Mental Health during the Pandemic
KEY TAKEAWAYS
More than two years after the COVID-19 pandemic began, moderate to severe anxiety in the U.S. was below its April 2020 level. During the second quarter of 2020, real gross domestic product contracted sharply. From March 2020 to April 2020, nonfarm payrolls fell by more than 20 million. And the number of unemployed jumped by more than 16 million. However, from April 2020 to December 2021, moderate to severe anxiety had four distinct periods of fluctuation.
- From May 14 to July 16, 2020, anxiety increased by 7.8 percentage points.
- It then fell by 4.4 percentage points from July 16 to Sept. 2, 2020.
- Anxiety then increased by 6 percentage points between Sept. 2 and Nov. 11, 2020.
- Finally, from Nov. 11, 2020, to May 26, 2021, anxiety fell by 12 percentage points.
For the rest of 2021, anxiety was relatively stable. Changes in public health restrictions, COVID-19 cases and deaths, and economic conditions all significantly explain these four periods of changes in anxiety, with the greatest contribution coming from changing economic conditions.
Read the paper “Mental Health over the Pandemic.”
The views expressed in this paper are solely those of the authors and do not necessarily reflect the views of the Federal Reserve Bank of St. Louis or the Federal Reserve System. Any errors or omissions are the responsibility of the authors.
The COVID-19 pandemic disrupted the daily lives of millions. The associated public health measures aimed at minimizing the spread of the virus created isolation, while illness, death and economic hardship over the pandemic resulted in major loss. These are all triggers for poorer mental health outcomes.
Once the U.S. was through the pandemic’s initial phases, the economy began to recover and many public health restrictions were lifted, although COVID-19 ebbed and surged over time. Between April 2020 and December 2021, moderate to severe anxiety increased, fell, increased and then retreated (see the figure below). Headwinds of COVID-19 case severity and tailwinds of economic expansion and reduced restrictions may help explain this variation.
Anxiety Rate over Time, April 2020-December 2021
SOURCES: U.S. Census Bureau’s Household Pulse Survey and authors’ calculations.
NOTES: Weighted mean anxiety by week, HPS weeks 1-40 (April 2020-December 2021). Sample size is 2,541,405.
The evidence reported here demonstrates the challenges U.S. workers and their families faced and describes the key contributors to recent patterns in anxiety. First, we estimate the key contributors to anxiety using the full period (April 2020-December 2021). Second, we create four subperiods. During these subperiods, we see a pattern: Anxiety rises, falls, rises and then falls again. We generate predicted changes in anxiety in each of these periods that are associated with changes in the economy’s health and changes in an index capturing a state’s policy stringency and COVID-19 severity. This allows us to assess the importance of these factors in explaining the patterns in anxiety. Finally, we replicate that strategy across socioeconomic and demographic characteristics throughout the entire period.
How much did the strong labor market and softened public health restrictions contribute to improving mental health? How much did COVID-19 surges mitigate the improvement? And did these factors have a disparate effect across demographic and socioeconomic groups?
Why do the answers to these questions matter? Consider that health, including mental health, is a determinant of labor participation and productivity. Poor mental health, like lack of access to affordable child care, can serve as a barrier to reentering the labor market and might explain a portion of the workers lost since the start of the pandemic. Additionally, at work, the labor productivity of particularly affected individuals will diminish, resulting in lower wages and fewer advancement opportunities at the individual level, as well as lost production and economic growth.
We highlight a few of our findings below. You can read more about them in our working paper or see our related work pertaining to labor market barriers facing some groups.
Women, Young Adults and Low-Income Groups Have the Highest Anxiety Rates
The average adult anxiety rateWe define anxiety rate as the average share of the sample exhibiting moderate to severe anxiety. between April 2020 and December 2021 was 31.2%. Some groups experienced rates much higher than the average, including women, young adults and low-income groups. The figure below shows anxiety by age group. Several studies have found that younger people’s mental health declined more than the mental health of those in older age groups, and that younger people have been seeking help online in increasing numbers since the start of the pandemic.See American Psychological Association (2020) and Mental Health America (2023).
Anxiety by Age Group, April 2020-Decemer 2021
SOURCES: U.S. Census Bureau’s Household Pulse Survey and authors’ calculations.
NOTES: Weighted mean anxiety by age group, HPS weeks 1-40 (April 2020-December 2021). Sample size is 2,541,405.
As the following figure illustrates, the mental health of low-income individuals deteriorated significantly more than the mental health of those in higher-income groups, partly because of financial constraints.See Akay (2022). The layoffs and discharges early in the pandemic and the subsequent hiring boom were inconsistent across industries. In April 2020, the largest reduction in nonfarm payroll occurred in the leisure and hospitality sector. This sector also experienced the largest gains in employment between May and June 2020 as the economy began to reopen. These volatile labor markets early in the pandemic disproportionately hit lower-income workers, impacting their household finances and mental health. Yet, the anxiety rates of women and low-income individuals declined more than the rates of men and those with higher incomes, indicating substantial relief over the period. However, the anxiety rates of young adults saw little change.
Anxiety by Income Group, April 2020-December 2021
SOURCES: U.S. Census Bureau’s Household Pulse Survey and authors’ calculations.
NOTES: Weighted mean anxiety by income, HPS weeks 1-40 (April 2020-December 2021). Sample size is 2,541,405.
Poor Mental Health Is a Hindrance to Labor Supply and Productivity
Mental health influences labor supply, and thus the potential employment level and other labor market outcomes, such as productivity. COVID-19 impacted workers’ lives through illness, death, job loss, reduced autonomy and increased economic uncertainty. If mental health declined over the pandemic, worker productivity and labor force participation likely also declined.
Although we do not explore labor supply or productivity changes, we hope to offer valuable insight into which groups experienced the greatest anxiety levels and changes in anxiety rates, and how the labor market, public health policies and COVID-19 severity contributed to those changes. This information is key to understanding potential labor productivity and participation changes.
Additionally, suppose young adults experienced high anxiety rates and received little relief in those rates. In such a case, they will experience the greatest penalty in lifetime earnings due to reductions in participation and productivity, since they have the longest work life ahead of them.
The Economy Contributed the Most to Changes in Anxiety
Even when we control for socioeconomic and demographic characteristics, state unemployment rates, public health policy stringency and COVID-19 severity remain key predictors of moderate to severe anxiety for the overall population. These measures are also good predictors of women’s anxiety. For all 25- to 34-year-olds, COVID-19 case severity predicts their anxiety. For low-income groups, state public health policy, COVID-19 case severity and the state unemployment rate explain their anxiety.
To measure the ability of each factor to explain the 4.2 percentage point reduction in anxiety from April 2020 to December 2021, we calculate the predicted change in anxiety solely due to each factor. The decline in the unemployment rate explains roughly one-third of the actual change in anxiety.
The weakening of policy stringency explains 22% of the actual change in anxiety. These factors offset a predicted 0.7% increase in anxiety due to a worsening in case severity over the period.
We estimated similar contributions for women, 25- to 34-year-olds and low-income groups. For women, a lower unemployment rate is the largest contributor to the reduction in their anxiety. For 25- to 34-year-olds, the fall in case severity is the largest contributor to their anxiety rate’s decline. Finally, for low-income groups, weakened public health policies and a decline in the unemployment rate contributed to the fall in their anxiety rate. Similar to the overall sample, the rise in COVID-19 case severity placed upward pressure on their anxiety rate.
Conclusion
Between April 2020 and December 2021, anxiety was high among young adults, women and lower-income individuals. Over the same period, anxiety exhibited a pattern of fluctuation: an increase, followed by a decrease, an increase and another decrease. Public health restrictions, COVID-19 cases and deaths, and unemployment all significantly explained changes in anxiety, with the greatest effect coming from the unemployment rate.
Groups with the highest anxiety rates—women, lower-income groups and young adults—faced structural barriers to returning to the labor market (Rodgers and Kassens, 2022). Structural impediments include mothers’ lack of access to child care, which reduces their job-matching efficiency.The evidence presented here complements our recent work on the Beveridge curve over the same period. The research presented suggests that the deterioration in mental health over the pandemic serves as an additional barrier to labor market participation. Even when such barriers are overcome, they can impact labor productivity (Boesch et al., 2021).
Understanding the disparate effects of the pandemic on mental health could help communities develop effective and targeted strategies. Additionally, understanding the relative impact of economic conditions, public health policy and COVID-19 severity on mental health could assist communities as they plan for future incidents (such as other health emergencies or natural disasters).
References
Akay, Alpaslan. “The Local and Global Mental Health Effects of the COVID-19 Pandemic.” Economics and Human Biology, 2022, Vol. 45, Article 101095.
American Psychological Association. Stress in America 2020: A National Mental Health Crisis. Washington, D.C., 2020.
Boesch, Tyler; Grunewald, Rob; Nunn, Ryan; and Palmer, Vanessa. “Pandemic Pushes Mothers of Young Children out of the Labor Force.” Federal Reserve Bank of Minneapolis, Feb. 2, 2021.
Mental Health America. Mental Health and COVID-19: Information and Resources. Alexandria, 2023.
Rodgers, William M.; and Kassens, Alice L. “The Beveridge Curve and Structural Barriers in the Labor Market.” Federal Reserve Bank of St. Louis, On The Economy, Oct. 20, 2022.
Notes
- We define anxiety rate as the average share of the sample exhibiting moderate to severe anxiety.
- See American Psychological Association (2020) and Mental Health America (2023).
- See Akay (2022).
- The evidence presented here complements our recent work on the Beveridge curve over the same period.
Subscribe to receive insights, research-based essays and upcoming events from the Institute for Economic Equity.
Views expressed are not necessarily those of the St. Louis Fed or Federal Reserve System.