SecProbe.io

CORRESP
1
filename1.htm

corresp

February 1, 2010

VIA
FAX AND EDGAR

Jim B. Rosenberg

Senior Assistant Chief Accountant

Division of Corporate Finance

United States Securities and Exchange Commission

100 F Street N.E.

Washington D.C. 20549

    Re:

    Aspen Insurance Holdings Limited

Form 10-K for the Fiscal Year ended December 31, 2008

Form 10-Q for the quarterly period ended June 30, 2009

(File No. 001-31909)

Dear Mr.
Rosenberg:

Further to your letter of December 23, 2009, we have set out below our responses to the questions
raised, proposed disclosures and where necessary provided additional analysis. This letter follows
our prior letter from Michael Cain, Aspen’s Group General Counsel, dated December 24, 2009 where we
requested an extension to the timetable for our response due to the holidays.

For your convenience, we set forth in this letter (the “Response Letter”) each comment from your
letter in bold typeface and include the Company’s response below each such comment.

Item 7. Management’s Discussion And Analysis Of Financial Condition And Results Of Operations

Critical Accounting Policies

Reserving approach, page 98, IBNR claims, page 99, Actuarial range of gross reserves, page 100 and Uncertainties, page 100

    1.

    We believe your disclosure in the Critical Accounting Estimates section of MD&A regarding the
estimation of the reserve for loss and loss adjustment expenses could be improved to better
explain the judgments and uncertainties surrounding this estimate and the potential impact on
your financial statements. We believe in order to meet the principal objectives of MD&A this
disclosure should enable the investor to understand 1) management’s method for establishing
the estimate; 2) whether and if so to what extent and why you have adjusted your assumptions
used to determine the estimate from the assumptions used in the immediately preceding period
and 3) the potential variability in the most recent estimate and the impact this variability
may have on reported results, financial condition and liquidity. Please keep these points in
mind in providing us your responses to the comments listed below. Please revise your
disclosure to include the following information and also consider providing any additional
information to achieve this objective.

    a.

    Please describe the methods you used to determine your reserve for loss and
loss adjustment expense. In your disclosure, include a description of the method you
use to calculate the IBNR reserve for each material line of business. For example, we

    understand that some companies may calculate this reserve by estimating the ultimate
unpaid liability first and then reducing that amount by cumulative paid claims and by
case reserves, but there may be other methods as well. In addition, to the extent your
interim procedures for determining the reserve for loss and loss adjustment expense
differs from the annual procedures, describe the differences.

    b.

    As it appears you utilize actuaries in estimating the liability for unpaid
claims and claims adjustment and may make adjustments to the actuarial estimates,
provide the following information:

1) Describe the method used by management to determine the adjustment and the extent to
which you rely on objective versus subjective determinations. Such adjustments may
include, but not be limited to, an incremental provision, a reduction in the liability,
or a reversal of a previously recorded adjustment.

2) When such adjustments or reversals are made, include MD&A disclosure that identifies
the amount of the adjustment or reversal, the method you used to determine it, and the
specific underlying reasons that explain why you believe the adjustment or reversal is
necessary.

a. In response to the Staff’s specific comments above we intend to supplement our existing
disclosures in our forthcoming Annual Report on Form 10-K for the year ended December 31, 2009 in
the appropriate place within the “Critical Accounting Policies – Reserves” section of the MD&A, as
follows:

“The need for IBNR reserves arises from time lags between when a loss occurs and when it is
actually reported and settled. By definition, we do not have specific information on IBNR claims
so they need to be estimated by actuarial methodologies. IBNR reserves are therefore generally
calculated at an aggregate level and cannot generally be identified as reserves for a particular
loss or contract. We calculate IBNR reserves by line of business and by accident year within
that line. Where appropriate, analyses may be conducted on sub-sets of a line of business. IBNR
reserves are calculated by projecting our ultimate losses on each line of business and subtracting
paid losses and case reserves.”

We also propose to add the following narrative to our forthcoming Form 10-K for the year ended
December 31, 2009 in response to the Staff’s specific comments:

“The main projection methodologies that are used by our actuaries are:

    •

    Initial expected loss ratio method (“IELR”): This method calculates an estimate of ultimate
losses by applying an estimated loss ratio to an estimate of ultimate earned premium for each
accident year. The estimated loss ratio is based on pricing information and industry data and
is independent of the current claims experience to date.

2

    •

    Bornhuetter-Ferguson method (“BF”): The BF method uses as a starting point an assumed IELR and
blends in the loss ratio, which is implied by the claims experience to date using benchmark
loss development patterns on paid claims data (“Paid BF”) or reported claims data (“Reported
BF”). Although the method tends to provide less volatile indications at early stages of
development and reflects changes in the external environment, it can be slow to react to
emerging loss development and can, if the IELR proves to be inaccurate, produce loss estimates
which take longer to converge with the final settlement value of loss.

    •

    Loss development (“Chain Ladder”): This method uses actual loss data and the historical
development profiles on older accident years to project more recent, less developed years to
their ultimate position.

    •

    Exposure-based method: This method is used for specific large typically catastrophic events
such as a major hurricane. All exposure is identified and we work with known market
information and information from our cedents to determine a percentage of the exposure to be
taken as the ultimate loss.

While our actuaries calculate the IELR, BF and Chain Ladder methods for each line of business and
each accident year, they provide a range of ultimates within which management’s best estimate is
most likely to fall. This range will usually reflect a blend of the various methodologies. In
general terms, the IELR method is most appropriate for lines of business and/or accident years
where the actual paid or reported loss experience is not yet mature enough to modify our initial
expectations of the ultimate loss ratios. Typical examples would be recent accident years for
lines of business in the casualty reinsurance segment. The BF method is generally appropriate
where there are few reported claims and a relatively less stable pattern of reported losses.
Typical examples would be our treaty risk excess line of business in our property reinsurance
segment and marine hull line of business in our international insurance segment. The Chain Ladder
method is appropriate when there are relatively stable patterns of loss emergence and a relatively
large number of reported claims. Typical examples are the U.K. commercial property and U.K.
commercial liability lines of business in the international insurance segment.

There are no differences between our year-end and our quarterly internal reserving procedures.”

b. In response to the Staff’s comment above, we propose to add the following narrative to our Form
10-K for the year ended December 31, 2009:

“Management has not made any adjustments to actuarial estimates. Actuarial projection
methodologies involve significant subjective judgments reflecting many factors such as changes in
legislative conditions, changes in judicial interpretation of legal liability policy coverages and
inflation. Our actuaries collaborate with underwriting, claims, legal and finance in identifying
factors which are incorporated in their range of ultimates in which management’s best estimate is
most likely to fall. While management can select its best estimate outside the range provided by
the actuaries, management has not made any adjustment to the estimates
provided by the actuaries.
Management, through its Reserve Committee, then reviews the range of estimates and other evidence
and selects its best estimate of reserves for each line of business and accident year. This
provides the basis for the recommendation made by management to the Audit Committee and Board of
Directors regarding the reserve amounts to be recorded in the Company’s financial statements. The
Reserve Committee is a management committee consisting of the Chief Risk Officer, the Group Chief
Actuary, the Chief Financial Officer and senior members of our Underwriting and Claims staff.”

    2.

    In order to show investors the potential variability in the most recent estimate of your loss
reserve, quantify and present preferably in a tabular format the impact that reasonably likely
changes in the key assumptions identified may have on reported results, financial position and

3

    liquidity by each line of business. Explain why management believes the scenarios quantified
are reasonably likely.

In our current disclosure we provide the lower and upper end of the actuarial range of gross
reserves along with management’s best estimate by segment.

As disclosed in our Annual Report on Form 10-K for the year ended December 31, 2008, the actuarial
ranges are not intended to include the minimum or maximum amount that the claims may ultimately
settle at, but are designed to provide management with ranges from which it is reasonable to select
a single best estimate for inclusion in our financial statements.

In respect of the sensitivity analysis we have provided below the range of distribution from our
stochastic modelling which provides a probabilistic approach to the analysis of reserves. We
believe that our proposed disclosure more accurately addresses the Staff’s comments as it provides
a distribution against which selected reserves can be assessed for which we show the probability of
various outcomes relative to the actuarial mean estimate.

The amounts shown for the lower and upper ends of the actuarial range are different from the sums
of the corresponding amounts for the four segments. The difference, which we show in the tables as
‘potential variation’, takes into account the fact that at the higher end of the actuarial range we
do not expect all segments to deteriorate at the same time (hence the variation credit) and,
conversely, at the lower end of the actuarial range not all segments will improve simultaneously
(hence the debit).

We propose supplementing our current disclosure taking into account the above in response to the
Staff’s comment with the following discussion of key reserving assumptions and related sensitivity
analysis to be included in our forthcoming Form 10-K for the year ended December 31, 2009:

“Loss Reserving Sensitivity Analysis: The most significant key assumptions identified in the
reserving process are that (1) the historic loss development and trend experience is assumed to be
indicative of future loss development and trends, (2) the information developed from internal and
independent external sources can be used to develop meaningful estimates of the initial expected
ultimate loss ratios, and (3) no significant losses or types of losses will emerge that are not
represented in either the initial expected loss ratios or the historical development patterns.

The selected best estimate of reserves is typically in excess of the mean of the actuarial reserve
estimates. The Company believes that there is potentially significant risk in estimating loss
reserves for long-tail lines of business and for immature accident years that may not be adequately
captured through traditional actuarial projection methodologies. As discussed above, these
methodologies usually rely heavily on projections of prior year trends into the future. In
selecting its best estimate of future liabilities, the Company considers both the results of
actuarial point estimates of loss reserves as well as the potential variability of these estimates
as captured by a reasonable range of actuarial reserve estimates. In determining the appropriate
best estimate, the Company reviews (i) the position of overall reserves within the actuarial
reserve range, (ii) the result of bottom up analysis by accident year reflecting the impact of
parameter uncertainty in actuarial calculations, and (iii) specific qualitative information on
events that may have an effect on future claims but which may not have been adequately reflected in
actuarial mid-point estimates, such as the potential for outstanding litigation or claims practices
of cedants to have an adverse impact.

In order to show the potential variability in the Company’s estimate of loss reserves, the internal
actuaries use stochastic modelling techniques around their mean estimate. We believe that
stochastic modelling provides a distribution against which selected reserves can be assessed for
which we show the probability of various

4

outcomes relative to the actuarial mean estimate. Stochastic modelling provides a range of
potential outcomes as reserve movements will be caused by any number of factors, and as such it is
unlikely that only one factor will change in a given period; stochastic modelling techniques will
reflect the impact from many factors. The output from the stochastic modelling is more meaningful
at a segmental level and is therefore not provided at a line of business level. We show in the
following table, the 10th percentile, 25th percentile, actuarial mean
estimate, 75th percentile and 90th percentile together with the actual
percentile that the selected loss reserves represent.

    As at December 31, 2008

    Management's

    Selected

    Reserve

    Percentile

    10th

    25th

    Mean

    75th

    90th

    Gross Reserves

    ($ in millions)

    (%)

    ($ in millions)

    Property Reinsurance

    $
    488.5

    65
    %

    $
    352.8

    $
    400.2

    $
    466.9

    $
    517.7

    $
    579.7

    Casualty Reinsurance

    1,311.1

    76
    %

    930.5

    1,034.1

    1,181.2

    1,299.1

    1,433.6

    International Insurance

    1,117.4

    66
    %

    804.6

    905.8

    1,076.1

    1,180.2

    1,361.0

    U.S. Insurance

    153.3

    68
    %

    87.3

    106.1

    132.9

    166.5

    201.0

    Potential Variation

    —

    70
    %

    357.2

    200.5

    —

    (182.9
    )

    (382.7
    )

    Total Gross Losses and Loss
Expense Reserves

    $
    3,070.3

    83
    %

    $
    2,532.4

    $
    2,646.7

    $
    2,857.1

    $
    2,980.6

    $
    3,192.6

The 10th percentile represents a 1 in 10 chance that, for example, the property
reinsurance reserves will be at or lower than $352.8 million. The 90th percentile
represents a 1 in 10 chance that reserves will be at or greater than $579.7 million. The
potential variation reflects the fact that not all the segments are perfectly correlated; that is,
we would not expect all lines of business to run off better than or worse than the mean at the same
time.

If the ultimate liabilities equate to the mean actuarial estimate, then the impact from the change
in loss reserves would be to increase net income before tax by $213.2 million (being the difference
above between the selected loss reserves of $3,070.3 million and the mean v