Effects of COVID Lockdowns, Curfews, and Business Closures on Sex Behavior and Realized Fertility

• In Event: Cyber-Risky Behaviors

Abstract

This paper uses rich longitudinal data to model how fertility decisions and trends in fertility evolved before and during the COVID pandemic. It uses a standard economic model of fertility decisions to frame the analysis. The model needs almost no extra adaptation to include measures of COVID-mitigation policies. I exploit cross-state and temporal variation in COVID policies that potentially affected the time couples spend together. The model predicts differential effects for people with an existing partner versus those not in a marriage or marriage-like relationship. The COVID policies could have affected fertility rates, relative to pre-pandemic rates, if they increased (decreased) the rate of sexual intercourse, led to more (fewer) conceptions, and more (fewer) of those women carried their pregnancies to full-term. Because factors work in both directions, one must investigate this question empirically. To establish empirical patterns and disentangle potentially offsetting predicted effects, I use main survey Panel Study of Income Dynamics (PSID) and Transition to Adulthood (TA) survey data. The analysis exploits temporal and cross-state differences in COVID policy interventions. I investigate how partnership formation, rates of intercourse, use of contraception, future birth rates of US residents changed when their state governments ordered residents to “stay-at-home,” imposed curfews, and closed businesses during the COVID-19 pandemic. I exploit variation across states that did and did not intervene. I also exploit the fact that governments enacted and repealed the interventions on different dates. These two features of the COVID-lockdown/curfew restrictions generate geographic and temporal variation in whether or not and when residents faced COVID interventions. To test for differences in birth rates, I use TA on sexual activity and PSID data on births that occurred between January 2018 and December 2021. I test whether sexual activity across people who experienced or did not experience lockdowns in the prior 30 days and whether birth rates differed between people in states that did and did not enact COVID-19 restrictions nine months earlier. I also estimate models that compare differences in birth rates of people in the same year and same month who experienced or did not experience an intervention. Results suggest that the COVID-19 interventions changed both sexual behavior of unpartnered people and the probability of a birth in opposite ways for people with and without a partner. When state governments ordered people to stay-at-home, imposed curfews, and closed businesses related to social activities (bars, restaurants, etc) sexual ctivitie falls among singles and, nine months later, the probability of a birth increased for people with a partner and fell for people without a partner.

Author

• Dean R Lillard, The Ohio State University
  Presenting Author