Professor Becker and I are both past the normal retirement age -- yet neither of us is retired. Both of us have, in fact, lifetime tenure, though both of us can, in principle, be forced to retire if we become incapacitated. Most Americans, and we suppose most people the world over, want to retire, generally by their early sixties. But there are dramatic exceptions, including Popes, Supreme Court Justices, and dictators.

The exceptions are becoming increasingly problematic with the extension of life brought about by modern medicine. If the only effect of medicine were to postpone incapacity, so that instead of becoming incapable of work at 70 one became incapable of work at 80, increased longevity would not pose a social problem. The problem arises from the fact that medicine increases the average period in which a person remains alive despite physical or mental incapacity. Medicine does not merely increase length of life; it also increases length of sick life.

We should distinguish physical from mental incapacity -- but not too sharply. Physical incapacity often reduces the amount of a person's working time, and if the person has a demanding job, the effect on his or her output can be great. If he or she is a key person -- such as a leader, or the occupant of a chair in a university department that has no additional slots -- the reduction in his or her output can greatly reduce the output of an entire organization. And that organization could be an entire nation. Mental incapacity is more dramatic and often more harmful; it may or may not reduce the amount of time that the afflicted individual works, but it can reduce the value of his or her output to zero.

The problem would not be acute in a system of mandatory retirement at a fixed age, whether for judges, professors, or religious and political leaders, but the entire trend is against fixed retirement ages. This seems a mistake. It is premised on the correct observation that people decay at different ages, so that setting any age of mandatory retirement designed to make sure that people incapacitated by age are removed will remove some people who are not incapacitated. But that observation merely identifies one cost of mandatory retirement. The benefits must also be considered. They are in fact great, because there is a cost entailed with removing people on the basis of individualized determinations of incapacity. Not only are such determinations often difficult to make (unless the individual has deteriorated to an extremely low level), but there is a justified concern that they may be used selectively, to remove political enemies or professional rivals. This is also a risk with mandatory retirement that allows exceptions for preferred individuals -- hence the pressure to forbid such exceptions.

Here is my suggestion that would achieve the principal benefits of mandatory retirement without the principal costs: beginning at age 70, require every life-tenured professor and every life-tenured judge to take a test of mental acuity every five years. The test results would be available only to members of the professor’s department or the judge’s court. The results would not be a basis for a determination of incapacity, nor admissible in a competence hearing. The expectation rather is that a poor test result would persuade the individual (perhaps by persuading his or her colleagues or family who would in turn persuade the individual) to retire voluntarily.

The design of such a test would of course be difficult. It could not be just an IQ test, because the abilities that enable a person to score high on such a test are not necessarily essential to competent performance in a given job, even an intellectual job. Moreover, older people can to an extent substitute experience for mental agility as a means of solving problems, and an IQ test is not a test of experience -- it is supposed to abstract from experience. In addition, experience is not only knowledge, but know-how, a trained intuition, qualities that may not be easily articulated and therefore may be difficult to test for. But not impossible. A professor, or a judge, could be given a problem in his or her field and asked for the solution. For a judge, the question might be: what is the correct decision in the following case? For the English professor: what does this poem mean?

One model could be the driving test that some states now require of elderly persons. Of course, it is easier to test for basic driving skills than for the skills required in high professional positions, especially for the skills implied by such expressions as the "wisdom of experience." Still, it should be possible to devise such a test as a collaborative project of experts in the field, psychologists, and educators. Few judges or professors would flunk. Rather, judges or professors who feared they might flunk would retire rather than take the test. The mere existence of the test would be a powerful signal to elderly people that, whatever their distinction, and even if they retain much of their acumen, they should be careful not to overstay their welcome.

While it would have been a shame had Justice Holmes been forced to retire from the Supreme Court at the age of 80, such examples are greatly outnumbered by cases in which an elderly person delayed retirement too long. The occasional false positive (pushing a person to retire when he or she is still competent) will rarely have any social significance because there will almost always be an equally competent younger candidate ready to step in. I intend no disrespect in expressing doubt that the Supreme Court Justices are the nine most competent confirmable candidates for the nation’s highest court, in a nation with hundreds or even thousands of highly competent judges. As for professors, retirement only stops their teaching, not their academic research or writing.

Finally, I do not suggest that the test requirement I have proposed be imposed by law on universities and judiciaries. The proposal is addressed to the institutions themselves, which could impose such a requirement by rule. Sanctions might not even be required for the rule to be obeyed. As long as some professors and judges took the test, those who did not might be suspected of fearing that they could not pass. Such suspicion would induce many of those individuals to either take the test or retire.