“When the capital development of a country becomes a byproduct of the activities of a casino, the job is likely to be ill-done.” – John Maynard Keynes, cited by Washington Post columnist David Ignatius
In the first presidential debate, the candidates vied for primacy in foreign policy expertise by demonstrating their ability to pronounce the name of Iran’s president, “Ahmandinejad.”
I call it a tie: Both successfully managed all five syllables enough times to prove it wasn’t just luck. It was almost as exciting as a spelling bee, although with fewer surprises.
Speaking of current events, last week’s column explored some of the lessons IT can apply from the current financial crisis. After these last licks we’ll take a break from breaking news for awhile:
It’s never quite that simple: Last week I said wishing doesn’t repeal arithmetic: “Aggregating a thousand loans, each with a one percent probability of failure, creates a very safe investment. Risk calculations don’t work that way, as anyone who has taken Statistics 101 knows.”
The real problem requires Statistics 201. As several readers pointed out, the risks of the thousand loans are correlated, so as mortgages started to default they crashed housing prices, causing more defaults.
Kevin O’Donnell drew the Lesson for IT: When performing a risk analysis, ask yourself, “What could make several of these things fail at the same time?” Don’t put your backup server on the same UPS (better, don’t put it in the same data center).
And ask your multiple ISPs about facilities sharing. Some Silicon Valley companies have spent painful hours out of the eMarketplace because their risk management plan arranged Internet access from two separate providers . . . who shared the same pipes.
Know your sources’ biases. In a recent story about the crisis, The Economist called the financial sector “the economy’s brain.”
Brain? Id is more like it — uncontrolled “I want what I want when I want it” behavior. As Freud might have explained, every id needs a superego to keep it in check.
The Economist has, for all its strengths, a clear regulation-is-bad/private-enterprise-solves-all-problems slant. Hence the overly flattering “brain” metaphor.
Lesson for IT: CIOs need to analyze the bias of every source of information they use (except, of course, Keep the Joint Running, which is unfailingly objective in its accounts of things).
Those with no skin in the game break the game. Deep down in the core of the financial meltdown lies a very soft bedrock of bad mortgages. They were sold by companies that didn’t care at all if they were good mortgages. They sold the paper and moved on to the next borrower.
My business partner suggests a regulation requiring mortgage originators to hold every mortgage for three years before selling it. An alternative: An outside regulator selects five percent of the mortgages sold every month at random, to be serviced by the mortgage originator for the life of the debt.
The result: No underestimated risk, no “liar loans,” no financial meltdown.
Lesson for IT: Give the rest of the business a direct stake in the health of the IT organization and the company’s technical architecture. This isn’t easily achieved, and the most common tactics — IT treating the rest of the business as its internal customer, and charging back for its services — achieve the exact opposite result.
Urgency isn’t crisis. In a true crisis, speed matters most — any decision now is better than a perfect decision too late. When the situation allows for thought and debate, thought and debate are worthwhile investments.
Treasury Secretary Paulson’s original “plan” was, at three pages, shorter than my consulting company’s standard contract, even though the immediate crisis had passed and the situation was merely urgent, not critical. No matter what happens next, I give everyone … including Paulson … credit for thinking, debating and compromising, instead of acting like the crisis is still unfolding.
Lesson for IT: Autocracy works for crisis and art, and I’m not that sure about art. When a decision is important, find ways to involve smart people in its making.
Every boom ends. Even booms that aren’t bubbles eventually come to a close. And yet, company after company … and individual after individual … rely on the money being real and the income sustainable. So when one, the other or both go away, the results are catastrophic rather than inconvenient.
Lesson for IT: Use temporary corporate affluence to invest in sustainable improvements, not in expansions of service.
Lesson for those who work in IT … and everyone else, too: When you’re making good money, put enough away that if you find yourself unemployed, or needing to walk away from your job for any reason, you can.
Fail to do so and you’ve given up control of your own life.
* In Forbidden Planet, the Krell succumbed to “Monsters from the id.” Rent the DVD. It’s old, but the special effects still hold up and the plot is remarkably sophisticated.