It might be tricky to work in without introducing other distracting questions, but maybe something to do with the dollar auction game? The central issue is brought about because the second-highest bidder has a large sunk cost.
The central issue is brought about because the second-highest bidder has a large sunk cost.
That game is about credibly committing to match raises. The non-social way is to make one's incentive to match greater than one's opponent. The first player can do this by paying 51 or more cents.
The other players can try to signal insanity by continuing, or greater insanity by bidding on top of that, but this is all signalling.
In real life, sunk costs are most important in situations where outcomes are uncertain for reasons other than social. I think betting on who drew the higher card off of a deck, with an ante, raises, calls and folds as in poker, would be better, or something similar.
A couple of years ago there was an online auction site that a number of people semi-seriously described as "pure evil". Items would appear for sale, and you could bid a small amount, maybe $5 for a stereo or something. But you gave up that $5 there and then. People would eventually buy a $300 item for $60 or so, but the site would take in $1000 for that $300 item. Wish I could remember the name of that site...
This is much like a game I saw: the experimenter auctions off $20, with the catch that the second highest bidder has to pay whatever they bid. I think the intent of the example was that there are some games where the only winning strategy is not to play, but I think it fits into sunk cost well, too.
I think there are a bunch of these sites now such as http://www.bidzillion.com/ and http://www.offandaway.com/
I did the math and these sites are making far more money per item than if they sold them at the inflated prices they list them at.
This is true even of the money used on the website. E.g.: the right to place 50 bids had 124 bids placed on it, bids which would have cost 50 cents each if bought in the $500 bulk pack.
Brazen. If people spend their money after seeing that, there's nothing for it.
Real-life example (recently happened to me): you order a nifty new gadget at bargain prices from a foreign ecommerce site that ships worldwide. They ding your credit card and ship the item. Here's the catch: you forgot that imports are often subject to customs and other kinds of taxes, often apparently capricious and arbitrary. People find out when the delivery shows up at their door, and they're faced with the choice of paying the extra tax or giving the item up.
How to work this into a class simulation: make the disclosure of the extra amount gradual instead of sudden; introduce elements of chance to the accumulating amount (e.g. "your item was recently classified as hazmat, with a 5% extra fee for safe handling"); crucially, have the final tax amount sometimes exceed the item's original price.
Sunk cost psychology predicts that people will prefer forking out tax, over giving up on the item and keeping the money to purchase a new one.
This can be done demo-style (ask who would give up at various points by show of hands), group or individual exercise.
This is an example not a demo, but I always liked this example
In the case of a movie ticket that has already been purchased, the ticket-buyer can choose between the following two end results if he realizes that he doesn't like the movie: Having paid the price of the ticket and having suffered watching a movie that he does not want to see, or; Having paid the price of the ticket and having used the time to do something more fun.
Poker might give a good demo where you slowly learn that your opponent likely has a better hand.
Grad school.
If anyone knows how to play Risk, the deployment phase should end in five minutes with some obviously doomed troops in Australia.
If you don't know of a good example of such a game, odds are that any game you find will not be known by students either. Why not create a game designed to have sunk costs and be easy to learn?
Some ways to modify existing games to introduce sunk costs:
The first would be a good modification of the card game War played for money/chips, the second would be a good modification of bingo or blackjack, the third would be a good modification for blackjack.
I'm hoping to find something that can be done in 5 minutes or so, as a classroom demonstration (for the rationality curricula).
I find sunk costs have a large effect in the board game "Go" (so that beginners are instructed "not to throw good stones after bad"), and I assume it also does in poker, but both of those games are too long and too full of distractors to be used in a simple demo.
Thanks for any suggestions!