Over a period of time, with proliferation of internal operational data and external data banks, risk analysts, market intelligence, rating analytics and other agencies aiding risk management evolve innovative tools in banks to develop more robust resilience. Some prove more effective.
Some redundant tools have to be phased out whereas some more are to be modified to suit the change in risk management architecture. In order to test whether a risk management tools possesses the degree of resilience to withstand risk shocks, stress testing by developing simulated risk exposure, will be essential.
With change in business priorities and increasing global connect, the complexity of risk is increasing. In order to better apprehend the impact of such market gyrations and de-risk the bank’s balance sheet, it will be essential to put systemic controls in place.
A constant prowl on the stress in risk is to be measured to take suitable measures to mitigate it and also plan to transform the policy framework and business dynamics to suit the changes in the work environment. More important will be to disseminate the assessment of stress culture among line management.
Stress testing system
In this context, development of data science and data analytics led to more use of stress testing as an institutionalized format to manage and assess resilience of tools of risk management. Stress testing is a simulation technique often used in articulating operational business to test its efficacy in the banking industry.
It is also used on asset and liability portfolios to determine their reactions to different financial situations – maturity profile of assets and liabilities. Additionally, stress tests are used to gauge how certain stressors will affect a bank, bank’s constituents such as a company, industry or specific portfolio. Stress tests are usually computer-generated simulation models that test hypothetical scenarios; however, highly customized stress testing methodology is also often utilized.
In investment portfolio management, stress testing is also commonly used for determining portfolio risk and setting hedging strategies to mitigate losses. Portfolio managers use internal proprietary stress testing programmes to manage and test their portfolios against market occurrences and potential events.
Asset and liability-matching stress tests are also widely used in business and investment management. Banks and their constituents/companies to ensure proper internal controls and procedures can use this format of stress testing. Retirement and insurance portfolios also greatly utilize stress testing to ensure efficient streams of cash flow and payout levels.
Basel Committee on stress testing
In 1996, the Basel Committee published criteria for the calculation of market risk capital requirements for banks using internal models. Stress testing has a significant position under these requirements, as well as in the draft amendments to the European Union directives.
Stress testing in draft directives is examined more closely in a consultation paper of the Committee of European Banking Supervisors (CEBS). The issue of stress testing is also being addressed by central banks through various specialist groups and organising conferences to disseminate better knowledge in risk management.
The entry of stress testing is to a large extent related to the introduction of models into bank risk management. The models themselves represent a simplification of reality and may be used to measure bank risks under certain assumptions. The assumptions differ depending on the particular model and the risk measured, though most models are based on the assumption of normal or ordinary development of risk factors.
However, the real market situation includes extreme and exceptional situations, which these models and their assumptions are not able to take account of. The main aim of stress testing is therefore to analyse any such gaps in bank risk management.
Core principles of stress testing
While banks should have an internal policy of stress testing designed in terms of central bank guidelines specifying the areas of risks to be covered, periodicity, data base, peer bank status, industry profile, business dynamics, macroeconomic data and so on.
Large-size banks with more complex portfolios may be expected to adapt suitable approaches to stress testing that are more demanding on knowledge and data. Then only the utility of stress testing can be enhanced to manage risk. Each stress testing process begins with the identification of the risks to which the stress testing will be applied.
The identification is based on the analysis of the bank’s portfolio, whose structure is to a large extent determined by the materiality of particular risk categories. Also important is the analysis of the external environment affecting the exposure of bank positions.
Once the material risks have been identified, it is necessary to select the relevant risk factors. They may be defined, for example, on the basis of past experience. Therefore, the identification of relevant risk factors may be carried out using an analysis of the reasons for historical losses and the circumstances in which they occurred. The number of selected risk factors depends mainly on the complexity of the portfolio and the nature of the risk to which the bank is exposed.
The calibration of stress scenarios is to a large extent a subjective matter for the bank. Unlike risk measurement by means of various models, stress scenarios do not have a defined probability of occurrence. As for its starting point, the bank may use past market events (historical scenarios) or create hypothetical scenarios based on its own expert assessments and own conception of the risk.
The severity of the scenarios or the size of the shocks is therefore in line with the bank’s ‘risk appetite’, in other words, the level of risk, which it is willing to accept. As regards the calibration of stress scenarios, it is generally the case that the scenarios are based on exceptional but plausible events. Central to this calibration requirement is the designing of scenarios that may realistically occur in regard to the bank’s external environment and portfolio.
Summing up merits of stress testing
Stress testing is a globally proven tool to keep a watch on the future risks by simulating risk environment, finding out the probability of loss, working out strategies to mitigate such emerging risks. Since it is a holistic process of factoring all probable events with defined consequences on bank’s risk, it enables better preparedness to manage risks, which have not yet been encountered.
Since a stress test framework includes a common infrastructure well designed to automate stress testing and support multifactor and complex models to measure the effectiveness of risk management tools, it can be better aligned to the bank. Customization of stress testing is the crux of its role in risk management. The better aligned it is, the more useful it can be to mitigate risk. It all depends on banks to build a robust data bank with external data well factored to make stress testing a comprehensive exercise of merit to mitigate risks. With data integrity standards progressively improving, the reliability of stress testing as risk management tool is also concomitantly increasing.
(Dr. K. Srinivasa Rao is Director, National Institute of Banking Studies and Corporate Management – NIBSCOM, Noida, National Capital Region – NCR, Delhi. India.
The views are his own)