The Coase Theorem Rules at NYU Law

The most fundamental principle in the field of law and economics is the Coase Theorem, initially put forth by my friend and former colleague Ronald Coase, who is still intellectually active in his nineties.

The Coase Theorem says that — absent large transaction costs — resources will end up being efficiently allocated, regardless of who holds the initial property rights.

NYU Law School is providing its students valuable real world experience with the Coase Theorem, according to this ABA Journal article.

Class assignments are made by lottery. There are no waiting lists for classes. This gives students an incentive to sign up for the most popular classes, even if they don’t want to take them. If they win a seat in one of the most sought-after classes, they can work out a deal to sell their seat to another student. (The way this is done is by the person holding the winning lottery ticket withdrawing from the class and the other person signing up for it a few seconds later; since there isn’t a waiting list, this scheme will work as long as no one else happens to sign up for the class in that few-second gap.)

In the end, the students willing to pay the most for classes are the ones taking them, which is efficient.

As in other areas (like organ donation), using cash to determine who gets into classes upsets some people. One of my students got into trouble for trying to sell her spot in my class, for instance.

Most likely NYU’s response to this publicity will be to change its policies, hopefully to a bidding system that will also lead to an efficient allocation. I would argue, however, that they should maintain the current system as a means of teaching the students about the Coase Theorem in a way they are sure to learn it.

The other reason that NYU might want to keep the system is that now that prospective students know about it, there might be a surge of applications. Rumor has it that students are willing to trade not just money, but also sex, to get access to popular classes.

(Hat tip: Mauro Tucci, Jr.)


Allocating desirable classes by lottery is a silly way to do it. All students are paying for an education, not for a chance at an education. Being able to sell a spot in a class provides an additional incentive to waste time spent in college.

Betting points for popular classes makes a lot more sense. Granting the better students extra points might also work.


It seems to me there's an aspect of the problem that hasn't yet been addressed. This is a resource allocation problem, but what, exactly, is the scarce resource being allocated? If I'm the professor, then ultimately, what is scarce is my time and attention. (I can answer the questions and grade the papers of 30 students, but not 100). Why, then, should I have no say in how it is allocated? Does anyone want to argue that a market, however it is arranged, can allocate my time better than I can?

A modest proposal, then: if a class is overenrolled, each student wanting a seat writes a one page essay explaining why he or she should get it. The professor then decides who gets in.

Now that's efficient.


A quick scan of the comments suggests that no-one has looked at the big win for the NYU Law school in allowing places to be traded for money. The really rich get the classes they want. Thus raising the likelihood that they and their families will provide endowment money. How could any scheme providing equity for poor students compete with that benefit?


(1) You aren't allowed to sell classes for $$.

(2) The listserv where people offer classes is called "Coases." :)

(3) I really like the system, but we're getting rid of it.

K in LA

I think they ought to extend the bidding idea to the admissions process. Currently, "legacy" students are given an enormous advantage over the rest of the applicant pool. Furthermore, schools are notoriously reticent about discussing average GPAs and test scores of legacies vs. those of the incoming class, nor do they disclose what percentage of admissions are legacies in any given year. The stated rationale is that this system encourages alumnae to donate generously to the school. Why not then allow everybody to play? Allow anybody to "donate generously" and gain the preferential treatment previously reserved for legacies. The fair method would be to set aside 10% of admissions for the offspring of exceptionally generous parents, and simply auction thesse seats off to the highest bidder.

Michael Wayne Harris

Jeez, this is what I hate. Levitt makes two claims:

1) That the distribution of classes is efficient, thus being an example of the Coase Theorem

2) The current state of affairs is desirable and should be kept.

So when people say things like "Why is efficiency measured in terms of cash? That is not the goal of the University. The goal should be to educate the masses in order to make the world a better place," it just shows they don't actually know what's being said. What unit efficiency is measured in is unrelated to what the university's goal is. Here, "efficient" doesn't mean "doing it's job well," "efficiency" means something very precise as a piece of economic jargon. There is no value judgement attached to efficiency. "Efficient" and "inefficient" in the way Levitt is using the term doesn't mean "good" and "bad."

Saying the system is "efficient" is simply a matter of description. You could like the system and I could hate the system, but we could both agree on an accurate description of it. For example, people who advocate universal health care in the US and people who are opposed to it can agree that it does not currently exist. Saying "the system is efficient" is the same as saying "the drinking age in the US is 21," "the voting age in the US is 18," etc. You can disagree about whether this is how the world SHOULD BE, but you can't disagree that this is how the world IS.

Statements about equality aren't really helpful either, such as "This is what is wrong with this whole “freakonomics” way of thinking: it values “efficiency” or whatever you call it over other things. Even if this system IS more efficient (which I don’t think that it is), is that all that matters? What about equity? " Freakonomics doesn't value efficiency, freakonomics shows that the system is efficient. Freakonomics doesn't answer questions like "should things be efficient" or "should things be equal", it answers questions like "are things efficient" and "are things equal." Freakonomics doesn't value anything other than coming up with an accurate way of looking at how the world is. Levitt might himself have an opinion about how the world should be, but that's not the same as freakonomics valuing something.

Please, folks, get the distinction between "description" and "prescription" straight. Levitt was being descriptive, he wasn't saying anything should be a certain way. (Okay, he did provide a cutesy claim not to be taken seriously, and he did provide a few arguments that someone who would support the status quo might make.) The point is that, even if we disagree about whether we should change the status quo, we should be able to agree about what the status quo is.



Untamed Primate,

Assume Jim, the poor student with only $200, won the lottery pick for the class he values at $200 and Bob did not get in.

Bob pays Jim $400 for his spot in the class.

Jim increases his net worth by 200%, and the $400 is obviously worth more (namely, $200 more) than Jim valued the class in the first place.

Bob wins because he gets the class he wants, Jim wins because he gets an extra $200 (the $400 minus the value he puts on taking the class).

End result, Bob gets the class, Jim does not.


Now assume Bob wins the spot in the class in the lottery.

Bob values the class at $400, Jim only values it at $200.

Jim cannot buy the class from Bob.

End result, Bob gets the class, Jim does not.


In either case, Bob gets the class, Jim does not, and this IS because Bob values the class more than Jim.

If Jim valued the class more, like say $500, then he would not sell the spot in the class to Bob in the first example, and he would start offering other classes, favors, etc., to Bob...or sell his classes of lesser value until he has the cash he needs to buy the spot from Bob.


James K

The only reasonable way to allocate classes is first come, first served. If some students get allocated too many classes, they will have to see a staff member to say which classes they wish to drop, and why. Newly available places would be offered to the next people on the list.

"Efficiency", the way economists abuse it, is meaningless here for all the reasons given above. It's pretty much meaningless in every other context, too.

As regards fairness, there is only one resource that all students have in common: their time. Make students queue up IN PERSON to register for classes. They will register first for the class most important to them.

The same process would work for online applications too - server crashes aside.

The Untamed Primate

#15: I'm fine with Jim and Bob making deals on their own -- if Jim really values doubling his worth more than the class, then he should sell the spot. That's Jim and Bob's business, and we shouldn't stand in the way of it. But is that transaction really solely Jim and Bob's business, does the school play a role here?

I'm only concerned with how the school allocates spots. And in this case, the school is responsible for this auction system -- even though the school doesn't sanction the system -- because the school created conditions that would predictably cause the auction system to arise.

Strictly speaking, the school is not auctioning off spots -- it's merely distributing the spots by lottery. But by letting those with winning lottery tickets drop the class and letting others take their place, the school has provided a clear incentive for every student to put in his or her name for coveted classes. Hence the school has created conditions that predictably gave birth to a de facto auction system -- any economist could have pointed this out. If markets can rise, they will.

We wouldn't need to debate the particulars of this auction system if the school fixed its allocation system: Don't allow students with winning lottery tickets to withdraw from the class. In other words, if you put your name in for the class, you had better be willing to take it, because you can't back out if you win. Wouldn't that make sure the students who want a spot get one?

#14: Borrowing money won't answer here, unless the lenders don't function rationally. Let's say Jim, who's worth $200, borrows some money. Well, Bob, who's worth $2,000, can borrow some money, too. And Bob will be able to borrow more, unless the lender bases his or her decisions on some vague sense of Jim's goodness or desire, rather than how much collateral Jim can put up.



#46, with all due respect to you and the "intelligent people" who considered education to be a right, that's completely irrelevant to this conversation.

At the university I attended, there were a few highly sought-after classes. In most respects, these were prized because they were ENTERTAINING (multi-media, prestigious guest lecturers, terrific lecturing style by the professor, etc.). How can you say people have a right to elite higher education that is also entertaining?

The simple fact is that these classes are a limited commodity and there isn't enough to go around to the whole NYU student body. I don't care how intelligent you are, you can't eliminate the scarcity of something just by declaring it a right. If that were so, why don't we just declare oil a right so we don't have to keep paying so much for our gasoline?

Conor - Ireland

Apologies in advance for not reading every post, but no-one has really considered the university's position in these arguments. If indeed, classes are distributed on a lottery basis and students can then exchange at market prices, it ensures that students who are willing to give up their place on a course for a fee will be paid a sum equal to their valuation of it, or else they would not sell and simply take the course. Sounds efficent until you consider the university's (and society's) loss... The equilibrium reached will ensure the best teachers are teaching the richest students, not the BEST students.

Even if the poor students were able to borrow (and in reality there is an upper constraint on the borrowing power of teenagers) it does not ensure the BEST students are in the best (most popular anyways)teachers' classes. I'm not going to get into the permutations here, but it is easy to see there could be loss to the university (reputational) in having poor quality graduates.

In Ireland, university fees are FREE and places are allocated based on achievement in state exams, this ensures the highest achieving students get their desired college. Generally, we don't have limits on class sizes after that, if there are too many people for a course, they move the class to a bigger lecture theatre, simple. If we did have limits on class numbers, I would suggest that for the long-term benefit of the university, and society, that places should continue to be awarded on the basis of demonstrated academic ability.


J. Sweeney

If the school auctioned seats in classes for dollars the most popular classes would become profit centers for the school leading to more supply. Initially it will seem unfair, but like cell phones which were once very expensive and rare, more sections of the desirable courses will be offered and their cost will decrease making them affordable. This increase in production will continue until an equilibrium is reached.


Why not have a waiting list? Couldn't the same theory apply, that is, if a person wants badly enough to get into a class, he or she can pay for it? What problem is this solving? You still have the problem of more students than seats in the most popular classes, except now, there are people signing up that don't even want to be in the class. The traditional measure of whether or not someone wants to be in a class is that he is enrolled in the college and he signed up for it. Of course you need to shake down those who don't really want to be there, because there is incentinve for them to sign up! No incentive, no waste. This seems less efficient to me, from the standpoint of the student.


Wouldn't the Endowment Effect come into this?

I think once a student has a property right to a seat they would irrationally place a higher value on it than any other student who wants to buy the seat, because it is a right that they own.

It's unlikely they do not to want to go to the lecture at all as they put their name down, unless they only put their name down to try to get a seat to sell.

This could mean that the students with the property rights don't exchange it because they (irrationally) convince themselves that they value the seat more than what can be exchanged for it.

Therefore, the outcome might not really be that efficient.

I'd be pleased if someone would explain to me why this wouldn't have a significant affect?


Efficent? Can someone please explain how you can determine the 'efficency' of particualr person over another in taking a course?

what is the goal of University?

you Americans are always stuffing up these perfectly nice systems like free education with your absurd flakey 'science' of the free market.

Punditus Maximus

Yes, the difficulty comes precisely from the assumption that each dollar gives roughly the same marginal utility. This is probably true for a pair of businesses deciding who gets access to a given resource. It is, of course, profoundly untrue for persons of very different means.

Efficiency is irrelevant, though. The Coase Theorem fails in this case on a more basic level, as initial allocations do determine (or strongly affect) final outcomes in highly plausible cases.


What is it with people and equality and equity? When was the last time anything in life was truly equitable?

Sooner or later some top quality professors (probably of economics) are going to band together and start their own school, perhaps Market University. Seats will go to the highest bidder, the professors will be paid based on the revenue they generate from their classes. People will complain that they're corrupting education by making it all profit based, meanwhile the underfunded, overworked professors at state schools will wonder why their students can't keep up with the students at Market U.


If I'm not mistaken, a lot of business schools have auction systems. At least mine (Wharton) does, albeit not one based on USD. One of my friends takes the bidding process to a rather extreme extent though. He wrote an algorithmic trading program to bid for him in the auctions both to profit and to get him into the classes he wanted. I have to say, my only question about his behavior is why he feels the need to be in business school at all.


@30- Because a long time ago, some very intelligent people in this country decided that education should be a right to all and not a privilege. While a law school education is not necessarily a right, the idea that education should not be a commodity (though in so many ways it is already treated as such) is historically rooted in our society. If you don't like that, go to a country without the educational legacy and history we have. And, yes, our history is also marred with horrible education decisions, but the fact remains that one of the backbones of our education system is that all are entitled to opportunities to learn.


Why is efficiency measured in terms of cash? That is not the goal of the University. The goal should be to educate the masses in order to make the world a better place.