Giving Less Than Your All

As promised or, depending on your point of view, threatened, I want to revisit Steven E. Landsburg’s new More Sex Is Safer Sex, this time addressing his contention that, given certain assumptions, it is preferable for a person to give his entire charitable contribution to whatever he deems the most worthy charity rather than parcel out his charitable contributions among various worthy charities. (Title reference: Landsburg discusses this in his chapter, "Giving Your All.")

Here is the basic logic of the argument. Begin with the key assumption that among the various charities under consideration, they are all sufficiently large and address sufficiently large problems that, however much your contribution may be, it will nonetheless represent only a very small increase in their endowments and, when spent, similarly address only a very small part of the problem they seek to solve. Landsburg uses CARE and the American Cancer Society as examples, so I will, too. The thinking here is that your $10 or $100 or $1,000 isn’t in and of itself going to be the determining factor in finding a cure for cancer, nor will it feed all the hungry children in the world.

Let’s say you plan on contributing $1,000 to charity and, as a preliminary matter, thought you’d make your contributions in $100 increments. If you deem feeding hungry children a better cause than cancer research, then your first $100 will go to CARE. Landsburg’s argument, in the proverbial nutshell, is that however much good your $100 did to feed one or more hungry children, the number of hungry children is vastly larger, the other children (metaphorically) waiting in line to be fed next are equally deserving of your charity and so your next $100 should go to CARE for the same reasons your first contribution did.

The size of the problem and of the charity is critical. Looking at small scale charitable contributions, e.g., should you contribute $100 toward fixing up a playground for children or toward fencing in a neighboring dog park (my examples), even if you like dogs more than children, at some determinate point the fencing gets paid for and it makes sense to contribute to the playground as well. That is, as Landsburg claims, you can make a real dent in small scale problems whereas your contribution, viewed in isolation, cannot make such a dent in the overall problem of world hunger or medical research.

So far, so good. Of course, we’re simplifying matters here by considering only two charities, whereas the world is filled with other possible objects of your charitable attention. (The Ridgely Early Retirement & World Cruise Fund springs to mind here.) In principle, however, you could rank the worthiness of every such charity and one would eventually come out on top. If you really couldn’t decide which of your top two charities was worthier, Landsburg says “flip a coin and give everything to the winner. If the two causes are equally worthy, sending $200 to either is just as good as sending $100 to each – and it will cost you just one postage stamp instead of two.”

Well, no. Landsburg “does the math” in an appendix to make his point. The math is good; the assumptions underlying the math, not so good.

Landsburg’s argument depends on distinguishing between the satisfaction, however derived, one gets by giving to charity and the good such contributions do for others. Analytically, that makes perfect sense. Insofar as we are capable of drawing that distinction and focusing solely on the latter, the math works out just fine. Unfortunately, however, his “defense of pure reason” (which is more Spockian than Kantian) presupposes that people are capable of arriving at moral conclusions by reason alone; that is, that they are capable of and should be willing to set aside the self-serving motives of charity and to do the research required to crunch the numbers.

In fairness, Landsburg acknowledges both the reality and usefulness of self-serving motives and the limits to which one can, should or will incur the search costs of ferreting out charitable bang-for-buck. But it seems to me he significantly underestimates them both. As with voting, information costs can be formidable and so there is a real element of rational ignorance involved in deciding among charities for most of us. But, okay, let’s say that putting in some time and effort sorting out charities is legitimately a part of our overall charitable contribution.

It may be true that, having thus invested a reasonable amount of time and effort into investigating not only, as in Landsburg’s oversimplified model, the endowment of the American Cancer Society and CARE but also their relative overhead costs, likely other sources of income (Landsburg says it shouldn’t matter if I know you are also going to give $100 to CARE, and he’s right. But what if I discover that some billionaire has just left his entire estate to CARE the day before I write the check. Might not that matter? Assuming you were dumb enough to contribute to NPR in the first place, might not Joan Kroc's $200 million contribution a few years ago have rationally swayed your coffee mug purchasing "membership" elsewhere?) and so forth, I determine that $200 to one is as good as $100 to each. Oh, and forget the stamp, they send pre-franked envelopes and there’s always the internet to give through, anyway.

Well, then, it would be irrational (and in that strictly utilitarian sense, immoral) at that point for me not to consider self-serving reasons why I might wish to split my contributions. I would, as economists say, have failed to maximize utility, would in effect have decreased the net wealth of the world by not taking my own happiness into account. To tart up the point with a bit of slightly misused economics jargon, once I truly am indifferent regarding the two charities in terms of the good they will do for others, it certainly doesn't follow that I should be indifferent as to other distinguishing factors.

I further question the underlying assumption that there is anything approaching an objective answer to the question: which is better, curing cancer or feeding hungry children? Landsburg blithely sets up the dilemma as one of blind instinct versus logical analysis, but logical analysis gets us to interpersonal utility comparisons and all sorts of other messy concerns. There is such a thing as the illusion of objectivity, too; and one of the most notorious sorts of such illusions is the mathematical formula which, upon close enough inspection, turns out to be using unmeasurable or incommensurable factors. I admit, however, that these concerns would require a more extended consideration than I am giving them here.

Viewed as a matter of economic logic, Landsburg’s key insight is that among two unequally worthy major charities, the marginal utility of one’s subsequent contributions to the most worthy would not be decreased sufficiently to justify giving to the second charity instead. Sure. It’s a great exam question, but it may still be highly questionable considered as real ethical advise.