The Stress Test
One of the ironies of the recent oil spill debacle in the Gulf of Mexico is that the oil and gas industry is most often credited with devising and putting to use so-called scenario planning. This management tool is meant to anticipate major changes in the environment – from disaster to depression, to blown pipes in deepwater rigs – and to enable organizations to create plans for immediate strategic response.
Formal scenario planning for companies got going in the 1970s, presumably with the rise of management consulting businesses. Though oil prices had remained stable since World War II, leaders at Royal Dutch Shell worried that disruptive change could happen with severe adverse effects on their business. Among the disruptive events they feared was a sudden increase in the price of oil caused by the emergence of the Organization of Petroleum Exporting Countries (OPEC) cartel.
The Yom Kippur War of 1973 strengthened Arab opinion against “the West”. Furious at the emergency re-supply effort that had enabled Israel to withstand Egyptian and Syrian forces, the Arab world imposed the 1973 oil embargo against the United States and Western Europe and as a consequence oil prices shot up dramatically.
Many oil companies struggled with the effects of the new competitive dynamics. Shell on the other hand thrived. They had prepared a plan – a scenario plan – for what they would do as these circumstances unfolded, and they implemented their plan while others were just gathering to deliberate on their next actions.
I had lunch with my Dad in London last week and he talked about how BP, who were a client of his, regularly (twice yearly) used to run intense “disaster recovery” situations for their senior management – this was back in the 1990s. These situations would be set-up with actors and the whole movie industry paraphernalia, and would be an institutionalised part of senior management’s diaries. We presumed that given the recent deepwater drilling shocker, and Tony Hayward’s foot-in-mouth media soundbites, that these sessions must have fallen in their importance or by the way-side completely.
In one of these trial scenarios there was a simulated explosion at a refinery pipe on the south coast of England which had left an unspecified number of BP workers dead, and as it turned out at a later briefing session, a primary school in danger from further explosions. My Dad’s client, who had been part of this scenario response, told him that things had gone very well until he got into a lift while taking a break. He had started chatting to someone about how they were getting on and told this person that solving the primary school issue was going to be a tricky public relations tester. The lady in the lift, who was an actress, then told him that she was a reporter for the Daily Mail and that his little insight would be tomorrow’s front page news.
His lesson for the real disaster situation had been learned – be careful who you speak to at all times. All painfully real and as a consequence very effective management training because of the seriousness with which the scenario was taken.
Before Lehman Brothers collapsed in late 2008 its risk management teams would have been performing scenario analysis of a kind on a daily basis. Aside from the daily Value at Risk metrics for calculating the daily potential losses across its balance sheet, they had disaster recovery plans in place ever since the World Trade Center attacks in 2001. These plans would ensure business continuity in the event of another such incident – crucial in a trading business where hundreds of millions of dollars are at risk on a day-to-day basis.
One of the fundamental issues with many of the theoretical “what ifs” that many organisations, like Lehman, test themselves with is that too many “what ifs” are dismissed as so unlikely as to be worthless in their testing. If senior management at Lehman had been tested with the questions: What if housing prices fall by 15% year on year? What if money markets suddenly seize up, so that interbank liquidity is virtually nil? What if a AAA rated counter party suddenly declares massive fraud and goes into bankruptcy proceedings?
In the middle of 2007 all of these questions would have been dismissed as “six sigma” events, i.e. so unlikely as to not be worthwhile thinking about as there was no statistical precedent. With hindsight, I’m sure there are many people who wished they had prepared answers to these questions, and put a framework in place to deal with such unlikely probabilities. Even if they had merely asked the questions then the answers would have made them better at considering the direction the business should be taken.
Our systems are never as resilient as we would like to think they are.
The latest budding failure in the scenario planning world is developing around the stress tests that are being imposed on 91 of Europe’s biggest banks. As a recent Credit Suisse report noted: “We doubt that stress tests would be announced if they were going to disappoint the market.”
Stress-tests are the Catch-22 of banking supervision. Their purpose is to demonstrate that banks are safe. To do so, the tests must be rigorous. Yet because their purpose is to bolster confidence, their rigour is bound to be doubted. Unless the tests reveal that the banks are in a terrible mess. In that case, the stringency of the tests would be proven, but the primary mission of reassurance would have failed horribly.
The tests apparently don’t look at the what would happen if Greece defaulted, nor if there was a liquidity breakdown in the interbank markets as a consequence. Yet these are two scenarios that the markets perceive to be highly probable.
The trick is, as with the folks at Royal Dutch Shell in the 1970s, to do your stress testing early and often. For the European banking system this is all coming too late – the answers will be either too unpleasant or too unrealistic to be of any great use.
Leave a Reply
Random Posts
On the third Friday of every month the US Bureau of Labor Statistics releases its preliminary data on non-farm payrolls, which are seen widely as a barometer of the broader economic picture in the US. Markets move significantly on the basis of these figures, so there is always a keen sense of anticipation on trading floors about what "the number" will be.
//
Read More
