The CFO in the Corporate Brain

A Nobel Prize-winning economist has some ideas that help explain how a CFO should function as part of the decision-making team.
Andrew SawersJune 7, 2012

How do you think? How do you decide? One man who has a few ideas is Nobel Prize–winning economist Daniel Kahneman, author of the best-selling book Thinking, Fast and Slow, who is said to be the father of behavioral economics. At a recent lecture given at the London School of Economics, he put forward some thought-provoking notions that provide some insight into the role of the CFO as part of the “corporate brain.”

At the heart of Kahneman’s theories lie two types of thinking:

  • System 1 produces fast thinking. It is intuitive, automatic, responsible for most of our decision-making, much more influential than we realize, and, crucially, it sometimes makes mistakes. It can cling to things it thinks it knows or trusts. Knowing that 2+2 = 4 or driving along a quiet, familiar road calls upon System 1 thinking.
  • System 2 produces slow thinking. It is deliberate and controlled, requires attention, and really kicks in when System 1 thinking receives a knock that disrupts it. Making sense of driving in a strange, foreign country on the other side of the road calls for System 2 thinking.

System 1, in other words, is a system for jumping to conclusions — which is great as long as it jumps to the right conclusions. In truth, Kahneman argued, most of our thinking is of the System 1 type: the point is, we are not, after all, the rational creatures we like to think we are.

Without delving into the science of it all, a number of lessons come out of this. One is that when important decisions are being made, fast System 1 thinking has to give way to slower System 2 thinking. “You give more weight to solid evidence and less weight to your impressions,” Kahneman explained. “Think slower for important decisions. You can recognize clues that tell you you are making a mistake and slow yourself down, but avoid paralysis through analysis.”

This can help make boards more effective, for example. As Kahneman explained, “A board is in the business of producing decisions. We have quality controls for other parts of [the business]. So our first instincts are that quality control means, in part, reviewing the evidence. But there is another aspect of quality control, which is the process: what went on as the decision was made? What [cognitive] biases were likely as the decision was made?”

These cognitive biases, it turns out, include such things as clinging to a failing strategy rather than scrapping it and adopting a new one, or the belief that particular successes are due to skill while a notable failure is sheer bad luck. Kahneman talked about a need for some sort of independent oversight while recognizing that boards don’t like having someone “looking over their shoulder.”

Another interpretation of Kahneman’s approach is that, in a corporate context, the CEO could be said to be a System 1 type of person in the corporate psyche while the CFO is System 2, which is probably why investors and journalists are often swept along by the visionary spiel of the CEO rather than the evidence-based presentation of the CFO.

It’s also apparently true that people are convinced by stories, not statistics. Having made a decision using System 2 thinking, it seems to be important to use the language of System 1 thinking to communicate that decision. People are said to be more influenced to take action by being told that 94% of people pay their taxes on time than they are by the threat of fines for late payment, for example.

In other words, the CFO has to not only master the numbers but also understand and be able to communicate what those numbers are saying. What, in a nutshell, is the numerical narrative, for that is what will have the most effect on the listener. “If we’d anticipate intelligent gossip as a reaction to our decisions rather than malicious gossip,” said Kahneman, “we’d all make better decisions.”

Andrew Sawers is editor of CFO European Briefing, a CFO online publication.