|
|||||||||||||||||||||||||||
|
Kevin Drum finds Steven Landsburg's defense of killing patients who can't afford medical care "morally cankered." (I'll have to remember that phrase.) That surely won't bother Landsburg, who seems to glory in his persona of economist-as-Mister-Spock.
And Landsburg's general point -- that when non-poor people try to help poor people, they should give the poor people what the poor people actually want, not whatever the rich people would feel best about providing for them -- is substantially correct. Expenditures that are prudent when you have lots of money can be imprudent when you don't, and we shouldn't pretend otherwise just to make ourselves feel good. (This sort of question comes up all the time in the context of safety regulation.)
But it should bother Landsburg that his analysis is plainly wrong on the level of technical economics, because it ignores the difference between "identified lives" and "statistical lives," which can also be thought of as the difference between small risks and big risks.
Here's the intuition: someone who would spend $10 to avoid a one-in-a-million risk of sudden death (which is about normal for a middle-class American) would rationally also accept $10 in return for facing such a risk. But that same person wouldn't accept $10 million in return for certain death, because if he died he wouldn't be around to enjoy the money. By the same token, he wouldn't accept $5 million to risk his life with probability one-half; not only does the risk of not being able to spend it make it less valuable, but the principle of diminishing marginal utility means that $5 million isn't worth 500,000 times as much as $10. So his "small-risk value" of avoiding death is much smaller, per unit of risk, than his "large-risk value."
(The Ron Howard paper that makes this point doesn't seem to be available on line; instead, I've posted a prosified set of my lecture notes on the topic.)
Thus Landsburg's demonstration that the Texas woman whose respirator was just shut off because she couldn't pay her medical bill probably would have chosen, back when she was healthy, some other benefit with the same cost over an insurance policy to pay for a ventilator if she needed it doesn't prove what he thinks it proves: that by letting the poor woman (dying of cancer, but conscious) die of hypoxia first the good people at Baylor Hospital were just respecting her own wishes. Of course, if Landsburg's moral intuitions hadn't been damaged by associating with his fellow market-worshippers, he would have noticed that his economic conclusion didn't feel right and gone looking for the technical error he could have easily found in the literature.
In other domains, our willingness to spend public money to save an identified life -- some particular idiot who's gotten snowbound climbing a mountain or wrecked his motorboat while drunk -- is all out of proportion to our willingness to spend public money on things like highway median barriers. That seems "irrational" in economic terms unless we notice that the identified lives are people facing large risks, not small ones. At that point, if we refuse to help because it costs "too much" money, we're not respecting their choices. We're just expressing our judgment that their lives aren't worth much.
Of course, Landsburg is free to believe that. And he's free to believe that keeping someone on ventilator support until the cancer kills her isn't really doing that person a favor.
The rest of us, by the same token, are free to believe that in the richest country in the history of the world no one who wants to keep breathing should be forced to stop breathing for lack of money.
Footnote Predictably, none of the busybodies who made a national crisis out of the Terri Schiavo case seem to care about the people being killed for their poverty under the Texas "futile care" rules, signed into law by Gov. George W. Bush. A Google search for the name "Tirhas Habtegiris" finds only the Left Coaster and YucatanMan at Daily Kos.