presentation Capital Allocation Fundamental Analysis by mikeholy

VIEWS: 3 PAGES: 33

									2003 Thomas P. Bowles Jr. Symposium



          The Key Issues
           and Mission


         Shaun Wang, Ph.D., FCAS




                             April 10, 2003
The Agenda


   Reality Check

   What Risks to Measure?

   Benchmark Capital

   Fair Value of Liabilities

   Our Scientific Program

                                Shaun Wang

                                             2
          Reality: Poor ROE Performance
          P/C Insurers vs. All Industries 1987–2002
    20%


    15%


    10%

                                                                                        13
     5%
                                                                                        pts

     0%


    -5%
           87   88   89   90    91   92   93   94   95   96   97   98    99   00   01   02E 03F

                               US P/C Insurers           All US Industries

Source: Dr. Hartwig at Insurance Information Institute; Fortune
                                                                        Shaun Wang

                                                                                                  3
  Source: A.M. Best, ISO, Reinsurance Association of America, Insurance Information Institute


      Combined Ratio:
      Reinsurance vs. P/C Industry
                                   Reinsurance                             All Lines Combined Ratio




                                                                                                                                        162.5
170

160                                    Year 2001: Reinsurers did
                                       even worse
150

140
                   126.5




130
                                                          119.2
                        115.8




                                                                                                                                                115.7
                                                                                                                  114.3




                                                                                                                                                        114.4
                                                113.6




120
         110.5




                                                                                                                              110.0
        108.8




                                             108.5




                                                                                                              107.7
                                   106.9




                                                                  106.5




                                                                                                                            106.5
                                                                             105.8




                                                                                                    105.6
                                  105.0




                                                                                                                                                                104.9
                                                                            104.8


110                                                                                    101.6
                                                                                       100.8


100                                                                                              100.5
      1991       1992           1993       1994         1995              1996       1997      1998         1999          2000        2001 2002*
                                                                                                            Shaun Wang

                                                                                                                                                                        4
Reality Sounded A Wake-up Call


   Where were the actuaries during years of severe
    under-pricing and under-reserving?

   Have some of the financial theories contributed
    to market irrationality?

   How can we maintain the continued viability of
    the actuarial profession – the #1 ranked
    profession?

                                    Shaun Wang

                                                      5
In Search for Answers, We Must …

   Get out of comfort zone --- traditional actuarial
    mindset

   Go to the deep water by understanding the risk
    drivers and market dynamics
       How can we project underwriting results without
        knowing the level of market competition?




                                         Shaun Wang

                                                          6
What Risks to Measure?

      Traditional P&C           New Horizon
                                   Business Process Risk
       Risk Analysis
                                   Competitive Game
   Expected Loss
                                   Market cycle
   Loss frequency & severity
                                   Quality of Information
   Correlation between risks
                                   Reaction time
   Concentration of exposure
                                   Incentive misalignment

                                   Multiple Perspectives

                                           Shaun Wang

                                                             7
Focus on “Business Processes”

   Loss Modeling Is Only a Part of the Whole Story
    1. “This company has the brightest actuaries, so it got to be
       good …” --- Naïve thinking

    2. One company had the state-of-the-art actuarial pricing model,
       but in the end still lost so much money

   Need to quantify the Business Process Risk
       Top-line growth in a soft market poses a major risk

       Over-crowded competitive market poses a major risk



                                               Shaun Wang

                                                                       8
A Model of Market Competition

   Financial Result = Min{Quote1, …, Quotek}  Loss

              where Quotek  Normal(k, k)
    1. For long-tailed lines, delayed info  higher k 
       higher chance of premium deficiency

    2. more bidders k  higher chance of premium deficiency

   The Winner’s Curse: In insurance competitive pricing,
    the lowest price gets the business, but may be cursed
    with financial losses


                                          Shaun Wang

                                                              9
Competitive Game of Asset-Liability
Management
   Insurers are competing in two fronts:
       managing assets
       managing liabilities

   Prolific asset management is an “offensive
    play” that necessarily weakens defense
       Can score big during market boom
       In the recent market meltdown, EU insurers were hurt
        the most due to high concentration in stocks

                                           Shaun Wang

                                                               10
        US Insured CAT Losses (in $billion)
        and Rate On Line Index (1989=100)
      Source: Guy Carpenter & *III Estimate            ROL showed big jump after
250                                                    major CAT losses, and then
                                                       came down gradually …
                                                                                           $28.1

200                        $22.9

                                          $16.9


150                                                                  $10.1
      $7.5                                        $8.3 $7.3                  $8.3
                                   $5.5                                                            $5.8
                    $4.7                                                            $4.3
             $2.7                                             $2.6

100
      89     90     91      92     93      94     95    96    97      98      99    00      01     02*
                                                                           Shaun Wang

                                                                                                          11
Market Cycle & Risk Premiums

   Hefty investment gains in the 1990s helped
    insurance capital accumulation

   Pre-Sept 11 oversupply of capital triggered very low
    risk premiums

   The depletion of insurance capital due to Sept 11
    terrorist losses and investment losses
       After Sept 11, the expected hurricane losses had not
        changed, but the insurance rates jumped by more
        than 30%


                                         Shaun Wang

                                                               12
Quality of Information

   Poor Quality of Information is a major risk for
    (re)insurers

   Information asymmetry -- major hurdle for
    securitization (and reinsurers)

   Value of Information?
     Think about the US search for Al Qaeda

   Do we have a measure for “quality of information”?



                                        Shaun Wang

                                                         13
Reaction Time

   “Reaction Time” is an important aspect of risk
         XOL reinsurance has a higher severity volatility than
          proportional reinsurance. However, the reaction time
          for rate increase is quicker for XOL

         Rate increase delays in some regulatory jurisdictions

   For long-tailed liabilities or long-term guarantees:
    the ability to re-act is much limited.
         You have a stack of policies written in the past

         Too late to re-act

                                              Shaun Wang

                                                                  14
Incentive Misalignment

   Many “risks” are created by misalignment of
    incentives
       Underwriters short-term goal v.s. long-tailed liabilities

       Managers’ expansion of his/her own kingdom

       CEO’s compensation linked to growth and acquisition

   Trial Attorneys and the U.S. legal dynamics
       Lawyer Contingent Fees & Punitive Damages should be
        put in a trust fund for public good


                                             Shaun Wang

                                                                    15
Multiple Perspectives of Risk

   Entity-specific value versus Market price
        Market prices tend to exhibit local linearity

        Catastrophe risk to an entity may increase more than
         proportionally

   Volatility
        Outsider view: stochastic and random walk

        Insider view: trend and direction

   Risk of being short-sighted and losing perspective
        NASDAQ bubble;              Variable Annuity Guarantees

                                                   Shaun Wang

                                                                   16
The set of major risks depends on the
specific business /market

                  For Life Insurers

Traditional Risk            New Horizon
 Analysis
                               Asset management
   Mortality/Morbidity
                               Embedded guarantee
   Lapse                       (VADB hedging/reserving)

   Disintermediation          Competitive Game
                                (distribution, expenses)

                               …

                                        Shaun Wang

                                                           17
Risk Measures for Deciding
Capital Requirement and Fair Value
   New Basel Capital Requirement for Insurers (IAIS)
        Parallel to the Banking Basel Accord II

   Movement toward Fair-Value Accounting (FASB)
        Profound implications and heated debates

   Internally, companies are desperately looking for better
    ways of measuring risks and performance
        Companies launched capital Allocation projects
        Lot of confusion, misconception & practical difficulties

                                                   Shaun Wang

                                                                    18
Capital Allocation, or really
Capital Consumption?
   For high-risk low-return business, we want to allocate
    less capital to it, but the capital consumption is high!

   The capital consumption increases more than linearly
    for correlated risks and high-impact losses

   Knowing the capital consumption by business units can
    help manage the business!

   Many allocation methods rely heavily on superficial
    assumptions about diversification between LOBs


                                          Shaun Wang

                                                               19
Superficial Diversification Is Dangerous!


   The pure loss generating process may show a low
    correlation and high diversification benefit

   From business standpoint, playing two different games
    is much harder than playing just one competitive game

   The contagion (or drag) effect may overwhelm any
    diversification

   Over-diversification increases the risk of losing touch
    of reality for executives (and making bad decisions)


                                          Shaun Wang

                                                              20
Right and Wrong Diversifications

   Years of under-pricing were partially caused by the “low
    correlation” argument by some multi-line players

   Diversification needs to match with areas of expertise
        Renaissance Re, a mono-line CAT-writer, achieves
         diversification by geographic region and by peril

   Expanding to a new line of business is very risky

   Citigroup spun-off Travelers; GE selling ERC




                                             Shaun Wang

                                                               21
Benchmark Capital


   Other players’ capital allocation can affect you!

   To avoid artificial effects of diversification, industry
    benchmark capital charge is badly needed

   Parameters are more important than the model

   Benchmarks should reflect the inherent risks of the
    business, regardless of risk portfolio
   It will take a lot of fundamental analysis, expert opinion,
    and timely updates

                                            Shaun Wang

                                                                  22
Did “U.S. Risk Based Capital” Help?

   U.S. Benchmark RBC has only limited success:

       Factor based reserve charges ignored the bigger issue of
        reserve adequacy

       Incentives for putting up inadequate reserves

       Same capital charge factor for premium written in a hard
        market versus in a soft market

   Limitations due to a point-in-time measure, without
    reference to future direction and sensitivity over time


                                              Shaun Wang

                                                                   23
It Is Coming! -- Fair Value of Liabilities


   Actuarial Standards Setters are pushing for fair value of
    liabilities

   Motivated by consistent accounting treatment of assets
    and liabilities

   For actively traded assets, market values are readily
    available

   For insurance liabilities, there is no “active traded
    market” --- “fair value” creates big challenges and
    opportunities
                                           Shaun Wang

                                                                24
Challenges of Fair Value Accounting


   Fair value will introduce more volatilities on paper
        Are we prepared for the “consequences”?

        Is it better to enlarge or dampen the underwriting cycle?

   Heated debate on the credit standing of the liability
    holder:
        There seems to be a conflict between “financial theory”
         and “public interest”

        We will learn a lot more today from the speakers!

                                              Shaun Wang

                                                                     25
Financial Theories for Fair Value


   “CAPM with Zero beta” does not reflect reality

   The Link between Insurance Stock Price and
    Individual Loss Distribution is WEAK!

   Insurance equity prices tend to reflect more of the
    quality of company management
        Renaissance Re --- mono-line writer for catastrophe
         insurance, but very stable stock price appreciation




                                             Shaun Wang

                                                               26
Fair Value: Arbitrage-free versus
Actuarial Models

   Two different models, how do we reconcile them?

   Frictional costs are the missing link
    1. Actuarial models should be modified to reflect
       available hedging in the capital market

    2. Arbitrage-free models assumed complete market and
       zero transaction cost (which are often not the case)




                                          Shaun Wang

                                                              27
Reserve Deficiency for Long-tailed
Liabilities
   Before tackling the fair-value question, we have a more
    fundamental problem of reserve deficiency
   As of 2002, P&C Industry reserve deficiency is estimated
    at $120 billion – Morgan Stanley
         Recently a flurry of billion$ reserve increases
   In 2002, the top 300 EU companies have unfunded
    pension liability > $267 billion -- WSJ
   Reserve uncertainty for Long-Term-Care & Annuity
    Guarantees?
                                             Shaun Wang

                                                               28
Cycle Nature of Reserve Estimates


   The adequacy of reserve estimates showed a
    clear cycle over the years, coupled with the
    pricing UW cycle
        Pressure on short-term performance
        Following the competitors
        Tax smoother for some players
        A slow-death sentence for many companies


                                         Shaun Wang

                                                      29
Recent Dramas in Actuarial Reserve
Opinions
   Mechanical actuarial methods can produce a wide range
    of reserve estimates
   In the past the lowest reserve estimates were often being
    used
   Recently we saw large increases in reserve estimates
        Trigged by lawsuit against professional actuaries
        Dramatic increases in reserve estimates may push struggling
         companies off the cliff

   Nowadays actuarial consulting fee is rated on the
    potential legal liability of the project
                                                  Shaun Wang

                                                                       30
Fair Value and Benchmark Capital Are
Intimately linked
   The fair value of reserve liability necessarily contains a
    risk margin --- (see Steve Philbrick 1994 paper)

   These risk margins should be reflected in the capital
    charge for reserve uncertainty risk

   Otherwise, we create disincentives that would distort the
    fair value calculation




                                            Shaun Wang

                                                                 31
Our Scientific Program

   The Bowles Symposium Call Paper Program:

        An overwhelming response of 25+ paper proposals

        We selected 15+ proposals, which were subsequently
         developed to papers

   This Symposium

        Joint efforts of Georgia State University, the CAS, and the
         Actuarial Foundation

        International event: participants from 9 countries

        Industry-Academic Partnership: 5 universities


                                                   Shaun Wang

                                                                       32
We Need Your Participation!


   In this meeting room we have many bright minds with
    deep research and industry experience

   Interactive discussion is a key feature of this Symposium

   It is important that you share your insights and
    perspectives

   Please try to be concise and clear in making your points

   Have fun!


                                           Shaun Wang

                                                                33

								
To top