Promotions: The Effect of ‘Labelling’ on Customer Behaviour

I've just had the great pleasure of writing a guest blog for The Marketing Directors (the marketing consultancy) and here it is: Promotions: The Effect of ‘Labelling’ on Customer Behaviour

How not to play Poker

Demonstrating that behavioural economists are no more rational than anyone else I committed the same mistake playing poker that I did about a year ago...

I acted irrationally by seeing different colour chips as qualitatively different as well as quantitatively different. Despite the fact that one black chip was worth five white chips, I preferred betting five white chips than one black chip because I saw the higher value black chips as in some way better. This led me to be more risk averse when I had lost my white chips and had only black chips to bet.

Because I saw equivalent amounts of money as different I acted in a non-fungible manner. I'm not blaming this violation of fungibility for my poor performance (I lost, badly) but I am surprised to see myself commit the exact same mistake that I even wrote a blog about before. Some people never learn...

Little Jimmy

The 4G auction in the UK used a highly complex mechanism called the combinatorial clock auction. Here is  a simple example of how an combinatorial auction works (using the the Vickrey-Clarke-Groves rule).

Little Jimmy has two chickens to sell. His friends are little Dave, little Fred and big Phil. Little Dave wants one chicken and is prepared to pay £5 for it. Little Fred wants one chicken and is prepared to pay £3 for it. Big Phil wants both chickens for a total of £7, but it not prepared to buy just one on its own.

How do we decide who gets the chickens? How do we create an auction where Dave, Fred and Phil have an incentive to bid their maximum valuation?

Answer: Maximise the value of the bids and make the winners pay the opportunity cost of winning. (Price = total bids of all winning bids if winner didn't bid - total bids of all other winning bids)...

We maximise the sum of the bids by having little Dave and little Fred winning one chicken each. But how much do they pay?

 

Dave: Fred is the other winner with utility £3. If Dave was removed Phil would win both with £7. (7-3=4)

Fred: Dave is the other winner with utility £5. If Fred was removed Phil would win both with £7. (7-5=2)

Therefore...

Little Dave wins one chicken and pays £4

Little Fred wins one chicken and pays £2

Big Phil wins nothing. Big Phil is sad. Big Phil is slightly less big now.

The Stable Marriage Problem

How does one go about matching husbands with wives?

Is there a way to do it that will result in stable marriages? 

If we define stable marriages as the situation where no couple would prefer each other to their current partner, then the answer is yes: a stable matching does exist. We can create a situation where no new couple would be created at the cost of a current couple.

Let's say there is a finite number of singles, who all go to a dance together.

We now use something called the deferred acceptance algorithm (Gale & Shapely, 1962):

Step 1. Each man proposes to his preferred woman.

Step 2. Each woman records all the men who propose to her.

Step 3. Each woman then rejects all the men except her most preferred suitor.

Step 4. Each rejected man then proposes to his next most preferred woman.

Step 5. Repeat Step 2 onwards until all men and/or all women are matched.

Ta da! We have a stable matching. There is no new couple that could be created. If a man preferred a different woman to his partner, well, the woman would already have rejected him.

Romance versus efficiency: which will win??

Blackberry Crumble

On Monday Blackberry announced it had provisionally accepted a bid from the Canadian private equity firm Fairfax Financial Holdings worth $9 (U.S.) a share, only 6% of the peak share price recorded in June 2008. Why has Blackberry, a firm with operations in 175 countries, $2.8bn in cash and whose corporate clients amount to 90% of the Fortune 500 been forced to accept such an offer?

Fundamentally, Blackberry misread market trends and did not follow consumer demand. It stood still while competitors HTC, Samsung and most notably Apple reshaped the mobile handset market. Blackberry isn't the only firm which failed to adapt (remember Nokia?) but its decline is swift and coming to a head. So why did Blackberry become so big in the first place?

Blackberry rapidly gained market share by doing two things really well. It had the best security and the best instant messaging system (BBM). This made it the favourite of both governments and corporate IT departments across the globe, whilst also appealing to young people who quickly became attached to BBM. However, Blackberry neglected three key trends in consumer demand.

Firstly, Blackberry ignored the fact that consumers want something fun, not just functional. It dismissed touch screens as unnecessary, not grasping that people found them alluring and  engaging enough to pay for. Blackberry has now brought out some touch screen models, but they have performed poorly and are too little too late. Blackberry continued catering for those few who saw buttons as indispensible while ignoring the direction of the rest of the market.

Secondly, Blackberry handsets are not sufficiently App friendly. Handsets were incapable of hosting Apps as recent as 2009, and even now there is a greatly reduced variety of available Apps. It almost goes without saying that fans of Apps do not buy Blackberry handsets. Again Blackberry found itself forgetting that consumers expect to enjoy using their phones. Ironically, one of the few Apps available on the Blackberry operating system is WhatsApp, an App that has done more than any other to erode the status of BBM as the best instant messaging service.

Thirdly, the recent trend of bring your own device (BYOD) has wrong-footed the Blackberry business model of appealing to IT departments. With employees often deciding which handset they use, security is no longer king in the corporate handset market leaving Blackberry out in the cold.

In summary, Blackberry had a winning formula but refused to adapt the formula in the face of significant trends in demand. Blackberry is now about to be taken into private ownership where it can be reshaped with less public scrutiny. This reshaping is widely predicted to involve turning Blackberry into a niche handset maker for the corporate world, and with large reserves of cash and intellectual property, there is an opportunity here. If Blackberry can adapt to key consumer demand trends, once again the future could be bright.

A Chocolate Gift

In the post yesterday I received a welcome surprise... a thank you note from the sailing summer camp I am a leader on complete with a chocolate bar!

I was touched, and as I was rather peckish at the time I tucked straight into the chocolate. That got me thinking. Why did I value the chocolate bar so much? If the thank you note had included the cash value of the bar I would have been rather puzzled (it's just not the done thing) and maybe even insulted. I certainly would not have felt as appreciated as I did as I devoured the Milky Way.

Tons of research has shown that material gifts are appreciated more than the cash. This is strange; cash gives us more flexibility. A recent paper by Kube, Marechal and Puppe (2011) concludes that it really is the thought that counts. The more thought that has gone into a gift, the more we appreciate it, almost regardless of how useful the gift is to us.

So there we go; if anyone feels like thanking me in the near future... send a chocolate bar...

Winter Fuel Payment

In the UK everyone over the age of 60 is given the Winter Fuel Payment each year by the government. ('Winter Fuel' refers to heating and electricity costs.) Over the last 10 years the Winter Fuel Payment has varied between £200 and £250 per person over 60, with the payment for those over 80 varying between £300 and £400. The Winter Fuel Payment is usually paid in one lump sum in November or December. There are no legal requirements or official guidelines over how the Winter Fuel Payment should be spent by the recipients.

On average, what percentage of the Winter Fuel Payment do you think is spent on fuel? 

Imagine the hypothetical scenario where everything stayed the same, but the Winter Fuel Payment was called 'The Annual Payment'. On average, what percentage of the Winter Fuel Payment (now called The Annual Payment) do you think would be spent on fuel?

The answers are 41% and 3%, respectively! (according to Beatty, Blow, Crossley and O'Dea, 2011) The label 'Winter Fuel' alone causes 38% of the Payment to be spent on fuel. This is called the labelling effect and it is another example of non-fungibility.

Does this information change whether you think pensioners should receive the Winter Fuel Payment?

Does the Child Benefit?

In the UK all parents receive Child Benefit from the government (equivalent to Child Tax Credit in the US). It's worth £20.30 per week for the eldest child and £13.40 per week for other children. The government spends 1% of GNP of Child Benefit. So it seems relevant to ask how Child Benefit is actually used by parents...

a) Parents spend a greater proportion of the Child Benefit on their children than they do with other sources of income
b) Parents spend the same proportion of the Child Benefit on their children as they do with other sources of income
c) Parents spend a lower proportion of the Child Benefit on their children than they do with other sources of income

(This data is accurate for the years previous to the recent change: all parents got the whole Child Benefit regardless of income)

The answer is c!

Blow, Walker and Zhu (2012) found that parents spent less of the Child Benefit on their kids than they do with other sources of income. Does this mean that British parents are uncaring?

Well, it gets worse... Parents typically spend nearly half of the Child Benefit on alcohol!

Oh dear. That's what children drive you to. 

- My Dad

But, in fairness, upon delving into the data Blow et al. discovered that parents have already insured their children's consumption out of primary sources of income. So an alternative explanation of the data is that parents don't count the Child Benefit when doing their budgeting and then treat it as an extra to be spent frivolously, safe in the knowledge that their children are cared for.

This is non-fungibility, but just not in the expected direction. The label 'Child Benefit' has a rather perverse effect. 

So, what do you think, does this prove Child Benefit should be axed?

Easy Money

Does the way you spend your money differ according to how it came your way?

More specifically, are you more or less prudent with income that you didn't directly earn (such as an inheritance or a government grant)?

A recent paper by Christiaensen and Pan (2012) analysed household spending in rural China and Tanzania and found that different sources of income are used differently. 

Earned income is more likely to be spent on food staples or education, while unearned income is more likely to be spent on more luxury goods such as alcohol, tobacco or clothing. While money is quantitatively the same, it is viewed as qualitatively different. This non-fungibility has implications for whether government money is distributed by employment generating programmes or cash transfers.

So have a think, are you less careful with unearned income? Could you better manage your finances by paying more attention to how you spend 'easy' money?

The Disposition Effect

A person who has not made peace with his losses is likely to accept gambles that would be unacceptable to him otherwise.

(Kahneman and Tversky, 1979) 

Stock brokers prefer to sell stocks that rise in price than stocks that fall in price. The preference for 'winners' over 'losers' is driven only by the desire to realise gains over losses. This is called the disposition effect, and it is likely to lead to lower profits.

This is because attitude to risk is different for losses than gains. Behavioural economics has shown that people tend to be risk seeking when it comes to losses, but risk averse when it comes to gains. For example, a stock that depreciates in value will be seen as a loss, making the stock broker more risk seeking and therefore more likely not to sell it (it may go up in value again). But if the stock rises in value then the stock broker is more risk averse and therefore more likely to sell it (to avoid the risk of it falling in value).

The disposition effect increases taxable income (Odean, 1998). If stock brokers realise 'winners' they have to pay tax on the gain. But stock brokers do not have to pay tax on 'losers'. Thus stock brokers could put off paying tax (and thus earn money) by holding 'winners' for longer. And by selling 'losers' taxable income reduces; if stock brokers sell the 'losers' and buy almost identical stocks taxable income actually falls. Thus the disposition effect is irrational for stock brokers (but good for the Inland Revenue!).

The disposition effect is a violation of fungibility because investors view units of money either side of the gain/loss boundary as qualitatively different. This is an example of non-fungibility causing market failure. If investors were aware of this non-fungibility they might be less likely to exhibit it.

Putting the Fun into Fungibility

My apologies for my recent silence, I've been working hard on my MSc dissertation (now successfully finished). The topic of my dissertation was 'fungibility'. My next few posts will explore fungibility and why being aware of it may help you make more rational decisions.

Fungibility is the principle whereby economic agents treat all units of money as equal and as perfect
substitutes, regardless of where they came from.

Violations of fungibility occur when individuals use money differently because of the way it was earned, stored or labelled. Violations of fungibility can cause us to make suboptimal decisions and are thus termed irrational.

Take the example of an €8 gift voucher at an expensive restaurant. For some customers the voucher can be spent on beverages (the 'labelled' voucher) while for others the voucher can be spent on either food or beverages (the 'unlabelled' voucher). As almost all customers spend at least €8 on beverages the gift is 'nondistortionary'.

Johannes Abeler and Felix Marklein (2013) ran this field experiment and found that customers who had the labelled voucher spent on average €3.90 more on beverages than those with the unlabelled voucher.

The label attached to the money had changed behaviour (even though, rationally speaking, both groups should have spent the same amount on beverages as the voucher was nondistortionary). This 'labelling effect' violates the principle of fungibility.

So the next time you receive a labelled voucher think to yourself 'Would I have spent that much on ... anyway?' 'How much was I originally prepared to spend on ...?'

It might just help you avoid be more rational in how you spend your money.

Air Regulation

A recent paper by two economists (Aguirregabiria and Ho) caught my eye.

Typically, in America, airlines operate 'hub and spoke' networks. That is, an airline will have a main airport (the hub) and all the other cities in the network (the spokes) only fly to the hub. Why do airlines tend to do this?

There are three main possibilities that the paper explores:

1. Passengers prefer large airports which tend to be more efficient, so are prepared to fly via the hub airport.
2. Airlines benefit from lower running costs (economies of scale) at a hub airport.
3. Airlines find it easier to deter competition by operating out of a hub airport.

Aguirregabiria and Ho find that airlines do benefit from costs savings of a hub airport. But they also find that airlines running a hub and spoke network tend to run more loss making services. They interpret this as a way of large airlines stopping smaller competitors from starting to run those 'spokes'. They also find that by having a hub airport an airline can drive up the cost of other airlines running services to that particular airport.

Thus hub and spoke networks could be a way airlines deter competition. 

If proven, air industry regulators should look into how airlines manage their networks and whether they illegally deterring competition that would benefit passengers.

Who Are We Really?

Behavioural economics finds it's significance from the way it reframes the economic perspective of human beings. We are no longer simply called homo economicus. We are allowed to selfless, confused and cooperative.

But that is not entirely satisfactory. The economic paradigm is still in place. We are still primarily consumers. We exist to consume, and consume to exist. We may occasionally give some of our consumption to others, but our primary purpose is unchanged.

There are, as I see it, two main drivers of this consumption complex. Our philosophical leaders (economists and their mindless followers; politicians) who tell us that consumption is everything. And us, who seemingly have an innate drive to get more stuff. More stuff than we had yesterday, more stuff than our neighbours have today.

The problem is that consumption isn't everything. We are more complex than that. Happiness is not just a function of consumption (c.f. friends, family). We cannot be reduced to one-dimensional, consuming robots.

And if that were not enough to make us think twice, limitless consumption is not actually achievable. Our environment has limits. The planet has limits. There are only so many fish in the sea. As Stewart Wallis (of the think tank the new economics foundation) recently said in a TED talk, we need to move on from seeing ourselves as consumers to stewards. Based purely on pragmatism, a serious change in our self-image is needed.

Temptation

The neoclassical economics way of viewing people involves assuming that people are fully capable of making the best decisions for themselves. If a neoclassical economist were to see someone (lets call him Nick) blowing all their savings in a Las Vegas casino he would simply assume that Nick had done all the sums and had concluded that his own personal benefit was maximised by gambling his life savings away.

That may be the case.

But behavioural economists tend to think not.

What if Nick knows what is best for him but chooses not to do it?

What if Nick has finite powers of self-control?

What if Nick hasn't even done the sums properly? He might be aware it's not the best option for him, but not just how disastrous it is.

Behavioural economists use the term 'bounded rationality' to refer to instances where people are clearly less than perfect (often by their own admission). If Nick may overly value short term fun over long term welfare. He knows he's being irrational, but he might need a helping hand to maximise his long term welfare. This is why people are increasingly offered commitment devices - in a rare moment of clear thinking Nick could opt into a scheme which bans him from casinos.

Corporate decision making can also fall foul of self-control issues. A timely example is the over-fishing of EU waters. If politicians really had the long term welfare of fishermen (and of the fish!) in mind then they would restrict the amount of fish that can be caught.

 

Taxing Rebate

If people are rational they know how to best spend their money, regardless of how and when they get it. So a recent study by Naomi Feldman (2010) raises some interesting questions.

Feldman examined whether a seemingly unimportant change in US tax law changed savings rates. In 1992 the George H. W. Bush administration changed how income taxes were collected. The amount of the taxes stayed the same, but less was collected each month. This tended to reduce tax rebates at the end of the financial year, but increase monthly net income. It didn't change the amount anyone paid in tax, but changed when they pay it. It meant there was a shift in income. Instead of receiving a sizeable yearly tax rebate, households had a larger net monthly income. Traditional economics would not predict any change in behaviour.

However, Feldman found that people saved significantly less after the change.

The yearly rebate was large enough to be seen as special and so people carefully considered how to spend it, typically saving a high percentage. The small monthly increase in income, however, was just added to household consumption budgets and so typically not saved. This is an example of mental accounting - the process by which people sort income into different budgets and tend not to shift money between budgets.

Thus we have a textbook example of the law of unintended consequences - a minor change in tax law causing huge shifts in behaviour. An argument, if ever there was one, for the use of behavioural economics in policy making!

Heston

You are called Heston. You need to make a bar of chocolate. The process to make this bar of chocolate uses three machines, one after the other. All three need to work for the chocolate bar to be made. Each machine has a success rate of 90%. What is the probability that the whole process is a success and the chocolate bar gets made?

Answer = 73%. (0.9 x 0.9 x 0.9 = 0.73)

People often guess a higher success rate than that. This is an example of the confirmation heuristic, whereby people overestimate the probability of conjunctive events. A 'heuristic' just means a rule of thumb, we use them all the time and mostly that's okay, but sometimes they lead us to make systematic mistakes just like the one above.

Percieved Intentions

This interview with the founder of Wikipedia about the takeover of Tumblr by Yahoo touches on how internet advertising works, and why it is important to get it right.

Jimmy Wales (BBC)

As he says, 'it's a social contract'. That is, things that aren't strictly financial, like intentions, matter. For example, if I think a website is taking advantage of me by throwing too many adverts at me, I might be offended or put off using that website. But if I perceive the website as just using fair means to make ends meet I might put up with the same amount of adverts. In summary, perceived intentions matter. (And as we speak behavioural economists are beavering away to include intentions in standard economic models...)

Coca Cola

I recently noticed an interesting pricing strategy followed by Coca Cola: at my local One Stop grocery store, a 2 litre bottle costs £1.99. But if you buy two bottles, together they cost just £2.50. The second bottle cost 51p - a quarter of the first one!

Needless to say, I bought two. Who wouldn't? 51p for two litres of coke is a great deal! How do we know it's a great deal? Because it would usually cost £1.99!

Very, very clever pricing strategy. (Charging £1.25 per bottle, regardless of quantity, would probably result if far fewer sales.)

They increase sales by anchoring our view of the innate value of a bottle of coke high (£1.99) before smashing it with a 51p deal - of course it's a great deal! (And the great deal feeling will probably increase the likelihood of buying coke again.) Additionally, I knew I only needed one bottle, but I felt that after investing £1.99 I might as well reap the rewards and put in another 51p. (The sunk cost fallacy).

Cold Beer

The following question was originally posed by Richard Thaler (1985):

You are lying on the beach on a hot day. All you have to drink is ice water. For the last hour you have been thinking about how much you would enjoy a nice cold bottle of your favourite brand of beer. A companion gets up to go make a phone call and offers to bring back a beer from the only nearby place where beer is sold (a fancy hotel). He says that the beer might be expensive and so asks how much you are willing to pay for the beer. He says that he will buy the beer if it costs as much or less than the price you state. But if it costs more than the price you state he will not buy it. You trust your friend, and there is no possibility of him bargaining with the bartender. What price do you tell him?

Now imagine instead of there being a fancy hotel there is only a small, run down grocery store. What price do you tell him?

In the experiment half those questioned were told it was a fancy hotel, half were told the grocery store. Interestingly, the answers differed.

The median response for the hotel was $2.65 while for the store it was only $1.50 (in 1984 dollars).

This contradicts standard economic theory where our preferences are supposed to be stable, regardless of who we interact with. I should value a beer the same regardless of who sells it to me. But as the example above shows, we use reference points. What we are willing to pay for a beer is not only based on our thirst, but also on our perception of a fair price or a good deal.

Passport Prices

I recently went through the process of getting a new passport and was interested to see the government using some price discrimination: different prices for different people. People self-selected into the different pricing categories. Why? Because the government charged different prices according to how quickly you wanted your passport:

4 weeks - £72.50
1 week - £103
1 day - £128

Thus those who are in a rush and thus prepared to pay more, pay more. While those who are in no hurry and thus prepared to pay less, pay less. Clever. Much cleverer than just one standard price. And it is used by firms/governments all the time. Price discrimination by time is ubiquitous because there will always be some impatient people.

The Confirmation Trap

Example taken from Bazerman and Moore (2009) based on Wason (1960):

Imagine the following sequence of numbers follows a rule, and that your task is to diagnose that rule. When you write down other sequences of three numbers, your instructor will tell you whether or not your sequences follow the rule.

2-4-6

What sequences would you write down?

Commonly guessed patterns include "numbers go up by two" and "the difference between the first two two numbers equals the difference between the last two numbers". In fact, the rule was much broader: "any three ascending numbers". But people had only tried sequences that tested their hypothesis by accumulating evidence that confirmed it, but didn't test for wider rules. They fell into the confirmation trap by just seeking to confirm their suspicions by trying sequences like 1-3-5 and 22-24-26 instead of trying, for example, 1-2-3 or 1-2-10.

This is called the confirmation bias. In all walks of life, from politics to business, people seek to confirm their beliefs rather than really test them.

People ask "May I believe it?" rather than "Must I believe it?"

Econitus

The following problem is adapted from Bazerman and Moore (2009) Judgement in Managerial Decision Making:

Lisa is worried about her health. Her doctor tells her not to worry too much as there is only a 1 in 1,000 chance that women of her age has the dreaded Econitus virus. Nevertheless, Lisa remains anxious about this possibility and decides to obtain a test that can detect Econitus. The test is moderately accurate: When someone has Econitus it delivers a positive result 86% of the time. But there is, however, a small 'false positive' rate: 5% of people produce a positive result despite not having Econitus. Lisa takes the test and obtains a positive result. What are the chances that she has Econitus?

0-20 percent chance
21-40 percent chance
41-60 percent chance
61-80 percent chance
81-100 percent chance

The correct answer is 1.7%!

If you answered 86% then you fell into the common trap of ignoring 'base rates'...

If 1,000 women like Lisa take the test, 999 will not have Econitus. But the false positive result means 50 will be told they have Econitus. Therefore there is only 1.7% chance that Lisa has Econitus (trust me!).

What this demonstrates is that people ignore background information in favour of more salient information about a specific case.

Hopefully this will be useful the next time you are given statistics by a doctor...

Boy Or Girl??

The following problem is taken from Bazerman and Moore (2009) Judgement in Managerial Decision Making:

You and your spouse have had three children together, all of them girls. Now that you are expecting your fourth child, you wonder whether the odds favour having a boy this time. What is the best estimate of your probability of having another girl?

6.25% (1 in 16), because the odds of getting four girls in a row is 1 out of 16

50% (1 in 2), because there is roughly an equal chance of getting each gender

A percentage that falls somewhere between the two estimates (6.25-50 percent)

The answer is 50%. (The sperm that determines gender of the child does not know the gender of previous children! In other words, the gender of each child is independent of that their siblings.)

So what explains our potentially wayward intuition here?

According to Kahneman and Tversky (1974) "Chance is commonly viewed as a self-correcting process in which a deviation in one direction induces a deviation in the opposite direction to restore the equilibrium. In fact, deviations are not corrected as a chance process unfolds, they are merely diluted."

So there you go, hope this helps when you're expecting your fourth child...