The Flaw of Averages: Why We Underestimate Risk in the Face of Uncertainty

CHAPTER 29

Real Portfolios at Shell

In a leather jacket and jeans, riding his Harley Davidson up St. Charles Avenue after work, he blends seamlessly with the New Orleans scene of the mid-1990s. But looks can be deceiving. In his previous five years, he has worked in the Netherlands, Brunei, Sarawak, Thailand, Vietnam, China, Oman, Nigeria, and Canada. Daniel Zweidler is a geologist working for Shell, and he is Swiss.

The Planets Were Aligned

I met Daniel in the fall of 2004, when Stefan Scholtes and I were assisting with a course for Shell executives at the Judge Business School at Cambridge University. We were planning to present portfolio management in the course and had sought Daniel’s input. But Daniel was also working on Shell’s real-life exploration portfolio as well, and invited me to help.

By then several firms had developed software focused on this area (some based on the methods published by Ben Ball and me). But, as discussed earlier, enterprisewide risk models could not easily be built up out of subcomponents. Daniel had experience with some of these complex systems, which, although technically sound, had not gained traction within the decentralized decision-making structure of the firm. This time the planets were aligned.

First, Shell had recently drawn its exploration investment decisions back into their headquarters at The Hague. Matthias Bichsel, then the executive vice president of Shell International Exploration and Production, describes the problem: “I had ...