Focus: Enterprise Risk Management (ERM)
Focus: Enterprise Risk Management (ERM)
Focus: Enterprise Risk Management (ERM)
October 2009
Summary
Deation and ination:
Consequences for the insurance industry
and capital management
10
14
20
28
32
36
46
Workshops
52
Extreme scenarios
Emerging risks
Risk responses
Atlas V
Compliance
Mortality bonds and swaps
US Cat business
Financial crisis management
52
55
58
59
60
61
62
63
Conclusion
64
66
Preface
GIVEN THE INCREASING COMPLEXITY and interdependency of
risks, the concept of Enterprise Risk Management has become
a core issue for the insurance industry.
This risk-oriented approach is essential for the management
of identified risks and the anticipation of new exposures, so
in June 2009 we brought together senior managers and senior
Risk Management professionals at a client seminar we organized
in Paris to share and discuss different points of view and visions
in terms of what ERM is and should be.
In the current troubled environment, we cannot ignore the major
macroeconomic issues at hand, especially those that may have
a significant impact on our industry, such as inflation and
deflation. Nor can we ignore their potential consequences for
Risk Management, as highlighted in the presentation given by
Denis Kessler.
On the operational side, our seminar dealt with ERM as applied
to practical issues involving emerging risks as well as extreme
risks such as terrorism, natural catastrophes or pandemic, and
their related risk responses.
We hope that this publication, which is based on the presentations delivered and the debates we had during our client seminar
will provide a consistent approach to ERM frameworks and will
help to enhance the Enterprise Risk Management culture in our
profession.
HEDI HACHICHA
CHRISTIAN MAINGUY
2,000,000 -
1,500,000 -
1,000,000 -
500,000 -
01999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
Source: Ecowin
CONCLUSION
We are exactly in the case of decision making in times
of uncertainty, with two hypotheses:
return of inflation;
no return of inflation.
The choice between protecting or not protecting the
company against inflation is made through a rational
evaluation of the cost of protection, which is a function
of 3 variables:
The probability distribution of the risk (x%);
The net present value of the loss amount if the risk
happens (L);
The risk premium.
The values associated with the 3 parameters are
dependant on the Groups view of the economic envi-
RISK HAPPENS
INFLATION
Yes
Probability
of x%
1: protect
2: do not protect
No
Probability
of 1-x%
No Cost
EXPECTED NPV-)
OF THE STRATEGY
x*L
( 0.2b to 0.3b)
( 0.4b to 0.5b)
1
SETTING THE SCENE
JEAN-LUC BESSON
Chief Risk Officer SCOR SE
It was the first time after September 11th that the possibility of a systematic risk for the insurance industry
was mentioned. Indeed, systematic risk in the banking
industry was very common, but according to different
stakeholders usual positions towards the insurance
industry, that systematic risk was excluded or very
remote. The main question came from the accumulation of these events of low probability and high
severity.
Eventually, many stakeholders came to the conclusion,
especially some regulators and the IMF, that the
insurance industry needed a capital increase but also a
more efficient and effective Risk Management
approach. This was the beginning, to some extent, of
the official story of Risk Management in insurance.
From the regulatory point of view, when the NAIC set
the RBC framework for regulation in the USA, they
envisaged a second step based on scenario testing and
dynamic models. Consequently, Dynamic Financial
Analysis (DFA) was very fashionable in the P&C industry
during this period but, due to the silo organisation of
most of the companies and the lack of appropriate
softwares, amongst other things, it did not prove very
successful.
Furthermore, Cat models after hurricane Andrews met
with success. All of these were implemented between
1990 and 1997. ERM now is benefiting from these first
two approaches. But in reality, there were also big
incentives, if we want to remain moderate, coming
from investors, market analysts, rating agencies and,
to a lesser extent, from the regulators.
Following the U.S. RBC framework, many countries in
the world have moved to a more risk-orientated
regulation:
Canada, Australia, UK with FSAs ICAS in 2004;
Switzerland with the Swiss Solvency Test, which is
already in place;
The European Union solvency regulation, which is
known today as Solvency I, does not consider the
different risks and is based on global considerations which do not require a need for better Risk
Management. The Basel II project launched
in 1999, with its three-pillar approach for the
banking industry regulation, led the launch of
the insurance Solvency II project.
Rating agencies, like many companies, largely anticipated this phenomenon. For example, Standard and
Poors officially promoted ERM in 2005 as a critical
component of its rating methodology. It seemed to be
a very important signal not to begin but to have a more
industrialized and common view for the different companies. We are moving to a world with an internal
model approach for the solvency capital requirements
with more sophisticated modeling than the factorbased methodology.
This new approach is defined by different concepts:
risk appetite;
risk tolerance;
diversification;
evaluation of extreme scenarios;
anticipation of emerging risks.
As far as the IFRS accounting rules are concerned, they
have been adopted as the standard for European
Union-based companies. They introduce higher profit
volatility, and the notion of Best Estimate for the
reserves is putting a lot of pressure on the tradition of
a very prudent and conservative approach. It does not
encompass the security margin, as was the case in the
past, in the European Union or in continental Europe.
There were two consequences in the insurance
industry:
Apart from the UK, it has triggered the fear of a need
for raised capital due to the tradition of prudence and
a conservative approach, particularly on the P&C side,
in the setting and the calculation of reserves in
Continental Europe. So to some extent, due to the
way that the IFRS considers the reserves, this level of
prudence has disappeared or will disappear;
Finally, there was also in continental Europe a certain
scepticism about the fact that we need to increase
the capital for the various companies because we
have not experienced any real bankruptcy or big failure in the insurance industry in the past. Many
stakeholders saw that Solvency I was not really
adapted to the industry, but to some extent there
were no real failures during this period. This stems
mainly from the fact that there was some comfort in
the reserves.
SOLVENCY II
Pillar 1
Pillar 2
Pillar 3
Solvency Capital
Requirement
Regulatory Controls
Risk-based
Internal Model
Standard
Formula
Internal Control
Risk Management
Governance
Compliance
Regulatory Reports
Communication
Transparency
to the Regulators
and the Markets
A three-pillar approach
With the Solvency II reform, the EU project is organised
through three sets of regulatory requirements
(pillars):
Pillar 1 is largely based on, or parallels, the Basel II
approach for banks. It is a calculation of the Solvency
Capital Requirements (SCR), with a risk-based
internal model and/or a standard formula. In the
framework of this regulation, reinsurers will be
obliged to apply the standard and will be able,
according to their capacity and their willingness, to
have an internal model. Nevertheless, during two
years after approval by the regulators, they will be
obliged to also run the standard rule;
European cultures. For Latin countries, some application problems or application issues will probably arise
because these countries are used to having legal texts
and playing with the legal aspects in the court. For
Germans and Northern Europeans, it is also a challenge. So some adaptation is necessary to be able to
fulfil the requirements.
After a long debate, it appears that these Solvency II
rules are generally accepted. However, a relevant question for the future remains the one-year time horizon.
Indeed, if you need to have a multi-year model, it is
very complex because you also need to have a lot of
hypotheses which makes the control of this modeling
very difficult. The regulator has decided in Solvency II
to have just a one-year time horizon. Consequently, it
is supposed to calculate what will happen in one year,
which is quite artificial for long-tail business. Indeed,
the way in which you calculate the one-year time horizon does not consider the fact that, in real life, there
is some monitoring of this type of long-tail business.
Board of Directors
Cat Committee
ALM Technical Committee
CEO
COMEX
Security Committee
Large Loss Committee
Asset Management
Risk Ofcer
IT
Risk Ofcer
Human Resources
Risk Ofcer
2
ENTERPRISE
RISK MANAGEMENT
FOR INSURERS: THE RATING
AGENCYS VIEW LAURA SANTORI
Senior Director, Head of ERM Europe
Standard & Poors
This article, based on the presentation given during SCORs Campus Seminar in
June 2009, sets out Standard & Poors view on
ERM, explains the link between ERM and ratings,
and gives an overview of the findings so far for
EMEA insurers and reinsurers.
I. What is ERM?
What is ERM in Standard & Poors view? Managing risk
is at the heart of what an insurance company must do,
because it is its business. A good Risk Management
program means that the company is able, first of all to
identify, then to measure risks, to project, to set limits
and to keep losses within these limits for all its major
risks. So basically, Risk Management is a protection of
the downside of the balance sheet, avoiding losses that
are outside the tolerance of the company.
The difference between Risk Management and
Enterprise Risk Management is the letter E, the word
Enterprise. It means actually doing the same thing consistently across all the risks in the enterprise, using the
same measures, using the same approach. The difference between Risk Management and Enterprise Risk
Management is what constitutes a competitive advantage for the insurance company. A company that
knows how to manage and measure its risks consistently can also choose the risks that bring better
risk-adjusted return, putting itself in a better position
when compared to other less advanced companies.
ERM components
The elements of ERM that S&P looks at can be depicted
as a temple. At the bottom is the Risk Management
culture, the pillars are Risk Control processes, Emerging
Risk Management, and Risk & Economic Capital
models. These three pillars support the overarching
Strategic Risk Management, which is the ultimate
point of arrival of a good ERM program.
Risk tolerance
Setting risk tolerance and limits is a process. There are
risk preferences. Any company that has been in force
for a number of years does not have to think about
risk preferences anymore. It could be that the companys expertise is in Life, so they do Life and these
sorts of products; or the expertise is P&C; or they do
not want to take a significant interest rate risk. These
are the risk preferences expressed at some point, but
now embedded in the business model of the company.
These risk preferences need to translate into a statement of risk appetite. This could be something like
We want to maintain single A ratings or We do not
want to have a negative net income over three or four
years, or We want to have a certain excess of capital
under S&P/Solvency I/Solvency II measures. This qualitative definition of risk appetite will then be translated
into a very quantitative statement of risk appetite that
we call risk tolerance, and this will then be translated
into risk limits. This is the way, in an ideal world, in
which risk tolerance and risk limits should be set up.
At the beginning, economic capital was used to define
the risk tolerance (for example, we want to hold capital
in excess of 200% of our EC). But then companies
realized that this approach meant managing to the tails
of the distribution. Then the concept of risk tolerance
moved a little bit more to the middle of distribution.
Besides economic capital and surplus targets, risk toler-
a risk adjusted product pricing, which is not well-established yet. Capital budgeting means allocating capital
to activities or business units based on the expected
risk returns and expected risk. Performance recognition
and incentive compensation is another very important
part of spreading the cultural risk throughout the
company.
Overall, Strategic Risk Management is effectively the
competitive advantage that comes out of an ERM
program.
EMEA Insurers
ERM Scores Distribution 31/12/2008
Weak
Adequate
Adequate w/str r/c
Adequate +
Strong
Excellent
11%
1%
9%
13%
6%
60%
Most of the companies that we see are adequate companies. They manage the risk in a very traditional way,
silo-based, somebody does ALM, somebody does
underwriting, and somebody does reserving. Often the
management of risk is not very sophisticated.
The excellent companies are leading edge on everything that concerns risk controls; with much more
established ERM programs, in place for longer with
much less variation, more stable and less liable to
changes: in sum a fully risk aware company. These are
normally companies where you can see risk adjusted
performance measurement, and they have had them
for quite a long time.
Conclusion
How is ERM evolving? In Europe, Solvency II is definitely
helping, although the pace of development that
Standard & Poors would have expected to see in
Europe did not materialize. Since Standard & Poors has
started rating insurers ERM, the distribution has been
3
FROM RISK MANAGEMENT
TO ERM WAYNE RATCLIFFE,
I. Traditional Risk
Management
Traditional Risk Management is often focused on risk
identification, assessment, diversification and mitigation of predominantly operational risks. In traditional
Risk Management, risk identification means looking at
causes and consequences, but with a strong focus on
management of specific risks, only. For instance, the
legal department would deal with legal risks, and the
underwriters would deal with treaty risks or contract
risks, but without interaction between departments.
The risk assessment, in traditional Risk Management,
often done with the help of risk maps as a visualisation
tool, tended also to be silo based. The concept of correlations and dependencies between lines or units was
not very well practiced.
Where internal models existed, e.g. for Life, Non-Life,
credit risk, or asset risk, they were not combined into
one group model. Creating an overarching model was
partly also a challenge due to the insufficient level of
hardware and software available ten years ago. As a
consequence, only the diversification within particular
silos could be modelled properly, and full benefits
across the whole company could not be reaped.
Likewise, the mitigation of risks through hedging
strategies, insurance, reinsurance and other means
was not coordinated across separate areas of
business.
Strategic
Reputation
Legal
Business
operations
Credit
Market
Strategic
Reputation
Legal
Business
operations
Credit
Market
No Risk
Management
After traditional Risk
Management
Insurance
Assumed
insurance
risk
No Risk
Management
After traditional
Risk Management
Legend
LTCM
Ideas
Events
New tools
Controlling bodies
Holistic
Strategic
Black Swans
Risk Culture
Traditional
Risk
Management
ALM
Insurance
Harvard
Business
Review
Douglas
Barlow
Cadbury
Report
on Gov.
Workers
Comp.
DFA
NAIC RBC
Sharpe
S&P
Advanced
Computer
Technology
Financial
Crisis
WTC
Captives
Basel II
Deregulation
Risk Metrics
COSO I
Solvency II
TM
COSO II
CRO
RM
Standards in
Australia
SOX
Enron
Derivatives
Securitisation
Hurricane
Andrew
Katrina
Solvency I
1990
2000
A holistic view
To ensure a holistic view is taken, all the actions from
traditional Risk Management across the company are
consolidated in a central area, in order to obtain an
overall view of all risks in the company. As an example,
extreme scenarios would take into account the exposures from all areas of the business. Also, the impact
of a new business opportunity may be acceptable for
the entity concluding the business but the additional
exposure may accumulate with exposures from other
parts of the Group so that the Groups risk tolerance
for a particular risk is exceeded. This approach should
be applied to all risks including, in particular, reputation
risks. Again, modern computing technology and data
management have significantly improved companies
ability to provide this holistic view.
September 11th is an example of how a single event
could impact on many areas of insured business. A
company writing several lines such as Life, Property,
Aviation, and Workers Compensation would have
been severely affected.
V. ERM implementation
and maintenance
Industry situation
ERM implementation is still a challenge for the insurance industry, although significant progress has been
made, as studies from Pricewaterhouse Coopers
show. In 2004, only 18% of companies interviewed
strongly agreed that ERM was an important part of
their interaction with regulators, rating agencies and
investors. 19% strongly agreed that their organisation had clearly defined standards for risk taking
activities and just 10% had a risk function in place for
at least three years. Four years later, the picture had
ST
RA
TE
G
IC
PE
RA
TI
O
N
RE
S
PO
RT
IN
CO
G
M
PL
IA
N
CE
Objectives
Internal Environment
Objective Setting
Event Identication
Risk Assessment
Risk Response
Control Activities
Information & Communication
tio
Monitoring
isa
an
g
Or
of risks. A gap analysis was performed for each mechanism. Consequently, projects were initiated where
necessary, to close any identified gaps. To keep the
system up-to-date, the framework is continuously
monitored, maintained and enhanced.
The ERM framework:
Provides transparency;
Facilitates communication, within the company as
well as with third parties, such as regulators and
rating agencies;
Provides a strategic overview of all Risk Management
mechanisms within the company;
Indicates the level of maturity of each Risk
Management mechanism;
Provides a bridge to ensure strategic goals and operational activities are aligned;
Improves the risk culture. By using this framework, it
is assured that a common language is applied. The
framework is also the foundation for internal training
and e-learning sessions.
Operations
Reporting
Compliance
Internal
Environment
Risk Culture
Risk Appetite
Embedded Governance
Risk Tolerance
Organisation Structure
Human Capital
Communication Policy
Employee Motivation
Objective
Setting
Strategic Goals
IT Strategy
Operational Plan.
Limits & Guidelines
Operational Performance
Management
Reporting Goals
Compliance Plan
Event
Identication
Economic &
Market
Intelligence
Risk
Assessment
ALM
Capital Model
Extreme Scenarios
Risk
Response
Risk Reporting
Risk Enquiry
Emerging Risks
Process Risk Landscape
Control
Activities
Compliance
Landscape
Compliance
Response
Information &
Communication
Communication
of Strategy
Internal Communication
External Communication
Communication
of Compliance
Monitoring
ICS Assurance
Compliance Dashboard
SCORs ERM dashboard is basically a mapping of the Risk Management mechanisms within the ERM framework
on two dimensions: Risk management components and objectives.
VII. Implementation
challenges
The implementation of a successful ERM is a challenging exercise. One sine qua non is the explicit commitment
of the executive management. But senior management
must also be closely involved. An ERM system must be
comprehensive and consistent, meaning that no part
of the company must lag behind. For instance, an
excellent insurance Risk Management that is fully
developed and integrated is of limited use if at the
same time the asset Risk Management is not embedded into the groups overall ERM framework. Since
there are no off-the-shelf ERM systems that can easily
be rolled out in a company, it must be developed by
the company itself, based on needs, culture and
resources. It is obvious that the installation of a wellfunctioning ERM framework is a major change
management task.
Conclusion
Risk Management has been evolving rapidly in the last
10-20 years with many companies having taken it to
the next stage Enterprise Risk Management.
ERM enhances traditional Risk Management by ensuring that a holistic view of all the companys risks is
taken as opposed to a silo view. ERM also encompasses
upside as well as downside issues thus ensuring that
strategic and operational profitable opportunities (in
relation to risk taken) are exploited. ERM also has a
strong influence in encouraging a consistent and
strong risk culture within the company where clear
ownership of risks is assigned. The Risk Management
function has moved away from a controlling role to
one of a catalyst for change and coordinator of Risk
Management initiatives, with risks being clearly owned
by those responsible for taking risk-related decisions.
4
RISK IDENTIFICATION
AND ASSESSMENT BERND LANGER,
Risk identification and assessment are two important steps in the ERM cycle.
Events that might have an influence on the objectives of a company have to be identified and
evaluated, in order to decide on appropriate
responses.
I. Risk identification
and assessment techniques
The risk identification needs to be comprehensive,
spanning all areas in which the company operates as
well as various timescales. It is, therefore, appropriate
to involve a broad range of people in this process,
including management and subject matter specialists.
Objective Setting
Is our framework
functioning?
Do we have to
reconsider steps?
Information
& Communication
Do we have
the right
management
information?
How do we
communicate?
Internal
Environment
How
do we
respond ?
What are
our control
activities?
Risk Response/
Control Activities
What are
our objectives?
Which
events
have
an inuence
on our
objectives?
How often
do they occur
and what is their
impact?
Risk
Assessment
Event
Identication
Assessed risks have to be measured against the companys stated risk tolerance and risk appetite. For each
risk exceeding pre-defined limits either with regard to
the loss potential or the frequency of occurrence,
responses will need to be implemented as a next
step.
There is a broad range of possible identification and
assessment techniques, a selection of which will be
presented in this article. Many of them combine the
two steps identification and assessment, so they
will be treated together.
V. Emerging risks
Emerging risks or known unknowns are difficult to
grasp, vague in their potential scope and impact, but
dangerous insofar as they have the potential to quickly
develop into large claims for insurance. Many companies already have experience with emerging risks that
turned into major problems, such as asbestosis claims.
The professional identification and assessment of
emerging risks across the company is a central element
of a sophisticated ERM, and one of the aspects that
sets ERM apart from traditional Risk Management.
Conclusion
The best approach for a solid risk identification and
assessment consists of a multi-component system.
Built to fit the specific situation, exposure and resources
of a company, the system can provide a comprehensive
overview of the risks that the company is exposed to,
as well as methods of dealing with them. At the same
time, the process can serve as an important component
of the companys risk culture.
5
RISK APPETITE
AND TOLERANCE
FRIEDER KNUPLING
Deputy Chief Risk Officer - SCOR SE
IV. Addressees
II. Features of risk appetite
and tolerance statements
An effective risk appetite and risk tolerance statement addresses all key stakeholders of the company,
such as clients, policyholders, shareholders, regulators, rating agencies, employees, and brokers;
The statement should be understandable, meaningful, and helpful for stakeholders in forming a view on
the company;
The risk appetite and risk tolerance statement should
be set by the Board of Directors. This statement is a
part of the corporate strategy, and a key part of the
risk strategy of the company, so it needs to be discussed with the Board of Directors;
The companys Risk Committee should be involved in
this discussion on an ongoing basis, and should formally endorse it.
Most publically traded companies declare target profitability rates, either by reference to a risk free rate or
as an absolute amount. Profitability targets and a risk
appetite statement cannot be viewed in isolation; they
have to be looked at in combination and need to be
consistent with each other. For instance, it does not
make much sense to define a very narrow risk tolerance and at the same time set an earnings target of
is the maximum of the internal capital, the rating capital and the regulatory capital, with the internal capital
being computed at the 99% Tail-Value-at-Risk value.
The second one is the buffer which serves for the purpose of avoiding recapitalizations above the defined
frequency.
15% risk free. The willingness to take on risks within
a defined range leads to constraints on the achievable
profitability. Large risks are often correlated with high
expected (nominal) returns, but if risk tolerance requires
a certain level of capital to be held against those risks,
this may limit the achievable return on capital.
Signalling Capital
3.6
3.6
4.3
0.5
4.1
3.4
Take the maximum
of the two denitions
to satisfy all stakeholders
Internal
RBC
A range
rating capital
requirement
Required
Capital*
Buffer
Capital
Target
Capital
Available
Capital**
Risk/Return trade-off
RoE incl. CoC before tax
(above RFR)
Target Capital
range
12.7%
11.9%
11.2%
10%
9.3%
8.2%
7.5%
Opportunity
for capital
redeployment
Replenish
buffer
3
4
5
(33%) (25%) (20%)
7.5
(13%)
10
(10%)
15
(6.7%)
20
(5%)
Probability of consuming
the Required Capital
One objective of the buffer is to fulfil SCORs risk tolerance statement that extreme scenarios with the
probability of 1 in 250 years or less should not consume more than 15% of the available capital. To this
end, SCOR analysed a range of extreme scenarios and
estimated their impact on the economic capital. In
2008, the available capital was around EUR 4.5 billion,
so the limit of 15% translates into EUR 675 million.
None of the events of probability higher than 1 in 250
listed and analysed exceeded this limit. Only the scenario of a global pandemic came very close to the
limit, and for this reason, SCOR took out a mortality
swap in order to keep the risk within the stated limits.
Of course, events that come close to SCORs risk tolerance are monitored particularly closely.
Conclusion
An articulate set of risk tolerance and risk appetite
guidelines is an essential building block for developing
a consistent risk culture in the company. It is also a
good way to keep the level of diversification of risks
acceptable, and to favour interaction between the different business units of the company by sharing the
same principles.
Of vital importance is that the profitability targets of
the company are in line with the companys risk tolerance and risk appetite.
6
DESIGN AND
IMPLEMENTATION
OF AN INTERNAL
MODEL MICHEL DACOROGNA
Head of Financial Analysis and Risk Modeling,
Group Risk Control - SCOR SE
Internal models are fast becoming indispensable instruments for the management
of the insurance and reinsurance industry. And
yet, for many years, the companies in this sector
have existed and thrived without such models.
We therefore need to ask ourselves what, over
the last few years, has led to their development
and generalization.
I. The development
of internal models
There are several reasons why internal models have
developed:
The first reason is the fact that peak risks in insurance
have grown. There are now more risks, occurring
simultaneously and across a wider range.
Furthermore, stakeholders, particularly the companies shareholders, have become much more
demanding in their attitude towards insurance companies and financial institutions. The increase of
importance of other players, such as the rating agencies or supervisors, has strengthened this demand:
- financial stability has become an essential and
decisive criterion for the insured;
- the regulators are looking at insurance again (SST,
Solvency II);
- investors are better informed (return on equity,
new accounting regulations).
without caring for their surroundings. Later, the portfolio theory by Markovitz and even earlier by the Italian
actuary de Finetti, appeared and we began to group
risks together; this approach lasted from the 1960s
until the 1980s. Then in 1993, JP Morgan made public
their quantitative approach to Risk Management and
popularized Value-at-Risk (VaR) as a risk measurement
tool. It was at this time that attempts began to be
made to consider and assess the overall risk of the
portfolio.
Value Protection
Collection of sub
models quantifying
parts of the risks
Quantication of
different risk types
Value Sustainment
Risk
Model 1
Portfolio Data
Risk
Model 2
Operational
Risk
Financial Instruments
Insurance
Risk
Valuation
Model 3
Credit
Risk
Valuation
Model 2
Modeling of
underlying risk
drivers
Risk Factors
Financial Instruments
Distributional and
Dependency
Assumptions
Portfolio Data
Scenarios
Management Strategy
IGR
Management Strategy
Market
Risk
Valuation
Model 1
Value Creation
Valuation Engine
Balance Sheet
Market
Risk
Credit
Risk
Insurance
Risk
Distributional and
Dependency Assumptions
Total Risk
Today, we have reached a form of modeling that is not only interested in the tail of the distribution, but in the whole
of the distribution, with the aim not only of identifying the risk, but also of managing the portfolio.
The objective now is to model the distribution in the centre as well as in the tail to be able to answer questions such
as: how likely is it that the plan defined at the beginning of the year will be reached by the end of the year?
Investment
strategy
Planning
Protability analysis
RBC allocation
What-if analysis
ALM
Solvency
Supervisors
Rating agencies
Risk
Management
Risk mitigation
Reinsurance
optimisation
testing
Hurricane
Earthquake
Total
Expected
62
16
78
Std. Dev.
84
60
104
VaR (1%)
418
332
544
VaR (0.4%)
596
478
690
tVaR (1%)
575
500
678
tVaR (0.4%)
700
598
770
>
<
2005
Underwriting Years
ea
nd
Known today
2006
2007
sk
Ri
2008
Plan for
next UWY
fo
n
re
y
ar
of
x
ne
le
ca
Over this particular portfolio, the mathematical expectation refers to the average losses. For earthquakes,
this value is very low, but looked at from a Value-atRisk at 1% or a Value-at-Risk at 0.4%, the phenomenon
becomes much more significant. If we look at the
portfolio across the board, the Value-at-Risk at 0.4%
and the Tail-Value-at-Risk at 1% are more or less
equal. The distribution allows us to calculate both,
but when allocating capital, it is better to use the
Tail-Value-at-Risk.
For the insured and the regulators, the risk for an
insurance company is that it will not be able to meet
Control/Optimisation
Analysis/Presentation
Output Variables
Company Model
Strategy
Risk Factors
Scenario Generator
Calibration
The SCOR model consists essentially of four basic components. There is a model for Life insurance; a model
for Non-Life and a model for investment, and to these
three models, we then apply economic scenarios. It is
a model for generating these economic scenarios in a
consistent way. This model allows us to decide on different hypotheses:
Inflation hypotheses;
Stock Exchange growth hypotheses;
Interest rate hypotheses.
Y1
This last model cuts across the other three for all of
them use the same economical hypotheses. The final
stage is the production of an internal report.
In this model, the idea of dependency is retained.
Indeed, we have mentioned 500 sources of risk. If they
were all independent, there would be an extraordinary
diversification in the model. Now, this is not the case
Y2
Dependence between
lines of business
Y3
X22
X11
X12
X15
X33
X31
X14
X13
X21
X23
X34
X32
Gumbel - = 2.07
Scatter Plot
Scatter Plot
Scatter Plot
1.0
1.0
0.9
0.9
0.9
0.8
0.8
0.8
0.8
0.7
0.7
0.7
0.7
0.6
0.6
0.6
0.6
0.5
0.5
m4
1.0
0.9
m4
1.0
m4
m4
Scatter Plot
0.5
0.5
0.4
0.4
0.4
0.4
0.3
0.3
0.3
0.3
0.2
0.2
0.2
0.2
0.1
0.1
0.1
0.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
m3
0.8
0.9
1.0
0.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
m3
0.0
0.0
0.1
0.1
0.2
0.3
0.4
0.5
0.6
0.7
m3
0.8
0.9
1.0
0.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
m3
The difference lies in the fact that in the rank correlation the dependency is the same everywhere, whereas
in reality the dependency changes with the size of the
risk. We model such dependency behaviour using
Archimedean copulas (Clayton or Gumbel).
Furthermore, there is also a wealth of dependencies
between the economy and the companys liabilities. In
most of the models, earthquakes, for example, do not
have any effect on the economy. However, in extreme
situations, such events may have important consequences on the economy. For instance, September 11th
accelerated the decline of the stock market and the fall
in interest rates. Conversely, an economy in difficulties
or with weaknesses has an effect on insurance companies commitments. To operate in this type of situation,
it is important for the insurance company to have a
reliable model of these dependencies.
Fig. 18: The full circle of dependency of the balance sheet on the economy
Cash ow
Assets
Liabilities
Investments
Cash & Short term
investments
Accounting
Fixed Income
Economic
Indicator
Equities
Real Estate
LoB2
LoB9
Alternative
Investments
Economy
Equity indices
GDP
Yield curves
Forex
Share of RBC
800
1,200
700
400
15
210
26%
38%
22%
13%
1%
-
3,325
100%
6,170
46%
In EUR millions
Fig.20: Optimization of the composition of P&C and Life business in SCORs group
Minimises risk
taking
rpe
~50% Life
~80% Life
Pri
tio:
isk
he r
ce
: Pri
erm
in t
ra
tio
e ra
r
s of
ft
ce o
Sha
rn
etu
Optimises return
on risk taking
he
of t
risk
in
P&C only
s
term
turn
of r
rp
Sha
Life only
We should never lose sight of the fact that the riskbased capital remains a theoretical estimation. Models
are tools that insurers can use, but they must be aware
of their limitations. It is often impossible to verify the
model statistically. How many 1/100 year events do
we have to check the statistical reliability of our tail
estimates? Nevertheless, we should test it to make sure
that it is a reasonable model. Here are few ways of
doing this:
Sensitivity test: variation in the parameters to see if
the results have meaning;
Prediction test: do the predictions come true?
Developing independent scenarios to check the
model outcomes against them.
Conclusion
These models can considerably improve transparency
and Risk Management. They enable us to assess the
performance of different risks in relative terms. They
facilitate informed discussion at management level:
decisions that go against the model are of course
possible but, insofar as the model exists, will require
justification.
Models are therefore going to play an ever greater role
in the insurance decision-making process, as long as
they continue to be developed, and we manage to
produce more dynamic models which also integrate
the strategies of the companys management.
Debate
Can we say that each company has its own model?
Mr Dacorogna replied that yes we can. Each company
should have its own model, but obviously the methods
used must be the same. It is a question of adapting the
model to the company, mainly to take account of its
complexity and sophistication. Mr Dacorogna considers that it is absurd to begin with a very complex model,
if people are not ready to take on board or comprehend
the results of this model or are not capable of
recognizing its limits and questioning it.
Another participant emphasized the fact that there are
risks that we can avoid, others that we can transfer and
others that we can diversify. But what should we do
about inevitable risks? How do we measure them?
How can we allocate assets and capital to them?
Mr Dacorogna pointed out that this was a pertinent
question. In fact, the company chooses to have, on top
of the capital calculated by the model, a protection
or buffer capital. There are three reasons for this.
The first is the uncertainty of the model. The second is
that we do not wish to be entirely dependent on the
investors, constantly going back to them every couple
of years to increase the companys capital. Thirdly,
this buffer capital is justified by the possibility of an
important crisis occurring.
The other possibility is to set very strict, absolutely clear
limits and define the level of risk covered, going
beyond the model, which as we have said, we know
to be inadequate in certain situations.
7
UNDERSTANDING AND
MANAGING RISK: A COMPETITIVE
ADVANTAGE FOR A GLOBAL
INSURER JEAN-CHRISTOPHE MNIOUX
Chief Risk Officer - AXA Group
Retranscription by SCOR SE
Threat perception
Opportunity realization
Safety-oriented attitude
Innovation-oriented attitude
Survival-driven
Incentive driven
Ambiguous risk acceptance
min
Risk in opportunity
max
Opportunity in risk
Local Risk
Management
LE
LE
GRM: 70 people
Local Risk Management: ~300 people
The central team is organised by type of risk:
A financial risk team is working to define an ALM
approach at group level and to coordinate ALM in
each of the companies;
A P&C risk team is responsible for auditing the reserves,
fixing indicators and metrics for measuring P&C performance, defining the groups reinsurance policy,
monitoring new products and emerging risks;
Its equivalent for the Life risk;
A smaller team on Operational Risk that coordinates
Understanding risks
Data collection
- Customer behaviour
- Market historical data
- Volatility, correlations
Governance
- Clearly dened
responsibilities
- Internal and external audit
- Alignment of interest
Risk drivers
- Sensitivities to risk factors
- Marginal contribution
Risk transparency
- Financial and risk disclosure
- Peer reviews
Dynamic management
- Mitigation actions
- Hedging solutions
Anticipation
- Surveillance
- Plausible stress scenarios
- Emergency plans
Financial risks
The overview of AXAs balance sheet (Fig. 24) shows
the distribution of its EUR 400 billion of assets, including virtually 80% in fixed income (split between
corporate bonds and government risks). The rest is
divided between different classes of assets. It is worth
noting that equity exposure is quite low.
8%
3%
3% 11
11
4%
32
5%
19
4%
17
17
3%
36%
11
134
34%
137
Insurance risks
AXAs Risk Management rests on five pillars:
Product Approval Process (PAP);
Risk appetite;
Reinsurance and retention policy;
Independent review of the reserves;
Long-term vision of emerging risks.
100%
90%
80%
70%
60%
50%
40%
We are here
30%
20%
10%
0%
Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08 Jan-09
US Market (total returns) 1825 2008
450
400
350
300
250
200
150
100
50
0
Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08 Jan-09
Asset swap spread ER30
ITAX Europe
0
Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08 Jan-09
EUR -10Y
USD-10Y
8
WORKSHOPS
In this section, two types of risks
were discussed among the participants.
The first two examples, Terrorism and Pandemic,
are extreme scenarios related to potentially catastrophic type of event. The third and fourth
examples, the effect of magnetic fields and cyber
risks are potential emerging risks.
EXTREME SCENARIOS
TERRORISM
XAVIER COLONNA, Treaty Practice Leader - SCOR Global P&C SE
JEAN-PAUL PERRIN, Property & Construction Practice Leader - SCOR Global P&C SE
PANDEMIC
IRENE MERK,
Head of ERM Methodology & Risk Monitoring - SCOR Global Life SE
Although common source infections can cause outbreaks, their source is usually quickly identified and
eliminated, so they usually do not develop into a pandemic. The pandemic potential in propagated diseases
depends on the contact rate between carriers, and on
the diseases virulence.
HISTORIC EVENTS
Pandemics in recent history were mostly triggered by
influenza. The Spanish Flu from 1918 comes to most
peoples minds when talking about historic pandemics,
and in total there have been between 10 and 13 influenza pandemics with new virus forms since the 18th
century. This is the reason why influenza-triggered pandemics are currently the main subject of concern for
companies and countries dealing with emergency
preparations. The picture was very different in the past,
with many major events caused by other infectious
agents. For instance, the Black Death (probably
caused by Yersinia pestis) wiped out a third of the
European population in the 14th century. The Roman
Empire was victim to the Antonine plague (probably
Date
Death toll
Asiatic or
Russian Flu
1889-1890 1 million
Subtype
H2N2 (not
confirmed)
H1N1
Asian Flu
1957-1958 1 - 1.5
million
H2N2
Hong
Kong Flu
1968-1969 0.75 - 1
million
H3N2
MODELING
In order to assess the potential development of a
disease, and from this the impact on a companys
insured portfolio, a standard epidemiological model
for infectious diseases called the SIR model can be
used, which allows transitions between the states
Susceptible, Infected and Recovered/Removed.
Parameters for a full model are Population, Number
infected, Contact rate, Transmission efficiency, Mortality
rate, and Recovery rate.
A simpler approach is to only model extra mortality.
Historic data can be used as a starting point, but needs
to be adapted to take into account positive factors
such as changes in medical progress as well as exacerbating factors, for example increased international
travel. Excess mortality can be estimated either as a
relative increase, e.g. a doubling in mortality, or as
absolute loading, e.g. 1 extra mortality for a 1:200
event. Guidance for estimating the impact of an
extreme pandemic can be taken from international
bodies, like the CEIOPS in their advice on the calibration of SCR and MCR in the Quantitative Impact Studies
(QIS). The QIS4 calibration of the mortality and disability catastrophe risk uses an estimate for excess
mortality within an insurance portfolio between 1 and
1.5 deaths per 1000 lives in most developed countries
for a pandemic with a level of severity expected once
every 200 years.
Generally, excess mortality is a function of return periods. SCOR uses a Pareto distribution for this purpose.
Sophisticated modeling also includes morbidity and
mortality assumptions per age (group) and per status
of insured.
EXPOSURE
In Life and health insurance, all products will be
impacted, albeit in varying degrees. For the extreme
scenario calculation, SCOR analyzed the sum at risk
under life covers by country, by age and by gender. The
gross impact was reduced by natural hedges, e.g. from
existing disability and Long Term Care claims, and by
mitigation measures such as the mortality swap.
The repercussions of a pandemic would also be felt in
Non-life insurance, but generally to a lesser extent. In
extreme scenarios, an impact on Credit & Surety,
Casualty, Medical Malpractice, Marine and Energy
amongst others should be analyzed. Natural hedges
such as reduced claims under Motor and Aviation lines
will act as a counterbalance.
On the asset side, financial markets and economies
react very strongly to a pandemic. Equity and corporate
bonds may lose value due to the pandemic affecting
entire economies, with sectors such as leisure and
transport hit harder than others. Default rates on corporate bonds from the vulnerable sectors will
increase.
Long-term interest rates are expected to fall, generating an impact on the economic capital that depends
on a companys duration matching.
Business operations will be negatively affected, with
cost of disruption due to absenteeism of staff.
Temporary reasons include employees being directly
affected by the disease, caring for a sick family member,
caring for children when schools are closed, or being
unable to reach the workplace due to break-down or
restrictions of public transport. Permanent absence of
employees as fatalities from the pandemic has to be
taken into account as well. When planning business
continuity and deciding about critical core processes,
temporary unavailability of infrastructure has to be
modelled, on a first level (water, electricity) as well as
on a second level (intranet, telephone, postal
services).
CONCLUSION
A holistic extreme scenario for a pandemic will allow a
company to decide on mitigation measures to install
upfront, such as reinsurance and a well-rehearsed BCP.
But it will also allow the company to actively steer
through the crisis, and thus enable the company to
reap the benefits of a strong ERM.
Although the progress of medicine, pharmaceutical
research and prevention might make a severe pandemic less likely, complacency would be dangerous.
The production of drugs and vaccines are limited and
highly localized, with drug distribution a huge logistical
effort, and the evolutionary arms race between humans
and pathogens means that the best we can do is to be
one step ahead.
EMERGING RISKS
Emerging risks include new
unknown risks and also changing risks that are
already known but evolving; they are marked by
high uncertainty and by being difficult to
quantify.
In order to avoid unexpected losses or unforeseen
claims developments and take advantage of
potential strategic opportunities, SCORs Emerging
Risks Governance & Process unit has created
an observers community across the whole Group,
as well as designing an emerging risk identification
system (the so-called EchO, Emerging or Changing
Hazards Observatory) that enables the observers
to share and discuss their perceived emerging
risks.
The observers are expected to develop their own networks in order to gather (local or global) early risk
ELECTROMAGNETIC
FIELDS DOMINIQUE FORT, Risk Management - Underwriting Audit - SCOR Global P&C SE
FABIEN GANDRILLE, Group Risk Management - SCOR SE
RALF ROESCH, Practice Leader Casualty Treaty - SCOR Global P&C SE
Assessment
Strong changes in all relevant areas (science, legal,
etc.) except social security;
Claims frequency considerably increased in all countries, all lines of business and all activities considered;
Legal costs substantially increased due to class
actions, compensation per person is also increased;
Deep pockets, large corporates like tobacco companies are sued, which could also be extreme;
Bottom-up valuation: we consider our respective
exposure per type of policy and our book exposed to
the major insurers active in the industries mentioned
above. We retain the main aggregate limit per year of
the period of reference (16 years) per country and
amounts are aggregated with all lines of business
combined worldwide (excluding USA).
CONCLUSION - RECOMMENDATIONS
The extent of policy coverage has generally been
contractually restricted through a specific EMF
exclusion or by introducing specific limits of liability,
especially for critical occupancies. Moreover, proper
monitoring and good aggregate control of EMF
exposures must be implemented.
CYBER RISKS
9
RISK RESPONSES
With the aim of respecting its
risk tolerance and the limits attached to it, and in
order to satisfy its profitability objectives, the
SCOR group has implemented various different
solutions relating to the risks it takes on. A few
examples of these are set out below:
The Atlas V cat bond, designed to protect the North
American portfolios with regard to earthquakes and
wind perils, has enabled SCOR to renew the portfolio
inherited from Converium within the Groups defined
risk appetite as shown below, whilst ensuring the
greatest level of diversification in both inward and
outward reinsurance and strengthening the teams
know-how with regard to securitization solutions.
ATLAS V
EMMANUEL DUROUSSEAU
Retrocession Manager - SCOR SE
SCORs motivations:
To secure a source of multi-year retrocessional capacity from the capital markets;
To obtain fully collateralised protection to complement SCOR Global SEs retrocession program as the
available market has been shrinking;
To provide SCOR with alternative capacity as a cycle
management tool;
Investors in Cat Bonds require transparency in the
underlying risks which they are taking, and therefore
SCOR agrees to carry some basis risks as the tradeoff for accessing their capacities. Some basis risk,
even if only very small, is inherent to any synthetic
cover. The principle was to maintain the risk level
in SCORs risk appetite. The goal, however, is to
minimize this basis risk under 1%: the LAZR
methodology gives the best possible compromise
(PCS loss distributed by AIR from state to county
level).
COMPLIANCE
THOMAS LANZ
Group Compliance Officer - SCOR SE
As a response to meeting
increased and tightened regulations and compliance requirements, SCOR has implemented various
compliance initiatives to strengthen the compliance culture and framework.
In order to achieve this goal a new position of a Group
Compliance Officer has been created and compliance
functions were embedded in the Hubs to monitor
changes in the compliance environment and to implement necessary measures. Besides these organisational
measures increased efforts have been made to raise
awareness amongst staff on compliance risks and obligations by using multiple channels such as quarterly
legal and compliance newsletters, e-learning and training seminars. Finally, a new online platform was
implemented to ensure easy access to Group and Hub
compliance policies, news and compliance resources.
Given the financial and economic crisis, increased vigilance needs to be given to the following key
compliance risks, amongst others, as highlighted in
several recent surveys:
IMPROPER PAYMENTS
The World Bank estimates that around USD 1000 billion a year is paid in bribes and a recent Ernst & Young
study* amongst 1 200 major companies in 33 countries revealed that 23% of respondents knew someone
in their company who had been approached to pay a
bribe to win or retain business. Over a third of respondents felt that the problem of bribery and corrupt
business practices is getting worse and 41% of
respondents stated that corrupt practices are prevalent
within the insurance sector. This risk situation requires
the strengthening of a zero-tolerance stance on bribery
and appropriate awareness raising.
FAIR COMPETITION
Training and other efforts need to be considered to
reduce the risk of employees being tempted in light of
the difficult market environment to enter into anticompetitive arrangements with competitors instead of
competing with them. Group-wide policy and training
need to highlight risks related to illegal arrangements
and proper business conduct required. Fines for anticompetitive conducts are increasing. The European
Commission imposed fines of EUR 5.6 billion for cartels
in 2007/2008 (compared to EUR 2.6 billion in
2005/2006).
KNUT PELLNY
Risk Modeling Actuary
SCOR SE
Interest Payments:
LIBOR/EURIBOR + Spread
SPV / SPRV
Collateral account
Fixed investment
returns based on
LIBOR/EURIBOR rates
At Maturity:
Return of remaining
principal amount
Investors
Proceeds
100%
80%
60%
40%
20%
0%
100%
105%
110%
115%
120%
125%
130%
Index Value
US CAT BUSINESS
PAUL HERTELENDY
Chief Underwriting Officer - Specialty Lines 1 SCOR Global P&C SE
FINANCIAL CRISIS
MANAGEMENT MICHLE LACROIX
Chief Investment Officer SCOR Global Investments SE
In spite of an unprecedented
turbulent context, SCOR has anticipated and
successfully weathered the financial storm
through three successive periods:
Cautious asset management: the strategic asset
allocation followed a strict and conservative
ALM Process combined with a conservative fixed
income portfolio and limited exposure to subprime and monoliners;
Cash accumulation: the anticipation of the crisis
led to accumulate cash, shorten duration and
restrict equity exposure. Bonds maturing were
reinvested in cash to cut the liquidity threat and
protect the value of the portfolio;
Crisis management: bank exposure was drastically reduced and closely monitored. Weekly
meetings of the Group Investment Committee
were established for counterparty risk reviews.
7,000
16
14
12
CAC 40 (pts)
March 2007:
subprime crisis
6,000
5,000
4,000
10
3,000
8
6
2,000
1,000
2
0
Cautious Asset
Management
Cash Accumulation
Crisis
Management
Q4 06 Q1 07 Q2 07 Q3 07 Q4 07 Q1 08 Q2 08 Q3 08 Q4 08
Equities
Cash & ST
Bonds
CAC40
Conclusion
NORBERT PYHEL,
Deputy Chief Executive Officer of SCOR Global Life SE
ERM may be seen as the end result of a progressive evolution of the governance of companies,
from Management to Risk Management and from
Risk Management to Enterprise Risk Management.
It has also to be seen as an achievement that is
unstable and in danger by nature and that may not
last long enough to produce all its benefits unless its
maintenance and refinement or enhancement are
being continuously taken care of with all due and
required consideration. For it to be efficient, it must be
kept as dynamic and evolving as the risks themselves,
both known and or yet emerging. Since ERM takes
time to be put in place, to become routine, to produce
its effects and to be proven by being tested, it is
not compatible with ever-changing organisations and
business models.
The transformations of a company towards being
Enterprise Risk Managed are not given and can only
be obtained by the means of a major effort by the
entire organisation. It implies a strong steering from
the top management and an efficient top-down
communication to mobilize energies, create the degree
of appropriation of the objectives and the initiatives at
all levels and generate the bottom-up contributions
and behaviours that are the conditions of its success.
When the decision is made to deploy ERM, it has to
materialize as a major project for (and carried by) the
company and it has to be handled as such, including in
terms of priorities, not to lose momentum. Otherwise
it has the risk of not being bought internally, facing
inertia and delays and ultimately failing at a substantial
cost, which goes beyond large amounts of money
having been wasted.
Not only does the ERM project serve its main purpose
of educating and securing all the decision-making processes but it solidifies the links between all the entities
of the company by increasing their awareness of their
respective contributions and by pooling them around
common thought processes and goals. It also implies
openness to human errors, and, in particular to those
errors that did not give rise to losses and used not to
be disclosed, so that the entire organisation can learn
from the near misses and progress.
Such a project is an integrator as well as a big push
towards longevity in business, sustainable development
and best-in-class performance.
SCOR believes that its investments in ERM are beneficial to its clients too. It views this first seminar dedicated
to ERM as an opportunity to communicate and obtain
feedback in order to improve and adapt its management of changes.
Guest speakers
LAURA SANTORI
Senior Director, Head of ERM Europe Standard & Poors
Laura Santori is a Senior Director at Standard & Poors Financial Institutions Ratings Europe.
Laura is the head of the Paris and Milan insurance teams and, since May 2005, head of Enterprise
Risk Management Europe.
Laura joined in February 2001 from Bacon & Woodrow, an actuarial consultancy firm based in
London. She is a qualified actuary specialised in life insurance.
Before joining Standard & Poors, Laura spent four years with Generali, three of which in Hong
Kong, and three years Bacon & Woodrow, an actuarial consultancy, in London.
JEAN-CHRISTOPHE MENIOUX
AXA Group Chief Risk Officer
Jean-Christophe Menioux is Chief Risk Officer for the AXA group, in charge of Risk Management
for P&C and Life risks, financial and operational risks, and capital management as well. The Group
Risk Management area has approximately 70 people centrally, covering a broad range of areas
including asset liability management, the reserving process, application of the product approval
process, risk monitoring, operational risk, economic capital and Risk Management systems.
Previously he was Director of Financing and group Treasurer. His responsibilities involved the centralized financing of the AXA group, including all types of Capital markets solutions such as securitizations,
subordinated debt or rights issues.
Prior to joining AXA in 2001, Jean-Christophe started his career in 1992 in HSBC France (former
Crdit Commercial de France) as head of trading on Interest rate derivatives and then as Chief Risk
Officer for market risks.
Jean-Christophe holds a M.Sc. in Economics from cole Centrale Paris, and followed an Executive
Development Program in INSEAD.
Jean-Christophe was also professor of International finance in La Sorbonne University.
SCOR speakers
DENIS KESSLER
Chairmain and Chief Executive Officer of SCOR SE
Denis Kessler is a graduate of HEC business school (cole des Hautes tudes Commerciales)
and holds a PhD in economics as well as advanced degrees in economics and social science. He has
been Chairman of the Fdration Franaise des Socits dAssurance (FFSA), CEO and Executive
Committee member of the AXA Group and Executive Vice-President of the MEDEF (Mouvement
des Entreprises de France). He joined SCOR as Group Chairman and Chief Executive Officer on
4 November 2002.
JEAN-LUC BESSON
Group Chief Risk Officer - SCOR SE
Jean-Luc Besson, Fellow of the Institut des Actuaires (France), holds a PhD in Mathematics and has
served as a University Professor of Mathematics and as Senior Vice President of Research, Statistics
and Information Systems at the FFSA (Fdration Franaise des Socits dAssurance - Federation
of French Insurance Companies). He was appointed Chief Reserving Actuary and Member of the
Executive Committee of the SCOR Group in January 2003 and has been Group Chief Risk Officer
since 1 July 2004.
VICTOR PEIGNET
Chief Executive Officer of SCOR Global P&C SE
Victor Peignet, a Marine Engineer and graduate of the cole Nationale Suprieure des Techniques
Avances (ENSTA), joined the Facultative Department of SCOR in 1984 from the offshore oil sector.
From 1984 to 2001, he held various positions in the underwriting of Energy and Marine Transport
risks at SCOR, first as an underwriter and then as Branch Director. He has led the Groups Business
Solutions (facultative) division since it was created in 2000, as both Deputy Chief Executive Officer
and then as Chief Executive Officer since April 2004. On 5 July 2005, Victor Peignet was appointed
manager of all Property & Casualty Reinsurance operations at SCOR Global P&C SE. He is currently
Chief Executive Officer of SCOR Global P&C SE.
NORBERT PYHEL
Deputy Chief Executive Officer of SCOR Global Life SE
Norbert Pyhel serves as Deputy CEO of SCOR Global Life Group and is a member of the SCOR Group
Comex. He joined the group in 2006 with the take-over of Revios, which was an off-spring of former
Gerling Global Re, where he started his reinsurance career in 1981 as a marketing actuary and
member of the German Actuarial Association. Over time he has been involved in business in all
leading life reinsurance markets in the world and the expansion of the global network of local
representations herein. Today his immediate focus is again on the development of SCOR Global
Lifes franchise structured by the four market units and extended over 6 hubs with 27 local offices
serving the clients in more than 80 countries around the globe.
MICHEL DACOROGNA
Head of Financial Analysis and Risk Modeling, Group Risk Control - SCOR SE
Michel Dacorogna is a member of senior management of SCOR SE and heading the group Financial
Analysis and Risk Management modeling team. His main respinsabilities are to develop the Asset
and Liability Management models for the group and on this basis to assess the risk-based capital
of the company and determine the best strategic asset allocation. The coauthor of: An Introduction
to High Frequency Finance, he has also published numerous articles in scientific journals. He is an
associate editor of Quantitative Finance. He received his Habilitation, Ph.D. and M.Sc. In Physics
from the University of Geneva in Switzerland.
FRIEDER KNPLING
Deputy Group Chief Risk Officer - SCOR SE
Frieder Knpling holds degrees in Mathematics and Physics from the Universities of Gttingen and
Freiburg. He worked as a lecturer and research assistant until he received a Ph.D. in Economics from
the University of Freiburg based on research into the econometric modeling of macroeconomic and
financial data. From 1999-2002 he worked for Gerling-Konzern Globale Rckversicherungs-AG and
the former Revios group on group-wide actuarial and financial topics. Since 2007 he has been
heading the Corporate Actuarial Department at SCOR. In December 2008 he was appointed Deputy
Group CRO of SCOR. Frieder Knpling is a fellow of the Deutsche Aktuarvereinigung (DAV).
BERND LANGER
Chiel Risk Officer - SCOR Global P&C SE
Bernd Langer is a graduate of the University of Applied Science, Cologne, where he studied business
economics (subject: insurance). He began his reinsurance career in 1996 at Gerling Global Re as an
underwriter for non-life reinsurance. From 2000 to 2002 he worked as risk controller and internal
auditor for Gerling Group. Bernd joined Converium Ltd., Zurich, in 2003 as internal auditor.
In 2005/2006 he was in charge of Converiums Sarbanes-Oxley project as Head of Internal Control.
In 2007 he lead Converiums SAP FS-RI implementation project. He was appointed Chief Risk Officer
of SCOR Global P&C in October 2007.
WAYNE RATCLIFFE
Director Group Risk Management - SCOR SE
Wayne Ratcliffe is a Fellow of the Institute of Actuaries (UK) and graduate of the University of
Cambridge, where he studied mathematics and operations research. He began his insurance career
in 1981 in Life & Pensions (Equity & Law then Prudential), as a product development and marketing
actuary. In 1994 he set up the London office of SCOR Vie before moving to the Actuarial department in Paris in 1998 as Head of anglo-saxon markets and subsequently as Deputy Actuarial
Director. After a brief spell at XL Re in 2006 he returned to SCOR in 2007 as Head of Group Risk
Management.
FRANK SMER
Chief Risk Officer SCOR Global Life SE
Frank Smer is a graduate of the University of Cologne, where he studied Mathematics. He began
his reinsurance career in 1995 at the Life & Health department of Gerling-Konzern Globale
Rckversicherungs-AG, Cologne. After a number of years as Executive Director in charge of a market
unit profit centre at the company (which became Revios Re in 2003 and SCOR Global Life in 2006)
he was head of Strategic Development and Risk Management at Revios as of 2004. He was
appointed Chief Risk Officer of SCOR Global Life in November 2007.
DEJAGLMC Imprimeur
All rights reserved. No part of this publication may be reproduced
in any form without the prior permission of the publisher.
SCOR has made all reasonable efforts to ensure that information
provided through its publications is accurate at the time of inclusion
and accepts no liability for inaccuracies or omissions.
SCOR
1, avenue du Gnral De Gaulle
92074 Paris-La-Dfense CEDEX France
www.scor.com