Act Three: Capital Stress Testing Scenarios And Swan Lake – Taking On The Most Fiendish Of Double Roles

By Mahim Mehra, Senior Risk Advisor, AxiomSL

When the great Russian ballerina Ana Pavlova made her American debut dancing the roles of the white and black swans Odette/Odile at the beginning of the 20th century, she caused a sensation. Most in the audience had never seen anything quite like the technical and artistic demands required of dancing a ‘double role’ such as the one required of the prima ballerina in Swan Lake – a tall order that must have caused Ms. Pavlova a fair amount of stress indeed. But it seems that stress can be caused by more than just a stage performance in which swans are personified. As the current global economic situation evolves post Covid-19, financial institutions are being forced to take a critical look at their capital stress testing practices and tackle a ‘double role’ of their own. While institutions simultaneously look back at what they have had to accommodate in past crises and look forward at how to predict where the global situation may be leading, they must aim to adjust their capital stress testing models accordingly.

Like Knowing The Dance Steps, The Need For Stress Testing Is Critical

In the first act of our miniseries – “It’s Stressful Living With A Black Swan” – we spoke about the need for stress testing scenarios, processes, and technologies that financial institutions can leverage to prepare for systemic or idiosyncratic forces that can severely disrupt capital and liquidity management, as has occurred during this global pandemic black-swan event. We outlined four steps that financial firms can consider as they seek to build industrial-grade risk, liquidity, and capital stress testing processes. Without knowing the steps, institutions are left dancing blindly as they address:

  • Exploring the network of factors that drive institutions’ capital and liquidity needs
  • Implementing best practices to bolster credit and market risk management capabilities
  • Understanding capital conservation and countercyclical capital buffers in times of extreme economic stress
  • Extending connections across the management of operational, credit, and capital risks

Defining The Role Of Capital Stress Testing: As Elusive As Playing The Role Of A Swan?

In Act Two: Liquidity Risk Stress Testing Scenarios – Dealing With Odile, That Villainous Black Swan, we discussed the complex choreography involved in liquidity risk stress testing scenarios and explored why big data is good, but ‘small’ data is what you really need for robust liquidity risk management – especially when the villainous black swan Odile is dancing.

In Act Three, we will explore how to accommodate factors such as changing definitions of default, the role of capital buffers and risk appetite when adjusting models, and important differences between point-in-time (PIT) model inputs versus through-the-cycle (TTC) ones. However, with no exact historical precedent for the coronavirus black-swan scenario, institutions are left wondering if their current models, which are mostly based on IFRS 9 requirements, can accurately predict capital adequacy. They are asking themselves:

  • Can we monitor/configure/adjust our risk appetite accurately to account for both near-and long-term stress testing scenarios?
  • When will regulators halt payment moratoriums for consumers and how will this affect our capital adequacy?
  • We understand that expected credit loss (ECL) and risk will be most affected in current stress testing models, but what other calculations should we consider?
  • How effective is the application of expert judgment on our current models?

The big question is: how can we perform robust capital stress testing with any certainty in the current unpredictable economic environment?

While it may be tempting to talk about the year that was, this year is not over. There will certainly be an Act Four to this challenging ballet. Even as the curtain falls on the black swan Odile, Odette may have to continue to face the music given that it could be nearly impossible to predict whether the economy’s recovery will be V, U, L or K-shaped, or something else entirely.

We Can Assume A Princess Can Turn Into A Swan, But Adjusting Risk Models Will Require More Than Just Fairy-Tale Assumptions

 Your Partner In Handling A Chameleon-Like Swan: Risk and Capital Stress Testing with AxiomSLHistorically, financial institutions have developed probability of default scales using the TTC methodology for assessing return-on-risk capital and expected losses. IFRS 9 methodology, on the other hand, uses a PIT probability of default to mark a loan to market. This approach is based on the impact of current macroeconomic factors, rather than a look through over time that normalizes economic peaks and troughs. Given the black-swan nature of the current pandemic, this thought process has clearly been turned upside down. Firms must intervene in their model inputs to gain accurate outputs that incorporate the current black-swan event – likely needing a healthy dose of expert judgment.

While audiences can suspend their disbelief for the duration of Swan Lake – to accommodate the idea that a good princess turns into a swan while a sorcerer creates an evil mirror image of her with the intention of confusing a well-meaning prince – blindly marking loans to market is not the type of magic financial institutions want to count on. But unless the current pandemic is adequately incorporated in modelling scenarios, such a situation may occur.

At First, Odile Seems Just Like Any Other Swan

When the villain Odile first makes her entrance, she is so enchanting that we are taken in. We almost hope that she is not the evil creation of a mad sorcerer. She seems to look like the virtuous Odette and dances many of the same steps. Alas, there is no foundation to her goodness and we quickly realize she is a fraud. To avoid such a pitfall with capital stress testing models, they must be built on a strong foundation of transparent and validated data. This strong foundation feeds several elements of a capital stress testing infrastructure –
including definitions of default, capital buffers, and risk appetite.

Are Costume Changes Akin To Default Definition Changes?

Capital adequacy measures and related reporting requirements such as the standardized approach to counterparty credit risk (SA-CCR) and the fundamental review of the trading book (FRTB) provide a system-driven mechanism to report wholesale credit and market risk on a financial institution’s books. In these uncharted times, the very definition of default is changing. These changes are the result of:

  • Government interventions on default forbearance programs
  • Economic stimulus funding efforts
  • Regulatory mandated rule changes


Unfortunately, unlike a costume change for the prima ballerina from white feathers to black (to switch from being Odette to being Odile), the aforementioned influences are almost impossible to execute in a model. Thus, future capital stress testing processes for deteriorating credit will have to cater for exceptions, given the effects of Covid-19. To model this, it makes sense for an institution to have a system-driven perspective of their credit and market risk profile. Having this type of information available in a transparent, data-driven platform – complete with daily calculations – will provide the best-case inputs for capital stress testing scenarios.

A Brief Intermission For The Ballerina To Catch Her Breath: Capital Buffer Reprieves

Due to many coronavirus-driven regulation reprieves, the industry has gained some time to reflect on how best to address upcoming Basel IV requirements. Many of the concepts contemplated within the Basel framework can be implemented in capital stress testing processes that take future regulations and scenarios into account. They are outlined as follows:

  • Capital conservation buffer – designed to avoid breaches of minimum capital requirements.
  • Countercyclical capital buffer – aimed to ensure that capital requirements consider the macro-financial environment.


During a crisis period, including the current one, a financial institution’s top line is stressed. At the same time, there is a hugely increased market demand for credit. Decisions by the European Banking Authority (EBA) and National Central Banks (NCBs) to provide fiscal or monetary help to the real economy translates into credit growth. The same phenomenon is true for North America and in most economies around the world. This means financial institutions need to provision for both buffers with stress scenarios and understand their impact in a challenging economic environment. This is especially true as moratoriums expire and institutions assess how credit deterioration will affect their capital adequacy, risk appetite, and business models.

Some Dance Steps Are Trickier Than Others: Adjusting Risk Appetite Assumptions

There is yet another topic to consider: the assumption that although Odette and Odile are different colors, in all other ways they seem to look and dance alike. This is rather like naïvely assuming that because all crises requiring stress testing follow the same pattern, risk appetite should be assessed as a percentage of annual earnings, as it always has been. The annual earnings percentage could be thought of as the level of loss a bank can sustain in a financial year and continue functioning. That amount typically ranges from one to two quarters of earnings, but perhaps it is time to revisit that assumption.

The pandemic’s immediate effect and ongoing aftershocks have shown that a drastic reduction in global economic activity in a very short timeframe is entirely possible. Therefore, it follows that a perspective on risk appetite in consideration of a stressed economic scenario is a desirable approach. What is important to bear in mind, however, is that this perspective could differ from institution to institution depending on what risks they are most prone to. Consequently, it could be dangerous to assume that there is only one set of factors to model.

Taking On Odette/Odile Requires Technical Preparation And Advanced Artistry

Just as a ballerina does not arrive at the rank of ‘prima’ ready to take on fiendishly difficult roles such as Odette/Odile without a formidable technique, institutions will require robust preparation to take on black swans of their own. There are many elements to consider when establishing a comprehensive capital stress testing framework including: understanding the moving target definition of default, assessing capital needs, and managing risk appetite amid unpredictable scenarios. Given the additional element of accommodating the double role of historical look back and predictive future modeling, financial institutions are tasked with a tall order. Organizations should examine the following when building stress testing foundations:

  • Enterprise stress testing – to evaluate internal processes and better inform decision making for risk and capital adequacy
  • Tactical vs. strategic modelling – to use tactical expert judgment in model scenarios short-term, but integrate adjusted inputs and recalibrate models in the long-term
  • Model proliferation – to avoid a veritable model muddle and manage risk within a robust execution framework


When institutions can adequately assess the above factors and stress scenario influences, they are able to flexibly adapt their capital stress testing to the current situation and be prepared for the future, regardless of what type of swan appears on stage.

Your Partner In Handling A Chameleon-Like Swan: Risk and Capital Stress Testing with AxiomSL

AxiomSL’s powerful solutions provide the framework for a wide spectrum of complex Basel-related calculations, including the analytics to produce reports and breakdowns of stress testing scenarios across the risk spectrum. Underpinned by its flexible data dictionary architecture, AxiomSL’s ControllerView® data integrity platform delivers the calculations, calculation histories, optimizations, granular results, analytics, and regulatory reports that satisfy compliance requirements and provide insights for capital and liquidity risk management – via secure RegCloud® deployment or on premises. In addition, the current pandemic-driven black swan scenario presents financial institutions with an opportunity to address the issue of model proliferation by strategically leveraging IntegratedModelView’s platform-native capabilities. IntegratedModelView, running natively on the platform, is a governance framework that can be adapted for credit model management and execution purposes – both for regulatory and internal needs.

The platform holistically delivers:

  • Highly scalable data management leveraging AxiomSL’s unique extensible data dictionary architecture and holistic model management with IntegratedModelView
  • Monitoring of risk appetite and limits across industries, asset classes, and counterparties
  • Fast processing of large daily data and calculation volumes, end to end
  • Results visualization via interactive dashboards


AxiomSL risk solution teams are collaborating to assemble a single cohesive stress testing framework that allows clients to define scenarios across all relevant risk and reporting domains, including credit risk, market risk, liquidity, and balance sheet views. In addition, solutions are being enhanced with advanced data technologies including Spark, Hadoop, and Redshift, to provide high-speed processing and intuitive, user-friendly dashboards powered by AnalyticView.

At the end of her luminous dancing career during which she was inexorably linked to that most famous of double-swan roles, Ms. Pavlova said of the demands of the ballet, “to float so lightly about the stage is the hardest of labors”. Clearly, just as becoming Odette/Odile necessitates both technical and dramatic prowess, risk management, liquidity and capital stress testing frameworks that can handle all varieties of swans will require a sophisticated artistry of their own. AxiomSL’s capabilities can provide financial institutions with the required technical foundation and artistic finesse.



We use cookies in order to give you the best possible experience on our website. By continuing to use this site, you agree to our use of cookies.
Accept