On Divorces, Estates and Algorithms

Photo by Vic, used under a Creative Commons License

Given my particular area of practice, a few friends–as well as my own dearly beloved–have brought this article to my attention in the last day or so. It is  about Laurier math prof Marc Kilgour, who has come up with a system that formulates the fairest division of property in a matrimonial dispute or in the division of an estate, based on each party’s ranking of most to least desired items.

It’s an interesting idea, and it may be useful in a family law practice from time to time. More likely, it might be useful at the “pre-lawyer” stage, if there are squabbles about who gets what from the shared home–and if both people are engineers or theoretical types who will be satisfied by a theoretical, algorithmically-derived “fairest” division of assets. Kilgour himself acknowledges this challenge in the article, pointing out that the biggest issue is likely to be buy-in, once the algorithm has done its calculations, the results come in, and it turns out that your sister is getting mom’s coveted heirloom vase, even though you ranked that first, second and third on your list.

In other words: just think about how disappointing it can be to lose a “heads or tails” flip of the coin, then turn that feeling up to eleven, in the context of objects that carry deep, emotional attachments.

Still, I could see some genuine usefulness, for a subset of engineer/CS/highly rationalist folks who truly would find it mollifying to know that the shared assets were divided in the fairest way possible.

But, from a legal practitioner’s perspective, the fact is that by the time most couples are seeking legal counsel, they have, often as not, already divided the personal property–or have agreed to figure that out between themselves.

It’s rare that it’s the couch, or the bookshelf, that is the true point of contention. To put it another way, if a former couple are each paying lawyers hundreds of dollars an hour to argue over a bookshelf, it is stating the obvious to say that the dispute is probably not really about the bookshelf.

An algorithm that divides assets according to theoretical principles of fairness is not going to help, much though we all might wish it otherwise (because contrary to popular belief, most lawyers I know truly do not want to spend their time drafting letters about who gets the leather couch, or the IKEA bookshelf, even if that time is billable. We are as conscious as anyone that the money spent arguing about the couch could otherwise be used to pay for the client’s children’s university educations, or for the client’s retirement).

Pets are trickier, but if both parties really adore their dog, then it’s unlikely that any amount of inanimate objects will compensate for the other person getting to keep Rover, so I’m not sure how this ranking algorithm would help in that instance, either.

Still, I don’t want to minimize the theoretical achievement of the work. And, as mentioned, I can see some personality types really getting some use out of it.

People who are profoundly rational, but big picture, and don’t want to be bothered with the details of actually performing the allocation or division of the assets (because if they did, they also wouldn’t need the algorithm–they’d just divide it themselves) would likely appreciate such a system. As well, there are some people who are strongly motivated by concepts of fairness–namely, the types of people who willingly suffer inconvenience or worse, in service of the fairest result–but get emotional and resistant about the specifics. Someone like that might be willing to be bound by such a system, which would allow him or her to avoid clashing with the ex over emotional attachments to particular objects, even if the result is personally sub-optimal or represents a compromised personal outcome, so long as he or she believed that it was totally fair. Rare, I know, but it does happen.