Behavioral Game Theory

Any health service needs to ration the care it provides somehow - there are not enough resources to pay for all and every course of treatment that may be of benefit. The UK has, in principle, a highly institutionalised way of rationing care through the National Institute for Health and Clinical Excellence. The basic idea is that NICE studies the medical evidence and makes an informed cost-benefit judgement on the desirability of a particular treatment. If the social costs look bigger than the social benefits then they provide guidelines to the National Health Service that the treatment should be provided. The advantage of this system is that it promotes systematic, objective, universal judgements.       But, this week we saw some of the problems with such a system. NICE announced that IVF treatment should now be available to women aged up to 42, rather than the current 39. It was decision that prompted a lot of discussion. Most discussion seemed to revolve around two distinct issues

It has been a bad season for the mountain rescue teams in Scotland - another avalanche yesterday, coming just days after a student from Leeds University died. I know nothing about these individual cases so would not want to comment on them. But, they are a reminder that the most important skill in mountaineering is not reading a map, using an ice axe, or knowing how to tie onto a rope - it is knowing when to ditch plan A and go onto plans B, C and D. The human propensity for time-inconsistency, however, makes this a surprisingly difficult skill too master.    For example, I remember many years ago attempting a quick trip up Snowdon in Wales. I'd been marking essays or something and needed some respite. It was a gnarly day - the wind was very strong, visibility poor, snow, ice etc. I met a mountain guide half way up with a few clients and he confidently predicted that no one was getting to the top today. I carried on and got within 200m of the top before finally turning back. Thi

The Premier League has recently agreed to new financial controls. Part of the deal is a cap on player wage bills. Economists usually criticize attempts to control market prices. But, sometimes a cap has merit, and I think this is one example.    First, let’s go through the standard argument against a cap on salaries. Suppose that a player called Hotshot would add £10 million in revenue to a club. Then, competitive logic says Hotshot will be paid £10 million. Any less than this and other clubs would have an incentive to buy him and pay him a higher wage. Hotshot’s marginal product is £10 million and so that’s what he will be paid. A cap on salaries would just force clubs to find some roundabout way to reward the player for his worth. For example, they could offer him a longer contract, or perks. All a cap does is to distort the market.     The flaw in this logic is that football is a zero-sum game. One team’s loss is another’s gain. The situation is, therefore, more akin to an a

Ask any self-respecting economics graduate the definition of a public good and they will tell you the answer: it is a good that is non-excludable and non-rivalrous. Which means that once it is provided everybody gets to consume it, and that one person’s consumption of the good does not detract from another person’s enjoyment of it. Simple enough, one might think. But, I sense a lot of disagreement about the nature of public goods. For example, I presented a paper recently on step-level public goods, and the audience was keen to dispute that a step-level public good ‘really is’ a public good. So, what’s the problem?    The definition of a public good only tells us about the consumption side. It leaves a complete blank as to how the public good will be produced. And this is where the differences start to surface. To most economists, if you say public good, the picture that first comes to mind is of a smooth linear or concave production function; the more people contribute towards t