Approximating Capital Requirement Due to Name Credit Concentration Risk

Abstract: Since banks provide various forms of credit, they consequently expose the business as well as society to risks. Therefore, global banking regulations exist to ensure that banks keep enough capital for the risks they are taking. One such risk is name credit concentration risk, arising when exposures in a portfolio have a skew distribution across different counterparties. Regarding the treatment of this particular risk, Finansinspektionen gives guidance on how Swedish banks and financial institutions could approximate it. They do so by proposing two methodologies, one within a Standardised Approach and one within the more advanced Internal Ratings-Based (IRB) Approach. A comparison between the resulting capital requirements for Finansinspektionen’s two proposed methodologies do not publicly exist. Performing such a comparison for portfolios with different characteristics, including a third parameter, would add to previous research. The comparison would also provide Swedish banks and institutions with additional knowledge of the differences in resulting capital requirements when applying Finansinspektionen’s two proposed methodologies on their portfolios. Thus, the authors have compared these methodologies. The thesis is performed through a quantitative study of generated realistic portfolios and of Ikano Bank’s leasing portfolio, where two methodologies for approximating capital requirement for name concentration risk are applied on the portfolios. Theory used is the methodologies proposed by Finansinspektionen, i.e. the Standardised Approach and an IRB Approach; a Granularity Adjustment according to Michael Gordy and Eva Lütkebohmert. The major conclusion of the thesis is that the relationship between a portfolio’s large and small exposures in terms of assigned PDi is of great importance for how the two methodologies’ resulting capital requirements differ. To further clarify, it is this relationship that affects the difference in capital requirements between the two methodologies, rather than the actual levels of PDi for the big and small exposures.