Here is an excerpt from an article written by Justin Fox for the Harvard Business Review blog. To read the complete article, check out the wealth of free resources, and sign up for a subscription to HBR email alerts, please click here.

*     *     *

So it was Lyme disease that did it! The tick-borne illness kept JPMorgan Chase’s Ina Drew out of the office for extended periods in 2010 and 2011. And it was during Drew’s absences, according to a richly detailed account in The New York Times, that the bank’s chief investment office, which she ran, began to get into trouble:

The morning conference calls Ms. Drew had presided over devolved into shouting matches between her deputies in New York and London, the traders said. That discord in 2010 and 2011 contributed to the chief investment office’s losing trades in 2012, the current and former bankers said.

Whether this really was the main reason for JP Morgan’s $3 billion (and growing) trading loss or not, it does at least sound like it could be true. Managing risks — especially the hard-to-pin-down, moving-target risks that any financial trading operation has to cope with — inevitably involves arguing. Which is why managing those arguments, as Ina Drew appears to have done brilliantly during the financial crisis but wasn’t around to do for the past couple of years, is so important.

The words “risk management” usually evokes less subjective, more data-driven pursuits. But data and objectivity can only get you so far. Philosopher Karl Popper famously proposed that to be scientific, a theory had to be falsifiable: that is, it had to make predictions that could be tested and possibly shown to be wrong. Popper spent a lot of time thinking about this definition of science and the burgeoning science of probablility, which he called propensity. (This summary is from Wittgenstein’s Poker, a book I’ve been reading):

As far as falsification is concerned, he thought that statements involving stable propensities — such as, ‘The die has a one in six chance of landing on six’ — could be tested by looking at what happens in the long run. But isolated statements of propensity — such as ‘There is a propensity of 1/100 that there will be a nuclear holocaust before the year 2050’ may resist testing, and to that extent exclude themselves from science.

Routine risks like worker safety and even some day-to-day trading hazards can thus be managed successfully with a mechanistic, scientific approach. But the kind of big-picture bets that JP Morgan’s chief investment office made could never be tested, or managed, in that way. Decisions either worked out or they didn’t; given the small sample size it was impossible to test what the true probabilities were.

To navigate such unquantifiable hazards, then, you need to make judgment calls. And that’s where argument (or discussion, or conversation, if you prefer) comes in. You want diverse, even opposing viewpoints. You want to manage their interactions in a way that allows the quieter, less-senior, less-predictable voices to be heard. You probably do want to accord different weights to the arguments of different people, although deciding how to do so (past track record? clarity of argument?) is hard.

*     *     *

To read the complete article, please click here.

Justin Fox is editorial director of the Harvard Business Review Group and author of The Myth of the Rational Market: A History of Risk, Reward, and Delusion on Wall Street.