Docstoc

insurance-risk-study-aon-benfield

Document Sample
insurance-risk-study-aon-benfield Powered By Docstoc
					INSURANCE RISK STUDY
Risk Quantification in a Connected World
About the Study

Rating agencies, regulators and investors today are demanding that insurers provide detailed assessments of their risk tolerance and quantify

the adequacy of their economic capital. To complete such assessments requires a credible baseline for underwriting volatility. In order to

help its clients address these needs Aon Benfield began a detailed study of underwriting volatility in 2003 leading to the Insurance Risk

Study, which is published annually in September. The Study provides our clients with an objective and data-driven set of underwriting

volatility benchmarks by line of business and country as well as correlations by line and country. These benchmarks are a valuable resource

to chief risk officers, actuaries, and other economic capital modeling professionals who seek reliable parameters for their models.


Using the factors in the Study as a complement to existing industry or proprietary frequency and severity curves, insurers can assess the volatility

of their business using the same metrics as catastrophe models. For example, it is possible to estimate the aggregate loss potential over an accident

year, both with and without the impact of the pricing cycle, analogous to catastrophe model aggregate Probable Maximum Losses (PMLs).

Aon Benfield’s Prime/Re insurance simulation tool is available to clients to carry out the required calculations. Prime/Re is described on
                               TM




the last page of this booklet.


The Study quantifies the systemic risk or volatility of each line of business for 17 countries comprising over 75 percent of global premium.

Systemic risk in the Study is the coefficient of variation of loss ratio for a large book of business. Coefficient of variation (CV) is a commonly

used normalized measure of risk defined as the standard deviation divided by the mean. Systemic risk typically comes from non-diversifiable

risk sources such as changing market rate adequacy, unknown prospective frequency and severity trend, weather-related losses, legal reforms

and court decisions, the level of economic activity and other macroeconomic factors. It also includes the risk to smaller and specialty lines

of business caused by a lack of credible data. For many lines of business systemic risk is the major component of underwriting volatility.



Asset portfolio risk                                                             insurAnce portfolio risk




                      Portfolio Risk
                                                                                                                   Empirical Data
                                                                                 Insurance Risk
Portfolio Risk




                                                                                                                                                       Systemic
                                                                                                                                                      Insurance
                  Systemic                                                                                                                               Risk
                 Market Risk
                                                                                                                                                Naïve Model

                                    Number of Stocks                                                              Account Volume



Modern asset portfolio theory teaches that increasing the number of stocks in a portfolio will diversify and reduce the portfolio’s risk, but will

not eliminate risk completely; the systemic market risk remains. This is illustrated in the left chart, above. In the same way, insurers can reduce

underwriting volatility by increasing account volume, but they cannot reduce their volatility to zero. A certain level of systemic insurance risk will

always remain, due to factors such as the underwriting cycle, macroeconomic factors, legal changes and weather, see right chart. The Insurance

Risk Study calculates this systemic risk by line of business and country. The Naïve Model on the right plot shows the relationship between

risk and volume using a Poisson assumption for claim count – a textbook actuarial approach. The Study clearly shows that this assumption

does not fit with empirical data for any line of business in any country. It will underestimate underwriting risk if used in an ERM model.




2
Aon Benfield Insurance Risk Study

Aon Benfield’s 2008 Insurance Risk Study provides the industry’s leading
publicly available set of risk parameters for modeling and benchmarking
underwriting risk. The factors in the Study are an essential input for
management to accurately quantify and manage underwriting portfolio
risk, and to communicate the results credibly with rating agencies,
regulators, investors and boards of directors. Additionally, the Study will
help companies considering mergers and acquisitions determine post-
transaction underwriting risk and optimal capital structures. Our Investment
Banking Group professionals work with Aon Benfield actuaries to incorporate
our proprietary research into their analysis and structuring advice.

The latest edition of the Study has a greatly expanded scope, including:

> Results from 17 countries comprising                       > Correlations between countries, measuring
  over 75 percent of global premium                            an international pricing cycle

> Results for eight core lines of business, with             > Indication of the state of the pricing cycle by country
  more granular line splits varying by country

> Correlations between lines within a country



20 08 key findings

Highlights of this year’s study include:
> Risk in growing and emerging markets is greater            > There are strong correlations between countries,
  than in established markets, highlighting the need           driven by an international pricing cycle. These
  for region-appropriate risk management initiatives.          are especially evident between the major
                                                               economies. They further dilute diversification
> Commercial lines underwriting risk is
                                                               benefits for international writers.
  substantial across the Study.
                                                             > Eight of the 12 countries with GDP exceeding
> Underwriting risk for personal lines is 10 to 15
                                                               USD 1 trillion, and seven of the G8 countries,
  points lower than for commercial lines and is
                                                               have declining insurance premium-to-GDP ratios,
  often driven by natural catastrophe risk.
                                                               indicating a softening pricing environment.
> Correlations between lines within a country
  vary by country, but are typically moderate to
  substantial. These correlations can significantly
  dilute diversification benefits for national writers.

The Insurance Risk Study applies sophisticated techniques from risk theory to over 1.5 million data points, spanning
17 countries and 178 different line/country combinations. Data sources include regulatory filings, statistical agents
and rating agencies.




                                                                                                                         3
  insurance risk study



  Global Risk Parameters
  The Insurance Risk Study quantifies the systemic risk for companies operating in selected countries in the Americas,
  Asia Pacific and Europe. Systemic risk is measured as the coefficient of variation (CV) of gross loss ratio. All the
  risk factors are calculated on a consistent basis for each line of business in each country and indicate the baseline
  volatility inherent in that line.


  u n d e r w r i t i n g c yc l e A n d i n s u r A n c e p e n e t r At i o n

  In addition to the line of business detail for each country, the 2008 Study also includes an analysis of countries’
  insurance penetration as measured by the ratio of gross written premium to GDP. These statistics provide
  insight into how “insured” countries are and show different levels of insurance penetration for developing
  versus mature markets. This analysis includes historical premium-to-GDP data from 31 countries.



  co m m o n t h e m e s                                                          generAl liAbility,
                                                                                  coefficient of vAriAtion (cv) of gross loss rAtio
  A table of the systemic risk factors for major lines appears
                                                                                   140%                                                                                             140%
  on the following page. Some highlights
                                                                                   120%                                      Mal                                                    120%
  from the study include:                                                                                            HK
                                                                                   100%                                                                                             100%

                                                                                   80%                          Mex                                                                 80%
  > Commercial lines are more volatile than personal
                                                                                   60%                        Tai                                                                   60%
    lines. With few exceptions, personal motor is the                                                               Colo
                                                                                                                                Braz
                                                                                   40%                                                                                 US           40%
    least volatile line across the countries surveyed.                                                                                       Can           Sp
                                                                                                                                              Ger     Fr               Kor
                                                                                   20%                                                          Aus             UK                  20%
                                                                                                                                 Jp
  > Personal property (homeowners) risk is generally                                0%                                                                                               0%
    comparable with the estimated U.S. homeowners                                     0.0%      0.5%         1.0%        1.5%         2.0%     2.5%   3.0%      3.5%         4.0%      0.0%   0.5
                                                                                                               Premium as a Percentage of GDP
    non-cat risk CV of 25 percent, with the exception
    of Brazil which shows a higher level of risk.
                                                                                  commerciAl property/fire,
  > As an insurance market matures its                                            coefficient of vAriAtion (cv) of gross loss rAtio
140% underwriting volatility tends to decrease.                                   140%
                                                                                                              Tai
 The charts to the right compare the level of insurance
120%                         Mal                                                  120%
                       HK
 penetration for each country against the risk of its general
100%                                                                              100%
                    Mex                                                                                Mex      Sing
 liability and commercial property business. The charts
80%                                                                                80%
                                                                                                       Grc          HK
                  Tai
 show that countries with insurance penetration exceeding
60%                           Braz                                                 60%                                      Braz
                      Colo                                                                                     Colo                       Can
                                                         US
 1.7 percent generally have less systemic volatility than
40%                                   Can          Sp
                                                                                   40%
                                                                                                                                 Jp          Aus           UK
                                                                                                                                                                US
                                                                                                                                                                       Kor
                                       Ger      Fr
20%                                                   UK Kor                       20%                                                  Ger
 other countries. This result is intuitive; countries with
                                Jp
                                          Aus                                                                        Mal                         Fr        Sp
 0%                                                                                 0%
                                   have 2.5% 3.0% 3.5% 4.0%
 larger insurance markets 2.0% more accurate data, more
   0.0%   0.5%  1.0%     1.5%                                                         0.0%     0.5%      1.0%         1.5%            2.0%    2.5%    3.0%      3.5%        4.0%
                    and can diversify more. This implies that
 precise pricing Premium as a Percentage of GDP                                                                Premium as a Percentage of GDP

 the risk for developing markets contains a component
 that can be diversified away as these markets mature.                                    Australia     Aus              Greece              Grc      Singapore        Sing
                                                                                          Brazil        Braz             Hong Kong           HK       Spain            Sp
 Finally, the systemic volatility is broadly consistent across                            Canada        Can              Japan               Jp       Taiwan           Tai
                                                                                          Colombia      Colo             S. Korea            Kor      UK               UK
 the mature markets, supporting the idea of common                                        France        Fr               Malaysia            Mal      US               US
 baseline volatility for common lines of business.                                        Germany       Ger              Mexico              Mex




  4
                                                                                                          Aon benfield




underwriting volAtility for mAJor lines by country,
coefficient of vAriAtion of gross loss rAtio for eAch line


AmericAs
                                     br Azil            cAnAdA          colombiA       mexico                u.s.
 Personal Motor                       12%                 17%             15%               11%              14%
                  (Personal)          42%                 22%                                                49%
 Property                                                                 53%               79%
                  (Commercial)        54%                 48%                                                33%
 Liability                            55%                 37%             55%               84%              36%
 Accident & Health                    49%                 40%             100%              8%               47%
 Marine & Aviation                    118%                82%                                                40%
 Cargo                                13%
 Workers Compensation                                                                                        27%
 Credit                               76%                 84%                               22%



AsiA pAcific

                                 Austr AliA    hong kong        JApAn   mAlAysiA   singApore      s. koreA    tAiwAn
 Personal Motor                     3%            45%             2%       13%        33%           8%              7%
                  (Personal)        15%
 Property                                         62%            29%       20%        81%           33%         123%
                  (Commercial)      30%
 Liability                          18%          105%            11%      117%                      26%            60%
 Accident & Health                                27%             3%                                               39%
 Marine & Aviation                  14%           53%            18%       37%        84%           40%         310%
 Cargo                                            62%                                 53%                          40%
 Workers Compensation                             99%             5%       76%        37%                          48%



europe
                                    fr Ance             germAny          greece         spAin                u.k.
 Personal Motor                       14%                 9%              39%               15%              10%
                  (Personal)                              23%                               10%              19%
 Property                             22%                                 68%
                  (Commercial)                            25%                               20%              33%
 Liability                            29%                 26%                               29%              25%
 Accident & Health                    26%                 23%             78%               43%               4%
 Marine & Aviation                                                        75%               29%              50%
 Cargo                                                    18%             93%               21%              22%
 Credit                                                                                     12%




                                                                                                                         5
insurance risk study




s tAt u s o f t h e p r i c i n g c yc l e                                       gross written premium (usd million)
                                                                         U.S.                                                                                                                                  483,406
Fluctuations in the gross premium-to-GDP ratio provide                   U.K.                                                                                                             69,477
                                                                       Japan                                                                                                           67,962
a gauge of the pricing cycle. For each country, we                   Germany                                                                                                          66,247
                                                                       France                                                                                               59,951
compared the latest year against the previous year                       Italy                                                        40,023
                                                                                                                                 34,606
and the five-year average to determine whether the                      Spain


                                                                                                                                            fpo
                                                                  South Korea                                               29,642

level of insurance premium is increasing or decreasing,               Canada
                                                                        China
                                                                                                                             29,581
                                                                                                                         26,260

and whether a change in the cycle is near.                           Australia                                  17,880
                                                                                                               17,246
                                                                  Netherlands
                                                                        Brazil                          14,156

The premium-to-GDP ratio gives a similar result for                    Russia
                                                                     Belgium
                                                                                                        14,128
                                                                                                       13,134

a maturing insurance market as for a hardening                    Switzerland                       9,713
                                                                                                  8,691
                                                                      Austria
market – something worth remembering for countries                    Sweden                     7,943
                                                                      Norway                     7,455
such as India and China.                                              Mexico                6,435
                                                                     Denmark                6,184
                                                                       Turkey               6,122
                                                                      Poland                5,859
                                                                        India              5,223
premium-to-gdp rAtio
                                                                     Portugal              5,045


                 lAtest previous
                                   five
                                                     pricing                     0                       20000                        40000                       60000                        80000                      100000
                                  yeAr                                           gross written premium As A percentAge of gdp (%)
                  yeAr    yeAr                        cycle
                                 Aver Age                         Switzerland                                                                                                                                     4.37%
                                                                  Netherlands                                                                                                                                  4.25%
 Australia        2.37%        2.46%         2.58%      Soft      South Korea                                                                                                                  3.66%
                                                                         U.S.                                                                                                                 3.61%
 Austria          3.13%        3.23%         3.24%      Soft                                                                                                                        3.34%
                                                                         U.K.
                                                                      Austria                                                                                                  3.13%
 Belgium          2.91%        2.89%         2.91%    Trough
                                                                        Spain                                                                                                3.04%
                                                                                                                                                                          2.91%
 Brazil           1.59%        1.56%         1.61%    Trough         Belgium
                                                                       France                                                                                     2.82%
 Canada           2.16%        2.45%         2.65%      Soft        Denmark                                                                                       2.81%
                                                                     Portugal                                                                                 2.62%
 China            0.81%        0.71%         0.71%    Hard /             Italy                                                                             2.52%
                                                                     Australia                                                                         2.37%
                                                     Maturing
                                                                    Germany                                                                         2.26%
                                                                                                                                                    2.26%
 Denmark          2.81%        2.75%         2.77%   Hardening       Sweden
                                                                       Russia                                                                       2.25%
 France           2.82%        2.89%         2.92%      Soft          Canada                                                                     2.16%
                                                                      Norway                                                                  2.02%
 Germany          2.26%        2.36%         2.42%      Soft           Japan                                                      1.69%
                                                                        Brazil                                                 1.59%
 India            0.53%        0.49%         0.51%   Hardening        Poland                                                1.46%
                                                                                                                    1.12%
                                                     / Maturing        Turkey
                                                                      Mexico                                      1.03%
 Italy            2.52%        2.55%         2.54%   Softening          China                                  0.81%
                                                                        India                          0.53%
 Japan            1.69%        1.73%         1.76%      Soft
                                                                                 0                              1                           2                             3                            4                    5
 Mexico           1.03%        1.02%         1.06%    Trough                     gross written premium per cApitA (usd)
                                                                         U.S.                                                                                                                          1,596
 Netherlands      4.25%        4.58%         4.56%   Softening                                                                                                                                         1,586
                                                                      Norway
                                                                  Switzerland                                                                                                         1,304
 Norway           2.02%        2.14%         2.23%      Soft
                                                                     Belgium                                                                                                      1,239
 Poland           1.46%        1.52%         1.54%      Soft             U.K.                                                                                             1,148
                                                                     Denmark                                                                                             1,139
 Portugal         2.62%        2.81%         2.83%      Soft          Austria                                                                                    1,047
                                                                  Netherlands                                                                                    1,045
 Russia           2.25%        2.18%         2.18%    Hard /           France                                                                              981
                                                                                                                                                     899
                                                     Maturing         Canada
                                                                     Australia                                                                      871
 South Korea      3.66%        3.42%         3.20%     Hard           Sweden                                                                       863
                                                                     Germany                                                                 802
 Spain            3.04%        3.07%         3.08%      Soft            Spain                                                               766
                                                                         Italy                                                        680
 Sweden           2.26%        2.40%         2.31%   Softening    South Korea                                                   609
                                                                        Japan                                             531
 Switzerland      4.37%        4.46%         4.39%   Softening       Portugal                                          476
                                                                      Poland                     154
 Turkey           1.12%        1.03%         0.96%     Hard            Russia               99
                                                                       Turkey              81
 U.K.             3.34%        3.65%         3.64%   Softening          Brazil             75
                                                                      Mexico              61
 U.S.             3.61%        3.78%         3.91%      Soft            China        20
                                                                         India    5

                                                                                 0               500                  1000
                                                                                  Summary of premium statistics for top 25 countries by 1500                                                                               2000
                                                                                 written premium. Years reported are the latest available.

6
                                                                                                                         Aon benfield



Reserve Risk
r e s e r v e r i s k i n c A p i tA l m o d e l i n g            i m pAc t o f h i g h e r n e t r e t e n t i o n s o n r e s e r v e r i s k

Quantification of reserve risk is another essential part          Over the past three years many primary insurance
of assessing the adequacy of economic capital.                    companies have offset the impact of the softening
Aon Benfield has developed a simple, yet robust model             market on direct premium by purchasing less
to simulate changes in booked ultimate reserves over a            reinsurance. The resulting increase in net retentions
one-year period. Based on observed volatility in a paid           will impact the volatility of reserves for these insurers.
or incurred development triangle, the method can be               As retentions increase, the volatility of the actuarial loss
applied mechanically to a large number of triangles to            development factors will increase, leading to wider
produce indicative ranges for risk. The table bellow,             confidence intervals for the actuarial “best estimates”
based on a sample of 86 U.S. property-casualty writers,           and greater reserve uncertainty. In turn, this will drive
shows average volatility by total outstanding reserves            greater capital allocations and cost of capital charges.
for personal lines and non-personal lines companies.
                                                                  Increased reserve uncertainty will manifest itself
                          personAl                                in two ways:
 outstAnding                                 multi-line
                            lines                         totAl
 reserves                                    reserve cv
                         reserve cv
                                                                  > New large losses will have a leveraged effect on the
 < $1B                        6.7%                10.0%   9.0%      ultimate loss indications produced by net actuarial loss
 $1-2.5B                      6.1%                10.0%   9.7%      development triangles. Using the standard actuarial
 $2.5-10B                     5.3%                 7.4%   6.7%      link ratio and Bornheutter-Ferguson development
 $10B+                        4.0%                6.6%    6.1%      methods, the impact of a large shock loss on the
 All                          6.0%                9.2%    8.4%      technical actuarial ultimate loss indication will be
                                                                    two to six times the reported loss for casualty lines.

Reserve volatility is substantially lower than prospective        > Higher net retentions will increase the uncertainty
                                                                    of the average severity of net losses reported.
underwriting volatility, as expected. It decreases with
                                                                    This increased uncertainty, coupled with the
reserves outstanding and is lower for personal lines
                                                                    leveraged effect of reported large losses, will
predominating companies. The detailed data shows that
                                                                    increase the potential for reserve shortfalls.
reserve risk is idiosyncratic, with individual companies
often producing results consistently better or worse              For example, for a typical D&O insurance book we
than the average.                                                 estimate that in one year out of ten reserves will be
                                                                  understated by 22 percent assuming a $2 million net
We believe our reserve risk model is superior to                  retention. If reinsurance is eliminated, this modeled
other available approaches: it provides a one-year                shortfall would be 88 percent, or a 300 percent increase.
view of risk rather than a full run-off view of risk,             For a typical umbrella book the shortfall increases
consistent with Solvency II and rating agency                     from 43 percent to 59 percent, or nearly 40 percent.
frameworks; it is sensitive to conservatism reflected
in a company’s ultimate reserve selections; and it
back-tests well against historical reserve changes.




                                                                                                                                             7
insurance risk study



Detailed United States Results
The U.S. portion of the Insurance Risk Study uses data from seven years of NAIC Annual Statement data for
2116 individual U.S. groups and companies. The database covers all 21 Schedule P lines of business and contains
1.1 million records of individual company observations.
                                                                                                                                                                                                                                                                                                                                                                                                          351%

coefficient of vAriAtion of gross loss rAtio | 1992-2007
        125 %
                                                                                                                                                                                                                                                                                                                                                                                   112%
                                                                                                                                                                                                                                                                                                                                                                     107%

        100 %
                                                                                                                                                                                                                                                                                                                                      87%            89%

                                                                                                                                                                                                                                                                                                                 73%
                  75 %                                                                                                                                                                                                                                                                         69%
                                                                                                                                                                                                                                                                             64%

                                                                                                                                                                                                           47%            47%                               49%
                  50 %
                                                                                                                                                                     40%                40%
                                                                                                                            36%                  36%
                                                                                           31%              33%
                                                          25%            27%
                  25 %
                             14%         17%


                      0%
                              Pri         Au                   Co            Wo             Me               Co             Oth                        Me                Sp               Oth               Pro                       Oth                       Ho            Re                  Fid               Int                Re              Re               Pro                 Sp                Fin
                                 vat           to                 mm            rk              dic               mm             er                          dic            eci                 er                du                                  er           me                 ins             elit             ern                  ins           ins                du               eci                  an
                                    eP              Ph                 erc           ers           al M               erc                  Lia                    al M             al L              Lia            cts                                                 ow               ura                ya                  ati            ura            ura               cts                    al P             cia
                                       ass               ysi              ial              Co                             ial                   bil                                     iab              bil              Lia                                               ne               nc                nd                  on             nc                 nc                Lia                  rop              lG
                                          en                cal                 Au           mp             alp                 Mu                     ity            alp                  ilit             ity                          bil                                     rs              e-                 Su                 al            e-                 e-                  bil                ert             ua
                                              ge                  Da              to             en            rac                                           -O            rac                  y                 -C                                 ity                                              Lia                 ret                               Fin              Pro                 ity                y             ran
                                                    rA              ma                                sat           tic              lti                       ccu                tic                                  laim                                -O                                            bil                 y                                  an                pe                   -C                            ty
                                                      uto               ge                               ion           e-                  Pe                                        e-                                                                                                                      ity                                                  cia                 rty
                                                                                                                            Oc                   ril                 rre                  Cla                              s-M                               ccu                                                                                                      l                                   laim
                                                                                                                                cu                                       nc                   im                                                     ad           rre                                                                                                                                             s-M
                                                                                                                                  rre                                         e                     sM                                                 e             nc
                                                                                                                                           nc                                                          ad                                                               e                                                                                                                                               ad
                                                                                                                                             e                                                             e                                                                                                                                                                                                              e




           co m m o n t h e m e s

> Loss ratio volatility, driven by pricing and term and condition changes through the underwriting cycle,
  increases risk by up to 50 percent or more, drives correlation between lines and reduces the benefits of
  underwriting diversification.

> Homeowners insurance is the most volatile major line in the 16-year period of the Study, due in large part to
  the active 2004 and 2005 Atlantic hurricane seasons. Excluding catastrophe losses, homeowners has a risk
  level comparable to commercial auto. The impact of catastrophe losses is to double the volatility for this line.

> Reserve development in long-tailed liability lines has produced upward revisions in estimated
  volatility since 2001. Even favorable development, as seen this year in medical malpractice claims-
  made, can exacerbate the cyclicality of underwriting results and thus increase volatility.

> Personal automobile liability and auto physical damage consistently produce the lowest volatility results.


finAnciAl guArAnty: A blAck swAn event                                                                                                                                                                                        directors & officers (d&o): systemic risk driver
                                                                                                                                                                                                                                                     160%
                                                                                                                            351%
                                                                                                                                                                                                                                                     140%
                      125%
                                                                                                                                                                                                                              Industry On-Level LR




                                                                                                                                                                                                                                                     120%

                      100%                                                                                                                                                                                                                           100%
CV Gross Loss Ratio




                               72%                                                                                                                                                                                                                    80%
                      75%                                      68%           69%                               68%
                                             64%                                                63%                                                                                                                                                   60%

                      50%                                                                                                                                                                                                                             40%

                                                                                                                                                                                                                                                      20%
                      25%
                                                                                                                                                                                                                                                           0%
                                                                                                                                                                                                                                                             0%             2%              4%           6%                8%               10%         12%               14%
                       0%
                               2001          2002              2003          2004               2005              2006          2007                                                                                                                                                              S&P 500 SCA Frequency


Sub-prime lending significantly impacted mortgage insurers’ 2007                                                                                                                                                              Securities class action suits are a significant driver of overall D&O loss
results. In past years the volatility of the financial guaranty line                                                                                                                                                          experience. Historically they have been very predictive of future D&O
fluctuated between 60 percent and 75 percent, among the more volatile                                                                                                                                                         loss ratios, providing a helpful “fast track” view of experience. Aon Re has
Schedule P lines. But the catastrophe event that emerged in 2007                                                                                                                                                              observed a 94 percent correlation between securities class action suits for
caused volatility to quintuple, to 351 percent, and 2008 will likely see                                                                                                                                                      S&P 500 companies and the Public D&O industry ultimate on-level loss
a further increase. Volatility is itself volatile, and risk managers must                                                                                                                                                     ratio data from 1996 to 2006. Based on 28 suits filed as of July 1, 2008,
carefully consider the impact of extreme events on their portfolios.                                                                                                                                                          this historical relationship indicates a 120 percent loss ratio estimate for
                                                                                                                                                                                                                              the first six months of 2008. For comparison, there were 29 claims in all of
                                                                                                                                                                                                                              2007, 18 in 2006, and 60 in 2002, which was the worst on-level year on
                                                                                                                                                                                                                              record. The relationship projects a 76 percent loss ratio for all of 2007.



8
                                                                                                                                                                    Aon benfield



i n s u r A n c e r i s k co m pA r e d to s to c k m A r k e t r i s k         One goal of ERM is to reduce this “self-inflicted” volatility
The 2008 credit crisis has caused considerable                                  penalty, allowing companies to write at greater leverage
volatility in stock prices, as measured by the CBOE                             and to lower their cost of capital over the long term.
VIX. Nevertheless, many insurance lines have shown                              Personal lines, which are more formula rated, show a
greater volatility than stocks in all but the most                              much lower cycle effect, with personal auto volatility
exceptional periods of the last 20 years. For example,                          only increasing by 8 percent because of the cycle.
while the S&P 500 reached a CV of 69% on October
10, 2008, fidelity and surety insurers have experienced                                                                                                           impAct of
                                                                                 line
                                                                                                                                                                pricing cycle
this level of volatility over the 2000-2007 period.
                                                                                 Reinsurance – Liability                                                                       88%
                                                                                 Other Liability – Claims-Made                                                                 61%
                                                                                 Workers Compensation                                                                          49%
                                                    compAr Able
                                                                                 Medical Malpractice – Claims-Made                                                             44%
 line                  volAtility                 period of stock
                                                  price volAtility               Special Liability                                                                             41%
 Private                     14%                        Early 2007               Other Liability – Occurrence                                                                  40%
 Passenger Auto
                                                                                 Commercial Auto                                                                               38%
 Commercial                  25%                      February 2008
 Auto                                                                            Commercial Multi Peril                                                                        26%

 Workers                     27%                        Early 2008               Homeowners                                                                                    18%
 Compensation                                                                    Private Passenger Auto                                                                        8%
 Commercial                  33%                     March 17, 2008
 Multi Peril                                      (day after Bear Stearns)
 General Liability           36%                       Autumn 2002              w h e n w i l l t h e c yc l e t u r n?

 D&O                         40%                   September 19, 2001           The following chart shows the long-term behavior
 Homeowners                  49%                  46% achieved during           of the U.S. pricing cycle. Industry net written
                                                1998 Russian financial crisis
                                                                                premium as a percentage of gross domestic product
 Fidelity and                69%                   October 10, 1987
 Surety                                         (S&P 500 falls below 900)       has fluctuated over the last 38 years, but when the
 Reinsurance -               89%                     80% achieved on            percentage falls below 3 percent it has signaled the
 Property                                            October 27, 2008           beginning of a hard market. We forecast values of 3.06
 Special Property           112%                    150% achieved on            percent for 2008 and 3.01 percent for 2009, which
                                                    October 19, 1987
                                                                                suggest that the current soft market will continue
 No P&C Line                <14%                  Late 2004 - end 2006
                                                                                into 2010 in the absence of a major cat event.

                                                                                industry nwp As percent of gdp

t h e u. s . u n d e r w r i t i n g c yc l e                                   4.50%

                                                                                4.25%
Volatility for most lines of business is increased by
                                                                                4.00%
the insurance underwriting and pricing cycle. The
                                                                                3.75%
underwriting cycle acts simultaneously across many                                                                                                   Hurricane Andrew

                                                                                3.50%                                                                                                      Long-term average
lines of business, driving correlation between the
                                                                                3.25%
results of different lines and amplifying the effect of
                                                                                3.00%                                                                                                 2007, 3.13%
underwriting risk to primary insurers and reinsurers.                                          1974, 2.98%               1984, 3.02%
                                                                                                                                                                  2000, 2.96%
                                                                                2.75%
The Insurance Risk Study shows that the cycle increases                                   Malpractice crisis           Liability crisis                          Soft market & WTC
                                                                                2.50%
volatility substantially for all major commercial                                       1970         1975       1980            1985          1990       1995           2000        2005         2010

lines, as shown in the following table. For example, the
                                                                                          Forecasts (Assuming No Major Cat Event)
underwriting volatility of other liability claims-made,
                                                                                                        NWP         GDP                    NWP
which includes directors’ and officers’ liability claims,                                 Year
                                                                                                       Change      Growth                 to GDP
increases by 61 percent, workers compensation by                                          2008          -0.7%          1.5%               3.06%

49 percent, commercial auto liability by 38 percent                                       2009          -0.7%          1.2%               3.01%

and other liability occurrence by 40 percent.



                                                                                                                                                                                                    9
insurance risk study




Correlation and the Pricing Cycle
co r r e l At i o n o f u n d e r w r i t i n g r e s u lt s


When modeling aggregate underwriting risk it is impossible to overstate the importance of understanding correlation
and dependencies between different lines of business. Modeling is invariably performed using an analysis-synthesis
paradigm. In most applications results are more significantly impacted by the correlation and dependency
assumptions than by all the detailed assumptions made in the analysis step.
The Study determines correlations between lines within each country and also between countries.



co r r e l At i o n b e t w e e n l i n e s

Correlation between lines is computed by examining the results from larger companies that write pairs of lines
in the same country. The tables below show a sampling of the results available, for the U.S., U.K., Germany and
Japan. In each table the correlations shown in bold are statistically significantly different from zero at the 90 percent
confidence level.

u.s.
                                                                                                                     malPraCtiCe
                                                      CommerCial



                                                                    CommerCial
                                                      multi Peril




                                                                                                        ProduCts
                                         Personal




                                                                                 workers



                                                                                           liabilit y
                                         liabilit y




                                                                                                        liabilit y




                                                                                                                                   liabilit y
                                                                                                                       mediCal
                                owners
                                 Home-




                                                                                             otHer




                                                                                                                                     otHer
                                                                                  ComP
                                                                       auto
                                           auto




                                                                                              oCC



                                                                                                           oCC



                                                                                                                         Cm



                                                                                                                                      Cm




 Homeowners                    100%        7%         20%            8%          -9%         -2%         10%           -9%           -4%
 Personal Auto Liability        7%        100%        25%           28%          31%        30%          27%          31%           37%

 Commercial Multi Peril         20%       25%         100%          53%          43%        50%          43%          59%           42%

 Commercial Auto                8%        28%         53%           100%         63%        69%          75%          73%           45%

 Workers Comp                   -9%       31%         43%           63%          100%       64%          60%          76%           57%

 Other Liability Occ            -2%       30%         50%           69%          64%       100%          64%          80%           61%

 Products Liability Occ         10%       27%         43%           75%          60%        64%         100%          80%           23%

 Medical Malpractice CM         -9%       31%         59%           73%          76%        80%          80%          100%          73%

 Other Liability CM             -4%       37%         42%           45%          57%        61%          23%          73%          100%




10
                                                                                                                        Aon benfield




u.k.




                                                                                commerciAl
                                                      & domestic
                                                      household




                                                                   finAnciAl




                                                                                               property
                                                                   personAl
                           Accident
                           & heAlth




                                                                                                            liAbility
                                        privAte




                                                                                  motor
                                        motor




                                                                      loss
 Accident & Health         100%         57%           22%           -25%        82%          56%          57%

 Private Motor             57%          100%          44%           -14%        81%          17%          55%

 Household & Domestic      22%          44%           100%          31%         30%          58%          44%

 Personal Financial Loss   -25%         -14%          31%          100%         -10%         -3%          22%
 Commercial Motor          82%          81%           30%           -10%        100%         23%          64%

 Property                  56%          17%           58%            -3%        23%          100%         33%

 Liability                 57%          55%           44%           22%         64%          33%          100%


germAny
                                                                                             protection
                             property




                                                                                  Accident
                                        liAbility




                                                                    liAbility
                                        gener Al




                                                      dAmAge




                                                                                                          mArine
                                                      motor



                                                                     motor




                                                                                                          cArgo
                                                                                               legAl




 Property                  100%          1%           53%            0%         -17%         -32%         33%
 General Liability          1%          100%          -1%           10%         -9%          -9%           5%
 Motor Damage              53%          -1%           100%           3%         -4%          -53%         48%

 Motor Liability            0%          10%            3%          100%          2%          -22%         50%

 Accident                  -17%         -9%           -4%            2%         100%         40%           2%
 Legal Protection          -32%         -9%           -53%          -22%        40%          100%         -79%
 Marine Cargo              33%           5%           48%           50%          2%          -79%         100%



JApAn
                             property




                                                                                                          personAl
                                                                                                          Accident
                                          liAbility




                                                                                               AviAtion
                                                                                  mArine
                                                                      Auto
                                                        wc




 Property                  100%         -6%            8%           28%         12%           4%          12%
 Liability                 -6%          100%           3%           -23%        -4%           1%          -3%
 WC                         8%           3%           100%          -2%         -19%         -2%           2%
 Auto                      28%          -23%          -2%          100%         12%          12%          18%

 Marine                    12%          -4%           -19%          12%         100%          9%          29%

 Aviation                   4%           1%           -2%           12%          9%          100%         11%
 Personal Accident         12%          -3%            2%           18%         29%          11%          100%




                                                                                                                                  11
insurance risk study




co r r e l At i o n b e t w e e n co u n t r i e s

The following tables show underwriting cycle correlation between countries. These are computed using premium-
to-GDP ratios as a proxy for overall pricing adequacy. Correlation coefficients in bold are significantly different from
zero at the 90 percent confidence level.

europe




                                                                                                                                                                       switzerlAnd
                                                                                                     netherlAnds




                                                                                                                                        portugAl
                                            denmArk




                                                                    germAny
                                 belgium




                                                                                                                    norwAy
                   AustriA




                                                                                                                                                             sweden
                                                                                                                              polAnd




                                                                                                                                                                                      turkey
                                                         fr Ance




                                                                               greece




                                                                                                                                                    spAin
                                                                                           itAly




                                                                                                                                                                                                u.k.
 Austria         100%           70%        0%           86%        85%        -59%        -56%      -13%           88%       -80%      15%         13%      25%       -27%           -68%      77%

 Belgium          70%           100%       51%          53%        45%        17%         20%       36%            81%       -19%      57%         -3%      33%       12%            -47%      44%
 Denmark          0%            51%        100%         47%        -38%       37%         43%       70%            -16%      5%        42%         48%      86%       16%            15%       12%
 France           86%           53%        47%          100%       57%        -25%        -7%       36%            59%       -67%      37%         50%      61%       21%            -34%      69%

 Germany          85%           45%        -38%         57%        100%       -48%        -70%      -33%           76%       -71%      15%         -15%     -23%      -52%           -69%      60%

 Greece          -59%           17%        37%          -25%       -48%       100%        71%       63%            -70%      66%       66%         23%      1%        53%            60%       -51%

 Italy           -56%           20%        43%          -7%        -70%       71%         100%      87%            -56%      54%       52%         70%      43%       85%            73%       -10%
 Netherlands     -13%           36%        70%          36%        -33%       63%         87%       100%           -34%      12%       73%         73%      61%       66%            58%       3%
 Norway           88%           81%        -16%         59%        76%        -70%        -56%      -34%           100%      -61%      -20%        -7%      12%       -32%           -83%      85%

 Poland          -80%           -19%       5%           -67%       -71%       66%         54%       12%            -61%      100%      -28%        -21%     2%        21%            46%       -58%

 Portugal         15%           57%        42%          37%        15%        66%         52%       73%            -20%      -28%      100%        37%      21%       47%            25%       -4%
 Spain            13%           -3%        48%          50%        -15%       23%         70%       73%            -7%       -21%      37%         100%     67%       82%            52%       41%

 Sweden           25%           33%        86%          61%        -23%       1%          43%       61%            12%       2%        21%         67%      100%      43%            10%       40%
 Switzerland     -27%           12%        16%          21%        -52%       53%         85%       66%            -32%      21%       47%         82%      43%       100%           73%       17%
 Turkey          -68%           -47%       15%          -34%       -69%       60%         73%       58%            -83%      46%       25%         52%      10%       73%            100%      -34%
 U.K.             77%           44%        12%          69%        60%        -51%        -10%      3%             85%       -58%      -4%         41%      40%       17%            -34%      100%

AsiA pAcific
                                            hong kong




                                                                                                     singApore
                   Austr AliA




                                                                               mAlAysiA




                                                                                                                              tAiwAn
                                                                                           russiA




                                                                                                                   koreA
                                                                                                                   south
                                 chinA




                                                                    JApAn
                                                         indiA




 Australia       100%           -82%       -28%         -83%       80%        38%         -63%      -11%           -17%      -14%
 China           -82%           100%       4%           78%        -76%       -63%        75%       -21%           69%       -1%
 Hong Kong       -28%           4%         100%         35%        0%         19%         10%       87%            -23%      86%

 India           -83%           78%        35%          100%       -68%       -37%        87%       14%            46%       16%
 Japan            80%           -76%       0%           -68%       100%       29%         -65%      14%            -24%      7%
 Malaysia         38%           -63%       19%          -37%       29%        100%        -15%      56%            22%       26%
 Russia          -63%           75%        10%          87%        -65%       -15%        100%      2%             74%       16%
 Singapore        -11%          -21%       87%          14%        14%        56%         2%        100%           -19%      70%

 South Korea     -17%           69%        -23%         46%        -24%       22%         74%       -19%           100%      -5%
 Taiwan          -14%           -1%        86%          16%        7%         26%         16%       70%            -5%       100%




12
                                                                                                                                                      Aon benfield




AmericAs




                                                              colombiA
                               cAnAdA


                                         mexico


                                                   br Azil
                     u.s.




 U.S.               100%     49%        47%       11%        61%

 Canada             49%      100%       41%       -7%        51%

 Mexico             47%       41%       100%      -9%        34%
 Brazil             11%       -7%       -9%       100%       37%
 Colombia           61%      51%        34%       37%        100%


countries with gdp greAter thAn usd 1 trillion
                                                                          germAny
                               cAnAdA




                                                   fr Ance




                                                                                                                                            br Azil
                                                                                                                                  russiA
                                                                                                                chinA
                                                                                                     JApAn
                                                              spAin




                                                                                                                         indiA
                                                                                          itAly
                                         u.k.
                     u.s.




 U.S.               100%     49%        45%       58%        9%           9%          9%            -19%        27%     61%      42%       11%
 Canada             49%      100%       39%       53%        57%          25%         62%           -24%        -3%     35%      37%       -7%
 U.K.               45%      39%        100%      69%        41%         60%          -10%          36%        -13%     10%      -28%      66%

 France             58%      53%        69%       100%       50%         57%          -7%           34%         7%      20%      8%        35%
 Spain               9%      57%        41%       50%        100%        -15%         70%           -29%        56%     61%      58%        6%
 Germany             9%      25%        60%       57%        -15%        100%         -70%          92%        -67%     -58%     -53%      38%
 Italy               9%      62%        -10%      -7%        70%         -70%         100%          -65%        79%     83%      83%       -34%
 Japan              -19%     -24%       36%       34%        -29%        92%          -65%          100%       -76%     -68%     -65%      41%
 China              27%       -3%       -13%       7%        56%         -67%         79%           -76%       100%     78%      75%       -31%
 India              61%       35%       10%       20%        61%         -58%         83%           -68%        78%     100%     87%       -33%
 Russia             42%       37%       -28%       8%        58%         -53%         83%           -65%        75%     87%      100%      -61%

 Brazil             11%       -7%       66%       35%        6%           38%         -34%          41%        -31%     -33%     -61%      100%




i m pAc t o f co r r e l At i o n o n d i v e r s i f i c At i o n b e n e f i t

Assessing and incorporating correlation into underwriting risk and economic capital modeling is essential.
Today actuaries appreciate there is more to assessing dependency than a simple measure of linear correlation –
although it remains a very important input. The table below shows a computed 100-year PML from a realistic
portfolio of five lines of business where the lines are aggregated under different dependency assumptions.
Prime/Re can reflect many subtle types of correlation in its modeling and is built to facilitate stress-testing
correlation assumptions.

                                                                                                  dependency Assumption
                                                                                                                  50%                  50%
                                                                            50%
                                                                                                             correlAtion          correlAtion
                                                                         correlAtion                                                                   perfectly
                                         independent                                                          t-copulA, 8          t-copulA, 3
                                                                           normAl                                                                      dependent
                                                                                                              degrees of           degrees of
                                                                           copulA
                                                                                                               freedom              freedom
 100-year PML (millions)                           82                               175                          182                       193            252
 Diversification Benefit                          67%                               31%                          28%                       23%            0%




                                                                                                                                                                   13
insurance risk study


Aon Benfield Analytics
Capital Modeling Capabilities
Aon Benfield Analytics’ team of actuaries, cat modelers,
accountants and rating agency experts works closely          fA s t A n d e A s y-t o - u s e

with brokers to provide relevant and timely analysis and
                                                             > Intuitive model structure interface.
to ensure clients are equipped to make the best possible
risk transfer and risk financing decisions. Technical        > Provides instantaneous feedback on model
experts are involved in all stages of the broking               parameterization, without the need for simulations.
process, from structuring and coverage evaluation
                                                             > Fast, detailed risk simulation engine.
to submission and placement. We believe this close
integration between analytics and broking allows us to       > Distributed processing to improve simulation speed.
design and implement the best solutions for clients

Prime/Re and ReMetrica, our award winning Dynamic
                                                             co m p r e h e n s i v e A n d s o p h i s t i c At e d
Financial Analysis capital modeling software, are
currently used by many of the world’s leading insurance,     > Models all types of risk transfer financing: traditional
reinsurance and actuarial consulting firms. Aon Benfield        and variable feature reinsurance, insurance linked
Analytics has a global functional alignment, allowing           securities, sidecars, cat bonds, swaps, etc.
us to serve clients consistently serve clients. We provide
                                                             > Flexible modeling of all types of underlying insurance
analysis to global reinsurance clients through our local
                                                                business including ground-up, excess of loss,
offices using a common methodology and approach.
This provides the best possible backup to the global-           large deductible, layered and shared property.

local reinsurance purchasing decisions that are              > Correlation and dependency through common
becoming increasingly common in the industry today.             mixing variables and copula methods, including

Our analytics are implemented in Prime/Re and                   more flexible extreme tail correlation.
ReMetrica through our award winning Dynamic                  > Automatic treatment of attritional losses,
Financial Analysis capital modeling software. Our tools
                                                                retaining correlation with larger losses and
are currently used by many of the world’s leading
                                                                incorporating loss ratio uncertainty.
insurance, reinsurance and actuarial consulting firms.
Our software is a stand-alone application, developed         > Enhanced loss payout algorithm provides
and supported by a team of experts, including                   systematic pattern uncertainty and fractional,
actuaries, mathematicians and software developers,              lump-sum and annuity payouts.
based in London and Chicago. The stochastic
                                                             > Incorporates results from all commercial
and scenario-based capabilities simulation engine
                                                                catastrophe models.
enables the evaluation of the whole spectrum of
risk financing – whether at the business unit, line of
business, company, or group level. Using our extensive
library of pre-developed components, coupled with
our comprehensive insurance risk parameterization
research, realistic and transparent risk models can be
built quickly for any specific business need. Key client
value features of our capital modeling software include:




14
                                                         For more information on the Insurance
c u s to m i z A b l e A n d f l e x i b l e             Risk Study, Prime/Re, ReMetrica, or
                                                         our analytic capabilities please contact
> Aggregates data from other models
                                                         your local Aon Benfield broker or:
    (investments, reserving, life).

> Able to import/export large models to/from Microsoft
    Excel and to link to other corporate data systems.   Americas
                                                         s t e p h e n m i ld e n h A ll
> Multi-user capability.
                                                         Head of Aon Benfield Analytics, Americas
> Customizable to the client’s metrics for               t: +1 312 381 5880
    risk, volatility, and capital benefit.               e: stephen.mildenhall@aonbenfield.com

                                                         International
                                                         John moore
b u i lt- i n f u n c t i o n A l i t y
                                                         Head of Aon Benfield Analytics, International
> More than 95 “out of the box” modeling components.     t: +44 (0) 20 7522 3973
                                                         e: john.moore@aonbenfield.com
> Multi-year models.

> A.M. Best and S&P current and enhanced
    capital adequacy models.

> U.K., French, Canadian, Australian and
    other regulatory capital models.

> Capital allocation supporting commonly
    referenced VaR, TVaR, and risk-adjusted
    probability based methodologies.

> Income statement, balance sheet, and cashflow
    analysis under multiple accounting standards.

> Reinsurance risk transfer analysis.                       s o u r c e s: A . m . b e s t, A xco
                                                            i n s u r A n c e i n f o r m At i o n s e r v i c e s ,
> All common European, London and                           n A i c A n n uA l s tAt e m e n t s
    Australian indexation clauses.                          (u. s .), fs A r e t u r n s (u k ), b A f i n
                                                            (g e r m A n y ), m sA r e s e A r c h
> Consideration of reinsurer credit default risk.           (c A n A dA), s u s e p (b r A z i l), h ko c i ,
                                                            insur Ance reseArch institute
> Diagnosis and audit tools.                                ( J A pA n ), ko r e A f i n A n c i A l
                                                            s u p e r v i s o ry s e r v i c e , b A n k
                                                            n e g A r A m A l Ays i A , m o n e tA ry
                                                            Au t h o r i t y o f s i n g A p o r e ,
                                                            tA i wA n i n s u r A n c e i n s t i t u t e ,
                                                            A n n uA l f i n A n c i A l s tAt e m e n t s .




                                                                                                                  15
              200 East Randolph Street, Chicago, IL 60601
   t: +1.312.381.5300 | f: +1.312.381.0160 | www.aonbenfield.com

Copyright Aon Benfield 2008 | Published by Aon Benfield Marketing and Communications | #1306 – 08/2008

				
DOCUMENT INFO
Shared By:
Categories:
Stats:
views:42
posted:3/24/2011
language:English
pages:16
qihao0824 qihao0824 http://
About