UNITED STATES of America, Plaintiff-Appellee, v. Louis E. JOHNSON, Defendant-Appellant, 685 F.3d 660

POSNER, Circuit Judge. The defendant, a former bailiff and former nightclub owner, pleaded guilty to possession of a firearm by an unlawful user of a controlled substance, 18 U.S.C. § 922(g)(3), and possession of an unregistered firearm. 26 U.S.C. § 5861(d). The judge sentenced him to 78 months’ imprisonment, the bottom of the applicable guidelines range, which was 78 to 97 months. The defendant’s lawyer has filed an Anders brief, seeking leave to withdraw on the ground that he can’t find a colorable ground for an appeal.

The only possible such ground is the judge’s decision not to give a below-guidelines sentence despite the defendant’s age, a question discussed at length at the sentencing hearing, where his lawyer argued that the defendant should get a shorter sentence than 78 months (six and a half years) because he is (or rather was, at sentencing) 70 years old (he is now 71) and so might die before he was released from prison. The judge consulted the Census Bureau’s life-expectancy table and found that the life expectancy of a black male aged 70 is 12.4 years.

So even without any time off for good behavior, which would reduce his time served by a maximum of 10 months and thus to 5 years and 8 months, the defendant’s sentence does not exceed his life expectancy. Which raises two questions: the bearing of old age on sentencing, and the bearing of life expectancy on sentencing. The propensity to engage in criminal activity declines with age, and is, on average, sharply lower for persons over 70— although persons 65 and older are 13 percent of the population, they account for only seven tenths of one percent of arrests.

FBI, “Crime in the United States: Arrests by Age, 2010,” www.fbi.gov/about-us/cjis/ ucr/crime-in-the-u.s/2010/crime-in-the-u.s.2010/tables/10tbl38.xls; U.S. Census Bureau, “The Older Population: 2010” 2 (Nov. 2011), www.census.gov/prod/cen2010/ briefs/c2010br-09.pdf (both visited on June 1, 2012); see also United States v. Bullion, 466 F.3d 574, 577 (7th Cir.2006); Catherine F. Lewis et al., “A Study of Geriatric Forensic Evaluees: Who Are the Violent Elderly,” 34 J. Am. Academy of Psychiatry & Law 324, 324 (2006); Richard A. Posner, Aging and Old Age 128-33, 310-15 (1995).

There is both an aging effect and a selection effect: the cost of acquiring criminal skills increases with age, and career criminals, who already possess such skills, are likely to retire from crime before reaching old age because repeated crimes bring increasingly heavy sentences. Id. at 132-33. Persons convicted of a crime committed when they are 70 or older are thus unlikely to commit further crimes even if released after a short term of imprisonment. Cf. U.S.

Sentencing Commission, “Measuring Recidivism: The Criminal History Computation of the Federal Sentencing Guidelines” 12, 28 (May 2004), www.ussc.gov/Research/Research_ Publications/Recidivism/200405_ Recidivism_CriminaLHistory.pdf (visited June 1, 2012). And so the Sentencing Guidelines state that “age may be a reason to depart downward in a case in which the defendant is elderly and infirm and where a form of punishment such as home confinement might be equally efficient as and less costly than incarceration.” U.S.S.G. § 5H1.1 (Age Policy Statement); see also United States v. Bullion, supra, 466 F.3d at 576.

But this general rule must have exceptions. The 70-year-old criminal is a rara avis, and by engaging in criminal activity at such an age provides evidence that he may be one of the few oldsters who will continue to engage in criminal activity until they drop. This may well be the case of our defendant, as the district judge explained in a thoughtful sentencing statement. The defendant is devoted to guns and drugs. He hosts parties in his home at which he and his guests consume crack cocaine and marijuana, and he not only buys guns but admits to sometimes selling them in his crime-ridden neighborhood.

There is the ominous fact that because of frailty the elderly are more prone than young criminals to use guns in crime, rather than less lethal weapons, Lewis et ah, supra, at 330, though our defendant does not appear to be a violent criminal — moreover he is not frail! He is 6'7" tall and weighs 230 pounds, and is in good health (or at least was when sentenced less than a year ago). He is physically capable of continuing indefinitely to engage in the illegal activities for which he was convicted and sentenced.

And the likelihood of a criminal’s committing further crimes when released from prison is of course not the only consideration that a judge should weigh in deciding how long a sentence to impose. Incapacitating a defendant — -preventing him from committing crimes by keeping him in prison for a prescribed period — is only one way of reducing the incidence of crime. Another is deterrence. The threat of imprisonment is a deterrent, and the threat of a longer imprisonment should have a greater deterrent effect than the threat of a shorter one.

Suppose there were a rule that a person who commits a crime after his seventieth birthday can be sentenced to no more than six months in prison, lest he die there. Then those oldsters who like the defendant in our case do not terminate their criminal careers upon reaching that milestone will have only a weak disincentive to commit further crime — especially if the probability of apprehension of and conviction for the crimes in which he habitually engages is low. For the expected cost of punishment is a function of the likelihood of being punished as well as of the severity of the punishment if imposed.

This discussion should make clear that the sentence imposed by the district judge was reasonable and that the defendant’s current lawyer was quite right to disclaim the existence of a colorable ground for challenging the appeal. But what of life expectancy, and specifically of tables of life expectancy? What role should they play in sentencing? Life-expectancy tables lump together large numbers of people who have only a few things in common, such as, in this case, age, race, and sex. The set of black men aged 70 to 74, estimated by the Census Bureau to have included 345,000 men in 2010, U.S.

Census Bureau, “Table 15. Projections of the Black Alone Population by Age and Sex for the United States: 2010 to 2050,” www.census.gov/population/ www/projections/files/nation/summary/np 2008-tl5.xls (visited June 1, 2012), actually contains a range of life expectancies. Some of these men will die within the year; some will live to 100 or even a few years beyond that; the rest will die in between the extremes.

All it means to say that the defendant when he was 70 had a life expectancy of 12.4 years is that the average person in the heterogeneous group to which he belongs (black 70-year-old men) can be expected to live 12.4 more years, implying that approximately half will die sooner and the other half later. (It would be exactly half if 12.4 were the median rather than average age of death of members of the group.) It doesn’t mean that the defendant will not die in prison; he very well may; the probability that he will die before he is released can be calculated, either from statistics concerning the experience of his group, or, with greater accuracy, from more refined statistics that would narrow the group to black 70-year-olds whose physical condition is similar to the defendant’s.

But even the most refined statistical calculation of his life expectancy will leave considerable residual uncertainty. We have wrestled in previous cases with the question whether life expectancy statistics should figure in sentencing for offenses for which Congress has not authorized a life sentence.

Our court has concluded, as have other courts, that a sentence which although it is a term of years is likely or even certain to be a de facto life sentence because of the defendant’s age is improper if the statute under which he was convicted provides that only a jury can authorize a life sentence (18 U.S.C. § 34, applicable to certain drug offenses). United States v. Martin, 63 F.3d 1422, 1432-34 (7th Cir.1995), abrogated on other grounds by Jones v. United States, 529 U.S. 848, 850-51, 120 S.Ct. 1904, 146 L.Ed.2d 902 (2000); United States v. Martin, 115 F.3d 454 (7th Cir.1997); United States v.

Martin, 100 F.3d 46, 46, 48 (7th Cir.1996); United States v. Prevatte, 66 F.3d 840, 846-49 (7th Cir.1995) (concurring opinion); United States v. Tocco, 135 F.3d 116, 131-32 (2d Cir.1998); United States v. Gullett, 75 F.3d 941, 950-51 (4th Cir.1996); United States v. Williams, 775 F.2d 1295, 1299 (5th Cir.1985); United States v. Hansen, 755 F.2d 629, 631 (8th Cir.1985). This is not such a case. The defendant’s age and physical condition do not make his sentence a de facto life sentence.

And if it did, it would just be one more consideration that the judge might be asked to weigh in determining the sentence, properly so if the prospect of dying in prison is thought to make a sentence of otherwise appropriate length harsher. The motion to withdraw is granted and the appeal is Dismissed.