Improving Operational Risk Management Systems by Formalizing the Basel II Regulation with Goal Models and the ISO/IEC 15504 Approach

Abstract

The bankruptcy of financial institutions shows the rapid changes in the risks profiles of financial systems and processes. Although financial institutions have always managed the operational risks, the profile of this kind of risks is changing due to the increasing international competitive pressure and the evolution of the financial institutions' operational systems relying more and more on IT systems. This paper reports the results of the joint research with the CSSF [1] focusing on the formalization of both the Basel II Accord and compliant operational risk management (ORM) systems implementations. This formalization uses concepts of the ISO/IEC 15504 process assessment standard and the concepts of strategy and policy. This structure of the model ensures the traceability between the Basel II Accord and compliant ORM systems implementations, improves the formal validation of those systems and is more adequate to represent all organizational levels of financial institutions.

1 Introduction

In Luxemburg, the stability of the financial system is at the core of the economic stability of the country. The CSSF [ 1 ], which is the official authority for financial institutions supervision, has the responsibility to define financial regulations and ensure their fulfillment. This task is not easy because more and more international regulations are introduced, such as the IFRS [ 2 ], Sarbanes-Oxley Act (SoX) [ 2 ] and the Basel II Accord [ 3 ]. Audit managers, risk managers (including security managers), and compliance managers have developed standards addressing those regulations. For instance, COSO [ 2 ], CobIT [ 2 ], ITIL [ 19 ] and ERM [ 2 ] are governance and risk management standards. However, up to now there is nearly no integration between the regulations themselves and also between those standards. A joint research with the CSSF aims at defining a method for ensuring a correct implementation of financial systems compliant to Basel II regulation. The results [ 6,21 ] are based on quality methods and techniques, mainly goal-based models and analyses used in goaloriented requirements engineering (GORE) [ 4 ]. The originality of the work lies in the formalization of the Basel II Accord and Operational Risk Management (ORM) systems by using concepts of the ISO/IEC 15504 process assessment standard [ 7 ] and the concepts of strategy and policy. This gives an adequate structure of the models at all organizational levels of financial institutions, ensures the formal traceability between the Basel II Accord and ORM systems, and improves their formal validation.

This paper summarizes and extends the results of the joint research with the CSSF, focusing on the formalization of both Basel II Accord and compliant (ORM) systems implementations. For a deeper understanding of the concepts presented here, see complementary information on the research results, the ISO/IEC 15504 standard, the Basel II Accord, and other standards such as ITIL on the CSSF website [ 1 ] (freely available). The next section presents the main goals of this research and the preliminary results. Section 3 shows the technique that has been created in the context of the real case study concerning the Basel II Accord regulation and its implementation in financial institutions. The last section summarizes the main results of this project and presents the future works planned in the follow-up research projects. 2 The Implementation of ORM Systems Compliant to Basel II. The Basel Committee has defined the operational risk as follows: it is the risk of loss resulting from inadequate or failed internal processes, people and systems or from external events. (§644 in [ 1 ]). As such, the operational risk encompasses all risks occurring at the operational and technical levels (see Fig. 1), in particular, all risks of the IT Software Engineering (SE) Processes (risks that concern project management, requirements analysis, design, security, ...). The methods used in IT SE (e.g. for safety and security analyses) do not cover the analysis of this very broad scope of risks.

The need for practical techniques is critical in order to help business units’ manager to efficiently implement the core business processes that are under their responsibility. Indeed, not only the Basel II Accord is imposing constraints on those core financial processes, but also the other regulations (e.g. SoX, IFRS) are interfering on the same processes. Moreover, each regulation stresses the importance on different but inter-related aspects. For instance, SoX stresses the importance on the reporting system also concerned by the ORM of Basel II Accord. In addition to that, decisions about ORM system implementation must be made at all organizational levels : strategic, tactical, operational and technical. The existence of operational risks within every business process imposes a tight integration between new ORM systems and each business process, increasing the complexity of modeling and implementing ORM systems. Last but not least, those regulations are difficult to understand due to their lack of structure and lack of completeness. For instance, in the Basel II Accord there is no definition of important concepts such as “ORM system”, “loss”, “loss event”, “unexpected loss”, ...

Requirements engineering and goal-oriented methods. The GORE methods can overcome the difficulties presented in the preceding section by formalizing the Basel II Accord and the implementation of ORM systems. These methods can be used to analyze and model systems at all organizational levels, from Business Models up to architectures [ 4 ]. Goal-oriented modeling languages are appropriate for that broad range of models and they support formal analyses [ 5 ].

However, in the case of the ORM system, it is difficult to manage all of those large models and complex analyses. Moreover, for validation purposes, it is important to refer to the concepts used in organizations, such as strategic objectives, strategies and plans, key indicators, policies, SLAs, ... Within the context of the Basel II Accord, additional structuring mechanisms have to be created on top of the usual goaloriented concepts.

Tleecvhelnical

Strategic B level

Tactical Operaletivoenlal level usiness value usiness Bprocesses

Procedures E.g. in the IT domain: applications, components, … The general framework given in the Figure 1, represented by the pyramid, is a standard view of the organization [ 8 ] used in financial institutions (and other activity sectors). The four organizational layers [ 9 ] – strategic, tactical, operational and technical levels – use concepts adapted to handle decisions at their corresponding abstraction level. For each level, from top to bottom, those concepts are mainly: business value [ 10 ], business processes, procedures and technical artifacts [ 11 ] (such as IT applications in the IT domain).

ISO/IEC 15504 process assessment model. A first part of the structure is given by separating the description of the core activities of each business process from the activities related to the generic aspects (capabilities) of the business process such as activity planning, work product control, management of the documentation concerning the business process itself, performance measurement, performance improvement, ...

As explained in [ 6 ], the benefits of this separation of concerns has proven to be very useful for the design of the goal models and during their verification and their validation. This separation of concerns is formally defined in the ISO/IEC 15504 standard. This new standard has been designed to be applicable to any business processes for assessments purposes [ 21 ]. Objectives, strategies, policies and indicators. Those concepts (bottom of Figure 1) detail complementary aspects needed for designing business value, business processes, procedures and technical artifacts. They are similar to organizational concepts needed in order to structure and formalize the links between each of the organizational levels [ 12 ].

When refining models at the higher levels of the hierarchy into models at the lower levels of the hierarchy and when verifying the link between two successive organizational levels, it is necessary to distinguish the main objectives to be fulfilled from the strategy describing the approach to fulfill these objectives and from the roles and responsibilities (policies) of the resources that will implement the strategies. Indicators are defined when there is a need for some monitoring, control, supervision or measurement concerning objectives, strategies or policies. Strategies and policies must be consistent with each other and they must fulfill the objectives.

The formal definition of those 4 concepts is based on goal-oriented models [ 4,5,6 ]. For the indicators, our work is based on the Goal-Question-Metric method (GQM) [ 13 ]. Policies give a description of the roles and responsibilities (in accordance with [ 14 ] and policy management [ 15,16,17,23 ]) and allow detailing the authorizations, obligations (and their delegations), accountabilities, and separations of duties [ 23, 24 ]. Strategies give a description of the main approach or steps to fulfill given objectives. Our work follows [ 18 ] where strategies are integrated with goal-oriented analysis. For the sake of separation of concerns, responsibilities (and related aspects) are not defined in strategies but only in policies. Note that in financial institutions, the description of policies recalls its related objectives and strategies. This is also sometimes the case for strategies that gives a short description of their corresponding policies (i.e. description of roles and responsibilities). However, it is found essential to separate those descriptions when designing and analyzing those policies and strategies.

Example: contribution of ITIL to the implementation of the ORM system. This example shows the usefulness of the separation of concerns when analyzing the Basel II regulation. The main problem for financial institutions is not to comply with the Basel II regulation, but to have efficient business processes fulfilling business goals that are also compliant to the Basel II regulation.

The current case study describes a financial institution that implements ITIL [ 19 ], an IT service management standard, for aligning the service provided by IT applications to the business goals through the use of Service Level Agreements. The analysis aims to answer the following question: what are the contributions of the IT service management implementation to the Basel II regulation?

In order to answer that question, the contributions of the IT service management goal-model to the Basel II goal-model must be analyzed. Only the core of each goal models is to be compared. Indeed, the generic aspects (capabilities) add only quality aspects concerning the ORM and the IT service management and do not address the main goals of the ORM and IT service management processes. The analysis of the goal model is simplified. This can be seen in the Figure 2: the left part of the diagram shows a part of the Basel II Accord formalization of ORM and the right part presents a partial IT service management system implementation using ITIL. The diagram shows only a part of the models from the strategic level (topmost) up to the operational level (bottom). Only objectives are shown for the strategic level and the operational level. In between, at the tactical level, the objectives and indicators of business processes are shown.

The links between the two models are formally analyzed [ 6 ]. For instance, Basel II and ITIL share the business goal of reducing the impact of incidents. When drilling down from the strategic level to the tactical level, the traceability links offered by the goal refinements shows that the ITIL goal imposing the incidents are tracked and recorded contributes to the Basel II indicator of the operational risk assessment. In this case, our structuring mechanisms allowed an efficient analysis because the focus excluded generic aspects and also excluded the policy and strategy aspects of both Basel II and IT service management. 4

Conclusions and Future Works

Building upon a method that has been defined within the setting of a real-case study in financial institutions, the Basel II Accord, new results are presented in this paper aiming at giving a simple but integrated set of concepts – goals, indicators, policies and strategies – which can be used to design financial systems compliant to regulations and structure their analysis in relationship with the artifacts commonly used in financial institutions – business value models, business processes models, procedures and technical artifacts. The formalization of goals, indicators, policies and strategies independently from each other allows analyzing and recording the design decisions across all organizational levels, making easier the link with the regulation. The main advantage of this method is that it keeps the structuring power of the ISO/IEC 15504 capability model that can be used to discover weaknesses and operational risks in the business process implementation with the method explained in [ 20 ]. Based on the same techniques as in [ 16 ], a prototype implementation is under development.

The current and future works of the authors focus on a constructive method aiming at giving an effective support for financial business process design (compliant to regulations), establishment, assessment, improvement, governance and benchmarking [ 6 ]. In particular, a risk and value analysis method is under development adapted to process assessment, improvement and governance. Some support is also given to another research made by experts in DPM [ 22 ]. The aim of those experts is to ground digital policy management in sound non-federated distributed IT systems that enforces policies fulfillment even outside the traditional IS frontier of each institution. Finally, the current project with the CSSF is still in progress with results that are extended to the IFRS [ 2 ] concerning the management of unquoted assets (IFRS-IAS39) [ 2 ]. In addition to model this regulation and the systems compliant to it, the relationship between IFRS-IAS 39 and Basel II can be analyzed and alternative compliant implementations of integrated systems can also be designed.

References

1. CSSF: Commission de Surveillance du Secteur Financier. The firsts results of the joint project are freely downloadable at http://www.cssf.lu/index.php? id=130 (accessed April 2006 ).

2. IFRS: International Financial Reporting Standards, IASCF , USA. SoX: Sarbanes Oxley Act of 2002 , USA. COSO: Internal Control - Integrated Framework , CSOTC , USA. CobiT®: Control Objectives for Information and related Technology, ISACA, USA . ERM: Enterprise Risk Management - Integrated Framework , CSOTC , USA.

3. Basel Committee on Banking Supervision, “International Convergence of Capital Measurement and Capital Standards”; BIS; Basel, June 2004 .

4. A. van Lamsweerde , "Goal-Oriented Requirements Engineering: A Guided Tour" . Invited minitutorial, Proc. RE ' 01 - International Joint Conference on Requirements Engineering, Toronto, IEEE, August 2001 , pp. 249 - 263 .

5. P. Giorgini , N. Maiden , J. Mylopoulos , E. Yu (eds.), “Tropos/i*: Applications, variations and Extensions”, Cooperative Information Systems Series , MIT Press, 2006 (forthcoming).

6. André Rifaut , “ Goal-Driven Requirements Engineering for Supporting the ISO 15504 Assessment Process” , EuroSPI 2005 , Budapest.

7. ISO/IEC 15504, “Information Technology - Process assessment”, (parts 1-5) , 2003 - 2006 ( see website [1] for details about this standard).

8. Anthony , R. N. Planning and Control Systems: A Framework for Analysis . Harward University, Boston, USA, 1965 .

9. Henderson , J. and Venkatraman , N., “ Strategic alignment: Leveraging technology for transforming organizations” . IBM Systems Journal , 1999 , 38 .

10. Osterwalder and Pigneur. An Ontology for e-business models . In “ Value Creation from EBusiness Models” , Wendy Currie ed., Butterworth-Heinenmann. Apr 2005 .

11. Robson W , Strategic Management and Information Systems , Pitman, 1997 .

12. Chaffey et al. ( 2005 ) - Business Information Systems: Technology, Development and Management for the E-business , Prentice Hall.

13. Van Solingen, “The Goal/Question/Metric Method: A Practical Guide For Quality Improvement of Software Development” , McGraw-Hill , Jan . 1999 .

14. René Wies , “ Using a Classification of Management Policies for Policy Specification and Policy Transformation” . In Proc. ISINM '95 , Santa Barbara , California, May 1995 .

15. N. Damianou , N. Dulay , E. Lupu, and M. Sloman , “ The ponder policy specification language ” In Morris Sloman, (ed), Proc. of Policy Worshop , 2001 , Bristol UK , January 2001 .

16. A. Schaad and J. Moffett . “Delegation of obligations.” In IEEE Policy Workshop , 2002 .

17. Qingfeng He and Annie I. Antón , “ Deriving Access Control Policies from Requirements Specifications and Database Designs” , TR-2004-04 , Department of Computer Science, North Carolina State University Raleigh, NC 27695 -8207 USA, September 02 , 2004 .

18. Rolland C. , N. Prakash , A . Benjamen, “ A multi-model view of process modeling” , Requirements Engineering Journal , p. 169 - 187 , 1999 .

19. ITIL: IT Infrastructure Library - Service Support , Service Delivery, published by OGC, London. (see website [1] for details about this standard).

20. A. Rifaut , M. Picard and B. Di Renzo , “ ISO/IEC 15504 Process Improvement to Support Basel II Compliance of Operational Risk Management in Financial Institutions” , International Conference SPiCE 2006 .

21. B. Di Renzo , M. Hillairet , M. Picard , A. Rifaut , C. Bernard , D. Hagen , P. Maar , D. Reinard , “ Operational Risk management in Financial Institutions: Process Assessment in Concordance with Basel II” , International Conference SPiCE 2005 .

22. J.-H. Morin and M. Pawlak , “ Towards a Global Framework for Corporate and Enterprise Digital Policy Management” , SoftWars conference, Las Vegas, USA, Dec 11 , 2005 .

23. Sandhu , R. S, “ Separation of duties in computerized information systems .” In Database Security, IV: Status and Prospects , 1991 .

24. David F. Ferraiolo , Ravi Sandhu, Serban Gavrila , D. Richard Kuhn and Ramaswamy Chandramouli, “ Proposed NIST Standard for Role-Based Access Control” , ACM Transactions on Information and System Security , Vol. 4 , No. 3, August 2001 , Pages 224 - 274 .