Predictably Irrational Chapter 3 – The Cost Of Zero Cost

To a neo-classical economist zero is just another price. To the average consumer it brings the magical connotations of free. We are always trying to get something for nothing so if something is free then consumers impulsively take the option. Ariely shows how this impulse comes with hidden costs that debunk the myth of rational consumers. Whether it’s from eating too much free food or accumulating worthless free pens, clickers etc, people are always trying to get a free lunch.

As usual (this is what is so great about the book) Ariely did an experiment to find the answer (neo-classical economists take note). He set up a stall with offering (one per customer) two piles of chocolates, the first being Lindt truffles (luxury chocolate) and the second being Hershey kisses (standard chocolate). He sold the Lindt for 15 cent and the Hershey for 1 cent. Due to their superior quality 73% choose Lindt and 27% chose Hersheys. So far, so good. An economist would say that Lindt was obviously a superior product and as the benefit from it outweighed its cost better than Hershey did, it sold better.

So Ariely knocked one cent off the price of each chocolate, so that the Lindt truffle cost 14 cent and the Hershey kiss was free (people were still limited to one). If Lindt was a superior product then it should hold its place as its relative price hadn’t changed, it still cost 14 cent more than Hershey. Sure the proportions had changed, but one cent is the same amount in your pocket regardless of proportions. You don’t buy goods in proportions; you buy them in absolute amounts. Any textbook will tell you that all decisions are done on the basis of cost-benefit analysis, so if the benefit of the chocolate outweighs the cost, then consumers will buy it. Seeing as the benefit (pleasure) of eating Lindt didn’t change and the cost relative to Hersheys also didn’t change, we should expect the same result. Instead the numbers were flipped and 69% choose Hersheys and only 31% choose Lindt.

Most of the time, choosing the free option doesn’t do any harm. However, when we have to choose between two options (like the above experiment) choosing the free option can make us worse off. Another example is Amazon who noticed that by adding the offer of free shipping to an additional purchase, got people to buy something they wouldn’t have otherwise bought. Some people may not have even wanted the extra product but the magic word free was too enticing. This slight change greatly increased sales, except in one country, France. That’s because instead of offering free shipping, the French division charged one franc. Although it was essentially free, it didn’t contain the magic word and therefore people didn’t fall under its spell.

Quick fire experiment. If you had the choice between a gift voucher worth $20 that costs $7 or a gift voucher worth $10 for free? Answer quickly and impulsively!

Which did you choose? Odds are you choose the free voucher, even though it was the wrong answer. The $20 voucher that costs $7 leaves you with a net of $13 and is better option, but most people are swayed by the prospect of getting something for free.

The magic allure of getting something for free can blind us to its hidden costs and lead us to make bad decisions. We buy the more expensive car because it comes with a free oil change (something that doesn’t cost much especially relative to a car). We choose the credit cards that are initially free before hidden costs kick in. Zero calorie or sugar free drinks feel like the healthy option, more than even one calorie drinks. Simply adding the prospect of free can greatly boost sales. This can also be used for public policy. Want to encourage people to get a health check, use environmentally friendly transport or get more education? Then don’t make it very cheap, make it free.


Filed under Books, Economics

8 responses to “Predictably Irrational Chapter 3 – The Cost Of Zero Cost

  1. is there anything that can be entirely free?

  2. If you approach the gift voucher question logically, it is still possible to prefer the free $10 voucher to buying a $20 voucher for $7. This is because a voucher could be lost, be forgotten and expire before you get a chance to use it. For example, if I estimate that there is 30% probability that I forget to use the voucher before it expires, it means there is 30% chance that I’ve lost $7 and 70% chance I got free $13. The total expected value would be $7, which is exactly the expected value for a free $10 voucher. If I think there is a higher than 30% chance that I lose the voucher, the $10 free voucher would be a better deal. If it’s a question about cash, which you can just put in your wallet and use whenever for whatever purpose without any special provisions, then the $20 for $7 is definitely a better deal and more people will take it.

    The candy example is a little trickier, but there is a similar logic in it too: when you get Lindt for $.15 or Hershey for $0.01, you have a chance in both cases to pay a certain amount of money for a candy, and there is a chance you will be disappointed in the candy (e.g., you decided that you didn’t like it, or immediately regretted that candy going right into your fat layer, or put it into the pocket to find it melted and no longer edible, etc.). So in this case you can assume that you are not getting your money worth, and would consider your payment of $0.15 or $0.01 to be completely or partially wasted. On the other side, there is a estimated monetary premium you can expect to receive from the candy: for example, if to you a Lindt is worth $0.10, you wouldn’t buy it for $0.15, and if you would have paid $0.25, you get a premium of $0.10. So you could also calculate your monetary expected value just as you did for the voucher, assuming a certain probability of disappointment and a certain premium. Depending on how you estimate the probability of disappointment and premium values, your expected value could be positive or negative. That would be true for $0.15 Lindt, $0.01 Hershey, and $0.14 Lindt.
    But a magical thing happens when you consider a free Hershey Kiss: because you are paying $0 for it, your expected monetary present value of the transaction will never be less than zero: you are either disappointed and lost no money, or you are not disappointed and receive the premium. So while in the $0.15 Lindt vs. $0.01 Hershey case you are looking at two choices both involving possible risks and rewards, in the second scenario $0.14 Lindt vs. free Hershey, you have one choice carrying possible risks and rewards (Lindt), and another choice where only rewards are possible. So if you are at all risk-averse, and don’t have a strong preference for Lindt, you should take free Hershey. In the first scenario where both candies aren’t free, you may go more on personal preference since risk/rewards comparison is very difficult to actually calculate in numbers. (I mean, how do you even calculate the probability of regretting something??:)
    Finally, in this risk vs. rewards explanation, it is actually the proportion of the prices that matters more than the difference between the prices. This is because when you compare these two candies, you are more likely to think “I’ll exchange a Lindt for X Hershey kisses”, than “I value Lindt as a Hershey kiss + Y cents”. You could easily foresee that if you raise the price for both Lindt and Hershey, while keeping the same $0.14 difference between them, you would see higher proportion of people choosing Lindt over Hershey.

    • Wow. That’s what you call well thought out response. I’ll admit, I did not see that coming. The inclusion of uncertainty and risk is an excellent point that never crossed my mind. Absolutely, if there was a risk of losing the gift voucher (which happens a lot) then the 0 investment might be the better option. However, in Ariely defence few people would take that into consideration and generally buy with the expectation of consumption. Its only businesses that include risk of loss or theft into their purchases.

      Again I didn’t think to include risk into the calculations about chocolate. I suppose while you are certainly right in your argument, that would only apply the first time when there is uncertainty. For repeat purchases of chocolate (i.e. most of them) there would not be this uncertainty.

      Ariely does include calculations in the book (which I excluded for space reasons). If the utility of Lindt is say 50 then the net benefit is 35 (50-15). If the utilty of Hersheys is 5 then the net benefit is 4 (5-1). In this case Lindt has higher utilty therefore it will be chosen. By reducing the price and therefore increasing benefit to 36 and 5 respectively, Lindt still has a higher utility and should still be chosen.

      • I’ve made this argument just to give an example of rational reasoning which appears as irrational behavior on the surface. I do realize that many people would not outline the argument exactly like that, but I think the instinctive desire to avoid the negative outcome (which is the main idea of my argument) would lead to the same decision.

        I agree that if you allow multiple transactions, a lot of risk goes away. This is why $20 for $7 would always be preferable to $10 for free if it only involves cash rather than vouchers, since the money will be mixed in with the rest of your cash.
        When you include utility values and allow for multiple purchases (not necessarily at the same time), it still makes more sense to buy Hershey (because when you will have spend 15 cents on Hershey, you will have received the net benefit 60 = 15 x 4) rather than 35 for 1 Lindt for the same 15 cents. In other words, you get more bang for your buck on Hershey. But if Hershey and Lindt price go up by 3 cents each, so that price difference remains, Lindt will have more net benefit per dollar spent.
        Of course, when you are limited to just one item, you could base your choice on just the net benefit, but then the uncertainty is back.

        • I suppose the point that Ariely is making is not that people are stupid or illogical, but rather they act differently to how we would expect. They are not rational in the traditional sense, but still follow an internal logical. Hence their actions are irrational in a predictable way.

  3. Pingback: SSL Connectivity for all Central Repository users Underway | Sonatype Blog

  4. Pingback: Grill – 'Normal' Activities

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s