We’ve addressed the rent-splitting problem before. Now, Harvard astrophysics grad student Jonathan Bittner is seeking to solve the problem with a rent calculator to help roommates approximate fair prices in this “market.” The web tool uses a formula designed by Bittner (informed by a survey of his friends) to calculate how much each renter should pay based on room size, windows, closets, but also more detailed elements such as noise from street and whether the room has an awkward layout. Bittner writes on his site: “The idea is that you should pay the ‘market’ rate for these amenities (or get a market discount for annoying things) without having to think too hard…I like formulas, I like surveys, I like fairness, and I wanted to make something on the web. There’s something both fun and funny about using a mathematical way to decide what is fair.”
A New Formula for the Rent-Splitting Problem
TAGS: Housing
and 
. You can also watch it on 
I would have thought an auction set up would work just as well. Let the roommates bid on the rooms and it avoids the need for a formula altogether.
“Fair” is one of those words which has six milliard different meanings, isn’t it? There’s no way a randomly selected group of youngsters are going to agree on the value of an extra square foot of floor space compared to a west-facing window.
The value of this system, if it has any, is not that it’s fair, but that it’s a set of rules proposed by someone – anyone – who isn’t actually living in that house. Having the landlord come in and tell everyone what they’re going to pay would probably work just as well. Probably better, as those dissatisfied with their lot wouldn’t then waste time inventing a “better” calculator….
As mentioned by a previous commentator, an auction is an efficient mechanism.
However, I would not auction rooms, but the ranking of picking rooms. I.e. You bid for the right to pick a room first, second, third (to n-1 where n is the number of rooms, as the right to pick the last available room has zero value).
If you wanted to get fancy, you could auction rights on a second-price basis to avoid the winners’ curse.
If you wanted to get fancier still, you’d run this auction every month – preferences may adjust with improved information about what it’s like to actually live in a room.
Bittner points out at his website that an auction is a great way to split the rent, but notes there are reasons/situations where people would prefer not to do it this way. Also, while auctions are generally fair, they do give an advantage to those with better gaming skills (if you don’t believe that then invite a novice to your fantasy baseball draft). In many aspects of life it is okay to “beat” others because of better gaming skills, but for many (including frequently the “winner”) this would not work well in a shared living arrangement. I think it is a brilliant idea that at least gives roommates an equitable option for rent allocation.
@Ross Parker – can you explain when your system would result in a better outcome than simply auctioning off rooms? if i have idiosyncratic preferences (say, i really want the room with no windows), i probably know that, so i’d rather wait and get the room for cheaper than bid high to ensure i had the chance to pick the room. the resulting prices, if different at all, would reflect the participants risk aversion more than the relative quality of the rooms.
also, a second-price auction (which, when talking about economically efficient auctions) usually means seled bids. Since there is a total rent number you need the bids sum to, i don’t think it would make sense here.
@Ryan – You’d auction choice positions rather than places because in a situation with three rooms and three flatmates, you have two choices for auctioning:
1. Simultaneous room auctions
2. Auctioning the rooms in turn
Simultaneous room auctions are problematic. Say I want room A or B, and I really don’t like C. Do I bid hard for A and B at the same time? What if I win both? Do I pick? Re-auction the least favourable (to me)?
Auctioning rooms in turn gives me another problem. What if I exit bidding on room A, thinking I will have a better chance at Room B, but miscalculate other’s preferences?
Auctioning room choice rankings avoids these dilemmas. If I want room A or B, I know I have to bid for at least the right to pick second. If I want just room A, I know I need to bid for the right to pick first.
Also, I don’t think there ought to be “a total rent number you need the bids to sum to”.
You agree on 1/3 rent each at the outset. The room auctions are a secondary round of redistribution.
You can’t confuse the overall rent level negotiation between the collective and the landlord with the distribution of resources within the collective.