Just as on the college market, this relationship exists because grade-twelve girls have to compete for scarce grade-twelve boys, not only because boys are less likely to finish high school but also because grade-twelve girls have to compete with girls in younger grades who are happy to have relationships with older boys. This market power of older boys not only puts pressure on girls to have relationships that involve sex (even when they would prefer that they didn’t), but it also hands bargaining power over to boys when it come to the decision to consistently apply protection against disease (i.e., wearing a condom).
Since women face higher risks of infection from unprotected sex than do men, giving men bargaining power over condom use leads to lower rates of condom usage because those making the decision face a lower cost of unprotected sex and, it could be argued, have a greater benefit from not using a condom.
Add to this the fact that older boys are not only transmitting infections to girls in their own year but also to girls in the cohort below them, and you have the making of an epidemic as diseases are passed through the “generations” of students.
This brings us to the issue of race that I raised earlier in the chapter, particularly the observation that girls who are black have significantly higher rates of STDs than girls who are either white or Hispanic. The authors of this paper observe a strong same-race preference in matching among high school students; specifically they find that over 86 percent of the couples in their sample are ones in which both people are of the same race. Black women were significantly more likely to be in a match with a black man than a man of any other race; 99 percent of black women were having sex with black men. Also, they find that the black teenage men in their sample are more likely to be having a relationship with someone of a different race than are black teenage women; 11 percent of black men were matched with women of a different race.
When we consider the observation that the recent high school completion rates for black men are between 7 and 12 percentage points lower than for black women (according to James Heckman and Paul LaFontaine), all the evidence suggests that black teenage women are competing with each other and with women of other races for far fewer black men on the high school market.
The evidence from the Arcidiacono paper I discussed earlier implies that this gender imbalance will result in black women having sex at much higher rates than would be their preference had there been a more even playing field. It also suggests that the bargaining power handed to black men as a result of their market power will lead to much higher rates of unprotected sex.
ABSTINENCE-ONLY PROGRAMS IN AFRICA INCREASE AIDS RISK TO TEENS
Kenya has an HIV education program that teaches young women moral values, refusal skills, and abstinence until marriage. Despite this call for abstinence, 21 percent of grade-eight girls and 48 percent of grade-eight boys report that they are sexually active. There is no mention of condom use in this program and, perhaps as a result, the HIV prevalence rates for youth in Kenya are high: 3 percent for girls between the ages of 15, 19.9 percent for women between the ages of 20 and 24, and 13 percent for women between 25 and 29.
One of the reasons why HIV is so prevalent among girls in Kenya—four times higher, in fact, than among boys of the same age—is that girls are having sexual relationships with much older men. For example, among girls who have become pregnant within a year of grade eight, 49 percent report the father is more than five years older and 16 percent report that he is more than ten years older.
The economic explanation for this phenomenon is that older men are acting as “sugar daddies,” negotiating unprotected sex with young girls in exchange for money and gifts.
Stanford economist Pascaline Dupas recently published the results of a novel experiment in which researchers introduced an educational program into a randomly selected subset of schools in Kenya that gave students one simple piece of information: they informed them of the prevalence of HIV by age and gender in the closest nearby city. They stayed within national guidelines and did not raise the issue of condom use, but they did answer questions that were raised by students.
In the year following the program, the pregnancy rate of girls who were informed that sex with older men exposed them to greater risk of HIV transmission was 28 percent lower than in the control group that had only the standard abstinence-only program. The biggest decline was in girls who were impregnated by men more than five years older than themselves: the number of women who fit that description fell an incredible 61.7 percent. Girls in the treatment group were also 36 percent more likely to report having used a condom in their last act of intercourse relative to the control group.
We don’t know if this simple-to-implement program reduced the incidence of HIV—just that it lowered the rate of unprotected vaginal intercourse with older men—but in just one forty-minute presentation in seventy-one different schools, these researchers managed to avert the birth of thirty children born to 15-year-old mothers; it brought about a significant change in student behavior.
Gender imbalances in high school leads to higher rates of risky sex for black teenage women, leading to more pregnancies and more sexually transmitted diseases.
One more thought on this issue. You might remember the evidence on black male incarceration rates we discussed in chapter 4 that showed that high rates of incarceration are reducing marriage rates of black women. In a similar paper, by Stéphane Mechoulan, the evidence suggests that a 1-percentage-point increase in the adult black male incarceration rate leads to an increase in the expected average age at which a black teenage woman gives birth by about seven extra months.
In the face of high incarceration rates for young black males, teenage women are postponing becoming sexually active; they are doing a better job at protecting themselves again pregnancy, presumably in response to concerns that their children’s fathers will not be in a position to provide them with support in the future.
ABSTINENCE MAY MAKE THE HEART GROW FONDER, BUT DOES IT INCREASE GRADES?
U.S. federal guidelines for abstinence-only education have imposed the requirement that students in abstinence-only programs be taught that “sexual activity outside of the context of marriage is likely to have harmful psychological and physical effects.” So, even without the unintended consequences of pregnancy and disease, being sexually active in high school is emotionally damaging to teenagers and undermines their chances of academic success, right?
The truth is that if teen sexuality is psychologically harmful, researchers have been unable to find any evidence to substantiate that claim.
A paper by Joseph Sabia and Daniel Rees tackles this issue using a large nationally representative sample of U.S. students taken in three stages: 1995, 1996, and 2001. They find that delaying sexual intercourse by just one year increases a teenage woman’s probability of graduating from high school by 4.4 percent. This effect of early loss of virginity on academic success, however, was exclusively among white teenage women. They found no relationship between virginity loss and high school graduation rates for either Hispanic or black teenage women.
If the cause of failing to graduate from high school is psychological harm, then early sexual activity is proven to be harmful only to white teenage women. The problem with this conclusion, however, is that while the authors control for a variety of factors that might influence both sexuality and high school completion, they have left out one thing: teen pregnancy. Given that a girl is likely to get pregnant in a given year only if she is sexually active, then it must be true that becoming sexually active one year earlier increases the chance that she will become pregnant at some point in high school.
We can’t use this evidence to decisively say that early entrance to sexuality is psychologically harmful for girls when we already know that the responsibilities of parenting a young child during adolescence makes it very difficult to complete high school.
The same paper finds that delaying having sex by one year has no statistically significant effect on
the high school graduation rate of teenage men, regardless of race. You may be tempted to conclude from this evidence that sex is psychologically less damaging for men, but the reality is that it is much easier for a man who has impregnated his girlfriend to finish school than it is for the girlfriend who has experienced childbirth.
Without controlling for childbirth, we can’t say for certain whether or not there is any high school completion effect for early entrance to sexuality among girls who can effectively control their own fertility, and the speculation that girls are more psychologically fragile than boys when it comes to sexuality is just that—speculation.
One interesting recommendation made by the authors of this paper is that abstinence-only programs should abandon the claim that premarital sexual activity has psychological effects and replace it with a more “nuanced message.” I am sure you share my curiosity as to what that might look like in the classroom, given these empirical results.
For example, perhaps teachers should give a stern warning to their students that sex is psychologically harmful, followed by the clarification that the boys can have sex without fear that it will negatively affect their school performance, but only with either black and Hispanic teenage women who also appear to be unaffected.
That message should, at least, get the kids’ attention.
One more question: if teen sex is psychologically harmful, then shouldn’t teens who witness the ill effects suffered by their close friends who have become sexually active learn from that friend’s mistake and avoid similar consequences?
A recent paper by David Card and Laura Giuliano taps into the unusual feature of the Add Health survey, which allows researchers to identify a teen’s peer group within a high school. They use this evidence to answer the following question: If a student’s best friend engages in a risky behavior (sex, smoking, marijuana use, or truancy), then what is the probability that the friend will engage in the same behavior?
SELLING THEIR MOST VALUABLE ASSET
Teenage women are repeatedly told not to give away their most valuable asset—their virginity. A few years ago, a brothel in Nevada announced it was selling the virginity of Natalie Dylan, a 21-year-old college graduate, for a remarkable $3.8 million. This has made many women wonder if there really is value to this “asset” and, since that time, many have tried to replicate Natalie’s success. But market forces are at play, even with this unusual service, and profiting from a virginity sale is unlikely in the future.
A recent paper by Fabio Mariani argues that cross-societal variations in the value of virginity are closely tied to opportunities that women have on the marriage market. His argument goes something like this: If a wealthy man meets a poor woman with whom he falls in love, he might be willing to marry her if she is a virgin despite the fact she is poor. If she is not a virgin, however, he might prefer to marry a wealthy woman who is a virgin even if he doesn’t love her.
The return on virginity for a poor girl, therefore, is the probability of marrying a rich man multiplied by his income. Anything that reduces this expected return to virginity in marriage will reduce the compensation a woman will need to receive in order to encourage her to relinquish her virginity outside of marriage. This compensation is her reservation price for selling her virginity.
Higher income inequality among men, like that observed in developed nations, should lead to higher-reservation value for virginity as women have more to gain by remaining a virgin when the gap in income between rich and poor men is greater.
At the same time, the virginity market has only one barrier to entry: participants must be chaste up until the point of the sale. This almost-free entry to a market implies competition in the market will be so high that the price will be driven down to a fair market value. I suspect that the price that men are willing to pay on a perfectly competitive virginity market is actually far below the reservation price of most potential sellers.
One final element that almost certainly eliminates the possibility of profitable virginity sales is the competition would-be virginity sellers in the West face from women in markets where virginity has a much lower expected value. The price of a virgin in a U.S.–Mexican border town is only about $400, well below the millions Western women are asking to prostitute themselves in this way.
Best friends tend to come from similar family situations, for example, be the same race and age, have similar educational goals, and similar attitudes toward risk. The authors are able to control for these factors and still find that if a student’s best friend has had sexual intercourse, then she or he is 4.5 percentage points more likely than the average student to have sexual intercourse the following year. If their friend had “intimate contact” (essentially making it to third base), the probability their friend will have intimate contact as well is 4 percentage points higher than the average student.
The best-friend effect is as significant as living in a single-parent household or of having two parents who did not finish high school—both of which increase the probability that a teenager is sexually active.
One possible explanation for the best-friend effect on sex is that teens smoke pot together and that pot smoking leads to a higher rate of sexual activity. Or it is that they drink together and that drinking leads to higher rates of sexual activity. It turns out, though, that pot use doesn’t increase entrance to sexual activity in teens, so even though having a best friend who smokes pot increases (to a small degree) the probability that the student also smokes pot, this can’t explain the increase in sexual activity. And while alcohol use increases sexual activity (significantly), having a best friend who drinks alcohol does not increase the probably that a student drinks more than the average student.
This leaves us with the perhaps unpalatable explanation that teenagers find sex pleasurable and that at least some teens observe their friends having sex without any of the expected negative consequences, and that encourages them to do the same.
There is one other paper that might explain the best-friend effect. Jesús Fernández-Villaverde, Jeremy Greenwood, and Nezih Guner argue that as contraceptives have become more effective, and the risk associated with premarital sex has fallen, parents have reduced their investment in preventing teens from becoming sexually active.
One way parents prevent their teenage children from becoming sexually active is by teaching them that premarital sex is shameful behavior. Using the same Add Health data we just discussed, these authors find that while teens are more likely to have already had sex if their friends have had sex, they are less likely to have had sex if they believe that sex is a shameful thing to do (with shame measured using the answers to questions like “How would you feel if your mother knew you were having sex?”).
This suggests that the reason why teens are more likely to follow in their friends’ footsteps—having sex if their friends have sex—is that having a sexually active friend reduces the sense of shame, or stigma, attached to being a sexually active teenager.
FINAL WORDS
Remember the young man I talked about at the beginning of this chapter, the one who told me he is going to wait until he is 21 to have a girlfriend? The funny part of that story is that his mother is bitterly disappointed that her son is not out there getting some action, just as she was when she was his age (and, in fact, far younger).
Not everyone remembers the sex they had when they were teens as fondly as my friend does, but it is interesting to note that the current generation of parents with teenage children might very well be the first generation in which the parents were more sexually active in high school than their kids.
Personally, I remember realizing in my early 20s that my cohort was on the tail end of the sexual revolution and that subsequent generations wouldn’t take the same sexual liberties that my generation had when we were young. At the time, my belief that the sexual revolution was over came from the growing awareness around HIV/AIDS. In hindsight, fear of disease has probably played a role in decreasing teen promiscuit
y, but the steady persistent decline in teen promiscuity over twenty years suggest that an economic explanation is warranted.
Recent media reports on declining teen promiscuity suggest that fear is playing a major role in that change in behavior and I think they are right. It isn’t just fear that life will be hard with a baby; teenagers didn’t need reality TV to inform them that trips to the mall aren’t the same with a crying baby. It is a fear of being left behind in an economy in which the only workers who have seen real increases in their standard of living over the past thirty years are those who continued their education after high school. It is fear of being at the bottom of the income distribution when those at the top lavishly consume while everyone else races to keep up. It is an understanding that childbirth early in life leads to a permanent decrease in lifetime earning ability that has encouraged both young men and women to be more cautious regarding their sexuality.
Of course, for these economic incentives to bring teen pregnancy rates in the United States to the low levels observed in other countries they would have to be available to everyone. All teenagers, for example, would have to believe that their lifetime earnings would be reduced by a teen pregnancy in order for that incentive to change their risk-taking behavior. But the reality is that the lifetime earning ability of a low-skilled worker is completely unaffected by having a baby as a teenager. And moving from unskilled to skilled worker requires an investment than many low-income families cannot afford for their children, regardless of their sexual behavior.
I started this chapter with a description of the way in which economics has shaped social norms regarding teen sexual behavior. I have one thought to add: Social norms within the socioeconomic groups that have been the most disadvantaged by the modern economy have evolved to take a more permissive view of teen sexuality. This economic perspective is important because without it we are tempted to believe that the causality runs in the opposite direction—that people have become economically marginalized because of their sexual behavior. That skewed perspective conveniently ignores the reality that teenagers in high-income families are not less promiscuous because they have higher moral standards, but because they face an entirely different set of economic incentives that have shaped the way that standards have been set within their communities.
Dollars and Sex Page 19