Newsletter #168 Home
Newsletter Contributors
Previous Issues
BARRA Home




Developing and
Implementing
Risk Management
Systems



[ Case Study ]
Fortis Group
Case Study



[ Equity Analytics ]
American Depository
Receipts in The
Global Equity Model



[ Equity Analytics ]
Analyzing the
Performance of
Crossing Networks

The Market
Impact Model™ — Part 4



Brainteaser
The BARRA Brainteaser
 for Fall 1998

Solution to the Summer
1998 Brainteaser



Fortis Group Case Study:
Technology Solutions for Central Risk Management

by Richard J. Moran, vice president, BARRA ERM and Dr. Matthijs van den Adel, managing director risk and ALM, Fortis Bank Netherlands

Portions of this article appeared in the November Enterprise Risk Management Supplement of Risk Magazine

About The Fortis Group

The Fortis Group has more than 100 companies operating internationally in insurance, banking and investments. It was created in 1990 from the merger of the Dutch banking and insurance combination AMEV/VSB and Belgium's largest insurer, AG, in what was the first European cross-border consolidation in the financial sector. Since its inception the group has both grown internally and expanded through acquisitions, mainly in the Benelux countries and the United States. Mergers over the last five years include: the Belgian bank, ALSK-CGER, in 1993; NMKN-SNCI in 1995; the Dutch merchant bank MeesPierson in 1996; and the Belgian bank Generale Bank in the second quarter of 1998. (See FIGURE 1.)

FIGURE 1: Fortis Group Balance Sheet Assets
[ Figure 1 - graph ]

Fortis Group provides insurance, banking and investment products and services in the Netherlands and Belgium. In the United States, it concentrates on the insurance and investment sectors. The group is active also in Spain, France, Luxembourg, the United Kingdom and Australia, and in various countries in Asia and the Caribbean.

Planning for Integrated Central Risk Management

The Fortis Group's strategy of expansion through acquisition and growth across financial services posed significant and unique central risk management (CRM) challenges, including the consolidation of an enterprise-wide risk system across a range of risk categories, organizations, and Information Technology systems.

The Dutch component Fortis Bank Nederland (FBN) was assigned the task of designing a risk management system and needed to align its overall business culture, practices and applications. This would allow FBN to leverage its existing resources while deploying new advanced risk management methodologies. It opted for an incremental evolutionary approach to this formidable risk management task, one that allowed it to achieve significant CRM goals while simultaneously containing the risk of unexpected setbacks.

Prior to the MeesPierson acquisition in 1996, FBN had fortified its systems and methodologies to optimize the measurement and control of its portfolio risks. It employed a basic parametric Value at Risk (VaR) application to monitor and control risk for the bank's linear financial products. With the integration of MeesPierson into the FBN central risk management function, the company decided to invest in an enterprise risk management (ERM) software solution.

This decision was based on two key motives. First, the introduction of Dutch Central Bank (DNB) regulatory reporting requirements based on the 1988 Basle Accord meant that FBN had to determine the correct figures for regulatory capital allocation purposes. Second, the MeesPierson banking business included derivatives trading on behalf of itself and its clientele which, when added to the FBN portfolio, created a significant increase in the total number of positions held in complex instruments.

FBN realized that choosing the right software was crucial to the development of a suitable risk management system. If it opted for a solution that met only one of the selection criteria, such as the need to meet minimum regulatory capital requirements, then this could negatively impact the firms overall disposition towards an evolutionary risk management strategy. FBN had to choose a system that was flexible enough to meet both its present needs and its future needs. The next step was to conduct a rigorous vendor selection process.

Building on a Solid Foundation: Advanced ERM Selection Process

The decision to select and deploy an advanced ERM solution at MeesPierson and FBN was motivated by a number of objectives reflecting various escalating ambition levels for risk management. FBN firmly believed that if it adopted an incremental approach, the banking business would be compelled toward a pro-active risk practice that would, in turn, lead to optimized capital allocation and, ultimately, to maximized returns on shareholder equity. Moreover, successful implementation of an advanced ERM solution would provide the foundation for the application of advanced risk practices across all of the insurance and investment businesses within the Fortis Group.

FIGURE 2: Key risk management objectives at Fortis Bank Nederland
Objective Ambition Level Key Benefit
1 Meet BIS/CAD requirements set forth by DNB Measurement (Regulatory)
2 VaR limits
Pre-deal analysis
Client VaR (e.g. market makers)
Counterparty VaR.
Monitor and Control
3 Risk/Reward Capital Allocation
4 Central Risk Management across Banking,Insurance, and Investment lines of businesses Optimized shareholder value

Objective 1 (See FIGURE 1) was a product of the DNB regulatory requirements for Capital Adequacy Directive (CAD) and Value at Risk (VaR) reporting based on the Bank of International Settlements guidelines.

Objective 2 included VaR limits, pre-deal analysis, client (e.g., market makers) VaR and counterparty VaR. This objective pushed the central risk management function beyond measurement and reporting to the monitoring of the actual and potential risks of FBN's positions and those of its clients, particularly clients who were market makers to whom FBN rendered clearing services.

Objective 3 used the earlier phases for the deployment of a Risk-Adjusted Performance measurement tool. This enabled management to allocate and transfer capital efficiently, balance the bank portfolio in line with policy, and assist in meeting long-term return on equity goals.

Objective 4 applied the practices and systems implemented by FBN across other business areas.

In selecting a software vendor, FBN identified business function, technology, and an appreciation of its evolutionary approach as essential criteria. In addition, FBN required a solution that would be agile enough to evolve through increasing levels of sophistication. With these core criteria in mind, the FBN risk project team established eight selection requirements for the software itself.

Financial Product Coverage—the full range of financial instruments in the FBN portfolio had to be covered, including a variety of equities, fixed income products, commodities, FX products, and derivatives.

Regulatory Requirements—the product had to provide modules that met BIS reporting requirements, e.g., VaR via historical simulation, stress testing, and backtesting.

Model Maintenance & Operation—functional capabilities such as enhanced stress and backtesting as well as sophisticated data management features were needed to ensure the efficient deployment and operation of the application.

Flexibility & Extensibility—the platform had to be able to accommodate new pricing models, risk methods and factors, dynamic drill-down structures, data sources, and customizations rapidly to enable the product to keep up with the bank's growth and to enable CRM ambition levels to be realized. This was one of the most heavily weighted requirements in the selection process.

The other selection requirements were Performance & Scalability, Platform & Hardware Requirements, Vendor Relationship & Support, and Product Maturity and Stability. (For detailed selection criteria for the Enterprise Risk Management Solution, see FIGURE 3.)

FIGURE 3: Detailed Enterprise Risk Management
Solution Selection Criteria


BARRA's Redpoint TotalRisk was selected as the ERM solution at Fortis Bank Nederland based upon its ability to meet a rigorous set of requirements established by the MeesPierson and FBN project teams.

Financial Product Coverage 
Equities & Derivatives
Fixed Income & Derivatives
Foreign Exchange & Derivatives
Commodities & Derivatives
"Best Practice" Pricing Models
Regulatory Requirements 
Value at Risk computation
  – VaR computed daily using Historical Simulation
  – 99th percentile one-tailed test
  – 10 day holding period
  – at least 1 years historical observations
  – empirical correlations within broad risk categories
Integrated into management procedures
  – Design & implementation of risk management system
  – Day-to-day risk management
  – Internal trading & exposure limits
  – Daily reports of exposure vs. limits
Unique risk of options
  – Non-linear characteristics
  – Full 10 day price shock
  – Capture vega risk
Capture market risks
  – Interest rates, include. spreads
  – Exchange rates
  – Equity prices, including general & specific risks
Rigorous & comprehensive stress testing
  – Low probability events for market, credit & operational risks
  – Liquidity risks
  – Develop worst case scenarios
P&L accounting
  – Daily P&L along the portfolio hierarchy
  – Handling fees and unwinding costs and commissions etc.
Regular back-testing program
Verify data integrity
  – Accuracy & completeness of positions
  – Consistency, timeliness & reliability of data sources
  – Accuracy & appropriateness of volatility & covariance assumptions
Model Maintenance & Operation 
Maintaining the portfolio hierarchy
Storing & computing portfolio positions from transactions
Maintaining tables of financial products & reference data
Utilizing different pricing models
Maintaining current rates for risk factors
Ad-hoc computation using Var-Covar
Routine computation using Historical Simulation
Stress testing using Monte Carlo Simulation
Back Testing Using Hypothetical P&L
Regulatory reporting of back testing, stress testing, & capital requirements
Security and authorization
Flexibility & Extensibility 
Programmable input & output interfaces
Support for multiple pricing models
Support for alternative risk methods
Ability to add new financial products
Ability to add new risk factors
Ability to handle the EURO and Y2K
Flexible data warehouse
Availability of existing interfaces for transaction/position source systems
Support for trading limit structures & pre-deal analysis
Support for counterparty & client risk with netting
Support for RAPM & RAROC using risk-adjusted returns
Support for Asset & Liability management
Performance & Scalability 
Meets daily peak level
EOD position/transaction capture
Daily VaR computations
Daily back test against actual P&L
Routine regulatory reporting
Ad-hoc VaR requests
Regular Stress tests
Intra-day analysis (2 hour window)
Product Maturity & Stability 
History of system development and upgrade releases
Current development program & expected release schedule
Current schedule of faults & service requests from fault log/database
The system must be capable of encompassing advances in risk management techniques and regulatory developments
Testing of software
Vendor Relationship & Support 
Financial & management stability of system vendor
Number/list of existing clients and operational installations
Reference sites & contact persons
Adequate development and support personnel for anticipated demand in Europe
Nature of Service Level Agreements in-place/available
Help desk available
Full user & systems documentation
Regular upgrades & release schedule
Training support
Platform & Hardware Requirements 
Hardware & software platforms ideally should be familiar to MP IT Department
Network requirements should be compatible with existing networks at MP
System should comply with IT standards

FBN conducted an extensive selection process that included a thorough review of 15 vendor solutions and, in the end, chose the Redpoint TotalRisk product offered by the BARRA Enterprise Risk Management business unit. This choice was based on the vendor's ability to meet key coverage and business function requirements, and most importantly, to provide a solution that enabled the bank to realize its evolutionary approach to the CRM challenge.

Redpoint TotalRisk delivered the range of business functions required and offered sufficient flexibility and scalability to keep pace with FBN's growth. Redpoint TotalRisk was first implemented at MeesPierson. This rapid and successful deployment provided FBN with the confirmation that this product and this client services organization were the ideal foundation for the CRM function within FBN. Today, FBN is in the final stages of implementing TotalRisk.

Beyond Banking: Risk Across the Enterprise

The ultimate objective of CRM at Fortis Bank Nederland is to optimize return on equity across all businesses and thus increase shareholder value. This is made possible by the sharing of quantifiable risks among the bank, the insurer, and the asset manager. The central risk management approach hinges on measurement techniques that can be applied by Fortis Bank Nederland's three components—asset management, insurance, and banking—despite the fact that they are involved in different sectors of the financial services (See FIGURE 4).

FIGURE 4: Central Risk Management Across Business Lines
[ Figure 4 - drawing ]

These three business lines deal with the same quantifiable risks for which they must keep sufficient capital in reserve for unexpected losses. The capital placed at risk of loss with a given level of confidence over a specified time under the determined risk/reward profile could be calculated individually for each of the three branches, but for minimizing the capital-at-risk it is more effective to adopt a central approach to calculating the potential volatility of outcomes, e.g., potential changes in value of assets. A consolidated firm-wide view allows for the allocation and transfer of non-optimized capital-at-risk among the three segments of the financial/insurance/asset management industry.

The differences between the individual business lines notwithstanding, it is clearly more efficient to share the development and use of data management and measurement systems among these lines. Though the life insurer focuses on future liabilities and an assured minimum return on investment and the bank focuses on market and lending risks, benefits to both businesses can be gained by sharing a common risk management system. For example, systems that manage internal databases concerning contractual insurance obligations, market exposure, duration of equity, gap analysis, mismatch positions, and internal credit ratings can provide a consolidated data source for the entire range of front-to-back office applications. The same is true for external data such as market factors—interest data on financial markets (rates, prices, and volatilities), correlations (markets, industrial sectors, and currencies), default and currency rates.

Thus, the key to FBN's success in central risk management is its ability to leverage such common platforms across its three business lines.

Summary

The Fortis Group's strategy of growth through acquisition and expansion across financial services posed a unique CRM challenge. By adopting an evolutionary incremental approach, FBN achieved a consolidated enterprise-wide view of risk across a broad range of categories, organizations, and systems.

As part of this strategy—and upon acquiring MeesPierson in 1996—a series of central risk ambition levels was set to drive the process forward. FBN believed that the implementation of each level would advance its banking business toward a forward-looking, proactive risk practice that optimized the allocation of capital. Furthermore, FBN determined that a successful banking implementation would provide the foundation for the application of these advanced risk practices across the insurance and investment businesses, maximizing returns on shareholder equity.

Analysis of the requirements of the various ambition levels revealed the need for an advanced ERM application, one which would pave the way for the realization of the ambitious goal of central risk management within FBN. Through its deployment of Redpoint TotalRisk, FBN has achieved its objectives and now faces the next hurdle in the evolution of risk management—the integration of Generale Bank and the expansion of risk management practices and systems across its non-banking business areas. 





[client support]   [portfolio management]   [investment data]   [trading  services]
[model  &  market information]   [research resources]   [about BARRA]  

[online product center]

[search]   [site map]   [contact us]   [home]  

Any questions or bug reports regarding this service should go to contactus@barra.com.
© 1995-1999 BARRA, Inc. All rights reserved. Terms of Use.