Robust Computation of XVA Metrics for Central Counterparty Clearing Houses

Leif Andersen and Andrew Dickinson

13.1 INTRODUCTION

This chapter continues the discussion of central counterparty clearing houses (CCPs) that we began in Chapter 12. Our focus here will be on practical techniques for calculation of XVA metrics (credit valuation adjustment (CVA) and margin valuation adjustment (MVA), specifically), for both house and client positions of a clearing member. As we saw in Chapter 12, detailed information about the inner structure of a CCP is rarely available, so we emphasise methods with a high degree of robustness and simplicity. Given that default exposures to CCPs are triggered by rare events, the chapter also pays special attention to the modelling of distribution tails.

Most of the key ideas we shall invoke in this chapter originate in Andersen and Dickinson (2018) and can be characterised as a creative use of scaling relationships to condense a large number of unknown micro-structure variables into a few intuitive macroscopic “levers” that can be estimated conservatively from available data and then applied to exposure computations that are easy to execute on a standard exposure calculation engine. A key building block for this approach is the analytical proportionality