Risky business: why do companies take more risks when faced with losses?

When making a risky decision, companies assume they are weighing up their best alternatives. But psychology suggests our choices are biased towards taking risks when dealing with losses, writes Dr Ed Mitchell

How humans make risky decisions is a central topic of the burgeoning field of behavioural economics. It has found that, far from being the perfectly rational decision-makers, our decisions are riddled with bias, inconsistency and irrationality.

One of the most significant contributions of the field has been that humans (and other animals) make very different decisions when faced with potential gains to when they are faced with potential losses.

Losses play a major role in many business activities, for example insurance underwriting, trading in bear markets, and environmental health and safety (accidents, pollution incidents, working hours lost, and so on). These areas of business activity have one thing in common: they are striving to minimise losses, rather than maximise gains.

TV gameshow

Imagine you’re a winning contestant on a TV gameshow. The host offers you two options for your prize. You can either have £15,000 for sure, or select a gamble on the flip of a coin, which will give you £10,000 if heads and £20,000 tails. Do you want the £15,000 for sure, or do you want to gamble?

If you’re anything like the writers of this article, you’d find the thought of taking the gamble and losing absolutely awful – you’d go home with £5,000 less than you could have. Indeed, most people in this case will prefer the £15,000 for sure – we prefer to take the fixed option, rather than variable option (the gamble).

Now imagine again that you’re on the gameshow. This time, the host gives you £20,000. He then offers you the choice of giving back £5,000, or flipping the coin, in which case he’ll take back nothing if heads, or £10,000 if tails. Most people in this situation take the gamble.

However, these two examples are (in terms of their economics) exactly the same – either take home £15,000 for sure, or gamble for £20,000 or £10,000. So why do we take the sure thing (the fixed option) when decisions are framed as gains, and take the gamble (the risky or variable) option when faced with losses?

In 1738, Daniel Bernoulli mapped the ‘human utility function’. Bernoulli reasoned that the price of an object is the same for everyone, but the value – the amount of utility it gives us – differs depending on our circumstances.

One hundred pounds is worth far more to a pauper than it is to a millionaire. Therefore, increasing amounts of wealth bring progressively less utility, represented by the black curved line in the diagram A (on page 26).

Now imagine a 50-50 gamble. We will be given £x if we win (x), but have £x taken away if we lose (-x). Due to the flattening of the curve as we gain more wealth, we stand to lose much more utility (z) if we lose, than the utility gained by winning (y).

Therefore it doesn’t seem worth taking the gamble; we would rather maintain the status quo or take a surer gain. In other words, we are risk averse. In other words, risk aversion is due to losses looming larger than gains.

Not the whole story

The psychologists Daniel Kahneman and Amos Tversky realised that risk aversion was not the whole story (Kahenman won the 2002 Nobel Prize in Economics for this work, shown in the graph over the page). They extended the utility function into the domain of losses, and reasoned that gains and losses always take place from a reference point.

In the same way as increasing gains bring progressively less utility, so increasing losses bring progressively less disutility. Diagram B shows that doubling one’s losses (going from -x to -2x) brings only a small amount of further disutility. However, there is a huge amount of utility to be gained from ‘getting back to the reference point’. Therefore people gamble when faced with losses.

This inconsistency in our decisions (taking the sure thing when faced with gains, and gambling when faced with losses) can have serious implications for businesses where we deal with losses – we will be inclined to make more risky decisions.

Imagine an explosion at your factory starts leaking poisonous gas into the surrounding community. You are the head of environmental health and safety. There are 600 people that will die if you don’t take action. You have the choice between two programs. Program A will save 200 lives for sure.

Program B has a one third probability that all 600 people will be saved, and a two-thirds probability that no lives will be saved. Which do you choose? Most people choose Program A in this situation.

Now imagine an example where the same situation occurs but now you can choose between Program C, which will result in 400 people dying, or Program D, which has a one-third probability of nobody dying, and a two thirds probability of all 600 people dying. Which do you choose?

Most people now choose Program D. However, both examples are exactly the same, except one is phrased in terms of gains (lives saved) and one in terms of losses (lives lost).

Simply switching how we frame the problem causes us to gamble with lives.

Established science

Kahneman and Tversky’s theory (dubbed Prospect Theory) is not the only theory to be able to explain this pervasive phenomenon.

Another framework, called Scalar Utility Theory also explains risk aversion for gains and risk proneness for losses, and unlike Prospect Theory is based on established scientific knowledge about how humans and other animals perceive amounts.

While theories may differ as to how to explain risk proneness when faced with losses, the phenomenon itself is very robust. If we can be so easily induced to gamble with lives, it is unsurprising that Enron executives took on more and more financial and personal risk as their losses became greater and greater.

Banks, such as Barings, are brought to their knees by traders who rack up progressively greater losses in increasingly futile attempts to get back to the reference point – to break even. But what about more day-to-day examples of decision-making?

Much riskier

It’s clear that areas of business that work with predominately negative Key Performance Indicators (KPIs), for example Environmental Health and Safety (number of train passengers killed per year, barrels of oil spilled, working hours lost due to on-the-job illness) may be induced to take much riskier decisions than if their KPIs were positively framed. This is to say nothing of the depressing and unmotivating working environment when constantly dealing in losses.

The impending Basel II financial regulations require banks to assess their risks in a much more formal way.

It is clear that the regulations only concentrate on downside risks (losses) rather than the upside. This may have a major impact upon the way banks choose to implement the regulations.

By understanding the psychology of risk, decision consultants such as ERM and ORRA can also point the way towards better decisions.

For example, both can alter the reference points of decision, plot utility functions making decision criteria explicit, and profile the risk prone-ness or risk aversion of decision-makers.

We may be hard-wired not to play it safe but we can overcome our innate tendencies.

· Dr Ed Mitchell is a research fellow at Pembroke College, Oxford and a co-founder of Oxford Risk Research and Analysis. Co-author Alex Kacelnik is Professor of Behavioural Ecology at Oxford University and Dr Tom Woollard is from ERM and heads the Corporate Advisory Services business in the UK

Action inspires action. Stay ahead of the curve with sustainability and energy newsletters from edie