Department of Labor Logo United States Department of Labor
Dot gov

The .gov means it's official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Article
November 2022

Empirical evidence for the “Great Resignation”

This article empirically assesses the observed increase in job resignations during the coronavirus disease 2019 (COVID-19) pandemic and examines the pandemic’s uniqueness from prior macroeconomic events. The article shows that, compared with the dot-com recession of 2001 and the 2007–09 Great Recession, the pandemic produced unique quits rates, and this finding holds across U.S. census regions. In addition, the results show that, during the pandemic, quits rates in firms with fewer than 1,000 employees were higher than quits rates in firms with more than 1,000 employees. A regression analysis assessing the antecedents of the pandemic’s quits rate also reveals that while the rates for hires and job openings had a positive effect on quits rates, hourly earnings and the unemployment rate exerted a negative effect. The article empirically confirms the “Great Resignation” phenomenon, which is characterized by record job quitting during the pandemic, and suggests that this phenomenon may be ameliorated by increasing hourly earnings, thereby increasing employees’ switching costs. However, if the phenomenon persists, it is conceivable that labor-saving investments that were hitherto economically infeasible will become feasible, altering the nature of work and the workplace.

The coronavirus disease 2019 (COVID-19) pandemic had a sudden, rapid, and unprecedented impact on many national and local economies. In response to the pandemic, governments shut down many segments of their economies, except those deemed “essential.”1 In the United States, the economy “troughed” for a few months immediately after the implementation of these policies. However, the persistence of the pandemic through new COVID-19 variants, coupled with multiple waves of infection, kept the U.S. economy from returning to its prepandemic condition.

A major observation associated with the COVID-19 pandemic is the “Great Resignation” phenomenon, which has received significant attention. This phenomenon, whose moniker was coined by Anthony Klotz, involves record rates of job quitting during the pandemic.2 As noted by one author, return-to-office mandates, attractive job offers from competing employers, and revelations about better work–life balance have motivated a “record-breaking departure from jobs in a shockingly small window of time.”3 Using a global survey of 4,000 companies and more than 9 million employee records, a recent study found that resignations increased the fastest among midcareer employees (i.e., those between 30 and 45 years of age).4 These resignations have also been attributed to people making changes to their work–life balance.5 A Public Broadcasting Service documentary on the future of work explored the potential effect of the COVID-19 pandemic on “American ‘workism,’” observing that, compared with men, women are leaving the workforce more rapidly and in larger numbers for a variety of reasons, including gaining access to childcare and providing care for family.6 However, research conducted prior to the pandemic shows that hires, job openings, and quits all reached new highs in 2018.7 This finding challenges the attribution of the Great Resignation to the pandemic, demanding further empirical investigation that would confirm this attribution.

Consistent with this goal, this article compares U.S. labor market dynamics during the COVID-19 pandemic (here, the pandemic period spans from March 2020 to January 2022) with those from the two previous recessions—the Great Recession (December 2007 to June 2009) and the dot-com recession (March 2001 to November 2001). Because it has been reported that the 9/11 terrorist attack, which occurred prior to the end of the dot-com recession, adversely affected people’s work attitudes, the end of the dot-com recession is extended to March 2002.8

The COVID-19 pandemic, the Great Recession, and the dot-com recession all had significant adverse economic impacts on individuals and organizations. Although the pandemic added sickness, death, and economic shutdowns to its toll, financial challenges and job losses during the previous two downturns produced their own mental anguish.9 The extent to which these impacts might have affected people’s willingness to quit their current employment has not been fully investigated, but there is some evidence that they might have engendered changes to the nature and type of work some people were willing to consider.10 Adding to the fallout from the dot-com recession, the impact of 9/11 exacerbated the ongoing financial challenges faced by some people, especially white-collar workers in certain parts of the country.11 In addition, it has been reported that the increased xenophobia that emerged in some areas after 9/11 has affected the sense of security of workers of certain religions and ethnicities.12

In comparing the aforementioned macroeconomic events, the article does not compare the pandemic’s labor market dynamics with those associated with other recent coronaviruses, because none of these viruses had comparable reach and economic impact. Neither the severe acute respiratory syndrome (SARS), which broke out in China in 2002–03, nor the Middle East respiratory syndrome (MERS), which was first observed in Saudi Arabia in 2012, produced a macroeconomic response comparable to that of COVID-19. SARS occurred in only five countries and registered less than 8,500 cases, including 73 in the United States.13 MERS occurred in 27 countries and recorded 2,519 cases and 866 deaths.14 No SARS-related deaths and only two MERS-related deaths occurred in the United States. In contrast, as of March 2022, the COVID-19 pandemic had caused nearly a million deaths and infected about 80 million people in the United States.15 Globally, COVID-19 had infected more than 455 million people and killed over 6 million by March 2022.16

For its main analyses, the article uses monthly data from the U.S. Bureau of Labor Statistics (BLS) Job Openings and Labor Turnover Survey (JOLTS).17 The JOLTS dataset includes information on job openings, hires, and separations. Separations are organized into three major components: (1) layoffs and discharges, which cover involuntary separations initiated by the employer, (2) quits, which are employee-initiated voluntary departures from current positions, and (3) other separations, defined as departures arising from retirements, transfers to other locations, disability, or death. The period of analysis, which spans from December 2000 through January 2022, is chosen to capture the three macroeconomic events of interest.

The article’s research question is addressed along three dimensions: national level, regional level, and firm size class. The national-level analyses test the hypothesis that the average quits rate during the pandemic was the same as the rates for the other macroeconomic events. To ascertain the Great Resignation phenomenon, the alternative hypothesis is that the average quits rate during the pandemic was higher. The analyses use the JOLTS dataset, whose data elements are defined in the BLS Handbook of Methods.18 Consistent with these definitions, the term “level” refers to a total number associated with a data element (e.g., job openings, hires, total separations, quits, layoffs and discharges, or other separations), whereas the term “rate” involves dividing the level for a data element by total employment and multiplying the quotient by 100, facilitating accounting for economic growth. The regional comparisons consider differences in quits rates across four U.S. census regions: Northeast, South, Midwest, and West.19 Finally, the comparisons by firm size use six BLS firm size classes (detailed later), hypothesizing that the quits rate during the pandemic was the same across the six firm size classes and across the three macroeconomic events within each firm size class.

The literature on job quitting: a brief overview

An economic theory proposed by Kenneth J. McLaughlin suggests that flexible wage rates increase employees’ desire to leave their current employers and employers’ desire to keep current employees when the wage rate is below a certain level (and vice versa).20 When an adverse economic event (e.g., a recession) occurs, the wage-to-profitability ratio increases and triggers layoffs. When the event abates and profitability begins to increase, the wage-to-profitability ratio decreases, motivating employers not only to keep their current employees but also to recruit new ones. Increasing demand for employees introduces a wrinkle in McLaughlin’s theory, because employers’ negotiations with employees become complicated by competitors’ wage offers to the same employees, increasing the latter’s options and strengthening their negotiating position.21

Better wage offers often motivate workers to change jobs without going through a period of unemployment.22 Research has established a procyclical relationship between quits and wage rates, providing a good predictive power for inflation on the upswing.23 However, the sticky-down nature of wages prevents wage adjustments and creates what George A. Akerlof, Andrew K. Rose, and Janet L. Yellen have described as a “vacancy chain.”24 Other researchers have showed that, besides wage rates, factors such as employment rates, employee job satisfaction, perceptions about compensation fairness, and ease of movement between jobs also contribute to job switching.25 Arguing that the foregoing factors are important but insufficient variables in explaining quits, Tae Heon Lee et al. have focused on a different variable—unsolicited offers from competing employers in tight labor markets.26 It has also been noted that quits may occur only after employees have reached certain milestones, such as receipt of stock options or retention bonuses.27 In the context of the COVID-19 pandemic, additional factors emerging as covariates in explaining quits are employee burnout from daily routines (e.g., wake up at 5 a.m., fight traffic, and sit in a cubicle all day) and the conscious decision to change one’s work–life balance.28 These psychological factors, which have received little attention in the literature, may be partially responsible for the unprecedented number of business applications during the pandemic. For example, the average number of monthly business applications between March 2020 and January 2022 was nearly 418,000, compared with about 209,000 during the Great Recession.29

Studies of labor market dynamics commonly assume that wage rates drive labor force participation rates.30 However, the relatively high number of people quitting their jobs and transitioning to self-employment during the pandemic challenges this assumption. It opens the possibility that the decision choice of employees takes into account not only working for wages but also working for self. Some market watchers believe that the challenges of sustaining business performance and income needs might entice some of those who quit their jobs to start their own businesses.31 The foregoing discussion suggests that the factors influencing quits during the COVID-19 pandemic could differ from those of earlier periods, because of the pandemic’s uniqueness in time, scope, and reach.

Quits trend in the United States

The COVID-19 pandemic ended the longest employment and economic expansion in U.S. history.32 Uncertainty about the pandemic’s potential public health effects unleashed an abrupt “closure” of the U.S. economy and the economies of most countries.33 The policy response to the pandemic altered the wage-to-profitability ratio virtually overnight, leading to the most massive and rapid layoff in the last 100 years of U.S. history. Total nonfarm employment went down from about 152.5 million in February 2020 to about 130.2 million by April 2020, a decline of 14.62 percent.34 Chart 1 shows the trends in monthly rates for hires, job openings, layoffs and discharges, and quits in the U.S. nonfarm sector from December 2000 through January 2022, a period that encompasses the three major macroeconomic disruptions considered in this article (the dot-com recession, the Great Recession, and the COVID-19 pandemic). The chart shows that, over the past two decades, the quits rate was higher than the rate for layoffs and discharges, except for the latter half of the Great Recession. The chart also shows that the gradient of the quits rate during the pandemic was steeper than in any prior period.

The number of quits from February 2020 to January 2022 was forecast by using the historical trend from the period following the Great Recession, to assess what that number could have been without the pandemic. The results put the projected quits at about 3.9 million in January 2022, which compares with an actual quits level of almost 4.3 million. Thus, despite the massive drop in quits at the onset of the pandemic, the level of quits “recovered” rapidly, surpassing the number of quits that could have been expected under the historical trend, without the pandemic.

Hires include all additions to payroll during an entire month, including workers who have been rehired on a full-time, part-time, permanent, short-term, or seasonal basis. Job openings include all open positions (full time, part time, permanent, short term, or seasonal) on the last business day of the month and refer to positions for which the employer is actively recruiting from outside the establishment. Adding context to chart 1, job openings and hires averaged about 5 million per month from the end of the Great Recession to February 2020, growing at average monthly rates of 0.80 percent and 0.33 percent, respectively. From April 2020 to January 2022, job openings and hires averaged (respectively) 8.32 million and 6.20 million per month, with the former growing at an average monthly rate of 3.86 percent and the latter remaining virtually flat. Never in the last 20 years of data reviewed for this article has the gap between job openings and hires been this wide. The data show that job openings surpassed hires back in December 2014 and fell below the level of hires only in May and June 2020. Indeed, by January 2022, job openings had exceeded hires by 4.81 million, which is higher than the quits level of 4.20 million in the same month. To break this trend, real wages have to reverse their current downward trend, which would provide superior net benefits from working for wages and entice those working for self to consider switching back to working for wages.35

Table 1 reports summary statistics for the rates of different data elements, enabling an assessment of the effects of each examined macroeconomic event after accounting for economic growth. The job openings rate is computed by dividing the number of job openings by the sum of employment and job openings and multiplying that quotient by 100. The rates for hires, layoffs and discharges, and quits are computed by dividing these data elements’ respective levels by employment and multiplying that quotient by 100. The hires rate for the entire period of analysis averaged 4.10 percent per month (a minimum of 3.10 percent and a maximum of 7.10 percent), with a standard deviation of 0.43 percent. The quits rate averaged 2.20 percent per month (a minimum of 1.30 percent and a maximum of 3.40 percent), with a standard deviation of 0.38 percent.

Table 1. Summary statistics for rates of hires, job openings, layoffs and discharges, and quits, by macroeconomic event, December 2000–January 2022
Macroeconomic eventJOLTS elementNMeanStandard deviationMinimumMaximum

None

Hires1994.050.303.304.80
Job openings1993.440.811.805.10
Layoffs and discharges1991.530.161.202.00
Quits1992.150.331.302.80

Dot-com recession

Hires134.280.184.004.60
Job openings133.130.332.603.70
Layoffs and discharges131.750.131.602.00
Quits132.410.172.202.70

Great Recession

Hires193.670.343.104.10
Job openings192.620.551.703.50
Layoffs and discharges191.850.241.602.30
Quits191.930.301.502.40

Pandemic

Hires234.820.693.507.10
Job openings235.741.343.607.50
Layoffs and discharges231.912.270.909.90
Quits232.710.481.703.40

Total

Hires2544.100.433.107.10
Job openings2543.571.101.707.50
Layoffs and discharges2541.600.700.909.90
Quits2542.200.381.303.40

Note: For this analysis, the dates of the three macroeconomic events of interest are March 2001 to March 2002 for the dot-com recession, December 2007 to June 2009 for the Great Recession, and March 2020 to January 2022 for the coronavirus disease 2019 (COVID-19) pandemic. JOLTS = Job Openings and Labor Turnover Survey.

Source: Author's estimates based on data from the U.S. Bureau of Labor Statistics Job Openings and Labor Turnover Survey.

Because the macroeconomic events considered in the analysis are of different durations, the Scheffé test is used to test the hypothesis that the means of their quits rates are equal. The alternative hypothesis is that the mean for the pandemic is higher than the means for both the dot-com recession and the Great Recession. If the hypothesis of mean equality is not rejected, then the observed trends in quits rates during the pandemic have been merely a continuation of historical trends. Conversely, if the hypothesis is rejected, then the pandemic’s effect on quits rates has differed from the effects of the two prior macroeconomic events.

Table 2 presents pairwise macroeconomic-event comparisons, showing contrast results for the average monthly rates of hires, job openings, layoffs and discharges, and quits. The results indicate that the pandemic’s average monthly quits rate was higher than and statistically different from the average monthly quits rates for the other macroeconomic events (p < 0.001). The hypothesis of equality of average monthly quits rates across the different macroeconomic events is, thus, rejected in favor of the alternative hypothesis. Similar conclusions are reached for hires and job openings, but not for layoffs and discharges. These results suggest that the pandemic did not just alter the quits rate but also profoundly affected two of the three other data elements. However, the results also suggest that the rate for layoffs and discharges, unlike the rates for the other data elements, did not change because of the pandemic.

Table 2. Pairwise macroeconomic-event comparisons for rates of hires, job openings, layoffs and discharges, and quits, December 2000–January 2022
Macroeconomic-event pairContrastStandard errorScheffé tScheffé P > t95-percent confidence intervalSignificance
Lower boundUpper bound

Hires rate

Dot-com recession vs. none

0.230.102.250.171-0.060.51[1]

Great Recession vs. none

-0.380.08-4.500.000-0.62-0.14***

Pandemic vs. none

0.770.089.930.0000.550.99***

Great Recession vs. dot-com recession

-0.610.13-4.790.000-0.97-0.25***

Pandemic vs. dot-com recession

0.540.124.450.0000.200.89***

Pandemic vs. Great Recession

1.150.1110.530.0000.851.46***

Job openings rate

Dot-com recession vs. none

-0.310.24-1.300.642-0.990.36[1]

Great Recession vs. none

-0.820.20-4.080.001-1.39-0.25***

Pandemic vs. none

2.300.1812.450.0001.782.82***

Great Recession vs. dot-com recession

-0.510.30-1.690.417-1.360.34[1]

Pandemic vs. dot-com recession

2.610.298.970.0001.793.43***

Pandemic vs. Great Recession

3.120.2612.000.0002.393.85***

Layoffs and discharges rate

Dot-com recession vs. none

0.230.201.140.731-0.330.78[1]

Great Recession vs. none

0.320.171.950.285-0.140.79[1]

Pandemic vs. none

0.380.152.520.098-0.040.81*

Great Recession vs. dot-com recession

0.100.250.400.984-0.600.80[1]

Pandemic vs. dot-com recession

0.160.240.660.932-0.520.83[1]

Pandemic vs. Great Recession

0.060.210.280.994-0.540.66[1]

Quits rate

Dot-com recession vs. none

0.260.102.640.075-0.020.53*

Great Recession vs. none

-0.220.08-2.720.062-0.450.01*

Pandemic vs. none

0.560.077.550.0000.350.77***

Great Recession vs. dot-com recession

-0.480.12-3.920.002-0.82-0.13***

Pandemic vs. dot-com recession

0.310.122.610.081-0.020.63*

Pandemic vs. Great Recession

0.780.107.470.0000.491.08***

[1] Not significant.

Note: *p < 0.10; **p < 0.05; ***p < 0.001. For this analysis, the dates of the three macroeconomic events of interest are March 2001 to March 2002 for the dot-com recession, December 2007 to June 2009 for the Great Recession, and March 2020 to January 2022 for the coronavirus disease 2019 (COVID-19) pandemic.

Source: Author's estimates based on data from the U.S. Bureau of Labor Statistics Job Openings and Labor Turnover Survey.

Quits rates in U.S. regions

In its JOLTS dataset, BLS organizes U.S. states and territories into four census regions: Northeast, South, West, and Midwest.36 This section explores differences in quits rates across these regions, testing two broad hypotheses: (1) in each region, the average quits rate during the COVID-19 pandemic was not statistically different from the quits rates during the dot-com recession and the Great Recession, and (2) the average quits rate during the pandemic was not statistically different across the four U.S. census regions.

The summary statistics for quits rates by region and macroeconomic event are presented in table 3. The table shows that the average quits rate during the pandemic was 2.78 percent in the South region, 2.49 percent in the Midwest region, 2.31 percent in the West region, and 1.83 percent in the Northeast region. Table 4 presents the test results for the two broad hypotheses. For the first hypothesis, the table shows that, in the Midwest region, the average quits rate during the pandemic differed statistically from the quits rates for the dot-com recession (p < 0.05) and the Great Recession (p < 0.001). The results are similar for the South region. For the Northeast and West regions, the average quits rate during the pandemic was not statistically different from the quits rate for the dot-com recession, but it was statistically different from the quits rate for the Great Recession (p < 0.001). The hypothesis that the average quits rates for the pandemic and the other macroeconomic events were equal was rejected in all cases for the Midwest and South regions, which suggests that the effect of the pandemic on quits rates in those regions was different from the effects of the two previous economic shocks. The results provide evidence that the pandemic produced statistically higher quits rates in all regions relative to the Great Recession and in two regions relative to the dot-com recession.

Table 3. Summary statistics for quits rates, by census region and macroeconomic event, December 2000–January 2022
Macroeconomic eventRegionMeanStandard deviationMinimumMaximum

None

Midwest1.870.311.102.60
Northeast1.500.230.802.00
South2.140.341.302.80
West1.960.341.102.60

Dot-com recession

Midwest2.160.261.802.50
Northeast1.800.151.602.00
South2.270.102.102.40
West2.160.211.902.60

Great Recession

Midwest1.620.261.201.90
Northeast1.350.241.001.80
South1.880.331.402.40
West1.780.291.202.20

Pandemic

Midwest2.490.411.603.10
Northeast1.830.321.302.30
South2.780.461.803.40
West2.310.461.402.90

Total

Midwest1.920.371.103.10
Northeast1.530.260.802.30
South2.180.401.303.40
West1.990.361.102.90

Note: For this analysis, the dates of the three macroeconomic events of interest are March 2001 to March 2002 for the dot-com recession, December 2007 to June 2009 for the Great Recession, and March 2020 to January 2022 for the coronavirus disease 2019 (COVID-19) pandemic.

Source: Author's estimates based on data from the U.S. Bureau of Labor Statistics Job Openings and Labor Turnover Survey.

Table 4. Pairwise macroeconomic-event and regional comparisons for quits rates, December 2000–January 2022
Macroeconomic-event and regional pairsContrastStandard errorScheffé tScheffé P > tSignificance

Midwest region

Dot-com recession vs. none

0.300.093.300.014**

Great Recession vs. none

-0.240.08-3.260.015**

Pandemic vs. none

0.630.079.080.000***

Great Recession vs. dot-com recession

-0.540.11-4.800.000***

Pandemic vs. dot-com recession

0.330.113.040.028**

Pandemic vs. Great Recession

0.870.108.970.000***

Northeast region

Dot-com recession vs. none

0.300.074.490.000***

Great Recession vs. none

-0.150.06-2.700.066*

Pandemic vs. none

0.340.056.510.000***

Great Recession vs. dot-com recession

-0.450.08-5.370.000***

Pandemic vs. dot-com recession

0.030.080.430.980[1]

Pandemic vs. Great Recession

0.490.076.710.000***

South region

Dot-com recession vs. none

0.130.101.340.616[1]

Great Recession vs. none

-0.250.08-3.040.028**

Pandemic vs. none

0.640.088.440.000***

Great Recession vs. dot-com recession

-0.390.12-3.100.024**

Pandemic vs. dot-com recession

0.510.124.250.001***

Pandemic vs. Great Recession

0.890.118.350.000***

West region

Dot-com recession vs. none

0.200.102.080.232[1]

Great Recession vs. none

-0.170.08-2.090.228[1]

Pandemic vs. none

0.350.084.640.000***

Great Recession vs. dot-com recession

-0.380.12-3.040.028**

Pandemic vs. dot-com recession

0.150.121.230.679[1]

Pandemic vs. Great Recession

0.520.114.910.000***

Pairwise regional comparisons

Midwest vs. Northeast

0.660.116.000.000***

Midwest vs. South

-0.290.13-2.220.032**

Midwest vs. West

0.180.131.420.163[1]

Northeast vs. South

-0.940.12-7.990.000***

Northeast vs. West

-0.470.12-4.040.000***

South vs. West

0.470.143.450.001***

[1] Not significant.

Note: *p < 0.10; **p < 0.05; ***p < 0.001. For this analysis, the dates of the three macroeconomic events of interest are March 2001 to March 2002 for the dot-com recession, December 2007 to June 2009 for the Great Recession, and March 2020 to January 2022 for the coronavirus disease 2019 (COVID-19) pandemic.

Source: Author's estimates based on data from the U.S. Bureau of Labor Statistics Job Openings and Labor Turnover Survey.

Table 4 also shows that, during the pandemic, the average quits rate for the Midwest region was higher than the quits rates for the Northeast and West regions (by 0.66 and 0.18 percentage point, respectively), but it was lower than the quits rate for the South region (by approximately 0.29 percentage point). The difference between the pandemic’s quits rate for the Midwest region and the quits rate for the Northeast region was positive and statistically significant (p < 0.001). The difference between the pandemic’s quits rate for the Midwest region and the quits rate for the South region was negative and statistically significant (p < 0.05). However, the difference in average quits rates between the Midwest and West regions was not statistically significant. Therefore, the hypothesis that the Midwest region’s average quits rate during the pandemic was equal to the quits rates of the Northeast and South regions is rejected, but the hypothesis concerning the pairwise comparison between the Midwest and West regions cannot be rejected. The results in table 4 also reject the hypothesis that the Northeast region’s average quits rate during the pandemic was equal to the quits rates of the South and West regions, and the same finding holds for the hypothesized equality of average quits rates with respect to the South and West regions.

Quit rates by firm size

Firm size class is defined by number of employees. Firms of size classes 1, 2, and 3 are those with, respectively, fewer than 10 employees, 10–49 employees, and 50–249 employees. Firms of size classes 4, 5, and 6 are those with, respectively, 250–999 employees, 1,000–4,999 employees, and 5,000 or more employees. Using the average employee number by size class between 2010 and 2021, one finds that, over that period, firms of size classes 4, 5, and 6 accounted for less than 1 percent of the total number of firms and 53.7 percent of the total number of employees.37 The monthly share of quits by size class is defined as the number of quits for each size class divided by the total number of quits in the nonfarm sector in each month. Chart 2 shows a positively skewed distribution of an increasing share of quits by size class. The distribution peaks at size class 3 for both the pandemic and the dot-com recession and at size class 2 for the “none” event. Firms of size classes 2 and 3 had the same average quits rates during the Great Recession. The chart confirms the findings of prior research indicating that firms of size class 6 tend to have lower quits rates, primarily because large firms are more likely to be unionized, have higher wages, and provide superior employment benefits.38

Table 5 shows that, for firms of size classes 4 and 5, the quits rate during the pandemic differed statistically from the quits rates of all other macroeconomic events (p < 0.001). For firms of size classes 1, 2, and 3, the pandemic’s quits rate did not differ statistically from the quits rate of the dot-com recession. For firms of size classes 2 and 3, the pandemic’s quit rate differed statistically from the Great Recession’s quits rate (p < 0.001). For firms of size class 6, there were no statistically significant quits-rate differences among the pandemic, the dot-com recession, and the Great Recession. The average pandemic quits rate for firms of size class 6 was statistically different from the rate for the “none” event, higher than the rate for the Great Recession, and lower than the rate for the dot-com recession. The same pattern holds for the average quits level for size class 6.

Table 5. Pairwise macroeconomic-event comparisons for quits rates, by firm size class, December 2000–January 2022
Macroeconomic-event pairContrastStandard errorScheffé tScheffé P > tSignificance

Size class 1 (fewer than 10 employees)

Dot-com recession vs. none

0.030.100.340.990[1]

Great Recession vs. none

-0.060.08-0.780.894[1]

Pandemic vs. none

0.210.082.770.056*

Great Recession vs. dot-com recession

-0.100.12-0.790.889[1]

Pandemic vs. dot-com recession

0.180.121.470.538[1]

Pandemic vs. Great Recession

0.270.112.570.088*

Size class 2 (10–49 employees)

Dot-com recession vs. none

0.230.112.070.234[1]

Great Recession vs. none

-0.320.09-3.430.009***

Pandemic vs. none

0.470.085.530.000***

Great Recession vs. dot-com recession

-0.550.14-3.940.002***

Pandemic vs. dot-com recession

0.240.131.800.357[1]

Pandemic vs. Great Recession

0.790.126.590.000***

Size class 3 (50–249 employees)

Dot-com recession vs. none

0.490.114.380.000***

Great Recession vs. none

-0.250.09-2.590.084*

Pandemic vs. none

0.620.097.150.000***

Great Recession vs. dot-com recession

-0.740.14-5.210.000***

Pandemic vs. dot-com recession

0.130.140.930.835[1]

Pandemic vs. Great Recession

0.870.127.090.000***

Size class 4 (250–999 employees)

Dot-com recession vs. none

0.160.111.560.490[1]

Great Recession vs. none

-0.250.09-2.810.051*

Pandemic vs. none

0.840.0810.300.000***

Great Recession vs. dot-com recession

-0.410.13-3.110.023**

Pandemic vs. dot-com recession

0.670.135.250.000***

Pandemic vs. Great Recession

1.090.119.490.000***

Size class 5 (1,000–4,999 employees)

Dot-com recession vs. none

0.060.090.720.916[1]

Great Recession vs. none

-0.140.07-1.970.278[1]

Pandemic vs. none

0.870.0713.090.000***

Great Recession vs. dot-com recession

-0.200.11-1.880.317[1]

Pandemic vs. dot-com recession

0.810.107.720.000***

Pandemic vs. Great Recession

1.010.0910.820.000***

Size class 6 (5,000 or more employees)

Dot-com recession vs. none

0.220.073.180.019**

Great Recession vs. none

0.010.060.150.999[1]

Pandemic vs. none

0.160.053.040.028**

Great Recession vs. dot-com recession

-0.210.09-2.430.120[1]

Pandemic vs. dot-com recession

-0.060.09-0.690.923[1]

Pandemic vs. Great Recession

0.160.082.040.247[1]

[1] Not significant.

Note: *p < 0.10; **p < 0.05; ***p < 0.001. For this analysis, the dates of the three macroeconomic events of interest are March 2001 to March 2002 for the dot-com recession, December 2007 to June 2009 for the Great Recession, and March 2020 to January 2022 for the coronavirus disease 2019 (COVID-19) pandemic.

Source: Author's estimates based on data from the U.S. Bureau of Labor Statistics Job Openings and Labor Turnover Survey.

While many organizations have implemented various forms of telework and remote work for years, the pandemic provided a different context for assessing these flexible work arrangements, affecting and giving voice to a broader group of workers and employers. Having no choice but to work at home, many employees had the time to assess their lifestyles and work–life balance. This introspection may have intensified with the large-scale death and disease unleased by the pandemic, engendering a new reality about work—one probably not seen since the genesis of Taylorism.39 Various employees—from those working on assembly lines and in cubicles to those occupying executive suites—had a chance, for the first time, to reflect on what they do, why they do it, and how they can balance their work–life choices given the pandemic’s existential realities.40 These considerations may explain why the pandemic has had a stronger effect on job quitting than the dot-com recession and the Great Recession, regardless of the level at which this effect is evaluated (national, regional, or firm size class).

However, the antecedents of higher quits levels and rates during the pandemic have also been influenced by practical constraints. For example, work exhaustion, fear of illness, and limited access to services such as childcare have contributed largely to the new situation.41 Data from the U.S. Census Bureau Household Pulse Survey (May 26 to June 7, 2021) show that nearly a quarter of adults in responding households took unpaid leave, cut work hours, or used vacation or sick days for childcare. About 17 percent of responding households had an adult who quit a job for childcare reasons, and another 8.4 percent had an adult who lost a job for the same reasons.42

Many frontline and other essential workers reported exhaustion from being overworked, a condition partly resulting from coworker sickness, exposure to COVID-19, or taking time off to care for loved ones.43 In all areas of essential work, replacing sick and deceased workers has become nearly impossible, even with increases in wages and signing bonuses.44 This exhaustion burden, it turns out, has caused frontline and emergency workers to become the single largest segment of workers quitting their jobs.45 For example, JOLTS data show that, during the pandemic, healthcare and social assistance workers quit their jobs at an average monthly rate of 2.71 percent, or 464,044 people per month. At the of end January 2022, more than 10.67 million healthcare and social assistance workers had voluntarily quit their jobs, compared with a total of 80.87 million workers for all industries combined. The share of total quits for this group of frontline workers during the pandemic was 13.20 percent, compared with 10.60 percent during the Great Recession and 9.24 percent during the dot-com recession. In addition to experiencing work burnout and exhaustion, many frontline workers now consider their compensation to be incommensurable with the risk in their workplace and the commitment being demanded from them during the pandemic.46

Hires, job openings, unemployment, wage and salary rates, and quits rates

Assessing whether the quits rate during the COVID-19 pandemic differed from the quits rates seen in prior macroeconomic disruptions was the primary motivation for this research. The results so far have shown that although the quits rate had been increasing since the end of the Great Recession, it accelerated soon after the onset of the pandemic. That the average quits rate during the pandemic differed statistically from the quits rates of earlier macroeconomic events (at the national and regional levels and by firm size class) confirms the Great Resignation phenomenon. This section identifies the antecedents of this phenomenon, providing insights into how it can be ameliorated.

Economies with flexible wages are susceptible to employees switching jobs when they receive superior offers from other employers.47 It is argued here that, during the pandemic, the “freedom” employees had away from their traditional workspaces afforded them the opportunity to consider their work situation and explore alternatives, including working for self. The analysis that follows assesses whether the antecedents of the quits rate behaved differently during the three macroeconomic events considered for this article. Using insights from the relevant literature, one can hypothesize that the quits rate in period t, qt, is determined by the hires rate, the unemployment rate, hourly earnings, and the job openings rate. It is conceivable that employees decide to quit their jobs by considering available information on these variables. The deployment of this information suggests a lag between the reception of the information and the quit decision. Experiments done for this article indicate that this lag consists of two 1-month periods (a two-period lag) for all independent variables, except for the categorical variable describing the macroeconomic events. The structural equation defining the problem is specified as follows:

qt = β0 + β1ht−2 + β2ut−2 + β3wt−2 + β4jt-2 + β5E1 + β6E2 + β7E3 + εt,

where h, u, w, and j represent, respectively, the hires rate, the unemployment rate, hourly earnings, and the job openings rate; E1, E2, and E3 represent the three macroeconomic events; and εt is the regression error term. The a priori expectation, based on the relevant literature and the foregoing statistical analyses, is that all coefficients are positive, except the coefficients on the unemployment rate and hourly earnings, which are predicted to be negative. This means that an increase in the hires rate is expected to increase competition among employers for working employees, leading to increasingly attractive offers from potential employers to entice workers from their current positions. Likewise, an increase in the job openings rate is expected to increase the pool of potential employers, signaling to current employees that the labor market is tightening and increasing their likelihood of quitting their current positions. The coefficient on the unemployment rate is hypothesized to be negative because an increasing unemployment rate signals a softening labor market, which encourages people to hold on to their current jobs. In addition, the coefficient on hourly earnings is hypothesized to be negative because increasing hourly earnings increase employees’ satisfaction with their current positions and reduce their likelihood of switching jobs.

While the hires rate and the job openings rate are taken directly from the JOLTS dataset, the monthly unemployment rate (from BLS) was obtained through the Federal Reserve Bank of St. Louis.48 Hourly earnings are total earnings divided by the total number of hours for which private sector employees received pay during a pay period, and include overtime pay and recurring cash compensation. Data on monthly hourly earnings from December 2000 through January 2022 are from the Organisation for Economic Co-operation and Development and were retrieved from the Federal Reserve Bank of St. Louis.49 All data are seasonally adjusted, rates are in percent, and hourly earnings are in U.S. dollars per hour.

To address any inherent serial correlation in the data, the model estimation used the Prais-Winsten procedure of Stata (version 17.0).50 A VIF (variance inflation factor) test indicated the absence of multicollinearity in the model. However, a Ramsey RESET test suggested the presence of omitted variables (p < 0.001), an expected result given that the quits rate is associated with and determined by additional variables not included in the model. Stata’s link test, which tests the hypothesis that a properly specified model should not find additional predictors except by chance, indicated the absence of misspecification (_hat (p < 0.001); _hatsq (p < 0.174); _const (p < 0.185)).

The regression results, presented in table 6, show a coefficient of determination of 0.896 and an F-value (7, 243) of 322.40 (p < 0.001). The Durbin-Watson statistic (transformed) is 2.01, suggesting the absence of autocorrelation. The regression results show that all coefficients have the theoretically expected signs. Holding all other variables constant, a unit increase in the unemployment rate decreases the quits rate by about 0.08 percentage point (p < 0.001), whereas a unit increase in the two-period lagged job openings rate increases the quits rate by approximately 0.21 percentage point (p < 0.001). In addition, a unit increase in the two-period lagged hires rate is associated with a 0.10-percentage-point increase in the quits rate (p < 0.007), and a $1.00 increase in hourly earnings is associated with a 0.03-percentage-point decrease in the quits rate (p < 0.001). The results also show that the categorical variable for the COVID-19 pandemic shifted the intercept up by more than 0.3 percentage point (p < 0.056), producing a quits rate of 2.42 percent.

Table 6. Regression results for quits rate response to two-period lagged hires rate, job openings rate, unemployment rate, and hourly earnings
VariableCoefficientSemirobust standard errortP > t95-percent confidence interval
Lower boundUpper bound

Hires rate

0.100.042.710.0070.030.18

Unemployment rate

-0.080.02-4.400.000-0.12-0.05

Hourly earnings

-0.030.01-4.630.000-0.04-0.02

Job openings rate

0.210.054.510.0000.120.30

Dot-com recession

0.010.050.150.881-0.090.10

Great Recession

-0.030.03-1.160.248-0.080.02

Pandemic

0.300.161.920.056-0.010.62

Constant

2.130.316.820.0001.512.74

Note: N = 251; F(7, 243) = 322.40; Prob > F = 0.000; R-squared = 0.869; Root mean squared error = 0.093; rho = 0.269; Durbin-Watson statistic (original) = 1.520; Durbin-Watson statistic (transformed) = 2.01. For this analysis, the dates of the three macroeconomic events of interest are March 2001 to March 2002 for the dot-com recession, December 2007 to June 2009 for the Great Recession, and March 2020 to January 2022 for the coronavirus disease 2019 (COVID-19) pandemic.

Source: Author's estimates based on data from the U.S. Bureau of Labor Statistics Job Openings and Labor Turnover Survey.

The elasticities associated with the explanatory variables provide more intuitive insight into the sensitivity of the quits rate to changes to those variables. The results indicate that, holding all other variables constant, a 1-percent increase in the two-period lagged hires rate increases the quits rate by about 0.19 percent (p < 0.007). Similarly, a 1-percent increase in the two-period lagged job openings rate increases the quits rate by nearly 0.34 percent (p < 0.001). Notably, a 1-percent increase in hourly earnings reduces the quits rate by 0.28 percent. This result suggests that employers are likely to succeed in reducing the quits rate by increasing hourly earnings, the only variable employers can manipulate directly. By increasing hourly earnings, employers increase the switching costs of current employees who may receive offers from other employers or who are considering self-employment. The hourly earnings elasticity of the quits rate for the pandemic period is −0.3 percent (p < 0.001), indicating a small absolute increase of about 6.4 percent from the estimated hourly earnings elasticity for the overall model. However, the job openings elasticity of the quits rate for the pandemic period is 0.43 percent, a 26.4-percent increase from the overall model’s value of the same indicator. When estimated specifically for the pandemic period, the elasticities for the hires rate and the unemployment rate decrease by 5.04 percent and 7.49 percent, respectively. The foregoing results provide further evidence that the quits rate behaved differently during the pandemic than in prior macroeconomic events.

Conclusion

This article sought to verify the Great Resignation phenomenon that unfolded in the United States during the COVID-19 pandemic. After controlling for economic growth, the analysis showed that the levels and rates of quits during the pandemic (up to January 2022) were statistically different from those seen in the Great Recession and the dot-com recession. The higher levels of quits during the pandemic were established for the whole economy, for the four U.S. census regions, and by firm size class. A regression analysis assessing the collective effect of the hires rate, the job openings rate, hourly earnings, and the unemployment rate on the quits rate revealed regression coefficients that were statistically significant and carried theoretically expected signs.

Given the empirical veracity of the Great Resignation (relative to prior macroeconomic disruptions), a continuing pandemic pressure (at sustained or higher levels) on the wage-to-profitability ratio may motivate accelerated investments in labor-saving solutions, such as those driven by technology. Presumably having become sensitized to the possibility of another pandemic in the future, public and business leaders may increase their investments in labor-saving technologies that would prevent the recurrence of economic impacts similar to those of the current pandemic.51 Therefore, organizations need to reimagine work and redesign the workplace to accommodate the postpandemic employee in a way that meets their risk management and performance objectives.

Suggested citation:

Vincent Amanor-Boadu, "Empirical evidence for the “Great Resignation”," Monthly Labor Review, U.S. Bureau of Labor Statistics, November 2022, https://doi.org/10.21916/mlr.2022.29

Notes


1 “Guidance on the essential critical infrastructure workforce: ensuring community and national resilience in COVID-19 response” (U.S. Department of Homeland Security, March 2020).

2 Arianne Cohen, “How to quit your job in the great post-pandemic resignation boom,” Bloomberg Businessweek, May 10, 2021, https://www.bloomberg.com/news/articles/2021-05-10/quit-your-job-how-to-resign-after-covid-pandemic?leadSource=uverify%20wall.

3 Arthur Woods, “The Great Resignation doesn’t have to threaten your DE&I efforts,” Harvard Business Review, September 1, 2021, https://hbr.org/2021/09/the-great-resignation-doesnt-have-to-threaten-your-dei-efforts.

4 Ian Cook, “Who is driving the Great Resignation?” Harvard Business Review, September 15, 2021, https://hbr.org/2021/09/who-is-driving-the-great-resignation.

5 Jack Kelly, “Workers are quitting their jobs in record numbers, as the U.S. experiences a booming job market,” Forbes, June 14, 2021, https://www.forbes.com/sites/jackkelly/2021/06/14/workers-are-quitting-their-jobs-in-record-numbers-as-the-us-experiences-a-booming-job-market/.

6 Future of work (Public Broadcasting Service, 2021).

7 Lawrence S. Essien and Montgomery McCarthy, “Job openings, hires, and quits reach historic highs in 2018,” Monthly Labor Review, July 2019, https://doi.org/10.21916/mlr.2019.18.

8 Amy E. Hurley-Hanson, Cristina M. Giannantonio, Heidi Carlos, Jessica Harnett, and Melanie Jetta, “The effect of the attack of 9/11 on organizational policies, employee attitudes and workers’ psychological states,” American Journal of Economics and Business Administration, vol. 3, no. 2, February 2011, pp. 377–389.

9 Miriam K. Forbes and Robert F. Krueger, “The Great Recession and mental health in the United States,” Clinical Psychological Science, vol. 7, no. 5, 2019, pp. 900–913.

10 Huynju Kang, Jaevin Park, and Hyunduk Suh, “The rise of part-time employment in the Great Recession: its causes and macroeconomic effects,” Journal of Macroeconomics, vol. 66, no. 2, December 2020.

11 Mark A. Schuster, Bradley D. Stein, Lisa H. Jaycox, Rebecca L. Collins, Grant N. Marshall, Marc N. Elliott, Annie J. Zhou, David E. Kanouse, Janina L. Morrison, and Sandra H. Berry, “A national survey of stress reactions after the September 11, 2001, terrorist attacks,” The New England Journal of Medicine, vol. 345, no. 20, November 2001, pp. 1507–1512.

12 Cynthia Miller-Idriss, “From 9/11 to 1/6: the War on Terror supercharged the far right,” Foreign Affairs, September/October 2021, https://www.foreignaffairs.com/articles/united-states/2021-08-24/war-on-terror-911-jan6; and Muhammad Safeer Awan, “Global terror and the rise of xenophobia/Islamophobia: an analysis of American cultural production since September 11,” Islamic Studies, vol. 49, no. 4, winter 2010, pp. 521–537.

13 For more information on SARS, see “Severe acute respiratory syndrome (SARS),” Health Topics (World Health Organization, 2021), https://www.who.int/health-topics/severe-acute-respiratory-syndrome#tab=tab_1; and “The disease,” Frequently Asked Questions (Centers for Disease Control and Prevention, 2005), https://www.cdc.gov/sars/about/faq.html.

14 For more information on MERS, see “Middle East respiratory syndrome coronavirus (MERS-CoV),” Health Topics (World Health Organization, 2021), https://www.who.int/health-topics/middle-east-respiratory-syndrome-coronavirus-mers#tab=tab_1.

15 “COVID-19 dashboard” (Johns Hopkins University and Medicine, Johns Hopkins Coronavirus Resource Center, 2022), https://coronavirus.jhu.edu/.

16 Ibid.

17 Job Openings and Labor Turnover Survey data are available at https://data.bls.gov/cgi-bin/dsrv?jt.

18 “Chapter 18. Job Openings and Labor Turnover Survey,” Handbook of Methods (U.S. Bureau of Labor Statistics), https://www.bls.gov/opub/hom/pdf/jlt-20130314.pdf.

19 “Geographic information” (U.S. Bureau of Labor Statistics), https://www.bls.gov/regions/.

20 Kenneth J. McLaughlin, “A theory of quits and layoffs with efficient turnover,” Journal of Political Economy, vol. 99, no. 1, February 1991, pp. 1–29.

21 Tae Heon Lee, Barry Gerhart, Ingo Weller, and Charlie O. Trevor, “Understanding voluntary turnover: path-specific job satisfaction effects and the importance of unsolicited job offers,” The Academy of Management Journal, vol. 51, no. 4, August 2008, pp. 651–671; and Loren M. Solnick, “Promotions, pay, performance ratings and quits,” Eastern Economic Journal, vol. 14, no. 1, January–March 1988, pp. 51–62.

22 Demetrio Scopelliti, “Job switching: a prelude to wage growth?” Monthly Labor Review, October 2015, https://www.bls.gov/opub/mlr/2015/beyond-bls/job-switching-a-prelude-to-wage-growth.htm; and J. Peter Mattila, “Job quitting and frictional unemployment,” American Economic Review, vol. 64, no. 1, March 1974, pp. 235–239.

23 Jason Faberman and Alejandro Justiniano, “Job switching and wage growth,” Chicago Fed Letter, vol. 337 (Federal Reserve Bank of Chicago, 2015), https://www.chicagofed.org/publications/chicago-fed-letter/2015/337.

24 George A. Akerlof, Andrew K. Rose, and Janet L. Yellen, “Job switching and job satisfaction in the U.S. labor market,” Brookings Papers on Economic Activity, vol. 2, 1988, pp. 495–594, https://www.brookings.edu/wp-content/uploads/1988/06/1988b_bpea_akerlof_rose_yellen_ball_hall.pdf.

25 Arindrajit Dube, Laura Giuliano, and Jonathan Leonard, “Fairness and frictions: the impact of unequal raises on quit behavior,” American Economic Review, vol. 109, no. 2, February 2019, pp. 620–663; Maury Gittleman, “Declining labor turnover in the United States: evidence and implications from the Panel Study of Income Dynamics,” Monthly Labor Review, January 2019, https://doi.org/10.21916/mlr.2019.1; Melvyn G. Coles and Dale T. Mortensen, “Equilibrium labor turnover, firm growth, and unemployment,” Econometrica, vol. 84, no. 1, January 2016, pp. 347–363; Michael W. L. Elsby, Ryan Michaels, and Gary Solon, “The ins and outs of cyclical unemployment,” American Economic Journal: Macroeconomics, vol. 1, no. 1, January 2009, pp. 84–110; Will Felps, Terence R. Mitchell, David R. Herman, Thomas W. Lee, Brooks C. Holtom, and Wendy S. Harman, “Turnover contagion: how coworkers’ job embeddedness and job search behaviors influence quitting,” Academy of Management Journal, vol. 52, no. 3, June 2009, pp. 545–561; and Paul Sullivan and Ted To, “Search and nonwage job characteristics,” Journal of Human Resources, vol. 49, no. 2, 2014, pp. 472–507.

26 Lee et al., “Understanding voluntary turnover.”

27 Thomas W. Lee, Terence R. Mitchell, Lowell Wise, and Steven Fireman, “An unfolding model of voluntary employee turnover,” Academy of Management Journal, vol. 39, no. 1, February 1996, pp. 5–36.

28 Jerry Carnes, “Why are so many people leaving their jobs during the ‘Great Resignation’?” 11alive.com, September 15, 2021, https://www.11alive.com/article/news/local/outreach/why-guy/great-resignation-of-2021-why-people-quit-their-jobs/85-8af4d53b-1270-412f-9e0e-1ce70b5170a1.

29 Lee et al., “Understanding voluntary turnover”; and “Business applications: total for all NAICS in the United States” (FRED, Federal Reserve Bank of St. Louis, September 14, 2022), https://fred.stlouisfed.org/series/BABATOTALSAUS.

30 Mitra Toossi, “A behavioral model for projecting the labor force participation rate,” Monthly Labor Review, May 2011, pp. 25–43, https://www.bls.gov/opub/mlr/2011/05/art3full.pdf.

31 Cohen, “How to quit your job.”

32 Ryan Ansell and John P. Mullins, “COVID-19 ends longest employment recovery and expansion in CES history, causing unprecedented job losses in 2020,” Monthly Labor Review, June 2021, https://doi.org/10.21916/mlr.2021.13.

33 Fatema Turkistani and Aseel Bin Sawad, “Fighting COVID-19 pandemic: analysis of the Swedish model vs. lockdown model,” International Journal of Pharma and Bio Sciences, vol. 11, no. 4, July 2021, pp. 70–76.

34 “Employment, hours, and earnings from the Current Employment Statistics survey (national)” (U.S. Bureau of Labor Statistics, 2021).

35 See “Table A-1. Current and real (constant 1982–1984 dollars) earnings for all employees on private nonfarm payrolls, seasonally adjusted” in Real earnings—September 2022, USDL-22-1991 (U.S. Department of Labor, September 2022), https://www.bls.gov/news.release/pdf/realer.pdf.

36 “Geographic information” (U.S. Bureau of Labor Statistics), https://www.bls.gov/regions/.

37 “Table F. Distribution of private sector employment by firm size class: 1993/Q1 through 2021/Q1, not seasonally adjusted” (U.S. Bureau of Labor Statistics), https://www.bls.gov/web/cewbd/table_f.txt.

38 Solnick, “Promotions, pay, performance ratings and quits”; Charles Brown and James Medoff, “The employer size-wage effect,” Journal of Political Economy, vol. 97, no. 5, 1989, pp. 1027–1059; and Peter A. Groothuis, “Turnover: the implication of establishment size and unionization,” Quarterly Journal of Business and Economics, vol. 33, no. 2, spring 1994, pp. 41–53.

39 Omar Aktouf, “Management and theories of organizations in the 1990s: toward a critical radical humanism?” Academy of Management Review, vol. 17, no. 3, July 1992, pp. 407–431; Maurice de Montmollin, “Taylorism and anti-Taylorism,” International Studies of Management and Organization, vol. 5, no. 3, 1975, pp. 4–15; and Sigmund Wagner-Tsukamoto, “An institutional economic reconstruction of scientific management: on the lost theoretical logic of Taylorism,” Academy of Management Review, vol. 32, no. 1, January 2007, pp. 105–117.

40 Paul Sullivan, “How the pandemic has changed attitudes toward wealth,” The New York Times, May 21, 2021, https://www.nytimes.com/2021/05/21/your-money/wealth-attitudes-pandemic.html; and “CEOs are joining the ‘Great Resignation,’” CNN Business, November 18, 2021, https://www.cnn.com/2021/11/18/investing/ceo-turnover/index.html.

41 “Week 31 Household Pulse Survey: May 26–June 7” (U.S. Census Bureau, June 16, 2021), https://www.census.gov/data/tables/2021/demo/hhp/hhp31.html; Gwen Moran, “6 ways to reduce anxiety about going back to the office,” Fast Company, June 2, 2021, https://www.fastcompany.com/90641007/6-ways-to-reduce-anxiety-about-going-back-to-the-office; and Sara Berg, “What doctors wish patients knew about pandemic fatigue” (Chicago, IL: American Medical Association, January 29, 2021), https://www.ama-assn.org/delivering-care/public-health/what-doctors-wish-patients-knew-about-pandemic-fatigue.

42 “Week 31 Household Pulse Survey: May 26–June 7” (U.S. Census Bureau, June 16, 2021).

43 “Essential workers more likely to be diagnosed with a mental health disorder during pandemic” (Washington, DC: American Psychological Association, March 11, 2021), https://www.apa.org/news/press/releases/stress/2021/one-year-pandemic-stress-essential.

44 Adam K. Raymond, “‘It’s not worth it to me:’ how the pandemic is changing the jobs of essential workers,” Spectrum News, July 16, 2021, https://spectrumnews1.com/ky/louisville/news/2021/07/16/how-the-pandemic-changed-the-jobs-of-essential-workers.

45 Brad Tuttle, “Workers are quitting these 4 kinds of jobs in droves,” Money, October 15, 2021, https://money.com/kinds-of-jobs-workers-are-quitting/.

46 Raymond, “‘It’s not worth it to me’”; and Tuttle, “Workers are quitting these 4 kinds of jobs in droves.”

47 Coles and Mortensen, “Equilibrium labor turnover”; and McLaughlin, “A theory of quits and layoffs.”

48 “Unemployment rate” (FRED, Federal Reserve Bank of St. Louis, October 7, 2022), https://fred.stlouisfed.org/series/UNRATE.

49 “Hourly earnings: private sector for the United States” (FRED, Federal Reserve Bank of St. Louis, September 14, 2022), https://fred.stlouisfed.org/series/LCEAPR01USM189S.

50 Stata 17.0 standard edition (Stata Corporation, 2021).

51 For example, a SnapLogic survey indicated that, because of the pandemic, 78 percent of IT leaders in the United States and the United Kingdom plan to increase spending on automation initiatives focusing on employee productivity and cost savings. See Scott Behles, Cameron Burke, and Jeff Drew, “78% of IT leaders plan to increase spending on automation initiatives post-pandemic,” BusinessWire, March 30, 2021, https://www.businesswire.com/news/home/20210330005284/en/78-of-IT-Leaders-Plan-to-Increase-Spending-on-Automation-Initiatives-Post-Pandemic.

article image
About the Author

Vincent Amanor-Boadu
vincent@ksu.edu

Vincent Amanor-Boadu is a professor in the Department of Agricultural Economics at Kansas State University, Manhattan, Kansas.

close or Esc Key