Internal and external controls of banking risk: towards a gradualist regulatory scheme.

Summary The objective of this thesis was to analyze the optimal weighting between internal and external control of banks in the context of a greater degree of freedom granted by the new banking regulation to banks. External regulation is based on procedures and techniques developed by practitioners within the institutions. In a context of liberalization of financial markets and marketization, internal control can only develop, but the question of its greater effectiveness then arose. The link between external control, which has shown its limits, and internal control, which allows for a more detailed assessment of banking risks, became a key issue. The proposed approach is based on three stages. The first is a theoretical and institutional analysis of internal control, then of external control. An analytical and empirical study is conducted on the VaR (value at risk) model, which is used as a risk quantification tool by practitioners. The second step proposes a reflection on the articulation between internal and external control. The use of internal models raises the question of their reliability and brings to light the notion of "model risk", which leads to the justification of the conservative approach adopted by the regulator, which consists of applying a multiplicative factor to the results of the VaR in order to calculate regulatory capital. Moreover, it is possible to combine the VaR method for normal periods and stress tests to take into account abnormal events. The last part of the thesis proposes a regulatory scheme based on "flexible and dynamic regulation". It is indeed possible to question the traditional dichotomy between market risks and credit risks, thanks to the use of new instruments, notably credit derivatives.