Stress Test Scenarios for Banks & The Challenges They Face
By Lee Thrope, Head of Risk Business Solutions
Stress test scenarios for banks & the challenges they face
Since the early 1990s, large financial institutions have been using stress tests (a series of complex tests designed to evaluate how a bank’s balance sheet responds to harsh economic changes) and scenario analysis to understand how they would cope in hypothetical crisis situations.
These stress tests are simulated using a complex library of models and the stresses applied can be light to severe to produce more dynamic results. This approach enables banks and financial institutions to understand how they would perform under different circumstances.
Of course, stress testing is nothing new: before the financial collapse of 07-09, stress tests were typically undertaken as a process of self-evaluation, but regulatory bodies decided that, in order to ensure the stability of the banking sector, they would develop their own stress tests that banks would have to apply.
As a result, banks and financial institutions are now expected to undertake more comprehensive analysis of critical risks, meaning stress test scenarios for banks are more complex and structured. Credit risk, market risk, liquidity risk, capital risk (and others) are all analysed to determine the bank or financial institution’s ability to withstand crisis situations in hypothetical scenarios. In addition to this, risk and financial information are blended together to provide a more holistic view.
With increasing levels of compliance and reporting requirements, banks and financial firms constantly need to evaluate not only their technology investments to deliver stress testing and scenario analysis, but also assess how they are organised.
All of this makes the process of stress testing increasingly complex – and with regulatory bodies requiring different standards, financial institutions need to be able to apply stress tests with consistency across the enterprise.
What are the challenges banks and financial institutions face?
Managing data and legacy systems
Different tools with different definitions will make it difficult to obtain a single version of data across all departments. Subsequently, when it comes to the pooling of data, it’s incredibly difficult to aggregate it, analyse it and develop a consistent and repeatable process.
For financial institutions to interrogate the data and develop accurate, actionable insights, that data needs to be evaluated for quality and then consolidated into a single, central platform. The platform should be both flexible and scalable to meet current and future stress testing needs, as well as capable of: creating data hierarchies, aggregating data for major risk models across the organisation, running stress tests with multiple parameters, providing an enterprise-level view of financials (including market, credit and liquidity risk) and include tools for the
documentation, management and enhancement of models. With all the above integrated, stress test scenarios for banks can be run much easier.
Inconsistent processes and model development
Stress testing methods and analytical outcomes need to not only be consistent with how the organisation looks at risk and reporting, but also meet regulatory guidelines. However, a major issue that many banks face in this process is inconsistency. This typically arises from legacy systems and siloed processes throughout the model lifecycle, and insufficient infrastructure to facilitate the development of those models.
Enterprise-wide stress testing starts with a single platform, such as the SAS Stress Testing Workbench, that operates across the organisation, is supported by built-in workflows that allow for the review and approval of results, and can automate stress test scenarios for banks. The results acquired from those stress tests can
then validated by financial experts who can provide data-driven insights.
Changing or increasing regulatory expectations
Every regulatory change forces banks and financial institutions to reconsider their approach to risk analysis – and
regulators too expect banks to use stress testing as part of ongoing management and not just for regulatory requirements.
However, for most banks, running stress tests on a routine basis is costly and time-consuming – the resource needed to undertake such an activity is perhaps out of reach for many. Therefore, a cost-effective solution, one which can automate the process, manage models and apply those models in accordance with regulatory requirements is essential.
Banks struggle due to reliance on manual processes and issues related to developing, benchmarking and executing models. Conducting stress tests manually takes far too much time and effort, and the results themselves may be subject to inconsistencies. For some banks, the solution has been to assign more people to the process in the hope of speeding it up, but more people can result in more problems.
Instead, banks can use automation to streamline and scale the process. With frequent stress test scenarios, banks move it from being a regulatory-driven exercise to a standard and routine practice that allows them to better understand their risk profile and plan for the future.
How can SAS help?
The complexities around the stress testing process, as well as increased regulatory pressure, has made it ever more difficult for financial institutions to manage.
To help financial institutions be compliant with regulations and run effective stress testing, we have developed a suite of integrated, enterprise-grade tools for stress test scenarios for banks.