The puzzle of sponsorship

THE PROBLEM: People give money to charity because it makes them feel good to do kind acts. But most people would not, spontaneously, give £50 to a random charity. But most people would sponsor someone to run the London Marathon, or do something else equally arduous, for about the same amount - even though the end result (charity gets £50) is the same. And anecdotal evidence indicates that for many such sponsors, the charity receiving the money is unimportant - they may not even know which one. What explains this behaviour?

THE ANSWER: Part of this could be explained by the pressure involved in collecting sponsorship: there's social pressure not to look mean when a colleague or friend announces that they're looking for sponsorship. Part, but not all: simply going round collecting donations would not produce as much money as going round looking for sponsorship. (I have no evidence on this but it seems fairly obvious). The nature of the arduous event - the Marathon or whatever - must play some role.

The theory I have arrived at is based on a casual remark by a colleague who described paying sponsorship as "a tax on friendship". This is close, but not actually true, because the runner does not profit from running the event - he doesn't keep any money, and it doesn't seem as though his friendships will be strengthened by his asking his friends for money.

I propose, instead, that running a sponsored event is a form of signalling. By announcing that he plans to cause himself significant discomfort by running a charity marathon, the runner is essentially saying: look at me, I am an altruistic person, and thus I am the sort of person who would make a good friend. Someone who agrees to sponsor him is basically exchanging money for a friendship with an altruist - a valuable asset - and also signalling back to the runner that they value his friendship enough to pay £50 for it. Essentially, it is an act of mutual potlatch - an action of conspicuous up-front destruction of value designed to demonstrate commitment to a relationship. Rather like an engagement ring, except that both sides in this situation are signalling, rather than just one.

TESTS OF THE HYPOTHESIS:
1. The greater the cost (in exertion) paid by the runner, the more he should try to profit from the signal. Therefore we would expect runners to approach more potential sponsors for a marathon than for a 10km run, and even more for a more demanding event such as an ultramarathon. Note that many events demand a minimum level of sponsorship for entrants - this is a distorting effect and such events should be excluded from the analysis.
2. The expected value of the friendship purchased depends on the expected lifetime. Therefore in a more fluid social environment, average sponsorship payments should be smaller - if you expect the runner to move to another company in three months, his friendship is worth less than if you expect to be working with him for the next ten years. We should also expect the average distance run to be smaller.

This post is based on a monologue inflicted by the author on a colleague who approached him for sponsorship for the London Marathon. The monologue was delivered on the Jubilee Line, eastbound section, between Green Park and London Bridge stations.