Time to Grab Extreme Risk by the Tail

Five years ago, the Lehman bankruptcy triggered a crisis of greater scope than any other the modern financial world had seen. Not long ago, people assumed that extreme crises happened once every 100 years. Based upon their frequency over the last 30 years, another crisis may not be far away. Is the financial industry ready for it?

In each crisis, extreme tail risk played a significant role in causing the devastating impact on institutions. Tail risk refers to exposure from uncertainty that is so high that – if it materializes – it can turn into catastrophic losses and overwhelm institutions. Bear Stearns and Lehman are two poster children of extreme tail risk.

Post-2008 calls to "do something" have driven intense public-policy debates, regulatory proposals, and institutional initiatives. Much has been done, ranging from the Dodd-Frank Act to a new Basel III accord to refinement of risk models and new controls in risk management.

However, almost all the steps implemented or proposed to date amount to marginal improvements and don't get to the heart of a critical problem. Considering the near meltdown of the system in 2008, what is needed is a fundamental change.

Two post-2008 observations demonstrate that "doing something" can't be effective unless extreme tail risk is addressed by in a fundamental way.

First, there have been calls for higher capital requirements. This is based on the general idea that more capital is better than less. But how much more? And will higher capital prevent the experience of 2008?

No one, including regulators, has an objective gauge.

Studies show that meeting regulatory capital requirements is no assurance of survival in a crisis. Per "Federal law and OTS regulations … An institution is treated as well-capitalized when … its Tier 1 risk-based capital ratio is 6.00% or greater ... " said the 2007 annual report of Washington Mutual, which appeared to be in fine fettle with a Tier I ratio of 6.84% before its collapse.

The need for a larger cushion is real, but it must objectively relate to the extreme tail risk that creates this need. Simply requiring higher capital is no assurance that a fundamental problem has been addressed.

Second, in the absence of an objective measure of extreme risk, regulators have devised stress testing. It's a step in the right direction, but falls short, as it relies on subjective black-box assumptions. This can create a faux confidence that masks critical vulnerabilities.

The "London Whale" is one example that caught everyone by surprise. Although JPMorgan Chase's earnings could absorb the loss, combined with other London Whales in the portfolio, such a surprise could be devastating in a crisis. And not every institution can absorb such a hit. How many more London Whales are lurking out there?

Unless tail risk is specifically addressed, these steps to "do something" won't amount to a fundamental change. Similar to hurricanes, financial crises will develop periodically. However, their impact and thus the need for capital can be managed proactively. For the following reason, this requires a totally new approach to extreme tail risk that can be easily grasped and implemented alongside risk management.

Financial institutions traditionally manage risk with a one-dimensional focus, even though a second dimension arising from extreme tail risk has very different implications.

Traditional risk management, based on probability theory, works well for revenue models because – priced properly – compensation for quantifiable uncertainty can drive the revenue engine. And, the cost of being wrong may be the loss of profits.

However, the other dimension – extreme tail risk – needs to be handled differently because the cost of being wrong can mean the difference between life and death of the institution.

Today, in the absence of a distinct approach, tail risk is viewed through the prism of risk management, which is designed to manage and protect revenue opportunities. Because of the limitations of a probability-based approach, this prism is incapable of objectively refracting extreme tail risk exposure, which can be devastating.

What is needed is a distinct prism with its own objective parameters for effective tail risk management to protect institutions from becoming casualties in crises. This can be done by turning to the concept of "probable maximum loss," which objectively focuses on exposure from extreme tail risk and is commonly used in the insurance industry. PML is a simple and effective tool to proactively focus on extreme tail risk.

Financial crises are a part of economic cycles. However, their impact on financial institutions and the financial system can be managed. An unequivocal focus on extreme tail risk will enhance the going-concern sustainability of institutions in a crisis and thus lead to a stronger financial system. A fundamental change is needed.

Five years after the last crisis, it is time to do so with urgency.

Karamjeet Paul, managing principal of Strategic Exposure Group, is the author of "Managing Extreme Financial Risk: Strategies and Tactics for Going Concerns" (Academic Press, October 2013), from which this post is adapted.