While we have all been distracted, some researchers have noticed another thing to worry about: Americans these days report having sex less often than Americans did a couple of decades ago. What?! Is this not supposed to be the age of hookups, Tinder swiping, the pornography web, Viagra, and all that? Yet, the drop-off in sexual activity, though modest in size, is real–for teens, for young adults, for middle-aged people. (All this even before Covid.) For some observers, this decline has become the next social problem.
What’s happening–er, not happening?
Getting Any?
There are two broad categories of studies on sex, one about teens, the other about adults.
Researchers have been tracking teen sexuality–age at first intercourse, number of partners, pregnancy rates, and so on–for a long time. Experts and the public have long viewed teen sex as a social problem–and most still do.[1] CDC data show that in the early 1990s about 38% of high school students reported having had sex in the prior three months, but in the late 2010s, that rate was down to 28%. Other indicators also fell; for example, the percentage ever having had sex dropped from about 53% in the ’90s to about 40% in the ’10s (see also here and here). You would think that fewer teens having sex later would ease worried minds. Yet, for some, the decline in teen sex is a social problem.
☞ Pause for a methods question: Why should we ever believe what teens–or, for that matter, what adults–say in surveys about their sexual activities? Indeed, we should be skeptical about the topline numbers, skeptical for example that 60% of today’s high school kids are virgins; it may be higher or it may be lower. But we can be relatively confident about trends when we compare answers over time. We can be confident unless survey-takers’ dishonesty about sex has substantially changed in one direction over time. One might imagine that the expansion of sex talk and sex images in popular culture would have encouraged 21st-century survey-takers to reveal more of or to exaggerate their sexuality, rather than to downplay it. Systematic bias of that kind should have driven reports of sexual activity up. But they have gone down.
The adult data show a similar moderate decline. One source is this question asked since 1989 in the biannual General Social Survey (GSS): “About how often did you have sex during the last 12 months?,” with answer categories from not at all to four or more times a week. I simplify the results in the chart below. It shows the percentage of GSS respondents who reported having sex once a week or more often, categorized by age, by gender, and by decade.
The most dramatic aspect of the graph–the sharp downwardly-slopping lines–displays the simple fact that the older people were, the less likely they were to report weekly sex. (Cue here philosophical musings about aging.) The second most dramatic result–the blue lines versus the red lines–show that married respondents reported more frequent sex than did the unmarried. (So much for “swinging singles.”) But our focus here is on the difference between the solid lines (1989 through 1999) and the dashed lines (2010 through 2018): The dashed 2010s lines fall below the solid 1990s lines, at least among the married. For any given age (up to the seventies), married GSS respondents in the 2010s were less likely to report weekly sex than married GSS respondents in the 1990s.
Proportionally fewer Americans are married now than were married in the 1990s. So, one component of the overall decline in reported sexual frequency is that more Americans are on the red-line trajectory than before. But the other component of the decline in adult sexual activity is the drop from the 1990s to the 2010s among the married. The frequency of sex for the unmarried did not fall from the 1990s to 2010s. It is among the married where the action (or lack of it) was.[2]
Several researchers using these other American data have reported declining sexual frequency (e.g., here and here). (One analysis found that the proportion of under-35-year-old adults reporting no sex for the last year has notably increased.) Studies in several other countries also show a decline in reported sexual frequency since 2000 and also show that, as in the graph above, the drop-off is particularly marked for people with spouses or, overseas, live-in partners. This is not just American exceptionalism.
What’s (Not) Going On?
I’ve already discussed one of the reasons for the overall decline: Americans are marrying less and later. Why that has been happening is a major subject of its own that I set aside here. What else might be decreasing libido or at least its expression? A just-published study of young adults suggests their drinking less alcohol nowadays is part of the explanation. Less liquor, less disinhibition, less sex? That study and others (e.g., here) also suggest that spending more time on the computer may depress sexual activity Perhaps gaming or Instagramming is more exciting than–or too exhausting for–sex?
But: If online activity is a serious distraction from sex, then the introduction in the 1950s and 1960s of television–a faster-spreading and more engaging distraction–should have also reduced sexual activity. If the Baby Boom of that era is any indication, it didn’t. More systematic efforts to measure the influence of television on sex and birth rates suggest that any effects are small, mixed, and probably due to the content of the programs rather than to the time-sink of television.[3]
Compelling explanations for the decline in sexual frequency have been hard to establish. A few researchers have speculated that it is part of a general, growing interpersonal alienation in society. Another suggestion is that people turn to solo sex because duo sex has become too complicated and porn is so available.
Before we turn in that zeitgeist direction for an explanation, we should put the 21st-century decline into historical context. The fall-off in sexual activity since the 1990s seems to have followed decades of increase. I haven’t found data quite like that of the graph above for earlier periods, but statistics on adolescent sexual activity and age at first intercourse point to greater sexual activity from the 1960s into the 1990s (see, e.g., here, here, and here). So, perhaps what is remarkable was not the drop in sexual frequency from 1990s to now, but a rise in sexual activity from the 1960s to 1990s. This was the era of greater sexual liberation in behavior and attitudes and the era when the Baby Boomers came of age. Maybe Americans of the 2010s were just returning to sexual “normalcy.”
Another possibility is that people are having less sex because of something in the air… or in the water… or in the food. Medical researchers report that levels of testosterone and semen, which are associated with male virility, have been going down for the last few decades (see, e.g., 1, 2, 3, 4). Those levels may be dropping because of changes in lifestyle or in diet (one study points to soda pop) or in the chemical environment.[4]
The puzzle remains: Surveys suggest that youth and married adults (but not the unmarried) are having sex less often now than their predecessors did two decades ago. Maybe the reasons are similar for both groups–changes in norms, changes in recreation, changes in the environment, whatever–or maybe the explanations are different for teens and adults. Also a puzzle: Is this a new social problem?
——————–
Update, Jan. 20, 2021:
Here’s another possibility: Increasing obesity of women has led to lower fertility of their sons.
Update, June 6, 2021:
There is an argument that the decline in sperm counts may be a methodological artifact.
Update, August 4, 2021:
But then, this story: Male fertility is declining – studies show that environmental toxins could be a reason.
Update, January 25, 2023:
The trend continues, as discussed in Phil Cohen’s post. Here is graph he presents from the GSS data, controlling for age and gender. Note that the 2021 data need an asterisk; the GSS methodology was quite different than in prior years.
……………………………………………………………………………………………………………………………..
NOTES
[1] In General Social Surveys of the 2010s, the proportion of Americans saying that sex between 14-to-16 year-olds is “always wrong” was about 60 percent (though declining).
[2] Perhaps the non-change among the unmarried resulted from more of them cohabiting in the 2010s; perhaps they were sorta-married. Indeed, the proportion of Americans cohabiting rose from the 1990s to the 2000s, although it has been flat since. But this probably does not really change the story we see in the graph. I ran a regression of the full sexual frequency scale on year, age, gender, and year X categories of partnering (GSS’s “posslq”), 2000 through 2018. The results show that reported declines in frequency, the coefficients for year, were significant only for the married, not for those cohabiting, nor those with non-resident partners, nor those with “no steady partner.”
[3] A study using surveys of millions of people over many years in less-developed nations estimated that the effect of owning a television was to reduce the likelihood that women respondents reported having sex in the previous week by about 2 percentage points; there was no effect for men. Other studies suggest that sexual content on television encourages sexual activity (here and here). A study in Indonesia found that the arrival of television service in villages seemed to be followed by a decline in the birth rate because the content encouraged use of contraceptives. The researchers believe that the tv shows conveyed the idea that “modern” people have fewer children.
[4] I should also note the theory, popularized by Mother Jones blogger Kevin Drum, that the 1960-90 era was marked by high levels of lead in Americans’ blood, that lead reduces the brain’s executive function, leading to dis-inhibition, so that levels of lead in the environment explain the rise of crime and teen pregnancy in that period and their decline since.
MADE IN AMERICA 