Better Living Through Game Theory

Michael Giberson

From The Economist comes news of a game-theory based software tool that divorcing couples can use to divide property:

Researchers in Australia have developed a computer program that relies on a branch of mathematics known as game theory to produce a fairer outcome when dividing property. Instead of the traditional approach of dividing a couple’s property in half, the system, called Family Winner, guides the couple through a series of trade-offs and compensation strategies. According to John Zeleznikow, a computer scientist at Victoria University in Melbourne, who developed the software with his colleague Emilia Bellucci, the results are fairer because both parties end up with what they value most.

Hadn’t thought of this application before, but I’m sure you could design a combinatorial clock auction to do the same thing. I wonder how the “Family Winner” takes account of combinatorial complications — maybe it isn’t a serious issue for property divisions.

Hat tip to GeekPress.

2 thoughts on “Better Living Through Game Theory”

  1. There’s a very simple procedure to arrive at a mutually agreeable property division. You can do it with a very simple spreadsheet – no custom software needed. It’s based on the premise that all property has a price.

    1. List all property to be divided.
    2. By some method, assign the roles of cutter and chooser.
    3. The cutter assigns a dollar value to each item on the list, such that the cutter would be equally happy with the item, or the money.
    4. For each item, the chooser choses the item or the money.
    5. Total the value of the items the chooser chose, and items the cutter will be keeping. Pay half the difference (in cash) to the shorted party, so that the columns are now equal.

    In my particular case, it was used to divide kitchen gear. You could apply it to real estate if you had to though; you’d just have to come up with a creative way to finance the value differential.

  2. Looks simple, but effective.

    The cutter gets the harder job, and it may put him at a strategic disadvance – you are in essence asking the cutter to reveal his reservation price for each item. But so long as we are limited to miscellaneous household goods, the likelihood of the cutter being seriously disadvantaged is probably small.

    (But, you know, as an economic consultant I can’t see much advantage in recommending approaches that are so simple that they can be implemented without me. Combinatorial auctions, on the other hand, that’s the good stuff!)

Comments are closed.