The Risk One Cannot Afford to Take

You can expect to be reading and hearing a lot more about JPMorgan Chase in the coming weeks, as investigators seek to get to the bottom of things: How do you lose billions of dollars in a flash? The latest, from the New York Times, is an article suggesting that an absent boss was at least partly to blame.

Evidently, Ina Drew, who oversaw the bank’s chief investment office, was sidelined a lot because of a bout of Lyme disease. Back when she’d been at her best, “Ms. Drew attended the regular morning huddle with traders and forced them to defend positions and outline the risks they would face during the approaching trading day,” the Times reported. Once she was in the office less, however, “her unit was making riskier bets.”

As we’ve noted before, Peter Drucker had no objection to risk-taking per se. After all, it’s inherent to moving forward as a business. “A business always saws off the limb on which it sits; it makes existing risks riskier or creates new ones,” he wrote in Management: Tasks, Responsibilities, Practices. “Risk is of the essence, and risk making and risk taking constitute the basic function of enterprise.”

But to approach risk the right way, you have to know what kind or risk you’re taking. “Risks . . . need to be classified,” Drucker advised in Managing For Results. “A risk is small or big according to its structure rather than according to its magnitude alone.”

Specifically, Drucker outlined the following four kinds of risk:

1.     The risk one must accept, the risk that is built into the nature of the business

2.     The risk one can afford to take

3.     The risk one cannot afford to take

4.     The risk one cannot afford not to take

Apparently, the traders at JPMorgan did not approach their work with option three in mind. In any case, the point isn’t that making big bets is inherently bad. The point is that things are bound to go wrong absent a clear thinking about what type of risk is being taken.

What about you? How do you classify risk in your work?