And here’s a scene about the rationality of charitable donors, inspired in part by Landsburg’s “giving your all” theorem:

d.  Charitable Giving

Bennett and DiLorenzo (1994) argue that due to donors’ rational ignorance, the market
for charitable giving works poorly.  But given that a person finds it worthwhile to donate in the first place, it is unclear why search effort would be especially low: in either case, gathering
more information lets you get more for your money.  Perhaps some people only donate to charity in
order to feel better about themselves; but they will only feel better because they believe that their donation makes the world a better place.  Suppose, in other words, that self-image is one of the arguments in their utility function, and they maximize utility given their psychological “technology” for transforming the (subjectively) expected social value of their charity into self-image.  With rational expectations, even donors who care solely about their self-image would acquire information
and otherwise act as if they were trying to maximize the (objectively) expected social payoff of each of their charitable dollars.

Yet people often seem to allocate their charity in a suboptimal way given their moral preferences.  Landsburg (1997) shows that under weak assumptions a rational altruist will only donate to one
charity.  Intuitively, unless you are both extremely wealthy and extremely generous, your individual donations won’t change the most important cause in the world into the second most important.  In practice it is much more common for people to give small amounts to a variety of causes.  Similarly, if charities compete for contributions from rational altruists, fraud and deception by charities should be no more prevalent than in the for-profit sector; like consumers in other markets, rational altruists will somehow verify that they are getting what they pay for.  Yet in practice the level of malfeasance in the charitable sector seems unusually high. (Bennett and DiLorenzo 1994)

The model of rational irrationality offers a partial explanation.  If self-image arises from how much good you believe your donations do, then irrationally over-estimating your impact can be utility-enhancing.  Monitoring charities and supporting the most efficient one increases the impact of your charitable dollars on the world, but reduces the impact of your charitable dollars on your self-image.  As long as the donor can sustain the belief that the charitable market is working efficiently, he gets the greatest possible psychic payoff from his donation.

Consider the argument that the rational altruist should only donate to one charity.  Adopting this principle would enable any donor, whatever his viewpoint, to increase the beneficent impact of his donations.  But avoiding this insight has offsetting benefits: You need not acknowledge that your past donations were suboptimally allocated; neither do you indirectly oblige yourself to ascertain the identity of the most deserving cause in the world.

If I remember correctly, one of the reasons I cut this section was because Alex Tabarrok told me it was unconvincing.  I haven’t looked at it for about eight years, but I still think it’s on-target.  Your thoughts?

P.S. This will be my last post for a while.  If I have convenient Internet access and enough spare time, I may blog from Singapore and/or Malaysia.  Otherwise, the next time you’ll be hearing from me is around Thanksgiving.