In June of 2020, three months after the COVID pandemic began in earnest in the United States, we wrote a report suggesting that the public health crisis and associated recession would result in 300,000 to 500,000 fewer births in 2021. Six months later, we have been asked several times if we have an updated estimate. We have revisited the issue and stand by our initial prediction of a large reduction in births. Based on our previous methodology and a labor market that improved somewhat more quickly than we anticipated, we place more emphasis on the lower range of our original estimate, likely closer to 300,000 fewer births. However, additional factors that we did not incorporate into our model – in particular, ongoing school and day care closures – might very well mean a larger reduction in births than that.
It will still be several months before birth data will become available that will enable us to count the “missing” births. Additional survey evidence has come out, though, since we released our initial report that supports a coming baby bust. Corroborating evidence generated since our June report supports our prediction of a baby bust next year. A survey conducted by Laura D. Lindberg, Alicia VandeVusse, Jennifer Mueller and Marielle Kirstein of the Guttmacher Institute reveals that that 34 percent of American women have either delayed their plans to have a child or reduced the number of children they expect to have as a result of the pandemic. A different survey conducted by Francesca Luppi, Bruno Arpino, and Alessandro Rosina shows that European women similarly report that they plan to postpone giving birth or have fewer children.
Levels of sexual activity have also fallen. In one survey (conducted by Justin Lehmiller, Justin Garcia, Amanda Gesselman, and Kristen Mark of the Kinsey Institute), almost half of adults surveyed report a decline in their sex lives. In another (conducted by Devel Hensel, Molly Rosenberg, Maya Luetke, Tsungchieh Fu, and Debby Herbenick at the University of Indiana), those with young children and, particularly, those with school-age children report the largest declines in intercourse.
One way to gauge individual behavior is to examine what they search for in Google; these data are available through Google Trends. A study by Joshua Wilde, Wei Chen, and Sophie Lohmann based on these data supports our prediction of reduced fertility. The authors report that searches for pregnancy-related terms, such as “ClearBlue” (a pregnancy test), “ultrasound,” and “morning sickness” have fallen since the pandemic began. Based on the reduced searches for pregnancy-related terms, the authors of that study forecast a reduction of births on the order of 15 percent, an even larger drop than what we forecasted.
THE PACE OF THE ECONOMIC RECOVERY
In the six months that have elapsed since our original June report, labor market conditions have improved more rapidly than experts were predicting back in June. A key element of our forecast for declining births was based on our empirical analysis that found that a one percentage point increase in the unemployment rate is associated with a one percent drop in the birth rate. We applied that estimated relationship to the expectation of a seven to 10 percentage point increase in the unemployment rate, which was in line with Federal Reserve expectations.
From today’s vantage point, it looks more likely that unemployment will have risen by around 5.5 percentage points in the year following the start of the pandemic (April 2020 through March 2021) from 3.5 percent to roughly nine percent. This estimate is based on observed data from the Bureau of Labor Statistics for April through November and assumes little change in the next few months. Using this revised expected change in unemployment, we would predict a 5.5 percent reduction in births from the unemployment effect alone. Applying that to the number of births in 2019 (3.75 million) suggests 206,000 fewer births in 2021.
Our original forecast also incorporated an additional reduction in births coming from the anxiety and social conditions associated with the public health crisis. We incorporated this into our forecast by examining the experience of the 1918 Spanish Flu. Back then, every spike in the death rate attributable to the flu was associated with a dramatic reduction in births nine months later. We relied on that evidence to increase our forecast based solely on labor market conditions by one to three additional percent, or another 38,000 to 114,000 fewer births.
The public health crisis has unfolded much as anticipated in the spring and similar in magnitude to the 1918 experience. That pandemic led to 408,000 deaths in the year after it began. Currently 290,000 deaths have occurred as of December 8 and 539,000 deaths are forecast to occur by April 1. The population of the United States is three times as large today as it was in 1918, but medical care has significantly improved, which should have reduced the death rate for a similar level of disease. We see no reason to alter our forecast based on these data. Combining the updated impact of the recession with our earlier additional impact of the public health crisis indicates that we should expect a baby bust in the range of about 245,000 to 320,000.
Yet there are reasons to believe that our July prediction might understate the impact on fertility. Ongoing school closures are putting tremendous strain on families that may reduce their willingness to have more children. Restrictions on public gatherings and social encounters might mean fewer new couplings that could lead to pregnancies, intended or otherwise. The extended nature of this crisis also is likely to create large structural changes in the economy; a sizable share of the jobs lost will be permanent. The longer the duration of the income loss that workers expect, the more likely it is that delayed births will never happen. We did not attempt to make any predictions based on these additional factors because we had no previous context or data from which to draw empirical conclusions.
It will still be several months before data will be available on the number of post-pandemic births that we can use to begin to assess our forecast. In the meantime, we have revisited our prediction based on the most recent evidence available. As of now, we stand by our prediction of a COVID baby bust of around 300,000 fewer births. But the longer the pandemic lasts, and the deeper the economic and social anxiety runs, it is feasible that we will see an even larger reduction in births with an increasing share of them averted permanently.