Response to consultation on Guidelines on institution's stress testing
Go back
• The draft Guidelines on Stress Testing (GL) duly describe several types of stress tests for different purposes (ICAAP, ILAAP, Reverse Stress Test, Recovery Plan, Strategic Planning, etc.), often associated with different degrees of severity; this plurality of scenarios/severity extent calls for a system that is able to link up these various stress testing exercises, by comparing them and ranking them. What is required is a uniform and univocal measure that could well take the form of the Plausibility-of-Scenario concept. To this end, it would be desirable to have some guidelines for the Plausibility-of-Scenario concept in terms of “probability of occurrence”. Indeed, probability is a univocal quantitative measure, perfectly suited to risk analysis and allowing for equitable comparisons and ranking.
• To this regard, when providing details about how to estimate probability distribution, in addition to past experience and expert judgments, it would be appropriate to explicitly mention the possibility of resorting to simulative methods based on stochastic processes, such as the Monte-Carlo simulation, these being an effective way to assess probability distribution for relevant events.
• (13) Plausibility of scenario: means the degree to which a scenario can be regarded as possible to materialise in respect of the consistency of the relationship of that scenario with the current macroeconomic and financial variables, the support of the scenario by a coherent narrative and the backing of the scenario by probability distribution and historical experiences. Plausibility can be expressed in terms of probability of occurrence of the scenario and in order to assess probability distribution, institutions can resort to historical data and experiences, but also, depending of the own bank’s approach, to i) simulative methods based on stochastic processes (such as the Monte-Carlo simulation), ii) expert judgments that take into account changing risk environments (e.g. observed structural breaks) iii) stress events that were observed in similar risk environments outside the institution’s own direct historical experience which should play a key role.
2. Paragraph 2. e) the methodological details, including models used and possible links between liquidity stress tests and solvency stress tests, namely the respective magnitude of such dynamic interaction and capturing feedback effects
• Regarding this point and specific sound solutions:
a) to take into account links between liquidity and solvency ST
b) to capture feedback effects
ABI would like to have additional information on the possible range of solutions as well as some concrete examples .
In addition, we would like to point out that these issues are amongst the most challenging around. Therefore, the industry requires a longer period of time for implementation as compared to the usual timeframe for full GL compliance.
3. Reverse Stress test)
• We would like more information on the range of acceptable solutions in order to define RST for Business Model viability (which variables? Which level?)
• Should the ST program provide a sound framework for the probability of scenario-occurrence, we would like this to replace Reverse-Stress testing.
Question 1: Please provide comments on all proposals put forward in this paper
1. Plausibility of Scenario Definition• The draft Guidelines on Stress Testing (GL) duly describe several types of stress tests for different purposes (ICAAP, ILAAP, Reverse Stress Test, Recovery Plan, Strategic Planning, etc.), often associated with different degrees of severity; this plurality of scenarios/severity extent calls for a system that is able to link up these various stress testing exercises, by comparing them and ranking them. What is required is a uniform and univocal measure that could well take the form of the Plausibility-of-Scenario concept. To this end, it would be desirable to have some guidelines for the Plausibility-of-Scenario concept in terms of “probability of occurrence”. Indeed, probability is a univocal quantitative measure, perfectly suited to risk analysis and allowing for equitable comparisons and ranking.
• To this regard, when providing details about how to estimate probability distribution, in addition to past experience and expert judgments, it would be appropriate to explicitly mention the possibility of resorting to simulative methods based on stochastic processes, such as the Monte-Carlo simulation, these being an effective way to assess probability distribution for relevant events.
• (13) Plausibility of scenario: means the degree to which a scenario can be regarded as possible to materialise in respect of the consistency of the relationship of that scenario with the current macroeconomic and financial variables, the support of the scenario by a coherent narrative and the backing of the scenario by probability distribution and historical experiences. Plausibility can be expressed in terms of probability of occurrence of the scenario and in order to assess probability distribution, institutions can resort to historical data and experiences, but also, depending of the own bank’s approach, to i) simulative methods based on stochastic processes (such as the Monte-Carlo simulation), ii) expert judgments that take into account changing risk environments (e.g. observed structural breaks) iii) stress events that were observed in similar risk environments outside the institution’s own direct historical experience which should play a key role.
2. Paragraph 2. e) the methodological details, including models used and possible links between liquidity stress tests and solvency stress tests, namely the respective magnitude of such dynamic interaction and capturing feedback effects
• Regarding this point and specific sound solutions:
a) to take into account links between liquidity and solvency ST
b) to capture feedback effects
ABI would like to have additional information on the possible range of solutions as well as some concrete examples .
In addition, we would like to point out that these issues are amongst the most challenging around. Therefore, the industry requires a longer period of time for implementation as compared to the usual timeframe for full GL compliance.
3. Reverse Stress test)
• We would like more information on the range of acceptable solutions in order to define RST for Business Model viability (which variables? Which level?)
• Should the ST program provide a sound framework for the probability of scenario-occurrence, we would like this to replace Reverse-Stress testing.