Peter Singer on Starvation and Charity

Tim Worstall just mentioned Peter Singer, so I thought I'd try actually reading something by Singer. Google coughed up Famine, Affluence, and Morality (1972), wherein Singer says we should all be sending a significant portion of our income to starving Bengalis, or whoever's starving at the moment.

It makes no moral difference whether the person I can help is a neighbor's child ten yards from me or a Bengali whose name I shall never know, ten thousand miles away.

But in fact, giving to a neighbor is preferable, in Singer's terms, because he sets an upper limit on what people are required to give:

...if it is in our power to prevent something very bad from happening, without thereby sacrificing anything morally significant, we ought, morally, to do it.

If we help our neighbor, we benefit by the improvement in the neighborhood. In effect, it's a rebate, which lets us give more at the same net cost to ourselves. Therefore, we can reasonably give more to a neighbor than to some arbitrary Bengali. This is realistic; people pay extra to live in nicer neighborhoods all the time (though others prefer New York). If the cost to the giver is a legitimate consideration, proximity has to be a legitimate consideration as well. Again, this is in Singer's terms, which assume that writing somebody a check reliably translates into benefit for him.

So his conclusion on that one looks very wrong to me. Onward:

...anyone who accepts certain assumptions, to be made explicit, will, I hope, accept my conclusion.

He makes other assumptions which are not made explicit: First, that the only way to help poor nations is to write them checks. Second, that writing checks to poor nations actually helps them. Third, that relatively wealthy nations can give as much as he wants us to without affecting how much we'll able to give next year, nor affecting how much poor nations will need next year: Is it true that if we buy a new car only when the old one is absolutely worn out, UAW members and General Motors executives will still be able to give just as much as they can now? Is it true that if we stop buying new clothes "not to keep ourselves warm but to look 'well-dressed'", the poor in China and Central America will be no more needy next year than they are now? If we're morally obligated to help, we're morally obligated to keep our economies healthy enough that we can help. We should be able to establish an optimal mix of consumption and giving in the first world, one which would provide for maximum ongoing cash payments to dependent nations (ensuring that they remain dependent, as all good "ethicists" demand). Regrettably, this optimum would probably permit us to buy new shirts more often than Singer would like. This further implies that if Dr. Singer is consuming below the optimum level, he'd best get his ass over to Banana Republic and do his part.

An argument in his favor would be that if we give a significant chunk of our money to poor countries, then they'll be buying the cars and polo shirts we've forgone (and they'd be justified in doing so because they don't have old but still-serviceable polo shirts to hang on to). At the risk of pretending to be an economist, do the economies and living standards of major aid recipients really bear that argument out?

I mention polo shirts only as an example of something which is costly, yet ideally worthless. I wouldn't wear one at gunpoint myself.

UPDATE 9/6/2005: About the near vs. far thing, consider it this way: It's probably instinctive for us to care more about those nearest to us, and you'll get more done by working with the current than against it. Consider the positive social value of self-interest in a market economy, and how destructive it turns in a command economy where corruption is the only way to get ahead (not that the value of liberty would cut much ice with Singer). Instincts aren't a product of "logic", so people like Singer consider them "unscientific", and conclude therefore that it's "scientific" to pretend they don't matter. If the problem is maximizing donations to charity, you take into account all the factors that affect how much people donate. Like all "idealists", Singer makes rules designed for machines, but demands compliance from people.