COVID and the war in Ukraine caused the supply chains disruptions and the commodity market prices elevating to the levels never seen before. Some analysts are predicting upcoming severe recession, corporate and mortgage insolvencies, a house price crash, more sovereign debt crises and an all-round credit crunch.
Black swan events. These events are rare and have a high impact. We have seen a number of such events in the recent past, e.g. the crash of 1987, dot com bust of 99-2000, 9/11 or the great recession of 2008.
What makes an event a tail event?
Tail events are events with a very small probability of occurrence but at the same time with a huge impact if they occur. Put simply: worst case scenarios.
Tail risk represents the loss that the portfolio will face in case of an occurring tail event. Tail risk is a risk that can be applied to any type of risk, to credit, market, liquidity, or operational risk. When tail events occur, it depends on the nature of the event itself and on the business activity and portfolio of a company, which risk type is most affected by this tail event.
Following this definition, the COVID pandemic as well as the war in Ukraine which caused the energy crisis are tail events.
Does current market situation have an impact on the definition of Tail Event?
The current market situation reminds us that tail events can occur, and sometimes with a higher probability than expected. The fact that more than one tail event has occurred within a very short period leads to the question if we need to re-define Tail Risk Events, considering that events that were once defined as Tail Events now seem to be "rather normal“ in terms of their likelihood.
The current tail events are not worse than the ones we know from the history. But they seem to occur much more frequently, one after the other, which could mean that we need to consider a higher probability of their occurrence. This would lead to an extension of the very definition of Tail Risk.
How do we manage Tail Risk?
Currently, the management of tail risk is often largely or even completely neglected due to the very low probability of occurrence following the generally accepted definition of tail risk events.
With the tail events which occurred recently, the market has suffered from their drastic consequences and high losses are associated with it. Following the assignment of higher probability of occurrence to scenarios previously considers as unlikely tail events, tail risk needs to be integrated in the existing risk management frameworks.
Finding and maintaining the balance between the low probability of occurrence and the potential high losses always were a major challenge. But now, with the increasing probability of occurrence, the consideration of tail risk is more important than ever before.
Challenges in the measurement of Tail Risk
Predicting a "black swan event" is impossible by definition, but a procedure to quantify tail risk is still required. For this, reasonable VaRs method that best represent the nature of the underlying risk factors should be calibrated to a distribution.
Ways to measure Tail Risk
- VaR provides a threshold return level that separates the normal from an extreme regime, and thereby assists risk managers to efficiently prepare for such event:
- Historical VaR: based on very few return values, it tends to change incrementally over time
- Fat-Tailed-VaR: better models financial returns
- Expected Shortfall (or Conditional VaR): expected average loss in the event that the unexpected loss is greater than the value-at-risk.
Tail Risk in the Banking Industry
The transformed RWA requirements in Banking industry facilitated move from VaR and stressed VaR to Expected Shortfall. Under Basel II, the RWA calculation approach was based on VaR methodologies, using a 10-day liquidity horizon and 99% confidence level. 2008 crisis proved that insurance instruments guaranteeing the 1% tail events are indeed not risk-less, and a time frame of 10-days is not sufficient to exit or hedge trading book exposures without significant mark-to-market losses.
While expected losses are already taken into account and managed for credit risk purpose through the pricing of a portfolio, unexpected losses also have to be mitigated as well. For that the Expected Shortfall is calculated, which is widely accepted to appropriately model tail risk.
In addition to the quantification of unexpected losses reverse stress testing is a requirement for some financial institutions but is not currently used widely by other industries. Reverse stress testing can provide management with a different lens on going concern and risk assessments and increase their robustness.
What the financial crisis in 2008 was for the banking industry could now be the current market situation for energy trading companies. That means companies need to implement robust method to measure and manage tail risk for their portfolio.
Initial impact analysis
Trading firms should perform impact analysis to identify the risk types which are impacted in case of Tail Events, considering different Tail Scenarios and impact on their portfolio. Additional risks should be identified with the review of existing baseline scenarios. The purpose is to ensure that all possible outcomes and potential risks of a given scenario are considered.
“Nothing is impossible” scenarios
Risk managers should review existing adverse market scenarios and include into your framework “nothing-is-impossible-scenarios” by enhancing your worst-case considerations. They should also reconsider the assigned probabilities of occurrence to certain worst-case events.
Revision of implemented methods and strategies
Review and revise your contingency plans and hedging strategies to take new scenarios und probability of occurrence of certain events into account.
Adoption of new procedure and processes
Successful management of tail risk should include the adoption of adequate procedures and processes. These should include dynamic reduction of the VaR confidence interval in crisis mode or increase of probabilities for unexpected events.