The United States just hit a 40-year low in its fertility rate, according to numbers just released by the Centers for Disease Control and Prevention. The 2017 provisional estimate of fertility for the entire U.S. indicates about 3.85 million births in 2017 and a total fertility rate of about 1.76 births per women. These are low numbers: births were as high as 4.31 million in 2007, and the total fertility rate was 2.08 kids back then. The United States has experienced a remarkable slump in fertility over the last several years, as I’ve explained elsewhere. 

Since 2007, fertility has fallen the most for the youngest women, but in the last year, declines have set in for women in their 30s as well. Fertility declines increasingly seem to be about much more than just postponed fertility, or else these women must be planning to have some very fertile 40s.

At least through 2016, this trend appeared to be mostly driven by changes in marital status. Births to never-married women are down more than births to ever-married women: age-adjusted marital fertility is down 14% since 2007, while age-adjusted never-married fertility is down 21%, as of 2016. Preliminary data from several states suggest these trends are likely to continue in 2017.

When it comes to discussions about declining fertility, conservatives tend to “get it” right away: not having a next generation, or having a far smaller one, will cause problems down the line. In my experience, progressives tend to be more hesitant: is this a back-door argument to keep women out of the workplace? No; in fact, there’s robust empirical evidence most women want more kids. Is this some science-denying attempt to ignore climate change? Again, no; in fact, no plausible trajectory of U.S. fertility has any appreciable impact on carbon emissions.

The decline in fertility has been far greater among minorities than among non-Hispanic whites. If we take age-specific birth rates from the peak-fertility year of 2007 and apply them to each age cohort in 2008-2016, the most recent complete data, we can create a counterfactual scenario of how many babies would have been born if age-adjusted fertility rates had not fallen after 2007. From 2008 to 2016, the deficit turns out to be between 4.1 and 4.6 million missing babies: basically, an entire year’s worth or more of childbearing vanished.

Other factors are at work, too. Fertility has fallen somewhat more for less educated than for more educated women. Age-adjusted fertility has fallen 15% for women with a bachelor’s degree or less, versus just 7% for women with graduate degrees. On the whole, births to women with no bachelor’s have totaled 12% below what would be expected if 2007 fertility rates had continued, yielding 3.1 million missing births, while births to women with a bachelor’s degree are down 10% for 1.1 million missing births, and births to women with a graduate degree are down just 7%, or 300,000.

In relative terms, these differences are smaller than differences by age or race, suggesting that socioeconomic class is not the biggest driver of changing fertility rates. This is borne out using other variables: Having insurance or not doesn’t seem to make any difference in fertility changes since 2007, and household income also has no clear, linear trend. With more covariates, these factors might have a stronger effect, but they aren’t the dominant story of declining fertility. There are plenty of other factors that could be looked at, but the basic conclusion is pretty straightforward.

Race, ethnicity, marital status, and geography are the best predictors of changes in fertility over the last decade. Fertility declines are most strongly associated with factors that are race- or region-specific, not broadly class-specific, as different economic classes appear to have quite similar trends. This doesn’t rule out all economic causes: there are important interactions between race and socioeconomic class. But this association does suggest two key takeaways to be kept in mind when discussing declining fertility: it is disproportionately landing on minority moms, births have fallen most for unmarried women, and economically-oriented solutions may only have modest direct effects.

Full story at