Several different stress-test exercises are carried out on BBVA Group’s trading portfolios. First, global and local historical scenarios are used to replicate the behavior of an extreme past event, such as the Lehman Brothers collapse or the “tequilazo” crisis. These stress tests are supplemented by simulations in which the aim is to generate scenarios that have a significant impact on different portfolios, but without being anchored to any specific historical scenario. Finally, fixed stress tests are performed for certain portfolios or positions that significantly affect the market variables involved in these positions.
Historical scenarios
The benchmark historical stress scenario for the BBVA Group is Lehman Brothers, whose bankruptcy in September 2008 had a significant impact on the behavior of financial markets around the world. The most relevant effects of this historical scenario are as follows:
- Credit shock, mainly reflected in the increase of credit spreads and downgrades of credit ratings.
- Increased volatility in many of the financial markets, giving rise to major price changes in the different assets (foreign currency, equity, debt).
- Liquidity shock in the financial systems, reflected by a significant movement in the interbank curves, particularly in the shortest ends of the euro and dollar.
Simulated scenarios
Unlike the historical scenarios, which are fixed and thus do not adapt to the composition of portfolio risks at any time, the scenario used for the economic stress tests is based on resampling methodology. This methodology uses dynamic scenarios that are recalculated regularly according to the main risks held in the trading portfolios. In a sufficiently broad data window that can include different stress periods (data are taken from 1-Jan-2008 to date), a simulation exercise is carried out through resampling historical observations. This generates a distribution of losses and gains that allows an analysis of events that are more extreme than those occurring in the selected historical window. The advantage of this methodology is that the stress period is not pre-established. It is a function of the portfolio held at any time, and as there is a high number of simulations (10,000) an analysis of the expected shortfall can be made with greater richness of information than the information available in the scenarios included in the VaR calculation.
The main characteristics of this methodology are as follows:
- The simulations generated respect the correlation structure of the data.
- Flexibility in the inclusion of new risk factors.
- Allows a great deal of variability to be introduced in the simulations (desirable when considering extreme events).
