Show Summary Details

Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE ( © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

date: 03 June 2020

Considerations for Policymaking Affecting Adolescents in the Liberal Democracy

Abstract and Keywords

Adolescents are developmentally distinct from both children and fully mature adults in ways that are relevant to policymaking. This chapter discusses some of those developmental differences and shows how the decision-making context influences adolescent competence in different settings. Thus, depending on the nature of the decision and the context in which decision-making will occur, policies appropriately extend to adolescents the ability to exercise different rights at different ages. This chapter focuses on three areas of non-criminal regulation of adolescent life: voting, driving, and the purchase/consumption of alcohol.

Keywords: adolescence, driving age, age of consent, human development, autonomy, voting age, underage drinking, cognitive capacities

Adolescents are developmentally distinct from both children and fully mature adults in ways that are relevant to a range of policymaking contexts. This chapter discusses some of those differences, and how lawmakers might account for them in crafting policies affecting adolescents.

Infants and children, of course, require care and lack the capacity for autonomous decision-making. Accordingly, laws affecting them generally aspire to safeguard their welfare interests, or well-being. Exemplifying this aspiration is the best-interests-of-the-child standard, the lodestar that guides judicial decision-making involving the custody of children.1 By prioritizing children’s welfare, the state meets what I have argued are its dual obligations towards its immature citizens—safeguarding their well-being during the period of dependence, and creating conditions during the period of immaturity to ensure that individuals reach maturity able to make autonomous life-determining choices. To advance the first obligation, the state endeavors to protect the young during the period of dependency either directly (by exercising parens patriae power) or indirectly (by delegating to parents the authority to make decisions). To advance the second, states make mandatory a range of education- and health-related measures meant to encourage young persons’ development into capable citizens and self-determining adults.2

Policymaking can become more fraught as young people age, however. As they mature, young people develop decision-making capabilities and an interest in exercising those liberties of which they are capable. Because individual liberty is a core value of (p. 758) the liberal democratic state, policies ideally respect and facilitate individuals’ exercising the autonomy of which they are capable.

Advances in the developmental sciences over the last two decades have improved our understanding of developmental processes, and their implications for adolescents’ decision-making abilities. We know that by the time they reach adolescence, young people will have reliably attained certain decision-making competencies, while they will still be developing others. It is important that policymakers endeavor to understand the ages by which young people typically attain different capacities, and to account for both their capacities and ongoing deficiencies when crafting policies.

The criminal context is perhaps the most well-known area of law where rules recognize development-related differences between young people and adults.3 The juvenile justice system aims to shield young offenders from the harshest of the penalties it imposes on adult offenders.4 The justification for this approach is the developmental immaturity of young offenders—because of development-related characteristics that impair good decision-making in certain contexts, juveniles are presumptively less blameworthy and more amenable to rehabilitation than adult offenders.

Apart from the criminal context there remain, however, multiple areas of social life where rules extend or withhold rights according to a person’s age. This chapter focuses on some of those areas of non-criminal regulation of adolescent life that have received insufficient academic attention.

The next part of this chapter describes some of the characteristics that differentiate adolescents from both children and adults in ways that are relevant to policymaking. The chapter then examines several decision-making contexts by way of illustration, as well as how the law might better address them: the franchise, driving, and alcohol consumption.

1. Cognition and Decision-making During Adolescence

Adolescence is the developmental period that follows childhood. The onset of puberty at the end of childhood marks the beginning of the adolescent period, while its end point is marked by no comparably significant biological event. It is thus not a period defined strictly by chronological age, but instead spans the entire second decade of life from approximately ages ten or eleven to twenty-one. Some researchers divide adolescence further, into three stages—early adolescence (roughly ages ten to thirteen); mid-adolescence (roughly ages fourteen to seventeen); and late adolescence (roughly ages eighteen to twenty-one).

Researchers have begun focusing on adolescence as a distinct period because the brain undergoes radical changes during this period of development. For lawmakers, aspects of brain development will be relevant to the course of young people’s (p. 759) attainment of various capabilities. For that reason, the developmental category is also relevant from a policymaking perspective.

The relatively young field of developmental cognitive neuroscience has provided tremendous insight into adolescence. Developmental cognitive neuroscience is an interdisciplinary scientific field that emerged following a 1989 Philadelphia conference that brought together developmental psychologists, cognitive scientists, and neuroscientists.5 Those working in the field study brain development from the prenatal period through adulthood, focusing not on the individual patient but instead on typical development to adulthood. Their aim is to understand both psychological processes (cognitive, social, and emotional development) and their neurological underpinnings. What happens in the typical brain during development can influence cognition and cognitive processing, and in turn help explain observed behaviors.

The parts of the brain that control movement, vision, and the senses have largely matured by the end of childhood. Basic cognitive abilities improve more or less linearly throughout childhood, reaching mature levels by mid-adolescence. These cognitive capacities include, for example, the abilities to understand and logically reason from facts, to process information, and to assess the nature of a given situation.

Research has established that by age fifteen or so, adolescents perform as well as adults on tasks measuring logical reasoning, information processing, and risk perception. Generally, the reasoning and basic information-processing capacities of the typical sixteen-year-old are “essentially indistinguishable” from those of adults.6

At the same time, typical adolescent behavior includes increased propensities for risk-taking, sensation-seeking, and impulsivity—all behaviors that seem to be at odds with adultlike cognitive abilities. Researchers have identified ongoing activity in two neural systems in the brain that helps to explain the apparent inconsistency.

In mid-adolescence, activity in a neural system of the brain known as the socioemotional system leads to a peak in reward sensitivity.7 As a result, adolescents tend to experience rewards as more rewarding, increasing their motivation to pursue reward and increasing their affinity for exciting, novel, and risky activities. After peaking in mid-adolescence, reward sensitivity declines.8

The cognitive control or self-regulatory system, on the other hand, improves more or less linearly over the course of development. As this neural system develops, individuals’ ability to overcome impulsive or inappropriate responses increases, maturing in adulthood. Accordingly, the heightened risk-taking that characterizes adolescent behavior generally can be seen as the product of heightened reward-seeking and relatively weaker self-regulatory control.9

These insights into adolescent development and decision-making ought to be relevant to policymakers: Whether adolescents will be able to exercise adult-like decision-making will very much depend on the context in which they’re having to make those decisions. In ideal conditions (like research labs), their mature cognitive capacities enable competent, rational decisions. However, their capacities are susceptible to being confounded by the real-world contexts in which they are making decisions. At least partly due to the combination of heightened reward salience and weaker self- (p. 760) regulation, their decision-making suffers in contexts that are emotionally charged or pressured.10 They are thus more likely than adults to act rashly or make bad decisions in the heat of passion, under time pressure, or on the spur of the moment. The presence of peers also has been found to activate adolescents’ socioemotional systems, thus contributing to increased impulsiveness and risk-taking in those contexts.

Science alone ought not dictate policy, but our understanding about decision-making capacities at different ages and in different contexts can improve, and should inform, policymaking.

Legal rules must operate according to bright lines; there will be relatively few circumstances where individual assessments will be feasible, or even useful. It is nonetheless critical for policymakers to begin understanding adolescence as a developmental period, as well as the general nature of capacities/incapacities that accompany it.11

Thus when weighing policies involving adolescents, policymakers ought to include among their considerations the following:

  1. (1) The nature of the decision to be made and the knowledge/cognitive processes it requires;

  2. (2) The context in which the decision will be made: Is it one involving the sorts of pressure likely to compromise adolescent decision-making? Does it involve the presence of peers?

  3. (3) If an individual, by virtue of their immaturity, makes a bad decision, what are the costs? In other words, is it a purely self-regarding decision? Or are there externalities whereby individuals or society at large will bear the cost of a bad decision?

The remainder of this chapter will consider important policymaking contexts: the franchise, driving, and the drinking age.

2. The Franchise

Although the franchise is the core of participatory democracy, neither scholars nor lawmakers in the United States have focused much attention on voting age requirements—particularly, whether age eighteen is a good proxy for the attainment of electoral competence. Across the globe, on the other hand, a growing number of democracies have lowered their voting ages, or are considering doing so.12 Recently, activists in the United States have begun mobilizing support for similar measures.

In general, the states have broad power to establish voter-qualification standards, for both state and federal elections. At the time the Constitution was ratified, they all retained the English common-law voting age, twenty-one.13 After Congress lowered the age of conscription from twenty-one to eighteen during World War II, however, the voting age also began receiving attention. That young men might be drafted (p. 761) into the military yet ineligible to vote for their representatives to government troubled some lawmakers and prompted advocates to mobilize to lower the voting age as well to eighteen. Congress eventually acted, lowering the voting age to eighteen in 1970. The U.S. Supreme Court held, however, that Congress had authority to set the voting age for federal elections but lacked the legislative power to establish the voting age for state elections.14 Following the Court’s ruling, Congress approved a Constitutional amendment lowering the voting age to eighteen nationally, in order to avoid the logistical challenges of different voting ages applying for federal versus state elections. State legislatures moved rapidly to ratify the Twenty-Sixth Amendment in 1971, extending the franchise to eighteen-year-olds.

While the Twenty-Sixth Amendment prevents states from raising the voting age, nothing prevents the states from lowering it and enfranchising some additional cohort of their younger citizens. The nascent U.S. movement to lower the voting age to sixteen has gained some traction—three cities in Maryland have lowered the voting age to sixteen for local elections, for example, and Berkeley, California did so for school board elections. A 2016 ballot measure to lower the voting age to sixteen in San Francisco, California garnered national attention and significant support, though ultimately failed.15 Bills to lower the voting age are being introduced in a number of cities, and one-third of all states permit seventeen-year-olds to vote in primary elections if they will turn eighteen in time for the general election.16

Individual members of a democratic political community have a presumptive entitlement to political participation—typically, through the franchise. The exclusion of individuals from the franchise through voter qualification rules ought to occur only with legitimate reason. Political theorists have consistently identified two criteria for inclusion: (1) significant and ongoing connection to the community (typically reflected in citizenship and residency requirements); and (2) vote decision-making competence (typically reflected in age qualifications).17 The voting age operates as the primary voter qualification rule that aims to ensure that voters have attained the requisite intellectual independence and decision-making abilities required for electoral competence.

Young community residents share the same, if not greater, interests as their elders with respect to issues of public concern—public health, education, safety, etc. Indeed, they are more likely to bear the long-term consequences of public policy. They thus cannot legitimately be excluded from the franchise on grounds that they lack ongoing interest in and connection to their political communities. Instead, it is young people’s lack of the relevant competence that justifies their electoral exclusion.

While newborns and the very young lack electoral competence, the typical person acquires it at some point during the course of development. And because age and cognitive development correlate, there is a temporal element to the attainment of electoral competence, for which age is arguably the most reasonable proxy. Moreover, the impracticality of individual competence assessments makes an age-based qualification a reasonable means by which to ensure electoral competence.

In order to determine reliably the age by which young people will have achieved electoral competence, it is necessary to have a conception of what electoral competence (p. 762) entails. In other words, what is the nature of the decision to be made, and what sort of knowledge/cognitive processes are required to make it? Do voters need a certain amount of political knowledge, or life experience, or something else, in order to make competent voting decisions?

Rousseau believed a well-informed citizenry was necessary to determine and implement the public good.18 Yet data consistently show the typical voter to be far removed from the ideal citizen of classic democratic theory.19 Indeed, incorporating even basic levels of civics or political knowledge into a conception of electoral competence would, if translated into voter qualification rules, operate to disfranchise a significant proportion of the current adult electorate.20

Possessing limited knowledge, time, and motivation, voters instead rely on information shortcuts, or heuristics, to help them make decisions. Researchers have found that the use of heuristics in vote decision-making, such as party affiliation and group endorsements, effectively enables voters to make the same vote decisions they would have made under conditions of full information.21 These limited information decision strategies, moreover, work as well as (and sometimes better than) traditional rational decision strategies.

The findings that most adult voters lack substantive political knowledge yet are nonetheless able to make “correct” voting decisions (defined as the choice that the voter would have made under conditions of full information, given the voter’s subjective values and beliefs) militates against a conception of electoral competence that includes a certain level of civics or political knowledge. In other words, young people ought not be required to possess specific factual knowledge in order to be considered competent to vote.

Without requiring specific knowledge, it is possible instead to characterize the various cognitive processes (mental operations) typically involved in the voting decision. To cast a nonrandom, rational vote, a typical voter acquires relevant information (such as candidates’ names and party affiliations), encodes it (for example, candidate X is a Republican and Republicans tend to support lower taxes), and applies additional reasoning processes to reach a decision. While forms of reasoning include deductive, inductive, and analogical reasoning, rationality does not necessarily require applying formal logic to a set of premises. Instead, rationality merely requires “good enough reasons” to justify a choice.22 Thus the typical voter may reason that a tax reduction would benefit her and vote for the Republican candidate. While she may not have made a particularly well-informed choice, she has made a minimally competent one, and one that is likely to be correct (in that it is likely to be consistent with a decision she might have made had she possessed full information). And because there is no universal state of “mature” cognitive processing beyond adolescence, a pragmatic standard for electoral competence could require “adultlike” cognitive-processing capacities—i.e., the minimum levels of thinking and processing attained by developmentally normal adults.

A useful standard of electoral competence might thus provide that a “minimally competent voting decision involves an adultlike application and coordination of reasoning processes to make a choice that can be justified by a good-enough reason.”23 (p. 763) The issue then becomes whether it is possible to identify the age by which individuals will have reliably attained the relevant competence.

As noted above, researchers studying cognitive development have found that by mid-adolescence, thinking processes are adultlike. By ages fifteen or sixteen, adolescents are as able as adults to acquire, retain, and retrieve relevant information and apply reasoning processes that lead to justifiable conclusions. In ideal conditions, they make competent, rational decisions. At the same time, their capacities are more susceptible to being confounded by the real-world contexts in which they make decisions. Whether they possess electoral competence thus depends on whether there are factors present in the voting context that might undermine or interfere with their decision-making.

An examination of the characteristics of electoral politics reveals voting to be a decision-making domain in which midadolescents’ adultlike cognitive-processing abilities should remain uncompromised: Elections unfold over a period of time, allowing voters time to deliberate and evaluate their options without undue pressure. There are many sources of information readily available, which voters can use as sort of a scaffolding or heuristics to help them evaluate their choices. And finally, voting itself is done anonymously and in private, which diminishes the concern that adolescents’ choices will be unduly pressured or influenced by peers.

When considering whether to extend new rights to young people, policymakers should also consider the consequences of a bad decision. Even assuming that adolescent voters made irresponsible or “wrong” voting decisions at higher rates than older voters (though there is no evidence that this would be the case), errors would have to occur in a consistent direction in order to affect the outcome of a typical election. However, it is more likely that errors would be randomly distributed, canceling each other out and having no effect on electoral outcomes.

Voter qualification rules based on the presumed incompetence of citizens younger than eighteen thus misunderstand the nature of most voters’ decision-making processes, and misjudge the age by which young citizens attain electoral competence.

A very different decision-making context—and one in which adolescents arguably lack adultlike capacity—is driving.

3. Driving

The presumptive age of licensure is sixteen in the United States, making it the earliest-licensing nation in the developed world.24 Teen drivers aged sixteen to nineteen have four times the crash rate of adults aged twenty and older.25 Crash rates are consistently highest among sixteen-year-olds, and decline substantially with each year of increasing age. Indeed, sixteen-year-old drivers crash at rates 250 percent higher than that of eighteen-year-old drivers.26 Researchers have found, moreover, that differences in driving experience alone do not account for the differences in crash rates. As a (p. 764) group, novice drivers have higher crash rates than do experienced drivers, but at each month of driving experience, younger novices have significantly higher crash rates than do older novices. Thus the crash risk for fifteen-year-old beginners is much higher than that for seventeen-year-old beginners. Eighteen-year-old beginners, however, have only a slightly higher crash risk than twenty-year-old beginners. In other words, at younger ages, driving inexperience plays a role, but a secondary one. Age- and development-related factors predominate in the earlier years of adolescence, then decline relative to experience. For the youngest novice drivers, development-related-related factors compound the risk related to their driving inexperience.27

Some of the experience-related driving risk can be ameliorated through education programs or other interventions, but development-related driving risk cannot. New drivers can acquire the basic elements of driving—basic knowledge of traffic rules and vehicle-handling skills—within a matter of days, or hours. Driving, however, requires a host of other skills—regularly performing specific patterns of visual search, interpreting a constantly changing external environment; recognizing and rapidly/appropriately responding to potential hazards; and maintaining near-constant attention to the driving task.

In order to perform these skills competently, a driver must resist distractions, whether internal, in-vehicle (such as passengers), or external. The driving inexperience of fifteen- and sixteen-year-olds requires them to devote significant cognitive resources to the driving task. At the same time, immature self-regulatory capacities render them more susceptible to these distractions, as well as lapses in attention and impulsive actions.28

The heightened reward salience experienced during adolescence also leads to increased sensation-seeking. For young drivers, strongly desirable sensations associated with driving include excitement, power, and increased status among peers. A propensity for sensation-seeking, moreover, is associated with risky driving, traffic violations, and car crashes.29

Driving, then, presents a social context in which adolescents struggle to perform competently. The externalities of their poor decisions, moreover, are significant. Young drivers are at fault in 75 to 80 percent of the car crashes in which they are involved.30 And while traffic fatalities account for nearly 40 percent of all deaths of sixteen- to nineteen-year-olds, even higher numbers of non-drivers (passengers, pedestrians, etc.) lose their lives in crashes caused by teen drivers.31

What, then, is an appropriate regulatory response? For large numbers of Americans, driving is a necessity. The suburban sprawl that typifies the US landscape (such that cycling, for example, is not a feasible option), combined with relatively weak systems of public transportation, leaves American adolescents with few options compared to their European counterparts.

I have argued that the most appropriate regulatory response is lowering the age for learner’s licensure (to enable novice drivers to accumulate experience) and raising the age of full licensure. Studies have shown that extended learner’s periods (in which (p. 765) new drivers practice under adult supervision) significantly improve safety. One of the most common methods of educating new drivers, however—short-term drivers’ education programs—do not improve the safety of adolescent drivers. While perhaps counterintuitive, studies have shown that completing drivers’ education courses tends to lead to overconfidence, which in turn increases crash risk.32

More recently, states have implemented graduated licensing, where new drivers must comply with a range of restrictions for a period of time.33 These restrictions can include prohibitions on driving in situations known to increase crash risk—for example, nighttime driving or driving with peers. To the extent that graduated licensing offers adolescents the opportunity to gain practice driving in lower-risk conditions, they seem to be part of an effective approach to reducing adolescent driving risk. States that have adopted graduated licensing provisions have seen crashes among sixteen-year-old drivers decrease between 10 percent and as much as 30 percent.34

Finally, education efforts might help shift the cultural meaning of driving, especially for boys and young men. Currently, risky driving is associated with male identity, and male drivers overall have approximately 60 percent higher rates of traffic violations, car crashes, and fatal car crashes than their female counterparts.35 The difference is even more stark for teen drivers, with male teens having even higher crash rates than female teens.36 It is possible that replacing the conception that pairs masculinity with “responsible driving” rather than “daring or even skillful” driving may eventually reduce the disproportionate crash risk of young male drivers.

Voting and driving represent decision-making contexts that map relatively well onto what we know about the nature of adolescent cognitive development. Determining the age at which to permit young people to purchase and consume alcoholic beverages, however, in some ways presents a more complex social and regulatory context.

4. Legal Drinking Age

Voting and driving both occur in public spaces, which facilitates monitoring and the enforcement of legal rules. Alcohol consumption, on the other hand, commonly occurs in private spaces, out of view of the public and law enforcement. Thus while legal rules will operate to dissuade some individuals under the drinking age from consuming alcohol, individuals under the legal age who are sufficiently motivated tend to have little difficulty circumventing the law.37 Policymakers must accordingly take account of the inherent limits and unintended effects of legal rules that establish minimum drinking ages.

The current drinking age throughout the U.S. is twenty-one. Prior to the 1980s, the legal drinking age in most states was lower—eighteen, nineteen, or twenty. Advocacy groups, including Mothers Against Drunk Driving, worked to raise awareness of fatalities caused by drunk drivers, and in 1982, President Reagan appointed the (p. 766) Presidential Commission on Drunk Driving. After the commission recommended raising the minimum legal drinking age, Congress enacted the National Minimum Drinking Age Act (NMDA) in 1984.38 The act authorizes Congress to withhold a percentage of a state’s federal highway funds unless its minimum legal drinking age is at least twenty-one. The US Supreme Court upheld the constitutionality of the NMDA in South Dakota v. Dole, and by 1988 every state had raised its drinking age to twenty-one.39

Evidence suggests that the higher drinking age is associated with lower rates of alcohol use and fewer traffic fatalities.40 Some researchers have suggested that the decrease in traffic fatalities may be a result of a combination of factors other than (or in addition to) the drinking age. For example, advances in auto-safety technologies such as airbags, developments in trauma care, and stepped-up law enforcement and education efforts might also have contributed to declining traffic fatalities.41

Even if some part of the reduction in drunk-driving-related traffic fatalities is due to the effects of other factors, reducing access to alcohol among those younger than twenty-one will predictably contribute to a reduction in drunk driving, simply because it raises the cost of consuming alcohol for those who are under the legal age. A higher drinking age makes purchasing alcohol more difficult for underage individuals, and exposes them to the risk of legal sanction. Accordingly, raising the legal age even higher will result in some further reduction in drunk-driving-related crashes. While fewer traffic fatalities is an undeniable benefit of a higher drinking age, taking account of a range of factors—social contexts, health effects, etc.—can help inform policy. A more inclusive account might serve to reinforce the suitability of current policy, but it may also reveal costs that offset or undermine the benefits of current policy.

Research suggests that alcohol physiologically affects adolescents differently than adults, both in the short-term and long-term. A normal adult liver can safely process approximately 50 alcohol calories an hour (one ounce of 40-percent alcohol). Teenagers’ livers can process only half that amount. As a result, young people are more susceptible to the intoxicating effects of alcohol than are adults.42

Development-related factors also seem to make the adolescent brain more vulnerable to the addictive effects of alcohol. A 2006 study, for example, found a strong inverse correlation between the age at which people began drinking regularly and lifetime risk of alcoholism. The correlation between age of first exposure and risk of lifetime dependence is strongest at younger ages. The youngest drinkers, those who began drinking before age fourteen, had a lifetime risk of alcohol dependence of 47 percent. Those who began drinking at age seventeen had a lifetime risk of 25 percent, and those who began drinking at age twenty-one had a lifetime risk of 10 percent.43 It is possible that factors other than age of first regular exposure explain the correlation. For example, rather than being the cause of later-life alcoholism, early drinking may reflect a preexisting vulnerability to alcoholism.44

Another study sought to identify whether a higher legal drinking age correlated with less problem drinking in adulthood. It found that young men coming of age (pre-NMDA) in states with legal drinking ages lower than twenty-one had 20- to (p. 767) 30-percent higher alcohol consumption as adults, and experienced a 10-percent increase in fatal traffic accidents over adult males. For females, there was no similar association between the legal drinking ages of their home states and adult levels of alcohol consumption.45

Evidence suggests that a higher drinking age does limit alcohol consumption by those under eighteen. For these young people, who generally live with parents, a higher drinking age reduces their access to alcohol and makes it less likely that they will have peers of legal age to purchase alcohol for them. The percentage of twelfth graders reporting drinking at least monthly declined since the widespread adoption of twenty-one as the drinking age, from 70 percent in 1977 to 45 percent in 2007. Fewer high-school-aged students drink overall compared with the late 1970s. The same is not true, however, of high school graduates, especially college students.

More than 40 percent of full-time college students reported being heavy drinkers in 2000, and slightly fewer other young adults aged eighteen to twenty-two—36 percent—reported being heavy drinkers.46 Binge drinking (five or more drinks for males and four or more for females, reaching a blood-alcohol content of .08) increased among underage drinkers more than any other group as the last century drew to a close, climbing 56 percent between the 1990s and 2001.47

The social settings in which young people access and consume alcohol vary, and different settings correlate with different consumption behaviors. Setting the drinking age must account for the contexts in which young people consume alcohol, as well as the short- and long-term physiological effects of alcohol consumption on adolescents compared to adults.

College campuses present a decision-making context that taxes the abilities of adolescent students. For most youths, arrival at college marks the first time that they experience freedom from parental supervision, and they must navigate unfamiliar educational, geographic, and social environments. The exclusive presence of peers heightens reward salience and can further compromise decision-making. Older college students are able to legally purchase alcohol, moreover, which they can provide to underage counterparts, rendering drinking age restrictions merely hortatory.

Some college administrators worry that the twenty-one-year-old age limit has driven underage youth to engage in “goal-oriented” clandestine drinking in order to get drunk, out of view of the public or older adults who might supervise or model more appropriate behavior.48 While the legal drinking age may have contributed to a decline in traffic fatalities, these administrators point to the thousands of annual cases of alcohol-involved assaults, date rapes, property damage, and alcohol poisoning involving underage drinkers as evidence that the drinking age is ineffective policy.49 In response, a coalition of more than one hundred college presidents and chancellors founded Amethyst Initiative in 2008, which urges reconsideration of the drinking age and a more comprehensive and pragmatic approach to address problem drinking.

Members of Amethyst Initiative argue that current drinking-age laws prevent colleges from discussing moderate, responsible drinking. Instead, they are limited to abstinence-only messages, which they claim are starkly at odds with ingrained campus (p. 768) (indeed, national) drinking culture. A nonprofit founded by members of Amethyst Initiative to address drinking laws, Choose Responsibility, has proposed permitting eighteen-year-olds to consume alcohol with parents and granting them licenses to purchase alcohol after completing an educational course.50

Our current knowledge of the physiological effects of alcohol consumption at different developmental stages remains imperfect, and the various social contexts in which alcohol consumption occurs further complicate analysis. We do know that alcohol consumption is an ingrained aspect of adult national culture. Efforts like those embraced by Amethyst Initiative and Choose Responsibility attempt to alter the social context in which young people are introduced into this aspect of adult culture. The federal law imposing a minimum drinking age as a condition for certain funding, however, renders it prohibitively costly for states to explore regulatory regimes that may be more effective than current law. It is possible that, absent the passage of measures that free states to experiment, we will continue to question whether current policy is optimal.

5. Conclusion

Policymakers must regularly navigate the tension between respecting young people’s autonomy versus protecting their welfare interests irrespective of their subjective desires. The contexts in which young people have a claim to decision-making autonomy will vary. Moreover, as culture and society evolve, so too will the nature of these decision-making contexts. For example, if (when?) automakers develop technology enabling vehicles to reliably override human error, the considerations surrounding adolescent licensure too will change.

In short, policymakers can learn a great deal from science about adolescent capacities and deficiencies, and this knowledge can inform and improve policymaking. Social context will always matter, however, such that we must constantly endeavor to understand the range of influences on adolescent well-being and adolescent decision-making.


Arnett, J. J. “Developmental Sources of Crash Risk in Young Drivers.” Injury Prevention 8, no. 2 (2002): ii17–23.Find this resource:

Beckman, Ludvig. The Frontiers of Democracy: The Right to Vote and Its Limits. London: Palgrave Macmillan, 2009.Find this resource:

Braitman, Keli A., Bevan B. Kirley, Anne T. McCartt, and Neil K. Chaudhary. “Crashes of Novice Teenage Drivers: Characteristics and Contributing Factors.” Journal of Safety Research 39, no. 1 (2008): 47–54.Find this resource:

Branche, Christine, Allan F. Williams, and Dede Feldman. “Graduated Licensing for Teens: Why Everybody’s Doing It.” Journal of Law, Medicine & Ethics 30, no. S3 (2002): 146–149.Find this resource:

Chafetz, Morris E. “The 21-Year-Old Drinking Age: I Voted for It; It Doesn’t Work.” Huffington Post, August 18, 2009.Find this resource:

Craig, Stephen C., and Michael D. Martinez. “Voter Competence.” In The Electoral Challenge: Theory Meets Practice, edited by David B. Hill and Stephen C. Craig, 62–90. Washington, DC: CQ Press, 2011.Find this resource:

Dahl, Ronald E. “Biological, Developmental, and Neurobehavioral Factors Relevant to Adolescent Driving Risks.” American Journal of Preventive Medicine 35, no. 3 (2008): S278.Find this resource:

Diamond, Adele (ed.). “The Developmental and Neural Bases of Higher Cognitive Functions.” New York Academy of Sciences 608 (1990).Find this resource:

Hamilton, Vivian E. “Democratic Inclusion, Cognitive Development, and the Age of Electoral Majority.” Brooklyn Law Review 77, no. 4 (2012): 1447–1513.Find this resource:

Hamilton, Vivian E. “Immature Citizens and the State.” BYU Law Review 2010 no. 4 (2010): 1055–1147.Find this resource:

Holland, Carol, Jennifer Geraghty, and Kruti Shah. “Differential Moderating Effect of Locus of Control on Effect of Driving Experience in Young Male and Female Drivers.” Personality and Individual Differences 48, no. 7 (2010): 821–826.Find this resource:

Jacoby, Susan. The Age of American Unreason. New York: Vintage Books, 2008.Find this resource:

Keating, Daniel P., and Bonnie Halpern-Felsher. “Adolescent Drivers: A Developmental Perspective on Risk, Proficiency, and Safety.” American Journal of Preventive Medicine 35, no. 3 (2008): S272–S277.Find this resource:

Lau, Richard R., and David P. Redlawsk. How Voters Decide: Information Processing During Election Campaigns. New York: Cambridge University Press, 2006.Find this resource:

Levesque, Roger J. R. Introduction to Adolescence, Rapid Social Change, and the Law, 3–28. Bloomington, IN: Springer, 2016.Find this resource:

Mayhew, Daniel R., Herbert M. Simpson, and Anita Pak. “Changes in Collision Rates Among Novice Drivers During the First Months of Driving.” Accident Analysis & Prevention 35, no. 5 (2003): 683–691.Find this resource:

(p. 773) McCartt, Anne T., Eric Teoh, Michele Fields, Keli Braitman, and Laurie Hellinga. “Graduated Licensing Laws and Fatal Crashes of Teenage Drivers: A National Study.” Traffic Injury Prevention 11, no. 3 (2010): 240–248.Find this resource:

McMullen, Judith G. “Underage Drinking: Does Current Policy Make Sense?” Lewis & Clark Law Review 10 (2006): 333–365.Find this resource:

Minton, Michelle. “Lower the Drinking Age for Everyone.” National Review, April 20, 2011.Find this resource:

Moshman, David. Adolescence Psychological Development: Rationality, Morality and Identity. Mahwah, NJ: Lawrence Erlbaum Associates, 2005.Find this resource:

National Research Council, Transportation Research Board, Institute of Medicine, Division of Behavioral and Social Sciences and Education, Youth and Families Board on Children, Program Committee for a Workshop on Contributions from the Behavioral and Social Science in Reducing, and Preventing Teen Motor Crashes. Preventing Teen Motor Crashes: Contributions from the Behavioral and Social Sciences: Workshop Report. Washington, DC: National Academies Press, 2007.Find this resource:

O’Connor, Lydia. “Lower Voting Age Proposal Fails in San Francisco, Wins Smaller Victory in Berkeley.” Huffington Post, November 10, 2016. this resource:

Reyna, Valerie F., and Frank Farley. “Risk and Rationality in Adolescent Decision Making: Implications for Theory, Practice, and Public Policy.” Psychological Science in the Public Interest 7, no. 1 (2006): 1–44.Find this resource:

Roan, Shari. “Tempest in a Bottle.” Los Angeles Times, September 1, 2008.Find this resource:

Rousseau, Jean-Jacques. The Social Contract and Other Later Political Writings. Cambridge: Cambridge University Press, 2012.Find this resource:

Scott-Parker, Bridie, Barry Watson, Mark J. King, and Melissa K. Hyde. “The Influence of Sensitivity to Reward and Punishment, Propensity for Sensation Seeking, Depression, and Anxiety on the Risky Behaviour of Novice Drivers: A Path Model.” British Journal of Psychology 103 (2012): 248–267.Find this resource:

Shulman, Elizabeth P., Ashley R. Smith, Karol Silva, Grace Icenogle, Natasha Duell, Jason Chein, and Laurence Steinberg. “The Dual Systems Model: Review, Reappraisal, and Reaffirmation.” Developmental Cognitive Neuroscience 17 (2016): 103–117.Find this resource:

Steinberg, Laurence. “A Social Neuroscience Perspective on Adolescent Risk-Taking.” Developmental Review 28, no. 1 (2008): 78–106.Find this resource:

Steinberg, Laurence, and Elizabeth Scott. Rethinking Juvenile Justice. London: Harvard University Press, 2008.Find this resource:

Thebault, Reis. “Should 16-Year-Olds Be Able to Vote? A Majority of the D.C. Council Thinks So.” Washington Post, June 27, 2018. this resource:

Waller, Patricia F., Michael R. Elliott, Jean T. Shope, Trivellore E. Raghunathan, and Roderick Little. “Changes in Young Adult Offense and Crash Patterns over Time.” Accident Analysis & Prevention 33, no. 1 (2001): 117–128.Find this resource:

White, Melanie J., Lauren C. Cunningham, and Kristeen Tichener. “Young Drivers’ Optimism Bias for Accident Risk and Driving Skill: Accountability and Insight Experience Manipulations.” Accident Analysis & Prevention 43, no. 4 (2011): 1309–1315.Find this resource:

Williams, Allan F. “Young Driver Risk Factors: Successful and Unsuccessful Approaches for Dealing with Them and an Agenda for the Future.” Injury Prevention 12, no. 1 (2006): i4–i8. (p. 774) Find this resource:


(1.) Judge Cardozo explained that when adjudicating custody of a child, a judge “does not proceed upon the theory that the petitioner, whether father or mother, has a cause of action against the other or indeed against anyone. He acts as parens patriae to what is best for the interests of the child.” Finlay v. Finlay, 148 N.E. 624, 629 (1925).

(3.) See, e.g., Laurence Steinberg and Elizabeth Scott, Rethinking Juvenile Justice (London: Harvard University Press, 2008) (describing aspects of adolescent cognitive development that warrant treating juvenile offenders differently than adult offenders).

(4.) See, e.g., Roper v. Simmons, 543 U.S. 551, 574 (2005), in which the Supreme Court held unconstitutional the imposition of the death penalty for crimes committed by an offender younger than age eighteen.

(6.) Laurence Steinberg, “A Social Neuroscience Perspective on Adolescent Risk-Taking,” Developmental Review 28, no. 1 (2008): 78–106. Not all cognitive processes mature by midadolescence. Regions of the brain that control higher thinking, emotion control, and planning do not mature until the end of adolescence. Thus some higher-level processes, including certain aspects of working memory (involved in a number of complex mental abilities, such as the ability to filter irrelevant information and suppress inappropriate actions), continue to specialize and develop into adulthood, maturing only in the early twenties. Ibid., 94–96. See also David Moshman, Adolescence Psychological Development: Rationality, Morality and Identity (Mahwah, NJ: Lawrence Erlbaum Associates, 2005), 24.

(7.) The peaking of neural activity in the socioemotional system is closely tied to pubertal maturation and tends to occur slightly earlier in girls than in boys.

(8.) The process by which this occurs is not yet fully understood. One theory is that the reward system becomes desensitized to the effects of the pubertal hormones that flood the brain beginning in adolescence, especially in dopamine-rich limbic regions associated with reward processing.

(9.) More research is needed to better understand the details of what is still a relatively rudimentary model. Researchers have suggested variations, but the model gained widest approval is one proposed in 2008 by developmental psychologist Laurence Steinberg, which he termed the “dual systems” model. See Steinberg, “A Social Neuroscience Perspective,” 78–106. See also, Elizabeth P. Shulman et al., “The Dual Systems Model: Review, Reappraisal, and Reaffirmation,” Developmental Cognitive Neuroscience 17 (2016): 103, 104.

(11.) Even scholars seeking to advance our understanding of young people on occasion continue to conflate adolescence with chronological age. See, e.g., Roger J. R. Levesque, Introduction to Adolescence, Rapid Social Change, and the Law (Bloomington, IN: Springer, 2016), 3–28 (defining adolescents as “teens under 18, whom the law views as ‘minors’ or ‘juveniles’”).

(12.) Austria lowered its national voting age to sixteen in 2007, primarily to counterbalance the growing percentage of voters aged sixty-five and older. Because those under eighteen could not vote at all, and because older Austrians voted at rates higher than that of younger citizens, the demographic imbalance raised concerns that government would become less responsive to the interests of young people. Nearly half of the sixteen German states, and several Swiss states have also lowered the voting age to sixteen for local elections. The self-governing Crown dependencies in the British Islands (the Isle of Man and the Bailiwicks of Guernsey and Jersey) lowered their voting ages from eighteen to sixteen in 2006, 2007, and 2008, respectively. The Parliamentary Assemblies of both Scotland and Wales voted in favor of lowering the voting age to sixteen (and sixteen-year-olds voted in the Scottish referendum on independence in 2010), although neither is currently able to implement the change without approval of or delegation of authority from the central UK Parliament. See Vivian E. Hamilton, “Democratic Inclusion, Cognitive Development, and the Age of Electoral Majority,” Brooklyn Law Review 77, no. 4 (2012): 1465–1473.

(13.) The Fourteenth Amendment indirectly enshrined twenty-one as the national age of electoral majority, by threatening states that deprived “male inhabitants…twenty-one years of age, and citizens of the United States” with loss of representation in Congress and the electoral college. U.S. Const. amend. XIV, §2, amended by U.S. Constitution Amendment 26, §1.

(14.) Oregon v. Mitchell, 400 U.S. 112 (1970).

(16.) “17-Year-Olds and Primary Elections,” National Conference of State Legislature, updated May 9, 2018,

(22.) See Moshman, Adolescent Psychological Development, 25–26.

(23.) I discuss voter decision-making and propose this conception of electoral competence in Hamilton, “Democratic Inclusion,” 1501–1504.

(30.) Braitman et al., “Crashes of Novice Teenage Drivers,” 52.

(33.) Florida was the first US state to replace its conventional licensure system with graduated licensure, in 1996. Anne T. McCartt et al., “Graduated Licensing Laws and Fatal Crashes of Teenage Drivers: A National Study,” Traffic Injury Prevention 11, no. 3 (2010): 24.

(36.) National Research Council et al., Preventing Teen Motor Crashes, 11–12.

(38.) 23 U.S.C. § 158.

(39.) 483 U.S. 203 (1987).

(40.) Ibid., 342.

(42.) “Key Results: Minimum Legal Drinking Age Policy,” Substance Abuse Policy Research Program, updated 2009,

(43.) See McMullen, “Underage Drinking,” 350.

(44.) Ibid.

(45.) Robert Kaestner and Benjamin Yarnoff, “Long-Term Effects of Minimum Legal Drinking Age on Adult Alcohol Use and Driving Fatalities,” Journal of Law & Economics 54, no. 2 (2011): 325–363.

(46.) McMullen, “Underage Drinking,” 348.

(47.) Roan, “Tempest in a Bottle,” citing 2003 J.A.M.A. study.

(49.) Ibid.

(50.) Roan, “Tempest in a Bottle.”