Development of tools for the management of interest rate risk arising from the banking book for Banque Havilland

Abstract

This thesis was written as part of a project thesis carried out within the Risk Department of Banque Havilland. The projects it defines and presents have been designed as part of the development of interest rate risk management tools to meet the European Banking Authority's recent requirements in this area. The purpose of this thesis is to present this new regulation, document the tools used and issue conclusions based on the analysis of the results obtained. The overall project is divided into four major parts.

The resistance test The first step, after understanding the regulatory framework and the various legal texts relating to interest rate risk, was to measure the bank's exposure, in terms of economic value and future income, to various scenarios for changes in the yield curve. This measurement was carried out using a resistance test. To do so, we had to make different assumptions, follow the calculation methodologies recommended by the European Banking Authority and apply different scenarios for changes in the yield curve. The results of this test are presented in the conclusion section.

Interest rate risk management procedure Based on these measures, we have developed an interest rate risk management procedure, including the definition of the risk, the selection of the instruments concerned, the calculation methodologies, the bank's "Risk Appetite" in this area as well as the distribution of roles and the procedure to be followed in terms of governance.

Development of a main component analysis This main component analysis was developed to analyze the dynamics of the yield curve and with the final objective of finding other shock scenarios for this curve to apply to our stress test. In addition to the exploration of the theoretical concept, this thesis explains the methodology used, some problems encountered and the results of the analysis.

Design of new scenarios via Monte-Carlo simulation Two scenarios of changes in the yield curve were developed by simulating many linear combinations of these main components. The methodology followed and the results of this "Monte-Carlo" are presented in this thesis.