Farewell
So, farewell to Daniel Kahneman, who died aged 90 this week. I’ve mentioned before that in a previous Smartodds offsite, when Matthew Benham was asked to name his favourite books, Kahneman’s ‘Thinking, Fast and Slow‘ was top of the list.
Kanheman won the Nobel Prize in Economic Sciences in 2002. The motivation for the award was stated as
for having integrated insights from psychological research into economic science, especially concerning human judgment and decision-making under uncertainty
Thinking, Fast and Slow is a brilliant attempt to condense this theory into a text that’s not just readable, but also enjoyable from a non-expert perspective. The Economist is quoted on the back cover of the book in this way:
As Copernicus removed the Earth from the centre of the universe and Darwin knocked humans off their biological perch, Kahneman has shown that we are not the paragons of reason we assume ourselves to be.
The title ‘Thinking, Fast and Slow’ refers to Kahneman’s main thesis that our thought process operate on two levels, or systems as he describes them. System 1 is like human overdrive – our instinctive reaction to events and problems as they occur. System 2 can be interpreted more as reasoning; the analytical approach that might be required when intuition breaks down. Everything Kahneman describes in his book is then viewed through the prism of these two – sometimes competing, sometimes collaborating – processes.
Beyond this basic premise, there is endless wisdom of many types in every chapter of the book. Much of it concerns patterns of individual human behaviour in the presence of uncertainty, which is why it is relevant to both statistical thinking and to understanding market behaviour. prior to Kahneman’s work, it was accepted wisdom in economic theory that while individual’s might be biased in their behaviour and actions, populations of individual’s would behave unbiasedly. This is the premise that leads to the notion that markets are bound to be efficient. Kahneman argued that the various cognitive biases which arise in human behaviour attributable to the dual system of human thinking processes blows a hole in the efficiency argument.
An Example
Daniel Kahneman was working at Princeton University up until his death. In 2015 he gave an insightful interview to the Guardian. The headline of that article, which paraphrases Kahneman, is illuminating:
What would I eliminate if I had a magic wand? Overconfidence
Statistics provides a framework by which overconfidence can be avoided, using rational methods to measure risk and take actions accordingly. But Kahneman’s book is full of examples that illustrate the human tendency to act irrationally, even when armed with good statistical information. Here’s one taken from a chapter called ‘Risk Policies’.
You have to make two simultaneous decisions.
Decision 1: Choose between:
A: An immediate gain of $240.
B: A 25% chance to gain $1000 and a 75% chance to gain nothing.
Decision 2: Choose between:
C: A certain loss of $750.
D: A 75% chance to lose $1000 and a 25% chance to lose nothing.
Look carefully at the options and make your choices.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
It’s standard to approach problems of this type by thinking in terms of expected gains and losses. Choosing A leads to an expected gain of $240. In fact, you are certain to gain $240 this way. Choosing B leads to an expected gain of $250. To maximise your expected winnings you should therefore choose B over A. Both C and D have an expected loss of $750, so by the criteria of minimising expected loss, you do equally well choosing either, and could even flip a coin to make the choice.
However, Kahneman argues that humans have an attraction to certain gains and an aversion to certain losses, making A and C more attractive options to most people. He reports that in an experiment, 73% of respondents preferred A to B and only 3% of respondents chose the pair of responses B and C.
Now let’s play another decision game. You’ve got just one decision to make, this time between two options.
Decision 3: Choose between
X: 25% chance to win $240 and 75% chance to lose $760;
Y: 25% chance to win $250 and 75% chance to lose $750.
Again, make your choice.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
If you are at all rational in your behaviour, you should have chosen Y. The win/lose probabilities are the same in each case, while with Y you will win more if you win and lose less if you lose. In non-technical terms, choosing Y is a no-brainer.
But now look back at decisions 1 to 3 collectively. Option X in decision 3 is equivalent to making the two choices A and D in decisions 1 and 2, while option Y is the same as making choices B and C. The fact that you were asked to make decisions 1 and 2 simultaneously actually means that the option A and D is entirely equivalent to X, while the option B
So, although Y is the indisputable choice to make when the two decisions are combined, most people – perhaps you too – fail to make that choice when the options are broken down into two separate decisions.
The fact that most people tend to be risk averse when it comes to gains (i.e. preferring option A to B), yet risk seeking when it comes to losses (preferring option D to C) is due to what Kahneman describes as prospect theory – the general tendency to view gains and losses asymmetrically. But the fact that individuals tend to sway one way when decisions are broken down individually and another when they are combined points to even deeper limitations in the rationality of human behaviour.
Thinking, Fast and Slow is full of insights of this type. Learning from them – about our individual tendencies to act irrationally – can help us to act more rationally ourselves and also be better placed to benefit from the irrationality of others. For those of us working in an environment that aims to exploit inefficiencies in market behaviour, both these aspects are critical. The Guardian review of the book stated
Its truths are open to all.
It’s up to each of us to learn from those truths.
