We’ve long heard the theory, from Criminology 101 to the U.S. Supreme Court: teenagers are crime prone (even “deadly”), biologically and developmentally impulsive, peer-driven risk takers, heedless of consequences.
The statistics would seem unassailable: in every culture, ages 15-24 or so have higher crime rates than those 25 and older. Of course, authorities once held that excessive crime by African Americans was an innate feature of primitive racial biology and undeveloped culture.
Unchastened by history, modern theorists have failed to investigate whether “adolescent risks” are explained not by bio-developmental internalities, but by straightforward externalities. Poverty is linked to many high-risk behaviors, and adolescents and young adults are much poorer than older adults.
The Center on Juvenile and Criminal Justice initiated studies investigating this policy-laden question, a tough one since the economic status of offenders must be estimated from the age, race, and geography associated with outcome statistics — a consistent, comprehensive data base but with some methodological challenges. The results of our macro-level analyses of risk outcomes for ages 14-69 were consistent and compelling.
With regard to violent crime, for example, we found that two-thirds of 14-19 year-olds live in areas where youths’ poverty rates top 15 percent, and these areas account for a staggering 86 percent of teens’ arrests. In contrast, just 11 percent of 40-69 year-olds live in areas where middle-aged poverty rivals typical teenaged rates, but they account for 25 percent of middle-aged arrests.
Conversely, just 19 percent of teenagers reside in areas where teen poverty rates average less than 10 percent, but these accounted for just 7 percent of teenagers’ arrests. However, 66 percent of 40-69 year-olds live in areas where middle-agers have poverty rates below 10 percent, accounting for fewer than half of middle-aged arrests.
We found that where teens enjoy low “middle-aged” affluence levels, teens display low “middle-aged” crime, violence, gun homicide, and traffic fatality rates. Where middle-agers suffer high, “teenaged” poverty rates, middle-agers have high crime, violence, gun homicide, and fatal traffic crash rates.
The only difference between teen and middle age with regard to criminal and other risk propensities is that far fewer middle-agers than teens live in high-poverty environments. Teenagers, like Latinos and Louisianans, suffer certain risks because they’re poorer, not because they’re young. We replicated these findings using various years, locales, and outcomes with strengthened results over time.
Our findings were rejected by traditionalists. Three noted authors responded, arguing the National Longitudinal Survey of Youth’s (NLSY) self-reported offending from Wave 1 (1996) to Wave 7 (2003) showed crime peaking around age 15 and declining thereafter, even when self-reported poverty status was held constant.
But this analysis was invalid on its face. The NLSY is limited to a single, outdated cohort (those born in 1980-1984) and contains crippling flaws and inconsistencies. For one, overall crime declined sharply from 1996 to 2003; individuals would report much lower crime rates in 2003 than in 1996 regardless of age or aging. Further, subject attrition in later waves cost Wave 7 a large share of its highest-risk respondents, rendering it incomparable to Wave 1.
For another flaw, socioeconomic status was reported inconsistently by parents for their families (Wave 1), then by young people for their households (Wave 7). For a third, the study’s limited ages, 10-24, enhanced the “college effect” (students often suffer high, temporary poverty along with low crime rates). Yet, its authors posed this lone study confined to 10-24 year-olds employing a messed-up data base as definitive proof that “the age-crime curve in adolescence and early adulthood is not due to age differences in economic status” and constituted a “strong refutation” to our studies of risks spanning ages 15-69.
That teenage crime and poverty are inextricably related has profound implications for policy. We need fewer age-based remediations and more measures to improve youths’ economic status and college access.
American political, medical, and academic establishments took decades to abandon prejudices that people of color are biologically inclined to steal watermelons, Mexicans are innately “hot blooded,” and Native Americans are genetic savages. Likewise, giving up expedient “teenage brain” phrenologies and developmental biases is imperative to refocusing and modernizing juvenile justice.
Defendant age may be legally relevant for other reasons. The United States affords teenagers fewer rights than do other Western nations. Youths denied the adult right to control their environments should not be held as responsible as adults for crime fostered by those environments—an issue that should be drawing far more attention than resurrected 19th century bio-determinism.