While scanning Randy Cohen’s recent Ethicist columns for something to complain about, I found this query about allocating faculty offices:
I am a faculty member at a university undergoing major campus renovations, including new office spaces. Departments were asked to determine their own ways of assigning rooms, but the task is complicated by factors like seniority and rank — does someone with tenure deserve a better room? Some faculty members have greater teaching demands and might need larger rooms to meet with students. What is the most ethical way to allocate offices: seniority? Rank? Lottery?
True to form, Cohen has nothing interesting to say, and offers no rationale for his random suggestions. It never seems to have occurred to him that scarce resources tend to be allocated most efficiently by markets. If he’d done a little research, he might have found this charming account of how the economists at Arizona State solved the office allocation problem.
I’m a little disappointed by the ASU story, not because I disapprove of what they did — though partially because I think they should have expected that other people would disapprove — but because I was expecting something a bit more creative, probably using some implicit medium of exchange other than money. How does the department choose a chair? How are teaching schedules assigned? I would think there would be other items to barter. (These, of course, would all have dollar willingnesses-to-pay, and if any of them actually can be bought and sold, the medium of exchange is actually fungible with money, but a lot of people’s irrational repugnances probably still wouldn’t be triggered.)
Any allocation of draw order will give a Pareto efficient allocation of offices, so any “efficiency” gains beyond that should include some sort of valuation of intensity of feelings. With perfect information about other people’s preferences, the values attached to offices imply values attached to draw positions, but without such information it could well be that, for example, the first person to choose has weak preferences over the two best offices, while the second has a strong preference; I think it would be better, therefore, to have a trading round after offices have been selected; the assignments don’t “close” until a week after they’re chosen in the designated order, with people swapping offices during that week. The evasion of repugnances that they managed could probably be maintained, with dollars changing hands, only if nobody is allowed to profit; i.e. if I pay $200 for my spot, I can’t accept more than $200 to swap offices with someone else. I think swapping the order in which we choose which course to teach next semester would raise fewer eyebrows, though.
Yeah, that column got even more disappointing when I started dating a professional ethicist and realized how fascinating those kinds of discussions *could* be. Maybe I should get him to start a counter-blog, where he analyzes each question looked at in The Ethicist, but properly and interestingly.
Out of interest how do they decide on how teaching responsibilities are allocated? Like this? If not why not? i.e. what is a better way of determining how teaching responsibilities are assigned and why is it worse (than bidding) as way of determining office assignments???
I can imagine there being some graduate course where only 1 or 2 faculty members could possibly teach it but this strikes me as the exception rather than the rule. (Although I’m not sure how to get around it.)
Why not use a rank order ballot grid? Have room locations across the top (x-axis) and people’s names down the left (y-axis). Each professor could rank the rooms in order of their own preference, and rank the potential occupant in each room in order of preference, all on one handy grid. People could then trade their votes (or something more tangible for votes) in order to get the room they want. On a certain date, finalize the votes, determine the allocation of rooms to maximize overall satisfaction, and start moving in.
It might be difficult to find the peak utility order (probably NP-hard), but it should be manageable — you probably don’t have to worry about hundreds of professors, and that’s what computers are for. Plus, if a professor leaves, you might be able to determine more easily who gets his or her office.
As an interesting extension, it may be possible to come up with a similar way to match students, professors, periods, and classes, though that would be even more complex. It would be kind of fun to watch a course election, though, with groups lobbying for particular lectures at particular times, or banding together to get the professor they want.
Because while markets may be good at “efficiency” they are not good at “justice.”
The most effiecient answer would be the office space belongs to the university or company and it is theirs to allocate however they see fit, and employees are lucky to have a space at all. But space wars aren’t about efficiency, they are about justice.
A more just answer may lie somewhere else.
Quick example to illustrate the point. 2 offices, 2 professors. One office has a sweeping view, the other a glorified broom closet. One professor, Oprah Winfry, is on faculty, but teaches only occasionally due to her other obligations (she actually was on faculty at Northwestern’s business school). The other professor, Prize Winner, has been on the faculty for 25 years, and her research is what gave rise to the growing fame and prestige of the university. Professor Winner also happens to have a husband in the hospital and 2 kids in college. The efficient market would probably put Professor Winfrey in the office with the sweeping view. The more just solution would probably allocate the offices the other way around.
Michael pointed to an intersting example of students and professors and classes. A lot of universities have gone to a bidding system for classes. That works a little better because all students start off with the same number of points to allocate, so if a student really really wants a specific class, they can submit a “suicide bid” putting all their points on a top choice, leaving the rest of their schedule to whatever is left over.
To build on Izzydog’s point, the unfairness of this system (the Arizona case in particular) is resident in the unequal starting positions of the auction participants. The question of ethics does not degrade simply to some abstract notion of efficiency but instead includes notions of fairness.
‘But space wars aren’t about efficiency, they are about justice.’
The real estate market is a ‘space war’, isn’t it? And how does rationing by price work there?
I’m old enough to remember when ‘justice’ was a factor in allocating space in that market, it was called racial segregation, and was legally enforced through covenants in RE deeds.
How, I wonder, with a fixed inequality (the differing qualities of the offices) can there be a more “just” allocation in the distributional justice sense? Somebody wins and somebody loses, regardless. The only way to achieve distributional justice would be to equalize the qualities of the offices or make those who are lucky enough to receive high quality offices pay a transfer to those who get low quality.
Why would a lottery be just means of allocation? We think of inequalities arising from the “birth lottery” as injustices that should be righted by equalizing opportunities (making the offices more equal) or income redistribution (transfers from the lucky office recipient to the unlucky). If the “birth lottery” gives rise to distributional injustice, how can an office lottery give rise to distributional justice?
In this light, the AZU scheme of distributing high quality offices on bids to make transfers to a department public good may serve distributional justice as much as it does efficiency.
I like the bidding system in this case. As long as the worst office
is about as good as the best office, minus the view, it’s a
reasonable allocation (I’m not worried about “best” in this
instance.). It’s readily prone to abuse, though. Suppose the
worst office was a literal broom closet. Each person would be
forced to bid artificially high simply in order to avoid being the
lowest bidder.
Really, the fairest system would be to make selections in order of
how long the person has spent in the office over the last [selected]
time period. Of course, that has its problems, too. Suppose X
needs the nice view to impress people coming in to approve a
research grant.
What strikes me is that the offices were so cheap. It only cost $75 for rights in perpetuity to a window office? Wow, that’s cheap, even for 1983.
Assuming you have enough notice of leaving that you can sell your office to someone else, you’re paying only the opportunity cost of the $75. What were interest rates then, maybe 12% or so? That’s about $10 a year. It cost 4 cents per business day to have a window office.
You sure these were economists? :)
@those who think the auction doesn’t meet a criteria of justice
The problem with looking for a “just outcome” is that the final outcome doesn’t tell you the complete picture. A final allocation that seems just to you may not seem just to me. It’s far better to concentrate on having a fair process and accept the result.
A good process for room selection is an auction because a person’s willingness to pay is indicative of how much they value room choice. People have different initial wealths, but so what? Any other selection criteria that tries to place people in rooms according to how much they care about the room favors some sort of initial position.
Hmmm,
a sealed bid auction is not a market, which makes the point of the anecdote moot.
Office space has by definition 1 seller ( monopoly? ) and cannot be resold… To even think that an efficient market can be established in these circumstances is silly.
Grumpy,
TjD.