Why States Should Ban Adolescent Driving (cont’d)

My previous post argues for raising the driving age, ideally to 18. Thank you to those who commented. You raise many good points, so I continue the conversation here. (And I must especially thank Prof. Cunningham for reading my article, and for his kind words.)

Treating Teens as Presumptively Delinquent? One comment suggests that raising the driving age treats young people as “presumptive law breakers.” I resist that characterization. Raising the driving age recognizes adolescent incapacity in this context and aims to protect adolescents, and those with whom they share the roadways, from the too-often tragic consequences of that incapacity. Such policy reform is consistent with the state’s obligations to its young people. I briefly discuss those general obligations, then return to the driving context specifically.

The State’s Obligations to Adolescents, Generally. The state should safeguard both the welfare interests and autonomy interests of the young. (I elaborate on this state obligation elsewhere, drawing on the work of my colleague James Dwyer and others.) Welfare interests pertain to young people’s well-being, irrespective of any affirmative choice they make. Autonomy interests refer to their interests in making self-determining choices and having the freedom to exercise the liberties of which they are capable. Compelling evidence suggests that driving is a liberty that adolescents do not have the capacity to exercise competently. The state thus fails to guard adolescents’ welfare interests — and protect them from their deficiencies — by extending them this liberty despite their incapacity.

Policymaking affecting adolescents in general poses a major challenge for lawmakers. Young people attain different capacities at different stages in their development, and development correlates predictably (though not perfectly) with age. Identifying and extending to adolescents liberties in contexts in which they have attained competence can be a challenge. Further complicating the state’s task is that, even where adolescents may have achieved the ability to perform competently, certain real-world contexts predictably confound their capacities and impede their performance. Thus by mid-adolescence, individuals have reached adult-like information-processing and logical reasoning abilities. But the quality of their decision making suffers in situations that require adolescents to quickly assess and react to risk, to reason while highly stressed or in the heat of passion, to make decisions in unfamiliar circumstances, or to act in the presence/under the pressure of peers. The neurobiological processes that support decision making under these conditions do not fully mature until late adolescence or early adulthood.

Prof. Cunningham’s analogy to rules that allow minors to disclaim contracts is a good example of a policy choice that may be at odds with what we now know about adolescent decision-making capacity. (I discuss adolescent contractual capacity briefly in another article (at pp. 1851-57, which argues against adolescent marriage.) By mid-adolescence, individuals have the cognitive capacity to understand the rights, duties, and responsibilities of a contract, and in light of that understanding, are able to make a voluntary choice to enter it or not. Absent the same factors that would invalidate a contract entered by an adult (duress, etc.), there is a strong argument that the adolescent should be held to his or her bargain. But contract policy might also permit minors to disclaim contracts entered in the sorts of contexts likely to compromise their decision making (e.g., the typically pressured context of buying a used car?).

Back to adolescent driving:

How Serious a Public Health Threat? One commenter (SgtDad) notes that traffic fatalities have declined in recent decades, making adolescent driving an “ever smaller problem,” with policy changes in turn having an “ever smaller effect.” To what extent does adolescent driving remain a true public health problem? An estimated 48 thousand 16- to 19-year-olds will die in car crashes

between 2003 and 2012, and well over 2 million more will suffer crash-related injuries. And that is only half the story — literally. Because in crashes leading to fatalities, twice as many non-drivers as drivers (i.e., passengers or individuals traveling in other vehicles) lose their lives; a majority of all drivers survive crashes in which there are fatalities.

The crash-related death rate for 15- to 19-year-olds peaked in 1970 at 43.6 per million miles driven, then declined significantly (for the reasons noted by SgtDad — seatbelt laws, improved vehicle safety, etc.), to 33.1 by 1990. Graduated licensing laws, which delayed licensure and increased driving practice, help explain post-1990 declines. Crash-related death rates thus declined to 26.0 by 2000. But declines have since slowed or plateaued. Finally, consider this (at fn 1): in 2009, 3,487 13- to 19-year-olds died in car crashes; 2,027 died as a result of homicides. Although only a fraction of those homicides occur in schools, school systems, and state and the federal government more broadly, are currently devoting significant resources to preventing these tragic deaths. Adolescent car crashes and traffic fatalities may not be national news today, but their prevention too merits public attention and resources.

Gender. The young driver problem is a gendered one. About 2 our of every 3 teens killed in car crashes in 2011 were males. Males in general have higher sensation-seeking tendencies than do females, and some researchers have suggested that the higher levels of testosterone present in male midteens increase aggressive behavior generally, including aggressive and risky driving. Other researchers argue that socialization and conformity with gender norms are more significant factors. In other words, they argue that risky driving is a socially-created norm associated with masculinity, but it is one that can — and should — be replaced with more constructive norms. [For a discussion and studies, see here, pp. 21-23.]

Intermediate Measures/Tailored Regulations. Prof. Cunningham suggests that different rules might appropriately apply to different driving conditions. Indeed, less than a decade after the first Uniform Vehicle Code recommended age 16 as the minimum age for licensure in 1926, the Natl. Conf. for Street Highway Safety amended the Code to make just the sort of recommendation Cunningham suggests. It proposed an approach to age of licensure in which states would take account of state-specific factors, such as the types of hazards present on the state’s roadways, in-state crash records of minors, and other conditions particular to a state. From the outset, rural and farm states tended to adopt lower minimum licensing ages. Other states adopted or retained age 16. But the amended Code’s recommendation to tailor state age-related licensure requirements to in-state conditions went ignored.

Yet the driving conditions encountered by — and crash characteristics of — young drivers vary. For example, young drivers living outside of cities are less likely to crash overall compared to young urban drivers, but they are at greater risk of single-vehicle crashes, which tend to cause more severe injuries than to multiple-vehicle crashes.

Mobility needs vary as well, yet there is scant empirical work examining the necessity of adolescent licensure. One study (fn. 308) found that 1 in 5 parents reported that safety concerns had led them to attempt (usually without success) to delay their children’s obtaining a learner’s permit. Without question, delaying adolescent licensure will sacrifice some mobility and convenience. At the same time, researchers found (fn. 314) that in New Jersey, the only state with a licensing age above 16, “licensure at age 17 has little effect on the lifestyle or employment of New Jersey 16-year-olds while producing a substantial reduction in their crash involvement.” Where exactly to strike the balance between safety and mobility is a social policy decision, and ideally requires additional empirical studies of the necessity of adolescent mobility.

Prof. Banzhaf suggests that electronically reducing the maximum speeds at which cars driven by adolescents may travel may reduce the number and severity of car crashes. Along the same lines, a few states restrict novice drivers from driving on highways. These types of measures are likely to reduce fatalities, as young drivers are generally more likely than older drivers to exceed posted speed limits or speeds appropriate to driving conditions (e.g. in foul weather or on wet roadways). Speeding is only one of the causes of adolescent crash risk, however.  Young drivers’ crashes also result, not from intentional risk-taking behaviors, but also from failure to employ routine safe driving practices, such as failing to detect another vehicle or traffic control, or generally losing control of their vehicle.

The Elderly Driver. My research has focused on policy making affecting adolescents. I’ve not studied elderly driving but can report the following: In 2008, the rate of 70- to 79-year-old drivers involved in car crashes in which there was any fatality was 24 per million miles driven. The rate of 16- to 19-year-olds was nearly twice as high — 46. (The above paragraph “Public Health” reports a different statistic — crash-related death rates. The “Elderly Driver” statistics report driver crash involvement, but not necessarily driver death) Drivers aged 70 to 79 had lower fatal crash rate involvement than both 20- to 24-year-olds and 25- to 29-year-olds (whose fatal crash rates were 38 and 36, respectively). Drivers aged 30 to 69 did have lower fatal crash rates — 12. The only drivers whose fatal crash rate — 55 — was higher than those of teen drivers were drivers aged 80 and older.

I offer only this thought: After having extended a category of individuals a certain liberty based (in the case of driving) on both their presumed and demonstrated capacity, the state should be loath to withdraw that liberty absent individual wrongdoing or an individualized assessment of incapacity. I made an argument along these lines in a recent article and series of posts on this site on the voting age (arguing for lowering the voting age to 16, and distinguishing the voting rights of adolescents from those of elderly citizens). Younger adolescents lack, and eagerly anticipate, the freedom and independence that attends licensure. But the young person’s anticipation of acquiring that liberty must be an altogether different thing from the older person’s loss of it. I do not know if data supports a categorical finding of driving incapacity after a certain age, but I would hope that states would require such data to be compelling before they rely on it to justify categorical de-licensure.

Self-Driving Cars? One comment mentions their emergence.  I admit that I have no idea. But if the price of non-crashing, widely-available self-driving cars is that my research becomes irrelevant, I’ll happily pay up.


image: http://www.sxc.hu/browse.phtml?f=download&id=41224

You may also like...

3 Responses

  1. nidefatt says:

    You haven’t even come close to proving cause and effect here. The salient point is that millions of teenagers drive and no one dies. You would punish an infinite number of young people on the basis of frankly unimpressive statistics.

  2. PrometheeFeu says:


    My comment regarding treating teens as presumptive delinquents was aimed at John Anzhaf. I understood you to mean that teens are bad at making decisions under certain conditions and that therefore, we should reduce their exposure to such conditions. But bad decisions are not necessarily illegal decisions.

    John Anzhaf was arguing for high-speed cut-offs at or under speed-limits effectively saying that unless you physically prevent teens from speeding, they will speed. In other words, he was arguing that teens should be presumed to be law-breakers. (specifically breaking the speed limit)

    The comment I aimed at your post was mostly that I don’t understand how you justify the age of 18. Yes, teenagers and others are put at a higher risk because people under the age of 18 can drive. But this is on a continuum. 17 is safer than 16 and 18 is safer than 17 and 19 is safer than 18. The scale is logarithmic, but I didn’t see you make a cost-benefit analysis to explain why 18 instead of 15, 16, 17, 19 or whatever.

  3. Ron Miller says:

    I picked this up from Overlawyered…

    One theory from the comments there is the idea that 18 year old drivers with no experience is the same as a 16 year old with no experience I expect that the inexperienced 18 year old is a better driver than the inexperienced 16 year old driver. If I wrong about this, I would question my premise that something has to be done with 16 year old drivers.

    One difficult issue is how embedded driving at 16 is in our culture.