SotG, Morality and Economics

Some of the discussion on last week’s post has tempted me to discuss punishments within Ultimate. I don’t necessarily expect everyone to agree, but hopefully it’s an interesting way of looking at it that you may not have considered.

It’s not really another argument about referees/observers/self-officiation. Regardless of who makes the calls, there is a question of what punishment should result from a foul.

This will probably seem a little off-topic for a while, but bear with me…

The psychologist Dan Ariely and his colleagues performed a wonderful experiment selling chocolate truffles on college campuses. They varied the price, and studied the behaviour of their ‘customers’. At 10 cents each, they sold a bunch of chocolates. At 5 cents, they sold twice as many – more people took some, and each individual took more. At 1 cent, they sold twice as many again.

But when they gave them away for free, they ‘sold’ fewer, even allowing for the fact that yet more individual people took some. Each person who took any took far fewer – and the majority took only one each.

The situation had moved from an economic one – ‘I have x money and y other things to buy, so I will buy z chocolates’ – to a moral, social one – ‘People would think me greedy; I should leave some for others’.

Rationally, there’s almost no difference between 1 cent and free. At 1 cent, you really ought already to be considering your reputation or the enjoyment of others, because the economic cost is tiny. But, for the most part, you don’t. You just buy chocolates. You look at the cost and the reward, and decide that you should buy as many as you want to eat.

No matter how small the cost, there is a psychological difference between that and free.

What’s all this got to do with Ultimate?

Well, in Ultimate in most of the world, there are no punishments. Zero. Zilch. Cheating is ‘free’. Cheating is unequivocally a moral issue, and social norms govern the behaviour on the pitch. If you’ve ever wondered why SotG works, then this experiment may help to show you. It’s not because we’re all nice guys; it’s because we’ve specifically made the reward/cost ratio infinite, and hence kept the whole thing in the moral realm.

There is no ‘economic’ cost whatsoever to cheating, and there’s always a benefit. So whether it’s nudging your guy on the mark at 1-all or hacking down a thrower on game point, the economic calculation is the same – of course you should do it! There’s no cost. Every opportunity to cheat has exactly the same reward per unit cost – infinity! So economically we should cheat all the time. But we can’t cheat all the time, because the game wouldn’t work, and no-one would ever play.

Economics without morality gives no useful guide to behaviour, because its recommendation – cheat all the time – is unworkable.

Instead, our behaviour has to be governed by social norms, by what is considered reasonable, by thoughts of our reputation, by concern for others. This works. If you put people in a situation where their decisions are explicitly social and moral, they will care about how others see them.

If you add in punishments, they might not. There is then an explicit calculation to be made about whether it’s worth it, and they won’t see it in moral terms. That might not be rational – just as a one cent charge shouldn’t make you selfish, the possibility of a TMF for double-teaming shouldn’t make you a cheat. But it just might, if you buy the analogy I’m making between the cost of spending cash and the ‘cost’ of a punishment in Ultimate.

And once we take morality out of it, even severe punishments (like a penalty kick plus a red card in football) are sometimes worth it – witness Luis Suarez at the World Cup illegally preventing a certain goal in the last minute, in the hope that the resulting penalty would be missed. It was missed, and his team won the game.

Punishments will not eradicate bad behaviour, they will merely control some of its excesses. And the price we pay for this is that those same punishments legitimise any behaviour that is ‘worth it’. They turn the cheat-or-not decision into an economic one, and we know how humans respond to incentives in those situations. There are more experiments to show this.

In the book Freakonomics, the authors famously discuss the case of the Israeli day care centre who were fed up with children being collected late. They instituted a fee for lateness, and the number of late parents immediately increased. The parents no longer felt guilty, and they were happy to pay a small fine for the flexibility of leaving their kids there longer.

One response to this is simply to increase the fine, which would of course work until the local billionaire decided he couldn’t be bothered to pick up his child. But the point is not about the size of the fine – the point is that the fine did not add to the existing social pressures; it replaced them. It moved from the moral to the economic. A larger fine might do a better job of replacing the moral pressure, but that won’t change the fact that morality, guilt or social pressure have disappeared from the equation.

Charging a ‘late fee’ legitimised lateness. It gave it a definite, measurable cost, and allowed parents to make an economic decision rather than a moral one. Psychologically, it gave people a possible answer to the question ‘Why did you do that?’ which is not tied to their moral failings.

None of us likes to admit to ourselves that we’re filthy cheats, though we may often feel a desire to cheat. Our self-image is important, and if we can find some way to describe our actions that doesn’t hurt our self-image, then we’re far more likely to cheat. You might steal from the till because the boss mistreats you, or you might drive too close to cyclists because you don’t believe they should be allowed on the road. You might avoid throwing to women because they’re not good enough, not because you’re a terrible misogynist (oh no, not you, never . . . )

Very few people act like a**holes in the full knowledge that they’re being a**holes – they find a way to lie to themselves about their motives. Economic arguments offer us that excuse we’re looking for. Instead of saying, “Yes, I’m a filthy cheat!” we can say, “Look how clever I am, I got a net benefit out of this! The rules are stupid, man, and I beat them. I win.”

Punishments in Ultimate will allow people to think that way. That’s not rational, but it is human.

Just as all players have a responsibility not to take the easy option (fouling), rule-makers have a responsibility not to take the easy option (punishments). It’s often very tempting to respond to unwanted situations by creating disincentives, but doing so may reframe the question in economic rather than moral terms. As Ariely puts it, “Policy makers should be careful not to add market norms that could undermine the social norms.”

I personally don’t believe that adding small punishments for minor offences (e.g. TMFs in the U.S. or, even more so, yardage penalties in pro Ultimate) will remove those offences; it may make them worse, as it replaces the social pressure.

Either we stick to the principle of the do-over and never punish, or some people will start to look at the whole game in economic terms. People are very, very bad at mixing economic and moral incentives – the economic point of view will tend to win if you give it even half a chance.

Let’s not give it a chance.