The Impact of an Anticipated Fair Value Accounting Framework On US GAAP Reporting P

1
The Impact of an Anticipated Fair Value
Accounting FrameworkOn US GAAP Reporting PC
(Non-Life) InsurersEducational Presentation to
the IASB16 February 2005James C. Votta and
Thomas P. ConwayErnst Young LLP
2
Introduction

• Ernst Young (EY) was engaged by the Group of
  North American Insurance Enterprises (GNAIE or
  the Committee) to model the experience of
  property and casualty insurers (PC) under an
  anticipated fair value (FV) accounting framework.
• Together with the Committee, we identified the
  significant concepts underlying anticipated FV
  accounting, as outlined in the Draft DSOP,
  particularly those that differ from US GAAP.
  These concepts include the discounting of loss
  reserves, the provision for market value margins
  on loss reserves, the immediate recognition of
  profit and/or loss on written premium, and
  others. These concepts were modeled
  incrementally for hypothetical PC insurers in
  multiple reasonably possible economic scenarios
  and the results were quantified in five-year pro
  forma financial statement information. The
  separate FV and US GAAP balance sheet and income
  statement data were compared to demonstrate the
  impact of these FV concepts on the reported
  results of a typical US property and casualty
  insurer. The comparisons focused on trends in
  equity, underwriting income, and net income.

3
CAS Sponsored Research

• The EY approach of incorporating the Draft DSOP
  into pro-forma financial statement information
  differed from other FV research sponsored by the
  Casualty Actuarial Society
• (1) The CAS sponsored studies compared US GAAP to
  FV by restating actual historical US PC industry
  experience from publically available US statutory
  data to US GAAP and then to FV
• (2) This research focused on the impact to loss
  reserves and
• (3) Relied on actual historical interest rates
  and specific market value margin methodologies.

4
Key Findings

• The EY analysis demonstrates that anticipated FV
  accounting for PC insurers requires a
  significant number of assumptions and methodology
  choices, in addition to those required under US
  GAAP. Insurers who make different assumptions or
  methodology choices may report significantly
  different financial results, even if their US
  GAAP results would have been identical. The
  potential lack of comparability across insurers
  may make it difficult for third parties to assess
  company and industry performance.
• Following is a discussion of our key findings
  regarding anticipated FV accounting for PC
  insurers.

5
Equity Will Likely Increase

• The implementation of anticipated FV accounting
  will require an adjustment to the US GAAP opening
  balance sheet. For a typical US PC insurer,
  anticipated FV equity will likely increase over
  US GAAP equity. Under the reasonable assumptions
  we modeled, the expected profit from US GAAP
  unearned premium recognized under anticipated FV
  plus the net impact of discounting and market
  value margins on loss reserves will cause initial
  FV equity to increase over US GAAP equity.
• An insurers invested assets are generally
  recorded at fair value under US GAAP. Therefore,
  the valuation of these assets will not
  significantly affect the FV balance sheet.

6
The Revaluation of Discounted Loss Reserves
Causes Volatility in Underwriting Income

• In a steady state environment, the discount on
  current period incurred liabilities are offset by
  the unwind of discount in prior reserves as
  claims are paid. When interest rates change, the
  revaluation of discounted loss reserves causes
  volatility in underwriting results. This
  volatility is exacerbated when premium volume
  and/or loss ratios are changing.

Actual Historical Interest Rates
7
Changes in the Market Value of Assets Cause
Volatility in Net Income

• Under anticipated FV, changes in the market value
  of an insurer's assets are reflected in income.
  Based on average asset maturities for the
  industry, this creates an additional source of
  volatility in anticipated FV net income. This
  volatility becomes greater as the average asset
  maturity increases.

Actual Historical Interest Rates
8
Changes in the Expected Payment Pattern Cause
Volatility in Underwriting Income

• Even with constant interest rates, the impact of
  changes to the expected payment pattern
  underlying discounted claim liabilities causes
  volatility in anticipated FV underwriting
  results. If the expected payment pattern is
  lengthened, with no corresponding change to the
  estimated undiscounted loss reserves, there is a
  one-time increase in anticipated FV underwriting
  income. If the payment pattern is shortened,
  there is a one-time decrease in anticipated FV
  underwriting income.

9
Changes in the Market Value Margins Cause
Volatility in Underwriting Income

• Under anticipated FV, a Market Value Margin (MVM)
  is included in the loss reserves to reflect the
  market price of risk. The net change in the
  total MVM provision is included in anticipated FV
  underwriting income. In a steady state
  environment, the MVM on current period incurred
  liabilities are offset by the release of MVM as
  claims are paid. Therefore, there is little or
  no anticipated FV income effect in a steady
  state. However, in a dynamic environment, the
  total MVM provision may change as the result of
  changes in premium volume and loss ratios. The
  MVM provision can also change as the result of
  changes in methodology or assumptions. These
  changes cause additional volatility in
  anticipated FV underwriting income.
• Loss Ratio Increase in Year 2000

10
The Provision for Insurers Own Credit Standing
Will Impact Underwriting Income

• Under anticipated FV, an insurers own credit
  standing is reflected in the valuation of the
  insurers claim liabilities. A rating downgrade
  of the insurer leads to the reduction of the
  insurers FV liabilities and this change in
  liabilities flows directly through FV
  underwriting income. A downgrade of an insurer
  may lead to increased net income in the short
  term.

Loss Ratio Increase in Year 2000
11
FV Requires Increased Use of Management Judgment
and Subjective Estimates

• Anticipated FV accounting for PC insurers
  requires assumptions in addition to those
  required by US GAAP, and involves an increased
  use of management judgment that increases the
  subjectivity inherent in estimates of reserve
  liabilities. The additional assumptions include
• The approach and quantification of the MVM.
• The expected profit associated with US GAAP
  unearned premium.
• Explicit provisions for the amount and timing of
  claim and expense payments (payout pattern).
• Future interest rates.

12
FV Requires Increased Use of Management Judgment
and Subjective Estimates

• These, and other assumptions and judgments may
  lead to a range of reported results. For
  example, an insurer that adopts a relatively
  pessimistic financial outlook (assumes shorter
  payment patterns for loss reserve discounting,
  establishes a higher MVM and maintains a shorter
  maturity asset portfolio) will have net income
  that is less volatile from period to period than
  an insurer that adopts a relatively optimistic
  financial outlook (assumes longer payment
  patterns for loss reserve discounting,
  establishes a lower MVM and maintains a longer
  maturity asset portfolio). Different financial
  outlooks may result in different reported FV
  results even with identical underlying US GAAP
  results.

13
Changes in FV Judgments May Produce Trends in
Reported FV Results that are Inconsistent with US
GAAP

• Period-to-period changes in FV assumptions and
  methodologies may cause reported results to
  differ from those developed under a consistent
  financial outlook. For example, period-to-period
  decreases in the MVM could mitigate underwriting
  losses and vice versa. A change to a longer
  expected payment pattern underlying discounted
  loss reserves with no corresponding change in the
  undiscounted loss reserve could have a favorable
  impact on anticipated FV underwriting income.
  Such changes would affect anticipated FV reported
  results, but not US GAAP results.

14
Reliance and Limitations

• This presentation is a summary of an actuarial
  analysis of hypothetical insurer experience under
  various steady state and dynamic economic
  scenarios. The scenarios were constructed to
  demonstrate a particular accounting concept or
  concepts under US GAAP and FV accounting. The FV
  treatment of these concepts was selected because
  no formal standards have been promulgated. We
  cannot guarantee that eventual standards will be
  the same as the FV treatments modeled and
  analyzed in this Report.
• Our analysis included studies of historical
  interest rates and PC insurance industry
  experience. Although we believe this historical
  data is useful to analyze differences between US
  GAAP and FV, we have not projected future
  interest rates or PC insurance industry results
  in this Report.

15
Distribution and Use

• This Presentation has been prepared for the sole
  benefit and use of The Group of North American
  Insurance Enterprises (GNAIE) under an
  agreement with GNAIE. The Presentation may be
  shared with the IASB, FASB and the NAIC but we
  must be formally informed of whom at these
  standard setting authorities will receive our
  Presentation.
• Furthermore, EY has agreed with the IASB to
  allow access to this Presentation through the
  IASB website, on the condition that each
  Recipient agrees to the following
• Neither the Presentation nor any information or
  advice contained therein was intended to, nor may
  any of it, be relied upon in any way by anyone
  else, including the Recipient.
• The Recipient does not acquire any rights against
  EY or its personnel as a result of its access to
  the Presentation and EY does not assume any
  duties or obligations to the Recipient in
  connection therewith. The Recipient agrees not
  to bring, and hereby releases and discharges EY
  and its personnel from, any claims arising out of
  the Recipients access or relating in any way to
  the Presentation or EYs services for GNAIE in
  preparing the Presentation.
• The Recipient shall not refer to or quote this
  Presentation, in whole or in part, in any
  registration statement, prospectus, public filing
  or announcement, loan agreement or other
  agreement or document without the prior written
  consent of EY.