COVID-19 and the Future of US Fertility: What Can We Learn from Google?

Joshua K. Wilde, Max Planck Institute for Demographic Research (MPIDR)
Wei Chen, Max Planck Institute for Demographic Research (MPIDR)
Sophie Lohmann, Max Planck Institute for Demographic Research (MPIDR)

The future effect of the COVID-19 pandemic on birthrates is currently unknown. Fortunately, Google search data are a particularly appealing predictor of future births. With these data, this study creates the first state-level dynamic prediction of the pandemic’s effect on US birthrates. Specifically, we employ simple machine learning techniques to select conception, pregnancy, and economic uncertainty keywords, and a time-lagged fixed-effects regression model to demonstrate that periods of above-normal search volume for these keywords predict births in the following months. Our analysis suggests that between November 2020 and February 2021, monthly US births will drop sharply by approximately 15% – the largest birth decline experienced in the US in over a century. States with fewer individuals with a college education, lower household income, and larger minority populations are predicted to have larger declines in fertility. Births remain depressed through August 2021, suggesting the potential for an extended “baby bust”.

Keywords: COVID-19, Digital and computational demography, Fertility and childbirth, Population projections, forecasts, and estimations

See paper.

  Presented in Session 171. COVID-19 and Fertility