Unthinkable: Should college places be awarded by lottery?

Using a lottery is preferable to distributing goods based on ‘bad reasons’, argues political scientist Peter Stone

Making decisions by lottery is often seen as an affront to justice. Witness, for example, the furore over the Camogie Association’s plan to settle a match by a coin toss over the summer.

But are there circumstances where lotteries should be the first port of call rather than the last? Would it be better, for example, to elect parliamentarians by lottery – a practice of early democracies in ancient Greece – rather than have special interests and campaign finance affect the outcome?

Dr Peter Stone of TCD's political science department believes there's scope for greater use of lotteries in society as a way of keeping "bad reasons" out of decision-making. Rather than seeing lotteries as a failure of imagination, he argues that those who dismiss lotteries can "have the failure of imagination because they think there must be a good reason for distinguishing [between options], even though we haven't found it yet".

School admissions are a "classic case" where lotteries are preferable, he says, and he extends the argument to third-level admission. Criticising the recent CAO reforms, which try to minimise random selection in the points race, he provides today's idea: "Not only should we not be reducing the amount of random selection [in college admissions], we should be letting more in."


When are lotteries a good idea?

“In general terms, lotteries are good when we want to keep reasons out of decision-making because we think people are prone to various kinds of biases, prejudices or bad sources of decision-making.

“Now, if you use lotteries you might also keep good reasons out. So you have to have reached the point where you’re more concerned about keeping out the bad reasons for influencing decisions than letting in the good ones, and I think that kind of situation is reached more often than people might think.”

In college admissions, though, isn’t it the fairest thing to judge people on merit?

“Where you are dealing with third-level education, there is no doubt that merit is supposed to be the over-riding criterion that is relevant, and I would in no way challenge that. But how confident are we that someone with 415 [points] is that much worse than someone who got 420? At some point that becomes statistical noise.

“In the American context, you have extreme, competitive, meritocratic processes to get into universities, and one thing these processes are very good at doing is reinforcing the stranglehold that the middle and upper classes have on getting into university. Because who has the time to do endless preparation for interviews and extracurricular activities? It’s the student who has resources.

“There could be very good, smart students from working-class or poor backgrounds who might look just as good on a cursory look, but then when you say, ‘We’ve got to have a precise measure’, that’s when you allow in all these other factors, and then it’s the kids at the bottom who lose out.

"The psychologist Barry Schwartz has a very interesting proposal in this regard. He puts the following challenge: he says, 'Let's see how good the admissions procedures are, let's keep admitting 90 per cent of the students through the traditional way but let's admit 10 per cent through random selection – above a baseline – and after 10 years let's see who's doing better'."

Is the US philosopher John Rawls a key influence here?

“Rawls gets mentioned quite a lot because he recognises the fact that, when people do not know the outcome of a process, they are going to take a guarded attitude to it. You’re going to be very cautious about designing a rule if you can’t make sure the rule doesn’t benefit you. In the school admissions case, if you know you can’t guarantee your kid gets into one of the good schools, then you are probably going to be concerned to make sure that even the worst schools aren’t that bad, whereas if you know your kid can avoid the worst schools, then where is your incentive to care whether they go to hell or not?”

Is reluctance to accept lotteries a sign that fairness isn’t taken seriously?

“Unfortunately I think it is the case . . . Sadly there are in most countries, including my home country the US, lots of people who do take the attitude, ‘Well, it’s not going to hurt me, so what do I care?’ Ideally, the way you deal with that is to have rules that apply equally to everyone – so you say, ‘There’s no special favours for you’ – but sometimes where you can’t do that perfectly, lotteries are a kind of substitute.”


Question: Would academics do a better job running the country?

William F Buckley replies: "I should sooner live in a society governed by the first 2,000 names in the Boston telephone directory than in a society governed by the 2,000 faculty members of Harvard University."