“I never went to bed with an ugly woman. But I’ve sure woke up with a few.”
-- Country songwriter Bobby Bare
The failure of BP
to “top kill” the Deepwater Horizon well brings to mind Blaise Pascal’s difficult
question, how to strategize the unanswerable question whether God exists.
As he came out, to
assume the positive and be right is marvelous, and to be wrong is of little
effect; to assume the negative and be right is no worse than neutral – but to be
wrong is damning.
Looked on as a
matter of risk management – rather than as a much-debated philosophical or
spiritual conundrum – Pascal’s challenge sits on top of the list of lessons
from the events just ahead of the catastrophic blow-out in the Gulf of Mexico.
That is, if it were
a final examination question in my MBA class – where we search for the lessons
from failures in decision-making across a broad range of business and personal
activity -- it would be hard to improve on the Deepwater Horizon disaster. The
catalog of avoidable errors in analysis and performance is long and troubling –
for example:
- Increased
safety measures can mask heightened risk, especially where early signs of
trouble are ignored as “close calls.”
- Increased
complexity of technology alters the risk profile at the frontier of safety
margins.
- Management
over-ride of on-the-spot warnings discourages honest communication and conceals
bad news under the pressure to deliver success.
- Group-think
and herd behavior are exacerbated by the over-confidence that mistakes past
achievement for future safety.
Briefly put, the
central theme in BP’s failed process was the consistent mis-measure, in its
concededly risky choices, of Pascal’s potential up and downside gains and losses
– related to what the statisticians confuse us all by referring to as Type 1
and Type 2 errors.
Examples abound,
from the modestly consequential to the highly destructive. Consider:
A driver is low on
gas. If he chooses to stop immediately, he gains the certain security of a full
tank, foregoing the possibility of a modest savings. Whereas, if he carries on
in search of a station with a lower price, his maximum upside is a few dollars
gained, but his downside is to be marooned awaiting rescue by the roadside, or
worse.
Or this: The space agency
NASA must decide whether a shuttle launch can safely proceed in cold weather.
Postponement to a warmer day means a modest loss of time, for the sake of
mission safety – whereas pushing for a “go” decision can at best achieve budget
and reputational success, but only with the risk that the frigid conditions
will cause an explosive failure of the rocket’s booster (q.v., the 1986 loss of the shuttle Challenger and the lives of its seven astronauts).
How extensible are
the lessons of BP’s questionable decisions – the use of a riskier well casing,
under pressures of time and budget, or the premature substitution of lightweight
seawater for heavier drilling fluid, under delays costing $500,000 per day (here)?
At the level of
individual risk tolerance, consider two insurance scenarios: Taking a pass on
auto collision insurance saves the premiums, and puts at risk only the
replacement cost of a total wreck. But choosing not to buy long-term health
care insurance runs the uninsured risk of a debilitating and devastatingly
expensive illness.
Or for investors:
excess portfolio concentration in pursuit of incremental performance ignores
the question, “How bad could this possibly get?” – as learned to their sorrow
by Bernie Madoff’s clients and employees with their savings in Enron’s
retirement program.
Consequences vary.
Pick a bad restaurant, and the only loss is an evening’s sub-standard dining.
But fail to hedge a naked short, and the losses could be ruinous.
Corporate risk
management depends for success on similar decision-making dynamics: Is a
bet-the-company strategy acceptable? Is it even recognized?
Whatever the
assessments made by Bear’s Jimmy Cayne or Lehman’s Dick Fuld or Countrywide’s
Angelo Mozilo, they mis-evaluated their exposures to disastrous downsides – no
less than the bureaucrats of NASA or the over-seers at BP.
On the other hand,
betting Pascal’s wager the wrong way risks the wrath of an angry deity – which
in the long run may be worse.
Thanks for joining this dialog. Please share
with friends and colleagues. Comments are invited and welcome, and
subscriptions sign-up is easy – both at the Main page.
Comments