We’re all rational decision makers? Think again.

We say we make decisions based on an assumption of a rationality that maximizes. In reality, we are imperfectly rational creatures who make sub-optimal decisions. Time to re-visit our assumptions.

The model of rational choice makes several assumptions. It assumes that our goal is to maximize self interest. It assumes that different dimensions of our decisions are “commensurable” (that is, comparable on a common scale—say a “utility” scale). And it assumes that we act with complete information and can meaningfully assign probabilities to every outcome. (…)

What modern research tells us is that we are imperfectly rational in two different respects.

  • One is that we do the “math” wrong; we are bad at thinking about uncertainty and at creating relevant and accurate spreadsheets in our heads.
  • The second is more important: we often want the wrong things.

We mispredict how much satisfaction, or utility, a given outcome will give us. We discount future consequences of decisions too steeply (and thus eat and spend too much, and exercise and save too little). And we mispredict how long a given decision will satisfy us. That is, we tend to ignore the fact that we get used to good things so that they provide us with satisfaction for a much shorter time than we imagine.

The model [of rational choice] is not fine. Virtually all of the assumptions built into it about human beings and the world are false:

It assumes that people are self-interested. Well, yes and no. Self-interest is certainly part of what moves us, but we are also interested in the welfare of others (…), and even the world. And we are also interested in doing what’s right (…).

It assumes that there is a common scale of value on which everything can be compared. There isn’t. Sure, we can assign value numbers to things like salary, colleagues, being close to our families, and the like, but in doing so, we are only kidding ourselves that these numbers actually represent a common underlying metric. Tradeoffs are hard to make, and often can’t be made formulaically. (…).

It assumes that we can attach meaningful probabilities to outcomes. Sometimes we can, but life is not a roulette wheel or a series of coin flips, in which probabilities are well defined. The world is a radically uncertain place, and we deceive ourselves if we think we can always attach numbers to our uncertainty.

via Barry Schwartz.

Managers who claim to know the future are more often dangerous fools than great visionaries

As complex systems go, the interaction between the ball in flight and the moving fieldsman is still relatively simple. In principle, most of the knowledge needed to compute trajectories and devise an optimal strategy is available: we just don’t have the instruments or the time for analysis and computation. More often, the relevant information is not even potentially knowable. The skill of the sports player is not the result of superior knowledge of the future, but of an ability to employ and execute good strategies for making decisions in a complex and changing world. The same qualities are characteristic of the successful executive. Managers who know the future are more often dangerous fools than great visionaries.

(…) Good predictions may be available in structured, well-ordered, situations – but, even then, forecasts are properly conditional or probabilistic. There are few certainties about the future: but one is that hedgehogs who make confident statements on the basis of some universal theory will be as persistently misleading counselors in the future as in the past. And that the foxes (…) who scramble everywhere for scraps of information will provide better, if more nuanced, advice.

via John Kay.

Manager or Computer?

Do you think your high-paid managers really know best? A Dutch sociology professor has doubts.

The professor, Chris Snijders of the Eindhoven University of Technology, has been studying the routine decisions that managers make and is convinced that computer models, by and large, can do it better. “As long as you have some history and some quantifiable data from past experiences,” Snijders said, a simple formula will soon outperform a professional’s decision-making skills. “It’s not just pie in the sky,” Snijders said. “I have the data to support this.”

(…) Studies over the years have shown that models can better predict, for example, the success or failure of a business start-up, the likelihood of recidivism and parole violation, and performance in graduate school. They also do better than humans at making various medical diagnoses, picking the winning dogs at the racetrack and competing in online auctions.

(…) The main reason for computers’ edge is their consistency, or rather humans’ inconsistency, in applying their knowledge. (…) And critically, models do not get emotional.

They allow an organization to codify and centralize its hard-won knowledge in a concrete and easily transferable form, so it stays put when the experts move on. Models also can teach newcomers, in part by explaining the individual steps that lead to a given choice. They are also faster than people, are immune to fatigue and give the human experts more time to work on other tasks beyond the current scope of machines.

However,

(…) Many in the field of computer-assisted decision-making still refer to the debacle of Long-Term Capital Management, a high-flying hedge fund whose founders included several Nobel laureates. Its algorithms initially mastered the obscure worlds of arbitrage and derivatives with remarkable skill, until the devaluation of the Russian ruble in 1998 sent the fund into a tailspin.

“As long as the underlying conditions were in order, the computer model was almost like a money machine,” said Roger Pielke Jr., a professor of environmental studies at the University of Colorado whose work focuses on the relation between science and decision- making. “But when the assumptions that went into the creation of those models were violated, it led to a huge loss of money, and the potential collapse of the global financial system.”

The fact of the matter is

In such cases, “you can never hope to capture all of the contingencies or variables inside of a computer model,” he said. “Humans can make big mistakes also,” he said, “but humans, unlike computer models, have the ability to recognize when something isn’t quite right.”

Manager 1, Computer 0

As long as managers make a strength of their “weakness”: that they acknowledge and correct their mistakes.