COVID-19’s disruptions disproportionately hit child care workers

As the COVID-19 pandemic has underscored the importance of child care in supporting employment among the nation’s nearly 20 million working parents with young children, the child care occupation itself has suffered large, sustained employment losses.¹ Our data analysis shows that over 40 percent of child care workers who were employed in February 2020—just before the pandemic took hold in the U.S.—were unemployed in April 2020, and total employment in January 2021 remained 20 percent below pre-pandemic levels. These workers were already in a position of economic disadvantage prior to the pandemic. They are predominantly women, are disproportionately people of color, and earn low wages. Our results highlight the need to support both child care workers, who have suffered disproportionately during the pandemic; and the industry, which is critical to our economic recovery.

Workers in many occupations have had difficult pandemic labor market experiences, but there are three distinct reasons to pay special attention to workers in the child care sector. The first is that high-quality early childhood development yields social returns in the form of better outcomes for children in the future, and disruptions to early childhood education can undermine those gains. The second is that the child care sector supports current economic activity in a unique way, allowing mothers in particular to participate in the labor market.² And the third reason is that child care workers are unusually economically disadvantaged, as we detail below.

Child care employment fell sharply

The COVID-19 pandemic affected both demand for and supply of child care workers, who experienced a sharp initial decline in employment. On the demand side, many parents withdrew their children from child care due to fear of COVID-19, while other parents lost their jobs and no longer needed (or could no longer afford) child care. On the supply side, states placed additional safety requirements on child care providers, such as reduced group sizes, that increased operating costs and likely reduced capacity. Data from the Bureau of Labor Statistics Current Population Survey (CPS) show that as a result, employment in this occupation declined from 1.1 million to 600,000—a 47 percent reduction—from January to April 2020.³ By contrast, employment in all occupations fell by a considerably lower but still shocking 15 percent over the same period.

Considered “essential workers” during the pandemic, child care workers have experienced only a partial recovery in their employment since its low level in April 2020. Through the second half of 2020, employment remained around 30 percent below pre-pandemic levels. With the slight increase seen in January 2021, employment in this occupation was 21 percent lower than in January 2020, compared to 5 percent for workers across all occupations. (We use January numbers here to capture annual transitions across occupations and remove any seasonality in overall employment.) Additional months of data will shed light on whether the recent increase in employment represents sampling variation or the beginning of a trend.

Employment fell by much more for employees—that is, those who work for a child care provider—than for the self-employed. See Figure 1 above.⁴ Comparing January to April 2020, employment fell 51 percent among employee child care workers but only 32 percent among self-employed child care workers. This is consistent with the disparate impact of the pandemic on providers: by April 2020, an estimated 63 percent of child care centers—but only 27 percent of family child care providers—had closed across the country.⁵ Self-employed, home-based child care providers may have had an easier time complying with restrictions on class sizes or temporarily operating with fewer children. Additionally, demand for home-based care may have increased if parents viewed it as a safer option during the pandemic. By January 2021, there were 3 percent more self-employed child care workers and 29 percent fewer child care employees relative to January 2020.⁶ As a result, the self-employed now make up over 30 percent of the child care workforce, up from 24 percent in January 2020.

Child care workers, already economically disadvantaged, saw hours decline

Child care workers are economically disadvantaged relative to other workers in the economy. In 2019, child care workers’ median hourly wage was $11.50, which was only 60 percent of the median hourly wage across all workers. These low hourly wages—combined with high rates of part-time work—contribute to low levels of weekly earnings. Child care workers earned a median of $350 per week in 2019, compared to median weekly earnings of $800 across all workers.⁷

In addition to being low-paid, child care workers are nearly all women and disproportionately people of color, as shown in the table below. Disproportionately represented in this occupation, Black and Latino/a workers were especially exposed to child care employment losses. But even among child care workers, these groups saw larger employment losses than others: the Black and Latino/a shares of the child care workforce declined by 1.3 and 3.8 percentage points, respectively, from the fourth quarter of 2019 to the fourth quarter of 2020.⁸

Child care workers who remained employed during the pandemic worked fewer hours than before. See Figure 2. Throughout 2019, roughly 10 percent of child care workers reported having worked fewer than their usual hours. The share who reported working fewer than their usual hours doubled to 20 percent at the beginning of the pandemic, in the second quarter of 2020. Since then, that share has declined, approaching pre-pandemic levels by the end of 2020. The decline in those reporting fewer than their usual hours could result from workers aligning their expectations about their “usual” hours with a new normal of actual hours. Alternatively, employed child care workers’ hours may have returned to normal by the fourth quarter of 2020.

We see a similar pattern in child care workers reporting that they would prefer full-time employment but cannot find it. The share of workers reporting this “part-time for economic reasons” status increased during the early months of the pandemic and declined to 2019 levels by the fourth quarter of 2020. Interestingly, this contrasts with the overall economy, where the share of workers who reported working part-time for economic reasons remains elevated despite a much stronger job recovery.

Despite hardships, child care workers largely remain in the occupation

Every year, some child care workers will change jobs or leave the labor force. In 2020, the rate of transitions was unusual—beyond the normal rate—which is helpful for assessing the effect of the pandemic. To pursue a deeper understanding of the pandemic’s disruptions, we use longitudinal identifiers in the CPS data to follow pre-pandemic child care workers over time and categorize their pandemic employment into four mutually exclusive categories: remained employed in child care, became employed in another occupation, became unemployed, and left the labor force.⁹

We focus on both short-run transitions, from February to April 2020, and long-run transitions, from the fourth quarter of 2019 to the fourth quarter of 2020. The former provide a snapshot of child care workers during the worst months of the pandemic recession, while the latter allow us to see the continuing effects on these workers. In each case, we compare the transition rates with the corresponding rates from the year prior, before the pandemic disrupted the sector.¹⁰

During the early months of the pandemic, child care workers were much more likely than usual to become unemployed. See Figure 3. From February to April 2020, around 43 percent of child care workers became unemployed—far above the 2 percent who became unemployed over the same period in 2019. Perhaps surprisingly, only 3 percent changed occupations from February to April 2020, compared to 7 percent over the same months of 2019, which suggests that child care workers were unable or unwilling to move into another type of employment once the pandemic hit. A weak overall labor market may have made it unusually difficult to find work in new occupations, while increased personal caregiving responsibilities may have led to labor force exit. Another potentially relevant factor was the additional unemployment insurance compensation for eligible workers, the impact of which on job search and re-employment is debatable. Preliminary evidence suggests that the extra $600 weekly payments provided under the Coronavirus Aid, Relief, and Economic Security Act did not discourage job seeking or acceptance, but the effect may have been different for child care workers specifically.¹¹

Turning to the longer-run transitions, workers whom we observe employed in child care in the fourth quarter of 2019 were nearly 10 percentage points less likely to remain employed in the occupation when we observe them a year later, compared to those whom we observe employed in child care in the fourth quarter of 2018. See Figure 4. As of the fourth quarter of 2020, the child care workers whom we observe from 2019 to 2020 were 5 percentage points more likely to be unemployed and 4 percentage points more likely to have left the labor force relative to the child care workers we observe from 2018 to 2019.

What does the future hold?

One year in, the child care workforce may have reached a pandemic normal, remaining 20 percent below its pre-pandemic employment level. This level is depressed but relatively stable, consistent with the pattern of mothers’ labor force participation: after a large decline from November 2019 through April 2020, mothers of young children saw little recovery in labor force participation through the end of 2020.

Given the important role of these workers in maintaining high-quality, accessible child care, our analysis raises two interconnected concerns. First, how can we support child care workers—typically among the lowest-paid workers in our labor market—who remain unemployed or have left the labor force due to lack of job prospects? Some relief is on the way from the recent American Rescue Plan Act, which provides $39 billion to support the child care sector, including funds to cover pandemic-related costs and assist in reopening child care programs that closed due to the pandemic. As the pandemic continues past its one-year anniversary, many child care workers have been unemployed for extended periods of time. Many may no longer have a relationship with a specific employer. Both of these conditions tend to make re-employment more difficult.¹² Improving job opportunities and continuing unemployment support are important steps in addressing the disproportionate impact of the pandemic on low-wage workers.

Second, how can we ensure that high-quality early childhood education is available to support children and a rapid and full rebound in parental employment once the COVID-19 pandemic subsides? This is a critical component of our economic recovery that will pay dividends, both in terms of improving the long-run prospects of young children and supporting the labor force success of their parents.

The authors thank Rob Grunewald, Ben Horowitz, Michael Madowitz, Vanessa Palmer, Angela Rachidi, Bina Patel Shrimali, and Libby Starling for insightful feedback.

Endnotes

1 Calculated from Employment Characteristics of Families-2019, Table 4, from the Bureau of Labor Statistics.

2 See, for example, Madowitz et al. (2016), Rachidi (2020), and Shrimali (2020).

3 The CPS conducts monthly interviews of a representative sample of the U.S. population. The high frequency of the data allows us to study the child care workforce during the pandemic, but the response rates among sampled individuals were significantly lower during our sample period compared to historical rates, particularly in the early months of the pandemic.

4 The child care occupation includes nannies who, in most situations, should legally be characterized as employees but may be considered self-employed due to misclassification or low levels of employment.

5 National Snapshot of State Agency Approaches to Child Care During the COVID-19 Pandemic, U.S. Department of Health and Human Services, Office of Inspector General. September 2020.

6 These data represent employment at a point in time. We cannot determine the breakdown of how potential reasons contributed to the lower job losses in self-employment. These potential reasons include: self-employed child care workers being more likely to remain employed, new child care workers being more likely to start out as self-employed, and existing child care employees becoming self-employed.

7 Weekly earnings are calculated from the CPS outgoing rotation files using earnings weights. Due to data limitations, the earnings figures exclude the self-employed and only reflect the earnings of wage and salary workers.

8 The decline in Black employment share is not statistically significant. Pandemic employment losses were uneven in other respects. We see some evidence that workers with less education and those without a certificate or license were more likely to lose employment during the pandemic, although the differences were not statistically significant. Consistent with these findings and the reduction in part-time work, the average weekly earnings actually rose from $381 in 2019 to $541 during the pandemic in 2020.

9 The structure of the CPS places some limitations on the types of transitions we can study. The survey interviews workers for four consecutive months, skips eight months, and then interviews them again for four consecutive months. In our context, this implies that we can only observe short-term employment transitions through May 2020, and long-run transitions through the beginning of 2021. We are grateful to IPUMS for the use of its longitudinal identifiers. Unfortunately, any longitudinal matching using the CPS cannot observe individuals who move geographically during the sample period or individuals who leave the sample due to non-response. Both factors were likely more prevalent during the pandemic, and it is difficult to know how it could bias our estimates. Finally, smaller sample sizes for the transitional analyses mean the estimates following individual workers are less precise.

10 Every year some child care workers will—in the normal course of events—change jobs or leave the labor force, and we want to assess the effect of the pandemic by looking for transitions that occur to an unusual extent. An underlying assumption is that the pandemic is the reason for the differences between 2019 and 2020, but there are likely other contributing factors and general variation year-to-year in transition rates. Additionally, we focus on the change in transition rates because previous work suggests that transitions between occupations using longitudinal data in the CPS are overstated (Kambourov and Manovskii 2013).

11 See Petrosky-Nadeau and Valletta (2020) and Ganong et al. (2021) for work on the effect of the additional $600 unemployment insurance payments on employment.

12 See Krueger et al. (2014) and Kroft et al. (2016).