« A couple of links relevant to my principles class | Main | Social safety nets, opportunity cost, and the importance of understanding economics (Part II) »
January 25, 2006
Social safety nets, opportunity cost, and the importance of understanding economics (Part I)
It all started with a piece by Steven E. Landsburg in Slate which asked the question: "Do the Poor Deserve Life Support?" The subtitle was, "A woman who couldn't pay her bills is unplugged from her ventilator and dies. Is this wrong?" Landsburg argues that it is not.
This brought forth a response from Robert Frank who wrote in the New York Times,
Many commentators have attacked his argument as morally preposterous. Well, yes. But it is also economically preposterous.
Bloggers Arnold Kling and Brad DeLong have also noted the exchange, and it is indeed noteworthy. So if you haven't read the Landsburg and Frank articles. Do so now, and come on back.
So here we have two top-notch, respected economists who find themselves completely at odds on this question. Both use economic logic and reasoning. The average person on the street might find each of them quite persuasive individually. Taken together, how do you make sense of them when they are so clearly opposed?
Let's dig in and see the source of the disagreement. Landsburg's main arugment is that the person in question probably would not have placed a high value on hypothetical "ventilator insurance" ex ante and therefore should not receive the benefit ex post. Landsburg writes,
...for the same cost, we could give each of those people a choice between ventilator insurance on the one hand or $75 cash on the other hand. If it turns out that I'm wrong and they all want the ventilator insurance, so be it. But why not at least ask them?
You can't do that with every government service. You can't offer people a choice between police protection and its cash value, because police patrols tend to protect entire neighborhoods at once, not just specific individuals. You might not want to offer people a choice between a flu vaccine and its cash value, because you'd really prefer to have vaccinated neighbors. But critical life support isn't like that; the benefits are targeted to specific individuals. There's no reason those individuals shouldn't be allowed to choose different benefits if they want them.
And Frank writes,
Even those who are not poor recognize that catastrophe is only one unlucky break away. One might lose one's job and be unable to afford health insurance, for example, or be stranded by a mountain blizzard and unable to afford a helicopter rescue. With such prospects in mind, most people favor collectively financed rescue efforts. That a poor person would not, or could not, buy private insurance against such contingencies is entirely beside the point.
Frank compares the life support situation to a helicopter rescue. Landsburg draws a stark contrast between life support and government provided services like police protection. Landsburg is right that you can't let people opt out of police protection. There is a free rider problem. If the rest of the neighborhood pays I still get the protection, so it would be individually rational to want to opt out. So the government provides the protection and requires payment in the form of taxes.
The way most cities are set up, the same is true of fire departments, but it wasn't always so. Long ago, insurance companies formed fire departments for the purpose of protecting their interests. Each building had a plaque indicating which company provided them with protection from fire, as well as insurance against loss. Presumably, people could choose their fire company on the basis of location, price, etc. I would guess that people could choose to have no insurance at all. In that case, I suppose that the companies who insured buildings on either side would work to prevent the fire from spreading but would not necessarily concern themselves with the uninsured building.
Today, our fire departments are usually agencies of city or county government, making them functionally similar to police departments. They respond to all calls for help, without regard to your insurance company, or even whether you pay taxes to that juristiction (as in the case of responding to an auto accident or medical emergency). Through many years of voting and policymaking, we have decided that this is a good system--just as most people would agree that public police protection is superior to a patchwork system of private security guards.
And yes, those public servants are the ones who rescue us from mountain blizzards, mine explosions, terrorist attacks, and a million other things for which writing individual insurance policies would be difficult if not impossible. The question is whether life support, of the type required in the article, is one of those situations.
Landsburg says that the benefit of life support is targeted at a specific individual. Well, so is the mountain rescue operation, at least once you have an "identified life" at stake (to use a term employed by Frank). So that isn't the real difference. The real difference is much more fundamental. The life support in this problem is not for the purpose of saving a life, but for the purpose of prolonging it enough to grant a dying wish.
One can try to measure the economic benefit of saving a life. One might even come up with a "willingness to pay" ex ante to save a life from mortal peril. And obviously society puts a high enough price on this as to employ rescue squads, search and rescue teams, and so on. But how do you measure the willingness to pay to grant a dying wish to say good-bye to a family member? Should the poor be priced out of this insurance market?
More to the point, would society be better off in a world where granting such consideration (within certain bounds of reason) was a social norm? Should taxpayers reimburse the hospital for providing a few more days of care in this case?
Unfortunately, this question is neither as objective, nor as simple, as asking whether society benefits from police protection or a national defense. Well meaning people can disagree. Answering in the affirmative does not come without cost for society. And yes, as Landsburg points out, it would mean that there would be fewer resources available to help the poor in other situations, perhaps more life-threatening situations. That is true. But we should also consider the social benefits of such a norm. Frank appeals to Adam Smith in arguing for a consideration of sympathy and empathy. The Theory of Moral Sentiments is the place to look for such discussions.
How much would you be willing to pay for a social norm such as this?
To be continued...
Posted by William Polley at January 25, 2006 1:51 PM
Trackback Pings
TrackBack URL for this entry:
http://www.williampolley.com/cgi-bin/mt-tb.cgi/466