Enterprise Risk Management (ERM) A risk-based approach to ... - Scor

Focus 

November 2010 

Enterprise Risk Management (ERM) 

A risk-based approach 

to the management 

of a (re)insurance company

ENTERPRISE RISK MANAGEMENT

Summary 

Preface 3 

1 Implementation of an ERM framework 4 

2 Adapting solvency regulations to times of crisis 16 

3 Reinsurance optimization in the context 

of capital management 24 

4 Capital assessment beyond stochastic modeling 36 

5 Integrating risk modeling into the organization 44 

6 Investment and Risk Management strategies 

in a changing regulatory framework 52 

7 ERM and Economic Capital Models: The A.M. Best view 64 

8 Macro-economic standpoint 72 

9 How to deal with the impact of inflation 

on pricing and reserving 80 

10 The financial markets view on the Risk/Reward strategies 

in the (re)insurance industry 90 

ERM: a system to ensure the optimal security and efficiency 98 

Speakers’ biographies 101

2 - November 2010 - SCOR 

The views and statements expressed 

in this publication are the sole 

responsibility of the authors.

Preface 

In the current stochastic, financial and economic environment, 

which is characterized by shocks and instability, and in view of 

impending regulatory changes such as Solvency II and its equivalents 

under consideration outside the EU, Enterprise Risk Management 

or ERM has become a top priority for the (re)insurance industry. 

Following SCOR’s first ERM seminar last year, which presented the 

main concepts involved in ERM and how they could be a driving 

force for our industry, the Group held a second seminar in June. 

The themes discussed this time around took matters a step further 

than last year, focusing on how to implement an ERM framework, 

on integrating risk and economic capital management into 

a company, and on how ERM can contribute to the decision making 

process, for example in terms of reinsurance optimization. 

The seminar also looked at why we need to adapt solvency 

regulation to times of crisis and at possible investment and risk 

management strategies in a changing regulatory framework. 

More generally, the June seminar revisited the macroeconomic 

environment and the best way for risk takers to cope with it 

from an ERM standpoint, and presented the financial market view 

of the risk reward strategies currently being applied in the 

(re)insurance sector. 

SCOR Global P&C hopes that this publication, which is based on 

the presentations delivered and the discussions held during the 

June seminar, will help to promote ERM culture in our industry. 

HEDI HACHICHA 

Head of Strategy and Development 

SCOR Global P&C

1 

IMPLEMENTATION 

OF AN ERM 

FRAMEWORK 

This article presents the concept 

and objectives of Enterprise Risk Management 

(ERM). Often ERM is presented in abstract terms, 

with many definitions and few concrete applications. 

However, this article is intended to illustrate 

the meaning of ERM through SCOR’s day to day 

experience in this field and to describe how Risk 

Management works in practice. 

Fig. 1: Building blocks of ERM 

Risk and 

Economic 

Capital 

Modeling 

Standard and Poor’s way of describing ERM has been 

herein adopted. The corresponding graphical presentation 

is a Greek temple as shown in Figure 1. The base 

represents Risk Culture. Built on this foundation are 

3 specific pillars: 

• Risk and Economic Capital Modeling to deal with 

present and upcoming risks; 


MICHEL DACOROGNA 

Deputy Chief Risk Officer, SCOR 

WAYNE RATCLIFFE 

Director Group Risk Management, SCOR 

Strategic Risk Management 

Emerging 

Risk 

Management 

Risk Management Culture 

Before going into details, it is important to note 

that risk constitutes the core business of insurers 

and reinsurers and gives them the opportunity to 

provide insured people with protection against 

risks to which they are exposed. Clearly, risk in this 

context should not be considered just as a danger 

but also as an opportunity for insurers and reinsurers 

to provide a valuable service and receive a 

fair reward for taking such risk. 


Control 

Processes 

• Emerging Risk Management to deal with future risks; 

• Risk Control Processes to control and support the 

management of these risks. 

The implementation of these 3 elements supports 

the roof of the temple, a process referred to by 

Standard & Poor’s as Strategic Risk Management. 

This is the ultimate aim of ERM.

I. Risk Culture 

What does Risk Culture mean for a (re)insurer? In fact, 

Risk Culture forms the basis of a solid risk management 

policy within the company, as illustrated in the 

Greek temple. 

The foundation of Risk Culture is strong internal 

risk-based governance. At SCOR this governance is 

overseen by a Board Risk Committee which reports 

to the Board of Directors. The main responsibilities of 

this committee are: 

• Ensuring that the company has an effective ERM 

framework in place; 

• Proposing an appropriate risk appetite framework to 

the Board and ensuring this is clearly communicated 

to and understood by all stakeholders, in particular 

by staff; 

• Monitoring and reporting on the Group’s risk profile 

to the Board; 

• Monitoring and reporting critical risk issues to the 

Board. 

Risk Culture benefits from the appointment of a Chief 

Risk Officer (CRO) who is a member of the company’s 

Executive Management. He/she is responsible for the 

management of the above areas and is expected to 

provide regular updates to the company’s Executive 

Management (weekly at SCOR) and the Board Risk 

Committee (quarterly at SCOR). 

At SCOR, the day-to-day management of these 

areas is dealt with by the Group Risk Management 

(GRM) department which reports to the Group CRO. 

The operating divisions (SCOR Global P&C and SCOR 

Global Life) also have their own Risk Management 

organizations, headed by a Division CRO who has 

a dotted line reporting to the Group CRO. Both 

organizations work closely with GRM. 

From a governance point of view it is also imperative 

that a clear separation of roles between risk decision 

takers and risk managers exists. In particular the risk 

takers must be accountable for their business decisions. 

The various levels of decision making should also be 

risk-based, e.g. critical risks should be owned and 

managed by members of Executive Management. 

At SCOR, various risk-related committees, at or below 

the Group Executive Management level, provide 

formalized decision making forums which enable the 

views of risk decision takers and risk managers to be 

taken into account. For example the Group Asset 

Liability Management (ALM) Committee is in charge of 

capital allocation (to assets and liabilities) and strategic 

asset allocation. The Group Investment Committee is 

responsible for tactical asset allocation and ensures that 

the investment guidelines are respected. 

SCOR - November 2010 - 5

Fig. 2: Process to derive Risk Preference and Risk Profile 

COMEX/BoD decision: 

• Risk-Based Capital (RBC) 

for asset risk is less than 

25% of total RBC 

• S&P capital for asset risk 

is less than 15% of S&P 

Net Total Adjusted Capital 

Investment/ALM 

Committee decision: 

• Investment guidelines 

• Strategic asset allocation 

• Limit allocation 

Finally Risk Culture is reinforced by: 

Asset Risk 

Preference 

• A remuneration system which incorporates incentives/disincentives 

for management and staff to 

optimize risk and returns. The formula for SCOR’s 

staff bonuses incorporates a significant element in 

respect of individual performance which is based on 

objective evaluation criteria including a part which 

rewards individual contributions to effective risk 

management; 

Definition of 

Risk Appetite 

Total Capital 

Allocated 

RAC 

Assets 

Fig. 3: ERM affects the entire organization 

Risk monitoring and reporting 


RAC 

Liabilities 

Board of Directors 

Liability Risk 

Preference 

Risk reporting Governance and direction 

Risk reporting 

Performance monitoring 

Risk identification & assessment 

Mitigation plans + responsibilities 

Performance reporting 

COMEX 

Risk Management 

COMEX decision: 

• Peak limit decision 

• Peak limit allocation 

• Product selection (NPI) 

• LoB/geographic 

growth areas 

• Underwriting 

guidelines 

• Risk-based, Group-wide policies and guidelines in areas 

such as ERM, reserving, underwriting, accounting, 

asset management, human capital management, 

compliance, internal audit, etc.; 

• Risk-based internal control standards (including 

exposure limits) at the process level. 

Policies & Strategy 

Prioritization and threshold setting 

Appropriateness of risk mitigation plans 

Decisions, guidance and ”sponsorship“ 

Process and guidelines improvement 

Risk / tolerances adjustment / change 

Risk plan execution 

Operational Processes Operational Processes 

Mission and vision, risk policy and appetite

II. Risk and Economic Capital 

Modeling 

The first pillar of ERM is the Risk and Economic Capital 

Modeling of the company. In order to be efficient, the 

model has to include certain characteristics: 

• Consistency (of the model), which addresses all 

types of risk such as underwriting risk, market risk 

and credit risk. These three risk areas are considered 

to be the most significant for SCOR. 

• Quality of the data, which helps to resolve 

potential deficiencies through regular validation 

and processes. 

• Appropriateness of assumptions, ensured e.g. by 

stress tests. 

• Capacity for identifying the main risk drivers of 

the company. 

• Quality assurance of Risk Adjusted Capital. The RAC 

modeling process and the connection to ERM and 

planning processes i.e. the internal model should be 

embedded in the business processes of the company, 

thus reinforcing Risk Culture. 

• Accurate programming and solid change 

management go without saying but they are very 

important for securing the current status and for 

any future changes. 

Assets 

Investments 

 

Cash & Short-term 

investments 

Fixed Income 

Equities 

Real Estate 

Alternative 

Investments 

Cash flow 

Accounting 

Economic 

Indicator 

Economy 

Equity indices 

GDP 

Yield curves 

Forex 

SCOR includes all relevant risks in its Asset and Liability 

Management approach. Figure 4 illustrates the issue 

of consistency. On the balance sheet of an insurance 

or reinsurance company, there are assets on one side 

and liabilities on the other. Cash flows and accounting 

connect assets with liabilities. Broadly speaking, an 

insurance or reinsurance company receives a premium 

that it invests. When the company has to pay claims, it 

sells some shares or assets that it has in its possession. 

This is the way most of the internal models look at the 

dependency between assets and liabilities. Nevertheless, 

the Asset and Liability Management approach is more 

complicated in reality as it is impacted by the economic 

situation. In fact, all phenomena affecting the economy 

have an influence on the value of both assets and 

liabilities. 

Here are two cases illustrating the question of 

dependency: 

The first example refers to a life insurance company 

that is committed to paying liabilities in ten years, at an 

interest rate of 5%. If the economic climate deteriorates 

(for instance, if a crisis occurs), the market value of its 

liability will increase because the interest rate will have 

decreased. This is because a debt payable in ten years 

will appear bigger when the interest rate is low. On 

the other hand, the behavior of the assets backing the 

expected liability will depend on the nature of those 

assets. Usually a large proportion of assets will be 

invested in bonds, and if the bond portfolio is chosen 

to have exactly the same duration as the liabilities 

Fig. 4: SCOR integrates all models in its ALM approach (consistent model) 

Liabilities 

Lines of Business (LoB) 

LoB1 

LoB2 

LoB9 


then the value of the bond portfolio will also increase 

by the same amount as the liabilities. This so-called 

‘perfectly matched’ portfolio is based, however, on 

the assumption that the expected future liabilities 

can be estimated with precision. In practice this is not 

always possible – if the liability cash flows turn out 

to be far greater than originally estimated then there 

are no available assets backing the shortfall. But it is 

precisely this shortfall that determines the amount of 

required capital. If interest rates fall then the value of 

this shortfall will increase, thus increasing the amount 

of required capital for the company. 

The second example focuses on natural catastrophes. 

Not all natural catastrophes have the same influence 

on the economy. An earthquake in Tokyo is expected to 

impact the economy in many ways. Many studies have 

demonstrated that such a quake is likely to influence 

interest rates, the stock market, and the Japanese Yen’s 

foreign exchange rate. 

SCOR’s model deals with such big catastrophes. They 

are very important in terms of calculating the capital for 

e.g. 1/100 year events. It is thus essential to integrate 

all the impacts of this type of event into SCOR’s model 

in order to come up with a realistic estimate of the 

required capital. This is the type of information that 

determines whether or not the model is consistent. 

Named events / stress scenarios* 

• Analysis of clearly defined and described events 

• Events can have happened in the past 

or may be possible in the future 

• Events can consist of a single risk factor 

or of a combination of several risk factors 

• Definition of direct and indirect impact, 

management actions, and contingency plans 

Figure 5 presents the pros and cons of two modeling 

approaches: both stochastic models and extreme scenarios 

are an important part of modeling. But since both 

approaches have their limitations, it is better to use both 

approaches to complement each other. The stochastic 

models allow (re)insurance companies to explore many 

possible outcomes. Nonetheless, it is complicated to 

calibrate the interactions between different phenomena 

and to deduce the risks involved. The extreme scenarios 

explore only a few scenarios, so the range of risks is very 

constrained. Nevertheless, this enables companies to 

focus on the dependency and to build it into the various 

scenarios. From a catastrophe such as an earthquake in 

Tokyo, it is possible to determine the consequences on 

the stock market and the interest rate, depending on 

the parameters chosen. By comparing the results of both 

approaches, it is possible to verify the plausibility of the 

results. For instance, if the probability given by the internal 

model of the outcome of a scenario is consistent with 

the examined scenario, this makes it more likely that the 

stochastic model includes the same kind of cases. 

Stress tests combining both approaches are difficult to 

develop; nonetheless they are very useful because they 

reveal risks that are not detectable in the balance sheet. 

Thus, the best approach for these economic models is 

to combine stochastic models and extreme scenarios. 

Incidentally, this is what the Swiss Solvency test requires 

from insurance companies. 

. 

Fig. 5: Stress Testing: Two forms of worst case analyses should be part of the model 


Examples: 

- Financial distress 

- Severe adverse development in reserves 

- Tokyo earthquake 

- Retrocessionaires‘ default 

* Based on Lloyd’s, RDS, scenario catalog by the Swiss Solvency Test 

and SCOR specific scenarios. 

Extreme tail scenarios 

• Detailed analysis of the various risk drivers 

of the worst 5% scenarios from the economic 

scenario generator 

• Heavy-tailed extrapolation of distributions 

deliver truly extreme scenarios 

• Scenarios consist of a combination 

of Profit & Loss (P&L) and balance sheet impacts 

Risk driver examples: 

- Aviation 

- Credit & Surety 

- Marine 

- Foreign exchange rates 

- Interest rates

III. Emerging Risk Management 

The second pillar of the ERM concept is Emerging Risk 

Management. It is important to note that, ten years 

from now, nobody would have been in a position to 

foresee the emerging risks that have actually emerged. 

The main objective of Emerging Risk Management is 

not so much to predict what the future risks are as to 

be well-prepared for their emergence. This objective 

could be summarized in three key words: 

• Identify potential emerging risks by using internal and 

external sources and central information gathering; 

• Assess the relevance of emerging risks by identifying 

affected areas, estimating economic impact and correlation 

with other risks; 

• Mitigate emerging risks by hedging/retrocession 

strategies, setting exposure limits, changing terms 

and conditions and securing access to liquidity 

(contingent capital, securitization). 

Therefore, insurance and reinsurance companies need 

to develop methods to anticipate the potential risks that 

could emerge, and above all to be prepared to mitigate 

the risks in question. In the past, Risk Management 

has led insurers and reinsurers to introduce certain 

preventive measures. Many safety regulations for 

industry and traffic have been inspired by insurers. 

SCOR combines a monitoring system with observers 

(“ECHO”) and tools for collecting information from 

numerous sources (e.g. “SCORWatch”) to track emerging 

risks. If a new trend or phenomenon is detected, it 

will be analyzed to determine if it could be an emerging 

risk. In the case of an emerging risk, an assessment and 

mitigation process is triggered. 

Within SCOR’s working group over 80 risks have been 

identified so far as possible emerging risks. 

Sample emerging risks 

identified at SCOR 

• Global pandemic 

• IT / Network risks 

• Climate change 

• Legal / Regulatory shift 

• Nuclear/biological/chemical terrorism 

• Technology: nanotechnologies, genetically 

modified organisms, chemicals 

• Product Liability 

• New chronic diseases 

• Mega projects 

• Socio-economic breakdown 

What are the options and the practical measures 

that (re)insurance companies could take when facing 

emerging risks? Here is a sample list: 

• Update business continuity plans 

• Explore possible reinsurance options to mitigate 

this risk 

• Include pre-defined clauses in insurance 

contracts 

• Strengthen underwriting guidelines 

• Ensure the existence of back-up systems 

• Constantly update the Natural Catastrophe 

models 

• Examine the possibility of opting out of certain 

covers/products/features where the risks are 

becoming too high 

• Assess the risk of parameter uncertainty for longtail 

business, review dependencies calculation 

• Encourage geographical diversification 

All of these measures form part of Emerging Risk 

Management. 


IV. Risk Control Processes 

The third pillar of the Greek temple is the Risk Control 

Process. This involves updating and monitoring the processes 

that have already been implemented. The risk 

control process is centered on three key words: 

• Identify 

• Assess 

• Respond 

As mentioned previously, a part of the Risk Culture 

is to establish risk-based internal control standards 

(including exposure limits). One of the objectives of the 

control process is to make sure that these standards 

are followed by all staff and by the company as a 

whole. Every risk is measured regularly and then the 

current exposure is compared to the limits. If need be, 

actions are recommended. SCOR also delivers regular 

risk reporting to the major decision makers, Chief 

Underwriters, the Executive Committee and the Board 

Risk Committee. 

Certain risks must be controlled and measured very 

frequently, particularly investment risk. For instance, 

a bank computes the Value-at-Risk daily and weekly. 

Fig. 6: Loss Event Management 

Generic Headline Loss Process 

Fostering transparency of numbers and process 

Newsworthy 

event occurs 

High level 

estimate of loss 

coordinated by 

Claims 

Claims trigger the estimation process 

based on significant expected loss 

Proposed guideline: > 1.5m 

It is different for an insurance company. Indeed, frequency 

control has to be adapted to the time horizon 

of the risk in question. 

One of the processes involved in risk control is Loss Event 

Management. When a major loss occurs, the question 

is how the company will manage the consequences of 

the incident. Figure 6 gives an illustration of Loss Event 

Management. 

Taking the recent Chilean earthquake, as an example, 

the first thing that the insurance company needs to 

do is to evaluate its potential exposure in Chile. If the 

first loss estimate is higher than a certain threshold, 

then the process of event management is initiated 

over one month. During this time, the company 

refines its estimation, and checks the underwriting 

of the contracts. So it monitors and reports to the 

management the total of the claims involved. Different 

parts of the organization will be involved in this process, 

e.g. natural disaster specialists, actuaries, underwriters 

and claims management units. For a month, a weekly 

estimation of the potential losses is computed and 

claims are reported contract by contract. After this initial 

phase, the monitoring enters a more standard portfolio 

review process for as long as is needed to assess the full 

extent of the loss. 

Evaluate Initiate 

Monitor 

Loss estimation process 

is triggered with UW. 

Weekly estimation + reporting 

of that specific loss event by contract 

Event moves into standard portfolio 

review process for Headline Large Losses 

Online reporting and updating of all HLL events. 

Monthly reporting to management as a component 

part of claims reporting process 

48 hours 1 st month Ongoing 

Clearly defined interfaces between claims, underwriting, technical accounting and actuarial 

Figure 7 gives an example of growth on a company’s 

netting plan. 

This chart illustrates how a (re)insurance company 

manages its exposure through the setting of limits and 

the use of reinsurance. The company plans a certain 

level of exposure growth, and then it implements a 

reinsurance policy so that the peak exposure is slightly 

smoothed out, the portfolio remains balanced and 


Headline loss 

is not actively 

monitored 

the diversification of the entire company is preserved. 

In certain regions the company was not able, for 

business reasons, to reduce its exposure and ended up 

with exposure that was too high; nevertheless, it was 

able to compensate it with hedging. Thus, the dark blue 

portfolio was finally more diversified than the light blue. 

This is an integral part of risk control and processes.

Fig. 7: Limit setting and enforcement 

Gross to Net Impact - Illustrative example 

Gross and Net distributions (including RI premiums): Single-event claims, 250-year return period 

Europe 

Windstorm 

Turkey 

Earthquake 

Switzerland 

Earthquake 

Gross Net of Retrocession / RIPs 

Japan 

Earthquake 

Reinsurance companies negotiate their Treaty P&C contracts 

mainly over two months in the year (November 

and December for Europe, February and March for 

Japan). During this period, it is highly important to 

monitor the exposures that underwriters propose to 

their customers. The curve in Figure 8 shows a sample 

contract exposure. 

The company determines its authorized capacity. It is 

possible to propose more than the authorized capacity 

because not all contract proposals will be taken up by 

Fig. 8: Risk Monitoring/Limit Control 

Renewal Process: Risk View 

Limit Setting 

(Net Appetite) 

Constant 

Feedback 

Loop 

Risk Control 

(Outwards options, 

additional limits, 

scale-down) 

Risk Reporting 

(Renewal Report) 

Japan 

Typhoon 

06.11 

USA 

Earthquake 

USA 

Hurricane 

Caribbean 

HU 

customers. There is, however, a point where the proposed 

capacities, even though they are not all accepted, 

will run a high risk of ending up above the set limits. 

This is the point where the Chief Risk Officer must intervene. 

In this example, the Executive Committee decided, 

following intervention by the CRO, to authorize more 

capacity to underwriters. At the end of the renewal 

period the bound contracts had an accumulated exposure 

lower than the initially authorised capacity, but the 

risk was too high for the CRO not to intervene. Once 

the contracts are bound, it is too late. 

Renewal Process: Monitoring 

14.11 

Illustrative example: 

Chief Risk Officer Intervention 

Estimated 

Exposure 

24.11 

01.12 

08.12 

Authorized 

capacity 

15.12 

20.12 

27.12 

03.01 


V. Strategic Risk Management 

Strategic Risk Management constitutes the “roof” 

of ERM. Strategic Risk Management ensures that the 

company’s return on capital (upside) objectives and 

risk (downside) constraints are reconciled. It can be 

decomposed into three parts: 

• Definition of a risk appetite framework (which is an 

integral part of the company’s strategic goals); 

• Development of various business strategies that 

enable the company to satisfy the objectives and 

constraints set out in the risk appetite framework; 

• Constant management of the risk profile to ensure it 

remains aligned with the risk appetite framework 

RISK APPETITE FRAMEWORK 

A company’s risk appetite framework should be defined 

by the company’s Board of Directors or equivalent 

supervisory body. The risk appetite framework should 

be systematically reviewed when a new Strategic Plan 

is developed. At SCOR, the review entails discussions 

in the Executive Management Committee and in 

the Board Risk Committee of the key implications of 

potential changes to the risk appetite framework. 

Recommendations are then made to the Board. 

In exceptional circumstances the Board may vary the 

amount and the composition of risk which the Group 

is prepared to take. 

The risk appetite framework encompasses three 

concepts: risk appetite, risk preferences and risk 

tolerances. 

Risk appetite 

Fig. 9: Framework to Strategic Risk Management 


Allocation between 

- Lines of Business (LoB) 

- Perils 

- Markets 

- Regions 

- Contract types 

- Clients 

Retro strategy 

Risk appetite defines the quantity of risk which a 

company wishes to accept to achieve a desired level 

of profitability. This will determine where the company 

wishes to position itself on the risk-return spectrum 

(between extremely risk averse i.e. low risk-low return 

and an extreme risk taker i.e. high risk-high return). 

SCOR uses a target retained risk profile (probability 

distribution of economic profits and losses) and target 

expected profitability to provide a complete definition 

of its risk appetite. A comparison of SCOR’s actual and 

target retained risk profile and profitability is regularly 

reported to the Board via the Board Risk Committee. 

Risk preferences 

Risk preferences are qualitative descriptions of the risks 

which the Group is willing to accept i.e. the type of 

risk the company wants to take: 

• Motor or Property on the P&C side? 

• Mortality, Accident & Health or Disability on the Life 

side? 

• Low uncertainty or high uncertainty business? 

i.e. working layers or remote catastrophe layers? 

• Geographical areas? E.g. Europe or Asia? 

SCOR wishes to cover a wide range of reinsurance risks 

from many geographical areas. SCOR has no desire to 

take operational, legal, regulatory, tax or reputation 

risks (but this does not mean that SCOR is immune to 

these risks). 

This choice will determine the risks to be included in 

SCOR’s underwriting guidelines. 

Risk appetite, risk preference, risk profile, risk limits 

Strategic liability 

risk-return-management 

Strategic asset and ALM 

risk-return-management 

- Strategic asset allocation 

- Hedging strategies 

- Duration (mis-)matching 

- Currency (mis-)matching

Risk tolerances 

Risk tolerances define the limits required by the company’s 

stakeholders (clients, shareholders, regulators, etc). 

SCOR defines two types of risk tolerances as follows: 

• The amount of retained exposure for each Line of 

Business (LoB) and asset class is limited to ensure 

that the contribution of each LoB or asset class to the 

95% TVaR of the annual Group economic profit/loss 

probability distribution does not exceed 5% (7.5% 

DEVELOPMENT OF BUSINESS STRATEGIES 

THAT SATISFY THE OBJECTIVES 

AND CONSTRAINTS SET OUT IN THE RISK 

APPETITE FRAMEWORK 

(i.e. risk-return optimization) 

Situations where optimized business strategies are 

developed include: 

• Definition of the strategic plan, which may include a 

possible revision of the risk appetite framework. 

In September 2010 SCOR launched its new three-year 

Strategic Plan “Strong Momentum” which replaced the 

previous plan “Dynamic Lift”. The ”Strong Momentum“ 

plan has three targets over the cycle: 

- Increase moderately (relative to the previous 

“Dynamic Lift” strategic plan) the target risk 

profile in the middle of the profit/loss probability 

distribution with no increase on tail risks (after 

hedging). 

for CAT business) of the Group’s available economic 

capital. These limits are intended to avoid concentration 

of risk in specific LoBs or asset classes and as such 

help to maintain a well diversified portfolio. 

• The amount of retained exposure of the Group (from 

one or several LoBs or asset classes) to very extreme 

scenarios (with a probability of 1 in 250 years) is limited 

to a loss of 15% of the Group’s available economic 

capital. These limits are designed to limit the impact of 

extreme scenarios on the shareholders’ capital. 

Fig. 10: Strategic Risk Management means basing business decisions 

on Risk/Return analysis 

Analysis 

of the 

renewal 

results 

Modeling 

of current 

portfolio 

- Risk appetite 

- Limit definitions 

for exposures 

- Pricing parameters 

- New product reviews 

Timeline 

January June November 

ALM 

Business planning 

Strategic 

targets 

(capital 

allocation) 

Business 

planning 

Modeling 

of the 

planned 

portfolio 

(risk budget) 

- Cat limits 

- Retro strategy 

- Capital consumption 

Strategic 

asset 

allocation 

(investments) 

Renewal 

pricing 

against 

the planned 

portfolio 

- Alignment of the asset 

portfolio to the plan 

- Strategic Asset Allocation 

- Allocation of extracapacity 

- Increase Return on Shareholders’ Equity (RoE) to 

1,000 bps above risk-free rate over the cycle. 

- Reach an “AA” level of security(1) 

for clients 

by 2013. 

• Definition of business units’ annual underwriting 

plans 

Capital management is firmly embedded in SCOR’s 

business planning processes. Figure 10 shows how the 

process of modeling risks is incorporated in SCOR’s 

P&C business unit’s planning process. After the P&C 

renewal period the results are analyzed and the portfolio 

is modeled. From this, business targets, new 

pricing parameters and products are defined for next 

year, and all this information is brought together in the 

P&C business plan. Capital allocations by LoB are also 

defined for next year. This allows SCOR to measure 

the performance of a LoB and to reduce or increase 

the exposure according to the potential results of that 

particular LoB. The business plan also includes Natural 

(1) This reflects the level of security provided by SCOR according to 

the S&P scale; however it does not reflect any Rating Agencies’ 

opinions of the Group. 


Catastrophe limits, a retrocession strategy and a total 

capital consumption concerning the next year. If there 

is any extra capital, a decision is made about what to 

do with it. The projected portfolio is then modeled, 

enabling SCOR to define asset allocations (within the 

boundaries of the strategic asset allocation) for the 

next year. After this, the renewal period begins. 

• Strategic asset allocation 

Strategic asset allocation determines the optimal 

asset portfolio mix that will protect the liabilities. The 

strategic asset allocation should be reflected in the 

company’s capital model and used to define limits 

and risk budgets. 

• Individual business decisions e.g. pricing, terms 

and conditions, product design 

• Mergers and acquisitions 

Taken together, the above actions lead to portfolio 

optimization at the company level which, in turn, 

facilitates the development of a risk management 

culture within the company. 

MANAGEMENT OF THE RISK PROFILE 

Using various risk monitoring and response mechanisms 

within its ERM Framework, SCOR’s Executive 

Management and the Board Risk Committee periodically 

review the Group’s actual risk profile and are 

systematically informed of any deviation from the target 

risk profile. In particular actual exposures are monitored 

against risk exposure limits for different Lines of 

Business, derived from the Group’s risk tolerances. 


Variations in the Group’s retained risk profile may 

occur due to a change in the underlying asset and 

liability portfolios and changes in the level of riskiness 

of certain parts of the asset and liabilities, e.g. due to 

rapid or slowly emerging changes in the economic, 

financial, social, legal and regulatory environments. 

Risk monitoring and response mechanisms enable 

companies to: 

• Take mitigating actions to ensure that the company’s 

retained exposure to specific risks remains 

within the defined exposure limits 

The first part of SCOR’s Capital Shield Policy – active 

hedging of peak exposures through retrocession and 

Insurance linked Securities (ILS) – is a key mitigation 

mechanism. The level of retrocession is selected to 

ensure that the Group’s retained risk profile respects 

the Group’s risk tolerances. 

• Take capital management actions to ensure that 

the Group’s capital base remains aligned to the 

risk appetite framework 

The second part of SCOR’s Capital Shield Policy – 

holding of Buffer Capital in addition to the required 

solvency capital – is a key capital management 

mechanism. The level of Buffer Capital is chosen to 

be consistent with SCOR’s risk appetite framework 

and strategic targets and enables SCOR to absorb 

a significant amount of inherent volatility in annual 

results, without having to turn to the market too 

frequently to raise capital.

Conclusion 

As a conclusion, it can be asserted that ERM is nothing 

less than sound insurance practice: 

• It allows (re)insurance companies to steer the 

business; 

• It enables them to define the value drivers of 

insurance; 

• It allows them to measure the performance of the 

business; 

• It encompasses the whole organization and its 

processes; 

• It makes the company more transparent to all 

stakeholders. 

Fig. 11: Strategic Asset Allocation (SAA) based on efficient frontier 

Risk versus Return (efficient frontier) 

Expected return 

Scenarios of equity allocations 

0% equity allocation 

Optimum equity allocation 

Downside risk (based on expected shortfall) 

HOW IS THE STRATEGIC ASSET ALLOCATION 

DETERMINED? 

Figure 11 shows a typical risk/reward curve, the x-axis 

representing the downside risk based on the expected 

shortfall, while the y-axis corresponds to the expected 

return of the strategy. The curve is computed by 

varying the proportion of equity in the investment 

portfolio. Both the risk and the reward are computed 

using the full asset and liability portfolio. The red 

point on the graph has 0% of equity. This means 

the company has only invested in fixed income. 

From 5%, it can build what is called the efficient 

frontier, which is basically the point at which it could 

obtain the maximum return for the minimum risk. 

This curve represents the set of optimal strategies 

This is why SCOR firmly believes that ERM does not 

simply represent a passing trend but is a concrete way 

for an insurance company to become more professional 

and more competitive. As with everything in life, there 

is a price to pay: ERM requires a long-term commitment 

to excellence from the entire organization. 

The image used in this article to understand the structure 

of the ERM concept is a Greek temple, therefore we will 

let a Famous Greek philosopher, Aristotle, conclude: 

“We are what we repeatedly do. Excellence, therefore, 

is not an act, but a habit.” 

The investment strategy is based on: 

• Risk/return considerations 

for the entire shareholder’s equity 

(including liability risk) 

• And risk aversion as defined 

by top management 

depending on the amount of equity in the investment 

portfolio. To determine the optimal asset allocation 

of the company, it is necessary to determine a target 

return. With this return it is easy to construct the 

corresponding straight line. The point where this 

line touches the curve is the point of the optimal 

asset allocation in light of the company’s profile. 

However, optimal asset allocation is not sufficient for 

a complete strategic asset allocation, because given 

the fluctuations of financial markets it is not possible 

to follow it constantly. It is needed to determine 

bands around which the allocation can move. For 

this, the company also needs to have a maximum 

and a minimum so as to limit the risks and to allow 

for tactical asset allocation within those boundaries. 


2 

ADAPTING SOLVENCY 

REGULATIONS TO 

TIMES OF CRISIS MICHEL DACOROGNA 


With the advent of the economic 

and financial crisis, the question of solvency is 

increasingly under discussion. This article addresses 

the current financial crisis and its consequences 

on solvency requirements and takes a fresh look 

at crises and their characteristics. It ultimately 

presents concrete proposals to make the regulatory 

system more flexible in terms of responding to 

future crises. 

I. Financial crises form part 

of the economic system 

The current developments in the financial industry 

are among the worst economic crises in history, as 

described in the diagram below. This shows a large 

sample created by a historian on the movements of the 

New York stock market from the 1800s to 2008: 

Fig. 12: The current developments in the financial industry line up amongst 

the worst economic crises in history 

S&P index annual returns from 1800 to 2008 

What is next? 

We are here! 

1965 9.1% 

2004 9.0% 

1959 8.5% 

1886 8.5% 

1845 8.1% 

1968 7.7% 

1830 7.4% 

1921 7.4% 

1871 7.3% 

1993 7.1% 

1802 6.8% 

1872 6.8% 

1899 6.5% 

1864 6.4% 

1878 6.1% 

1821 6.1% 

1926 5.7% 

1870 5.6% 

1840 5.5% 

1824 5.1% 

1832 4.8% 

1820 4.7% 

1806 4.6% 

1992 4.5% 1885 19.8% 

1807 4.4% 1852 19.6% 

1981 -9.7% 1856 4.4% 1999 19.5% 

1803 -9.6% 1866 3.6% 1943 19.4% 

1877 -9.4% 2007 3.5% 1976 19.1% 

1914 -8.6% 1889 3.5% 1898 18.9% 

1883 -8.5% 1812 3.5% 1963 18.9% 

1865 -8.5% 1916 3.4% 1924 18.8% 

1811 -7.5% 1906 3.1% 1850 18.7% 

1884 -18.8% 1819 -7.5% 1835 3.1% 1880 18.7% 

1903 -18.4% 1953 -6.6% 2005 3.0% 1891 17.6% 

1842 -18.1% 1887 -6.6% 1881 3.0% 1983 17.3% 

1876 -17.9% 1990 -6.6% 1831 3.0% 1951 16.5% 

1941 -17.9% 1934 -6.0% 1912 3.0% 1918 16.4% 

1973 -17.4% 1825 -5.8% 1874 2.8% 1901 15.7% 

1814 -16.7% 1939 -5.4% 1815 2.7% 1905 15.6% 

1940 -15.3% 1822 -4.8% 1956 2.6% 1972 15.6% 

1932 -15.2% 1805 -4.4% 1987 2.0% 1844 15.5% 

1846 -14.5% 1804 -4.3% 1892 1.8% 1982 14.8% 

1957 -14.3% 1875 -4.1% 1869 1.7% 1986 14.6% 

1913 -14.3% 1816 -3.8% 1838 1.6% 1858 14.3% 1915 29.0% 

1890 -13.5% 1848 -3.6% 1867 1.6% 1900 14.1% 1936 27.8% 

1841 -13.3% 1818 -3.2% 1855 1.5% 1909 14.1% 1989 27.3% 

1966 -13.1% 1851 -3.2% 1984 1.4% 1860 14.0% 1998 26.7% 

2001 -13.0% 1827 -3.1% 1902 1.3% 1919 14.0% 2003 26.4% 

1873 -12.7% 1960 -3.0% 1847 1.2% 1944 13.8% 1955 26.4% 

1853 -12.7% 1882 -2.9% 1813 1.1% 2006 13.6% 1985 26.3% 

1839 -12.3% 1894 -2.5% 1978 1.1% 1834 13.0% 1991 26.3% 

1910 -12.1% 1888 -2.5% 1809 1.1% 1964 13.0% 1980 25.8% 

1929 -11.9% 1896 -2.3% 1911 0.7% 1897 12.6% 1904 25.6% 1958 38.1% 

1946 -11.9% 1810 -2.1% 1895 0.5% 1942 12.4% 1938 25.3% 1863 38.0% 

1962 -11.8% 1861 -1.8% 1970 0.1% 1988 12.4% 2009 23.5% 1928 37.9% 

1937 -38.6% 1836 -11.7% 1994 -1.5% 1808 0.0% 1979 12.3% 1925 23.3% 1908 37.4% 

2008 -38.5% 1974 -29.7% 1977 -11.5% 1923 -1.5% 1823 0.0% 1817 11.9% 1961 23.1% 1995 34.1% 1933 46.6% 

1907 -33.2% 1930 -28.5% 1837 -11.5% 1829 -1.1% 1826 0.0% 1952 11.8% 1950 21.8% 1975 31.5% 1954 45.0% 

1857 -31.0% 1920 -24.5% 1969 -11.4% 1833 -0.9% 1828 0.0% 1971 10.8% 1922 20.9% 1997 31.0% 1843 45.0% 

1917 -30.6% 2002 -23.4% 1859 -10.7% 1948 -0.7% 1849 0.0% 1868 10.8% 1996 20.3% 1927 30.9% 1879 43% 

1931 -47% 1854 -30.2% 1893 -20% 2000 -10.1% 1801 -0.1% 1947 0.0% 1949 10.3% 1967 20.1% 1945 30.7% 1935 41.5ç 1862 55.4% 

To -50% To -40% To -30% To -20% To -10% 0-10% 10-20% 20-30% 30-40% 40-50% 50-60% 

Source: globalfinancialdata.com 


2009 showed 

a +24% return

The cells (Figure 12) contain the yearly returns of the 

stock market from -50% to +50% to 60%, and are 

ordered by size. The 2008 crisis, at -38.5%, is the third 

largest yearly negative return in 210 years. The year 

1937 comes second, differing from 2008 by only 0.1%. 

The worst year is 1931 at -47%. 

The internet bubble that burst in 2000 caused a 

decrease of 10% (-10%) at the end of the year. The 

stock market returns of 2000-2002 provoked a decrease 

of 23% (-23%), while the crisis of 2008 generated 

a drop of 38.5% (-38.5%). In 2009, stock market 

returns recorded an increase of 24%; nonetheless the 

same phenomenon occurred for the other crises. This 

temporary recovery should not be taken as the end of 

the turbulence created by the financial crisis. 

The financial crisis has caused previously unimaginable 

wealth losses: In May 2008, the hundred largest banks 

in the world lost USD 384 billion. Among them were 

Citigroup: USD 43 billion, UBS: USD 39 billion and Merrill 

Lynch: USD 37 billion. The Gross Domestic Product (GDP) 

of all OECD countries dropped by 2% in the last quarter 

of 2008 and by 2.1% in the first quarter of 2009. 

AIG lost USD 100 billion during the last quarter of 2008. 

This corresponds to USD 300,000 per minute, 24 hours 

a day, 7 days a week! These are amounts that defy the 

imagination. 

Crises are a part of human history, whether they are 

man-made or natural. History shows that mankind 

has been exposed to many natural catastrophes and 

man-made crises, looking back over just two centuries: 

• Since 1825 the volcano Vesuvius has had 9 periods 

of eruption (1) . During the twentieth century volcanic 

eruptions caused fewer than 1,000 deaths per year. 

However, two events caused more than half the deaths 

recorded in the last century: 29,000 killed for the 1902 

eruption of Mont Pelée (on the island of Martinique) 

and 23,000 killed for the 1985 eruption of Nevado del 

Ruiz (in Colombia) (2) . More recently, the 2004 Asian 

Tsunami claimed more than 170,000 lives (3) . 

• Over this period, the S&P 500 index has lost 9 times 

more than 20% in a year. When looking at those 

facts, one is almost tempted to think that crisis is 

more the norm than the exception. 

Many crises have already been recorded; it is very likely 

that there are many more to come. Even if it is impossible 

to foresee such catastrophes, the idea is to be prepared 

and to adapt the financial system to face up to crises with 

as little damage as possible. This means that companies 

and regulators have to change their general mindset. 

The focus should be much more on how companies can 

survive these crises unharmed than on trying to predict or 

avoid crises. Avoiding crises is probably an unachievable 

goal, so let us see how to survive them instead. 

WHAT IS BEHIND THE CURRENT SEVERE 

CRISIS? WHAT ARE ITS CHARACTERISTICS? 

Existing measures have shown their limits in terms of 

preventing severe crises from recurring and preventing 

risk from asphyxiating the development of the financial 

system through excessive capital requirements and 

deleveraging. It is thus very important to analyze the 

current crisis and to come up with new ideas for how 

to handle such exceptional situations. 

Three characteristics come out of any financial crisis: 

• A crisis, almost by definition, comes as a surprise 

because it reveals links between risks that were 

previously neglected; 

• The risk appetite of market actors is irresponsibly 

high, and accompanied by high financial leverage; 

• There is an excessive concentration of aggregate risk 

in highly leveraged financial institutions. 

(1) Source Volcano Discovery. 

(2) Environmental Hazards: Assessing Risk and Reducing Disaster. 

(3) World Bank. 


The surprises that have the potential to trigger severe 

financial crises are not simply bad realizations within 

a known probabilistic environment. Rather, they are 

changes in the environment itself. It is this “rare event” 

feature that holds the key as it has the potential to 

trigger a sharp rise in perceived uncertainty and flight 

to quality. An example of such a surprise took place 

when the Reserve Primary Fund, a leading money market 

fund, did not reimburse at par after Lehman declared 

bankruptcy (breaking the buck). The authority thought 

that because Lehman did not have a very high Credit 

Default Swap (CDS) exposure and the CDS exposure 

of the Reserve Primary Fund was so high, its failure 

would not catch the market by surprise. Unfortunately, 

the Primary Fund had invested 1.2% of its assets in 

Lehman’s debt. 

Immediately after Lehman filed for bankruptcy, the 

fund suffered a massive run, with over USD 30 billion 

of redemption requests (about half of its total assets), 

before it stopped accepting redemption requests the 

following day. Money market funds had been considered 

extremely safe, and had indeed benefited from the 

flight to quality the previous year, growing by about 

USD 850 billion (34%) since mid-2007. The drop in the 

Reserve Primary Fund’s NAV caused investors to question 

the safety of the entire industry and the first under-par 

redemption from a Money Market fund since their 

creation in 1968. That week, there were USD 169 billion 

in redemptions from total investments of USD 3.4 trillion 

(5%), as well as a major shift from prime funds towards 

funds investing exclusively in government debt. 

In retrospect, the consequences of Lehman’s demise 

on the Primary Fund could have been predicted: public 

filings showed large investments in Lehman as early 

as November 2007. Anyone who took the trouble to 

connect the dots could, in principle, have foreseen 

what might happen. However, money market funds 

had a track record of stability that had always made it 

unnecessary to inspect their holdings. The realization 

that there might be further losses in previously 

unexamined places led investors to intensify their flight 

to quality. The main failure was in not understanding 

that relatively “small effects” could generate huge 

impacts and create confusion for the entire system. 

Surprises quickly trigger a chain of unexpected events 

from the panic they engender. In times of crisis small 

effects can generate huge impacts. 


The second important trigger for a severe crisis is the 

highly leveraged and interconnected sector of the 

economy, generally the financial sector, being exposed 

(directly or indirectly) to a surprise of the kind discussed 

earlier. Aggregate risks are those exposed to aggregate 

shocks impacting the entire economy. Investment in 

structured products has exposed financial institutions 

to more aggregate risk and surprises than in the past. 

In the current crisis, banks were holding mostly AAAtranches 

of a large variety of new ABS (Asset Backed 

Securities) (85% of assets held in securitized form). 

These tranches rely on protection from junior tranches 

and from the law of large numbers in order to reduce 

the risk of default enough to achieve an AAA-rating. 

The law of large numbers implies that the loss on a pool 

with a sufficient number of underlying assets, as was the 

case with most ABS, can only occur when an aggregate 

shock takes place. Furthermore, the higher up a given 

structure is situated, the larger the aggregated shock 

must be for it to pierce the protection offered by the junior 

tranches. Losses large enough to affect the AAA-tranche 

only occur in situations of severe aggregate shocks, 

but this is exactly what large surprises do! Therefore 

holdings of AAA-tranches of structured products 

exposed financial institutions to more systemic risks 

than their rating, when misinterpreted, would suggest, 

and certainly more than similarly rated “single name” 

corporate bonds. Corporate bonds are still affected by 

macroeconomic conditions but idiosyncratic factors play 

a larger role. Downgraded structured financial securities 

lost on average between 5 and 6 notches in the period 

2007/2008. By comparison, during the great corporate 

bond downgrade of 2001/2002 (30% of corporate 

bonds were downgraded) the average notch-loss 

was 1.8. The systemic consequence of this risk was 

that highly leveraged institutions were bearing more 

aggregate risks than would have been thought from 

simply observing the average ratings of their assets. 

Having the financial sector of the economy holding such 

risk with respect to aggregate surprise proved to be a 

recipe for disaster. 

Surprise and aggregate risks are major causes for 

financial crises. Neither element is likely to disappear 

soon. One can thus expect that the economy, despite 

all attempts to protect it, will suffer shocks in the future. 

The correct risk management policy is to prepare the 

organization for the occurrence of such shocks and to 

make sure that the company can survive them. Rather 

than trying to predict the next crisis or to avoid it, 

(re)insurance companies should concentrate on making 

sure that their risk management model integrates the 

occurrence of crisis with a reasonable probability. Only 

in this way will organizations be prepared to face major 

disruptions in the financial markets.

II. Consequences of financial 

crises on insurance capital 

requirements 

SCOR was actually very prudent during the crisis, building 

a huge cash position. Retrospectively, this proved to be 

the right thing to do, especially with a good model. 

However, this policy drew questions from the regulators 

because it departed from the usual ALM strategies. It is 

thus useful to look at the impact of the crisis on SCOR’s 

internal model to understand why the company acted 

like this and to see what a good alternative could be in 

terms of regulation during times of crisis. 

THE MAIN CONSEQUENCES OF THE CRISIS 

ON SCOR’S INTERNAL MODEL 

Good risk models would show an increased risk of the 

situation and thus come up with higher risk adjusted 

capital than in quieter times. This is natural because 

the model reflects inherent uncertainty, which naturally 

increases during turbulent times. Requiring companies 

to keep the same level of security as before the crisis 

would require a significant increases in capital. The 

following example examines the consequences of the 

crisis on SCOR’s internal model. 

Some consequences are directly linked to the fall of 

interest rates: 

Fig. 13: Risk-Free Rate development 

Historical 10Y Risk-Free Rates 

in % 

10 

8 

6 

4 

2 

Aug. 87 

Aug. 89 Aug. 91 Aug. 93 

10Y US T-bond 

• The life model required more capital for mortality 

due to the fall in interest rates. This translated to 

more capital for the whole SCOR portfolio. It is easy 

to understand that a liability payable in ten years 

appears bigger if the interest rate is low. 

• At the same time, SCOR’s available capital experienced 

a marked decrease. Using the same yield curve as 

before the crisis would have resulted in an increase in 

the available capital. Due to the fall in the discounting 

benefit of P&C reserves, the available capital looks 

smaller, even though aside from the interest rate 

nothing has changed in the portfolio. 

Other consequences are linked to the increased volatility 

of financial markets and increased credit risk: 

• During the crisis the volatility of stock returns more 

than doubled. 

• Credit spread skyrocketed for reinsurance to more 

than 2,000 basis points. 

Those cumulated effects significantly reduce the 

solvency ratio of companies. SCOR has seen a significant 

drop in its solvency ratio even though there were no 

significant changes in its liability portfolio and its asset 

portfolio was significantly derisked. Luckily SCOR was 

well capitalized; otherwise the company would have 

had to go to the financial markets to get more cash 

at a time of liquidity squeeze, thus accentuating the 

lack of liquidity. 

Aug. 95 Aug. 97 Aug. 99 Aug. 01 Aug. 03 Aug. 05 Aug. 07 Aug. 09 

10Y German Bund 

10Y French OAT 

10Y US T-bond: 3.68x 

10Y French OAT: 3.502x 

10Y German Bund: 3.26x 

Source: Reuters as of 22 Jan. 2010. 


OTHER ECONOMIC INDICATORS 

SUCH AS INFLATION MAY BE TAKEN 

INTO CONSIDERATION 

Figure 14, which comes from an Economic Scenario 

Generator during the crisis (last quarter of 2008), shows 

the importance of such an advanced indicator. This is 

an input in SCOR’s model. It refers to the distribution of 

inflation. The blue line corresponds to the distribution 

of the historical inflation index. The brown one is the 

forecast distribution. The chart indicates the change 

in the distribution. Considering that the interest rate 

and the inflation will decrease, the distribution has 

shifted. Nonetheless, the most important thing for 

the capital is the widening of the distribution. SCOR’s 

model forecast a higher uncertainty on the inflation. 

The higher uncertainty on inflation required a capital 

increase because of the uncertainty of the outcome. 

Fig. 14: The ESG reflects the current 

uncertainty on inflation 

Empirical density function of 1 year returns of CPI: USD 

15 

10 

5 

0 

-0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 

Historical data: annualized quarterly returns 

Simulated data: yearly returns 

Data: 

• Number of historical data: 83 (relatively small) 

• Number of simulations: 60,000 

The empirical distribution of the simulated inflation is: 

- Out of phase because the current interest rates are lower, 

- Wider because the current volatility is bigger (GARCH effect). 


APPLYING SOLVENCY RULES AFTER 

A CRISIS IS COUNTERPRODUCTIVE 

When a crisis occurs, regulators insist on respecting 

solvency requirements, but: 

• The economic situation is deteriorating, which could 

force companies to declare insolvency for claims they 

would have to pay far in the future; 

• By requiring a capital increase from insurance and 

reinsurance companies, regulators force companies 

to de-leverage their balance sheet, which would fuel 

the crisis. 

This type of action would immobilize a huge amount 

of additional capital, which increases the cost of 

the protections provided by insurers and reinsurers. 

Furthermore, it would dry out the capital available 

for the rest of the economy, further weakening nonfinancial 

companies, which would, therefore, reduce the 

quality of the asset portfolio of insurers and reinsurers, 

reinforcing the vicious circle. 

REGULATORS AND COMPANIES COME UP 

WITH AD HOC CHANGES IN THE RULES 

Several solutions were proposed to escape from the 

dilemma of pro-cyclicality: 

• Use of a swap rate to create liquidity premium, which 

allows companies to account for liquidity risks; 

• Give companies negative capital to compensate for 

the capital increase; 

• Suspend the solvency rules based on risk models to 

enable companies to survive. 

Such proposals could cast doubts on the validity and 

the usefulness of modeling the risks of the company. In 

fact, such measures contradict the risk-based regulation 

principles. On the one hand, such tricks are dangerous. 

On the other hand, the system has to develop a certain 

flexibility. It is worth noting that the parameters of 

models are very sensitive. So if companies start changing 

the parameters, they can obtain any result they want, 

and this is not what they should be doing. They should 

not act as if there were no crisis, because then their 

model will defeat its purpose, which is to look at the 

risk. On the contrary, they have to accept that there is 

more risk. They should accept this and they should live 

through it. After all, if companies have capital, it is to be 

used in times of crisis.

III. Giving the regulatory system 

sufficient flexibility 

to respond to the next crisis 

The system must have a certain amount of flexibility. 

In such confusing situations simple rules are necessary, 

simple rules that do not cast doubts on the outcome 

of the models. 

RECOGNIZING THE RISKINESS 

OF THE SITUATION 

The regulators’ answers to the crisis are to propose an 

indiscriminate increase of the target capital for insurance 

companies. Such requirements, if they go through, would 

have severe social consequences by drying up even more 

market liquidity and increasing the price of insurance. 

• What is required in such a context is simply the 

recognition that the situation is riskier. 

• Capital requirements should be adaptive and should 

change according to the economic situation. 

Moreover, empirical studies have shown that a measure 

like Value-at-Risk ex-post (1) is lower during crises 

because the market corrects itself and the probability of 

rebounds increases after a big downward movement. 

CHANGING THE RISK MEASURE THRESHOLD 

Solvency II requires that the capital be computed using 

the Value-at-Risk at a threshold of 99.5%. This threshold 

has been chosen arbitrarily to be in the tail of the 

distribution. There is no objective reason to take 99.5% 

instead of 99% or 98%. This threshold does not need 

to be constant all the time. 

120% 

100% 

80% 

60% 

40% 

20% 

0% 

28.02.1873 

Yearly volatility (monthly data) 

120% 

100% 

28.02.1877 

80% 

60% 

40% 

20% 

28.02.1881 

Return and volatility of US stock market (1870-2009) 

0% 

-150% -100% -50% 0% 50% 

YoY return 

28.02.1885 

28.02.1889 

Source: IMF (2009). 

28.02.1893 

28.02.1897 

28.02.1901 

28.02.1905 

26.02.1909 

28.02.1913 

28.02.1917 

100% 

28.02.1921 

150% 

27.02.1925 

28.02.1929 

28.02.1933 

26.02.1937 

28.02.1941 

• Moving from 99.5% to 99% is a simple and transparent 

way to recognize the reality of the economic 

situation: nobody can be so safe any more when the 

whole world is in turmoil. 

• Furthermore, this change will mainly compensate 

the increase in cost of capital due to the model’s 

recognition of the higher risks. 

In SCOR’s model, it would represent a decrease of roughly 

10% of the required capital, compensating the increase 

due to higher market volatility and lower interest rates. 

The supervisory authorities should set a threshold 

sufficiently remote to inspire confidence in the system 

by all stakeholders, within a predefined range, in order 

to react quickly if an indicator shows an economic 

degradation close to a pre-crisis. 

WHICH INDICATOR WOULD ENABLE 

AN OBJECTIVE ASSESSMENT 

OF THE CRISIS SITUATION? 

The volatility of stock indices is a relevant indicator to 

detect a crisis situation. Market volatility affects the risks 

generated by the Economic Scenario Generator (ESG) and 

thus ends up requiring higher capital from the model. 

Figure 15 presents the yearly volatility of the stock index 

since 1870. The volatility is more correlated with negative 

return than positive return. When a crisis occurs, 

the volatility increases faster than in periods of calm 

and the market reaches twice the average volatility of 

roughly 40%. This situation has only happened twice, 

in 1929 (120%) and in 2008 (80%). 

(1) VaR ex-post means the Value-at-Risk measured at a certain point 

in the past. 

Fig. 15: Market volatility has reached such values only twice in the past 140 years 

Yearly volatility of US stock market since 1870 (monthly data) 

28.02.1945 

28.02.1949 

27.02.1953 

28.02.1957 

28.02.1961 

26.02.1965 

28.02.1969 

28.02.1973 

28.02.1977 

27.02.1981 

28.02.1985 

28.02.1989 

26.02.1993 

28.02.1997 

28.02.2001 

28.02.2005 

27.02.2009 


A SIMPLE RULE TO MANAGE 

A CRISIS SITUATION 

This article proposes a simple rule to manage a crisis 

situation. Regulators should declare that companies 

would be allowed to use a Value-at-Risk (VaR) at 99% 

for the next year if the stock index yearly volatility 

reaches twice its historical average measured over a 

very long period. 

A year later, if the volatility is below this index, the 

regulators would then restore the 99.5% threshold 

and ask companies to refurbish their capital to comply 

with it. 

Such a rule would allow insurance and reinsurance 

companies to use part of their capital to face up to 

the tough economic situation without running the risk 

of becoming insolvent for liabilities they would have to 

pay in the distant future. 

Such a flexible system combines three advantages: 

1. It solves the dilemma of pro-cyclicality; 

2. It reduces the need to lock up useless extra capital; 

3. It enables companies to analyze a crisis situation with 

transparency and objectivity. 


Conclusion 

Crises are unforeseeable and reveal links that were 

previously underestimated. Reinsurance and insurance 

companies undoubtedly need to learn from all crises in 

order to reduce the risk of the same causes producing 

the same effects. 

It is essential to adapt the solvency regulations to the 

occurrence of crises and their seriousness with simple 

adaptive rules, such as modifying the threshold of the 

risk measure. Indeed, this is a simple way to restore 

confidence in the system because it recognizes the 

objective situation and gives companies time to adapt 

and refurbish their solvency ratio. 

It is worth utilizing an economic indicator such as 

the extreme volatility of financial markets because it 

avoids blaming any stakeholders for the decision and 

highlights the objective situation. 

Being prepared for the next crisis means integrating 

into the models a reasonable probability that a crisis will 

happen and developing dynamic strategies to survive it. 

The Romans used to say: “si vis pacem para bellum” (if you 

want peace prepare for war). One will paraphrase them by 

saying: “if you want to survive a crisis prepare for it”.


3 

REINSURANCE OPTIMIZATION 

IN THE CONTEXT OF CAPITAL 

MANAGEMENT EVA SCHLÄPFER DE MONTMOLLIN 

Senior Risk Consultant, SCOR 

MAGDALENA KLAPPER-RYBICKA 

Risk Consultant, SCOR 

This article outlines the different 

motivations an insurance company may have for 

purchasing reinsurance. It then introduces the 

capital management view, as a driver for structuring 

a reinsurance program. From this perspective, 

specific reinsurance valuation criteria are investigated 

in detail using an example portfolio. 

I. Various criteria for the 

valuation of a reinsurance 

program 

Insurance companies may have different reasons for 

buying reinsurance: 

• To enhance profit and portfolio management, to 

decrease the volatility of the portfolio, to increase 

diversification benefits, to protect profitability. 

• To support the company’s capital management strategy 

in order to achieve the solvency capital requested 

by the regulators, to satisfy the capital requirements 

of the rating agencies, or simply to provide better 

capacity. 

• The reinsurance purchase may also be cost driven: 

insurers buy reinsurance if expected recoveries exceed 

the reinsurance premium. 

• Various other reasons, such as: to continue the 

previous year’s reinsurance program; to benefit from 

the reinsurers’ know-how or services (e.g. product 

development, medical underwriting, etc.). 


The purchase of reinsurance, therefore, may also be 

considered as a risk management tool that enables 

insurers to lower their capital requirements in the short 

term. Insurers also have other means at their disposal 

to mitigate their overall business risk, such as: changing 

their investment strategy, raising capital and changing 

their underwriting policy. However, these methods take 

longer to implement. 

Figure 16: Risk Management 

for Insurance Companies 

Risk mitigation instruments 

Short-term 

Long-term 

Buy reinsurance 

Change investment strategy 

Raise capital 

Change underwriting policy 

Different possibilities for mitigating overall business risks 

according to the time needed for their application and effect. 

The various motivations for buying reinsurance determine 

the criteria to be applied to the evaluation and 

comparison of different reinsurance programs. 

Depending on the motivation involved, the best 

reinsurance program could be any of the following: 

• The program offering the best profit protection, 

the lowest volatility, the best diversification benefit or 

the highest profit for insurer and reinsurer, taking the 

cost of capital into account (all of which corresponds

to the portfolio management and profit optimization 

perspective). 

• The program with the lowest required capital resulting 

from the company’s internal model, the lowest Solvency 

II standard model capital requirement or the lowest 

rating agency capital (if the capital management view 

is taken). 

• The program with the lowest cost of reinsurance - 

premium minus recovery (cost driven policy). 

Each criterion defined for the evaluation of a reinsurance 

program (independent of its nature) should definitely 

incorporate the costs of the reinsurance program, in 

order to produce a meaningful and realistic result. In a 

further step this article will focus on reinsurance valuation 

as a part of capital management practices. 

II. Valuation of a reinsurance 

program in the context 

of capital management 

In order to look at reinsurance valuation from a capital 

management perspective, this section will start with 

some general definitions. 

Fig. 17: The capital of an insurance company 

Economically adjusted capital: 

Available Capital 

Discount in loss 

reserves 

taxes 

Unrealized 

capital gains Latent 

Capital as reported 

in financial statement 

An insurance company receives capital from its 

shareholders and utilizes that capital to guarantee 

that it will pay claims – up to a certain probability. The 

primary focus of capital is, then, not to provide finance, 

but to absorb the risks undertaken. Capital can thus be 

seen as a “commodity” used to produce the company’s 

business. 

It is necessary to distinguish between two definitions 

of capital: available capital and risk-based capital (or 

required capital). 

Available capital or economic capital is the capital a 

company has at its disposal (equity, long-term debts, 

unrealized gains, discount in loss reserves, latent taxes, 

etc). Conversely, the risk-based capital or required capital 

is the minimum amount of capital that the insurer 

needs to cover the risk taken. 

The difference between the two is called the “signaling 

capital” and corresponds to the economic resources 

available to develop more business or take on more 

risks. As illustrated in Figure 17, the amount of reinsurance 

involved has an impact on the balance between 

required and signaling capital. It also has a slight impact 

on the available capital, due to the cost of reinsurance. 

This can be seen on the left-hand side of the graph. 

Capital required given management’s 

risk appetite framework: Required Capital 

Signaling Capital 

Required Capital 

for underwriting risk 


for investment risks 


for other risks 

More reinsurance 

Less reinsurance 


The amount of required capital is determined by the 

insurance company’s risk appetite, in other words the 

amount of capital the company is willing to risk in a 

certain time horizon. Other elements are the proper 

definition of required capital and a capital allocation 

policy. These are prerequisites for the optimization of 

the portfolio’s profitability, because they build the foundations 

on which to measure the true performance of 

the business. Hence they have to be incorporated at the 

core of business processes. 

Further discussion of risk allocation would be outside 

the scope of this article, however the definition of 

required capital will be addressed in more detail. 

There are many different definitions of required capital. 

Clear definitions should ensure consistent use of terminology 

within the organization. The definition of required 

capital may be developed based on various mathematical 

concepts and should address the risk aversion of the company 

in question. SCOR has its own definition of required 

capital as used in its internal model. 

Let us begin with standalone risk capital. The asset and 

liability portfolio builds a set of risk classes: {Xi}, e.g. 

bonds, equities, Nat Cat, Motor, etc. The standalone 

(or undiversified) risk capital of a class is the difference 

between two values on the probability distribution of 

annual economic profit. These two values are: 

• The expected shortfall at 1% quantile of the distribution 

• The expected value of the distribution 

E(X i | X i < 1% percentile) – E(X i ) 

This definition is one of several possible definitions. 

It expresses the shareholders’ point of view, since it 

measures how far the company deviates from expected 

profit (see Figure 18). By way of comparison, the Swiss 

Solvency Test definition takes zero profit as its reference 

point. Thus, regulators represent the clients’ view and 

therefore seek to know whether or not a company can 

pay its obligations. 

Solvency II provides another definition of required capital. 

The directive applies the Value-at-Risk at the 0.5% 

quantile. This risk measure corresponds to a point on 

profit distribution (i.e. not the expected shortfall) and 

therefore does not incorporate the risk profile below 

the chosen threshold. One advantage of this is that it is 

easier to compute analytically, with no need to apply a 

Monte Carlo simulation. 

SCOR has based its definition of required capital on the 

expected shortfall at the 1% quantile. It can be demonstrated 

empirically that required capital measured as 

Value-at-Risk at 0.5% is close to required capital defined 

with expected shortfall at the 1% quantile, so that the 

risk capital as measured by SCOR is in line with the 

Solvency II regulatory view. The risk capital definition 

used by SCOR is coherent and its resulting mathematical 

features (additivity) allow the required capital to be allocated 

to the single risk classes. This leads to the definition 

of diversified capital. 

Figure 19 provides a closer look at the tail end of the 

profit distribution already presented in Figure 18. Since 

the profit distribution is the result of a Monte Carlo 

simulation, it is possible to analyze the single worst-case 

scenarios, and in particular the contributions to the 

scenarios of the various different risk classes. 

Diversified Risk Capital is defined as a contribution 

(measured in terms of average) of a class of risk to the 

expected shortfall of the company overall profitability 

loss distribution (Z): 

E(X i | Z < 1% percentile) – E(X i ) 

Diversified Capital is the capital amount that SCOR 

needs to keep for a certain risk class in a given portfolio, 

in order to be able to pay its obligations. 

Fig. 18: Definition of standalone capital as the difference between 

mean expected value and 1% expected shortfall 


Standalone Risk Capital 

1% ES 100% Expected value 

1% VaR 

10% 

Profit Distribution 

0% 

-220 -170 -120 -70 -20 30 80 

90% 

80% 

70% 

60% 

50% 

40% 

30% 

20%

Fig. 19: Diversified Risk Capital 

-250 -200 -150 -100 -50 0 50 

With this definition, the main risk contributors can be 

identified as shown in Figure 19. The total required 

capital for the portfolio is around EUR -90 million and 

can be divided into four risk classes. The highest risk 

comes from classes 4 and 1, with diversified risk capital 

of EUR -44 million and EUR -26 million respectively. 

The contribution for the other risks is lower. 

It is interesting to note that some risks in certain scenarios 

show a positive contribution to the overall risk. 

This may occur due to the low dependency of these 

risks on other risk classes. 

The example also demonstrates that the contribution 

of some risk classes may vary widely for different 

Diversified 

Risk Capital 

Definition of diversified risk capital as a contribution of a single risk class to the aggregated shortfall. Due to the graphical limitations, 

the diagram only contains around 20-30 worst-case scenarios, (not the whole 1%). The blue curve corresponds to the profit distribution. 

For each scenario the total risk may be split into certain risk classes. The mean is therefore computed separately for each risk class. 

Fig. 20: Risk drivers 

Risk 1 




Total 

scenarios. To better understand the contribution made 

by risk classes to scenarios of differing frequency, it is 

necessary to make a distinction between risk capital 

and risk drivers. 

Risk drivers are measured based on a 5% shortfall, 

rather than the 1% shortfall used for risk capital: 

E(X i | Z < 5% percentile) – E(X i ) 

They reflect more frequent risks, while risk capital 

focuses more on peak or extreme risks. The contribution 

made by different risk drivers may be quite different 

from their contribution to the required capital, as 

illustrated in Figure 20. 

0% 

-220 -170 -120 -70 -20 30 80 

Risk 1 Risk 2 

Risk 3 Risk 4 

1% ES 5% ES 100% Mean 

80% 

60% 

40% 

20% 

Risk Driver 


Risk driver and risk capital reflect risks of differing frequency. Various risk classes may contribute differently to more or less frequent risks. 

For example, risk class 4 (in pastel blue) contributes much more to the peak risk than to the risk driver, while risk class 2 (in dark blue) 

does the opposite. The same thing can also be seen in Figure 18. 

-26 

-9 

-7 

-44 

-86 


Having defined the risk measures and how to allocate 

risk capital to particular risk classes, one can now turn 

to the criteria for evaluating a reinsurance program. 

Reinsurance is a way to reduce required capital, and 

the valuation criterion proposed accounts for the costs 

of reinsurance including the cost of capital. 

How is a reinsurance program evaluated in this context? 

On the one hand, the insurance company purchasing a 

reinsurance program pays a corresponding reinsurance 

premium and expects some recoveries and related costs. 

On the other hand, a reinsurance program is a way to 

reduce capital requirements, which consequently also 

lowers the cost of the capital that the company needs 

to keep in order to pay its obligations. Therefore, this 

criterion compares the cost of reinsurance to the cost 

of capital to be kept for the portfolio. 

Fig. 21: Reinsurance valuation 

criteria used 

Cost of 

capital on 

required capital 

Cost of Reinsurance (RI): 

RI premium 

- RI Recoveries 

III. Example of two reinsurance 

structures 

This section will continue to look at valuation criteria 

using a practical example. 

Figure 22 shows a realistic example portfolio which is 

characterized by a large P&C book, mainly in Property, 

Casualty and Motor. The total required capital is around 

EUR 95 million, primarily relating to the P&C new 

business book. Assets are mostly invested in bonds 

and real estate and they do not constitute a very highrisk 

contributor. 

BEFORE APPLYING THE REINSURANCE 

STRUCTURE 

Let us briefly analyze the gross performance of the 

main risk contributor, the P&C new business book (see 

Figure 23). The standalone required capital for P&C 

new business amounts to EUR 92.9 million. Diversified 

with the whole portfolio, this risk class requires EUR 

52.4 million, giving a diversification benefit of 43.5%. 

The insurance company needs capital to keep this 

book, which involves some cost. Assuming that the 

cost of capital rate is 11.5%, the cost of the capital 

needed amounts to EUR 6.03 million. This cost should 

be reflected in the assessment of the P&C new business 

profit, which amounts to EUR 9.7 million. 

Fig. 22: Example portfolio – gross P&C book with main risk categories 

and composition of inverted assets 

Risk Category 

in EURm 


Reinsurance valuation 

criteria optimizes capital 

while also taking into 

account the pure cost 

of reinsurance. 

Exposure Fully Diversified Capital Exposure definition 

Gross Non-Life New Business 210 52.4 Written Premium 

Gross Non-Life Reserves 240 24.0 Claims Reserves+UPR 

Invested Assets 83 10.1 Market Value 

Credit 107 5.4 Available Capital 

Other Risks 107 0.9 Available Capital 

Operational Risk 107 1.5 Available Capital 

Total RBC 94.3 

Asset Class Market Value Fully Diversified Capital 

Equity 13 2.86 

Real Estate 40 6.06 

Fixed Income 29 1.18 

Total Assets 82 10.1

Fig. 23: P&C new business book – gross required capital, 

related cost of capital and profit (in €m) 

P&C NB 

Gross 

Standalone 


92.9 

P&C NB Gross Profit 

excluding CoC 

15.8 

P&C NB 

Diversification 

Benefit: 43.5% 

P&C NB Gross Profit 

including CoC 

9.7 

To gain a concrete view of the impact of reinsurance on 

diversification and capital requirements, a reinsurance 

program is applied to the P&C new business book with 

a structure identical to that shown in Figure 24. 

As already mentioned, this P&C new business book 

consists mostly of Casualty, Motor and Property lines; 

it therefore requires a specific reinsurance structure. 

The program is characterized primarily by a quota share 

contract for the Casualty and Motor lines, which is 

followed by XL cover. The Property line is protected with 

the 3-layer cover; the 2-layer Cat program completes the 

reinsurance structure. 

APPLYING THE REINSURANCE STRUCTURE 

TO THE P&C NEW BUSINESS PORTFOLIO 

An analysis of the net portfolio gives the following 

picture: net standalone required capital for the P&C 

P&C NB 

Gross 

Diversified 


52.4 

Total Gross Capital = 94.3 

Costs of Capital 

allocated to P&C NB 

6.03* 

Operational Risk 

Other Risks 

Credit 

Invested Assets 

Gross Non-Life Reserves 

Gross Non-Life New Business 

* CoC Rate = 11.5% 

new business book has obviously decreased due to the 

reinsurance structure and amounts to EUR 54.8 million. 

The net profit without considering the cost of capital is 

of course lower than the gross profit, since in this view 

reinsurance naturally appears as an additional cost. 

Within the overall portfolio, the net required capital for 

this risk class amounts to EUR 23.9 million, generating a 

better (compared to the gross portfolio) diversification 

effect equal to 56%. The overall portfolio is better 

balanced with this reinsurance structure. The cost 

of capital needed for the P&C new business book 

is now also reduced due to the lower amount of 

diversified capital, and amounts to EUR 2.8 million. 

With adjustment for cost of capital, the net profit of 

this book is slightly over EUR 10 million (see Figure 25 

for details). 

Fig. 24: P&C new business book and current reinsurance program (in €m) 

P&C Line Gross Premium RI Structure 

Casualty 70 QS 

Per Risk XL2: 60 xs 60 

Motor 50 

25% 


Property Cat 10 

Total Property 80 

Total P&C 210 

CAT XL2: 60 xs 40 

CAT XL1: 30 xs 10 




Portfolio 

characteristics: 

• Quota share before 

XL for Casualty and 

Motor combined 

• Property covers 

are XL 


Fig. 25: P&C new business book – net required capital, 

related cost of capital and profit (in €m) 

The question that arises now is whether or not this program 

is efficient due to the chosen valuation criteria. 

As mentioned, the application of the reinsurance 

program lowered the P&C new business book’s profit 

by around EUR 2.5 million (due to the costs of 

reinsurance). On the other hand the mitigation of the 

peak risks, which resulted in a lower standalone capital 

as well as in improved diversification with the rest of 

the portfolio, lowered the cost of the capital needed 

to retain this EUR 3.2 million risk. Both effects should 

be considered together, as shown in Figure 26. As a 

consequence the profit of this book including cost 

Fig. 26: Impact of the reinsurance program on the profit of the P&C 

new business book including cost of capital (in €m) 


P&C NB Net 

Standalone 


15.8 

54.8 

P&C NB Net Profit 

excluding CoC 

13.3 

13.3 

Gross 

Net 

P&C NB Profit 

excluding CoC 

P&C NB 


Benefit: 56.3% 

10.5 

P&C NB Net Profit 

including CoC 

Costs of RI 

2.5 

Total Net Capital = 71.2 


allocated to P&C NB 

2.8* 

< 

P&C NB Net 

Diversified 


23.9 

of capital (the value chosen as a valuation criterion) 

has increased, which justifies the application of this 

reinsurance structure. 

Even though the cost of capital that the insurer 

can save with this reinsurance program exceeds 

the cost of reinsurance, the agreement is probably 

also advantageous for the reinsurer. Reinsurers are 

presumably better able to diversify their risks than 

direct insurers, thus for the reinsurer, the cost of capital 

resulting from the transferred risk is generally lower 

than the cost of capital resulting from the same risk 

for the direct insurer. 

Saved CoC 

3.2 

6.03 

Operational Risk 

Other Risks 

Credit 


Gross Non-Life Reserves 

Gross Non-Life New Business 

* CoC Rate = 11.5% 

2.8 

Gross Net 

P&C NB 

Costs of Capital

Fig. 27: Optimal reinsurance, win-win situation (in €m) 

Reinsurer 

Portfolio 

Impact of new risk 

on reinsurer portfolio 

Costs of RI** 

2.5 

** CoC Rate = 12% 

P&C NB Gross 

Standalone 


Transferred 


P&C NB 

Net 

Standalone 


54.8 

< 

This concept is illustrated in Figure 27. Following the 

previous example, part of the P&C new business book 

is transferred to the reinsurer. On the right-hand side 

one can see the relief of the insurer’s diversified capital, 

which enables certain savings in the cost of capital. It 

should be kept in mind that this capital relief depends 

on the Insurer’s risk appetite, which determines the 

definition of risk capital used. It is also interesting to 

note that the capital relief for this risk class may (and 

probably will) influence the capital required for other 

risk classes, which can also be observed in Figure 27. 

The stronger diversification effect of the transferred risk 

within the reinsurer portfolio is visible on the left-hand 

side of the graph. All in all, the cost of capital for the 

insurer, even though advantageous, is greater than the 

cost of capital that the reinsurer needs in order to retain 

the transferred risk (with comparable cost of capital rates 

for both parties), which creates a win-win situation. 

APPLYING THE ALTERNATIVE REINSURANCE 

STRUCTURE TO THIS PORTFOLIO 

This reinsurance structure suffers from certain drawbacks. 

Cover for the Property line is limited to EUR 120 million, 

whereas the underlying gross model presupposes potentially 

higher losses. The cover is therefore “too short” 

for the underlying gross portfolio. The second problem 

arises from the unfavourable conditions of the quota 

share contract for Casualty and Motor lines (in particular 

due to the rather low ceding commission). Consequently, 

it is clear that this program could be improved. 

Change of 

diversified capital 


Saved with RI* 

3.2 

* CoC Rate = 11.5% 

Definition of Risk Capital reflects 

the company’s risk appetite, 

Costs of Capital saved with RI 

depend also on the chosen percentile 

P&C NB Net 

Fully Diversified 


Total Capital 

Net = 71.2 Gross = 94.3 

Reinsurer-Cedant win-win situation. Reinsurer benefits from the stronger diversification of the transferred risk within its portfolio. 

Fig. 28: An alternative 

reinsurance program for the P&C 

new business book 

P&C Line Gross 

Premium 

Adjusted RI Structure 

Casualty 70 Per Risk XL2: 60 xs 60 

Motor 50 Per Risk XL1: 40 xs 20 

Property Cat 10 

Total Property 80 

Total P&C 210 

CAT XL2: 60 xs 40 

CAT XL1: 30 xs 10 




In response to these drawbacks, a slightly modified alternative 

reinsurance structure is proposed, as shown in 

Figure 28. The reinsurance cover for Casualty and Motor 

lines has been restructured so that the risk mitigation 

impact of the quota share is (partially) compensated by 

the lower attachment point of the non-proportional program, 

i.e. EUR 20 million rather than EUR 30 million. 

Secondly, the limit of cover for the Property line has been 

increased from EUR 120 to EUR 200 million so that the 

extreme risks in the portfolio are also covered. 


Fig. 29: Comparing the two structures (in €m) 

Evaluation of both reinsurance structures using the proposed criteria. The adjustments to the reinsurance structure lowered the cost 

of reinsurance and at the same time enabled further savings in the cost of capital. 

The impact of these changes in the reinsurance structure 

should be analyzed, using valuation criteria previously 

proposed. As expected, replacing the quota share 

with the extended non-proportional cover significantly 

lowers the cost of reinsurance. Further extension of the 

Property cover to include peak risks lowers the amount 

of required capital for this line, and consequently also 

the total cost of capital for this portfolio (Figure 29). 

To conclude, the adjusted reinsurance structure is more 

appropriate for the considered P&C new business 

portfolio due to the selected valuation criteria and 

measures adopted for the required capital (particularly 

with respect to the level of risk aversion). 

Now, it would be interesting to look at the influence 

of both reinsurance structures on the solvency ratio of 

the company in question. 

Fig. 30: Solvency ratios (in €m) 

120 

100 

80 

60 

40 

20 

0 

114% 

94 

Gross 

151% 153% Solvency ratio 

71 

107 

Original Net Adjusted Net 

Total Required Capital Available Capital 

Solvency ratio for the insurer: gross and net with both investigated reinsurance structures. 


Original RI Structure 

Adjusted RI Structure 

P&C NB Profit 

excluding CoC 

13.3 

14.3 

2.5 

Costs of RI 

1.4 

Costs of RI 

70 

3.2 

Saved CoC 

3.4 

Saved CoC 


P&C NB 

2.8 

2.6 

As indicated in the initial portfolio characteristics, the 

company has available capital of EUR 107 million at 

its disposal, with total required capital amounting 

to around EUR 94 million. For simplification purposes, 

it is assumed that the reinsurance structures do not 

materially impact the available capital, however 

they reduce the total required capital to EUR 71 

and EUR 70 million respectively. The solvency ratio is 

computed by dividing the required capital by available 

capital, as shown in Figure 30. 

The initial gross solvency ratio of 114% appears to be 

rather low and significantly improves with reinsurance, 

rising to 151%. Further refining of the reinsurance 

structure improves it a little further.

Fig. 31: Impact of the reinsurance refinement on the standalone 

required capital by risk class (in €m) 

... With the original reinsurance structure ... With the adjusted reinsurance structure 

NB P&C Standalone Capital 

Property NonCat 

Prop. Cat 

Motor 

Casualty 

DETAILED ANALYSIS OF THE IMPACT 

OF THE REINSURANCE STRUCTURE 

ADJUSTMENTS ON THE REQUIRED CAPITAL 

First of all, it is necessary to look at the impact of the 

reinsurance refinement on the standalone required 

capital by risk class (see Figure 31). The overall standalone 

required capital for the P&C new business book 

has moderately decreased. The capital for the Property 

line has decreased naturally, due to the extended cover 

for peak risks. 

Restructuring the reinsurance for Motor and Casualty 

caused a decrease in the standalone capital for the 

Motor line and a slight increase in capital for the 

Casualty line. This interesting effect can be explained by 

the different characteristics of the underlying models. 

NB P&C Standalone Capital 


Prop. Cat 


Casualty 

0 10 20 30 40 50 60 0 10 20 30 40 50 60 

More severe losses recover better from the non-proportional 

structure than from the quota share. For more 

frequent and less severe events, the capital reduction 

from the non-proportional program is lower than from 

the quota share. The standalone capital for the Cat line 

remains unchanged, as expected. 

The reduction in diversified capital for the P&C new 

business book is mainly due to lower diversified capital 

for the Property line (see Figure 32). The restructuring 

of the Motor and Casualty reinsurance cover gives 

another detailed picture. The contributions made by 

these lines in a diversified portfolio increase slightly, 

especially for Motor. The explanation given above also 

applies in this case. 

Fig. 32: Impact of the reinsurance refinement on the diversified 

required capital by risk class for the new book P&C Diversified Capital (in €m) 

... With the original reinsurance structure ... With the adjusted reinsurance structure 


Prop. Cat 


Casualty 


Prop. Cat 


Casualty 

0 2 4 6 8 10 12 0 2 4 6 8 10 12 


More detailed results for the Property cover are shown 

in Figure 33. In the graph on the left side, the extreme 

risks exceeding the original reinsurance cover are visible 

(although not frequent). The 1% worst net risks after 

application of the adjusted reinsurance cover, presented 

in the graph on the right side, do not contain any losses 

with such a high gross loss amount. 

More detailed results for the Motor and Casualty cover are 

shown in Figure 34. In the original cover (the graph on the 

left) the quota share reduces all gross losses and the nonproportional 

reinsurance only impacts the peak risks. The 

adjusted non-proportional structure allows the retention of 

more small risks; the retention line has a sharper increase 

(the graph on the right). The non-proportional cover acts 

not only on the peak risks but also in the middle band. 

Fig. 33: Detail Results: Property Non-Catastrophe, aggregate losses 

Aggregate Losses - Gross vs. Net / 1% worst Net Losses 

Net Losses (EURm) 

160 

140 

120 

100 

80 

60 

40 

20 

Original RI 

0 

0 

50 100 150 200 250 300 50 100 150 200 250 300 

Gross Losses (EURm) 



160 

140 

120 

100 

80 

60 

40 

20 

Adjusted RI 

Refinement of the Property cover – the graphs present 1% worst aggregated net losses for both reinsurance structures. For each of these 

losses the corresponding gross and net loss amount is graphed in order to visualize the impact from the appropriate reinsurance program. 

Fig. 34: Detail Results: Motor, aggregate losses 

Aggregate Losses - Gross vs. Net 


90 

80 

70 

60 

50 

40 

30 

20 

10 

Original RI 

retention 

XL 

0 

0 20 40 60 80 100 120 140 160 


retention 

Refinement of the Motor and Casualty cover – the graphs represent aggregated losses for both reinsurance structures, in a gross vs. net loss 

amount view. The impact from quota share and non-proportional reinsurance can be clearly observed. Severe gross aggregate losses are 

significantly reduced if they are caused by rare but severe events (triggering the layer); and are reduced less if caused by the accumulation 

of more frequent and less severe risks. 



90 

80 

70 

60 

50 

40 

30 

20 

10 

Adjusted RI 

XL 

0 

0 20 40 60 80 100 120 140 160 

Gross Losses (EURm)

IV. Partial model and fully 

diversified capital 

The main difficulty that arises when assessing valuation 

criteria previously mentioned lies in the need to estimate 

the diversified required capital. Measuring the impact 

of structural reinsurance changes (in particular for Non- 

Proportional cover) on the diversified capital is not simple and 

usually requires an internal model based on a Monte Carlo 

simulation. This is not a straightforward process because 

the internal model is a holistic one, a “huge machine” that 

needs time to input data and compute results. 

SCOR has recently developed a Capital Deployment Tool 

(CaDeT), which allows it to forecast the changes to standalone 

and diversified capital caused by new exposure, 

using the standalone and diversified capital intensity 

benchmark. This capital forecast, although produced by 

a factor model, accounts for the non-linear impact of the 

change of diversification within the considered portfolio, 

determined by the size of the risk classes of which it is composed 

and by the corresponding change in exposure. 

Fig. 35: Assessment of fully diversified gross capital 

Gross NB P&C 

stochastic model 

Analysis 

Model 

Scenarios 

Data P&C NB model: 

• Gross Loss Models 

• Gross Premiums and Expenses 

• Dependencies 

• (Reinsurance structure for net view) 

Outputs 


Factors 

Benchmark 

Gross Portfolio 

This new methodology has been applied to the example 

used in this article, as shown in Figure 35. As a starting 

point it is necessary to use a partial stochastic internal 

model in order to gain a good picture of the diversification 

structure of the gross P&C new business book. 

The model produces gross diversified results purely 

within this portfolio, but it is much lighter and faster 

to run. These results, together with a number of 

suitable gross benchmark portfolios (reflecting 

diversification effects with other risk classes), were 

used as a starting point for the CaDeT model in order 

to obtain an approximated fully-diversified required 

capital by risk class. 

A similar approach has been followed in order to compute 

the corresponding net diversified capital Figures. 

For this case, however, the new partial model has been 

run for the P&C net business book, and fully diversified 

gross results were used as a benchmark for the diversification 

effects with other risk classes. The estimated net 

fully diversified capital has been utilized to compute the 

change in the cost of capital due to reinsurance. 

Non-Stochastic 

Capital Deployment 

Factor model: 

• Diversification correction 

• Projection of standalone capital 

• Aggregation of segments 

• Adjusted profit distribution 

Benchmark Data for the whole Portfolio: 

• Exposure 

• Standalone Capital 

• Diversified Capital 

• Strength of internal diversification 

Fully diversified 

Gross results 

Results: 

• Required Capital 

• Diversification Benefit 

• Sensitivity 

Assessment of fully diversified required capital applying CaDeT methodology. The graph illustrates the estimation of gross diversified 

capital. The corresponding net results were estimated analogically, applying the Net NB P&C partial model as well as gross results 

as a benchmark for diversification with other risks. 

Conclusion 

The valuation of different reinsurance options depends 

on the motivation for buying reinsurance. 

When calculating the cost of reinsurance, the cost of 

capital savings is a way of considering the impact of 

reinsurance on capital. The assessment of capital relief is 

relative and depends on the definition of capital used. 

A fully diversified capital view is essential to take into 

account the cost of capital relief (Solvency II standard 

model, internal model, rating agencies). 

Partial stochastic models allow the evaluation of capital 

impact for different reinsurance options in a more 

appropriate way. 

Factor models or models based on aggregate 

distributions (like rating agencies’ models and Solvency II 

standard model) are struggling to reflect the impact 

of non-proportional reinsurance, so the idea is to use 

the partial model and then to use a kind of capital 

assessment tool for diversification effects. 


4 

CAPITAL ASSESSMENT 

BEYOND STOCHASTIC 

MODELING 

After an introduction to Capital 

Modeling at SCOR, this article addresses the motivation 

behind the development of a factor-based 

methodology used to approximate capital assessments 

for different cases, as an extension of the 

stochastic model. It then elaborates on the main 

characteristics of this methodology – namely the 

consideration of dependencies and diversification 

– through examples, highlighting some of the 

challenges involved. 


EVA SCHLÄPFER DE MONTMOLLIN 




I. Capital Measurement 

and Allocation 

HOW DOES SCOR COMPUTE ITS CAPITAL 

REQUIREMENTS? 

SCOR uses a sophisticated stochastic model to calculate 

its risk-based capital by quantifying the combined 

capital impact of worst-case scenarios from various 

risk drivers. Assets and liabilities are modeled using 

simulation techniques and dependencies are applied 

both within and between portfolios. This ensures that 

dependent risks are properly accounted for and that 

conservative benefits from the diversification within the 

portfolio are taken into account. 

Risk mitigation measures such as retrocession, asset 

allocation and hedging are applied to ensure that the 

net risk profile is modeled. 

Figure 36 shows an example output of SCOR’s Group 

Internal Model. 

The diversification effect is the difference between the 

standalone capital and the diversified capital.

Fig. 36: SCOR’s capital requirements 

Standalone Risk-Based Capital and risk category contributions to SCOR’s total diversified Risk-Based Capital 

Risk-Based Capital 2009, SCOR Group 

in EURm 

RBC 

Standalone* 

RBC 

Diversified 

Share 

of RBC 

P&C New Business 1 595 834 26% 

P&C Reserves 1 646 1 249 39% 

Life 1 703 877 28% 

Assets 699 127 4% 

Credit Risk 280 38 1% 

Other Risks (inc. FX) 321 61 2% 

Operational Risk 214 214 7% 

Total RBC 6 244 3 186 

* Sum of Standalone RBC 

Diversification Benefit 49% 

SCOR calculates its risk based capital by quantifying the combined capital impact of worst case scenarios from various risk drivers. 

Diversification between risks is the basis of the business 

model for insurance and particularly reinsurance 

companies, for the following reasons: 

• Risk Diversification reduces a company’s need for riskbased 

capital. This is key to insurance, reinsurance 

and investments. 

• Nevertheless, risks are rarely completely independent: 

- Stock market crashes are usually not limited to 

one market. Recent market crashes show that 

global crises are interdependent. 

- Certain Lines of Business are affected by economic 

cycles, such as Liability, Credit & Surety and Life 

insurance. This creates an implicit dependency. 

- Major catastrophes can produce claims in 

various Lines of Business. 

• Dependency between risks reduces the benefits of 

diversification. 

• The influence of dependency on the aggregated riskbased 

capital is thus crucial and needs to be carefully 

analyzed. 

The processes and systems of the internal model are 

relatively complicated; running the model with the 

proper data is time consuming. The full model is very 

detailed but not very well suited to swift analyses. This 

is the main reason why SCOR has recently developed 

a simplified version of its internal model as a basis for 

dynamic decision making. 

This is a capital projection tool that is less complex than 

the internal model, and is useful for the following: 

• “As if” analysis. 

• Strategic planning support. 

• A useful benchmark for the Group internal model 

and standard formulas. 

• Identification of main contributors to the capital 

requirements. 

• Assessment of diversification benefit and change as 

a consequence of portfolio change. 

• Assessment of capital adequacy (solvency ratio). 

• Analysis of profitability. 


Fig. 37: Functionality and high-level architecture of CaDeT 

SCOR 

Group Internal Model Data 

Benchmark Data 

Data: 

New 

• Exposure 

exposure 

• Standalone Capital Adjustments 

• Diversified Capital 

• Economic Value Distribution 

Parameters: 

Strength of internal diversification 

Future: 

Mean time to payment (for discount) 

SCOR’S CAPITAL DEPLOYMENT TOOL (CADET) 

This capital projection tool is called CaDeT. It is a factorbased 

model that takes output from the stochastic 

Group Internal Model and is used to calculate new 

capital requirements under changed assumptions. 

A key feature of this tool is an adjusted diversification 

effect for a new portfolio. This is one of the main 

differences compared to other factor-based models, 

such as rating agency models for example. 

CaDeT is a light analytics tool that allows the user to 

change assumptions on exposure and to approximate 

capital and risk. It is a major breakthrough in methodology 

that blends stochastic modeling with factor-based 

methods. Due to high demand, ongoing extensions are 

planned to the functionality presented here. 

CaDeT requires some benchmark portfolio data such as 

exposure information, standalone capital information, 

diversified capital information at a certain granularity 

and an economic value distribution for the risk classes 

of interest. Then the user can adjust the exposure 

information, for example to reflect a new business plan. 

The tool extrapolates the portfolio benchmark, taking 

into account certain non-linear model assumptions 

on diversification behavior. In this way it projects new 

standalone and diversified capital amounts. CaDeT also 

delivers new diversification benefits, updated required 

capital and a sensitivity analysis on the profit distribution 

in just a few seconds. It is a good approximation of the 

Group Internal Model, but since it relies on simplified 

methods CaDeT is obviously less accurate than the 

Group Internal Model. 


CaDeT Result approximation 

Factor model: 

• Diversification correction 

• Projection of standalone capital 

• Aggregation of segments 

• Adjusted profit distribution 

Future extensions: 

• Retrocession optimizer 

• Profitability assumptions 

• S&P analytics 

Fig. 38: Example - original portfolio 

Risk Category 

in EURm 

• Required Capital 

• Diversification Benefit 

• Sensitivity 

II. Capital assessment beyond 

stochastic modeling 

A realistic portfolio can be analyzed in order to explain 

the methodology in more detail and to demonstrate the 

impact on diversification and capital. The table below 

shows exposure information for the main risk categories 

as well as their capital contributions in EUR millions. 

This portfolio is a P&C book with mainly Property, 

Liability and Motor. Assets are mostly invested in bonds, 

real estate and some equities. 

Exposure* Fully diversified 

Capital 

Non-Life New Business 210 52 

Non-Life Reserves 240 24 

Invested Assets 83 10 

Credit 107 5 

Other Risks 107 1 

Operational Risk 107 2 

Total capital 94 

Example portfolio – gross P&C book with main risk categories as 

well as contributions to the diversified capital. 

* Exposure means: Written premium for New business, Reserves, 

Market value of invested assets and Available Capital for credit, 

operational and other risks.

Fig. 39: Example – original exposure and capital 

Detailed view on P&C new business 

Premium 

Standalone Cap 

Property 

Non Cat 

Property 

Cat 


Casualty 

10 

0 20 40 60 80 0 20 40 60 80 0 20 40 

Total contribution to the fully diversified capital of P&C 

new business is EUR 52 million. Property Non-Cat is the 

largest Line of Business and consumes the most capital 

50 

70 

80 

29 

43 

53 

75 

Diversified Capital 

10 

10 

9 

23 

69% 

66% 

81% 

79% 

P&C New Business 

Book Capital: 52 

Example portfolio – Detailed view of the P&C new business book with premium volumes, standalone capital and contributions to the fully diversified 

capital. The diversification benefits per Line of Business in the full diversification is represented in orange. All amounts are in EUR millions. 

WHAT HAPPENS IF EXPOSURE CHANGES? 

Fig. 40: Standalone Capital (in €m) 

Example for Motor – doubling the premium 

Motor (New Business) 

160 

140 

120 

100 

80 

60 

40 

20 

0 

160 

140 

120 

100 

80 

60 

40 

20 

0 

~106% standalone capital intensity 

50 

Written Premium 

100 

2x 

53.4 

Standalone Capital 

in a standalone and a diversified view. The diversification 

benefit for smaller Lines of Business is larger, but 

Property is also well diversified. 

In soft market 

you can grow only 

by adding business 

with higher risk 

~106% standalone capital intensity 

160 

140 

160 

Motor Example, what can happen to the capital if the premium doubles. Three examples are presented showing the effects 

of diversification for different types of growth. 

3x 

Grow if diversity 

(healthy market), 

maximum of internal 

diversification: 

sqrt(2) = 1.41 

1.41x 

grow by adding 

weakly dependent 

business 

1.8x 

120 

100 

80 

60 

40 

20 

0 

160 

140 

120 

100 

80 

60 

40 

20 

0 

160 

140 

120 

100 

80 

60 

40 

20 

0 

100 

100 

100 

~75% standalone 

capital intensity 

75 

~100% standalone 

capital intensity 

100 

Diversification Benefit 


In the Motor Line of Business, the written premium is 

EUR 50 million and the standalone capital is EUR 53 million, 

which means that the capital intensity is around 100%. 

If the premium volume doubles, the capital should 

increase but it is not immediately clear by how much. The 

increase in capital actually depends on how the portfolio 

can be expanded. This can be achieved by increasing 

the share of business or writing new business, or a 

combination of both. For example in a soft market, the 

only way you can grow is through writing business that 

is worse than the existing book from a risk perspective. 

Then the capital intensity increases and the required 

capital subsequently increases over-proportionally. In the 

best-case scenario – i.e. in a healthy market – it is possible 

to expand so that the portfolio’s internal diversification 

improves significantly, increasing capital by a factor of 1,41 

(square root of two). If the portfolio were expanded with 

weakly dependent business, the capital would increase 

by a factor of around 1.8. 

Now, it is possible to explore the capital projections of 

the methodology if the portfolio changes. Two cases 

will be explored, one better-balanced portfolio and 

one extremely unbalanced portfolio, keeping the total 

premium volume constant. 

The first, balanced, portfolio has a new premium 

volume for Property Non-Cat of EUR 56 million 

(previously EUR 80 million), EUR 5 million for Property 

Cat (previously EUR 10 million) and EUR 99 million for 

Casualty (previously EUR 70 million). The Motor premium 

volume does not change. The total premium volume 

stays at EUR 210 million. 

In the second, extremely unbalanced portfolio, 

EUR 180 million of the new business premium is written 

in Motor, this is more than 80% of the total premium. 

The rest is written in the smaller lines. The total premium 

volume stays at EUR 210 million. 


TOTAL CAPITAL REQUIREMENTS 

AND PER RISK FACTORS 

Fig. 41: Capital requirements 

per risk factors (in €m) 

Capital 

in € 

200 

150 

100 

50 

0 

Main Risks 

Other 

Operational 

Credit 

Capital Deployment Overview 

94m 

52m 

89m 

46m 

Gross Gross Better 

Balanced 

170m 

127m 

Gross 

Unbalanced 


P&C Reserves 


Total capital (in orange) and main risk factors to the total 

required capital for the original example portfolio as well as 

the two modified portfolios. In black we show the 

contributions of the P&C new business capital. 

The total capital in the better-balanced portfolio 

drops to EUR 89 million, mainly due to the lower 

contribution of P&C new business. The extremely 

unbalanced portfolio has a much higher total capital 

of EUR 170 million, again mainly due to the much 

bigger impact of the P&C diversified capital. Capital 

consumption for the other risk factors only changes 

marginally due to the change in diversification. 

The impact of the different Lines of Business on the 

capital contributions for the balanced and extremely 

unbalanced portfolio should be analyzed more 

precisely.

Fig. 42: Example – balanced 

Increasing Premium volume for Casualty, decreasing for Property 

Property 

Non Cat 

Property 

Cat 


Casualty 

Premium 

5 

-5 

50 

0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 



As shown in Figure 42, the two Property lines diversify 

better in the more balanced portfolio. The diversification 

benefit increases from 69% and 66% for Property 

Non-Cat and Property Cat to 77% and 79% respectively. 

As expected, the diversification benefit for 

Casualty decreases from 79% to 68%, which is a natural 

consequence of a significant growth of this line. 

Below is a similar graph for the extremely unbalanced 

portfolio. 

56 

+29 

-24 

99 


14 

50 

53 

66 


Better balanced portfolio, P&C new business book with premium volumes, standalone capital and contributions to the fully diversified 

capital. The diversification benefits per Line of Business in full diversification are in orange. All amounts are in EUR millions. 

Fig. 43: Example – very unbalanced 

Basically a mono-line company 

Total 

Property 

Property 

Cat 


Casualty 

Premium 

2 

14 

14 

180 


11 

9 

9 

3 

11 

11 

21 

77% 

79% 

80% 

68% 

The standalone capital and the contribution of Motor 

to the total diversified capital increase significantly. The 

diversification benefit for Motor drops from 80.6% to 

38.7%. Diversification benefits on all lines except Motor 

are very large and the small lines are allocated very little 

capital. This effect can often be observed when there is 

a dominating risk category – “the winner takes all”. 

0 50 100 150 200 0 50 100 150 200 0 50 100 150 

208 


0 

1 

0 




96% 

88% 

39% 

127 

Extremely unbalanced P&C new business book with premium volumes, standalone capital and contributions to the fully diversified capital. 

The diversification benefits per Line of Business in full diversification are in orange. All amounts are in EUR millions. 

96% 



SOLVENCY RATIOS 

Finally, it is also interesting to observe the behavior of 

the solvency ratio between the original portfolio, the 

balanced portfolio and the unbalanced portfolio. The 

solvency ratio is defined as the quotient of the available 

capital of EUR 107 million and the required capital. 

Underwriting new business in a more balanced way 

improves the solvency ratio from 113% to 120%. The 

solvency ratio for the unbalanced case is below 100%, 

which means that this company would need to urgently 

raise capital or lower its required capital by de-risking the 

portfolio. Of course, in reality, there are limitations as to 

how fast significant portfolio changes can be achieved 

due to changes in the underwriting policy. Reinsuring 

part of the portfolio could be a faster solution in terms 

of moving to a better-balanced portfolio. This example 

shows in particular that having a well-balanced portfolio 

is important from a capital perspective. 

Fig. 44: Solvency ratios 

of the three example portfolios 

Div. Capital 

in €m 

200 

150 

100 

50 

0 



vs. Available Capital 

113% 120% 

Gross Gross Better 

Balanced 

63% 

Solvency ratio 

Gross 

Unbalanced Available 

Capital 

III. Modeling of Diversification 

and other Challenges 

HOW CAPITAL CHANGES WITH EXPOSURE 

This section will now show in more detail how capital 

changes due to changes in exposure. 

Fig. 45: Exposure against capital 

intensity and against capital amounts 

(in %) 

200% 

180% 

160% 

140% 

120% 

100% 

80% 

60% 

40% 

20% 

0% 

0 

(€m) 

1,000 

900 

800 

700 

600 

500 

400 

300 

200 

100 

0 

0 

1,000 2,000 3,000 4,000 

Exposure 

(€m) 

Standalone Capital Intensity 

Diversified Capital Intensity 

200 400 600 800 

Exposure 

LoB standalone Capital 

Diversified Capital 2009 

1,000 

(€m)

In Figure 45, exposure is shown on both X axes. The 

Y axis on the first graph shows capital intensity, while 

the Y axis on the second graph shows capital amounts. 

This model requires a number of parameters that need 

to be adapted and calibrated for different Lines of 

Business and markets. 

The behavior of the standalone capital intensity will be 

explained more in detail; diversified capital intensities 

behave similarly. In the second graph you can see 

that the capital amounts increase with the exposure. 

The development of standalone capital with growing 

exposure is modeled, assuming that a small portfolio 

can be expanded with improving internal diversification. 

This is where the capital intensities decrease 

with growing exposure. But there is a saturation 

point where diversified growth is no longer possible. 

The portfolio can only grow with worsening internal 

diversification, or because only riskier new business can 

be underwritten. From this saturation point the capital 

intensities increase once again. 

The parameterization of the curve depends on the 

Lines of Business, the portfolio and the market, and 

should be chosen in such a way as to reflect realistic 

behavior. 

CHALLENGES AND FUTURE DEVELOPMENT 

It is worth mentioning that there are many challenges 

and areas of future development surrounding this 

methodology. For example, the modeling of duration 

matching and interest rate sensitivity needs to be 

improved. If the asset allocation is changed significantly, 

this will change the duration of the assets, which may 

be better or worse matched to the duration of the 

liabilities. This should have an impact on the capital. 

Furthermore, the methodology needs some parameters 

– for example to describe the change in standalone 

capital when exposure changes, as discussed in 

the previous section. A realistic choice reflecting the 

portfolio and market conditions with appropriate 

parameters is challenging. 

To summarize, the methodology as it stands today is 

used for the following: 

• Approximate assessment of capital requirements, 

taking changing diversification into account. 

• Identification of main contributors to the capital 

requirements. 

• Assessment of capital adequacy (solvency ratio). 

• Approximate adjustments to the capital with changing 

exposure. 

Further planned developments include: 

• Profitability analysis of changing portfolio. 

• Impact of different reinsurance structures on 

capital. 

• Better modeling of interest rate sensitivities. 

This methodology can also be used to make the leap 

from a partial stochastic model analysing the impact of 

a reinsurance structure to fully diversified capital. 

Conclusion 

For quick approximations of capital requirements, 

a simple proxy of a complex model, in our case the 

internal model, is needed. The methodology presented 

here shows one way of building such a proxy tailored 

to SCOR’s needs. Since diversification is an essential 

concept, this methodology is focused on more realistic 

modeling than Solvency II’s quantitative impact studies 

or the rating agencies’ models. 

Obviously, this methodology cannot replace a complex 

internal model, but it is a very useful extension since the 

calibration is based on the underlying complex model 

for fast and simple analyses. This article has shown a 

way of calibrating this model with benchmark data 

based on the Group Internal Model. Using other benchmark 

data allows an extension in order to address other 

issues or test potential business changes with the same 

methodology. 


5 

INTEGRATING RISK 

MODELING INTO 

THE ORGANIZATION JANICE COWLEY 

Program Director of CoCPIT, SCOR 

CoCPIT (Cost of Capital, Pricing, 

IT) was a two-year program of change in terms of 

the cost of capital, pricing and IT, and was aimed 

at further integrating the internal model into the 

organization. The program concluded in the 

summer of 2010. At its conclusion SCOR was able 

to demonstrate the Group’s strong involvement in 

modeling and the connectivity of the results to its 

day-to-day activities. 

This article provides simple and straightforward 

examples on how SCOR has: 

• Cemented the process of modeling and governance 

within the organization; 

• Integrated the use of the Group’s internal model into 

the organization. 


I. Cementing the process 

of modeling and governance 

within the organization 

The CoCPIT approach is based on three components: 

• Data Ownership and Audit ability are increasingly 

important due to the upcoming Solvency II regulations. 

They enable to have the data that feed the 

models within the organization. The systems that have 

been developed are built on three components: 

- Owners sign off on their data; 

- Automation and solid design provides a robust 

audit trail; 

- The development of a centralized data repository 

is used as the basis for capital modeling 

and will be used in the future for Solvency II and 

standard models. 

• A robust governance structure constitutes the second 

component of CoCPIT, comprising: 

- A defined and transparent governance 

structure; 

- The clear establishment of roles and responsibilities; 

- The involvement of the Executive Team and 

Board in the process. 

• Use of expertise within the organization to make 

sure that: 

- Processes are designed and their results validated 

in partnership with the organization; 

- Processes are flexible in order to benefit from 

expert judgment and knowledge. It is essential 

to know the limitations of the models; 

- Transparency is ensured concerning limitations 

and assumptions.

A CLEAR GOVERNANCE FRAMEWORK UNDERPINS THE MODELING PROCESS 

Fig. 46: Modeling of risk and capital 

The involvement of the organization in modeling, validation, challenge and review 

is of paramount importance to SCOR 

Current 

Portfolio 

Plans 

Reserves 

Enter & Sign 

off data 

Retrocession 

& Hedges Economic 

Scenarios 

Model Life, 

P&C, Assets 

The process is structured so that the starting point is the 

provision of data and sign off before execution of the 

modeling of standalone and globally diversified results. 

The process is overlaid with the ALM Governance 

framework, which ensures the necessary technical and 

management reviews of results. 

Model Group 

Capital 

Risk-Based Capital 

& capital 

intensity 

Strategic Asset Allocation 

& duration matching 

Risk drivers 

& strategies 

Capital allocation 

Process 

Expert Review 

Groups 

ALM Technical 

Committee 

ALM Committee 

Board of Directors 

Info to regulators 

& Rating Agencies 


Fig. 47: A clear governance framework underpins modeling process 

The pyramid above illustrates the review and approval 

process at SCOR. 

Expert reviews are designed to ensure that results are 

reviewed at a division level by those familiar with the 

risks in that division. The review groups are composed 

of experts in that area. These division experts have a 

responsibility to ensure the technical and modeling 

accuracy of the results on a standalone level. The scope 

of the role includes ensuring the transparency of any 

limitations or assumptions within the model. 

The global technical review is performed by the 

ALM Technical Committee, which also reviews the 

consolidated results of the organization. Transparency 

on assumptions and limitations is provided and 

the impact of any changes to methodology is made 

explicit. The ALM Technical Committee comprises 

technical experts, including representatives from Group 

Risk Management division, Chief Reserving Actuaries, 

Pricing Actuaries, Head of Natural Catastrophes, Global 

Head of Retrocession, and so on. 

The ALM Committee, composed of the Group’s executive 

team, completes the management review of 

the results. The ALM Committee relies on the ALM 


Board of Directors via Risk Committee 

Endorses ALM results 

Approves Strategic Asset Allocation ”SAA” 

Endorses capital and buffer 

ALM Committee 

Optimal capital deployment to achieve risk/return 

Proposes risk based capital, buffers & SAA 

Sets risk mitigation strategies & manages to defined risk tolerances 

Affirms the appropriate capital allocation per portfolio for pricing 

ALM Technical Committee 

Alerts ALM Committee to any limitations of the model 

Assures accuracy and completeness of models + results 

Proposes the capital allocation per portfolio for pricing 

Assets, Life & P&C Technical Review Groups 

Assure the accuracy and completeness of standalone results 

Alerts the ALM Technical Committee to limitations, areas of 

judgement and decisions taken to assure accurate results 

Technical Committee for the accuracy of the results and 

to understand the drivers behind change, limitations 

and areas where judgment is required. As part of this 

process, the ALM Committee often focuses on areas 

requiring management action, for instance areas 

where risks are arising, where diversification is not 

optimized or where there are proposals worth pursuing, 

for example in relation to strategic allocation or risk 

mitigation. These proposals are then put to the Board 

of Directors. 

Finally the Board of Directors, via the Risk Committee, 

endorses the results of the modeling process, sets the 

strategic asset allocation and ultimately agrees on risk 

tolerance levels. 

TRANSPARENCY AND INVOLVEMENT 

ARE AT THE CORE OF SCOR’S APPROACH 

SCOR’s approach is based on three main features: 

• Advanced Models 

Governance 

Management/ 

COMEX 

Technical 

Expert 

These constitute a basis for decision. It is important to 

acknowledge that decision making needs to be based 

around the results, and processes must provide for that 

flexibility.

• Assumptions and Limitations 

It is crucial in terms of gaining the trust and acceptance 

of the organization to identify and communicate the 

limitations of the internal model. Assumptions are 

transparent, so their validity can be understood and 

challenged where necessary. Limitations are made clear 

so that informed decisions can be made. Judgement is 

applied as necessary. 

• Involvement and disclosure 

Involvement, and ensuring that a constructive dialogue 

takes place, are crucial to the success. The ability to 

analyze results is provided, thereby optimizing the use 

of the knowledge and expertise of the organization. 

These highlights illustrate how SCOR has cemented the 

process of internal modeling within the organization. 

• 

• 

• 

Risk/return decisions 

Financial & 

Economic Results 

Performance 

Measurement 

Business 

Actions 

Risk Tolerances & Limits 


Budgets 

II. Integrating the use 

of the Group internal model 

into the organization 

In this section are selected examples of how the results 

of the internal model are used at SCOR. 

THE INTERNAL MODEL IS A CORNERSTONE 

OF ERM AT SCOR 

Figure 48 shows how capital and risk are integrated into 

the organization, from strategy setting and business 

planning through to capital allocation, risk budgets, 

underwriting execution and performance evaluation. 

Strategy & Risk Appetite 

Capital and risk tolerance are key considerations when 

setting and affirming the strategy. 

• When considering the strategic plan, the implications 

on capital are confirmed and it is ensured that the 

Group does not exceed its risk tolerance. 

• When the results of the Group internal model are 

discussed, the Board Risk Committee ensures that 

SCOR’s target risk profile is respected and the 

expected return and volatility are understood. 

• The capital buffer is set in line with the capital shield 

strategy to avoid the need to regularly approach the 

shareholders for additional funding. 

Fig. 48: Integrating the internal model into the organization’s processes 

The internal model is a cornerstone of Enterprise Risk Management at SCOR 

• 

• 

Strategy & 

Risk Appetite 

Available capital & Risk Tolerance 

Target capital & Risk Profile 

Portfolio 

Planning & 

Optimization 

Capital 

Allocation 

• 

• 

• 

• 

• 

• 

• 

• 

Capital & Risk Awareness 

Trade off analysis 

Diversification benefit 

Sensitivity of result 

CaDeT – SCOR’s tool 

to support strategic planning 

Capital diversification benefits 

Principles of capital allocation 

Proactive operational risk allocation 


Fig. 49: Appropriate executive level governance over risk tolerance and adherence 

Risk Tolerance is set by the Board of Directors 

SCOR’s Board and Executive Management team regularly review the Group’s Risk Profile and Available 

Capital to ensure that they remain aligned with the Group’s Risk Appetite framework 

Change in Economic Value in EUR bn 

At a strategic and Board level, an understanding of the 

risk profile is important. The corporate level tolerances 

are taken and combined with operational level limits, 

so that an end to end or top to bottom approach to risk 

tolerance and appetite within the company is get. 

Portfolio Planning & Optimization 

The portfolio planning and optimization is carried out 

on the basis of: 

• Capital & Risk Awareness 

• Trade off analysis 

• Diversification benefit 

• Sensitivity of result 

• CaDeT – SCOR’s tool to support strategic planning 


Net income Expected 

growth in 

economic value 

+ Buffer Capital Required Capital Available Capital 

Return Period in Years 

CaDeT is an abbreviation for Capital Deployment Tool. 

CaDeT provides real time information on the impact 

of changed exposure (or business plans) on capital 

requirements and diversification benefit/optimization. 

The tool is the organization’s response to the need for 

real time information on capital and risk. 

Figure 50 demonstrates its internal usefulness. The 

numbers used in this example are just illustrative. It is 

possible to alter exposure, and the standalone capital 

and diversified capital will respond instantly. A key 

feature is the consideration of the changed exposure 

and diversification benefit across the entire portfolio. 

In addition to presenting capital and risk numbers, this 

example shows the probable volatility of results. 

For SCOR, CaDeT has represented a big step forward 

in terms of integrating modeling results into the organization. 

CaDeT has proven to be extremely beneficial 

for strategic decision making in terms of portfolio 

composition and looking at potential business 

acquisitions. It is also used to cement the understanding 

of capital and risk across the organization, because it 

is very simple to use and easily accessible.

Fig. 50: CaDeT example (Simplified example) 

Risk Factor Exposure Standalone Capital Diversified Capital Diversification Risk 

Monitor 

Current Change New Current New Change Current New Change 2009 Plan 

P&C New Business 1,500 50% 2,250 800 1,100 38% 415 700 69% 48% 36% 

P&C Reserves 5,000 5,000 800 800 0% 625 600 -4% 22% 25% 

Life 1,300 1,300 850 850 0% 450 415 -8% 47% 51% 

Investments 5,000 5,000 350 350 0% 65 55 -15% 81% 84% 

Other 2,000 2,100 150 150 0% 30 25 -17% 80% 83% 

Total 1,585 1,795 13% 

Changes in exposure 

(different granularities available) 

Available & Required Capital - Internal Model 

Available Capital 

2,000 

Current 

2,100 

New 

100 

THE INTERNAL MODEL IS THE BASIS 

FOR CAPITAL ALLOCATION 

Current 

Under CoCPIT, SCOR sought to cement an over-arching 

global framework around capital allocation. To achieve 

this, SCOR established a global multi-division, multifunctional 

team to look at this topic. 

The key principles to follow in Capital Allocation 

were defined at the outset and can be summarized 

as follows: 

• Align with the Group Internal Model: the objective 

was to make sure that capital allocation was 

connected to our Group internal model results. 

• Allocate Gross of Retrocession: a second key principle 

was to allocate capital gross of retrocession – whilst 

SCOR allows for retrocession benefit in the pricing of 

treaties, the capital allocated and basis for measuring 

return, is gross. 


Diversified (net) capital 

1,615 

1,825 

New 

210 

Are assessed real-time 

in terms of risk and capital 

Solvency Ratio 

= Available/Required 

124% 

115% 

2009 Plan 

-9% 

Solvency monitor: 

• Capital allocated reflects risk: the general principle 

regarding the comparative quantum of capital 

allocated was that the greater the risk, the greater 

the dependence or the greater the development 

time, then the greater the amount of capital allocated 

to a portfolio or a treaty. 

• Unit of measure: the idea was to try introducing a 

common unit of measure for capital allocation – by 

using capital intensities, ensuring that the exposure 

measure is appropriate to the underlying business. 

• Monitor and manage: it was agreed upfront that 

there must be a monitoring process to review the 

actual capital allocated within pricing vs. expectation, 

so that deviations are known and understood and if 

necessary action is taken. 


Fig. 51: In pricing, allocation of capital follows strategic principles 

Measurement of risk and allocation of diversification benefit 

RISK TOLERANCES AT THE OPERATIONAL 

LEVEL ENSURE CONNECTIVITY OF THE 

STRATEGY WITH DAY-TO-DAY OPERATIONS 

In day-to-day decision making, capital and return on 

that capital is a key factor that SCOR examines in terms 

of when the business is actually written. 

Risk budgets 

Risk tolerance is set on the basis of available capital and 

is implemented at all levels of the company: 

• At Group level, to ensure that the company is not 

over-exposed to a single event combining all Lines 

of Business and asset classes; 

• At Lines of Business level, to avoid concentration 

of risk in specific Lines of Business or asset classes 

and hence ensure that the diversification benefits 

are optimized. 

Fig. 52: Risk / Return decisions at the treaty level are assessed 

against capital allocated 

The capital allocated to the risks has a cost associated: the shareholder expects a return, the bondholder an interest. 



Expenses 

Terms & Conditions 

Pure Losses 


Expenses 

Terms & Conditions 

Pure Losses 

Economic 

profit 

Cost 

of capital 

Expenses 

Expected 

Loss 

Economic 

profit 

Cost 

of capital 

Expenses 

Expected 

Loss 

It is worth noting that risk mitigation measures such as 

retrocession, and other hedging are applied to ensure 

that risk tolerance levels are respected. 

Business actions and performance 

measurement 

A key factor in the decision whether or not to underwrite 

a contract is the return on the allocated capital. 

Thus, at the treaty level the results of the internal model 

are used as a basis for decisions. 

As an example: 

Capital allocated 

The capital allocated is based 

on the risk of each and every treaty 

Diversification benefit from 

the internal model is leveraged 

The charge to the contract 

is the cost of that capital 

The amount of capital allocated 

is monitored 

• If 100 in capital is allocated and a profit of 15 is 

expected to be made, then the expected return on 

risk-adjusted capital is 15%. 

• If, however, this were a more risky contract, potentially 

with a greater dependence or a longer tail, then it 

may have a capital allocated of 150. Consequentially 

the expected return would be 10%. 

Return on capital 

The capital allocated is based 

on the risk of each and every treaty 

The performance / return assessment 

is based on that capital 

Risk Capital: 100, Profit: 15 

Return on capital: 15% 

Risk Capital 150, Profit: 15 

Return on capital: 10%

Conclusion 

The results of SCOR’s internal model are important 

to the company’s organization. The CoCPIT team has 

therefore invested a lot of time to ensure that the 

Group internal model is connected to day-to-day decision 

making within the organization. 

The involvement of the management from the outset 

is of paramount importance in terms of integrating 

capital and risk management into the organization. 

Robust governance with executive involvement ensures 

the appropriate management of risk. 

The internal model provides important information to 

assess whether the Group’s risk profile is in line with the 

Group risk appetite framework. Nonetheless, it does 

not replace the need for decisions. 

Finally, building the trust of stakeholders, both internally 

and externally, is immensely important. CoCPIT 

team has therefore ensured that it is very transparent 

around its assumptions and also its limitations, so that 

people know when to rely on the results and in what 

areas they might have to make judgment decisions. 


6 

INVESTMENT AND RISK 

MANAGEMENT STRATEGIES 

IN A CHANGING REGULATORY 

FRAMEWORK JULIEN S. HALFON 

Head of Coverage 

P-Solve Risk Management Solutions 

The Solvency II Directive is now 

almost fully drafted and its content is relatively 

well defined. Notwithstanding, large-scale changes 

in regulation, even when they are known, can 

bring unforeseen consequences – positive or negative 

– often because of external factors such as 

market moves or changes in behaviors. The pension 

markets – especially in the UK and the Netherlands 

– have undergone similar changes over the last 

decade. The outcomes of these changes were not 

always what they were expected to be. 


The main question that arises in this article is: what 

happens to investment and risk management strategies 

when the regulatory framework changes? To answer 

that question, this article proposes to: 

• Review the changes that have taken place in terms of 

pension regulation in the UK and the Netherlands; 

• Understand the implications and challenges faced by 

UK and Dutch pension funds; 

• Analyze changes to investment and risk management 

strategies: How they have evolved and how pension 

funds have responded to regulatory changes. 

The following are the views of Julien S. Halfon and do not 

necessarily represent the views of P-Solve Investments 

Limited.

I. Regulatory changes 

Interestingly enough, the drivers behind all these 

changes have little to do with pensions. If anything, 

regulation has essentially been driven by specific 

events such as the collapse of the Enron and Maxwell 

Communication Corporation pension funds (which 

drove a lot of UK and US pension funds to change 

their investment framework) and by the attempt to 

create a single financial solvency framework (notably 

in the Netherlands) between Life insurance companies 

and pension funds. 

Early on, market specialists realized that pension funds 

behaved like captive Life insurance companies and had 

to be regulated in a similar way. Therefore, changes in 

regulatory frameworks have mostly, but not only, been 

driven by financial and solvency considerations rather 

than human resources considerations. 

Across the various markets, pension regulatory and 

accounting changes in the UK and the Netherlands 

(and also to some extent in Switzerland and the US) 

were aimed essentially at: 

• Applying fair value to both sides of the balance sheet; 

• Maintaining relatively high regulatory solvency and 

funding levels; 

• Increasing corporate sponsor responsibility; and 

• Increasing governance and education. 

These changes touched upon many aspects of pension 

funds’ financial management. It is worth noting 

that the European Solvency II Directive will also require 

similar changes. 

APPLYING FAIR VALUE TO ASSETS 

AND LIABILITIES 

Pension assets have traditionally been (and remain) 

invested in liquid financial securities such as bonds, 

equities and properties, because these assets are: 

• Easy to monitor; 

• Easy to periodically value; and 

• Easy to liquidate when required. 

Defined benefit pension liabilities are discounted using 

market-observable instruments rather than the fixed 

discount rates that were used for years: 

• Government bond yields or swap curves – for both 

interest rates and inflation – for solvency and funding 

purposes; and 

• AA corporate bond yields and breakeven inflation for 

accounting purposes. 

This has created a more consistent framework for both 

sides of the balance sheet. 

FUNDING LEVEL: THE CASE OF THE UK 

The changes in the UK funding framework responded to 

the need for pension funds to be fully funded or close to 

full funding at any point in time. If a pension fund is in 

deficit there is a requirement to get back to full funding 

within a relatively short period of time. Funding levels 

have become scheme-specific and depend on: 

• The size and nature of the liabilities; 

• The size of the deficit; and 

• The investment strategy. 

Given the existing funding deficits, most funds aim first 

for deficit reduction and try to achieve full funding in 

the medium to long term. 


What have been the main changes to the regulations? 

• Valuation assumptions now take into account the 

riskiness of asset allocation and the probability of 

achieving return objectives; 

• Pension scheme trustees maintain a strict fund 

governance between valuation dates; and 

• Additional contributions are requested from the 

sponsor and recovery periods were initially set at 

seven years (because of the Credit Crisis, this has 

been extended to ten to fifteen years). 

UK funding requirements have become more 

conservative. 

SOLVENCY LEVEL: THE CASE 

OF THE NETHERLANDS 

The same phenomenon occurred in the Netherlands, 

with a somewhat different outcome. 

In the New Financial Assessment Framework (nFTK) 

produced by the Dutch National Bank, pension fund 

solvency measures have also been made schemespecific 

and dependent on: 

• The size and nature of the liabilities; 

• The size of the deficit; and 

• The investment strategy. 

The objectives were to maintain high solvency levels 

and reduce the volatility of the net position by creating 

an asset buffer that protects solvency against sudden 

drops. In order to define the buffer, three solvency 

tests (minimum, solvency and continuity) have been 

established using two different methods. 

This approach is driven by the concept of economic 

shortfall risk. Broadly speaking, if a pension fund has a 

risky asset allocation, it should reserve more (i.e. set up 

a higher buffer) for solvency purposes as the probability 

to become insolvent increases. The specific constraint 

is that a pension fund should always remain funded 

over a twelve-month period on a 97.5 percentile 

probability basis. 

In practice, Dutch pension funds need about 25% - 

30% more assets than nominal liabilities to meet this 

requirement. The minimum funding requirement is 

105% on a nominal basis (i.e. without inflation) before 

triggering a reaction from the regulator. 

Solvency requirements have also become more 

conservative for Dutch pension funds. 


CORPORATE SPONSOR RESPONSIBILITY 

Corporate sponsors have been made more responsible 

from legal and financial standpoints in both countries. 

Their responsibilities and duties have been expanded: 

• Walking away from pension obligations has become 

very difficult; 

• Regulators in the UK and the Netherlands tend to 

frown upon hazardous investment strategies and 

penalize them; and 

• Additional contribution levels and recovery periods 

are set on the basis of the size of the deficit and the 

riskiness of the sponsor and its capacity to pay. 

Changes in regulation and accounting standards have 

significantly affected the sponsoring companies. 

Corporate pension obligations are also becoming more 

transparent and are being scrutinized internally and 

externally. For example, in 2003 Standard & Poor’s 

produced a list that they called the “dirty dozen”, 

which consisted of around ten European companies 

with pension problems that could negatively impact 

their credit rating. 

GOVERNANCE AND EDUCATION 

Increased governance is the last cornerstone of the 

regulatory changes. Pension funds have a duty to: 

• Draft investment principles and implementation 

guidelines; 

• Select investment and risk management advisors; 

• Monitor asset performance and the evolution of 

liabilities. 

Filing with bodies such as the Dutch National Bank or the 

(UK) Pension Regulator has also been made compulsory. 

Pension fund management bodies have therefore 

significantly invested in outside advice and education. 

They collaborate with sector associations, consulting 

firms, asset managers and investment banks offering 

advisory services and education programs to trustees 

and investment committees.

II. Implications & Challenges 

OVERALL IMPLICATIONS 

• The new regulations are moving towards one 

common solvency framework for all financial 

institutions, as is already the case in the Netherlands 

and in some Scandinavian countries. 

• The new regulations turned pensions into a 

problem scrutinized by equity and credit research 

analysts and by investors. Insurance companies and 

pension funds suddenly appeared on the radars of 

rating agencies, equity analysts and credit research 

analysts. 

• They also turned pensions into asset-liability and 

risk management problems, with long durations 

and potentially indexed liabilities. Before that, 

liabilities were fixed and pension funds had to 

manage assets in a way that was consistent with a 

solvent situation. 

• Regulators require pension funds to be or to 

remain well funded and solvent. The new regulations 

make it quite onerous to take risks in the 

Netherlands or to count on corporate sponsors in 

the UK. 

These four points demonstrate that changes in regulation 

have also changed the nature of the problem. From a 

human resources problem it has become a financial 

institution problem with investment, risk management, 

corporate finance and credit rating implications. 

CHALLENGES 

Fig. 53: Applying fair value to Assets and Liabilities 

UK Liability Valuation Drivers 

8.0% 

7.0% 

6.0% 

5.0% 

4.0% 

3.0% 

2.0% 

Dec-99 

Jun-00 Dec-00 Jun-01 

UK 30-yr Gilt Yields 

Source: Bloomberg. 

Dec-01 Jun-02 Dec-02 Jun-03 Dec-03 Jun-04 Dec-04 Jun-05 Dec-05 

UK 30-yr BE Inflation 

The market observable indicators now used to discount 

liabilities are volatile. Market volatility now 

affects both sides of the pension fund’s balance 

sheet as well as those of the corporate sponsors. 

As shown in Figure 53, corporate bond yields moved 

from 5% to 7% between December 2005 and 

December 2008. Assuming a liability duration (average 

liability cash flow maturity) of about 20 years, a 2% 

move in interest corresponds to a 40% change in the 

value of liabilities. 

Likewise, during the same period, inflation moved from 

3% to almost 4%, and dropped by 1 percentage point 

in 2007 and 2008. These changes are also in the range 

of 20% to 30% of the total amount of liabilities. 

Sterling AA Credit Spread over 15yr 

Jun-06 Dec-06 Jun-07 Dec-07 Jun-08 Dec-08 Jun-09 Dec-09 


By moving into a market-consistent framework for 

the valuation of assets and liabilities, the regulations 

brought large amounts of volatility to the pension funds 

and to the sponsors. This volatility is unrewarded as it 

applies to liabilities. 

The sponsors have become legally and financially 

responsible. They have to make their contributions 

according to the deficit and their financial capacity 

to help fund pension deficits. 

Figure 54 presents the additional pension fund 

contributions of 340 large public and private institutions 

in the UK. In the 90s these represented between 

GBP 500 million and 1 billion per quarter on average. 

Fig. 54: UK Corporate Sponsor Responsibility 

UK Sponsors’ Additional Contributions 

£ bn 

5.0 

4.0 

3.0 

2.0 

1.0 

0.0 

1992 

1993 

1994 

Special Contributions 


1995 

1996 

1997 

1998 

1999 

4 Quarter Moving Average 

340 public and private funds (including some DC contributions). 

Source: PPF & ONS. 

2000 

When the problem arose in 2001/2002, contributions 

started to increase significantly. When the new 

regulations came into force, the contribution levels 

increased to almost GBP 5 billion per quarter. 

In the mid 2000s sponsors could afford to pay high 

levels of contribution. However, since 2008, companies 

have had less cash flow, so they have not been able to 

contribute as much. Contributions decreased to GBP 

1.5 billion. The assumption that corporate sponsors 

would be able to help did not prove to be accurate. 

Funding levels have improved but assets need 

to increase and/or liabilities to decrease in order 

to return to full funding. 

2001 

2002 

2003 

2004 

2005 

2006 

2007 

2008 

2009

Fig. 55: UK defined-benefit pension Assets & Liabilities 

UK pension Assets & Liabilities 

Assets & Risk-Free Liabilities (£ m) 

1,500,000 

1,250,000 

1,000,000 

750,000 

500,000 

250,000 

0 

2001 

Corporate Schemes Corporate Scheme Liabilities 

Source: OECD, PPF, Watson Wyatt & P-Solve Estimates. 

2002 2003 2004 2005 2006 2007 2008 2009 

Figure 55 shows that corporate pension plans 

roughly hold about GBP 1.2 trn of liabilities for about 

GBP 1.0 trn of assets. The actuarial funding level is 

estimated at 85.9% (almost the highest in the last 

10 years). Interestingly enough, the shock in 2008 

was not as marked as people thought because at the 

same time as bond yields went up, inflation went down. 

This provided relief on the liability side. 

Dutch pension funds’ solvency ratios have decreased 

and appear to remain volatile. 

Solvency levels fell sharply as a result of the credit 

crisis and remain volatile. If anything, since the new 

Dutch regulations came into place, because of market 

issues, the funding level became even more volatile. 

Corporate Scheme Funding Level 

Fig. 56: Dutch defined-benefit pension Assets & Liabilities 

Dutch pension Assets & Liabilities 

Assets & Risk-Free Liabilities (€ m) 

1,000,000 

750,000 

500,000 

250,000 

0 

2001 

Assets Estimated Liabilities (Technical Provisions) 

Source: DNB & Dutch Insurers Association. 

100% 

80% 

60% 

40% 

20% 

0% 

As indicated in Figure 56, the funding level in 2007 

was almost 150%. At the end of December 2009, 

Dutch pension funds roughly held about EUR 610 bn 

of liabilities for about EUR 666 trn of assets. The overall 

solvency ratio went down to below to 97% or 98%, 

then went back up to 109% (with a lot of pension 

funds still below the limit of 105%). Complex solvency 

models did not take extreme events into account. 

These examples above show that, by themselves, 

regulatory changes did not bring all of the outcomes 

expected. They did not bring full funding or more solvent 

pension systems, in the UK or in the Netherlands. Nor 

did they bring a reduction in funding level volatilities. 

2002 2003 2004 2005 2006 2007 2008 2009 

Estimated funding ratio (%) 

150% 

100% 

50% 

0% 

Funding Level 

Funding Level 


III. Changes in Investment and 

Risk Management Strategies 

What actually happened is, in many ways, far more 

interesting. To respond to the challenges of the 

changing regulatory framework and environment, the 

investment and risk management strategies of defined 

benefit pension funds, as well as their benefit policies, 

have evolved dramatically over the last decade. 

Pension fund managers, trustees, investment committees, 

pension fund boards and corporate sponsors in 

both countries looked at the problem and understood 

that they had to change the way in which they thought, 

managed their risks and invested their funds. They also 

modified their benefit and contribution structures. 

Liability and Risk Management had become crucial 

words by the second half of this decade. 

APPROACH TO CHANGES IN INVESTMENT 

AND RISK MANAGEMENT STRATEGIES 

The traditional de-risking approach for pension funds 

(and their corporate sponsors) consisted in selling all 

or part of their liabilities to insurance companies (bulk 

annuity buyout). However, because of their prohibitive 

cost, “pension liability buyout” was out of the reach of 

most pension funds, which preferred to decompose the 

entire approach, starting with a risk budgeting exercise 

where risks would be identified and qualified. 

Figure 57 describes the various alternative stages to an 

annuity buyout. 

The process involves: 

• Reducing liabilities by closing down the defined 

benefit pension fund, limiting indexation or freezing 

the benefits. 

• Diversifying investments by moving out of listed 

local equities into new asset classes. 

• Hedging unrewarded liability risks such as interest 

rates and inflation and more recently longevity. 

The amount of risk reduced is similar to that of an 

annuity buyout, but the process is far less expensive. 

Fig. 57: Approach to changes in investment and Risk Management strategies 

Multiple instruments are used to achieve diversification and de-risking, starting with risk budgeting 

Increasing Risk 


Current Exposure 

Risk Budgeting 

Liability Reduction 


Interest Rate Hedging 

Inflation Hedging 

Longevity Hedging 

Insurance premium 

Increasing Cost 

Annuity buyout

RISK BUDGETING 

Pension funds and corporate sponsors are now 

developing pension risk budgets based on capital 

market and corporate finance models, in order to assess 

the implications of different investment and risk management 

strategies in terms of: 

• Risk-return profiles; 

• Solvency and funding; 

• Contributions; and 

• Corporate sponsor metrics (net debt, net income or 

free cash flow). 

Risk budgets tend to include: 

• Risk tolerance limits for both the pension fund and 

the sponsoring company; 

• Alternative investment and risk management strategies 

to be applied at varying levels of solvency 

through the use of trigger points; and 

• Differentiation in the treatment of rewarded and 

unrewarded risks. 

Fig. 58: Liability reduction 

Membership by member type - UK 

100% 

80% 

60% 

40% 

20% 

0% 

Actives 

Mar-06 

Source: PPF and P-Solve Estimates. 

100% 

80% 

60% 

40% 

20% 

0% 

Deferred 

Approaches to investment and risk management strategies 

are increasingly influenced by capital market and 

investment banking activities. 

LIABILITY REDUCTION 

Corporate sponsors started closing down their defined 

benefit funds to new members in the late 1990s, but 

accelerated in the late 2000s. They also changed or 

reduced the indexation policy. More recently, they also 

started freezing defined benefit pension funds. Closure 

to new members and an increase in deferred members 

has reduced the pension liabilities of many pension 

funds. Defined contribution plans have been set up as 

an alternative. 

Figure 58 shows the impact of closing down and 

freezing plans on the membership profile in the UK 

and the Netherlands. 

33% 33% 36% 36% 37% 

41% 42% 

Mar-07 Mar-08 Mar-09 

Mar-10 (E) 

Retirees 

42% 43% 44% 

26% 25% 22% 21% 19% 

Membership by member type - NL 

Actives 

Dec-05 

Source: DNB and P-Solve Estimates. 

14% 15% 15% 15% 15% 

49% 50% 51% 52% 54% 

37% 35% 34% 33% 31% 

Deferred 

Dec-06 Dec-07 Dec-08 

Dec-10 (E) 

Retirees 


The number of UK active members fell from 26% in 

March 2006 to 19% in March 2010. The number of 

deferred members rose from 41% in March 2006 to 44% 

in March 2010. The number of Dutch active members 

fell from 37% in December 2005 to 31% in December 

2009. The number of deferred members rose from 49% 

in December 2005 to 54% in December 2009. 

In five years there has been a 20% reduction in active 

members for both countries. 

Fig. 59: Diversification - The case of the UK 

Pension Funds’ Asset Allocation 

100% 

80% 

60% 

40% 

20% 

0% 

Dec-01 


Alt. 

7% 

Bonds 

26% 

Equities 

67% 

Dec-03 

UK Equities 

Overseas Equities 

Alt. 

5% 

Bonds 

31% 

Equities 

64% 

Alt. 

11% 

Bonds 

28% 

Equities 

61% 

Gilts 

Index-Linked Gilts 

Corporate Bonds 

DIVERSIFICATION 

The case of the UK 

There has been a shift in UK pension fund asset 

allocations. Most of the pension funds were massively 

invested in equities (70%) at the beginning of the 

decade. By the end of the decade, however, equities 

represent around 53% and the UK equities home bias 

has been reduced (less than 25% of total assets). Bond 

allocation increased significantly with allocations to 

index-linked and corporate bonds. However, the real 

change affected alternatives, which rose from less than 

5% to 16%. Pension funds invested in private equity, 

hedge funds, commodities, infrastructure and so on. 

Alt. 

11% 

Bonds 

29% 

Equities 

60% 

Alt. 

13% 

Bonds 

33% 

Equities 

54% 

Mar-06 Mar-07 Mar-08 Mar-09 

Mar-10 (E) 

Cash & Deposits 

Property 

Other Investments 

Alt. 

16% 

Bonds 

37% 

Equities 

47% 

Alt. 

15% 

Bonds 

32% 

Equities 

53%

Fig. 60: Diversification - The case of the Netherlands 

Pension Funds’ Asset Allocation 

100% 

80% 

60% 

40% 

20% 

0% 

Dec-01 

Alt. 

9% 

Bonds 

43% 

Source: DNB & OECD. 

Equities 

48% 

Dec-03 

Alt. 

11% 

Bonds 

43% 

Equities 

46% 

Euro Listed Equities 

Non-Euro Listed Equities 

Unlisted and Other Equities 

The case of the Netherlands 

Alt. 

8% 

Bonds 

43% 

Equities 

49% 

Dec-05 Dec-06 Dec-07 Dec-08 

Dec-09 

Eurozone Bonds 

Non-Euro Bonds 

Loans 

At the same time, Dutch pension funds started to 

change their asset allocation, drastically reducing Eurolisted 

equities in traditional markets to between 10% 

and 15% of the total. The remaining equity investments 

are now in private equity, hedge funds, or are directly 

invested in companies or infrastructure funds. Dutch 

pension funds have also been investing in derivatives 

such property derivatives. 

Fig. 61: Liability hedging 

Asset Matches liability 

structure? 

Mortality derivatives 

Bonds 

Swaps 

Alt. 

8% 

Bonds 

36% 

Equities 

56% 

Alt. 

11% 

Bonds 

38% 

Equities 

51% 

Cash & Deposits 

Property 

Other Investments 

LIABILITY HEDGING 

Alt. 

17% 

Bonds 

41% 

Equities 

42% 

Alt. 

11% 

Bonds 

27% 

Equities 

62% 

Risk management techniques and strategies developed 

on the capital markets are now regularly used by 

pension funds. Liability hedging generally means 

the use of bonds, interest rates and swaps to reduce 

unrewarded (liability) risk by matching assets to 

liabilities (i.e. liability-driven investment strategies). 

These instruments have been utilized by many pension 

funds in the Netherlands and in the UK. The choice 

between the instruments was dependent on: 

• liability structure; and 

• cost versus risk reduction benefits. 

Cost-effective? Currently used? 


LIABILITY-DRIVEN INVESTMENTS 

Fig. 62: Liability-driven investments 

LDI penetration in the UK 

40% 

30% 

20% 

10% 

0% 

17% 

31-Mar-06 

20% 

Source: PPF & P-Solve estimates. 

23% 

26% 

30% 

31-Mar-07 31-Mar-08 31-Mar-09 31-Mar-10 (E) 

Liability-driven investment strategies are being adopted 

by an ever-increasing number of pension funds. These 

strategies are based on using either long-term bonds or 

derivatives that will genuinely match the liabilities. 

Liability-driven investment penetration is estimated at 

30% in the UK (PPF & P-Solve) and at 70% in the 

Netherlands (Aegon Global Pensions). According to the 

same source, pension funds in the two countries are 

well over 50% hedged. 


LIABILITY BUYOUT MARKET 

Fig. 63: Liability buyout market 

Cumulative pension Liability buyouts in the UK 

Liabilities (GBP bn) 

20 

15 

10 

5 

0 

Source: LCP. 

2006 

2007 2008 2009 2010 - QI 

As may be expected, although the UK pension liability 

buyout market is growing, it has remained a limited and 

marginal phenomenon (about 1.5% of total corporate 

liabilities). According to the Dutch Insurers’ Association, 

pension contracts managed by insurance companies in 

the Netherlands represent up to 12% of total defined 

benefit pension liabilities. 

FIDUCIARY MANAGEMENT 

Finally, an increasing number of firms are offering 

fiduciary management services in the UK. A fiduciary 

manager is appointed by the pension fund to manage 

pension assets on its behalf, with varying degrees of 

freedom: 

• Traditional global / UK asset managers (BlackRock, 

SEI, Allianz Global Investors); 

• Dutch pension aggregators or risk management 

companies (Mn Services, Cardano, etc.); and 

• Investment consultants (Mercers, Hewitt, Towers 

Watson, P-Solve Asset Solutions). 

Although their approaches vary, they tend to focus 

on the execution of unconstrained asset allocations 

and the rapid implementation of investment and risk 

management decisions. They aim to: 

• Help pension funds access more complex products 

and strategies; 

• Help attain a high level of diversification; 

• Undertake rotation of asset classes; and 

• Provide a higher level of professionalism. 

In the Netherlands, according to Bureau Bosch, 77% 

of assets are managed by local and foreign fiduciary 

managers. Often, they also manage interest rate 

hedges.

Fig. 64: Dutch fiduciary management 

Assets and Hedges 

(€ m) 

750 

500 

250 

Conclusion 

- 

2001 

Dutch Fiduciary Managers 

Source: Bureau Bosch. 

WHAT LESSONS HAVE COMPANIES 

LEARNED? 

This article has demonstrated that regulatory 

changes do not always bring the expected outcomes. 

Nevertheless, they trigger changes in investment and 

risk management behavior, which take relatively little 

time to take shape. When pension funds are faced 

with such profound changes as risk-based capital 

and fair value for liabilities, plus governance, plus risk 

management, plus transparency, there is no other 

2002 2003 2004 2005 2006 2007 

2008 

Foreign Fiduciary Managers 

choice but to change themselves. In the UK and the 

Netherlands these changes are already in place. If the 

same pattern is applied to the insurance industry with 

Solvency II, the first major changes will be perceptible 

in 2017 at the latest. 

Education and governance are unquestionably key 

in terms of comprehending (and benefitting from) 

regulatory changes. Pension funds have made many 

efforts to change their behavior and to provide more 

professional advice, sometimes subcontracting their 

assets to someone else. Another fundamental change 

has been embedding their risk management approaches 

into investment strategies. 


7 

ERM AND ECONOMIC 

CAPITAL MODELS: 

THE A.M. BEST VIEW MILES TROTTER 

General Manager, 

Analytics Non-Life, A.M. Best 

This article proposes to explore 

ERM and economic capital from the rating agency 

view. In view of the current challenges, how do 

ERM and models fit into the A.M. Best rating 

process? 

This article begins with an overview of A.M. Best 

rating process, then it establishes the general background 

of ERM in the (re)insurance industry; last but 

not least it poses a certain number of questions concerning 

ERM and models in the rating process and 

their ongoing improvements. 

Responding to the insurance industry trends and 

practices, A.M. Best has been accompanying and 

supporting (re)insurance companies since 1899. 


I. The rating process 

Fig. 65: The rating process 

This above graph summarizes the rating process. 

Market Profile 

Operating Performance 

Balance Sheet Strength 

The A.M. Best rating process involves 3 components: 

• The balance sheet strength which is the bedrock 

for any rating; 

• The operating performance which supports the 

balance sheet strength through retained earnings; 

• The market profile which is the more subjective part 

of the rating, access to business is notably assessed 

in this part of the process. For instance, if a company 

has dedicated access to certain business and other 

companies cannot compete with it, this will protect 

its earnings and support its financial strength.

A.M. Best rating has six secure financial strength ratings 

which range from B+, the lowest and the first secure 

rating, to A++, the highest rating. The higher the rating 

level is, the more important the marketing profile is. The 

balance sheet strength and the operating performance 

are givens. The profile is the component which enables 

to distinguish between higher rating levels. 

THE BEST CAPITAL ADEQUACY RATIO (BCAR) 

A.M. Best has developed a proprietary capital model, 

the Best Capital Adequacy Ratio (BCAR). This ratio is 

used for the analysis of the balance sheet strength. 

This is a simple deterministic or factor-based model, 

i.e. a market surveillance type model that can be 

Fig. 66: A.M. Best’s Capital Adequacy Ratio (BCAR) 

+ (B1) Fixed-Income Securities 

+ (B2) Equity Securities 

+ (B3) Interest Rate 

+ (B4) Credit 

+ (B5) Loss Reserves 

+ (B6) Net Written Premium 

+ (B7) Off Balance Sheet Risks 

– Covariance 

AM Best’s Capital Adequacy Ratio (BCAR) 

rapidly populated with reasonably limited data. This 

benchmark contributes usefully to the rating process. 

The BCAR score expresses a percentage. As depicted in 

the tabular layout, the total adjusted capital is divided 

by the net required capital. The total adjusted capital 

corresponds to the available capital used to pay claims 

and the total required capital is obtained by applying 

factors to the different elements on the left of the 

below table. 

The score is obtained from the model and a one-toone 

relationship between total adjusted capital and net 

required capital is needed to achieve a secure rating 

with A.M. Best, which is B+ on its scale. 

Net required capital Total adjusted capital 

+ Shareholders funds 

+ Positive adjustments. Typically: 

• Discounting of reserves 

• Difference between market 

and book values 

• Value of In-Force Business (VIF) 

• Economic Reserves e.g. RfBs (1) 

• Hybrid Equity 

– Negative adjustments: 

• Deduction of one cat PML 

• DAC 

BCAR score = total adjusted capital / net required capital 

(1) Reserven für Beitragsrückerstattung - technical refund reserve under German life policies. 


II. Risk Management 

Risk Management tools and practices have advanced 

significantly in recent years. ERM has been the industry 

response to growing exposure to volatility in company 

earnings and capital. Reinsurance and insurance companies 

face emerging risks such as terrorist attacks, 

natural catastrophes such as hurricanes (four hurricane 

events in six weeks in 2004, three events in 2005) and 

more recently the credit crisis whose naughty child, 

the economic crisis, continues. The industry has been 

heading towards a higher risk profile. The (re)insurance 

companies have taken steps to manage their exposure, 

mitigate risks and preserve policyholder security. 

On the Non-Life side, catastrophe modeling is an 

example of this. 

ENTERPRISE RISK MANAGEMENT 

FRAMEWORK 

ERM has been a major addition to insurers’ vocabulary. 

Its foundations are rooted in traditional Risk 

Management practices and controls as shown in the 

pyramid. It constitutes a natural extension of these 

practices and only the enterprise view where the 

(re)insurers now identify, quantify and measure the 

different risks that expose the company on a holistic 

basis, is something new. 

ERM is based on three main areas: 

• Risk Culture: the company has to establish a risk 

culture that embeds risk awareness and accountability 

in daily operations; 

Fig. 67: Enterprise Risk Management 

framework 

Senior 


ERM 

and ICM* 

Capital 


Traditional Risk Management 

Practices and Controls 

* Establish Risk-aware Culture, including proper alignment 

of management incentives 

* Implement Improved Risk Identification and Management 

* Develop Sophisticated Risk Measurement Tools, e.g. ICM 

The pyramid illustrates the connection between ERM 

and (re)insurer’s traditional practices and controls. 


• Risk Identification and Management: the ability to 

identify key risks across an organization and set uniform 

controls and procedures to manage their impacts; 

• Risk Measurement: consists of using sophisticated 

tools and data collection to quantify risk, including 

the impact of correlation, economic conditions and 

extreme events. The results are then reported to senior 

management. 

ERM is very much about the aggregate view for the 

organization as a whole. 

Hopefully the hard lessons of the recent past will be 

learnt. Many governments have taken unprecedented 

steps to bail out, or in some cases not bail out, some of 

the largest institutions in the world. The current economic 

situation is very volatile and now the focus is shifting to 

sovereign debt issues and the possibility of a double dip 

recession. In any case, there is lag in foreclosures and it 

is likely that there will be another spike in foreclosures in 

2010/2011. 

Prior to 2008, many institutions felt that their ERM capital 

modeling processes were really good at getting them 

ahead of the risk curve for a more efficient use of capital. 

However, 2008 has definitely come as a lesson to the 

effect that tail events are not just scenarios to be modeled 

and statistically analyzed; they actually happen. 

DOES THAT MEAN THAT ERM IS A FAILURE? 

ERM at some institutions definitely failed, but this 

does not mean that ERM itself is a failure. The crisis 

impact on (re) insurers and other financial institutions 

demonstrates that there were gaps in their ERM 

programs, notably in the understanding of individual 

risks and risk correlation and dependencies. Insurers 

were used to having a high level of liquidity, a strong 

cash position. The crisis taught companies that cash 

management and liquidity analysis must be integrated 

into the Risk Management framework, including the 

stochastic analysis and modeling. And it is necessary 

to keep on challenging the risk inventory and looking 

for emerging risks. 

In the (re)insurance industry, some companies had some 

difficulties. The key points on the lessons they learned 

from the crisis could be summarized as follows: 

• Regularly recalibrate stress test assumptions and 

confirm their overall robustness; 

• Update assumptions concerning the dependency of 

securitized assets to reflect market conditions; 

• Ensure that the dependencies between the equity and 

the corporate credit risks are adequately modeled; 

• Enhance the financial acumen of the reviewing team 

to reflect the importance of judgment in reviewing 

model output.

Fig. 68: Why did insurance companies fail? 

It is worth listening to the experiences of institutions 

during the crisis. And one of the critical lessons to learn 

from the crisis is that the fundamentals matter. Specialists 

recommend back to basics: credit analysis, internal 

underwriting control, sound financial management, 

common sense and discipline. Most companies 

emphasize that modeling is important, but they have 

to use their judgment. ERM is a continuous process, 

a mindset, not simply a task to be performed. 

WHY DID INSURANCE COMPANIES FAIL? 

As shown in the pie chart above, the leading reason 

for failure is a combination of deficient loss reserves 

and inadequate pricing, followed by excessive growth. 

Should a company conclude from these elements that 

it must focus on reserving prudently, charging the right 

prices and may be not pursuing for too much growth? 

If ERM teaches anything, it is that the answer to that 

question is definitely no. 

Risk Management should cover the whole organization 

and any or all of these risks should be managed by a 

company if they are relevant to the company. Any of 

these reasons for failure could be attributed to a failure 

of Risk Management. 

It is worth noting that in the rating process Risk 

Management constitutes the common thread which 

links balance sheet strength, operating performance 

and business profile. Risk Management is used in the 

company’s strategic decision making process to define 

its profile but also in its financial management practices 

which dictate the sustainability of its performance, and 

ultimately its exposure to capital volatility. 

A failure in Risk Management 

Deficient loss reserves/Inadequate Pricing 

Rapid growth 

Alleged fraud 

Catastrophe losses 

Affiliate impairment 

Investment problems 

Miscellaneous 

Significant change in business 

Reinsurance failure 

Source: A.M. Best: 1969-2008 Impairment Review, 

Special Report, April 6, 2009. 

Deficient loss reserves, inadequate pricing, and rapid growth are the leading triggers. Investment and catastrophe losses play 

a much smaller role. This pie chart represents the impairment of the US Property and Casualty industry from 1969 to 2008. 

If a company has sound Risk Management and executes 

its strategy effectively, it will preserve and build its 

capital strength and perform successfully over the long 

term and is likely to have a strong rating. 

That is why A.M. Best does not produce separate 

ratings of companies for ERM. Risk Management is the 

element which supports the whole process. All these 

elements form a part of A.M. Best rating. 

Fig. 69: Financial Strength Ratings 

An opinion as to an insurer’s financial strength 

and ability to meet its ongoing obligations 

to policyholders 

Balance 

Sheet 

Strength 

Insurance Company 

Financial Strength 

Operating 

Performance 



Best’s 

Rating 

Business 

Profile 


Fig. 70: Traditional RM: silo approach 

Underwriting 


TRADITIONAL RISK MANAGEMENT 

As already mentioned, one has to explore traditional 

Risk Management in order to comprehend the recent 

developments of Enterprise Risk Management in the 

(re)insurance industry. Traditional Risk Management 

refers to the policies and procedures that companies 

have set to identify, quantify and measure each of their 

risks separately or individually. The main categories of 

risk for (re)insurers are: 

• Credit risk; 

• Market risk; 

• Underwriting risk; 

• Operational risk; 

• Strategic risk. 

Market 


The leading feature of traditional management is the 

separate assessment of risks which is often referred 

to as a silo approach. It differs from Enterprise Risk 

Management, which utilizes a holistic process. 

A.M. Best does not deem ERM to be the sole answer for 

(re)insurers. Each company may pursue its own approach. 

Thus, the traditional approach could be modified, using 

ERM concepts, to be more effective. A company has to 

understand the correlation between risks. If a company 

modifies its traditional approach in this way, it can still 

be considered strong by A.M. Best. 

Credit 


Traditional RM is typically implemented using a “silo” approach with little or no formal 

interaction, communication, or alignment among risk managers. 


Operational 


Strategic 


III. ERM in the rating process 

A.M. Best seeks to make an open-minded assessment 

of a company’s operating practices. The onus is on the 

company to show that its Risk Management process is 

tailored to its individual business needs, is flexible and 

adaptive and works. 

The crisis has proved that Chief Risk Officers, risk committees, 

corporate dashboards, corporate risk appetites 

and stochastic capital models are certainly all valid parts 

of a robust ERM program, but strong fundamental risk 

management in a sustainable business model comes in 

different shapes and sizes. 

Furthermore, the role of rating agencies such as 

A.M. Best is by no means to prescribe to companies 

what ERM should be or should not be. A.M. Best’s 

portfolio of rated companies is too diverse to impose 

such a measure.

Fig. 71: Catastrophe exposure in BCAR 

THE CATASTROPHE EXPOSURE 

Standard BCAR Two-Event Cat Stress Test 

EVENT – 1/100 Wind or 1/250 Earthquake 1 st EVENT – 1/100 Wind or 1/250 Earthquake 

ADJUSTMENT 

– Subtract after-tax net PML from Surplus 

As part of the rating assessment, A.M. Best uses its 

own model i.e. the BCAR model. For catastrophe risk 

management, BCAR-based tests are run. A 1/100 years 

windstorm after-tax net Probable Maximum Loss (PML) 

or 1/250 years earthquake after-tax net PML, whichever 

is the largest, is deducted from the surplus, the capital 

being therefore always viewed net of the largest 

PML. But following the events of 2005, a two-event 

catastrophe stress test was introduced to the model. 

As shown in Figure 71, in addition to the deduction of 

the largest after-tax net PML from the surplus for the 

first event, 40% of the pre-tax net PML is added to loss 

reserves, and 40% of the reinsurance recoverables are 

added to the credit risk. For the second event, there is 

a further deduction from surplus of the largest after-tax 

net PML. 

The second event for earthquake drops down to a 1 in 

100 event reflecting the likelihood that a second earthquake 

in the same place will be less severe. The idea is 

obviously to test capital for multiple events. The normal 

tolerance would be a 30-point drop in the BCAR from 

the minimum guideline. If a company’s BCAR decreases 

under this threshold, then there would be questions to 

ask the company. 

ADJUSTMENT 

– Subtract after-tax net PML from Surplus 

+ Add 40% of pre-tax net PML 

to Loss Reserves 

+ Add 40% of reinsurance recoverables 

to credit risk 

2 nd EVENT – 1/100 Wind or 1/100 Earthquake 

ADJUSTMENT 

– Reduce surplus by after-tax net PML 

Here are some of the questions that (re)insurers are 

likely to face from their A.M. Best analysts: 

• Describe the involvement of the Board of Senior Management 

within your risk management framework; 

• Has your organization established and communicated 

any risk management objectives to your employees, 

and other stakeholders? 

• How does your organization encourage good riskbased 

decision making? 

• What is your organization’s process for identifying 

and cataloging key risks? 

• What tools does your organization use to determine 

required capital? 

• How do you factor in correlation/dependency of 

individual risks? 

• How do you incorporate operational risk and strategic 

risk in your evaluation of required capital? 

• How are emerging risks identified and evaluated? 

• What are the top risks exposures your organization 

faces, and how are these risks managed? 

The questions focus on the usual areas: corporate 

culture, risk controls, risk models, particularly capital 

models, emerging risks. With reference to the last 

question shown in the list above, analysts will always 

ask for examples. If the example quoted is a mistake, 

the company will have to explain what it learnt from 

this mistake. Did it lead to any changes in their risk 

management process? 


Fig. 72: Risk Management and BCAR 

A.M. Best’s traditional approach 

BCAR 

HOW TO DETERMINE REQUIRED CAPITAL? 

As for the determination of required capital, it is based 

on an interactive dialogue comparing: 

• Company’s view of capital (e.g. ICM); 

• BCAR score and historical trends; 

• Holding company issues; 

• Other models or metrics. 

Weak Risk 


Strong Risk 


BCAR Guideline 

Low Exposure to earnings and 

Capital volatility 

High 

A.M. Best does not import results from companies’ 

internal models into its model but this does not mean 

A.M. Best ignores company views or analysis. These 

form an essential part of the decision making process 

for the rating. 

In many instances, companies reaching the same 

BCAR position may have different ratings based on the 

integration of other considerations, including financial 

management practices and operating elements which 

dictate performance sustainability and capital volatility. 

Risk Management is a key factor to determine the 

amount of capital required for a given rating. 

Companies with strong Risk Management and relatively 

low exposure to earnings and capital volatility, will have 

capital at the BCAR guideline level. For companies with 

weak Risk Management, it is difficult to maintain the 

fixed minimum of required capital. Therefore, they need 

to obtain a higher BCAR score, even if they maintain 

a low exposure to volatility. As exposure increases, the 

BCAR requirement also increases at a faster rate. 

Fig. 73: Risk Management and BCAR 

A.M. Best’s current approach 

BCAR 

What’s different… 

A.M. Best will consider 

allowing companies with 

STRONG Risk Management 

to maintain lower BCAR 

levels relative to 

the rating guideline 

Weak Risk 


Strong Risk 


BCAR Guideline 

Low Exposure to earnings and 

Capital volatility 

High 

The A.M. Best traditional view is depicted in the first chart above (Fig. 72). The vertical axis corresponds to the BCAR score, and the 

horizontal axis represents exposure to earnings and capital volatility. The second chart shows its current view. 


The A.M. Best view has evolved in recognition of 

increased sophistication in Risk Management practices 

in the industry. Lower BCAR scores relative to the 

guideline are considered for companies with a strong 

Risk Management. In doing this A.M. Best follows the 

improvements of the industry and adapts its rating 

process. 

THE INTERNAL CAPITAL MODELS (ICMs) 

APPROACH 

The A.M. Best approach to ICMs is similar to ERM in 

that it is handled as part of the rating process and not 

separately, although a supplementary review may be 

appropriate. This is a review, not an audit, with the 

objective of understanding the company’s modeling 

and comparing this with the BCAR. 

The BCAR is a deterministic model, built from historic 

analysis of industry level data. So for net premiums 

written, A.M. Best has historic data by class of business 

and arrives at a factor using that. With the company’s 

model, it will be based on its own risk data and maybe 

stochastic or may have stochastic elements to it. So the 

two cannot be mixed. Rather than trying to drop the 

company’s analysis into the BCAR model, A.M. Best 

simply tries to understand differences and why they are 

occurring. This sort of comparison and the conclusions 

that arise from it are very much a day-to-day part of 

the rating process.

To gain comfort in the integrity of the ICM process 

and results, A.M. Best seeks information on the following 

items: 

• Granularity of the model and data integrity; 

• Time horizons; 

• Risk metric (VaR… TvaR… CTE); 

• Assumptions and scenario testing; 

• Timeliness, availability and applicability of data; 

• Linkages to financial reporting and strategic 

planning; 

• Operational, strategic and emerging risks; 

• Disclosure (internal and external); 

• Internal and external “audit” findings; 

• Next steps in ICM development. 

A company may have a model that looks very good, but 

if it takes six months for the company to run it, then it is 

hardly going to be fully embedded within the company’s 

decision making. A.M. Best also wants to know what 

the company does about the more esoteric or difficult 

to measure areas of risk, whether operational, strategic 

or emerging risks and how widely disseminated the 

results are. Does the company see its model as the be 

all and end all of modeling for the company, or are 

there future developments expected? 

SOLVENCY II: THE FUTURE FOR RISK 

MANAGEMENT IN EUROPE 

Solvency II introduces risk-based regulation, the implications 

of which are summarized as follows: 

• This directive will lead to an increased focus on Risk 

Management and to an improvement of standards 

in Europe; 

• More companies are expected to use Internal Capital 

Models because the level of required capital is very 

likely going to increase. Therefore, they will have to 

seek approval and this will include demonstrating 

that the ICM is embedded in the organization. The 

time frame for obtaining approval is a tight one; 

• In the longer term, more and more companies 

will be using internal capital models that will result 

in strengthening risk management standards. 

The ORSA (Own Risk and Solvency Assessment) 

means that companies will have to incorporate risk 

into their strategic management. 

Conclusion 

A.M. Best will continue to look at ERM and ICM models 

within the existing rating process, which is an interactive 

one. 

Companies have to be practical in their approach, do 

what makes sense to them and advise A.M. Best why 

it works for them. 


8 

MACRO-ECONOMIC 

STANDPOINT MICHÈLE LACROIX 

Chief Investment Officer, SCOR 

This article presents the financial 

and economic difficulties caused by the current 

environment and the position adopted by SCOR 

in this context. The first section analyzes the crisis 

in all its phases; the second before examining 

SCOR’s view on the financial and economic issues 

involved in the second section. The last section 

highlights how the Enterprise Risk Management 

process is implemented into SCOR’s asset management 

decisions. 

I. The global economy is facing 

a critical transition phase 

To understand the current situation, it is necessary to 

explore the pre-crisis period, which will be referred to 

as the leveraging period. Saving rates decreased constantly 

during this period, which was itself split into 

3 phases: 

• Financing US economic growth: up until 2001, non-US 

residents financed the real economy, investing in 

corporate bonds and helping to implement real 

growth. 


• Financing US public debt: from 2001 to 2005, nonresidents 

invested in the US to finance the budget 

deficit and the war in Iraq. This period corresponds to 

what Alan Greenspan used to call a “conundrum”, 

i.e. a period with no inflation, low historical rates and 

strong incentives to borrow money. 

• Financing households: at this time, and as has often 

been the case in US history, savings rates were too 

low to finance consumer spending. In this context 

Alan Greenspan favored securitization, which 

enabled financial institutions to transfer their risks to 

other economic agents with fewer legal constraints. 

Securitization became the easiest way to finance 

local loans with non-US residents’ investments, which 

explains why the entire world was exposed to the US 

mortgage market and why the bursting of the US real 

estate bubble had global consequences. 

As shown in Figure 74, over the last three years the 

debt burden has simply moved from the private to the 

public sphere, staying at the same level and actually 

postponing global deleveraging of the economy.

Fig. 74: The deleveraging years 

Markets lost confidence 

in financial institutions 

➜ the first dip 

The end of the real estate rally clearly resulted in 

the collapse of the system: the deleveraging of US 

households came first, but the level of debt could not 

easily be lowered given the low level of inflation. 

The second step, i.e. the deleveraging of financial 

institutions, was more visible: the 2008 liquidity crisis 

showed how lack of confidence is critical in the financial 

markets and can create panic. Markets concerns were 

addressed by outstanding bailout programs, which 

resulted in the transfer of private debt to public debt; 

nevertheless the burden of the debt has not decreased, 

so the deleveraging process is far from over. 

Small increase in saving rates 

Subprime bubble Financial crisis Sovereign debt crisis 

➜ End of real estate rally 

➜ Deleveraging of consumers 

➜ Household have tried to 

reduce their level of debt 

but low inflation does not 

help such a move 

➜ Lehman’s and AIG’s bankruptcy 

and market turmoil 

➜ Deleveraging of financial institutions 

through balance sheet clean up 

➜ Markets concerns were calmed 

down thanks to outstanding 

bail-out programs 

➜ Transfer from private debt 

to public debt 

➜ Tightening spreads on corporate 

market vs higher spreads on 

government bonds debt 

Markets confidence seemed 

to be restored 

➜ the 2009 rally 

➜ The recovery seemed to be 

on tracks, but… 

➜ As long as the markets trusted 

the government, spreads were 

under control 

➜ Markets lost confidence in the 

sustainability of current public 

deficits in Europe 

➜ Local vs Paneuropean issuer 

➜ Lack of coordination 

➜ Time for a global answer 

Markets lost confidence 

in sustainability of public debt 

➜ the second dip? 

Debt burden flew from private to public but the deleveraging is not over 

The financial markets understood what was happening 

very quickly because, at the time of the bailout 

programs, corporate spreads on the markets were 

tightening, meaning that the markets once again trusted 

the corporate sector to cope with their debt burden. 

At the same time, government spreads were widening 

because everyone was beginning to understand that 

they had to pay for all this debt. 

Eventually the financial markets lost confidence in the 

public sector, wondering how such a level of debt could 

be sustained. 


Fig. 75: The global economy is facing a critical transition phase 

2007-2008: phase I 

Financial & economic 

collapse 

➜ Global recession 

➜ Liquidity crisis 

➜ Low interest rates combined 

with exploding credit spreads 

➜ Financial market disruptions 

➜ Fiscal and social deficits 

➜ Emergency interventions 

of Governments 

II. SCOR’s view on the financial 

and economic environment 

Figure 75 shows SCOR’s view on the economic and 

financial environment, split into 3 major phases. 

As already mentioned, the liquidity and financial crisis 

in 2007-2008 led to the global economic depression 

and to distressed financial markets: governments then 

decided to create historically huge amounts of money, 

which sooner or later will have to be dealt with. 

The years 2009-2011 may be referred to as the transition 

phase. This phase encompasses upcoming risks 

and throws up a number of questions: will interest rates 

decrease, remain stable or increase? Will the sovereign 

crisis extend to other countries after Greece, Spain and 

Portugal? When will the high volatility of exchange 

rates stop? Will social unrest develop if governments 

change in countries such as Greece? Will new regulations 

support the financial markets or not? 

These are major issues that need to be addressed. For 

its part, SCOR anticipates a return of inflation in the 

coming years, an increase in interest rates and a potential, 

limited increase in GDP, probably with currency 

adjustments. Investments will be chosen accordingly. 

Given that the economic and financial environment 

is so unstable, what choices are there for investors? 

It is worth noting that the financial markets rely on 

expectations: the issue is not whether or not there will 

be a double-digit inflation rate, but how long it will 

last. This is why the equity market is so bumpy in these 

uncertain times. 


2009-2011: phase II 

Key uncertainties 

in a stochastic world 

➜ Shape of recovery in front of 

global recession (L, V, W or √) 

➜ Uncertainty on monetary policies 

➜ Uncertainty on interest rates 

developments 

➜ Instability of exchange rates 

➜ Sovereign debt crisis 

➜ Tax and regulatory debates 

~ 2012 onwards: phase III 

Scenario 

of normalization 

➜ Stabilization of economic activity? 

With increased inflation? 

➜ Normalization of monetary 

policies? 

➜ Increase of nominal / real interest 

rates? Decrease of credit spreads? 

➜ Stabilization of equity markets? 

➜ Control of public deficits? 

➜ Stabilization of regulatory 

reforms? 

III. How are Enterprise Risk 

Management processes 

implemented into SCOR’s 

asset management decisions? 

To come back to Enterprise Risk Management, how do 

you deal with uncertainty? SCOR has tried to identify 

its main risks, and to decide whether the largest risk 

lies in equity, the return of inflation, the steepening 

of the yield curve or the effects of volatility on SCOR’s 

profitability. Once this has been done, a trade-off has 

to be made between protecting the value of assets and 

capturing the highest profitability: a certain amount 

of expected return must be relinquished in order to 

protect the balance sheet against the most significant 

risks. In times of high uncertainty, when it is difficult to 

anticipate upcoming developments, the safest course is 

to prepare for the worst-case scenario so that you can 

very quickly adapt to new disruptions. 

The following four case studies highlight SCOR’s 

handling of uncertainty within its investment policy.

CASE STUDY N°1: POSITION THE PORTFOLIO IN CASE OF INFLATION 

Fig. 76: Case study no.1: position the portfolio in case of inflation 

General concern: inflation is more damaging 

for reinsurers’ balance sheet than deflation 

➜ Accept the trade off between hedging 

and profitability 

➜ Inflation in developed countries may be postponed 

to 2012 

➜ Inflationary stimuli are in place 

• Unprecedented liquidity injection in the financial system will inevitably lead to surge in inflation 

and interest rates 

• Full sterilization of excess monetary creation likely to be impossible 

➜ What is currently missing in the market for further inflationary developments: 

• Shock in demand for credit more than supply 

• Deleveraging is not over yet 

• Growth expectations are weaker than a couple of months ago and commodities are directly impacted 

Inflation is more damaging for an insurer’s balance 

sheet than deflation. Reinsurers lose a lot of money in 

an inflation scenario, so they need to be prepared for 

inflation to rise. Even in the case of deflation, reinsurers 

should be positioned to tackle the return of inflation 

if it is identified as a major risk. Once again, there is a 

trade-off involved. This was the case in 2008, when the 

financial markets were geared more towards deflation 

than inflation. At that time SCOR decided to build an 

inflation-linked bond pocket in order to be prepared 

for a possible return of inflation. This bond pocket was 

increased up to EUR 1 billion in the first half of 2009, 

when inflation expectations were close to zero. SCOR 

bought securities with negative inflation breakevens, 

which means deflation expectations. The Group was 

absolutely out of step with the rest of the market, but 

it felt that it had to be protected against this risk, which 

is a major one for reinsurance balance sheets. Of course 

it is far less attractive to invest in linkers than in nominal 

bonds, but everything has a cost. 

SCOR also decided that, in order to be prepared for 

the return of inflation, it had to stay very liquid in its 

portfolio. However, being liquid does just not mean 

investing purely in cash. SCOR decided to implement 

SCOR’s answers 

Direct answer 

➜ The inflation linked bonds pocket has been 

increased up to EUR 1 bn in the first half of 2009 

when inflation expectations were close to zero 

➜ The carry is less attractive than that 

of nominal bonds 

➜ The portfolio is liquid to take advantage 

of any rise in interest rates 

Alternative answers 

➜ Diversify the portfolio to other asset classes that will 

benefit from inflation come-back (emerging markets, 

high yield, equities, real estate, commodities…) 

what is called the “rollover strategy”, which enables to 

receive cash from redemptions on a regular basis. Each 

quarter, the Group gets money back from redemptions, 

which can then be reinvested to prepare for a rise in 

interest rates. 

It now became clear that inflation expectations could 

be postponed for some time, whereas SCOR had 

initially expected sooner a return of inflation. This was 

the chance to diversify the portfolio into asset classes 

that could benefit from inflation, such as commodities, 

real estate, equity, emerging markets, high yield markets 

and so on. 

As indicated at the bottom of Figure 76, the inflationary 

stimuli are all in place. What is currently lacking 

in the market for this inflation to become a reality 

is that there is no demand for credits in the United 

States or Europe. This is why there is still no inflation, 

despite the high level of liquidity that has been pumped 

into the market over the past few years. So it is not 

a question of supply, but of demand. Nobody knows 

when it will happen, but as soon as demand returns, 

whether from industry, private companies or households, 

inflation could return to the markets. 


CASE STUDY N°2: STAY LIQUID 

Fig. 77: Case study no.2: stay liquid in our portfolio 

General concern: 

in case of markets collapse, cash is king 

➜ Accept the trade off between hedging 

and profitability 

➜ The panic has been pushed away thanks 

to outstanding bail-out programs 

➜ Current market are under pressure 

In the event of market collapse, cash held by companies 

is important. This was SCOR’s experience in 2007, 

2008 and the beginning of 2009, and it accepted 

the trade-off between hedging and profitability. As 

already mentioned, between early 2007 and the end 

of Q1 2009, SCOR accumulated up to EUR 4.6 billion in 

cash, which was an outstanding amount in relation to 

our balance sheet, representing more than 40% of the 

Group’s invested assets. So 40% of the invested assets 

were invested in an asset class with a close to zero 

return, which was highly costly for our investments, 

but this was the price of protecting the Group’s 

assets. There is, therefore, always a trade-off between 

the desire for profitability, which entails a risk, and the 

desire to stay as secure as possible, which will lead to 

less attractive remuneration. 

The Group has successfully protected the value of its 

balance sheets and its assets and has come through 

the financial crisis having suffered as little damage as 

possible. If you look at the net asset value of SCOR before 

and after the crisis, you will see that it has remained 

stable despite dividend distribution. This policy of “stay 

in cash” worked very well during the crisis. 

The panic has now subsided thanks to outstanding 

bailout programs. Since the counterparty risk involved 

was now easier to anticipate, SCOR began to look for 



Direct answer 

➜ SCOR accumulated cash early 2007 

up to EUR 4.6 bn in Q1 2009 

➜ The return of cash was close to zero 

and 40% of invested assets were in cash 

➜ SCOR successfully protected the value of 

its assets through the financial crisis (Cf NAV) 

Alternative answers 

➜ Since Q2 2009, strong focus on liquidity while 

seizing short term market opportunities: 

➜ Current liquidity is EUR 1.5 bn at end of Q1 2010 

➜ But roll-over strategy in place with EUR 4.4 bn 

of cash flows from the invested portfolio 

over the next 24 months 

➜ Short duration 

➜ SCOR progressively came back on selective credit, 

equities and other asset classes 

➜ Equities because of double dip risks 

➜ Currencies due to euro weakness 

➜ Rates due to sovereign debt concerns 

➜ Emerging markets because of risks of contagion… 

alternative ways to be able to face liquidity needs. SCOR 

lowered the amount of liquidity from EUR 4.6 billion 

at the end of Q1 2009 to EUR 1.5 billion at the end of 

Q1 2010, which means that over a period of one year, 

it placed EUR 3 billion of invested assets in more attractive 

asset classes. It did this very selectively, ensuring 

that the duration of our portfolio remained very short. 

It selected its high quality corporate bonds very carefully 

and had strong concerns about sovereign risks. 

The Group implemented this rollover strategy with a 

strong focus on the amount to be redeemed or received 

as coupons over the next 24 months. Instead of holding 

cash as it was done during the turmoil, SCOR is now 

in a position where it receives around EUR 4.5 billion 

cash that can be reinvested in the financial markets at 

the next available opportunity. 

The current situation is also a concern, with markets 

under pressure. Interest rates are stressed, equity values 

are stressed, foreign exchange rates are stressed, and 

something could explode in the coming months. The 

question is, what will burn out? Are companies going 

to face a contagious situation where one collapsed asset 

class infects the others, or will this be contained within a 

country, an area, or a specific asset class? Once again no 

one has the answer, so SCOR has decided to remain liquid 

and to continue to arbitrage in favour of liquid assets.

CASE STUDY N°3: REACT ACTIVELY TO THE SOVEREIGN CRISIS 

Fig. 78: Case study no.3: react actively to the sovereign crisis 

An early concern on sovereign risks 

➜ As extensively discussed during the July 2009 Investors’ Day, 

strong focus on sovereign risks 

SCOR regularly scorecards all its investments to preserve 

its financial strength 

4.00 

3.50 

3.00 

2.50 

2.00 

1.50 

1.00 

Investors’ Day – July 2009 

Watch List 

Sovereign risk scorecard: example of the November 2008 study 

South Korea 

Reduce 

Hungary 

Mexico 

0.50 

0.00 

Austria 

Sweden 

Switz 

-0.50 

Fin 

Hold Ger 

Watch List 

Norv 

-1.00 

Neth 

Greece 

Ireland 

Poland 

Italy Cz 

NZ 

SLV 

DK Port 

UK 

USA 

Can Esp Belg 

Jap Fr Lux 

Australia 

Singapore 

-1.50 -1.00 -0.50 0.00 0.50 1.00 1.50 2.00 2.50 

Macro-economic scoring 

Brazil 

Russia 

This case study is about reacting actively to the sovereign 

crisis. In Spring 2009, SCOR was already concerned 

about European public debt. In fact, SCOR started to 

become concerned at the end of 2008, at the time of 

the bailout program decisions. As you probably know, 

Denis Kessler is an economist, so his first reaction was 

to say, “It’s not possible to sustain such public debt. 

Keep an eye on sovereign names and on sovereign 

individual exposure to financial markets.” In view of 

this, the Group decided to create a kind of scorecard 

of all the big countries in which it was supposed to 

invest, and used this as a focal point at our quarterly 

Group Investment Committee meetings. SCOR had 

already decided at the end of 2008 to reduce its 

exposure to Spain and Italy, at a time when it no longer 

had any exposure to Greece. What you have to do in 

such difficult times is to be highly responsive. If you no 

longer have confidence in a country, you leave it. If the 

situation subsequently eases and you think that new 

elements should be taken into account, then you have 

to reassess the situation and you can reinvest from that 

point on. This is not a problem. With clients, it’s a very 

different story. When you want to run off an activity, 

you have to wait until the last client‘s policy has ended. 

When you deal with the financial markets, you can buy 

securities one day and sell them the next. 

South Africa 

Gvt CDS scoring COR regularly scorecards all it 

Source: SCOR calculation 

India 

China 

Iceland 

Turkey 

Financial 

Change/ 

Default risk 

Forbidden 

(High default 

risk) 

Exit 

(Significant 

default risk) 

No major 

change 

Hungary 

Greece 

Poland 

Czeck Rep. 

Monitor Italy 

(Intermediate Korea 

default risk) Portugal 

Hold (Hold 

default risk) 

Canada 

France 

Germany 

Japan 

Switzerland 

United States 

Rating 

downgrade 

risk 

Brazil 

Iceland 

Russia 

Belgium 

Denmark 

Ireland 

New Zealand 

Spain 

CDS 

spread risk 

India 

Turkey 

Mexico 

Slovakia 

Combination 

on the two 

risks 

China 

South Africa 

Australia 

United 

Kingdom 

Singapore 


Significant decrease in global exposure 

to euro peripheric countries… 

87 

288 

169 

248 

0 8 11 

5 3 

17 

Portugal Ireland Italy Greece Spain 

Total 31/12/2009 Total 19/05/2010 

… as well as exposure on government bonds 

& assimilated of euro peripheric countries 

105 

55 

0 0 3 7 

29 

0 0 0 

Portugal Ireland Italy Greece Spain 

Govies & assimilated 31/12/2009 

Govies & assimilated 19/05/2010 

SCOR was already addressing its concerns about the 

sovereign crisis in early 2009 and also did so at our 

Investors’ Day last year. As you can see from the top 

right-hand side of Figure 78, it still had some exposure at 

the end of 2009 to Italy, Spain and Portugal. By mid May, 

as shown in the bottom right-hand side above graph, 

SCOR had sold all the Portuguese, Irish and Greek debt 

and had considerably reduced exposure to Italy and Spain 

because we no longer had confidence in these risks. 

SCOR’s current concern is France. Should it continue to 

hold a larger amount of French vs. German bonds? Is the 

current weight of French government bonds sustainable? 

What is going to happen? This does not mean that the 

Group does not trust the French government. It simply 

means that there is a risk and that, in terms of ERM, 

it has to assess that risk, make a decision and put it 

into practice. So if it is decided that the bonds are 

safe, they will be kept. If it is decided that they are not 

sure whether or not they will be safe over the coming 

years, SCOR has to act accordingly. The trade-off is as 

follows: you sell French government bonds, you invest 

in German government bonds and you lose 50 basis 

points. Each time you arbitrage from risky to non-risky, 

you lose some kind of remuneration in one way or 

another, but you have to try to protect your assets. 


CASE STUDY N°4: DIVERSIFY THE PORTFOLIO 

Fig. 79: Case study no.4: diversify the portfolio 

General concern: high stress on financial markets 

makes it difficult to predict any evolution 

➜ Get ready for the worst 

➜ Implement strict investment controls 

process / stress test 

➜ Stay reactive 

The last case study addresses portfolio diversification. In 

order to be prepared for the worst-case scenario, SCOR 

has implemented a strict investment control process 

as well as stress tests. The way to cope with a stressed 

market is to enlarge the scope of your investments to 

ensure that you capture the maximum profitability, to 

be highly selective on each asset class, and to monitor 

every position closely and be ready to change your 

mind. Sometimes you are right, sometimes you are 

wrong. Maybe you should have kept this asset class 

because it eventually turned out to be safer and more 

attractive than you expected, but you have protected 

your assets and that was your main concern at the 

time of the decision. Once again, there is always a 

trade-off between profitability and sheltering your 

assets in safer harbours. 

IV. Positioning the investment 

portfolio to absorb large shocks 

while seizing short-term 

opportunities 

By way of illustration and to conclude this article, 

Figure 80 represents the profile of SCOR’s portfolio at 

Q1 2009 and at Q1 2010. The total invested assets 

are roughly the same. SCOR has placed a large part of 

its cash and investments in fixed income. At Q1 2009, 

40% of its portfolio was invested in cash and shortterm 

investments, which means investments of less than 

three months and government bonds with a maturity 

of up to one year. One year later, only 12% is in cash 

and short-term investments, and the fixed income has 

almost doubled. This reflects the inflation program that 

the Group started in the second quarter of 2009, when 

it thought that this crisis situation was over. 



➜ Enlarge the scope of investment to be sure 

to capture the maximum profitability 

➜ Be highly selective on each asset class 

➜ Monitor closely every position and be ready 

to change your mind 

➜ This remains the best way to capture profitability 

➜ Bet on various countries / asset classes / currencies enable to take opportunities 

and to be able to exit quickly one of the strategy in case of bad developments 

The Group lengthened the duration of the T-bills that 

were invested in cash into medium-term government 

bonds classified under fixed income, in order to 

benefit from the steepening of the yield curve with 

an additional 100 basis points. It was decided to move 

EUR 600 million of cash investments into corporate 

bonds in order to benefit from still very attractive 

spreads over government bonds. It was at the time of 

this long rally on the corporate market last year, when 

the market began once again to envisage economic 

growth, that the Group entered into this double dip 

risk scenario. Finally, EUR 150 million were invested 

in equities to take advantage of good opportunities 

on the markets. The rebalancing of these portfolios 

is still in process. The Group wants to adjust the level 

of cash to between EUR 1 and EUR 1.5 billion, and 

to stay highly selective on the fixed income and the 

equity markets. As already mentioned several times, 

this situation could change since SCOR is constantly 

adapting to new developments. 

What is not visible in these charts is the rebalancing 

SCOR has conducted on alternative investments – in 

Q1 2009 it had already requested the redemption of all 

the hedged funds it had in the portfolio, but this process 

takes time. These redemptions were initially requested 

at the end of 2008, but one year later it still had 

a lot of them in the portfolio because the redemption 

process was still underway. Incidentally, this gives a 

good indication of what an illiquid market can look 

like. At the end of 2009, the Group had EUR 100 million 

less in alternative investments, which were placed in 

Q1 2010 in assets classified as alternative investments, 

but which have nothing to do with hedge funds. The 

company also classifies commodities as alternative 

investments. It has notably invested a little in gold, raw 

materials, agricultural products and so on. These are 

the kinds of investment you select in times of inflation 

return risk, because this class will benefit from inflation.

Fig. 80: Positioning investment portfolio in order to absorb large shocks 

while seizing short-term opportunities 

Q1 2009 

Total invested assets (1) : EUR 11,360 million 


investments 

40% 

Impairments 

-5.4% +3.7% 

Real Estate 

3% 

Fixed Income 

48% 

Government & Assimilated 

25% 

Return on 

invested assets 

before impairments 

Equities 

6% Alternative 

Investments (2) 

3% 

Q1 2010 

Total invested assets (1) : EUR 12,656 million 


investments 

12% 

Real Estate 

3% 

Equities 

6% Alternative 

Investments (2) 

3% 

Structured products 

5% Structured products 

8% 

Corporate bonds 

12% 

Covered & Agency 

6% 

Return on 


after impairments 

Government & 

Assimilated 

38% 

Impairments 

Fixed Income 

76% 

Return on 


before impairments 

Covered & Agency 

7% 

Corporate bonds 

23% 

Return on 


after impairments 

-1.6% -0.5% +4.4% +3.9% 

(1) Excluding funds withheld. 

(2) Including hedge funds, infrastructure funds, private equity, commodities and non-listed equities. 

In other words, in times of uncertainty you have to be 

very careful in what you do, but you do not actually 

have to ban anything. Of course alternative investments 

are considered risky, but some selective assets within 

this classification could help to prevent the risks you 

are facing. 

The Group also tries to avoid investing in solutions 

that it does not fully understand. When it is offered 

new investments/solutions with which the Group is not 

familiar, it learns more about the asset class in question 

to ensure that it has ticked all the right boxes and that 

it knows exactly why this particular asset class is good 

for its balance sheet. SCOR makes sure that it knows 

why it should use this asset class over another one, so 

that it can face all the risks involved and be sure that it 

has made a good choice. Of course, this does not mean 

that at the end of the day the Group will have predicted 

what is going to happen with the investment. 

Once again, if there is no inflation in 2012 or 2013, 

or even up to 2020, it will have been costly to build 

this inflation-linked bond pocket, but the Group will be 

happy with it because the trade-off is infinitely preferable 

to being exposed to inflation with no protection. 

This is how SCOR has implemented ERM into its 

investment strategy. 


9 

HOW TO DEAL WITH THE IMPACT 

OF INFLATION ON PRICING 

AND RESERVING TONY NEGHAIWI 

Chief Pricing Actuary, SCOR Global P&C 

ERIC DAL MORO 

Chief Reserving Actuary, SCOR Switzerland 

This article addresses inflation, its 

origins and its impact on long-tail business lines. 

Based on SCOR’s experience, it shows how inflation 

could be integrated into pricing and reserving. 

I. Liability reserves 

and inflation 

For Liability business lines, inflation sources can broadly 

be classified into the following groups: 

Fig. 81: Sources of inflation for liability reserves 

Medical Price Index vs. All-item Price Index 

150 

140 

130 

120 

110 

100 

90 

80 

1996* 


• The Consumer Price Index (CPI), defined by the United 

States Bureau of Labor Statistics as “a measure of the 

average change over time in the prices paid by urban 

consumers for a market basket of consumer goods 

and services.” 

• Changes in regulation or legal interpretations in any 

jurisdiction. An illustration of this is the regulatory 

change in 1996 that completely modified reserve 

levels in French Non-Proportional Motor treaties. 

• Social changes: in fast-growing countries like China, 

the rehabilitation of victims is becoming increasingly 

expensive due to rapidly improving living standards. 

• Medical cost inflation: Figure 81 compares European 

medical cost inflation with the CPI in Europe. The 

medical cost rate is double that of the CPI. 

1997* 1998* 1999 2000 2001 2002 2003 2004 2005 

Medical price index All-item price index *Estimates. 

Source: Comité Européen des Assurances.

• Emerging risk inflation can be a source of inflation 

but is much more difficult to analyze (e.g. the possible 

impact of nanotechnology on future Liability claims). 

The precautionary principle can also be considered 

a source of inflation. This principle is being applied 

more and more and may therefore drive an increasing 

number of claims. This is illustrated by the recent 

eruption of the volcano in Iceland, for which some 

passengers would certainly have liked to file Liability 

claims. 

All of the above inflation sources may become drivers 

of superimposed inflation. Superimposed inflation is 

not an easy concept to grasp. In 2007, the Institute of 

Actuaries of Australia asked its members to come up 

with their best possible definitions for it. Two of these 

are set out below: 

• “The increase in claims costs that exceeds inflation, is 

unrelated to increases in claim frequency and ignores 

higher claims costs emanating from legislation changes 

and the like”. 

• “A measure of the increase in claim values in 

excess of the rate of wage inflation. Alternatively, 

a term dreamt up by actuaries to explain away their 

miscalculations!” 

Nowadays, insurers are facing these unpredictable 

factors and the question they are asking is: when 

will superimposed inflation start and how can it be 

stopped? 

The same survey by the Institute of Actuaries of Australia 

posed the question of how to stop superimposed 

inflation. Proposed solutions included: 

• Changing benefits in insurance policies. 

• Excluding/limiting the involvement of lawyers in 

the Liability system. Although this answer appears 

a good solution at first, it should be mentioned 

that most Liability claims are handled by insurers. 

Liability claims are only occasionally presented to a 

court, especially when the indemnities involved are 

small. In the U.S., however, the tort system, and class 

action suits in particular, allow courts to award large 

indemnity payouts. This can drive superimposed 

inflation and could be impacted by the limitation of 

lawyer involvement. 

• Changing the way in which claims are managed. 

In order to detect which claims are likely to lead to 

difficulties or fraud, it is helpful to implement a 

predictive model based on the claims data that is 

usually available when the claim is first entered into 

the IT system. On the basis of this model, the more 

complex claims can be handled by senior adjusters 

and the more simple cases by junior adjusters. This 

allocation of claims is a very important factor for 

insurers because for Liability, especially, moral hazard 

is a major driver. An optimal claims management 

system is therefore important to prevent claims 

inflation arising from undetected fraudulent claims. 


Fig. 82: Sources of inflation for liability reserves 

Example of the impact of inflation on Non-Proportional Liability 

Cover 

10 EURm 

Priority 

1 EURm 

Reinsurers are partially covered 

against inflation for the 

Non-Proportional business 

through indexation clauses. 

Figure 82 illustrates the potential impact of inflation 

on a Non-Proportional contract. For a treaty of 

EUR 10 m xs 1 m, one imagines that a claim reaches 

EUR 1.1 m. The reinsurance cost is EUR 0.1 m. 

With 10% inflation, the claim will reach EUR 1.2 m 

and the cost to the reinsurer will be 100% higher. 

So, even though stabilization clauses in Non-Proportional 

contracts exist in many markets, inflation can have a 

dramatic effect for reinsurers. 

NON-PROPORTIONAL MOTOR BUSINESS 

IN FRANCE 

Treaty 10 EURm xs 1 EURm 

Claim 

1.1 EURm 

Cost 

reinsurance 

0.1 EURm 

10% inflation 

French Motor Non-Proportional treaties provide an 

important example for examining legal changes and 

Claim 

1.2 EURm 

Cost 

reinsurance 

0.2 EURm 

Cost 

reinsurance 

+100% 

their impact on loss development triangles. In 1996, 

a legal change was introduced to the calculation of 

pension indemnities payable to bodily injury victims. 

This led to: 

• The introduction of more conservative mortality tables; 

• The application of more conservative discount rates. 

Insurers had five years to implement the new regulations, 

with a retroactive effect on open claims. 

Below is a triangle of chain-ladder coefficients for 

cumulative incurred losses on French Motor Non- 

Proportional lines. 

Fig. 83: Motor Non-Proportional in France 

Triangle of incurred chain-ladder coefficients for Motor Non-Proportional in France 

UWY N N+1 N+2 N+3 N+4 N+5 N+6 N+7 N+8 N+9 N+10 N+11 N+12 N+13 N+14 N+15 N+16 N+17 N+18 

1980 0.13 14.53 1.74 1.05 1.02 1.11 1.01 0.97 1.03 1.04 0.99 1.00 1.18 1.01 1.05 1.28 0.80 1.03 

1981 0.00 4.97 0.99 1.04 1.21 1.16 0.86 1.04 1.02 0.92 1.03 0.94 0.97 0.98 1.26 0.93 1.10 0.97 

1982 0.78 0.91 1.07 1.18 1.11 1.08 1.02 0.97 1.08 1.06 1.04 1.00 1.05 1.40 0.71 1.02 1.06 1.02 

1983 1.07 1.14 1.03 1.01 0.95 0.94 0.99 1.02 1.04 1.02 1.05 1.06 1.43 0.70 1.05 1.11 0.86 

1984 1.39 1.14 1.13 1.16 1.03 1.00 0.98 1.02 0.96 0.98 1.05 1.30 0.92 1.06 1.01 1.01 

1985 1.54 1.29 1.09 0.94 1.03 1.06 0.86 1.00 1.00 1.01 1.28 0.95 1.03 0.95 1.04 

1986 1.37 0.91 1.03 1.05 0.97 1.02 1.02 1.01 1.09 1.29 0.88 1.08 1.02 0.89 

1987 1.37 1.01 1.02 1.02 1.08 0.94 1.10 1.00 1.29 1.09 1.10 0.97 0.97 

1988 1.55 0.96 0.99 0.94 1.11 0.99 1.01 1.32 0.97 1.07 1.04 0.97 

1989 1.16 0.92 1.14 0.97 0.99 1.06 1.39 1.01 1.22 0.99 0.98 

1990 NA 1.34 1.11 1.01 1.11 1.04 1.43 0.91 1.16 0.99 0.97 

1991 1.41 0.87 1.19 1.01 1.34 1.02 1.08 0.98 1.00 

1992 1.29 1.13 1.10 1.48 0.94 1.10 0.94 1.11 

1993 1.21 1.03 1.39 0.91 1.05 0.98 0.99 

1994 1.15 1.59 1.17 1.13 0.98 0.96 

1995 4.39 1.54 1.16 0.96 1.04 

1996 1.40 1.28 1.20 0.96 

1997 1.35 1.12 1.01 

1998 0.96 1.13 

1999 1.10 

The impact of the change of regulation can be clearly seen on the chain-ladder coefficients. 

2000 

How to deal with such distortions? 

82 - November 2010 - SCOR

In the triangle (Figure 83), the impact of the regulatory 

change is visible in the diagonals of 1996, 1997 and 

1998. For 1996, instead of having lower chain-ladder 

coefficients (between 1.2 and 1.3), the coefficient 

reaches 4.3. This high coefficient is due to cedants 

adapting their bodily injury reserves to the new 

regulation and thus reporting higher incurred values. 

This is typical of a retroactive change in regulation: 

a huge effect is visible in the years following initial 

implementation. 

Following this example, the way how SCOR deals with 

such distortions will be presented. 

TREATMENT OF INFLATION IN RESERVING 

SCOR has two main methods for dealing with reserving 

inflation: 

• Stress testing: this consists of incorporating a loading 

for future unpredictable inflation into ultimate loss 

ratios, based on pricing information for the most 

recent underwriting years. Putting up reserves in this 

way can be costly and, as a result, requires a strong 

balance sheet. 

• Correction of triangles for embedded inflation: the 

correction of triangles for inflation consists of removing 

the inflation effects embedded in the triangles. 

This correction may involve removing chain-ladder 

coefficient diagonals affected by legal interpretation 

changes (e.g. French Motor Non-Proportional). 

It may also involve correcting the whole triangle 

for some base inflation (e.g. TME rate in France). 

The methods set out below provide two ways in which 

to automatically estimate this embedded inflation: 

- Separation method of Verbeek-Taylor 

(G. Taylor 1986), 

- Generalized Linear Model (GLM - England & 

Verrall 2002). 

The modified triangle is then projected and the final ultimate 

loss is calculated, factoring in future inflation. Future 

inflation estimates can be based on exposure and frequency 

determinants (if possible). In any case, the future 

claims inflation assumption remains judgmental. 

Stress tests 

In order to factor inflation into the reserves, the ultimate 

loss ratio used for reserving purposes is split into 

three layers: 

• The first layer consists of the expected claims level 

with no inflation assumptions; 

• The second layer consists of the additional loss ratio 

points stemming from the original price inflation 

assumption. This information comes from pricing 

actuaries; 

• The third layer represents the loss ratio points 

additional to the second layer, stemming from 

stressed inflation. 

Figure 84 illustrates the three layers concept: 

The first layer, claims without inflation, represents an 

80% loss ratio. 

The second layer factors in 5% inflation over a claim 

duration of 2 years: this would represent a further 8% 

on the loss ratio. 

Finally, the third layer factors in an additional 2% 

inflation, which leads to an additional 3%. This is the 

cost of inflation unpredictability. 

Fig. 84: Treatment of inflation in reserving 

Stress-test – Applies to recent Underwriting Years, based on pricing information 

1 st layer 

2 nd layer 

3 rd layer 

Pricing Ultimate 

Loss Ratio 

without inflation 

Δ Ultimate Loss Ratio 

when original price with 5% inflation 

Δ Ultimate Loss Ratio 

for higher inflation (7%) 

CoC Expenses 10% 

Claims without 

inflation 

80% 

8% 

3% 

Reserving ULR 101% 

Note on the calculation below: 

We assume a claim duration of 2 years. 

= 80% [(1+5%) 2 –1] 

= 80% [(1+7%) 2 – (1+5%) 2 ] 


Triangle correction for embedded inflation 

The correction of triangles for embedded inflation can 

be conducted cell by cell on applying the following 

factor: 1 

(1+x) i+j 

where: 

• x is the average estimated rate of inflation in the past, 

• i relates to underwriting year, 

• j relates to development years. 

Details about this method can be found in England 

and Verrall (2002). 

Once the triangles have been corrected by the factors 

above, a standard projection can be made to estimate 

uninflated IBNRs (e.g. using the chain-ladder method). 

The challenge is then to estimate the reinflated IBNRs 

taking future inflation factors into account. 

Usually, future inflation is determined by future exposure 

and future frequency. However, unlike Property 

insurance, the determinants of Liability exposures and 

frequency are not straightforward to define. 

Fig. 85: Treatment of inflation in reserving 

Worker’s Compensation loss ratio vs. unemployment 

140% 

120% 

100% 

80% 

60% 

40% 

1979 


Workers comp loss ratio (LHS) Unemployment rate (RHS) 

Source: J.P. Morgan, APRA, ISC. 

As examples of Liability determinants for inflation, in 2009 

the Insurance Services Office in the US proposed the following 

measures for some Liability business lines: 

• Contractors’ Liability exposure: the average hourly 

earnings of construction workers would seem to be 

correlated to inflation. 

• Manufacturers’ Product Liability exposure: future 

sales would be a good determinant even though 

they remain uncertain. 

• Frequency: unemployment rates and interest rates 

may be used. In Australia, for instance, the unemployment 

rate and the ultimate loss ratio of Workers’ 

Compensation insurance seem to be correlated, as 

shown in Figure 85. Therefore, using the predicted 

unemployment rate, it should be possible to estimate 

the future ultimate loss ratio, which would integrate 

some elements of superimposed inflation. 

To conclude, in the context of superimposed inflation 

reappearing and this future inflation being different 

to past inflation, insurers already have many tools to 

deal with this new environment, as described above. 

However, it should be noted that the impact of inflation 

on reserves can be huge; fortunately, modeling 

solutions exist to deal with it. 

1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 

12% 

10% 

8% 

6% 

4% 

2% 

0%

II. Pricing and inflation 

Projecting the cost of ultimate loss, and identifying and 

dealing with its drivers, is a common objective in both 

pricing and reserving. It also constitutes perhaps the 

most critical issue not just for these actuarial functions 

but also for the health and viability of the insurance 

and reinsurance industry as a whole. 

Fig. 86: Should Pricing/Reserving be concerned? Absolutely 

Reasons for US P/C Insurer Impairments, 1969-2008 

7% 

8% 

9% 

8% 

4% 4% 

8% 

14% 

38% 

As highlighted by Figure 86, between 1969 and 2008, 

38% of US P&C defaults were attributed to deficient 

loss reserves and inadequate pricing, by far the largest 

cause of insurer defaults. One may also argue that 

there is a strong link between pricing/reserving and 

some of the other causes on the chart, such as rapid 

growth. Unsurprisingly, frequent defaults usually follow 

turbulent times. 

Deficient loss reserves/Inadequate Pricing 

Rapid growth 

Alleged fraud 

Catastrophe losses 

Affiliate impairment 

Investment problems 

Miscellaneous 

Significant change in business 

Reinsurance failure 

Source: A.M. Best, 1969-2008 Impairment Review, Special Report, April 6, 2008. 

Deficient Loss Reserves and Inadequate Pricing are the leading cause of insurer impairments, underscoring the importance of discipline. 

Investment Catastrophe Losses play a much smaller role. 


Fig. 87: Some historical perspective on inflation 

Inflation is not double-digit now as it was in 1974-81 

Annual 

Inflation Rates 

14% 

12% 

10% 

8% 

6% 

4% 

2% 

0% 

-2% 

70 

60 

50 

40 

30 

20 

10 

0 

1969 8 

1970 

1971 

1972 

1973 

1974 

1975 

1976 

1977 

1978 

1979 

1980 

1981 

1982 

1983 

1984 

1985 

1986 

1987 

1988 

1989 

1990 

1991 

1992 

1993 

1994 

1995 

1996 

1997 

1998 

1999 

2000 

2001 

2002 

2003 

2004 

2005 

2006 

2007 

2008 

2009 

2010F 

Source: US Department of Labor, Bureau of Labor Statistics; Blue Chip Economic Indicators (4/2010 issue) (2010 forecast). 

If you started work in the insurance industry 27 years ago, you never experienced annual inflation as high as 6%. 

Fig. 88: Some historical perspective on Non-Life companies’ impairment 

P&C Insurer Impairments, 1969-2009F (forecast) 

15 

12 

7 

11 

9 

34 

9 

13 

12 

19 

9 

16 

14 

13 

36 

49 

31 

34 

50 

48 

Source: Insurance Information Institute. 

The number of imparments varies significantly over the P&C Insurance Cycle, with peaks occuring well into hard markets. 

Companies are currently facing economically turbulent 

times, where high inflation is one potential scenario as 

the world economy comes out of the crisis. This would 

have serious consequences on (re)insurance industry, 

especially on long-tail lines such as Liability. Even more 

serious is the fact that most of companies in the industry 

have not worked in a period of high inflation. 


55 

60 

58 

41 

29 

16 

12 

31 

18 

19 

5 of the 11 are Florida companies 

(1 of these 5 is a title insurer) 

49 

50 

47 

35 

18 

14 

15 

5 

7 

1970 

1971 

1972 

1973 

1974 

1975 

1976 

1977 

1978 

1979 

1980 

1981 

1982 

1983 

1984 

1985 

1986 

1987 

1988 

1989 

1990 

1991 

1992 

1993 

1994 

1995 

1996 

1997 

1998 

1999 

2000 

2001 

2002 

2003 

2004 

2005 

2006 

2007 

2008 

2009F 

11 

However, “economic inflation” in the form of the CPI is 

not the only driver of loss costs. In pricing, one ultimate 

loss framework is composed of frequency (number 

of claims), and severity (the cost per claim). This is an 

effective and simple framework, and helps to identify 

loss cost drivers. Inflation is mainly linked as a driver of 

severity, but it is only one of many. To illustrate this with 

real statistics, let us take a look at Liability in the UK.

EXAMPLE FROM THE UK MARKET 

Figure 89 comparing the average severity of a UK Liability 

portfolio to the CPI and wage inflation index clearly 

shows that the trend in claims costs far outstrips the 

“economic” inflation indices: superimposed inflation 

is real. In the UK Liability market, many past studies 

indicated that the overall loss trend – the average loss per 

claim – had grown by around 10% per year in a period 

where the underlying Consumer Price Index inflation 

was in the range of 2-3%. It is worth noting that this 

superimposed inflation does not necessarily need to 

correlate closely to the Consumer Price Index. 

Fig. 89: UK Example – Long-term relationship 

Severity Indices vs. CPI 

Index (1990 = 1.00) 

6.00 

5.00 

4.00 

3.00 

2.00 

1.00 

Median 

ACPC 

Flat increase 

2-yr av median 

4-yr av median 

CPI 

wage 

Economic Inflation is a broad basket of items, some of 

which are more relevant to loss inflation than others, 

such as health care costs, wage costs, and repair costs. 

Furthermore, there are many other factors affecting 

claim costs that are not directly related to “economic” 

inflation. To name just a few: changes in laws and 

regulations (such as the admission or non-admission 

of pain and suffering); benefit schemes (such as the 

Ogden tables in the UK); social attitudes (e.g. tendency 

to file claims). These additional drivers of loss inflation 

are not necessarily correlated to economic inflation. 

Figure 90 and Figure 91 correlate the one-year loss 

trends to the CPI and wage inflation index. 

0.00 

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 

Year 

Source: SCOR data. 

Severity Trends have far outstripped general inflation; year-on-year severity changes too erratic to provide reliable trends; long-term trends 

more stable, but require evaluation of past sources of superimposed inflation and future expectations. 


Fig. 90: UK Ex. – Short-term relationship 

Severity comparison to wage inflation 

2-yr median 

50% 

40% 

30% 

20% 

10% 

0% 

0% 

-10% 

1% 2% 3% 4% 5% 6% 7% 

-20% 

Source: SCOR data. Wage Inflation Source: SCOR data. 

Clearly, it is hard to find this direct correlation, and in 

fact the straightforward correlation to the wage index 

misleadingly implies a negative correlation, which 

practically, does not make much sense. This is not just 

due to the other drivers of the severity trend, but also 

to the random nature of claims severity, and the timing 

of claims. There are many “dates” affecting claims 

severity other than the date of policy inception: date 

of accident, date of settlement, date of payment, etc. 

When picturing the relationship between the severity 

index and wage inflation in the longer term, some trends 

become visible. As shown in Figure 92, loss inflation and 

wage inflation have indicated a positive correlation from 

1998 to 2009. Nonetheless, one can clearly see that 

factors other than inflation affect claims severity. There 

was a stronger relationship between the 4-year severity 

index and wage inflation, but there are also many other 

Fig. 92: UK Ex. – Long-term relationship 

4-yr median loss severity index against wage 

inflation index 

Loss Inflation Index 

4.00 

3.50 

3.00 

2.50 

2.00 

1.50 

1.00 

1.00 1.20 1.40 1.60 1.80 2.00 2.20 2.40 


Wage Inflation Index 


From 1998 to 2009, Loss and Wage 

Inflation have tracked and severely, 

but same trend was not apparent 

in period of four previous years 

2-yr median 

Fig. 91: UK Ex. – Short-term relationship 

Severity comparison to CPI 

50% 

40% 

30% 

20% 

10% 

0% 

0% 

-10% 

1% 1% 2% 2% 3% 3% 4% 4% 5% 

-20% 

elements to deal with. The general trend is shown in the 

data, but the relationship is weak. 

The purpose herein is not to try to minimize the effect 

of inflation, but to highlight the fact that there are other 

factors involved, and most importantly that a quick 

look at the statistics (especially when looking at short 

periods) may misleadingly imply the non-existence of 

such a relationship. 

MANAGING THE TREATMENT 

OF INFLATION IN PRICING 

CPI 

Loss inflation has more of an effect on long-tail lines 

such as Liability. In this example, if the inflation estimate 

is changed by 1%, the effect on the loss cost would 

reach 8% undiscounted and 7% discounted. 

Loss Inflation 4-yr median 

Fig. 93: UK Ex. – Long-term relationship 

4-yr median loss severity to wage inflation 

20% 

18% 

16% 

14% 

12% 

10% 

8% 

6% 

4% 

2% 

0% 

2.0% 2.5% 3.0% 3.5% 4.0% 4.5% 5.0% 


Wage Inflation

Fig. 94: Effect of Loss Inflation and Discounting on Pricing 

Cumulative Payments - base 100 

120 

100 

80 

60 

40 

20 

0 

1 3 5 7 9 11 13 15 17 19 21 23 25 

Year of Maturity 

10% Inflation 10% Inflation & 4% Discount 11% Inflation & 5% Discount 

11% Inflation 11% Inflation & 4% Discount 


Known inflation can be priced in and has a more 

leveraged effect on long-tail Lines of Business, but 

this leveraging effect is reduced after discounting for 

investment income. When and if inflation and interest 

rates move in tandem, price is not materially affected, 

but that is a big IF. 

On an excess of loss basis, however, the effect is more 

dramatic. The leveraging effect makes loss inflation 

(economic and superimposed) much more dangerous. 

There are two effects to consider here. First, for losses 

already above a certain threshold, the increase caused 

by inflation will affect the part above the threshold; 

for example, if a claim is valued at EUR 110 with a 

threshold of EUR 100 (making excess of loss EUR 100), 

the increase of 1% from EUR 110 to EUR 111 will affect 

the excess layer by 10%! From 10 to 11. Furthermore, 

one should consider losses that were formerly below 

the threshold but now exceed it due to inflation. 

OTHER POTENTIAL DANGERS 

FROM THE ECONOMIC CRISIS 

One should also consider other elements that do not 

really constitute inflation per se but rather are related 

to economic scenarios. Some of these may affect 

frequency (or the number of claims) rather than severity. 

Although these are currently purely speculative, one 

cannot dismiss them and should keep an eye on them in 

order to protect the viability of (re)insurance industry. 

Governments are accumulating more and more debt and 

there will come a point when they need to deal with it. 

Pattern Effect: 1% inflation Index Effect 

10% Inflation - Undiscounted 100 0% 

10% Inflation - Discounted at 4% 74 -26% 

11% Inflation 108 8% 

11% Inflation - Discounted at 4% 79 7% 

11% Inflation - Discounted at 5% 74 0% 

Inflation is one possible solution in this regard. Pension 

or unemployment reforms are another; however, while 

they may not lead directly to increased losses, the people 

affected may be inclined to find more reasons to file 

claims. Another possible solution is to cut the budgets 

of certain government departments, thereby increasing 

pressure to find alternative sources of revenue. Insurance 

is one potentially tempting source, with the possible 

option of allocating more cost burdens (hospitalization 

for example) to the insurance sector. 

Furthermore, high unemployment rates and low or 

reduced income could tempt people to turn to insurance 

as an alternative source of revenue. The increased filing 

of injury claims, whether legitimate or fraudulent, and 

the increase in arson and theft, are all examples of such 

dangers. Economic pressures, which could negatively 

impact safety and maintenance standards and therefore 

cause more accidents, are another example. 

The most important things to remember when pricing 

are to stay on the alert and to take whatever 

materializes in the projections into consideration. 

There are, however, other ways to manage these 

risks, such as: 

• Holding more capital to cover the increase; 

• Using more inflation-indexed investments; 

• Integrating indexation clauses; 

• Improving the quality of data used for reserving 

and pricing; 

• Involving regulators in the discussion; 

• Using Asset Liability Management, a key technique 

for analyzing the combined effect on assets and 

liabilities. 


10 

THE FINANCIAL MARKETS VIEW 

ON THE RISK/REWARD STRATEGIES 

IN THE (RE)INSURANCE 

INDUSTRY MARCO CIRCELLI 

Head of Group Corporate Finance & Financial Communications, SCOR 

This article presents the view 

of the financial markets (analysts and investors) 

with regard to the risk/reward strategies of 

(re)insurance companies. 

I. The investor’s reassessment 

of a reinsurer’s key strategic 

choices during the financial crisis 

Reinsurance companies are profiled on the basis of the 

key choices made around their strategic axes. These 

strategic axes enable us to determine the risk/reward 

profile of a (re)insurance company. 

As shown in Figure 95, which is taken from SCOR’s 

strategic plan, there are four key strategic cornerstones: 

Portfolio Management, Capital Management, 

Franchise Management and Cycle and Growth 

Management. Within these four strategic cornerstones 

there are four variables, to wit, volume, solvency, 

risk-based capital and profit. These variables have an 

impact on the Group as a whole, in terms of franchise, 

profitability, risk diversification and financial rating. 

Along these strategic axes, (re)insurance companies 

have pursued three key business models since the 

WTC in 2001. These three business models are called 


business models A, B, and C by default. They are based 

on the strategic cornerstones mentioned previously: 

• Business model A focuses more on traditional reinsurance 

business, and on high business diversification. 

It follows a cautious asset management approach 

and a restrained capital management strategy, with 

a consistent dividend policy, low financial leverage, 

and capital allocation concentrated on business risks 

with a focus on technical profitability. Concerning 

cycle and growth management, this business model 

pursues a long-term approach geared towards liquidity 

and operating cash flow. 

• Business model B follows a more opportunity- 

driven business approach based on rather limited 

diversification (monoliners, Cat players). With regard 

to capital management, the objective of this business 

model is to maximize returns through high, but 

volatile, dividend distribution and share buy-backs. 

A large amount of capital is allocated to Cat eventdriven 

business. 

• Business model C is similar to business model A. 

However, business model C focuses on new business 

lines, including innovative business lines such as banking 

business and Credit Default Swap (CDS) business. 

A large part of capital is allocated to an aggressive 

asset management policy. The capital management of 

business model C is geared towards the maximization 

of returns in the short term through share buy back 

and high leverage. Its financial strategy has a limited 

focus on liquidity and operating cash flow.

Fig. 95: The profile of a reinsurance company is a matter of key choices 

around its strategic axes 

Source: SCOR. 

IN THE (RE)INSURANCE INDUSTRY THERE 

WAS A CLEAR VALUE GAP BETWEEN 

THE BUSINESS MODELS 

Figure 96 compares share price to book value per share 

(P/BV) versus the return on equity (ROE) of the three 

business models. One allocated real companies to each 

business model, based on their risk approach, business 

diversification, asset strategy, etc. 

As demonstrated in the graph, in 2006, before the 

financial crisis, business models B and C were clearly 

better valued by the financial markets than business 

model A, with P/BV at a premium. 

Business models A and C had similar ROEs but business 

model C was valued higher in terms of P/BV, as many investors 

liked the innovative nature of its business strategy. 

Of all the business models, business model B had a much 

higher ROE and P/BV than the others, as monoliners 

were producing much higher returns, especially in years 

of low catastrophe events. 

Fig. 96: In the reinsurance industry there was a clear value gap 

between the business models - Pre-crisis multiples (2006) 

P/BV 2006 

1.45 

1.4 

1.35 

Portfolio 


Franchise 


Model A 

Higher 

diversification 

Model C 

Value gap 

1.3 

ROE 2006 

13% 15% 17% 19% 21% 23% 25% 

RISK BASED CAPITAL 

(economic capital) 

PROFIT 

Model B In 2006, before the financial crisis, 

business models B and C were clearly 

better valued by the market than 

business model A, with a higher P/BV 

at a premium. 

Business models A and C had similar 

ROEs but the business model C 

was higher valued in terms of P/BV. 

From all the business models, business 

model B had a much higher ROE 

and P/BV than the others. 

Source: Factset as of 03/06/2010. 

Capital 


VOLUME SOLVENCY 

(available capital) 

Lower 

franchise 

Higher 

franchise 

Lower 

diversification 

Risk diversification 

Franchise 

Rating 

Profitability 

Higher 

ROE 

Lower 

rating 

A+ 

Higher 

rating 

Lower 

ROE 

900 bps 

Cycle & Growth 



WHAT HAPPENED DURING THE FINANCIAL 

MARKET CRISIS? 

In fact, the financial crisis has led to a general reassessment 

of these business models, contradicting the 

strategic axes of reinsurance companies and impacting 

the four key strategic cornerstones previously quoted: 

• The portfolios of some reinsurance companies have 

suffered from a lack of diversification, especially some 

credit reinsurance companies and other monoliners. 

• Their franchise management has faced difficulties 

due to financial activities that have led to major issues 

(e.g. CDS exposure results in significant write-downs). 

• Aggressive asset management has led to substantial 

asset write-downs and impairments. 

• In addition, over-active capital management tools 

have led to stretched capital bases, low financial 

flexibility and questionable share buy-back programs. 

Shortly before the crisis, certain companies started to 

buy back their own shares when their share price was 

high, and leveraged the companies by issuing debt. 

• Cash flow management has been neglected, which 

has resulted in liquidity stress and has forced sales of 

assets and reputational problems. 

These factors have led to a general reassessment of the 

business models, particularly business model C. The 

latter appears to be less ‘suitable’ than business models 

A and B in a period when the financial environment 

is challenging. 

INVESTORS HAVE STARTED TO REWARD 

THE SECTOR FOR RISK MANAGEMENT… 

Investors have started to reward the sector for Risk 

Management. They have mainly distinguished between 

three key areas: 

• The investment margin is, on the Life side, a major 

driver for profitability; transparency is very important. 

It enables investors to focus on guarantees and the 

quality of hedging, and to track economic assumption 

changes. On the P&C side, the market can be suspicious 

of outsized investment returns. The key is therefore 

a balanced and clearly communicated investment 

strategy. 

• The risk margin is less of a driver but still constitutes a 

focus in terms of quality and track record; it is a useful 

tool for distinguishing between a volatile and a stable 

investment, especially on the Life side. Investors look 

for operating variances and assumption track records 

in the development of embedded value, and for 

sensitivity to mortality and morbidity scenarios. 

• The expense margin focuses more on the P&C side 

by annually reassessing the track record in terms of 

combined ratios, reserved track records, and sensitivity 

to large loss scenarios. 

Fig.97: Resulting in the repricing of reinsurers’ business models 

Pre-crisis multiples (2006) Post-crisis multiples (2009) 

P/BV 2006 

1.5 

1.4 

1.3 

1.2 

1.1 

1.0 

0.9 

0.8 


Model C 

Model A 

Model B 

0.7 

ROE 2006 

0.6 

0% 5% 10% 15% 20% 25% 


P/BV 2009 

1.5 

1.4 

1.3 

1.2 

1.1 

1.0 

0.9 

0.8 

0.7 

Model C 

Model A 

ROE 2009 

0.6 

0% 5% 10% 15% 

Model B 

20%

… RESULTING IN THE RE-PRICING OF REINSURERS’ BUSINESS MODELS 

Fig. 98: Differences in earnings expectations are also smaller 

Earnings Yield (1) 

40% 

30% 

20% 

10% 

Model A: 

Long-term average: 12.8% 

Post-crisis average: 14.4% 

Model B: 



Model C: 



Lehman Brothers' 

bankruptcy 

Re-pricing 

Model A: 14.4% 

Model B: 14.8% 

Model C: 14.0% 

0% 

Jan-06 May-06 Sep-06 Jan-07 May-07 Sep-07 Jan-08 May-08 Sep-08 Jan-09 May-09 Sep-09 Jan-10 May-10 

(1) Next 24 Months Earnings/Price calculated at 1/average. (Price/Next 24 Months Earnings). 


According to Figure 97, which has appeared already, 

business model A was previously valued lower than 

business models B and C. Whereas the reassessment 

and re-pricing of the insurance industry in 2009 led to 

business model A being valued the highest, even though 

it delivers lower returns than business models B and C. 

Business model C has clearly been under stress over the 

last few years, and has become the least valued. Earning 

expectations are not as compelling as before. 

Let us compare the earning expectations of the three 

business models at the beginning of 2006 with those 

at the beginning of 2010 using the graph below. This 

graph shows the share price versus the next 24 months’ 

earnings expectations (earning yields). 

Fig. 99: The best option is to consistently deliver over time 

It appears that in 2006 the earning yield, or earnings 

expectation, for business model A was lower, and 

that the yields for business models B and C were 

much higher. Before Lehman Brothers’ bankruptcy, 

the earning yields of all three business models were 

pretty stable. However, when the financial market 

crisis started, business model C’s earnings expectation 

drastically decreased and became very volatile. At the 

same time business models A and B were, compared 

to business model C, less affected by volatility. 

Given the return on equity ratios and the average for 

each business model, the best option for investors is to 

consistently deliver over time. Although business model 

A did not generate the most attractive ROE over the last 

five years, it turns out that it had the most consistent 

ROE in the standard deviation over the last five years. 

Company’s ROEs 2005 2006 2007 2008 2009 Average Standard 

Deviation 

Business Model A 5.7% 17.0% 14.7% 5.0% 15.1% 11.5% 5.7% 

Business Model B 1.8% 22.8% 20.5% 4.5% 18.8% 13.7% 9.8% 

Business Model C 1.9% 17.9% 19.0% -1.9% 9.2% 9.2% 9.3% 

Average 3.1% 19.2% 18.1% 2.5% 14.4% 11.5% 8.1% 

Source: Company’s data. 


II. Investors’ approach to (re)insurance company valuation 

Fig. 100: “The valuation life cycle” – how investors look at (re)insurance companies 

Methodology Pros Cons 

Appraisal value 

(EV (1) + multiple of life 

NBV (2) ) 

HOW DOES AN INVESTOR EVALUATE 

A COMPANY DURING THE FINANCIAL 

MARKET CYCLE? 

As shown in Figure 100, which is taken from a Goldman 

Sachs presentation, investors distinguish between three classic 

markets: the bull, the mid-cycle and the bear market. 

• During a bull market, investors focus on values such as 

embedded value and future profit, which characterize 

investors’ confidence and optimistic expectations. This 

means that investors allocate capital to Life and Non- 

Life. Based on the computed ROE and capital cost of 

each division, they round the allocated capital up for 

the valuation. This approach puts a higher valuation 

on companies than they are actually worth. 

• Conversely in bear markets, investors focus on tangible 

net asset values or equity excluding intangibles, or on 

classical Solvency I ratios. This is a period characterized 

by a fall in investment prices and pessimism, so people 

pay more attention to hard facts than estimates. 

• As for the mid-cycle market, which is marked by a 

period in which investors are more uncertain, they 

compare companies according to the ROE and the 

price to book value. 

“THE VALUATION LIFE CYCLE” – 

HOW INVESTOR LOOK AT (RE)INSURANCE 

COMPANIES 

• Simplicity 

• Unreflective of returns 

• Varying quality of assumptions 

• Arbitrary new business multiples 

Sum of the parts • Useful for composite businesses 

• Allows differential valuation of excess capital 

RoE (3) /RoEV (4) vs. P/B (5) / 

P/EV (6) 

• Good overall comparison 

• Links to BV growth, which is observable 

P/E (7) (7 

• Simplicity 

• Good for “earnings businesses” 

1x P/TNAV (8) P/TN 

• Focus on pure tangible assets 

for shareholders 

Solvency ency I 

• A more accurate measure of company 

stress 

• In line with local regulators 

• Reasonably simple to monitor 

Figure 101 confirms what we have just been discussing. 

In order to comprehend investor behavior, Merrill Lynch 

• Easy to misallocate capital between 

divisions 

• Need for x-cycle adjustments 

• Neither EV or BV(9) 

are fully comparable 

• Cyclicality and volatility of stated earnings 

• Distorted for fast growing life businesses 

• Growing inconsistencies in IFRS 

• Not a consistent measure 

• Liabilities not at fair value 

• Strip out DAC(10) 

? VOBA (11) ? 

• No allowance for returns 

• Not an economic measure 

• Ignores rating agencies 

• Ignores other triggers, e.g. leverage 

• Inconsistency of calculation 

Source: Goldman Sachs, October 2009, “Strategic relevance of changes in key insurance metrics”. 

(1) EV: Embedded Value. (2) NBV: Net Business Value. (3) RoE: Return on Equity. (4) RoEV: Return on Embedded Value. 

(5) P/B: Price to Book. (6) P/EV: Price to Embedded Value. (7) P/E: Price Earnings ratio. (8) P/TNAV: Price to Tangible Net Asset Value. 

(9) BV: Book Value. (10) DAC: Deferred Acquisition Costs. (11) VOBA: Value of Business Acquired. 


Bull Market 

Mid 

Cycle 

Bear Market 

led a study in which the bank asked several insurance 

and reinsurance investors a series of questions, most 

notably: what is the most important valuation metric 

that you utilize today? 

The bank first asked them this question in 2008, during 

the crisis, and then once again in May 2010. In 2008, 

investors were focused mainly on tangible book value, 

whereas in 2010, they concentrated on risk earnings 

and free cash flow yield. The development over the 

last twelve months in the graph on the right shows 

that attention to free cash flow yield has significantly 

increased over the last twelve months in comparison 

with the tangible book value. Overall, the focus from 

investors is more on earnings and free cash flow, which 

suggests a mid-cycle market. Whereas the Merrill Lynch 

study shows that in 2008 the focus was on tangible 

book value, which presupposes a bear market. 

PRICE TO ADJUSTED TANGIBLE BOOK VALUE 

IS THE MOST ADEQUATE VALUATION 

METRIC FOR (RE)INSURANCE COMPANIES 

Investors utilize three ratios in order to evaluate 

a (re)insurance company: 

• The price to book value is widely used by analysts 

across the reinsurance industry and enables them 

to compare a stock market value to its book value. 

It is computed by dividing the share price by 

the equity.

Fig. 101: “Valuation life cycle” confirmed by most recent investor survey 

What is your most important valuation metric today? 

Survey 2008 

13% 

14% 

Survey 2010 

16% 

19% 

11% 1% 

18% 

30% 

19% 

43% 

16% 

Tangible book value 

Embedded value 

IFRS earnings 

IFRS book value 

Free cash flow yield 

Dividend yield 

• The price to tangible book value aims to give a 

better valuation as it focuses on ‘utilized’ capital 

base and corresponds to the price of each share if 

the company liquidated all of its assets. This ratio is 

calculated by dividing the share price by the equity 

minus the intangibles. 

50% 

40% 

30% 

20% 

10% 

0% 

Tangible 

book value 

Almost half of the investors still look at IFRS 

earnings and tangible book value (P/TBV) 

– mid valuation life cycle? 

Free cash flow yield is more important, 

as it might give an early warning signal to 

mismanagement resulting in major issues 

Embedded 

value 

IFRS 

earnings 

IFRS 

book value 

Free cash 

flow yield 

Dividend 

yield 

Jun-09 Sep-09 Dec-09 May-10 

Source: BofA Merrill Lynch Global Research. 

• The price to adjusted tangible book value is the book 

value after having adjusted assets and liabilities to 

market value. The share price is divided by the equity 

minus the intangibles, which is added the value of 

business acquired. 

Fig. 102: Price to adjusted tangible book value the most adequate valuation 

metric for (re)insurance companies 

1 Price to Book Value (P/BV) 

Widely used by analysts across the reinsurance business P/B = 

2 Price to Tangible Book Value (P/TBV) 

Aims to give a better valuation as it focuses 

on “utilized” capital base 

3 Price to Adjusted Tangible 

Book Value (P/Adj. TBV) 

Removes the effect of goodwill 

on acquired life portfolios for 

recent acquisitive players 

P/Adj. TBV = 

Market Cap 

Shareholders’ Equity 

Market Cap 

P/TBV = 

Shareholders’ Equity – Intangibles (1) 

Market Cap 

Shareholders’ Equity – Intangibles (1) + VOBA (2) 

(1) Includes Deferred Acquisition Costs (DAC) and VOBA. (2) Value of Business Acquired (VOBA). 

Development over the last 12 months 


III. Investors’ view of the sector 

remains uncertain despite 

their reassessment 

WHAT ARE THE ORIGINS 

OF THIS UNCERTAINTY? 

Many analysts’ industry reports have been published 

recently, however the range of topics is very wide and 

opinions vary: 

• Global financial crisis effects 

• Solvency II 

• Capital Management 

• Inflation risk and scenarios 

• Dividend expectation 

• Pricing / Underwriting discipline 

• Preference for Life / Non-Life? 

• Reserve position 

This clearly shows that there are still many uncertainties 

emanating from various sources, notably the regulators. 

The most prominent concern is unquestionably the 

European directive Solvency II, which will have a 

significant impact on (re)insurers’ activities. 

UNCERTAINTY CAN BE SEEN IN THE HIGHER 

AMOUNT OF “NEUTRAL” VIEWS 

ON THE SECTOR, AND IN AN INCREASED 

PREFERENCE FOR LIFE 

The Merrill Lynch survey reveals that the investors 

questioned on their stance and preferences vis-à-vis the 

insurance industry placed more trust in the (re)insurance 

industry during the crisis in 2008 than they did in 2010. 

Moreover, the percentage of neutral ratings increases in 

2010, which indicates that investors have less confidence 

in this industry than before. During the crisis, there was 

a preference for Non-Life, whereas in 2010 this trend 

reverses with an increased preference for Life, which 

means that investors prefer to pursue a more stable business 

line. This is reflected in the following question: What 

will the most important driver of the sector be over the 

coming six to twelve months? It is interesting to note that 

regulatory development was not a concern before. In May 

2010, it became one of the key concerns for investors. 

PARADOXICALLY, INVESTORS ARE ALREADY 

ASKING FOR INCREASED RISK APPETITE 

AND MORE ACTIVE CAPITAL MANAGEMENT 

Indeed, 30% of the investors questioned would like to 

“raise the risk curve with regard to invested assets” over 

the next few months. 26% of them want to maintain 

a fortress balance sheet in view of the continuing 

uncertainty over Solvency II. And 21% of them would 

like to increase the risk appetite in terms of invested 

assets and dividends, which means being more offensive 

in terms of capital management. 

Fig. 103: Uncertainty can be seen in higher amount of “neutral” views 

on the sector and an increased preference on Life 

38% 

21% 


Survey 2008 Survey 2010 

11% 

What is your current 

sector stance? 

5% 

13% 

6% 

6% 

32% 

73% 


Very overweight 

Somewhat overweight 

Neutral 

Somewhat underweight 

Very underweight 

Within the sector, do you 

have a preference for life 

or non-life stocks? 

Life 

Non-Life 

No preference 

51% 

26% 

21% 

8% 

5% 

31% 

53%

Fig. 104: Regulatory developments are rising investors’ attention 

What will be the most important driver of the sector over the coming 6-12 months? 

19% 

50% 

40% 

30% 

20% 

10% 

0% 

8% 3% 0% 

22% 

Earnings 

estimate 

revisions 

GENERALLY, WHAT ARE THE KEY 

TAKE-AWAYS? 

• Prior to the financial crisis, analysts and investors 

largely favoured business models focused on a limited 

number of business lines, or financial products with 

more aggressive behavior towards capital and asset 

management. There was also a clear valuation gap 

between business models A, B, C and the traditional 

reinsurance business model. 

26% Earnings estimates revisions 

Valuation re-rating (up or down) 

Regulatory developments (e.g. Solvency II) 

Book value development 

M&A 

Capital action (right issues, buy-backs, positive/negative dividend surprises) 

Disclosure changes (presumably enhancements) 

22% 

Valuation 

re-rating 

Regulatory 

development 

Jun-09 Sep-09 Dec-09 May-10 

BV 

development 

M&A 

Capital 

action 

Disclosure 

changes 


Fig. 105: At the same time investors are already asking for increased risk appetite 

and more active capital management 

Given the recovery of balance sheets, do you believe that insurers should be starting to: 

13% 

21% 

10% 

26% 

30% 

Move up the risk curve, with regard to invested assets 

Maintain a fortress balance sheet given continuing Solvency II uncertainty 

Increase dividends/buy back stock 

Preserve capital – balance sheets remain stretched 

Consider strategic M&A, even if this results in capital issuance 


• In fact, the crisis has led investors to reassess and 

re-price the sector, and to start rewarding companies 

for good risk management by using different methodologies 

according to the “Valuation Life Cycle”. 

• Investors continue to focus on book value (including 

Goodwill), paying less attention to VOBA. 

• The financial markets are still uncertain about the reinsurance 

sector, which is also driven by increasing investor 

attention to regulatory developments (Solvency II). 

• Paradoxically, investors are already asking for increased 

risk appetite and more active capital management. 


Conclusion 

ERM: A SYSTEM TO ENSURE 

THE OPTIMAL SECURITY 

AND EFFICIENCY VICTOR PEIGNET 

Chief Executive Officer, SCOR Global P&C 

GILLES MEYER 

Chief Executive Officer, SCOR Global Life 

ERM is a fact of day-to-day life 

at SCOR: KYR (“Know Your Risks”) is as important 

as KYC (“Know Your Clients”) within the Group, 

and the combination of the two is what drives its 

operations. 

ERM stands for Enterprise Risk Management, although 

it could be argued that this acronym is short of one key 

word to fully represent the concept of ERM and what 

it means for the reality of company operations: this 

missing word is SYSTEM. 

For a global (re)insurance company like SCOR, being 

ERM(s) compliant means: 

• Operating with a single comprehensive Information 

System, 

• Being driven by capital allocation and return on 

capital, based on its own internal model, and 

• Having a strong Risk Culture embedded in its 

operations. 


The ultimate objective of an ERM system is to ensure the 

optimal security and efficiency of a company’s decision 

making processes and operations, without affecting its 

entrepreneurial spirit or innovative mindset, based on 

the best possible knowledge of risks and management 

techniques. 

By 2005, SCOR had managed to recover from its 2002-03 

crisis and had put ERM at the center of its recovery 

process. Although the Group had certainly regained 

the organizational structure of an entrepreneurial 

and ambitious company, having returned to good 

health thanks to a strong franchise, experienced 

and dedicated human resources, safe and efficient 

procedures and processes and a corporate culture, 

it had not yet reached the stage where it had a strong or 

fully operational ERM system, or a Risk Culture served 

by that system. 

Between 2005 and 2010, SCOR implemented and 

completed a 5-year project to introduce such a system, 

which is now fully deployed at all levels throughout 

the Group. This major initiative has had deep, 

structuring effects on the way in which SCOR operates, 

plans its business, dynamically monitors and controls 

its exposures and makes its decisions. 

Basically, this system has vastly improved both the 

quality of data and the way in which that data is 

treated, from how it is entered and warehoused, both 

before and after processing, to how its traceability and 

integrity are ensured.

In an integrated system like this, all the data centers (or 

the global data center), and all the software programs 

are linked and interconnected through automatic 

interfaces, transfers and downloads. Double entries are 

therefore minimized while automatic checks, balances 

and reconciliations are maximized. 

This single system constitutes the cornerstone of 

SCOR’s ERM policy. Without it, internal modeling would 

have to be conducted piecemeal, which would reduce 

efficiency and put the risk inherent to discontinuities 

into the modeling process. 

PRACTICALLY SPEAKING, WHAT DOES ERM 

MEAN FOR SCOR NOW? 

From an operational perspective, it means that treaties are 

underwritten using a single pricing tool, which is centrally 

controlled and parametered based on local analysis by 

Line of Business and by market. It also means that SCOR’s 

“deal teams” for given accounts or portfolios (underwriters, 

actuaries and modelers), along with the management 

team, have access to the dynamic follow up of: 

• The expected performance of priced and booked 

business, expressed either as ultimate underwriting 

ratios or as RoRAC (“Return on Risk Adjusted 

Capital”), which must exceed hurdle rates and be as 

close as possible to the Group’s target ROE; 

• The tail Value-at-Risk and the Value-at-Risk of this 

business, which must comply with pre-defined risk 

appetites; 

• The capital that has been used, compared with the 

allocations at any given time, and 

• Automatic year on year “as if” re-built price 

comparisons. 

This means that, during renewal seasons, the build-up 

of the following year’s portfolio can be monitored in 

real time against the plan and the budget and that, as 

renewals progress, adjustments can be made to the 

initial capital allocations by shifting capital to markets 

and Lines of Business with better than expected 

performances. With these management and reporting 

tools at their disposal, SCOR’s teams can behave like 

portfolio managers rather than deal makers, thereby 

optimizing their acceptances to meet portfolio target. 

WHAT MAY BE THE “DRAWBACKS” 

OF SUCH AN ERM SYSTEM? 

The first drawback is that there are risks attached 

to such a user-friendly underwriting system in terms 

of human behavior. These risks all involve the 

underwriters moving away from the traditional 

underwriting process and relying too heavily on the 

models to make their choices. From this point of view, 

it is equally important for underwriters: 

• To have a good understanding of the modeling process 

in order to achieve a greater commitment to and 

identification with it, and to become familiar with the 

tools involved, which enables them to work closely 

with specialized actuaries and modelers dedicated 

to their markets and Lines of Business within the 

“deal teams”, 

• To apply their underwriting expertise to their own 

assessment of the results of the models and to 

challenge these through plausibility checks and 

sensitivity analysis, and 

• To pay as much attention to trends as to final results, 

whilst respecting the Group’s values of consistency 

and continuity in their business relationships. 


The second drawback is probably that, being based 

on such a complex matrix of correlations, the internal 

model is so heavy to run that it challenges the 

calculation power of even the latest generation of 

machines. For the time being, therefore, it is practically 

impossible to run as many operational iterations 

as requested, whether for sensitivity analysis and optimization 

purposes at the planning and budget stage, or 

for unexpected large contracts with potentially material 

impacts over the course of a renewal season. 

In order to make the use of the internal model more 

dynamic, companies need to: 

• Upscale their in-house calculation capabilities, 

although these are limited to the power of the 

machines available on the market, or 

• Use the internal model for a limited number of detailed 

calculations per year, dedicated to final planning, 

budgeting and regulatory areas, and have a proxy 

model for fast runs and quick decision making. 

Our Group has followed the second route and developed 

a simplified model (CaDeT). This model can be used 

throughout the organization for checks such as “what 

if” analysis, the assessment of diversification benefits 

and the implications of portfolio changes. 

Another advantage of having a simplified model like 

this is that its broad, daily use by the teams will help 

to maintain and further reinforce the Group’s ERM 

culture. 

Since the implementation of its ERM framework, the 

SCOR Group has gained considerable experience and 

expertise. Eight areas contributed to this successful 

implementation: 

• Diversification that reduces the need for riskadjusted 

capital is the strategy followed by our 

Group. This strategy is based on SCOR’s twin business 

engines – Life and Non-Life – with the same weight 

allocated to both engines, which constitute the first 

level of diversification. It is worth noting that the 

same diversification techniques are applied within 

the two business engines. 

• The Risk Control system is a key component that 

SCOR has strengthened. SCOR Global Life and SCOR 

Global P&C are each the result of the recent mergers 

of several companies with different approaches to risk 

management. In view of this, the ERM framework 

has improved over the last three years with the 

development of one single approach for both Life 

and Non-Life, which has been embedded within the 

company. For example, SCOR Global Life and SCOR 

Global P&C have each put into practice a clear and 

defined underwriting and pricing process, having 

undergone a process of harmonization into one single 


system as opposed to the multiple pricing systems 

inherited when the entities were merged. Today, 

this enables SCOR to look at risks with consistency 

throughout the entire organization. 

• Stricter adherence to profitability targets forms 

part of this stringent risk control system. Any deviation 

is referred using a defined escalation process. 

• Capital Shield Strategy: buying retrocession protec- 

tion enables us to continue to reduce volatility even 

further. As an example on the Life side, the purchase 

of mortality swaps is also a part of this capital shield 

strategy. SCOR Global Life has decided to reduce its 

gross exposure by buying two swaps that are fully 

collateralized. 

• Internal Capital Model, which is completely embed- 

ded into the pricing model. This means that SCOR 

could prove to any stakeholder, such as the rating 

agencies for example, that any kind of pricing given 

by SCOR Global Life or SCOR Global P&C is consistent 

with the internal capital model. 

• Reserve adequacy: the reserves are checked on a 

regular basis, both locally and centrally, and are also 

reviewed by a third party consultant. 

• Risk enquiry, which is led by the SCOR Global Life 

and SCOR Global P&C Chief Risk Officers. Risk enquiry 

means that any type of risk that could be material to 

the Company is tracked on a regular basis. The risks 

are split into three categories – A, B and C – depending 

on their potential economic impact. This system 

involves a large number of people throughout the 

organization, which is to its advantage. 

• Solvency II: SCOR is preparing for the upcoming 

Solvency II regulations in Europe. One of the key elements 

of Solvency II is data integrity, which forms part 

of the Internal Capital Model. Compared to our peers 

in the reinsurance market, SCOR has a key competitive 

advantage: a single Group-wide IT system. 

All of the above demonstrates how the ERM process 

really is at the core of our business.

SPEAKERS’ BIOGRAPHIES 

Guest speakers 

JULIEN HALFON 

Head of Coverage, P-Solve Risk Management Solutions 

Julien Halfon is the Head of Coverage for P-Solve Risk Management Solutions. He advises European 

corporations on pension investment and risk management issues and European and American asset 

managers on marketing strategies for the European institutional investor markets. 

He joined P-Solve in 2008 from the investment banking division of Lazard & Co, where he led 

Lazard’s European corporate pension advisory initiative. Prior to joining Lazard, Julien was a Senior 

Investment Consultant at Hewitt Associates where he was part of the corporate advisory team and 

co-ran the infrastructure manager research group. He was also a Senior Associate in the Pension 

& Insurance Strategy Group of Goldman Sachs International in London. Julien holds a Masters 

degree in International Finance and Economics from the Dauphine University of Paris, an MBA from 

the Wharton School of the University of Pennsylvania and an M.A. from the School of Advanced 

International Studies of Johns Hopkins University. 

MILES TROTTER 

General Manager, Analytics Non-Life, A.M. Best 

Miles Trotter is general manager at A.M. Best’s London office. He is responsible for leading A.M. Best’s 

non-life analytical teams for Continental Europe and the United Kingdom, including Lloyd’s and 

London market companies. 

Miles has extensive experience in the London market. He began his career as a graduate trainee 

at Lloyd’s broker, Bain Dawes plc. Later he specialised in financial institutions insurance as a director 

of Holmes Johnson Lessiter Ltd, a subsidiary of Alexander Howden Ltd. In 1989, he moved to 

underwriting at Syndicate 1095, managed by Wellington Underwriting Agencies Ltd. 

After earning a master’s degree at the London Business School, Miles returned to Wellington to help 

start a new joint venture company, SBW Insurance Research Ltd. Prior to joining A.M. Best in 2001, 

Miles spent seven years developing this company’s research and analysis products. 



SCOR speakers 

GILLES MEYER 

Chief Executive Officer, SCOR Global Life 

Gilles Meyer is graduate of a French business school, he holds an MBA from the GSBA of Zürich. 

Gilles Meyer began his career as an underwriter at Swiss Re before taking over the optional department 

of the company La Baloise in Basel. After 23 years of experience in optional and treaty reinsurance, 

Gilles Meyer was CEO of Alea Europe from 1999 to 2006, in charge of both the Property & Casualty 

and the Life reinsurance and from 2005 to 2006, he was the Director of the Underwriting group of 

Alea. He joined the SCOR Group in January 2006 and directed the German-speaking markets of based 

in Hanover, Basel and Winterthur. 

End of 2006, he took the position of Director of the Business Unit 1 of SCOR Global Life. Since 

February 2008, he is CEO of SCOR Global Life SE. 

VICTOR PEIGNET 

Chief Executive Officer, SCOR Global P&C 

Victor Peignet, a Marine Engineer and graduate of the École Nationale Supérieure des Techniques 

Avancées (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 Group’s 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. 

MARCO CIRCELLI 

Head of Group Corporate Finance & Financial Communications, SCOR 

Marco Circelli is Head of Group Corporate Finance & Financial Communications at SCOR Group. 

He joined Converium’s Investor Relations in 2003. At the beginning of 2005 he took over the 

responsibility for the Business and Market Intelligence department where he was involved in the 

strategic development of the Company. Since August 2006, Marco Circelli was Head of Investor 

Relations at Converium. Prior to Converium Marco had worked in Management Accounting for 

Unilever and in various functions at UBS and Credit Suisse. He brings along more than 12 years of 

experience in Finance (of which almost 7 years in reinsurance), and holds a business degree from the 

University of Applied Science of Central Switzerland, and a MBA in Financial Services and Insurance 

from University of St. Gallen (Switzerland). 

JANICE COWLEY 

Program Director of CoCPIT, SCOR 

Janice Cowley is Program Director of CoCPIT – a global program focused on capital and risk management. 

Reporting to the Group CRO, she is responsible for leading large change projects for 

SCOR. Janice has more than 10 years experience in leading roles in both reinsurance and investment 

banking. Janice has a degree in Business from Victoria University of New Zealand and has completed 

post graduate studies and Harvard Business School, Boston.

MICHEL DACOROGNA 


Michel Dacorogna is a member of senior management of SCOR SE and heading the Group Financial 

Analysis and Risk Management modeling team. His main responsibilities 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 co-author 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. 

ERIC DAL MORO 

Chief Reserving Actuary, SCOR Switzerland 

Eric Dal Moro is an engineer of the École des Mines and graduated as an actuary from the French 

Institute of Actuaries. He is also a member of the Swiss Institute of Actuaries. 

He worked at AXA Direct Japan as Technical Director (1998 -2001). Then, he joined AXA Assicurazioni 

in Italy as Senior Actuary (2001-2002). He also worked for Ernst &Young in Paris and Zurich as 

Senior Manager (2002-2007). He is currently Chief Reserving Actuary and Appointed Actuary of 

SCOR Switzerland AG. 



Magdalena Klapper-Rybicka is a Risk Consultant in the Group Financial and Risk Modeling team 

of SCOR. Over the past two years, Magdalena contributed significantly to the development and 

implementation of an integrated quantitative risk management system for the SCOR Group. 

Magdalena Klapper-Rybicka studied computer science at AGH University of Science and Technology 

in Krakow, Poland. During her time at university, she worked as a scientist and on international 

research projects. She holds a PhD from the Department of Computer Science with a major in 

artificial intelligence. 

MICHÈLE LACROIX 

Chief Investment Officer, SCOR 

Michèle Lacroix joined SCOR in September 2008 as Chief Investment Officer of SCOR’s newly created 

asset management company, SCOR Global Investments. A graduate of HEC, she spent several years 

on the trading floor at the Caisse Centrale des Banques Populaires, first in charge of government 

bond activity and then covering all fixed income and hedging strategies. After 4 years of brokerage 

on fixed income she moved to Legal & General Asset Management (France) as a Portfolio Manager, 

managing the EURO fund and heading the development of new mutual fund strategies, before 

becoming Chairman and Chief Executive Officer of the company. 



TONY NEGHAIWI 

Chief Pricing Actuary, SCOR Global P&C 

Tony Neghaiwi joined Converium as Chief Actuary in 2006 and became in 2008 Chief Pricing Actuary 

and member of senior management of SCOR Global P&C with responsibilities of all P&C pricing 

and modeling operations. He has more than 23 years of experience in the insurance/reinsurance 

industry. He worked notably for Aetna Life & Casualty and Liberty Mutual in the United States, ZFS 

and XL in Switzerland. He held a Masters of Science in Applied Mathematics from University of 

Tennessee and completed post graduate studies in Mathematics from University of Houston and is 

a Fellow of the Casualty Actuarial Society. 

WAYNE RATCLIFFE 

Director Group Risk Management, SCOR 

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. 

EVA SCHLÄPFER DE MONTMOLLIN 


Eva Schläpfer de Montmollin is a Senior Risk Consultant in the Group Financial and Risk Modeling 

team at SCOR. For two years, she has been heavily involved in the development and implementation 

of an integrated quantitative risk management system at SCOR Group (CoCPIT). 

Prior to joining SCOR she held various actuarial roles at Zurich Financial Services where she has 

been leading a team of actuaries developing pricing methodologies and implementing a widely 

used pricing platform. She holds a PhD and a MSc in Mathematics from the University of Fribourg 

in Switzerland. She is a fully qualified actuary (SAV, DAV).

Paper copies limited to consultation. 

Distribution in electronic format only. 

To request an electronic copy, 

please email: scorglobalp&c@scor.com 

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.

Editor: Strategy & Development 

SCOR Global P&C 

scorglobalp&c@scor.com 

No. ISSN 1638-3133 

November 2010 

SCOR 

1, avenue du Général De Gaulle 

92074 Paris-La-Défense CEDEX – France 

www.scor.com 

November 2010 – Design: – Photo credits : Corbis, Getty Images, Inspoon, Shutterstock, Thinkstock.

Enterprise Risk Management (ERM) A risk-based approach to ... - Scor

Create successful ePaper yourself

Delete template?

Save as template?