The Economics of Peer Pressure

With teen pregnancy rates up for the first time in a decade and the Obama administration eliminating federal funding for abstinence-only sex-education programs, intense debates abound over how to delay and reduce teen sexual activity. Numerous psychological and sociological studies show that peer norms strongly influence sexual initiation among adolescents, and they suggest that interventions targeting social interactions could be effective. Economics doctoral student Seth Richards is bringing a new perspective to bear on how peers influence heterosexual teen sexual activity with research that defines these social interactions in terms of supply and demand.

Richards identifies demand as the decision to search for a sexual partner as influenced by peer norms—or the fraction of a student's same-sex, same-grade peers who are sexually active. Supply is the availability of opposite-sex partners—or opposite-sex members in a student's school that are actively seeking to become sexually active.

Using the economics framework of search and matching, Richards has developed a model that can quantify the distinct impacts supply and demand have on teen sexual activity. This distinguishes his research from existing econometric work on peer influences on risky behaviors because these studies have generally examined a single, composite effect of social interactions.

"The clarity that comes from trying to have a formal model for the behavior you’re studying forces you to be clear about the factors you think are involved." – Seth Richards

"You can imagine that supply and demand really are different factors in this behavior," Richards says. "It's not like studying teen smoking because you don't need to think about how Phillip Morris perceives changes in peer influence in a high school. But changes in peer influences of the opposite gender would affect the availability of partners for you. In terms of search and matching, the probability of finding a match—supply—depends on the search decisions of other people in your school—demand. To flip it around, their demand decisions affect your supply."

Richards used his model to analyze data on teen sexual activity from the National Longitudinal Study of Adolescent Health, which provides a nationally representative sample of U.S. high school students in the mid-1990s. He found that peer norms have a large effect on the timing of sexual initiation for both male and female students, but that the effect is 50 percent greater for boys than for girls. For example, in a counterfactual simulation that removed peer influence on search decisions, the number of students who initiated sex during ninth or tenth grade fell by 42 percent for boys and 22 percent for girls. Richards also discovered that increases and decreases in partner availability respectively increase and decrease the rate of sexual initiation for boys. However, availability does not significantly impact the sexual initiation rate for girls.

"One conclusion," Richards says, "is that boys seem to be more susceptible to these social mechanisms, whether it's the norms among their same-gender peers or the availability of partners. To the extent that people are thinking about interventions that work through social mechanisms, they should consider specifically how they're targeting boys."

Disciplines like psychology and sociology are best equipped to delve into how and why social interactions work they way they do, but Richards believes that the tools of economics are particularly good at isolating and measuring the impact of specific factors on social issues. "The clarity that comes from trying to have a formal model for the behavior you're studying forces you to be clear about the factors you think are involved," he says. "It also encourages you to limit the number of factors you're examining. Because you want to articulate mathematically how they work, you have to become very clear about what they do."

The discipline is also strong in sophisticated statistical methods that allow researchers to think carefully about factors that could be confounding their analyses of social issues. "If you see a correlation among outcomes," Richards explains, "for example, students in a school becoming sexually active at the same age—you can figure out whether they're actually influencing each other or if there's something else about the area that you're not taking into account."

Richards, who will join the Heinz College at Carnegie Mellon as an assistant professor this fall, is currently examining the impact of social norms on teen contraceptive use. The large decline in teen pregnancies between 1995 and 2005 is largely attributed to increased use of contraception, and Richards wants to explore how much this increase was affected by changes in peer attitudes and by school- and community-based interventions. His future research plans include studying network influences on medical decision making among both patients and doctors.