Did Leaded Gasoline Contribute to the U.S. Baby Bust?

The U.S. fertility rate has fallen to a 40-year low of 1.76 children per 100 women, according to the Centers for Disease Control and Prevention (CDC). Is lead to blame?

A new study by three Carnegie Mellon University (CMU) economists says it is. Noting that “increased lead exposure lowers the general fertility rate for women of childbearing age,” the trio argues that “reductions in airborne lead between 1978 and 1988 increased fertility rates and that higher lead in topsoil decreased fertility rates in the 2000s.” The might right, but another explanation for the falling fertility rate is more convincing.

Between 1973 and 1996, the U.S. completed the process of totally eliminating lead as an additive in gasoline. As a result, lead concentrations in the atmosphere have dropped by 99 percent since 1980. In their study, the CMU economists look at fertility rate trends in that portion of the population (about 30 percent) where they have adequate data for atmospheric concentrations of lead in the 1980s. They similarly seek to correlate fertility rates in about 70 percent of the population with their exposure to lead in the topsoil after the year 2000.

After taking into account possible confounders—local climate, housing, socioeconomic status, education, nearness to highways, race, poverty, unemployment rates—they calculate that the decline in airborne lead meant 95,000 additional babies would be born annually. They also estimate that cleaning up areas with high concentrations of lead in the topsoil to the median level of contamination would induce an additional 166,000 births annually.

The researchers do find a correlation between seeking out infertility services and living in states with above-average airborne lead concentrations. But CDC data show that the infertility rate for married women fell from 11.2 percent in 1965 to 10.3 percent in 1976 and further to 8.4 percent in 1982, and the use of leaded gasoline was increasing for most of that period. The CDC reports that the infertility rate among married women continued to fall, reaching about 6 percent in 2010, despite the fact that the CMU researchers identify ongoing exposure to lead in topsoil as a contributing factor to lower fertility rates. (I cite infertility rates of married women because that’s the longest consistent dataset offered by the CDC.)

Interestingly, data from the 1950s report atmospheric lead concentrations in major cities that were similar to those during the 1980s. Yet post–World War II fertility rates soared, resulting in the baby boom. U.S. total fertility rates fell steeply from the baby boom years, reaching their nadir of 1.74 children per woman in 1976. Total fertility rose to 2.08 children in 1990 and then held more or less steady at around 2 children per woman, rising to a pre–Great Recession peak of 2.12 children in 2007.

The CDC has tracked contraception use among married women from 1965 onwards (adding unmarried later). In 1965, some 63 percent of married women were using some form of contraception, including oral contraceptives, IUDs, and condoms. That rose to nearly 68 percent in 1976, as oral and intrauterine contraceptives displaced condoms. In addition, sterilizations increased from 7.8 to 18.6 percent of women during that decade. In 2008, the CDC reports, 78.6 percent of married women practiced some form of contraception.

Another way to parse U.S. fertility data is to consider just how many children U.S. women on average intended to have versus the actual number that they give birth to. A 2005 study looking at intended and ideal family sizes between 1970 and 2002 finds that both remained stable at just over two children during that period. “Reported fertility intentions of American women approximate the country’s contemporary period levels of fertility,” notes the study. Adding that their findings show “in the aggregate, both stable intentions across time and an ability to realize those intentions.”

A 2010 study that looked at the childbearing intentions of a late baby boom cohort (born between 1957 and 1964) found that the women aimed to give birth to an average of 2.22 children. By 2006, they had given birth to 1.97 children on average. So why did a substantial proportion of Americans in that cohort end up having fewer children than they intended? The researchers highlight the role played by “life-course factors,” finding that “both women and men who postponed childbearing or marriage were much more likely to have fewer births than they intended.” The study points out that postponement implicates declining fecundity with age, competing nonfamilial activities, and failure to find a suitable marriage partner.

Exposure to lead is undeniably harmful to human health. And perhaps exposure to lead has contributed to lower U.S. fertility rates. But it seems more likely to me that U.S. fertility rates have fallen largely because Americans are choosing to have fewer children.

For further background, see also my analysis of the claims that exposure to leaded gasoline increased U.S. crime rates.