5 rules for taking the stress out of stress testing

When you want to know how things really work, study them when they’re falling apart.

– William Gibson, Zero History

In the shadow of every portfolio lies the portfolio that could have been, had the portfolio manager made different choices. You can make choices, but you cannot choose outcomes. Even a portfolio built with the right choices can still turn out to have had the wrong outcome.  

In investing, only one thing is certain – you will experience drawdowns at some point. Nothing goes up forever, and returns aren’t linear. While drawdowns are inevitable, their size and your portfolio’s precise position on its path towards them are difficult to predict, even in hindsight. Some investors believe that drawdowns cannot be guarded against; that history is the best teacher. Others say history is to be learned from but does not repeat or even rhyme (sorry Mark Twain); that the future must be written anew. 

When considering the impact of multiple plausible futures on your portfolio, it’s essential to deconstruct before you reconstruct. Begin by breaking down the event into its key components: the ‘what’ and the ‘how’. Then, estimate the ‘when’ and the ‘why’. You’ll need a flexible platform like Axioma Risk to isolate each component, enabling you to make independent and relevant decisions for each. This approach helps you build a more representative and relevant stress test, providing real and actionable insights on how to handle those possible futures. 

The two types of stress tests we will discuss below focus on historical and transitive stress tests. Transitive stress testing is to historical stress testing what historical fiction is to history. It provides context, helping us navigate through history to gain insights. Flexibility in this context is essential to model the known aspects of the unknowable, allowing you to best quantify its potential impact on your portfolios based on what you know or can forecast with reasonable confidence. Anything beyond ‘reasonable confidence’ involves too much guesswork to be valuable. 

In this note, we call out five best practice ‘To-Dos’ to ensure your stress testing delivers actionable insights from this analysis.

Understanding and managing portfolio risk in an uncertain world

Effective stress testing is both an art and a science, requiring a thoughtful balance of historical data, forward-looking scenarios, and robust risk modeling. While both historical and transitive stress testing approaches have their unique advantages, the most robust framework combines both.  

By following these best practices, risk managers can build stress tests that not only highlight potential vulnerabilities but also provide actionable insights for portfolio management. The key is to maintain flexibility in stress testing frameworks while ensuring scenarios remain plausible and actionable.