Financial Services, Gambling, & Behavioral Economics: Busting The Myth Of Consumer Rationality
By Neal Cole
I’ve worked as a customer insight and research manager in financial services (FS) for most of my career. During this time I’ve noticed that colleagues often assume that consumers are more rational when buying financial products compared to other categories of goods and services. There is a perception that FS products are more ‘serious’ than your average consumer goods product (FMCG).
This view is sometimes supported by The Consumer Involvement Theory. The theory suggests FS purchases fall into the high involvement and rational segment of the model. This is due to the relatively high cost of FS products and that purchases are more about logic and less about emotion. You don’t buy a pension everyday!
But what does the evidence from experiments in behavioral economics and neuroscience indicate about rational decision making in the face of risk and uncertainty? Are consumers’ really discreet, self-determining individuals who make considered, rational decisions?
This view increasingly looks misguided and is probably a fallacy created by our own minds to make us feel in control of our behavior. As Mark Earls points out in his book Herd:
“Our failure to acknowledge the truth about human nature distorts our attempts to understand human behavior and frustrates our attempts to change it. Bad theory = Bad Plan = Ineffective action.” Mark Earls on Stephen Pinker, Herd
Behavioral economists Dan Airely and Nobel laureate Daniel Khaneman have uncovered strong evidence that rational decision making is often an illusion. That is not say people don’t behave differently when considering money issues. Dan Ariely found that just thinking about money makes people more selfish, self-reliant and less charitable. However, these traits don’t necessarily make people more rational in their FS decision making.
Insight identified from behavioral economists challenge many of the basic assumptions of traditional economics and related theories of decision making. Some of the Key insights are:
- Emotions – Human decision making is unconsciously driven by our emotions and social norms much more than we have appreciated in the past. This is due in part by our reliance on our fast, intuitive, but largely unconscious mind. Daniel Khaneman refers to this as system 1. This makes the majority of our decisions. But its frequent use of rules of thumb (heuristics) make people prone to biases that can lead to sub-optimal decisions.
- Answering an easier question – Because we find cognitive thought hard work, system 1 will often substitute an easier question for a difficult question to answer instead. It will do this automatically if we are unable to easily retrieve an answer to a hard question.
- Context dependency – Our state of mind and the decisions we make are heavily influenced by the environment within which we find our selves. This leads to inconsistencies in our decision making that we are largely unaware of.
- Memory – Our memory of events is unreliable and heavily biased towards the beginning, the peak of activity and the end of an event. We neglect the duration of an event and have little awareness of our true motivations.
- Illusion of understanding – Khaneman uses the acronym WYSIATI (What You See Is All There Is) to describe our tendency to think that the limited information we have about the world is all that there is to know. Humans create narrative fallacies in an attempt to make sense of what are often random events.
“Our comforting conviction that the world makes sense rests on a secure foundation: our almost unlimited ability to ignore our ignorance.” Daniel Khaneman, Thinking, fast and slow
- People herd – As Mark Earls points out humans are a “super social species”. Our behavior is unconsciously influenced by what other people do and more so than we realize or like to admit. In the face of uncertainty we look to how other people behave and will often follow their lead.
- Mark Earls takes this insight a step further and argues that market size and market share are primarily a function of consumer-to-consumer interaction. The implication being that rather than focusing on supply side factors, marketing should pay more attention to understanding and modelling interactions that generate mass behavior (i.e. consumer-to-consumer interactions).
”You have to understand the rules of interaction – the accepted behaviors and rules of thumb of the individuals whose interaction generates the complexity of behavior that you are studying – because these will shape the outcome of interactions.” Mark Earls, Herd
- Real people are also sometimes generous and willing to contribute to the good of the community. These are not the characteristics of a rational person described by traditional economic theory.
- So peoples’ decisions are mainly influenced by factors that they are not consciously aware of. Humans review and post-rationalize decisions. This suggests that our perceptions of a product or brand are likely to change after an action rather than before as implied by traditional marketing models like AIDA (Attention, Interest, Desire, Action). It should probably be changed to Context, Attention, Emotion/social norms, Action, Review, Memory (C.A.E.A.R.M). Not a great acronym, but I still find marketing people using the old AIDA model so we do need to encourage them to move on from it.
So what specifically does behavioral economics have to say about FS decision making? Risk and uncertainty is at the heart of Daniel Khaneman and Amos Tversky’s Prospect theory. Three cognitive principles form the basis of the theory:
- The perceived value of a decision outcome (the utility derived) is dependent upon the history of one’s wealth (the reference point). This may seem obvious, but traditional economics does not recognize that a poor person will perceive a gain of £1,000 as generating more utility than would a millionaire. A person’s reference point is often the current status quo.
- People experience diminishing sensitivity to both sensory changes (e.g. light) and to changes in wealth. So for example the subjective difference between £1,000 and £1,100 is much smaller than between £100 and £200.
- Humans are loss averse. When compared against each other people dislike losing more than they like winning. Thus losses loom larger than gains even though the value in monetary terms may be identical. This explains why investors find it painful to sell shares that are below their purchase price and find it easier to sell shares that are in profit. This is not rational behaviour.
Loss aversion is key to understanding how people perceive financial services, and gambling of course. Extensive research has been undertaken to estimate the psychological value of losses and gains. These studies have identified a loss aversion ratio of between 1.5 and 2.5. This means that a loss that is identical in money terms to a gain is valued up to 2.5 times more than the gain.
Interestingly, professional risk takers such as fund managers are more tolerant of losses. This may be because they are less emotionally aroused than the amateur investor. Loss aversion leads to predictable behaviours in a number of situations:
- If a potential loss could be ruinous or would threaten their lifestyle, people will normally dismiss the option completely. Only obsessive gamblers would normally consider this type of situation.
- Where people are presented with a situation where both a gain and a loss are possible people tend to make extreme risk averse choices. For example a person is presented with the choice between a small guaranteed gain over 5 years (e.g. a deposit based account) and a stock market linked product that carries a low risk of a large loss. People have a tendency to focus on the large potential loss and often select the former, less risky option. This is why advisers will focus on the large upside potential of a stock market linked investment and try to play down any potential for large losses.
- Where the choice is between a certain loss and a larger loss that is just a probability (i.e. there is a chance of no loss), diminishing sensitivity can result in excessive risk taking. This explains why private investors sometimes refuse to cut their losses on poorly performing shares and instead invest more money (to reduce the average purchase price) in the hope that the price will recover sufficiently to avoid a loss. This is known as the sunk-cost fallacy.
“Loss aversion is a powerful conservative force that favors minimal changes from the status quo in the lives of both institutions and individuals.” Daniel Khaneman, Thinking, fast and slow.
THE POSSIBILITY AND CERTAINTY EFFECTS:
- When considering FS decision making it also necessary to understand how consumer evaluate risks. There are two key biases that relate to the psychological value (weight) given by people to different probabilities or risks.
- The possibility effect results in highly unlikely (low probability) events being given more weight than they justify. This helps explain the attractiveness of both gambling and insurance policies that cover unlikely events (e.g. extended warranties).
- The certainty effect leads to events that are almost certain being given less weight than their probability justifies.
- Indeed, research shows that unlikely events (1% to 2% probability) are over weighted by a factor of 4. However, for an almost certain event the difference is even larger. In experiments a 2% chance of not winning was given a weighting of 13% (or an 87.1% chance of winning).
- Where the odds of an event are very small (e.g. around 0.001% or less) people become almost completely indifferent to variations in levels of risk. Rather emotional factors and how a risk is framed are the key drivers of how people react to these levels of risk. This explains why after a terrorist attack there tends to be more focus on whether insurances cover such risks even though the level of risk (to an individual) remains extremely low. It also helps to explain why people are often too willing to bet on extreme events happening.
“When the top prize is very large, ticket buyers appear indifferent to the fact that their chance of winning is minuscule.” Daniel Khaneman, Thinking, fast and slow
- Khaneman also found evidence that rich and vivid descriptions of an outcome (e.g. the lifestyle of a lottery winner) helps to reduce the impact of probabilities. In particular he found that people are more heavily influenced (in terms of weighting of probabilities) if an event is described by using frequencies (e.g. the number of people) than by abstract concepts such as chance or risk.
- Due to our use of intuitive thinking (system 1) and the laziness of system 2, most people have a tendency to evaluate individual risks separately and independently. People tend to make decisions when a problem arises rather than trying to look at the bigger picture. What Khaneman found was that this approach will almost always lead to sub-optimal decisions due to our focus on loss aversion. The best solution is to aggregate decisions together. A professional investor achieves this by always looking at individual shares as part of a balanced portfolio. This reduces the impact of loss aversion on our preferences.
- People hold their money in different accounts, some of which are real and some are only mental (e.g. money from my dad to buy my daughter a present). There is normally the everyday spending account, general savings, savings assigned for emergencies, maybe savings designated for private education and so on. People use mental accounting as an aid to self control. They have a clear hierarchy of willingness to use these accounts to cover their immediate needs and have an emotional attachment to the state of their mental accounts.
- Mental accounting is a form of narrow framing and can have disastrous consequences in financial services. It often leads to private investors to set up a separate mental account for each share they own. This results in investors wanting to close each account as a gain. So when they need money for their daughter’s wedding what do they do? They have a very strong preference to sell winners rather than losers. It also helps to explain why consumers might have an outstanding credit card balance of £2,000 (with an APR of around 20%), and yet have savings of £10,000 (paying just 4% interest). These are not rational behaviors.
- Emotions are also an important factor in how we evaluate gains and losses. Most theories of decision making assume that people evaluate available options in a choice separately and independently. This does not reflect human nature. People feel regret when the experience of an outcome is affected by an alternative option that was open to them, but they did not choose. Thus missing out on selecting the top performing managed fund may influence the perception of your investment choice.
ARE PEOPLE REALLY MORE RATIONAL WHEN BUYING FS PRODUCTS?
- The evidence clearly suggests no. People are prone to the same biases when purchasing FS products as they are when buying consumer goods. Indeed, FS decisions may sometimes be subject to more powerful disruptive forces (e.g. loss aversion and mental accounting) than other types of purchases. This demonstrates the importance of regulation and consumer education in the FS sector.
WHY DO FS MANAGEMENT BELIEVE IN RATIONAL CONSUMERS?
- Senior management in the FS sector is dominated by a series of very numerate professions who are highly skilled at estimating risks and calculating probabilities. There are actuaries in life & pensions, underwriters in general insurance and lending, bankers, accountants, and a smattering of economists. Given their training and experience of dealing with risk and uncertainty they are less prone to key cognitive biases such as risk aversion and mental accounting. For this reason FS management are less likely to appreciate how strongly consumer behaviour is influenced by these biases.
I observed an example of this when I worked for a large UK life assurance company. We developed a Guaranteed Capital Bond that protected your initial investment and provided some limited potential to benefit from any rise in the stock market. It researched well, but the CEO (who was an actuary) thought it wouldn’t sell. It didn’t offer enough upside potential if the stock market grew strongly. The Director of Sales & Marketing (a sales person) was supportive of the launch because he understood how loss averse people can be. I don’t need to say who won the argument when it went on sale.
IMPLICATIONS FOR MARKET RESEARCH:
I could write a whole post on the implication for market research arising from the above insights. Instead I would like to finish with just a few suggestions for consideration:
- Use analytics to better understand current customer behavior. In the digital age we can now use web analytics to track and measure online customer behavior. We also have the ability to conduct online experiments (i.e. A/B and Multivariate testing). But even in the off-line world there are many sources of data to explore and analyze before we need to conduct primary research.
- Fewer focus groups please! In some FS organizations focus groups appear to be the default research tool. In a previous post, Should focus groups carry a health warning, I pointed out my own concerns about this method of research. Interestingly John Kearon of BrainJuicer made a similar observation:
“Yes, they (Focus groups) can reveal powerful insights in the hands of a great researcher, but all too often they are just the lazy default of unquestioning research buyers and produce little or no insight on the subject at hand.” John Kearon, BrainJuicer
- Don’t ask direct questions but instead observe behavior. People are unreliable in their recall of why they make decisions. Insights are more likely to emerge from observing human behavior during key experiences than trying to ask direct questions. This can be carried out in a number of ways including ethnography, auto-ethnography and analysis of customer interactions (e.g. telephone calls) with customer facing staff.
- Covert monitoring of behavior. There is plenty of evidence to show that people behave differently when they know they are being observed. I used video mystery customers (using a hidden camera) to evaluate training and development needs for one company’s sales team. I was informed that almost all of them met agreed standards when they were accompanied on visits by a trainer. However, almost the opposite was observed when we analyzed the videos of the mystery customer appointments. Unless you have regular monitoring of service standards in place you can’t be sure what level of service your customers are receiving.
- Customer facing staff. Listening to sales people, advisers, brokers, telephone agents, people who speak with customers on a daily basis can very insightful. People are better at observing how other people behave than trying to explain their own behavior. Experienced sales people collect a wealth of knowledge about how customers respond to different strategies, what turns them off, what excites them, what confuses them and what appears to motivate them.
- Co-create. A collaborative approach to research encourages mutual respect and shared learning. Including social influencers (i.e people who shape attitudes and behaviours of their peers) in the process helps ensure the generation of more innovative ideas than would be the case with only experts and working parties involved. Collaboration also helps break down barriers between different stakeholders and speed up concept development and refinement.
- Crowd sourcing. There is growing evidence that asking large groups of people to participate in predictive markets can be a very good way of selecting winners. James Surowiecki’s book, The Wisdom of Crowds, has a mass of evidence to support this approach.
- Understand the ‘how-mechanic’ of groups of consumers. To really understand ‘herd’ behaviour companies need to focus on influencing consumer-to-consumer interaction rather than B2C interaction. Mass behavior occurs under certain circumstances and is based upon simple rules.
“By examining the interactions and behaviors that a particular group of people has, it is possible to identify the underlying rules that drive it.” Mark Earls, Herd
- The mistake many organizations make is to see Word of Mouth (WoM) as a channel rather than the way consumers interact and influence each other. To benefit from this insight it is necessary to understand the conditions of interactions (e.g. the environment) and the rules of interaction (e.g. how people engage with each other). By making small changes to either or both of these elements of interaction we may be able to significantly influence individual and ultimately group (e.g. private investors) behaviour.
Thank you for reading my post. I hope it challenged your thinking about consumer decision making and the implications of behavioural economics for market research.
Further reading: Thinking, fast and slow by Daniel Khaneman, Herd by Mark Earls (@Herdmeister), Influence by Robert B. Cialdini, PHD (@RobertCialdini) ; Predictably Irrational by Dan Ariely (@danariely); the Upside of irrationality by Dan Ariely; The Wisdom of Crowds by James Surowiecki; Consumer.ology by Philip Graves (@philipgraves); Nudge by Richard Thaler (@R_Thaler).