16

Reasonableness Checks: Frequency & Severity CLRS Basic Track II 17

Reasonableness Checks: Pure Premium CLRS Basic Track II 18

Reasonableness Checks: Pure Premium CLRS Basic Track II 19

Sensitivity Analysis: Current Year Analysis l Improvements in results may stem from: » » » l Higher rates Lower claim frequency Lower claim severity Better results would appear to be present if: » » » CLRS Basic Track II Claims were being processed or paid more slowly Case reserves were less adequate Mix of business is different 20

Sensitivity Analysis: Ratios l Review historical relationships » Losses – Reported losses to paid » Claim counts – Settlement rate – Ratio of claims closed with no payment to total closed claims » Losses and Claim Counts – Severities or average values CLRS Basic Track II 21

Sensitivity Analysis: Ratios - Examples CLRS Basic Track II 22

Sensitivity Analysis: Rate Level Adequacy Increases in average premium are primarily due to: Changes in the mix of business. Rate increases. If the changes in average premium in the latest two years are due to rate increases, then that would explain much of the improvement in loss ratios. If the changes are due to shifts in the mix of business, then the improvement in the loss ratios may or may not be real. Further investigation would be needed to understand what the shift was and whether the different business types have varying loss development characteristics. CLRS Basic Track II 23

Sensitivity Analysis: Claim Severity There is no consistent pattern in severity, except that it has generally increased over the years. This is typical, as we expect severity to increase due to inflation. The very small increase in severity that is forecast for the current year is unusual. In the same year, claim frequency has increased. Perhaps there is an increase in the number of small dollar claims? This would be a good question to ask the Claim Department. CLRS Basic Track II 24

Sensitivity Analysis: Claim Closure Rate In the past few years, claims have been closing more rapidly. This would imply that claims are being paid more rapidly and that the paid loss development factor is probably too high. One of the major assumptions of the PLDM (consistent payment patterns) has been violated. CLRS Basic Track II 25

Sensitivity Analysis: Case Reserve Adequacy CLRS Basic Track II 26

Sensitivity Analysis: Case Reserve Adequacy In general, we expect increasing numbers: 1. Across the rows because smaller claims settle more quickly; and 2. Down the columns due to inflation. It is important to understand the company’s case reserving philosophy and procedures to be able to interpret trends in the data. Many changes in case reserve procedures can be monitored by talking to the Claims Department. Changes in case reserve adequacy affect incurred loss development patterns. For example, if case reserves were less adequate in the current accident year, greater future development would be expected for those accidents than was typical in the past. Use of historical loss development factors in this situation would underestimate future development and lead to inadequate overall reserve estimates. CLRS Basic Track II 27

Sensitivity Analysis: Case Reserve Adequacy The fit of the average case reserves @ 12 months implies an annualized trend rate of 19%! This rate is substantially higher than industry trend rates for private passenger automobile liability, which are in the range of 8% to 10%. CLRS Basic Track II 28

Frequency/Severity Projection Method A line or another curve can be fitted through actual values for accident years 1995 through 1999. The fitted points for the current year can be used as estimates for the ultimate frequency and severity. R-squared is a measure of how well a fitted curve matches the data. The value can range from 0 to 1. 00, where 1. 00 indicates a perfect fit. CLRS Basic Track II 29

Frequency/Severity Projection Method CLRS Basic Track II 30

Selection of Tail Factors l How much difference does the tail factor selection make? CLRS Basic Track II 31

Selection of Tail Factors Ultimate losses increase by 2% or $1. 8 million. l Loss reserves also increase by $1. 8 million; however, the 2% increase in the tail factor represents a 7% increase in overall reserve levels! l IBNR reserves are increased by an even higher percentage as a result of an increase of 2% in the tail factor l CLRS Basic Track II 32