Unfairness of Splitting the Bill

This is part three on the economics of splitting the bill:
Part I - The inefficiency of splitting the bill
Part II - The inequality of splitting the bill
Part III - The unfairness of splitting the bill (this post)

In her comment to my post on the inequality of splitting the bill, Lise notes that people with smaller appetites or cheaper tastes suffer when you split the bill evenly. Notwithstanding the alleged trend for women to order steaks on dates (TimesSelect), she believes that women will suffer on average from splitting the bill. Note that this is not an efficiency issue; it persists even when people don't order more when they split the bill.

So why do we split 50-50 when it isn't "fair"? Peyton Young (who recently moved from Johns Hopkins to Oxford) has written several papers and a survey article on this topic. If we wanted people who ordered more to pay more, it would be natural for each person to pay their share. This is hard to get exactly right without a calculator and a fair bit of effort, as the bill doesn't calculate the tax on each item or cluster the bill. We like 50-50 because it is easy.

Of course, there are times when 50-50 isn't particularly easy. To me, splitting a $45 bill into $20 and $25 seems easier than splitting $22.50 each. This suggests another answer, which is that 50-50 is the default. If you aren't going to go with 50-50, then this requires a discussion of what the right proportions are in every case. This negotiation is rather costly, and could easily spoil an otherwise pleasant meal. Even if 50-50 isn't right in a specific case, deviating from it opens a can of worms few of us want to deal with.

While the stakes (steaks?) aren't too high in the case of restaurant meals, there are settings with larger consequences where people stick to an "unfair" 50-50. (Peyton Young discusses the proportion of crops that go to the farmer in sharecropping as an example.) Here, it seems that 50-50 is an anchor-point for negotiations.

Is this a big deal in the case of dinner? Even if 50-50 isn't the right split in a given case, in many situations it is probably right on average. I may order more than my companions at one meal; I may order less at another meal. As long as this nets out over the long run, who cares? Lise's concern was that it doesn't net out on average for women, who order less on average. This may be so, but this begs the question of whether the unfairness to women of splitting the bill nets out against the social norm that men pay for dinner on dates with women. And this begs the question of how men paying for dinner nets out against other gendered costs (e.g., clothing) and differences in income. This is a can of worms I don't want to open, and definitely a dangerous one to open on a dinner date regardless of who is paying.