Bubble Value at Risk: A Countercyclical Risk Management Approach, Revised Edition

Chapter 13

Market BuVaR

I can calculate the motion of heavenly bodies, but not the madness of people.

—Sir Isaac Newton, mathematician and physicist, (1642–1727)

13.1 WHY AN ALTERNATIVE TO VaR?

The 2008 credit crisis has challenged the foundation and the tradition of value at risk (VaR) as the standard risk measure and the de facto metric for minimum capital. As we have seen in Part Two, the weaknesses in VaR had been well known and debated in academia for some time; for example, see the warnings of Danielsson and colleagues (2001). Unfortunately they have been largely ignored by the industry. It was only during the global financial crisis (2008) that these weaknesses were put to the test. VaR was subsequently criticized by a populist movement and popular press; see Taleb (2007).

Three major weaknesses highlight the need for a better alternative:

1. VaR as a number is not very useful. It may have a meaning, but it is not very useful. Recall that a 97.5% VaR is not the quantity you could lose with 2.5% chance—it is the minimum you could lose with a 2.5% chance. The expected loss could be many times larger. You will be precisely inaccurate if you misunderstand this point! As an analogy: suppose you have a volcano warning system that tells you with precision that the minimum potential loss of life is 50,000 in the next eruption. The next day, the model says, due to a new tectonic reading, the minimum potential loss has gone down to 10,000. You would naturally feel relieved by this ...