Like the minimum wage and rent control, the market for human organs is a classic topic when teaching the basics of supply and demand. Organ markets are largely outlawed and, as a result, the demand for organs greatly outstrips the supply. For example, according to some estimates, as many as 4,000 people in the United States die each year while waiting for donor kidneys (some of which could, in principle, come from healthy donors).
As Dick Thaler notes in the New York Times today, the usual economist solution to this problem – allowing the buying and selling of human organs – is a political non-starter. Many people find the idea “repugnant,” as economist Alvin Roth has put it.
One solution, which Roth helped pioneer, is to create organ swaps rather than sales. Suppose, for example, that my wife needs a kidney and that I am willing to donate, but am not a match. And at the same time, a woman wants to donate a kidney to her sick brother, but also isn’t a match. That seems like a dead end (so to speak), but if I am a match for her brother, and she is a match for my wife, then we can arrange a swap – my kidney for hers. Two lives get saved, and there’s nothing repugnant about it.
Over time, this basic idea has expanded to include “daisy chains” of donations involving numerous donors and recipients (for a nice description see this recent article in Wall Street Journal).
Thaler considers another way to address the problem of organ supply (from individuals who become brain dead, not those who are healthy) using the insights of behavioral economics:
One strategy is to alter the default rules for signup. Most states, as well as many other countries, use an “opt in” or “explicit consent” rule, meaning that people must take a concrete action, like going to a public library or requesting and mailing in a form, to declare they want to be donors. But many who are willing to donate organs never get around to such steps.
An alternative approach, used in several European countries, is an “opt out” rule, often called “presumed consent,” in which citizens are presumed to be consenting donors unless they act to register their unwillingness.
“Opt-out” can lead to much higher rate of organ donations than ”opt-in”:
Consider the difference in consent rates between two similar countries, Austria and Germany. In Germany, which uses an opt-in system, only 12 percent give their consent; in Austria, which uses opt-out, nearly everyone (99 percent) does.
Of course, the idea of presumed consent is not without controversy. For many people, it seems disturbingly presumptuous for the government to assert some sort of property right to their organs.
In response to that concern, Thaler recommends the middle road of ”mandated choice” in which individuals would periodically be required to decide whether or not they want to be donors. In Maryland, where I live, that happens when you go to renew your drivers license.
With mandated choice, the government needn’t presume anything about your willingness to donate. By forcing you to declare a choice, however, it reduces the problem of aspiring good samaritans who never get around to signing up.
All of which leads me to the following question: can behavioral economics help policymakers think about the idea of insurance mandates and expanding health insurance coverage? My two recent posts about an individual mandate (here and here) have inspired a host of very thoughtful comments, many of which raise concerns (both moral and constitutional) about the idea of the federal government forcing Americans to buy insurance that they may not want.
My question (for which I don’t have an answer) is whether there is some way to adopt “opt-out” policy designs to increase insurance coverage without (overly) violating reasonable beliefs about the appropriate limits of government power. Suppose, for example, that there was some way to default uninsured individuals into relatively inexpensive insurance plans providing catastrophic coverage. The individuals would be allowed to drop out, but at least they would be initially enrolled. Would that be both effective and acceptable? (One concern would be the need for individuals to reveal their insured / uninsured status to the government, e.g., on their tax forms, which I know some of my readers would find troubling.)