Mercury Insurance 2008 Annual Report Download - page 42

32
• The average severity method analyzes historical loss payments and/or incurred losses divided by closed claims and/or
total claims to calculate an estimated average cost per claim. From this, the expected ultimate average cost per claim
can be estimated. The average severity method coupled with the claim count development method provide meaningful
information regarding inflation and frequency trends that the Company believes is useful in establishing reserves.
• The paid loss development method analyzes historical payment patterns to estimate the amount of losses yet to be
paid. The Company primarily uses this method for loss adjustment expenses because specific case reserves are
generally not established for loss adjustment expenses.
In states with little operating history where there are insufficient claims data to prepare a reserve analysis relying solely
on Company historical data, the Company generally projects ultimate losses using industry average loss data or expected loss
ratios. As the Company develops an operating history in these states, the Company will rely increasingly on the incurred loss
development and average severity and claim count development methods. The Company analyzes catastrophe losses separately
from non-catastrophe losses. For catastrophe losses, the Company determines claim counts based on claims reported and
development expectations from previous catastrophes and applies an average expected loss per claim based on reserves
established by adjusters and average losses on previous similar catastrophes.
There are many factors that can cause variability between the ultimate expected loss and the actual developed
loss. Because the actual loss for a particular accident period is unknown until all claims have settled for that period, the Company
must estimate what it expects that loss to be. While there are certainly other factors, the Company believes that the following
items tend to create the most variability between expected losses and actual losses:
• Variability in inflation expectations – particularly on coverages that take longer to settle such as the California
automobile bodily injury coverage.
• Variability in the number of claims reported subsequent to a period-end relating to that period – particularly on
coverages that take longer to settle such as the California automobile bodily injury coverage.
• Variability between Company loss experience and industry averages for those lines of business that the Company is
relying on industry averages to establish reserves.
• Unexpected large individual losses or groups of losses arising from older accident periods typically caused by an
event that is not reflected in the historical company data used to establish reserves.
These items are discussed in detail:
1. Inflation Variability – California automobile lines of business
For the Company’s California automobile lines of business, the bodily injury (BI) reserves comprise approximately 65%
of the total reserve; material damage, including collision, comprehensive, and property damage (MD) reserves make up
approximately 10% of the total reserve; and loss adjustment expense reserves make up approximately 25% of the total reserve.
The BI reserves account for such a large portion of the total because BI claims tend to close much slower than MD claims. The
majority of the loss adjustment expense reserves consist of estimated costs to defend BI claims, so those reserves also tend to
close more slowly than MD claims. Loss development on MD reserves is generally insignificant because MD claims are closed
quickly.
BI loss reserves are generally the most difficult to estimate because they take longer to close than most of the Company’s
other coverages. The Company’s BI policy covers injuries sustained by any person other than the insured, except in the case of
uninsured and underinsured motorist BI coverage, which covers damages to the insured for BI caused by uninsured or
underinsured motorists. BI payments are primarily for medical costs and general damages.
The following table represents the typical closure patterns of BI claims in the California automobile insurance coverage:
Claims closed Dollars Paid
BI claims closed in the accident year reported 35% to 40% 15%
BI claims closed one year after the accident year reported 75% to 80% 60%
BI claims closed two years after the accident year reported 93% to 97% 90%
BI claims closed three years after the accident year reported 97% to 99% 98%
% of Total