• Christian Schitton

Irrational behaviour from Behavioural economics angle

Intro

In one of the last articles we analysed the appropriateness of subjective decision making in complex market environments. Now, we go a bit ahead and have a look at how our emotions and our mind are creeping us into irrational behaviour. We take this look from the behavioural economics' point of view which will give a further notch of how important data driven decision making processes are.

To begin with, we have to ask ourselves: Do we act in a rational manner when exposed to difficult situations? Does experience give us the necessary cushion to keep afloat in distressed circumstances? Are all market facts properly incorporated when putting economic expectations into a new business? Can we trust our intuition when deciding on buying an asset? Do we even get when we start to act irrationally?

I am afraid, the answer is no.

I've prepared some showcases (and a joke about the upcoming winter) why business leaders cannot fully rely on intuition and experience in making decisions and why it is so important to be assisted by a well adapted management information system as a rational supporting frame. And as there are almost always future developments - in other words uncertainties- involved in decision making, an efficient data driven decision making process backed by the latest tools, artificial intelligence and causal inference may offer, can be a big advantage. If we really think about it - an essential one!


Loss Aversion

A short question:

  • You have a 95% chance of receiving EURO 10,000,-- or you have a 100% chance of getting EURO 9,499,--. What would you choose?

Another question:

  • You have a 95% chance of losing EURO 10,000,-- but a 100% chance of losing EURO 9,499,--. What would you choose now?

Test settings showed that confronted with the first question most people would decide to take the EURO 9,499,--. After all this is a sure thing. Though, the majority of respondents in the second question would bet on the 5% chance of avoiding a loss. Although, this would mean a 95% chance to realise a significant higher loss than in case of just accepting the sure loss of EURO 9,499,--. But why?

In principle, people are drawn to sure things over probabilities even when the probability is a better choice. This is called the Certainty Effect. On the other hand, in order to avoid losses people are ready to take greater risks than they would be ready to take in case of possible gains. This is called Loss Aversion. In other words, we start to play safe on gains and start to gamble when faced with losses! In fact, emotions are in full drive here and jump in as soon as people have to choose between options while risk is involved. This pattern was summarised under the term Prospect Theory by the behavioural economists Daniel Kahneman and Amos Tversky.




Just think how tempting it can be to put additional money into a project which already went south and incurred a loss when there is a chance to recover the situation. Although, while doing so there is also a significant chance that things still fail and incur an even bigger loss.


This is so unfair - the Ultimatum Game


Here is another example of how emotions can trick a person's mind, especially how the feeling of being treated fair/ unfair can lead to sub-optimal conclusions: The Ultimatum Game. The setup of this experiment is as follows:

  • There are two persons. One is the "proposer". One is the "accepter".

  • The proposer gets EURO 10,-- and has to offer an amount between EURO 1,-- and EURO 10,-- to the accepter.

  • The offered amount is purely at the proposer's discretion.

  • Though, the accepter may accept or refuse the offered amount.

  • In case the accepter refuses the offered amount, the proposer has to give back the EURO 10,-- and neither the proposer nor the accepter gets anything.

It turns out that things are not so easy... Surprisingly, in respective test settings there was no dominant example of how to split the EURO 10,--. Splits like 5/5, 6/4, 7/3, or 8/2 were evenly present which gives an indication that there was no clear opinion on what constitutes a "fair" or "rational" split. Rather, it gives a strong indication that splitting decisions were made irrationally and emotionally. Otherwise there would be not so many different (and evenly distributed) results of a split (a more detailed account on the Ultimatum Game can be found with Chris Voss and Tahl Raz: see below).

It is to be said that any proposer who offered more than EURO 1,-- acted emotionally. And, any accepter who refused such kind of offer ("...because it is so unfair that I do not get the equal part of a EURO 10 split..") acted emotionally as well. After all, EURO 1,-- is better than getting nothing. And still, just due to the unfair-argument refusals happened. By the way, how would the decision of a respective accepter change if he would not know that the proposer has a budget of EURO 10,--? In other words, when the reference point is unknown to the accepter, how does this change the emotional decision? The offer itself stays the same.

Now think about somebody who plays on you the fair/ unfair - game in a manipulative way during a negotiation. How would this go?


Some other emotions in play


The way our brain - better the way our emotions - trick us can be manifold. In a normal stage of mind, one has intuitive feelings and opinions about almost everything that comes along ones way. And in general, we generate intuitive opinions on complex matters. That means when a satisfactory answer is not found quickly our mind will find a related question that is easier to grab and will answer that easier question. This unconscious process called mental shotgun sometimes works fairly well and sometimes may lead to really serious errors.

One of my favourite "mind trickers" is the predilection for causal thinking (translated in a more laymen way: you need a narrative for everything). This can lead to serious mistakes in evaluating the randomness of truly random events, i.e. you see cause and effect in circumstances where actually there is just chance (...it just happened). Imagine the gender of babies born in a certain time sequence in a hospital. Consider the following three possible sequences (G...girl, B...boy):


BBBGGG GGGGGG BGBBGB


The question is: Are those sequences equally likely? And the intuitive answer --"...of course not..." -- is wrong. As those six births are independent from each other and because the outcomes 'boy' and 'girl' are approximately equally likely, any of those possible sequences as shown above is as likely as any other. And still - this remains counterintuitive as just the third sequence seems random.



Affect heuristics - This is another mean way to get fiddled. In principle, people let their likes and dislikes determine their beliefs about the world. For example, the political belief of a person determines the arguments that person finds compelling. This should give you some notion how open one is to counterarguments when at the same time being "hot" on a specific market or product.

Finally, let's have a short look at an issue called confirmatory bias. This is when people seek data and information which is compatible with the beliefs they currently hold. This bias favours uncritical acceptance of suggestions and exaggeration of the likelihood of extreme and improbable events. In summary, this does not seem to be a track to a rational and objective decision making.


Data Driven Decision Making


Most of the examples shown here were taken from Daniel Kahneman's book 'Thinking, Fast and Slow'. But don't make a mistake. There are so many more ways how mind and emotion can trick us into false conclusions and irrational decisions. Indeed, this short essay covers just 40 pages out of 450 pages full of potential flaws and biases in our mental and emotional stage. In case this is not enough then have a look at Dan Ariely's book "Predictably Irrational". Those are another 350 pages full of mind flaws.

And? Do you still think that you can make rational and logic decisions, especially when faced with complex or distressed situations? I would say that helpers are needed to get things under control.

The task of an efficient management information system is to provide a decision maker not only with the best possible picture of current circumstances or meaningful analytics for the task at hand, but to make sure that subjective expectations and opinions are counterchecked and put into relation with pure market facts and to keep the line against all sorts of emotionally induced opinion leads as well as heavily biased conclusions. In short: data driven decision making.

This is true for the Business Analytics part and it is even more true for the Predictive Analytics part. The more longterm a future development is to be assessed, the higher the uncertainty gets, therefore the more vulnerable a decision is towards bias, irrational expectations and emotional derailing.




The good news is that Business Analytics does already a great job in its part of the assignment. Even better news is that Predictive Analytics is increasingly backed by extremely strong tools in the field of Statistics, Artificial Intelligence, Network Dynamics or Causal Inference to cover all sorts of uncertainty in future developments.

The bad news is that emotion is a necessary element to decision making. There is no way around this fact. The final act of deciding is inevitably connected to emotions. Hence, it is vital to incorporate a "rational" and well fuelled supporting system which does its job up to the final act of deciding. Or, as Kahneman put it: "Intuition cannot be trusted in the absence of stable regularities in the environment."


A little joke - Will there be a cold winter?


Instead of a formal conclusion of this article, I would like to finish it with a joke. It is not only very funny (at least I thought it is funny) but shows where emotional guessing without a proper supporting system can lead to.

It is early fall and some Inuits ask their chieftain if he expects a strong winter this year. In this case they would have to start their winter preparations much earlier. The chief confirms a strong winter and already sends his people for collecting timber into the woods. Though, he is not sure at all about the upcoming winter, but cannot tell to his people as this would expose him as a weak leader. So, to be sure he calls the local weather forecast station for a respective forecast. After a short pause the duty officer of the weather forecast station confirms that indeed a strong winter is to be expected. The chieftain feels affirmed and is happy that he sent his people to collect wooden and to prepare for the winter.

Meanwhile in the weather forecast station, a colleague of the duty officer asks why the officer is so sure that there will be a strong winter. After all, it is the beginning of autumn and therefore still a bit early to make such a fast call without detailed considerations. The duty officer, who has no idea who was on the line before, says: "Yes, I know! But then I looked out of the window and saw Inuits already collecting timber. When they are collecting wooden this early means that they expect a strong winter. Those people have a lot of experience and do know!" :)

In the next article we will talk about the importance and necessity to incorporate the knowledge and experience of professionals in building up those supporting systems. This is vital on the part of the decision maker but it is equally important on the part of any consultant assisting a company to know, to understand and to have the experience in the business environment they are dealing with.

Sources

Thinking, Fast and Slow by Daniel Kahneman/ 2011

Predictably Irrational by Dan Ariely/ 2009

Never Split The Difference by Chris Voss and Tahl Raz/ 2016