Irrational Coursemates

This is an experiment used by one of my lecturers (Prof Seidmann) last week on 100 MSc Economics students at the start of a lecture...

We were told the following:

Choose a whole number between 0 and 99 (inclusive).

He will calculate the average (mean) of the numbers chosen by everyone here and divide by 2.

The winner is the person that chooses the number closest to this (half the average).

What number would you pick?

I picked zero.

This is because it is the rational thing to do. If the average is 50 then half the average will be 25. But if half the average is 25 then everyone should choose 25. Then half the average will be 12.5. And so on... Eventually you end up at zero.

My calculation, however, was flawed. I had assumed that MSc Economics students are rational. Further, I had assumed that MSc Economics students think that MSc Economics students are rational.

One person sitting near me was irrational and put 50 (possibly they misunderstood the instructions). Already I knew that my guess of zero was not going to be exactly correct.

In the end the correct answer was about 13.

My first response was "Just how stupid are my coursemates?!"

But then I realised that many in the room would have suspected that others were irrational and so guessed a positive number. For example, the person sitting next to me put 7 although he knew the rational thing to put was zero.

Some people may have been trying to second guess what people thought people thought would do! And so on. Thus we cannot (yet) conclude that all my coursemates are stupid (as well as me).

In conclusion, even if you are rational, you may not always act as economists might expect because you might be expecting others to be irrational. Funny old world.

The Endowment Effect

I am very pleased to introduce a guest blog by Alex Silk. Alex is somewhat of an expert on the endowment effect and I have been bugging him for months to write this post: Enjoy!

The endowment effect is demonstrated in a really simple experiment that was conducted by an economist called Jack Knetsch back in 1989. The experiment had three separate treatments. In the first treatment each participant was given a (identical) mug, they were told that this was a gift. They were then each given the option of switching the mug for a bar of Swiss chocolate (which could be bought at the same price as the mug). The second treatment was the reverse of this; each participant was initially given the chocolate bar and was then asked whether or not they wanted to exchange it for the mug. Standard economic theory predicts that the proportion of subjects who end up leaving the experiment with a mug should be equal in both treatments (allowing for random error) – this appears to be a fairly reasonable assumption. So what do you think happened?

Well what Mr Knetsch found was that in both treatments 90% of people kept the item which they were originally given. Furthermore, in a third treatment where each participant was given a straight choice between the mug and the chocolate bar 56% of people chose the mug (where again economic theory predicts the proportions should be the same as in the first two treatments).

What the experiment demonstrates is something called the endowment effect: people value a good more highly when they are in possession of it. While this is a significant violation of some important economic theories (something that for your sake I hope you are not too concerned about!), on one level this may not seem that surprising to you: a child would value her favourite teddy bear more than an identical one sitting on a shelf in a shop. However, what may be surprising is the fact that other experiments have shown that virtually as soon as you take ownership of a good you value it more (unless you expect to sell it in the near future).

So the next time you buy a can of baked beans remember that, subconsciously at least, you value that can slightly more than each of the cans you left behind you in the shop. Isn’t that useful to know?

- Alex Silk