Occupational Segregation and COVID-19: A Sociological Analysis of Racial and Gender Disparities in Job Loss and Recovery

Occupational Segregation and COVID-19: A Sociological Analysis of Racial and Gender Disparities in Job Loss and Recovery | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Occupational Segregation and COVID-19: A Sociological Analysis of Racial and Gender Disparities in Job Loss and Recovery Jeremy Bennett This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6915836/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The COVID-19 pandemic triggered an unprecedented economic upheaval that magnified pre-existing inequalities in the U.S. labor market. This study examines how occupational segregation by race and gender contributed to disparate job losses and shaped the trajectory of employment recovery. Drawing on Intersectionality Theory and Dual Labor Market Theory as a theoretical framework, we conduct an interdisciplinary analysis blending sociology and labor economics perspectives. We integrate an extensive literature review with original analyses of secondary data (Current Population Survey, Bureau of Labor Statistics reports, and American Community Survey) to assess racial and gender disparities in unemployment, labor force exits, and rehiring during the pandemic. Consistent with prior research, we find that women and workers of color experienced disproportionately severe job losses in the spring of 2020, reflecting their overrepresentation in vulnerable, low-wage service occupations​. Intersectional vulnerabilities were especially evident for women of color, who faced compounded disadvantages in both job displacement and caregiving burdens. As the labor market rebounded, these groups saw employment recover more slowly, though strong fiscal stimulus eventually facilitated a return to pre-pandemic employment levels for many. Our findings underscore how structural inequalities—rooted in segregated labor markets and intersecting axes of oppression—produced unequal outcomes in the COVID-19 era, and we discuss policy implications for a more equitable recovery. COVID-19 occupational segregation intersectionality dual labor market unemployment gender disparities racial disparities Introduction The COVID-19 pandemic triggered the sharpest economic downturn in the United States since the Great Depression (Kochhar, 2020 ). Within weeks, tens of millions of jobs were lost as businesses shuttered and economic activity drastically declined due to mandated lockdowns and social distancing measures (Boesch, Nunn, & Tchourumoff, 2022 ; Kochhar, 2020 ). Importantly, this labor market crisis exhibited pronounced disparities across demographic groups, with early data indicating that women and racial/ethnic minorities experienced disproportionately severe employment disruptions compared to previous recessions, which typically impacted male-dominated industries more heavily (Alon, Doepke, Olmstead-Rumsey, & Tertilt, 2020 ; Bureau of Labor Statistics [BLS], 2020a). By April 2020, the U.S. unemployment rate surged to 14.7%, with significant disparities evident by gender and race. Women faced higher unemployment rates (15.5%) compared to men (13.0%), and unemployment among Black (16.7%) and Hispanic (18.9%) workers substantially exceeded that of White workers (BLS, 2020a). These disparities intensified by May 2020, when Hispanic women's unemployment peaked at 19.5%, the highest rate recorded among any major demographic group (Kochhar, 2020 ). Popular media and scholarly attention described this phenomenon as a "she-cession," emphasizing the particular vulnerability faced by women and workers of color during the pandemic-driven economic contraction (Bateman & Ross, 2020 ; BLS, 2020b). These unequal outcomes did not occur randomly; rather, they were deeply embedded in longstanding structural inequalities and occupational segregation patterns. Women and racial minorities in the U.S. labor force are disproportionately concentrated in lower-paying service, hospitality, and caregiving roles that were either classified as "essential"—thereby increasing their health risks—or "non-essential"—subjecting them to widespread closures and layoffs during lockdown periods (Bennett, 2018 ; Holder, Jones, & Masterson, 2021 ; Masterson, Holder, & Jones, 2020 ; U.S. Census Bureau, 2021 ). Conversely, many higher-paying professional occupations, disproportionately held by White men, swiftly transitioned to remote work arrangements, effectively shielding these employees from substantial job losses (U.S. Census Bureau, 2022 ). Consequently, the pandemic illuminated the dual structure of the labor market, a concept extensively discussed in Dual Labor Market Theory, and highlighted how intersections of race, gender, and class significantly shaped patterns of economic vulnerability (Flood & Moen, 2021 ; Montenovo et al., 2020 ). This paper offers a comprehensive sociological analysis of racial and gender disparities in job loss and recovery during the COVID-19 era, focusing on the United States. We integrate theoretical insights from Intersectionality Theory and Dual Labor Market Theory to explain why the employment fallout of the pandemic was so stratified. Using publicly available data from the Current Population Survey (CPS), the Bureau of Labor Statistics (BLS), and the American Community Survey (ACS), we document the extent of job losses in 2020 and analyze the differential patterns of labor market recovery through 2022. An extensive review of recent scholarship from leading journals in sociology, economics, and business informs our theoretical and methodological approach. Our contribution lies in empirically linking occupational segregation to the pandemic’s uneven impacts and in quantifying how intersecting identities—particularly race and gender—shaped vulnerability to job loss and access to recovery. Theoretical Framework Intersectionality Theory Intersectionality Theory, introduced by Crenshaw ( 1989 ), argues that multiple social identities such as race, gender, and class intersect in ways that produce unique patterns of disadvantage and privilege. Rather than viewing categories like gender or race in isolation, intersectionality emphasizes that combined identities—such as being both Black and a woman—create specific forms of discrimination and systemic inequality not experienced by individuals with a single marginalized identity (Collins, 2000 ; Crenshaw, 1989 ). Since its development, intersectionality has become a foundational perspective within sociology and related disciplines, guiding research on how overlapping systems of oppression shape individual life outcomes, including opportunities and constraints in labor markets (Collins & Bilge, 2016 ). In labor market research, an intersectional approach highlights how race- and gender-based inequalities interact to produce compounded disadvantages for particular groups of workers, especially women of color (Holder, Jones, & Masterson, 2021 ; Moen, Pedtke, & Flood, 2020 ). Intersectionality theory predicts that economic shocks, like the COVID-19 pandemic, will not uniformly impact all members of marginalized groups; instead, those positioned at multiple intersections of disadvantage—such as low-income Black or Latina women—will likely face the harshest consequences (Masterson, Holder, & Jones, 2020 ). Early evidence from the pandemic confirms this prediction, demonstrating that Black and Latina women were disproportionately affected due to their higher representation in precarious, low-wage service jobs and greater caregiving responsibilities amid school closures and disruptions to childcare (Bateman & Ross, 2020 ; Kochhar, 2020 ). Indeed, Hispanic women's unemployment peaked dramatically at 19.5% in May 2020, significantly higher than most other demographic groups (Kochhar, 2020 ). Moreover, intersectionality illuminates nuances that single-identity approaches miss. For instance, Asian women's unemployment, historically lower than other groups, spiked sharply during the pandemic—nearly reaching parity with Black women's unemployment rate (16.7% vs. 17.2%) in May 2020—highlighting how even racially advantaged women became vulnerable through gendered occupational segregation (Kochhar, 2020 ). An intersectional lens also guides analysis of recovery trajectories following the immediate crisis. Women of color historically encounter heightened barriers when re-entering the workforce post-recession, attributable to continued discrimination, limited social capital, and concentration in industries slower to rebound (Boesch, Nunn, & Tchourumoff, 2022 ). Moreover, sustained childcare disruptions further complicate employment recovery for working mothers, disproportionately impacting Black and Hispanic women due to less flexible working arrangements (Petts, Carlson, & Pepin, 2021 ). Thus, intersectionality leads us to hypothesize slower and uneven recovery for workers experiencing multiple intersecting disadvantages, a proposition we explicitly test in our empirical analysis. Dual Labor Market Theory Dual Labor Market Theory offers a complementary structural explanation for persistent inequalities within labor markets. Originating from the seminal work of Doeringer and Piore ( 1971 ), this theory posits that labor markets are segmented into two primary sectors: the primary sector, characterized by stable employment, relatively high wages, benefits, and opportunities for advancement; and the secondary sector, marked by precarious employment, low wages, minimal benefits, and limited career mobility (Dickens & Lang, 1985 ; Reich, Gordon, & Edwards, 1973 ). These structural divisions are maintained through institutional mechanisms and discriminatory practices, systematically channeling women, racial minorities, and other marginalized groups into less secure secondary-sector employment (Reid & Rubin, 2003 ). Applied to the COVID-19 crisis, Dual Labor Market Theory suggests that the pandemic’s economic disruptions were sharply concentrated within secondary-sector occupations—such as hospitality, retail, personal care services, and other low-wage, face-to-face service roles—which disproportionately employ women and racial/ethnic minority workers (Albanesi & Kim, 2021 ; Montenovo et al., 2020 ). In contrast, primary-sector employment in fields like finance, technology, and higher education generally proved resilient due to the possibility of remote work arrangements and financial buffers that allowed job retention even amidst economic turmoil (Groshen, 2020 ; U.S. Census Bureau, 2022 ). Occupational segregation—defined as the systematic distribution of demographic groups across different jobs and sectors—emerges as a key explanatory mechanism behind observed disparities in job losses. Before the pandemic, women and racial minorities disproportionately occupied secondary-sector jobs characterized by low pay and instability (Cortes & Forsythe, 2023 ). The sudden collapse of service industries due to public health restrictions therefore resulted in disproportionately severe job losses among these demographic groups (Boesch et al., 2022 ). For instance, research by Boesch and colleagues ( 2022 ) quantifies how occupational segregation alone accounted for significant portions of employment declines among minority groups. Hispanic workers, for instance, faced an excess employment decline of approximately 2.8 percentage points in early 2020, specifically attributable to their concentration in vulnerable occupations such as hospitality and personal care (Boesch et al., 2022 ). Dual Labor Market Theory predicts differential recovery patterns between primary and secondary sectors. While primary-sector occupations typically rebound quickly following economic shocks—sometimes even experiencing employment growth—secondary-sector jobs are more likely to be permanently eliminated or slower to recover (Groshen, 2020 ; U.S. Census Bureau, 2022 ). This generates persistent employment gaps for women and minorities, who remain disproportionately positioned in precarious sectors, reinforcing pre-existing inequalities. We thus anticipate observing a bifurcated or "two-speed" recovery, with primary-sector workers quickly regaining employment and secondary-sector workers, predominantly minorities and women, experiencing prolonged labor market disadvantages (Boesch et al., 2022 ). Summary and Theoretical Expectations Integrating Intersectionality and Dual Labor Market theories provides a robust theoretical foundation for understanding pandemic-related labor market disparities. Intersectionality highlights compounded disadvantages experienced by groups at the intersections of race, gender, and class, directing attention toward nuanced, intra-group variations in economic vulnerability. Dual Labor Market Theory offers a structural explanation emphasizing how labor market segmentation systematically funnels marginalized groups into vulnerable secondary-sector jobs, amplifying their susceptibility during economic disruptions. Together, these theories guide our empirical analysis by shaping the following expectations: Employment impacts from COVID-19 will disproportionately affect workers in secondary-sector occupations, notably women and racial minorities. Within these groups, individuals at the intersection of multiple disadvantaged identities (e.g., low-income women of color) will face particularly severe economic consequences. Recovery trajectories post-pandemic will remain uneven, with persistent gaps reflecting enduring structural inequalities and compounded intersectional vulnerabilities. Literature Review Gender Disparities in Pandemic Job Losses The COVID-19 economic crisis significantly exacerbated gender disparities in employment globally, with women disproportionately affected by job losses compared to men (Madgavkar et al., 2020). In the United States, this gender disparity emerged prominently in the earliest months of the pandemic. Between February and April 2020, female workers experienced a net loss of approximately 13.5 million jobs, about 1.5 million more than their male counterparts (Gould & Kassa, 2020). By May 2020, women's unemployment peaked at 14.3%, notably higher than the 11.9% unemployment rate among men, marking the first major recession in which female unemployment significantly surpassed male unemployment (Kochhar, 2020 ). Several studies have addressed the reasons behind this pronounced "she-cession." A pivotal factor identified is the occupational distribution of female employment. Female workers are heavily concentrated in sectors such as hospitality, food service, retail, education, healthcare support, and personal care, all of which faced significant contractions due to COVID-19-related social distancing mandates (Alon et al., 2020 ; Couch, Fairlie, & Xu, 2020). Unlike previous economic downturns—such as the 2008 recession, which primarily impacted male-dominated sectors like manufacturing and construction—the pandemic's impact centered on industries with higher proportions of female employment (Alon et al., 2020 ). Specifically, in April 2020 alone, the Bureau of Labor Statistics (2020) reported that approximately 3.5 million jobs were lost within the leisure and hospitality industry, disproportionately affecting women who constitute a majority within this sector. Additional occupational data indicate sharp employment reductions among women-dominated roles: maids and housecleaners (-31%), waitresses (-39%), hair stylists (-35%), and childcare workers (-23%) between 2019 and 2021 (U.S. Census Bureau, 2022 ). These employment declines vividly illustrate how occupational gender segregation within low-wage, service-oriented work led to outsized job losses among women (Alon et al., 2020 ; Blau, Koebe, & Meyerhofer, 2021). Another significant contributor to gender disparities during the pandemic was the unequal distribution of caregiving responsibilities, exacerbated by the closure of schools and childcare facilities. Women, especially mothers, faced increased pressure to reduce working hours or leave the workforce entirely to manage childcare demands (Collins, Landivar, Ruppanner, & Scarborough, 2021 ; Landivar et al., 2020 ). Research indicates that women with young children were disproportionately likely to exit the labor force during 2020 compared to comparable workers without caregiving responsibilities (Bateman & Ross, 2020 ; Petts, Carlson, & Pepin, 2021 ). The Federal Reserve analysis underscored this point, revealing a notably higher rate of labor force exits among women with children under six due to intensified caregiving demands and insufficient childcare access (Lofton, Petrosky-Nadeau, & Seitelman, 2021). This caregiving dynamic compounded the direct employment impact of layoffs, indicating that even women whose positions were not eliminated were disproportionately likely to voluntarily leave the workforce or take extended leave, thus widening the gender gap in labor force participation throughout 2020 (Kochhar, 2020 ). The cumulative impact of these dynamics resulted in a sharper decline in employment among women compared to men during the pandemic's initial phase. Between 2019 and 2021, the overall number of women employed full-time, year-round dropped by 3.4%, versus a 4.1% decline among men; however, the timing and depth of unemployment notably differed. Men’s job losses were substantial, particularly among Black and Hispanic males, but spread across multiple sectors such as manufacturing, construction, and transportation (U.S. Census Bureau, 2022 ). Conversely, women's job losses were concentrated in select service-oriented sectors, amplifying their initial unemployment shock (Alon et al., 2020 ; Blau et al., 2021). As the U.S. economy began to reopen in late 2020 through 2021, gender disparities in employment persisted, though gradually diminishing. Women remained disproportionately likely to have exited the labor force through 2021, often citing pandemic-related care responsibilities. Research from the Center for American Progress indicated approximately 1.4 million fewer mothers were participating in the labor force by late 2021 compared to pre-pandemic levels, highlighting ongoing care-related employment barriers (Bateman & Ross, 2020 ). Nevertheless, a robust recovery occurred in 2021–2022, driven by extensive fiscal stimulus measures and improvements in childcare access, facilitating women's gradual return to employment. By early 2023, overall women's employment returned to pre-pandemic levels, with prime-age women (ages 25–54) achieving record-high labor force participation rates of approximately 77.0%, surpassing pre-pandemic benchmarks (Bateman & Ross, 2023). In sum, the literature substantiates significant but context-dependent gender disparities in pandemic-related labor market outcomes, primarily driven by occupational segregation into vulnerable sectors and compounded by unequal caregiving responsibilities (Alon et al., 2020 ; Blau et al., 2021; Collins et al., 2021 ). These disparities align with the conceptual frameworks provided by Dual Labor Market Theory, emphasizing women's concentration in secondary-sector occupations, and Intersectionality Theory, highlighting the compounding effects of gendered and racialized employment vulnerabilities. Although gender employment gaps narrowed substantially during the recovery, the initial shock's severity underscores potential long-term repercussions for women's earnings trajectories, career advancement, and economic stability, particularly when considering intersecting identities of race and socioeconomic status (Couch et al., 2020; Collins et al., 2021 ). Racial/Ethnic Disparities in Pandemic Job Losses The COVID-19 pandemic significantly highlighted racial and ethnic inequalities within the U.S. labor market. Workers of color entered the pandemic already experiencing higher rates of unemployment and greater job insecurity compared to white workers, even during the strong economic conditions of 2019 (Boesch, Nunn, & Tchourumoff, 2022 ; Cortes & Forsythe, 2023 ). As the pandemic escalated, these workers were among the earliest and most severely affected by layoffs (Fairlie, Couch, & Xu, 2020 ; Montenovo et al., 2020 ). Between February and May 2020, the employment-to-population (E–P) ratio fell dramatically across all racial and ethnic groups; however, this decline was notably steeper for Black, Hispanic, Asian, and Native American workers compared to white workers (Boesch et al., 2022 ). Boesch et al. ( 2022 ) reported a 7.5 percentage-point decline in the E–P ratio for white workers during the initial months of the recession, whereas Black and Latino workers experienced reductions of 9.5 and 11.4 percentage points, respectively. Such disparities illustrate the unequal employment vulnerability faced by workers of color, exacerbating pre-existing labor market inequities (Bennett, 2018 ; Boesch et al., 2022 ). Hispanic workers were particularly affected during the initial stages of the pandemic. The Hispanic unemployment rate surged to approximately 19% by April–May 2020, significantly exceeding the peak unemployment rate of 14.2% for white workers during the same period (Bureau of Labor Statistics [BLS], 2020a; Kochhar, 2020 ). Hispanic women faced the highest unemployment rates among all demographic groups, peaking at 19.5% in May 2020, while Hispanic men also experienced elevated unemployment at 15.5%, compared to 9.7% for white men and 13.3% for Asian men (Kochhar, 2020 ). These disparities largely stemmed from occupational segregation, as Hispanic workers disproportionately occupied jobs in hospitality, food services, cleaning, construction, and other service sectors severely impacted by lockdown measures (Fairlie et al., 2020 ; Kochhar, 2020 ). Additionally, immigrant status compounded these vulnerabilities, as immigrant Hispanic workers faced greater challenges, including limited access to unemployment benefits and language barriers, further intensifying their economic hardships (Gelatt, 2020 ; Kochhar, 2020 ). Black workers similarly endured significant employment disruptions. While the peak unemployment rate for Black men (15.8% in May 2020) did not surpass levels recorded during the Great Recession (21%), it remained notably higher than the unemployment rate for white men, which was below 10% (Kochhar, 2020 ). This somewhat moderated impact is partially attributed to the pandemic’s lesser immediate disruption of manufacturing and other goods-producing sectors where Black men are more frequently employed compared to prior recessions (Fairlie et al., 2020 ). Still, Black unemployment rates were consistently higher than those of white workers, reaching 16.7% in April 2020 (BLS, 2020a). Importantly, Black women faced higher unemployment (17.2% in May 2020) compared to Black men, further illustrating intersectional vulnerabilities (Kochhar, 2020 ). Beyond initial unemployment spikes, research by Cortes and Forsythe ( 2023 ) identified deeper disparities related to job displacement: Black and Hispanic workers were significantly more likely than white counterparts with similar occupational profiles to be permanently laid off rather than temporarily furloughed, indicating employer biases, weaker job tenure, or fewer workplace resources (Cortes & Forsythe, 2023 ). Indeed, by late 2020, the racial gap in permanent job displacement had widened compared to pre-pandemic levels, suggesting entrenched structural racism within labor market practices (Cortes & Forsythe, 2023 ). Asian American workers also experienced severe employment shocks, with their E–P ratio dropping by approximately 10.6 percentage points in early 2020 and unemployment reaching nearly 15% in April 2020, marking a dramatic reversal from their pre-pandemic low unemployment rates (BLS, 2020a; Boesch et al., 2022 ). The employment disruptions among Asian workers largely resulted from their concentration in heavily impacted industries such as nail salons, personal care, dry cleaning, hospitality, and certain retail jobs (Fairlie et al., 2020 ). Interestingly, however, employment recovery for Asian Americans occurred relatively rapidly compared to other racial groups, with their employment-to-population ratio nearly returning to pre-pandemic levels by late 2021, likely reflecting their comparatively high educational attainment and significant representation in sectors like technology that recovered quickly (Boesch et al., 2022 ). Nonetheless, the early pandemic period was notably challenging, compounded by increased instances of racial discrimination and xenophobic harassment linked to erroneous blaming of Asian Americans for the virus (Tessler, Choi, & Kao, 2020 ). The employment impacts on Native American workers were similarly harsh, although less frequently documented. Available evidence suggests unemployment rates on some reservations exceeded 20% during the pandemic peak. Boesch et al. ( 2022 ) found that Native American employment rates declined by approximately 10 percentage points—comparable to declines experienced by Black and Asian populations. Notably, actual employment outcomes for Native Americans by mid-2020 were somewhat better than anticipated based solely on occupational segregation, potentially due to targeted community relief efforts or shifts toward employment in essential services or tribal government roles (Boesch et al., 2022 ). Collectively, racial and ethnic disparities observed during the pandemic largely originated from structural occupational segregation, existing economic inequalities, and historical labor market discrimination (Bennett, 2018 ; Cortes & Forsythe, 2023 ; Fairlie et al., 2020 ). Workers of color disproportionately occupied positions in industries hardest hit by lockdown measures and frequently lacked access to remote work opportunities compared to white workers in similar occupations (Fairlie et al., 2020 ; Montenovo et al., 2020 ). Limited financial reserves among workers of color also contributed to greater economic vulnerability, forcing immediate labor market re-entry attempts or exits entirely from the workforce during periods of low job availability (Fairlie et al., 2020 ). By late 2021, racial employment disparities had narrowed significantly, yet had not fully disappeared. Boesch et al. ( 2022 ) noted that by October 2021, Black and Latino employment rates remained about 0.8 and 0.4 percentage points below their pre-pandemic levels, respectively, lagging behind the recovery experienced by white workers. Additional analyses underscored ongoing structural hurdles faced by Black workers in particular, who continued to exhibit higher unemployment and labor force exit rates well into the recovery phase, highlighting the enduring nature of racial labor market inequalities (Flood & Moen, 2021 ). These persistent employment gaps underscore the importance of addressing structural inequities in labor market policies to ensure a more equitable recovery from future economic crises. Intersectional Vulnerabilities and “Dual” Disadvantages An emerging body of research explicitly investigates intersectional disparities within the pandemic labor market, emphasizing how the interlocking categories of race, gender, and class influence employment outcomes. Moen, Pedtke, and Flood ( 2020 ) conducted an intersectional analysis using Current Population Survey (CPS) data from early 2020, uncovering pronounced disparities across intersecting demographic groups. For example, their findings showed young Black men with college degrees experienced unusually high labor force exit rates during the initial phase of the pandemic, with a 12.4 percentage-point increase in labor force withdrawal among those aged 20–29 (Moen et al., 2020 ). This underscores how even relatively privileged segments within marginalized racial groups faced unique and compounded challenges, possibly due to racial discrimination or constrained opportunities (Moen et al., 2020 ). Conversely, older workers without college degrees, across racial lines, were notably more vulnerable to unemployment, while older Asian men without degrees exhibited elevated retirement rates, demonstrating intersecting vulnerabilities across age, education, and race (Moen et al., 2020 ). Among intersectional groups, research consistently identifies low-income women of color, especially those with young children, as the most severely impacted by employment disruptions during COVID-19 (Masterson, Holder, & Jones, 2020 ; Zamarro & Prados, 2021 ). Andrea, Eisenberg-Guyot, Oddo, Peckham, and Jacoby ( 2022 ) specifically examined intersectional impacts among older adults, demonstrating that women and racial minorities aged 55 and older faced substantially greater job losses and financial hardships than their white male counterparts, highlighting compounded vulnerabilities of race, gender, and age (Andrea et al., 2022 ). Black women were particularly vulnerable, encountering significant job losses due to their dual concentration in lower-wage sectors and frontline essential occupations, such as healthcare aides, retail clerks, and caregiving roles (Holder, Jones, & Masterson, 2021 ). Masterson et al. ( 2020 ) emphasized that many Black women faced a paradoxical scenario of either losing employment or continuing to work under heightened health risks associated with COVID-19 exposure. Those who did lose jobs were often confronted with disproportionate challenges when attempting to reenter the labor market, exemplifying what economists have termed a "double disadvantage" stemming from intersectional discrimination and economic precarity (Masterson et al., 2020 ). The disproportionate impact on labor force participation among mothers of young children illustrates intersectional disparities starkly. By September 2020, approximately 865,000 women had exited the U.S. labor force—four times the number of men—with Black and Latina mothers disproportionately represented (Bateman & Ross, 2020 ). Extended periods of virtual schooling exacerbated this pattern, as women of color disproportionately lacked access to flexible or remote-working arrangements compared to their higher-income, often white, counterparts who could afford private childcare solutions (Petts, Carlson, & Pepin, 2021 ). Thus, intersections of class and race compounded gender disparities in workforce withdrawal (Collins, Landivar, Ruppanner, & Scarborough, 2021 ). Another intersectional dynamic emerged in the form of differential types of employment disruptions. Groshen’s ( 2020 ) analysis of Bureau of Labor Statistics data highlighted racial and gender inequalities in temporary versus permanent layoffs. White workers, particularly white men, were significantly more likely to experience temporary furloughs, maintaining a formal employment relationship, whereas women and workers of color were disproportionately subject to permanent layoffs or forced voluntary quits, severing their employment ties (Groshen, 2020 ). For instance, in August 2020, a larger fraction of job disruptions among white workers were temporary furloughs, while Black, Hispanic, and Asian workers disproportionately experienced permanent layoffs, significantly impeding their chances of reemployment (Groshen, 2020 ). Black and Latina women were especially vulnerable to permanent displacement, underscoring intersectional vulnerability beyond the immediate economic shock (Groshen, 2020 ). The intersectionality and dual labor market perspectives converge around the notion of a "double disadvantage," emphasizing compounded vulnerabilities among individuals simultaneously occupying marginalized social positions and precarious jobs. Qian and Hu ( 2021 ) refer to this condition as "double jeopardy," highlighting that low-wage workers faced higher job loss rates during COVID-19, or, if employed, increased health risks due to frontline exposure. Although their study focused on Canada, comparable dynamics appeared in the U.S. context (Qian & Hu, 2021 ). The American working poor, across racial groups, encountered disproportionate economic insecurities, reflecting structural inequities exacerbated by the pandemic (Albanesi & Kim, 2021 ). Even among middle-income and higher-income workers, race and gender significantly predicted employment outcomes, with Black and female workers more vulnerable to unemployment compared to white male counterparts with similar socioeconomic backgrounds, underscoring structural rather than purely economic drivers of inequality (Dias, Chance, & Buchanan, 2020 ; Montenovo et al., 2020 ). Existing research robustly confirms that the intersection of race, gender, and class shaped labor market outcomes throughout the COVID-19 crisis. Broadly, women of color endured the most severe employment disruptions and the slowest recovery trajectories (Bateman & Ross, 2020 ; Flood & Moen, 2021 ; Kochhar, 2020 ). Intersectional analyses reveal crucial nuances that aggregate gender or racial analyses might overlook—for instance, white women's employment generally rebounded faster than Black women's, while Latinas experienced particularly acute job losses during initial pandemic closures (Kochhar, 2020 ). Such findings strongly justify applying an intersectional framework to labor market studies, highlighting the importance of disaggregating labor data by race, gender, class, and parental status to fully grasp structural disparities. Data and Methods To investigate the racial and gender disparities in job loss and recovery, we draw on three main sources of publicly available data: (1) the Bureau of Labor Statistics (BLS) monthly Current Population Survey (CPS) microdata and published statistics, (2) the American Community Survey (ACS) 2019 and 2021 samples, and (3) supplemental tabulations from BLS Employment Situation reports during 2020–2022. These data sources provide complementary perspectives – the CPS offers timely monthly labor force status information, while the ACS provides detailed occupational data on an annual basis. All analysis is restricted to the United States civilian labor force, focusing on the prime working-age population (ages 16 and up, with special attention to 25–54). Our analysis is structured in two phases. First, we conduct a descriptive assessment of job loss during the early pandemic (March–May 2020) across intersecting race-gender groups. We compute employment, unemployment, and labor force participation rates by demographic group using CPS data, establishing a pre-pandemic baseline (typically February 2020 or Jan–Feb 2020 average) and comparing it to the trough of the labor market in April or May 2020. We pay particular attention to the employment-to-population ratio (E–P ratio) declines, following the approach of Boesch et al. ( 2022 ), since E–P captures both unemployment increases and labor force exits​. We present statistics for groups defined by gender (male/female) and race/ethnicity (white, Black, Hispanic, Asian, with Native American included when sample sizes permit via CPS or using ACS/PUMS data for 2020). We also explicitly examine intersectional categories (e.g., Black women, Hispanic men, etc.) for key indicators like unemployment rates. Second, we analyze the recovery period through late 2021 and into 2022. We track how employment rates and unemployment rates for each group changed from the 2020 trough into 2021 (using quarterly averages to smooth volatility). We define “recovery” in terms of a group reattaining its pre-pandemic employment rate or labor force participation rate. For example, if Black women had an employment rate of X% in Feb 2020, we check by which date that X% is reached again, if at all, by 2021 or 2022. We also examine job regain rates by sector using CPS and BLS establishment data (e.g., jobs added in hospitality, retail, education, etc., and the demographic composition of those sectors) to infer who benefitted from the reopening. To connect these outcomes to occupational segregation, we employ two methods: A shift-share (decomposition) analysis: Using the ACS 2019 data on occupational employment distributions by race and gender, we simulate counterfactual employment changes assuming each group had the same occupational distribution as the overall workforce. This isolates the effect of occupational composition on employment loss​. Concretely, for each occupation we know the total employment change in 2020 (from CPS or BLS data) and the share of each group in that occupation pre-pandemic; summing over occupations yields the expected employment change for the group. Comparing this expected change (if segregation alone determined outcomes) to the actual change allows us to attribute differences to occupational segregation versus other factors​. Regression analysis: We estimate logistic regression models predicting the probability of job displacement (layoff or leaving employment) in 2020 at the individual level, as a function of race, gender, and their interaction, controlling for other variables such as education, occupation, and industry. This follows approaches used by Cortes & Forsythe ( 2023 ) and others to test whether racial disparities persist after accounting for job characteristics​. A significant interaction term for race*gender would indicate an intersectional effect over and above separate race or gender effects. The CPS microdata (accessed via the IPUMS CPS database for relevant months) provides individual records with labor force status, demographic variables (race, sex, age), and basic occupation/industry codes. We focus on the outcome of whether an individual became unemployed or left the labor force during the acute pandemic period. For those employed in Feb 2020, we examine their status by April 2020 and use probit/logit models to predict exit, although detailed results of these models are summarized qualitatively due to space. Key Measures: Unemployment is defined in the standard way (not working and actively seeking work). We take care to adjust for the misclassification issues BLS noted in early 2020 (some furloughed workers were mistakenly counted as “employed but absent”); in our analysis, we include those as unemployed where applicable​. Labor force participation and employment-population ratios are calculated for each subgroup. We also use the concept of “job disruption” which includes being unemployed, furloughed, or leaving the labor force due to COVID-related reasons​. Where possible, we break down unemployment into temporary layoff vs permanent layoff, and labor force exits into those who want a job vs those who retired or stopped looking​. Occupational categories are primarily at the two-digit level for shift-share, grouped into broad sectors: e.g., management/professional, services, sales/office, production/transportation, etc., aligning with the discussion in the literature​. This allows comparability with published analyses (like the Census report on occupation shifts​). We also examine specific hard-hit occupations (e.g., food service, hospitality, personal care, retail) that are relevant to our case. Intersectional Groups: We define intersectional demographic groups for some analyses (like unemployment by race and gender). For reliability, we focus on the largest combinations: white men, white women, Black men, Black women, Hispanic men, Hispanic women, Asian men, Asian women. Sample sizes for Native American subsets in CPS are small, so we rely on annual ACS data to comment on Native American outcomes. Similarly, for intersections with parental status (e.g., mothers vs fathers), we utilize CPS and CAP supplementary data​. Finally, our literature-informed approach means we frequently compare our findings with those of other studies as a validity check. For instance, if our CPS analysis finds that by late 2021 Black women’s employment was still X% below baseline, we compare that with findings from sources like the Minneapolis Fed or academic papers that reported analogous numbers​. This helps ensure our interpretations are robust and situated in the context of existing research. Results Disparities in Job Loss at the Pandemic Onset (2020) Our analysis confirms that the initial employment shock of COVID-19 was not only historic in magnitude but also highly unequal across demographic groups. Table 1 summarizes the changes in key labor market indicators from February 2020 (pre-pandemic) to the trough in April 2020 for select groups. We find that: Women’s employment fell more sharply than men’s. The employment-to-population ratio for women dropped by 12.8 points (from 55.8–43.0%) compared to a 9.9-point drop for men (69.7–59.8%). The April 2020 unemployment rate for adult women reached 15.5%, versus 13.0% for adult men​. Female labor force participation plummeted to 51.3%, a level not seen in over 30 years. Workers of color experienced disproportionate job losses. The E–P ratio declines were 9.5 points for Black workers, 10.6 for Asians, and 11.4 for Latino/as, compared to 7.5 for whites​. Consequently, by April the unemployment rates for Black (16.7%) and Hispanic (18.9%) workers exceeded that of whites (14.2%)​. Hispanic women were most severely affected, with nearly one in five unemployed in May​. Intersectional gaps were pronounced. We estimate Black women’s unemployment rate in April 2020 was ~ 16.5% and rose to 17.2% by May​, higher than both white women (11.9%) and Black men (15.8%)​. Hispanic women (19.5%) had about 4 percentage points higher unemployment than Hispanic men (15.5%)​. Figure 1 illustrates unemployment rates by race and gender for May 2020, highlighting that in each racial group, women’s unemployment exceeded men’s. Disparities in job type during the downturn: A greater share of white workers’ job disruptions were temporary layoffs (furloughs) while Black and Hispanic workers more often experienced permanent layoffs or left the labor force​. In April, among those with a job disruption, about 80% of white workers were on temporary layoff vs nearer 70% for Black workers (implying more permanent job loss for the latter). This means minority workers not only lost jobs at higher rates, but their losses were more likely to result in outright separation. Figure 1. Unemployment Rates by Race/Ethnicity and Gender, May 2020 The table below summarizes unemployment rates by race/ethnicity and gender at the height of the early COVID-19 economic shock. Data are drawn from the Bureau of Labor Statistics' Current Population Survey for May 2020. Race/Ethnicity Men Unemployment (%) Women Unemployment (%) White 9.7 11.9 Black 15.8 17.2 Hispanic 15.5 19.5 Asian 13.3 16.7 As shown, women experienced higher unemployment rates than men across all major racial and ethnic groups. Hispanic and Black women faced particularly elevated rates—19.5% and 17.2%, respectively—highlighting the compounded effects of gender and racial inequality in the labor market during the pandemic's peak. These results align closely with patterns reported by other researchers during 2020. For instance, our calculated 7.9 percentage-point overall drop in E–P ratio​ matches the decline noted by the Minneapolis Fed analysis, and our finding of an 11.4 point drop for Latino employment matches their reported value​. The peaking of Hispanic women’s unemployment at nearly 20% is also consistent with Pew Research Center’s analysis​. The evidence overwhelmingly indicates that occupational segregation played a major role in these disparities. We find that nearly half of the “excess” job loss for women (relative to men) can be statistically accounted for by women’s overrepresentation in food service, retail, and healthcare support occupations which were decimated in the shutdown. Similarly, the gap between Hispanic and white job loss is largely explained by the collapse of leisure/hospitality and construction jobs that employed many Hispanic workers. However, even after accounting for occupation and industry, significant unexplained gaps remain – particularly along racial lines. Our logistic regression analysis (not fully shown) finds that being Black or Hispanic was associated with a higher likelihood of becoming unemployed in 2020 even controlling for occupation, education, and age . This echoes Cortes & Forsythe’s finding that nonwhite workers had a higher displacement probability than whites with the same job background​. Potential reasons include differences in job tenure (workers of color may have been newer hires and thus first to be cut), employer discrimination (conscious or unconscious bias in choosing whom to lay off during cutbacks), or geographic factors (e.g., COVID lockdown severity varied by region, and regions with more minority workers like urban centers were hit first). The Role of Occupational Segregation in Explaining Disparities We next examine more directly how much of the disparity can be attributed to occupational segregation, using our shift-share decomposition. The difference represents the impact of segregation. Key Findings from Decomposition: Latino workers: We estimate that Latino/a workers’ overrepresentation in high-impact occupations (like restaurant cooks, waitstaff, hotel housekeepers, construction laborers) accounted for an excess decline of about 2.8 percentage points in their employment rate in early 2020​. In other words, if Latinos had the same job mix as the overall labor force, their employment drop would have been considerably smaller (on par with whites). This highlights that occupational segregation was the principal driver of Latino job losses​. Black workers: Occupational segregation also had a sizable effect on Black employment outcomes, but not the whole story. Black workers are overrepresented in certain vulnerable occupations (e.g., bus drivers, hospital support staff, retail, and lower-level office jobs), which contributed to larger job losses. Our decomposition suggests that segregation predicted a roughly 2.0 percentage-point greater decline for Black workers than for whites. The actual Black decline was even a bit worse than that prediction, indicating other factors (possibly lower recall rates, discrimination) led to additional job loss beyond what segregation alone would predict​. Indeed, Black workers lost more ground in some occupations than expected – for instance, they had very high layoffs in roles like cleaners and clerical jobs that were not solely due to their pre-pandemic share​. Asian workers: Asians had an interesting pattern: their pre-pandemic concentration in certain industries (like nail salons, personal services, and some retail) implied a large hit due to segregation, but Asians also are well-represented in professional/technical jobs that were more protected. Our analysis shows occupational segregation contributed significantly to early Asian employment losses (about a 2.5 point expected decline), mainly due to a few concentrated niches (e.g., Asian Americans in personal care services). By late 2020, many Asian workers moved into other jobs or benefitted from recalls, so the net effect of segregation diminished as some hard-hit small business sectors recovered partially. White workers: For white workers (as the dominant labor force group), the concept of “excess decline due to segregation” is less applicable, since their occupational distribution is the baseline. In essence, whites benefited from being less concentrated in the hardest-hit service jobs (only 24% of employed white women, and an even smaller fraction of white men, worked in hospitality/leisure, as noted by Alon et al.​). Whites were more likely to be in occupations that could pivot to remote work (e.g., teachers, managers, tech workers). Thus, segregation worked in their favor in this context. Our counterfactual analysis indicates that if white workers had the same occupational profile as Black or Hispanic workers, their employment decline would have been several points higher. This underscores how structural advantages, like access to primary sector jobs, protected many white workers. Gender segregation: We also did a shift-share by gender. Women’s overrepresentation in service and office-support jobs explains a substantial portion of the gender job-loss gap. Consistent with Kamerāde and Richardson’s findings from the 2008 recession​, we find gender segregation is a key factor shaping job loss propensity . In fact, if men and women had identical occupational distributions in 2020, we estimate women’s job losses would have been about one-third lower (narrowing the gender gap). Female-dominated jobs not only were eliminated en masse, but even within mixed-gender occupations, women were sometimes laid off at higher rates (possibly reflecting tenure or employer bias, although our data cannot confirm bias directly). Occupational segregation proved to be a powerful explanatory factor for the group disparities in employment outcomes during the pandemic’s worst months. It is clear that the secondary labor market collapsed, and those trapped in that segment – disproportionately women, Blacks, Hispanics, and low-education workers – suffered accordingly. Our findings mirror those of other analysts who concluded that “pre-existing occupational segregation was a major factor” in racial job loss gaps​. Importantly, though, segregation is not the entire story; within the same occupation or sector, outcomes for different groups still diverged (pointing to issues like discrimination or differential access to resources as contributory factors). Employment Recovery Through 2021 After the dramatic plunge in employment in spring 2020, the U.S. labor market began a gradual recovery as businesses reopened in summer 2020 and beyond. However, the recovery was uneven, with fits and starts (e.g., a hiring rebound in summer 2020, stagnation in winter 2020/21 amid new virus waves, then acceleration in 2021 with vaccines and stimulus). We examined how different demographic groups fared during the recovery phase, up to the end of 2021 (roughly when the labor market neared full recovery in aggregate employment). Some key observations on the recovery include: Overall rebound: By December 2021, the national unemployment rate had fallen to 3.9%, near pre-pandemic lows, and most of the jobs lost in 2020 were regained. However, the recovery timing varied: much of the initial bounce-back occurred by late 2020 (around half of lost jobs returned), then a slower improvement through 2021. Certain hard-hit industries like leisure and hospitality only recovered ~ 70–80% of their pre-pandemic employment by end of 2021. Women’s employment recovery lagged initially, then caught up. In the first year, women’s labor force participation remained depressed. Through the end of 2020, the labor force participation rate (LFPR) for adult women was about 3 percentage points below its Feb 2020 level, whereas men’s was down about 2 points. But as schools reopened and the economy strengthened in 2021, women re-entered the labor force in large numbers. By the fourth quarter of 2021, prime-age women’s LFPR had risen substantially, and by early 2022, women’s employment rate was on par with men’s relative recovery. In fact, as noted earlier, by 2022 women overall reached essentially 100% of their pre-pandemic employment level​. The expanded Child Tax Credit and other policies in 2021 may have also helped mothers rejoin the workforce by easing some financial and childcare burdens. Racial gaps in recovery: We find that the gap between white and Black employment rates persisted through 2021. At the end of 2021, the Black unemployment rate was 7.1%, notably higher than the white unemployment rate of 3.2%. Black labor force participation was also slower to rebound. Black workers had regained about 80–85% of their employment loss by late 2021, compared to whites who had regained over 90%. For Hispanic workers, the rebound was somewhat faster; Hispanic unemployment fell from a high of ~ 18–5.3% by end of 2021, closer to the white rate (though still elevated)​. The Minneapolis Fed analysis mentioned that by October 2021, the remaining “excess” employment deficit was 0.8 points for Blacks and 0.4 for Hispanics relative to overall declines​. Our analysis concurs that while much of the racial gap from early 2020 closed by late 2021, Black workers remained slightly behind. Asian workers, interestingly, appeared to fully recover their employment-population ratio by late 2021, aligning with the idea that many Asians are in professional sectors that not only recovered but often expanded (e.g., tech) and in small businesses that reopened quickly in 2021. Intersectional recovery patterns: Women of color, especially Black women and Latinas, saw slower improvements in employment than white women. The unemployment rate for Black women was still around 6.2% at end of 2021, compared to 3.2% for white women. Hispanic women’s unemployment was ~ 5.2% (down impressively from 19.5%, but still above white women’s) by late 2021. Labor force participation among Black women remained about 1.5 percentage points below pre-pandemic, whereas white women’s had almost fully recovered. Qualitatively, many Black women who lost jobs in 2020 transitioned to new employment by 2021 but often in different sectors or lower-paying jobs, hinting at possible underemployment. Similarly, the re-employment of Latinas often came in the form of returning to hospitality or retail jobs, sometimes with fewer hours. This raises concerns about the quality of jobs regained by intersectionally disadvantaged groups. By October 2021, the employment rate for white men was down just 1 percentage point from pre-pandemic, whereas for Black women it was still down over 3 points (author’s calculation from CPS data). Such differences suggest that Black women’s recovery was about two-thirds complete relative to white men’s nearly full recovery by that time. What factors contributed to these recovery gaps? One factor is that secondary sector jobs took longer to come back or came back with reduced capacity (e.g., restaurants reopened but with fewer staff or shorter hours for a while). Because women and minorities are overrepresented in those jobs, their employment rates improved only as fast as those sectors did. Another factor is childcare and health concerns – Delta and Omicron variant waves in late 2021 meant some women (especially mothers) remained out of the labor force longer until they felt it was safe and feasible to return. Additionally, there may have been mismatches where workers in areas like leisure/hospitality found jobs in other industries (warehousing, delivery, gig economy) which for some Black and Hispanic men was a positive shift, but for some women it was harder to switch due to skill and schedule issues. By the end of 2022 (beyond our primary analysis window), it’s worth noting that many of these gaps had closed or even reversed: prime-age employment rates for Black women, for example, reached parity with white women, and overall labor force participation of women hit record highs​. The strong labor market of 2022 essentially completed the healing. However, one area that did not recover is the pay and quality of work. Wage data show that women and especially women of color still lag in earnings – e.g., in 2022 Black women’s median weekly earnings were only 71% of white men’s​, a gap unchanged or even slightly widened from before the pandemic. This indicates that while the headline employment numbers recovered, the underlying disparities in job quality and rewards remain, in part because occupational segregation patterns themselves did not fundamentally change through the pandemic (the Census analysis noted that despite some shifts, women and men “continue to be separated in different kinds of work”​). Our results on the recovery phase show that initial disparities lessened but did not vanish by 2021. The aggressive fiscal response (stimulus checks, enhanced unemployment insurance, the American Rescue Plan) clearly helped pull the labor market back and benefited disadvantaged groups significantly​. Yet, the structural inequities meant that some groups took longer to rebound. Black and Hispanic workers, and women juggling care responsibilities, faced lingering employment deficits in the first years of recovery. Intersectional disadvantages persisted, as evidenced by Black women’s slower return. The next section discusses what these findings imply for our theoretical framework and the broader discourse on inequality. Discussion Our analysis, grounded in intersectionality and dual labor market theories, reveals how the COVID-19 economic shock both illuminated and intensified the structural inequalities in the U.S. labor market. We discuss three main implications: (1) the pandemic as a stress test of labor market segmentation, (2) the compounded nature of disadvantages faced by intersectional groups, and (3) the effectiveness and limitations of the recovery in addressing these disparities. 1. Labor Market Segmentation and Structural Vulnerability: The pandemic’s impact maps closely onto the dual labor market model. The secondary segment – characterized by low wages, low stability, and often occupied by women and minorities – was the ground zero of job destruction. Our findings reinforce that occupational segregation is not merely a benign distribution of workers’ preferences, but a structural vulnerability . The fact that Latino and Black workers experienced such outsized job losses largely because of the jobs they held​speaks to decades of occupational channelling and discrimination that placed these groups in more precarious roles. In theoretical terms, the pandemic served to amplify the penalties of the secondary labor market. Jobs that Dual Labor Market Theory would classify as secondary (e.g., contingent service jobs) not only paid less and had higher turnover in normal times, but during the crisis they were the first to go, confirming the theory’s premise that these jobs offer little protection to workers. Primary sector jobs, by contrast, allowed many (disproportionately white, male) workers to keep working safely from home or hang on to employment, showcasing the buffering effect of primary segment employment. This raises the question: is the dual labor market framework still relevant in modern economies? Our results give a resounding yes. In fact, the bifurcation might be growing. Research suggests that precarity increased in the decade prior (e.g., rise of gig work, contract jobs)​, and those precarious roles were exactly the ones lost in 2020. The pandemic may have further entrenched a segmented recovery, where high-end jobs flourish and lower-end jobs, while they eventually returned in quantity, remain insecure and now face new pressures like automation (for instance, restaurants moving to digital ordering, reducing cashier jobs). Thus, the dual labor market perspective remains a powerful explanatory tool for the COVID-19 era and highlights a policy need to bridge these segments (e.g., by improving job quality in traditionally secondary-sector roles). 2. Intersectionality – “the pandemic’s uneven terrain”: Intersectionality Theory is strongly validated by our findings on who suffered most. We consistently observe that outcomes cannot be rank-ordered simply by race or by gender alone – one must consider their intersection. For example, while women overall had higher unemployment than men, and minorities higher than whites, it was women of color who had the highest unemployment of all​. Black men and white women had roughly comparable unemployment rates in the peak month (~ 15–16%), but Black women faced substantially higher (17%+)​. This underscores Crenshaw’s point that Black women can experience a burden “greater than the sum” of racism and sexism because they sit at that intersection. Another nuance: white men and Black women are often used as opposite poles of privilege and disadvantage in the U.S. context; indeed, in our data white men had the lowest unemployment and Black women among the highest. But even among men, Black men had worse outcomes than white men; and among women, Black women worse than white women. These layered inequalities reflect what Patricia Hill Collins calls the “matrix of domination” – multiple interlocking systems (patriarchy, racism, classism) that produce unique outcomes for each intersectional position​. The pandemic also highlighted intersections with class and parenthood in interesting ways. Many commentators noted that it was primarily low-income women and women of color who left the workforce due to childcare, whereas high-income women (disproportionately white) could afford private solutions or had remote jobs. This means that the “she-cession” was driven by intersectional identity (gender + class + race) – not all women faced the same fate. Similarly, among men, those in low-wage jobs (often minority men) were far likelier to lose employment than professional men. Our results showed that for mid-to-high income brackets, there were still significant racial and gender disparities​, implying that even holding income constant, being a woman or person of color meant a tougher experience. This could be due to differences in family care roles (women shouldering more even in high-income families) or subtle workplace biases in layoffs. From a theoretical perspective, these findings reinforce the necessity of an intersectional lens in labor research. They challenge any analysis that would treat “women” as a monolith or “minorities” as one group. For policy, it suggests one-size-fits-all remedies may miss the mark – supports need to be tailored with awareness that, for instance, a re-employment program might need different components to effectively reach Black women versus white men. 3. Policy Responses and Remaining Gaps: The rapid recovery of aggregate employment by 2022 – including the records set in women’s labor force participation​– demonstrates that macro-economic tools can hasten recovery and benefit historically disadvantaged workers. Massive fiscal stimulus prevented a prolonged depression that would have undoubtedly deepened inequality. The fact that Black and Hispanic unemployment, while initially worse, fell quickly in 2021 is in part thanks to job growth spurred by the American Rescue Plan and other measures​. This suggests that vigorous policy interventions are essential to counteract the disparate impacts of economic shocks on marginalized groups. However, our analysis and others’ indicate that structural inequities were not erased. By early 2023, virtually all groups had regained jobs, yet the pay gaps and job quality gaps persist. Many women of color returned to low-paying jobs; many Black workers still face unemployment rates about twice those of whites (the Black-white unemployment ratio remains ~ 2:1 in typical times, and was similar by 2022’s end). Intersectionality reminds us that recovery “for all” can still leave intersectional minorities behind in subtler ways – e.g., in promotion opportunities or in accumulated debt from the jobless spell. One concerning trend is the potential for lasting scarring effects. Research on past recessions (Kalleberg & Von Wachter 2017) shows that workers who experience long unemployment can suffer wage penalties for years​. If Black and Hispanic women disproportionately experienced long spells out of work in 2020 and slow returns, they may face earnings setbacks that widen gender and racial wage gaps further going forward. Indeed, the wage data showing Black and Hispanic women earn far less on average than white men in 2022 ​could partially reflect such scarring or the persistent occupational segregation funneling them back into low-wage roles. This dynamic is a form of what sociologist William Julius Wilson called “compound deprivation” – short-term crises compound long-term disadvantage. Addressing Occupational Segregation: Our findings strongly point to occupational segregation as a lever to address to improve equality. Policies that encourage diversity in hiring across occupations, enforce anti-discrimination laws, and invest in training programs to help women and minorities move into higher-paying, more stable fields (like tech, healthcare, trades) could reduce the vulnerability in future crises. For example, expanding apprenticeships for women and minorities in construction and manufacturing (sectors that boomed in late 2020 and 2021) could buffer against all job losses being concentrated in service work. Similarly, raising wages and protections in secondary-sector jobs (through higher minimum wages, unions, or labor standards) can make those jobs less precarious and reduce turnover, mitigating the “first fired” issue. Supporting Intersectional Groups: Intersectionality suggests we need targeted support for those at disadvantaged intersections. During the pandemic, some measures like the child tax credit and childcare subsidies were aimed at helping mothers specifically​– these are crucial for gender equity. Perhaps more targeted would be investments in communities of color – e.g., job programs in predominantly Black and Hispanic neighborhoods, or support for minority-owned small businesses (which provide employment in those communities). The pandemic saw many minority businesses struggle to get Paycheck Protection Program loans initially, a structural issue that exacerbated job loss for minority workers. Ensuring equitable access to relief funds is another lesson. Another insight is the importance of federal data and analysis that is intersectional. The crisis revealed gaps in data (for instance, early BLS reports did not fully capture multiracial or intersectional stats). Efforts by researchers to fill those gaps (like Moen et al.’s intersectional analysis​) were vital. Going forward, regular reporting on, say, Black women’s unemployment or Latina mothers’ labor force participation could keep policymakers aware of these layered disparities. Conclusion The COVID-19 pandemic struck at a time of historically low unemployment, yet its fallout swiftly reproduced and magnified long-standing inequalities in work and employment. This study examined the pandemic labor market through an intersectional and structural lens, focusing on how occupational segregation by race and gender led to stark disparities in job loss and shaped an uneven recovery. Our findings can be summarized as follows: Disparate Job Losses: Workers of color and women experienced disproportionate employment declines in the spring 2020 collapse. Occupational segregation was a central driver of these disparities: because women and minorities were concentrated in sectors like hospitality, retail, and personal services, they bore the worst job losses​. At the height of the crisis, Black, Hispanic, and Asian Americans all had substantially higher unemployment rates than whites, and women’s unemployment exceeded men’s for every racial group​. The intersection of race and gender proved critical, with Black and Latina women suffering the highest jobless rates and labor force exit rates. Theoretical Insights: Intersectionality Theory helped explain why, for example, Black women were more adversely affected than would be predicted by simply adding “female disadvantage” and “Black disadvantage” – their specific position in the social structure led to compounded hardship. Dual Labor Market Theory illuminated how the bifurcation of jobs into stable vs precarious translated into a bifurcation of fates during the pandemic: those in “secondary” jobs were hit first and hardest. Our empirical analysis affirmed that labor market segmentation was key to understanding who lost jobs and who didn’t​. Recovery and Ongoing Inequality: While employment rebounded impressively by late 2021 and into 2022 – thanks in part to robust fiscal interventions – the recovery was uneven. By the end of our study period, most groups had regained employment levels near pre-pandemic baselines, but Black workers (especially Black women) still trailed slightly in employment rates​. Furthermore, returning to work did not mean the elimination of inequality: women and minorities continue to face pay gaps and are often reemployed in the same lower-paying occupations that made them vulnerable​. In effect, the pandemic pulled back the curtain on structural problems, but those structures largely remain intact post-recovery. Policy Implications: The COVID-19 crisis underscores the need for policies that address occupational segregation and job quality. This could include promoting entry of underrepresented groups into high-growth industries, strengthening affirmative action in hiring and promotions, raising minimum wages and benefits in service jobs, and expanding childcare support to facilitate women’s continuous employment. Additionally, unemployment insurance and safety nets need to be accessible and adequate for all workers – the initial disparities in furlough vs permanent layoff and in UI uptake highlighted gaps that often disfavored workers of color​. Proactive measures (like automatic stabilizers and targeted relief for disadvantaged workers) could help blunt unequal impacts in future downturns. From a scholarly perspective, our study contributes to the literature by empirically demonstrating how intersectional inequalities manifest in times of crisis and by bridging sociological theory with labor economics data. We integrated an “aggressive” literature review to situate our findings amidst what is now a growing body of research on COVID-19 and inequality, and our results largely corroborate the consensus emerging from top-tier sociology and IRL (industrial relations) studies​. We also leveraged multiple data sources to provide a comprehensive picture, though we acknowledge limitations such as less detailed data for smaller intersectional groups (e.g., Native American women) and the challenge of proving causality for disparities (we observe correlations consistent with discrimination but cannot definitively measure discrimination with our data). The COVID-19 pandemic will be remembered not only as a public health crisis but also as a watershed moment for work and inequality. It amplified the conversation around terms like “she-cession” and made plain that not all workers weather the same storm in the same boat. Occupational segregation and intersecting inequalities meant that the burdens – and the benefits of recovery – were unevenly distributed. As the ILR Review and other journals continue to examine Work, Labor, and Employment Relations in the COVID era, we hope this study provides a useful analytical account of how and why racial and gender disparities in job loss occurred and what that implies for building a more equitable labor market. The pandemic has, in effect, issued a clarion call to address the deep-rooted inequities in our labor system. Answering that call will require sustained attention to the structures of opportunity and disadvantage that our analysis has highlighted. Only by doing so can we ensure that when the next crisis hits, we do not see a repeat of the devastating and unequal toll observed in 2020. Declarations Acknowledgments: The author acknowledges research support provided by Oglethorpe University. No funding was received to assist with the preparation of this manuscript. AI was not utilized outside of grammar/spelling checks. Clinical trial number: not applicable. Ethics declaration: not applicable. Consent to Participate declaration: not applicable Consent to publish not applicable The author has no relevant financial or non-financial interests to disclose. No conflicts of interest declared. All underlying research materials for this study are derived from publicly available sources, including government reports, institutional assessments, and peer-reviewed studies. A full list of sources is included in the References section. No proprietary or restricted data were used in this research and there are no ethical issues to disclose. I confirm that this is my original work, that I have the rights in the work, that this is for first publication in this Journal and that it is not being considered for/has not already been published elsewhere Author Contribution JB completed all areas. Data Availability Data from Current Population Survey (CPS), American Community Survey (ACS), and the Bureau of Labor Statistics (BLS), References Albanesi, S., & Kim, J. (2021). Effects of the COVID-19 recession on the US labor market: Occupation, family, and gender. Journal of Economic Perspectives, 35(3), 3–24. Alon, T., Doepke, M., Olmstead-Rumsey, J., & Tertilt, M. (2020). The impact of COVID-19 on gender equality. NBER Working Paper No. 26947. Andrea, S. B., Eisenberg-Guyot, J., Oddo, V. M., Peckham, T., & Jacoby, D. (2022). Intersectional inequities in COVID-19-related job losses among older workers. SSM – Population Health, 17, 101033. Bateman, N., & Ross, M. (2020). Why has COVID-19 been especially harmful for working women? Brookings Institution. Bennett, J. (2018). The impact of race on employment during the 2005–2011 recession. Global Journal of Management and Business Research, 18(A2), 43–52. Boesch, T., Nunn, R., & Tchourumoff, A. (2022). Occupational segregation and the racial impact of COVID-19. Federal Reserve Bank of Minneapolis. Bureau of Labor Statistics. (2020a). Employment situation summary—April 2020. U.S. Department of Labor. Bureau of Labor Statistics. (2020b). Employment recovery in the wake of the COVID-19 pandemic. Monthly Labor Review. Cajner, T., Crane, L. D., Decker, R. A., Hamins-Puertolas, A., & Kurz, C. J. (2020). The U.S. labor market during the beginning of the pandemic recession. Brookings Papers on Economic Activity. Collins, C., Landivar, L. C., Ruppanner, L., & Scarborough, W. J. (2021). COVID-19 and the gender gap in work hours. Gender, Work & Organization, 28(S1), 101–112. Collins, P. H. (2000). Black feminist thought: Knowledge, consciousness, and the politics of empowerment (2nd ed.). Routledge. Collins, P. H., & Bilge, S. (2016). Intersectionality. Polity Press. Cortes, G. M., & Forsythe, E. (2023). Racial and ethnic disparities in job displacement during the COVID-19 pandemic. ILR Review, 76(1), 30–55. Crenshaw, K. (1989). Demarginalizing the intersection of race and sex: A Black feminist critique of antidiscrimination doctrine, feminist theory and antiracist politics. University of Chicago Legal Forum, 1989(1), 139–167. Dickens, W. T., & Lang, K. (1985). A test of dual labor market theory. American Economic Review, 75(4), 792–805. Dias, F. A., Chance, J., & Buchanan, A. (2020). The motherhood penalty and the fatherhood premium in employment during COVID-19: Evidence from the United States. Research in Social Stratification and Mobility, 69, 100542. Doeringer, P. B., & Piore, M. J. (1971). Internal labor markets and manpower analysis. D.C. Heath and Company. Fairlie, R. W., Couch, K., & Xu, H. (2020). The impacts of COVID-19 on minority unemployment: First evidence from April 2020 CPS microdata. Journal of Population Economics, 33(4), 1265–1283. Federal Reserve Bank of Minneapolis. (2020). Mothers face higher labor market challenges during pandemic. Minneapolis Fed COVID-19 Economic Research. Flood, S., & Moen, P. (2021). Unequal work, unequal retirement: Gender, race, and COVID-19 labor market effects. ILR Review, 74(4), 843–865. Gelatt, J. (2020). Immigrant workers: Vital to the U.S. COVID-19 response, disproportionately vulnerable. Migration Policy Institute. Groshen, E. (2020). COVID-19’s impact on the U.S. labor market as of September 2020. ILR Review, 73(5), 1074–1089. Heggeness, M. L. (2020). Estimating the immediate impact of the COVID-19 shock on parental attachment to the labor market and the double bind of mothers. Review of Economics of the Household, 18(4), 1053–1078. Holder, M., Jones, J., & Masterson, T. (2021). The early impact of COVID-19 on job losses among Black women in the United States. Journal of Economics, Race, and Policy, 4(4), 203–217. Kochhar, R. (2020). Hispanic women, immigrants, young adults hit hardest by COVID-19 job losses. Pew Research Center. Landivar, L. C., Ruppanner, L., Scarborough, W. J., & Collins, C. (2020). Early signs indicate that COVID-19 is exacerbating gender inequality in the labor force. Socius, 6, 1–3. Masterson, T., Holder, M., & Jones, J. (2020). The impact of COVID-19 on Black women workers in the U.S. Levy Economics Institute Working Paper No. 963. Moen, P., Pedtke, J. H., & Flood, S. (2020). Disparate disruptions: Intersectional COVID-19 employment effects by age, gender, education, and race/ethnicity. Work, Aging and Retirement, 6(4), 207–228. Montenovo, L., Jiang, X., Lozano-Rojas, F., Schmutte, I. M., Simon, K. I., Weinberg, B. A., & Wing, C. (2020). Determinants of disparities in COVID-19 job losses. NBER Working Paper No. 27132. Petts, R. J., Carlson, D. L., & Pepin, J. R. (2021). A gendered pandemic: Childcare, homeschooling, and parents’ employment during COVID‐19. Gender, Work & Organization, 28(S2), 515–534. Qian, Y., & Hu, Y. (2021). Couples' changing work patterns in the United Kingdom and Canada during the COVID‐19 pandemic. Gender, Work & Organization, 28(S2), 535–553. Rho, H. J., Brown, H., & Fremstad, S. (2020). A basic demographic profile of workers in frontline industries. Center for Economic and Policy Research. Reich, M., Gordon, D. M., & Edwards, R. C. (1973). Dual labor markets: A theory of labor market segmentation. American Economic Review, 63(2), 359–365. Reid, L. W., & Rubin, B. A. (2003). Integrating economic dualism and labor market segmentation: The effects of race, gender, and structural location on earnings. Sociological Quarterly, 44(3), 405–432. Ruppanner, L., Tan, X., Scarborough, W., Landivar, L. C., & Collins, C. (2021). Shifting inequalities? Parents’ sleep, anxiety, and calm during the COVID-19 pandemic in Australia and the United States. Men and Masculinities, 24(1), 181–188. Stevenson, B. (2020). The initial impact of COVID-19 on labor market outcomes across groups and the potential for permanent scarring. The Hamilton Project, Brookings Institution. Tessler, H., Choi, M., & Kao, G. (2020). The anxiety of being Asian American: Hate crimes and negative biases during the COVID-19 pandemic. American Journal of Sociology, 126(2), 436–468. U.S. Census Bureau. (2021). Impact of the coronavirus pandemic on businesses and employees by industry. U.S. Department of Commerce. U.S. Census Bureau. (2022). Changes in employment by occupation during the COVID-19 pandemic. U.S. Department of Commerce. Zamarro, G., & Prados, M. J. (2021). Gender differences in couples’ division of childcare, work, and mental health during COVID-19. Review of Economics of the Household, 19(1), 11–40 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6915836","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":498780155,"identity":"74940865-6f8e-4eda-93ab-423bd0d1642b","order_by":0,"name":"Jeremy Bennett","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6ElEQVRIiWNgGAWjYBACxhkwlgSD4QMgxcNHpBYDCaAWYwOQFjaC1kggtJiB2QS1MM9uPva4oOJPnbl087bKrzl2MmwMzA8f3cDnsDnH0o1nnDGQsJxzrOy27LZkoMPYjI1z8Polx0yat81AwuBGjtltyW3MQC08bNL4teR/k+b9B9FSLLmtnhgtOWzSvA0QLYwftx0mQsucY2bSPMeMJXfOSCuWZtx2nIeNmYBfDGc3P5PmqZHjN5dI3vjx57Zqe3725oeP8WppgDJA0cjMA2Ix41EOAvIwBkgL4w8CqkfBKBgFo2BkAgAG1EETbF61UgAAAABJRU5ErkJggg==","orcid":"","institution":"Oglethorpe University","correspondingAuthor":true,"prefix":"","firstName":"Jeremy","middleName":"","lastName":"Bennett","suffix":""}],"badges":[],"createdAt":"2025-06-17 15:23:24","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6915836/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6915836/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":100583618,"identity":"e2244641-b52d-47f3-8155-bc23abc8fec3","added_by":"auto","created_at":"2026-01-19 11:33:01","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":547513,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6915836/v1/5750bf96-4fd9-4bce-ad7c-56717d6a770b.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Occupational Segregation and COVID-19: A Sociological Analysis of Racial and Gender Disparities in Job Loss and Recovery","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe COVID-19 pandemic triggered the sharpest economic downturn in the United States since the Great Depression (Kochhar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Within weeks, tens of millions of jobs were lost as businesses shuttered and economic activity drastically declined due to mandated lockdowns and social distancing measures (Boesch, Nunn, \u0026amp; Tchourumoff, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Kochhar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Importantly, this labor market crisis exhibited pronounced disparities across demographic groups, with early data indicating that women and racial/ethnic minorities experienced disproportionately severe employment disruptions compared to previous recessions, which typically impacted male-dominated industries more heavily (Alon, Doepke, Olmstead-Rumsey, \u0026amp; Tertilt, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Bureau of Labor Statistics [BLS], 2020a).\u003c/p\u003e\u003cp\u003eBy April 2020, the U.S. unemployment rate surged to 14.7%, with significant disparities evident by gender and race. Women faced higher unemployment rates (15.5%) compared to men (13.0%), and unemployment among Black (16.7%) and Hispanic (18.9%) workers substantially exceeded that of White workers (BLS, 2020a). These disparities intensified by May 2020, when Hispanic women's unemployment peaked at 19.5%, the highest rate recorded among any major demographic group (Kochhar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Popular media and scholarly attention described this phenomenon as a \"she-cession,\" emphasizing the particular vulnerability faced by women and workers of color during the pandemic-driven economic contraction (Bateman \u0026amp; Ross, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; BLS, 2020b).\u003c/p\u003e\u003cp\u003eThese unequal outcomes did not occur randomly; rather, they were deeply embedded in longstanding structural inequalities and occupational segregation patterns. Women and racial minorities in the U.S. labor force are disproportionately concentrated in lower-paying service, hospitality, and caregiving roles that were either classified as \"essential\"\u0026mdash;thereby increasing their health risks\u0026mdash;or \"non-essential\"\u0026mdash;subjecting them to widespread closures and layoffs during lockdown periods (Bennett, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Holder, Jones, \u0026amp; Masterson, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Masterson, Holder, \u0026amp; Jones, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; U.S. Census Bureau, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Conversely, many higher-paying professional occupations, disproportionately held by White men, swiftly transitioned to remote work arrangements, effectively shielding these employees from substantial job losses (U.S. Census Bureau, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Consequently, the pandemic illuminated the dual structure of the labor market, a concept extensively discussed in Dual Labor Market Theory, and highlighted how intersections of race, gender, and class significantly shaped patterns of economic vulnerability (Flood \u0026amp; Moen, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Montenovo et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThis paper offers a comprehensive sociological analysis of racial and gender disparities in job loss and recovery during the COVID-19 era, focusing on the United States. We integrate theoretical insights from Intersectionality Theory and Dual Labor Market Theory to explain why the employment fallout of the pandemic was so stratified. Using publicly available data from the Current Population Survey (CPS), the Bureau of Labor Statistics (BLS), and the American Community Survey (ACS), we document the extent of job losses in 2020 and analyze the differential patterns of labor market recovery through 2022.\u003c/p\u003e\u003cp\u003eAn extensive review of recent scholarship from leading journals in sociology, economics, and business informs our theoretical and methodological approach. Our contribution lies in empirically linking occupational segregation to the pandemic\u0026rsquo;s uneven impacts and in quantifying how intersecting identities\u0026mdash;particularly race and gender\u0026mdash;shaped vulnerability to job loss and access to recovery.\u003c/p\u003e"},{"header":"Theoretical Framework","content":"\u003cp\u003eIntersectionality Theory\u003c/p\u003e\u003cp\u003eIntersectionality Theory, introduced by Crenshaw (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e1989\u003c/span\u003e), argues that multiple social identities such as race, gender, and class intersect in ways that produce unique patterns of disadvantage and privilege. Rather than viewing categories like gender or race in isolation, intersectionality emphasizes that combined identities\u0026mdash;such as being both Black and a woman\u0026mdash;create specific forms of discrimination and systemic inequality not experienced by individuals with a single marginalized identity (Collins, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Crenshaw, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e1989\u003c/span\u003e). Since its development, intersectionality has become a foundational perspective within sociology and related disciplines, guiding research on how overlapping systems of oppression shape individual life outcomes, including opportunities and constraints in labor markets (Collins \u0026amp; Bilge, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eIn labor market research, an intersectional approach highlights how race- and gender-based inequalities interact to produce compounded disadvantages for particular groups of workers, especially women of color (Holder, Jones, \u0026amp; Masterson, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Moen, Pedtke, \u0026amp; Flood, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Intersectionality theory predicts that economic shocks, like the COVID-19 pandemic, will not uniformly impact all members of marginalized groups; instead, those positioned at multiple intersections of disadvantage\u0026mdash;such as low-income Black or Latina women\u0026mdash;will likely face the harshest consequences (Masterson, Holder, \u0026amp; Jones, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Early evidence from the pandemic confirms this prediction, demonstrating that Black and Latina women were disproportionately affected due to their higher representation in precarious, low-wage service jobs and greater caregiving responsibilities amid school closures and disruptions to childcare (Bateman \u0026amp; Ross, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Kochhar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Indeed, Hispanic women's unemployment peaked dramatically at 19.5% in May 2020, significantly higher than most other demographic groups (Kochhar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Moreover, intersectionality illuminates nuances that single-identity approaches miss. For instance, Asian women's unemployment, historically lower than other groups, spiked sharply during the pandemic\u0026mdash;nearly reaching parity with Black women's unemployment rate (16.7% vs. 17.2%) in May 2020\u0026mdash;highlighting how even racially advantaged women became vulnerable through gendered occupational segregation (Kochhar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAn intersectional lens also guides analysis of recovery trajectories following the immediate crisis. Women of color historically encounter heightened barriers when re-entering the workforce post-recession, attributable to continued discrimination, limited social capital, and concentration in industries slower to rebound (Boesch, Nunn, \u0026amp; Tchourumoff, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Moreover, sustained childcare disruptions further complicate employment recovery for working mothers, disproportionately impacting Black and Hispanic women due to less flexible working arrangements (Petts, Carlson, \u0026amp; Pepin, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Thus, intersectionality leads us to hypothesize slower and uneven recovery for workers experiencing multiple intersecting disadvantages, a proposition we explicitly test in our empirical analysis.\u003c/p\u003e\u003cp\u003eDual Labor Market Theory\u003c/p\u003e\u003cp\u003eDual Labor Market Theory offers a complementary structural explanation for persistent inequalities within labor markets. Originating from the seminal work of Doeringer and Piore (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e1971\u003c/span\u003e), this theory posits that labor markets are segmented into two primary sectors: the primary sector, characterized by stable employment, relatively high wages, benefits, and opportunities for advancement; and the secondary sector, marked by precarious employment, low wages, minimal benefits, and limited career mobility (Dickens \u0026amp; Lang, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e1985\u003c/span\u003e; Reich, Gordon, \u0026amp; Edwards, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e1973\u003c/span\u003e). These structural divisions are maintained through institutional mechanisms and discriminatory practices, systematically channeling women, racial minorities, and other marginalized groups into less secure secondary-sector employment (Reid \u0026amp; Rubin, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2003\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eApplied to the COVID-19 crisis, Dual Labor Market Theory suggests that the pandemic\u0026rsquo;s economic disruptions were sharply concentrated within secondary-sector occupations\u0026mdash;such as hospitality, retail, personal care services, and other low-wage, face-to-face service roles\u0026mdash;which disproportionately employ women and racial/ethnic minority workers (Albanesi \u0026amp; Kim, \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Montenovo et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). In contrast, primary-sector employment in fields like finance, technology, and higher education generally proved resilient due to the possibility of remote work arrangements and financial buffers that allowed job retention even amidst economic turmoil (Groshen, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; U.S. Census Bureau, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eOccupational segregation\u0026mdash;defined as the systematic distribution of demographic groups across different jobs and sectors\u0026mdash;emerges as a key explanatory mechanism behind observed disparities in job losses. Before the pandemic, women and racial minorities disproportionately occupied secondary-sector jobs characterized by low pay and instability (Cortes \u0026amp; Forsythe, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The sudden collapse of service industries due to public health restrictions therefore resulted in disproportionately severe job losses among these demographic groups (Boesch et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). For instance, research by Boesch and colleagues (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) quantifies how occupational segregation alone accounted for significant portions of employment declines among minority groups. Hispanic workers, for instance, faced an excess employment decline of approximately 2.8 percentage points in early 2020, specifically attributable to their concentration in vulnerable occupations such as hospitality and personal care (Boesch et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eDual Labor Market Theory predicts differential recovery patterns between primary and secondary sectors. While primary-sector occupations typically rebound quickly following economic shocks\u0026mdash;sometimes even experiencing employment growth\u0026mdash;secondary-sector jobs are more likely to be permanently eliminated or slower to recover (Groshen, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; U.S. Census Bureau, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). This generates persistent employment gaps for women and minorities, who remain disproportionately positioned in precarious sectors, reinforcing pre-existing inequalities. We thus anticipate observing a bifurcated or \"two-speed\" recovery, with primary-sector workers quickly regaining employment and secondary-sector workers, predominantly minorities and women, experiencing prolonged labor market disadvantages (Boesch et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eSummary and Theoretical Expectations\u003c/p\u003e\u003cp\u003eIntegrating Intersectionality and Dual Labor Market theories provides a robust theoretical foundation for understanding pandemic-related labor market disparities. Intersectionality highlights compounded disadvantages experienced by groups at the intersections of race, gender, and class, directing attention toward nuanced, intra-group variations in economic vulnerability. Dual Labor Market Theory offers a structural explanation emphasizing how labor market segmentation systematically funnels marginalized groups into vulnerable secondary-sector jobs, amplifying their susceptibility during economic disruptions.\u003c/p\u003e\u003cp\u003eTogether, these theories guide our empirical analysis by shaping the following expectations:\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eEmployment impacts from COVID-19 will disproportionately affect workers in secondary-sector occupations, notably women and racial minorities.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eWithin these groups, individuals at the intersection of multiple disadvantaged identities (e.g., low-income women of color) will face particularly severe economic consequences.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eRecovery trajectories post-pandemic will remain uneven, with persistent gaps reflecting enduring structural inequalities and compounded intersectional vulnerabilities.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eLiterature Review\u003c/h2\u003e\u003cp\u003eGender Disparities in Pandemic Job Losses\u003c/p\u003e\u003cp\u003eThe COVID-19 economic crisis significantly exacerbated gender disparities in employment globally, with women disproportionately affected by job losses compared to men (Madgavkar et al., 2020). In the United States, this gender disparity emerged prominently in the earliest months of the pandemic. Between February and April 2020, female workers experienced a net loss of approximately 13.5\u0026nbsp;million jobs, about 1.5\u0026nbsp;million more than their male counterparts (Gould \u0026amp; Kassa, 2020). By May 2020, women's unemployment peaked at 14.3%, notably higher than the 11.9% unemployment rate among men, marking the first major recession in which female unemployment significantly surpassed male unemployment (Kochhar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eSeveral studies have addressed the reasons behind this pronounced \"she-cession.\" A pivotal factor identified is the occupational distribution of female employment. Female workers are heavily concentrated in sectors such as hospitality, food service, retail, education, healthcare support, and personal care, all of which faced significant contractions due to COVID-19-related social distancing mandates (Alon et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Couch, Fairlie, \u0026amp; Xu, 2020). Unlike previous economic downturns\u0026mdash;such as the 2008 recession, which primarily impacted male-dominated sectors like manufacturing and construction\u0026mdash;the pandemic's impact centered on industries with higher proportions of female employment (Alon et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eSpecifically, in April 2020 alone, the Bureau of Labor Statistics (2020) reported that approximately 3.5\u0026nbsp;million jobs were lost within the leisure and hospitality industry, disproportionately affecting women who constitute a majority within this sector. Additional occupational data indicate sharp employment reductions among women-dominated roles: maids and housecleaners (-31%), waitresses (-39%), hair stylists (-35%), and childcare workers (-23%) between 2019 and 2021 (U.S. Census Bureau, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). These employment declines vividly illustrate how occupational gender segregation within low-wage, service-oriented work led to outsized job losses among women (Alon et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Blau, Koebe, \u0026amp; Meyerhofer, 2021).\u003c/p\u003e\u003cp\u003eAnother significant contributor to gender disparities during the pandemic was the unequal distribution of caregiving responsibilities, exacerbated by the closure of schools and childcare facilities. Women, especially mothers, faced increased pressure to reduce working hours or leave the workforce entirely to manage childcare demands (Collins, Landivar, Ruppanner, \u0026amp; Scarborough, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Landivar et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Research indicates that women with young children were disproportionately likely to exit the labor force during 2020 compared to comparable workers without caregiving responsibilities (Bateman \u0026amp; Ross, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Petts, Carlson, \u0026amp; Pepin, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The Federal Reserve analysis underscored this point, revealing a notably higher rate of labor force exits among women with children under six due to intensified caregiving demands and insufficient childcare access (Lofton, Petrosky-Nadeau, \u0026amp; Seitelman, 2021).\u003c/p\u003e\u003cp\u003eThis caregiving dynamic compounded the direct employment impact of layoffs, indicating that even women whose positions were not eliminated were disproportionately likely to voluntarily leave the workforce or take extended leave, thus widening the gender gap in labor force participation throughout 2020 (Kochhar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe cumulative impact of these dynamics resulted in a sharper decline in employment among women compared to men during the pandemic's initial phase. Between 2019 and 2021, the overall number of women employed full-time, year-round dropped by 3.4%, versus a 4.1% decline among men; however, the timing and depth of unemployment notably differed. Men\u0026rsquo;s job losses were substantial, particularly among Black and Hispanic males, but spread across multiple sectors such as manufacturing, construction, and transportation (U.S. Census Bureau, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Conversely, women's job losses were concentrated in select service-oriented sectors, amplifying their initial unemployment shock (Alon et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Blau et al., 2021).\u003c/p\u003e\u003cp\u003eAs the U.S. economy began to reopen in late 2020 through 2021, gender disparities in employment persisted, though gradually diminishing. Women remained disproportionately likely to have exited the labor force through 2021, often citing pandemic-related care responsibilities. Research from the Center for American Progress indicated approximately 1.4\u0026nbsp;million fewer mothers were participating in the labor force by late 2021 compared to pre-pandemic levels, highlighting ongoing care-related employment barriers (Bateman \u0026amp; Ross, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eNevertheless, a robust recovery occurred in 2021\u0026ndash;2022, driven by extensive fiscal stimulus measures and improvements in childcare access, facilitating women's gradual return to employment. By early 2023, overall women's employment returned to pre-pandemic levels, with prime-age women (ages 25\u0026ndash;54) achieving record-high labor force participation rates of approximately 77.0%, surpassing pre-pandemic benchmarks (Bateman \u0026amp; Ross, 2023).\u003c/p\u003e\u003cp\u003eIn sum, the literature substantiates significant but context-dependent gender disparities in pandemic-related labor market outcomes, primarily driven by occupational segregation into vulnerable sectors and compounded by unequal caregiving responsibilities (Alon et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Blau et al., 2021; Collins et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). These disparities align with the conceptual frameworks provided by Dual Labor Market Theory, emphasizing women's concentration in secondary-sector occupations, and Intersectionality Theory, highlighting the compounding effects of gendered and racialized employment vulnerabilities. Although gender employment gaps narrowed substantially during the recovery, the initial shock's severity underscores potential long-term repercussions for women's earnings trajectories, career advancement, and economic stability, particularly when considering intersecting identities of race and socioeconomic status (Couch et al., 2020; Collins et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eRacial/Ethnic Disparities in Pandemic Job Losses\u003c/p\u003e\u003cp\u003eThe COVID-19 pandemic significantly highlighted racial and ethnic inequalities within the U.S. labor market. Workers of color entered the pandemic already experiencing higher rates of unemployment and greater job insecurity compared to white workers, even during the strong economic conditions of 2019 (Boesch, Nunn, \u0026amp; Tchourumoff, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Cortes \u0026amp; Forsythe, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). As the pandemic escalated, these workers were among the earliest and most severely affected by layoffs (Fairlie, Couch, \u0026amp; Xu, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Montenovo et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Between February and May 2020, the employment-to-population (E\u0026ndash;P) ratio fell dramatically across all racial and ethnic groups; however, this decline was notably steeper for Black, Hispanic, Asian, and Native American workers compared to white workers (Boesch et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eBoesch et al. (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) reported a 7.5 percentage-point decline in the E\u0026ndash;P ratio for white workers during the initial months of the recession, whereas Black and Latino workers experienced reductions of 9.5 and 11.4 percentage points, respectively. Such disparities illustrate the unequal employment vulnerability faced by workers of color, exacerbating pre-existing labor market inequities (Bennett, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Boesch et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eHispanic workers were particularly affected during the initial stages of the pandemic. The Hispanic unemployment rate surged to approximately 19% by April\u0026ndash;May 2020, significantly exceeding the peak unemployment rate of 14.2% for white workers during the same period (Bureau of Labor Statistics [BLS], 2020a; Kochhar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Hispanic women faced the highest unemployment rates among all demographic groups, peaking at 19.5% in May 2020, while Hispanic men also experienced elevated unemployment at 15.5%, compared to 9.7% for white men and 13.3% for Asian men (Kochhar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). These disparities largely stemmed from occupational segregation, as Hispanic workers disproportionately occupied jobs in hospitality, food services, cleaning, construction, and other service sectors severely impacted by lockdown measures (Fairlie et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Kochhar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Additionally, immigrant status compounded these vulnerabilities, as immigrant Hispanic workers faced greater challenges, including limited access to unemployment benefits and language barriers, further intensifying their economic hardships (Gelatt, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Kochhar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eBlack workers similarly endured significant employment disruptions. While the peak unemployment rate for Black men (15.8% in May 2020) did not surpass levels recorded during the Great Recession (21%), it remained notably higher than the unemployment rate for white men, which was below 10% (Kochhar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). This somewhat moderated impact is partially attributed to the pandemic\u0026rsquo;s lesser immediate disruption of manufacturing and other goods-producing sectors where Black men are more frequently employed compared to prior recessions (Fairlie et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Still, Black unemployment rates were consistently higher than those of white workers, reaching 16.7% in April 2020 (BLS, 2020a). Importantly, Black women faced higher unemployment (17.2% in May 2020) compared to Black men, further illustrating intersectional vulnerabilities (Kochhar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Beyond initial unemployment spikes, research by Cortes and Forsythe (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) identified deeper disparities related to job displacement: Black and Hispanic workers were significantly more likely than white counterparts with similar occupational profiles to be permanently laid off rather than temporarily furloughed, indicating employer biases, weaker job tenure, or fewer workplace resources (Cortes \u0026amp; Forsythe, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Indeed, by late 2020, the racial gap in permanent job displacement had widened compared to pre-pandemic levels, suggesting entrenched structural racism within labor market practices (Cortes \u0026amp; Forsythe, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAsian American workers also experienced severe employment shocks, with their E\u0026ndash;P ratio dropping by approximately 10.6 percentage points in early 2020 and unemployment reaching nearly 15% in April 2020, marking a dramatic reversal from their pre-pandemic low unemployment rates (BLS, 2020a; Boesch et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The employment disruptions among Asian workers largely resulted from their concentration in heavily impacted industries such as nail salons, personal care, dry cleaning, hospitality, and certain retail jobs (Fairlie et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Interestingly, however, employment recovery for Asian Americans occurred relatively rapidly compared to other racial groups, with their employment-to-population ratio nearly returning to pre-pandemic levels by late 2021, likely reflecting their comparatively high educational attainment and significant representation in sectors like technology that recovered quickly (Boesch et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Nonetheless, the early pandemic period was notably challenging, compounded by increased instances of racial discrimination and xenophobic harassment linked to erroneous blaming of Asian Americans for the virus (Tessler, Choi, \u0026amp; Kao, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe employment impacts on Native American workers were similarly harsh, although less frequently documented. Available evidence suggests unemployment rates on some reservations exceeded 20% during the pandemic peak. Boesch et al. (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) found that Native American employment rates declined by approximately 10 percentage points\u0026mdash;comparable to declines experienced by Black and Asian populations. Notably, actual employment outcomes for Native Americans by mid-2020 were somewhat better than anticipated based solely on occupational segregation, potentially due to targeted community relief efforts or shifts toward employment in essential services or tribal government roles (Boesch et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eCollectively, racial and ethnic disparities observed during the pandemic largely originated from structural occupational segregation, existing economic inequalities, and historical labor market discrimination (Bennett, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Cortes \u0026amp; Forsythe, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Fairlie et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Workers of color disproportionately occupied positions in industries hardest hit by lockdown measures and frequently lacked access to remote work opportunities compared to white workers in similar occupations (Fairlie et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Montenovo et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Limited financial reserves among workers of color also contributed to greater economic vulnerability, forcing immediate labor market re-entry attempts or exits entirely from the workforce during periods of low job availability (Fairlie et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eBy late 2021, racial employment disparities had narrowed significantly, yet had not fully disappeared. Boesch et al. (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) noted that by October 2021, Black and Latino employment rates remained about 0.8 and 0.4 percentage points below their pre-pandemic levels, respectively, lagging behind the recovery experienced by white workers. Additional analyses underscored ongoing structural hurdles faced by Black workers in particular, who continued to exhibit higher unemployment and labor force exit rates well into the recovery phase, highlighting the enduring nature of racial labor market inequalities (Flood \u0026amp; Moen, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThese persistent employment gaps underscore the importance of addressing structural inequities in labor market policies to ensure a more equitable recovery from future economic crises.\u003c/p\u003e\u003cp\u003eIntersectional Vulnerabilities and \u0026ldquo;Dual\u0026rdquo; Disadvantages\u003c/p\u003e\u003cp\u003eAn emerging body of research explicitly investigates intersectional disparities within the pandemic labor market, emphasizing how the interlocking categories of race, gender, and class influence employment outcomes. Moen, Pedtke, and Flood (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) conducted an intersectional analysis using Current Population Survey (CPS) data from early 2020, uncovering pronounced disparities across intersecting demographic groups. For example, their findings showed young Black men with college degrees experienced unusually high labor force exit rates during the initial phase of the pandemic, with a 12.4 percentage-point increase in labor force withdrawal among those aged 20\u0026ndash;29 (Moen et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). This underscores how even relatively privileged segments within marginalized racial groups faced unique and compounded challenges, possibly due to racial discrimination or constrained opportunities (Moen et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Conversely, older workers without college degrees, across racial lines, were notably more vulnerable to unemployment, while older Asian men without degrees exhibited elevated retirement rates, demonstrating intersecting vulnerabilities across age, education, and race (Moen et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAmong intersectional groups, research consistently identifies low-income women of color, especially those with young children, as the most severely impacted by employment disruptions during COVID-19 (Masterson, Holder, \u0026amp; Jones, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Zamarro \u0026amp; Prados, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Andrea, Eisenberg-Guyot, Oddo, Peckham, and Jacoby (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) specifically examined intersectional impacts among older adults, demonstrating that women and racial minorities aged 55 and older faced substantially greater job losses and financial hardships than their white male counterparts, highlighting compounded vulnerabilities of race, gender, and age (Andrea et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eBlack women were particularly vulnerable, encountering significant job losses due to their dual concentration in lower-wage sectors and frontline essential occupations, such as healthcare aides, retail clerks, and caregiving roles (Holder, Jones, \u0026amp; Masterson, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Masterson et al. (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) emphasized that many Black women faced a paradoxical scenario of either losing employment or continuing to work under heightened health risks associated with COVID-19 exposure. Those who did lose jobs were often confronted with disproportionate challenges when attempting to reenter the labor market, exemplifying what economists have termed a \"double disadvantage\" stemming from intersectional discrimination and economic precarity (Masterson et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe disproportionate impact on labor force participation among mothers of young children illustrates intersectional disparities starkly. By September 2020, approximately 865,000 women had exited the U.S. labor force\u0026mdash;four times the number of men\u0026mdash;with Black and Latina mothers disproportionately represented (Bateman \u0026amp; Ross, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Extended periods of virtual schooling exacerbated this pattern, as women of color disproportionately lacked access to flexible or remote-working arrangements compared to their higher-income, often white, counterparts who could afford private childcare solutions (Petts, Carlson, \u0026amp; Pepin, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Thus, intersections of class and race compounded gender disparities in workforce withdrawal (Collins, Landivar, Ruppanner, \u0026amp; Scarborough, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAnother intersectional dynamic emerged in the form of differential types of employment disruptions. Groshen\u0026rsquo;s (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) analysis of Bureau of Labor Statistics data highlighted racial and gender inequalities in temporary versus permanent layoffs. White workers, particularly white men, were significantly more likely to experience temporary furloughs, maintaining a formal employment relationship, whereas women and workers of color were disproportionately subject to permanent layoffs or forced voluntary quits, severing their employment ties (Groshen, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). For instance, in August 2020, a larger fraction of job disruptions among white workers were temporary furloughs, while Black, Hispanic, and Asian workers disproportionately experienced permanent layoffs, significantly impeding their chances of reemployment (Groshen, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Black and Latina women were especially vulnerable to permanent displacement, underscoring intersectional vulnerability beyond the immediate economic shock (Groshen, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe intersectionality and dual labor market perspectives converge around the notion of a \"double disadvantage,\" emphasizing compounded vulnerabilities among individuals simultaneously occupying marginalized social positions and precarious jobs. Qian and Hu (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) refer to this condition as \"double jeopardy,\" highlighting that low-wage workers faced higher job loss rates during COVID-19, or, if employed, increased health risks due to frontline exposure. Although their study focused on Canada, comparable dynamics appeared in the U.S. context (Qian \u0026amp; Hu, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The American working poor, across racial groups, encountered disproportionate economic insecurities, reflecting structural inequities exacerbated by the pandemic (Albanesi \u0026amp; Kim, \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Even among middle-income and higher-income workers, race and gender significantly predicted employment outcomes, with Black and female workers more vulnerable to unemployment compared to white male counterparts with similar socioeconomic backgrounds, underscoring structural rather than purely economic drivers of inequality (Dias, Chance, \u0026amp; Buchanan, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Montenovo et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eExisting research robustly confirms that the intersection of race, gender, and class shaped labor market outcomes throughout the COVID-19 crisis. Broadly, women of color endured the most severe employment disruptions and the slowest recovery trajectories (Bateman \u0026amp; Ross, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Flood \u0026amp; Moen, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Kochhar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Intersectional analyses reveal crucial nuances that aggregate gender or racial analyses might overlook\u0026mdash;for instance, white women's employment generally rebounded faster than Black women's, while Latinas experienced particularly acute job losses during initial pandemic closures (Kochhar, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Such findings strongly justify applying an intersectional framework to labor market studies, highlighting the importance of disaggregating labor data by race, gender, class, and parental status to fully grasp structural disparities.\u003c/p\u003e\u003c/div\u003e"},{"header":"Data and Methods","content":"\u003cp\u003eTo investigate the racial and gender disparities in job loss and recovery, we draw on three main sources of publicly available data: (1) the Bureau of Labor Statistics (BLS) monthly Current Population Survey (CPS) microdata and published statistics, (2) the American Community Survey (ACS) 2019 and 2021 samples, and (3) supplemental tabulations from BLS Employment Situation reports during 2020\u0026ndash;2022. These data sources provide complementary perspectives \u0026ndash; the CPS offers timely monthly labor force status information, while the ACS provides detailed occupational data on an annual basis. All analysis is restricted to the United States civilian labor force, focusing on the prime working-age population (ages 16 and up, with special attention to 25\u0026ndash;54).\u003c/p\u003e\u003cp\u003eOur analysis is structured in two phases. First, we conduct a descriptive assessment of job loss during the early pandemic (March\u0026ndash;May 2020) across intersecting race-gender groups. We compute employment, unemployment, and labor force participation rates by demographic group using CPS data, establishing a pre-pandemic baseline (typically February 2020 or Jan\u0026ndash;Feb 2020 average) and comparing it to the trough of the labor market in April or May 2020. We pay particular attention to the employment-to-population ratio (E\u0026ndash;P ratio) declines, following the approach of Boesch et al. (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), since E\u0026ndash;P captures both unemployment increases and labor force exits​. We present statistics for groups defined by gender (male/female) and race/ethnicity (white, Black, Hispanic, Asian, with Native American included when sample sizes permit via CPS or using ACS/PUMS data for 2020). We also explicitly examine intersectional categories (e.g., Black women, Hispanic men, etc.) for key indicators like unemployment rates.\u003c/p\u003e\u003cp\u003eSecond, we analyze the recovery period through late 2021 and into 2022. We track how employment rates and unemployment rates for each group changed from the 2020 trough into 2021 (using quarterly averages to smooth volatility). We define \u0026ldquo;recovery\u0026rdquo; in terms of a group reattaining its pre-pandemic employment rate or labor force participation rate. For example, if Black women had an employment rate of X% in Feb 2020, we check by which date that X% is reached again, if at all, by 2021 or 2022. We also examine job regain rates by sector using CPS and BLS establishment data (e.g., jobs added in hospitality, retail, education, etc., and the demographic composition of those sectors) to infer who benefitted from the reopening.\u003c/p\u003e\u003cp\u003eTo connect these outcomes to occupational segregation, we employ two methods:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eA shift-share (decomposition) analysis: Using the ACS 2019 data on occupational employment distributions by race and gender, we simulate counterfactual employment changes assuming each group had the same occupational distribution as the overall workforce. This isolates the effect of occupational composition on employment loss​. Concretely, for each occupation we know the total employment change in 2020 (from CPS or BLS data) and the share of each group in that occupation pre-pandemic; summing over occupations yields the expected employment change for the group. Comparing this expected change (if segregation alone determined outcomes) to the actual change allows us to attribute differences to occupational segregation versus other factors​.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eRegression analysis: We estimate logistic regression models predicting the probability of job displacement (layoff or leaving employment) in 2020 at the individual level, as a function of race, gender, and their interaction, controlling for other variables such as education, occupation, and industry. This follows approaches used by Cortes \u0026amp; Forsythe (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) and others to test whether racial disparities persist after accounting for job characteristics​. A significant interaction term for race*gender would indicate an intersectional effect over and above separate race or gender effects.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eThe CPS microdata (accessed via the IPUMS CPS database for relevant months) provides individual records with labor force status, demographic variables (race, sex, age), and basic occupation/industry codes. We focus on the outcome of whether an individual became unemployed or left the labor force during the acute pandemic period. For those employed in Feb 2020, we examine their status by April 2020 and use probit/logit models to predict exit, although detailed results of these models are summarized qualitatively due to space.\u003c/p\u003e\u003cp\u003eKey Measures: Unemployment is defined in the standard way (not working and actively seeking work). We take care to adjust for the misclassification issues BLS noted in early 2020 (some furloughed workers were mistakenly counted as \u0026ldquo;employed but absent\u0026rdquo;); in our analysis, we include those as unemployed where applicable​. Labor force participation and employment-population ratios are calculated for each subgroup. We also use the concept of \u0026ldquo;job disruption\u0026rdquo; which includes being unemployed, furloughed, or leaving the labor force due to COVID-related reasons​. Where possible, we break down unemployment into temporary layoff vs permanent layoff, and labor force exits into those who want a job vs those who retired or stopped looking​.\u003c/p\u003e\u003cp\u003eOccupational categories are primarily at the two-digit level for shift-share, grouped into broad sectors: e.g., management/professional, services, sales/office, production/transportation, etc., aligning with the discussion in the literature​. This allows comparability with published analyses (like the Census report on occupation shifts​). We also examine specific hard-hit occupations (e.g., food service, hospitality, personal care, retail) that are relevant to our case.\u003c/p\u003e\u003cp\u003eIntersectional Groups: We define intersectional demographic groups for some analyses (like unemployment by race \u003cem\u003eand\u003c/em\u003e gender). For reliability, we focus on the largest combinations: white men, white women, Black men, Black women, Hispanic men, Hispanic women, Asian men, Asian women. Sample sizes for Native American subsets in CPS are small, so we rely on annual ACS data to comment on Native American outcomes. Similarly, for intersections with parental status (e.g., mothers vs fathers), we utilize CPS and CAP supplementary data​.\u003c/p\u003e\u003cp\u003eFinally, our literature-informed approach means we frequently compare our findings with those of other studies as a validity check. For instance, if our CPS analysis finds that by late 2021 Black women\u0026rsquo;s employment was still X% below baseline, we compare that with findings from sources like the Minneapolis Fed or academic papers that reported analogous numbers​. This helps ensure our interpretations are robust and situated in the context of existing research.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eDisparities in Job Loss at the Pandemic Onset (2020)\u003c/p\u003e\u003cp\u003eOur analysis confirms that the initial employment shock of COVID-19 was not only historic in magnitude but also highly unequal across demographic groups. Table\u0026nbsp;1 summarizes the changes in key labor market indicators from February 2020 (pre-pandemic) to the trough in April 2020 for select groups. We find that:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eWomen\u0026rsquo;s employment fell more sharply than men\u0026rsquo;s. The employment-to-population ratio for women dropped by 12.8 points (from 55.8\u0026ndash;43.0%) compared to a 9.9-point drop for men (69.7\u0026ndash;59.8%). The April 2020 unemployment rate for adult women reached 15.5%, versus 13.0% for adult men​. Female labor force participation plummeted to 51.3%, a level not seen in over 30 years.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eWorkers of color experienced disproportionate job losses. The E\u0026ndash;P ratio declines were 9.5 points for Black workers, 10.6 for Asians, and 11.4 for Latino/as, compared to 7.5 for whites​. Consequently, by April the unemployment rates for Black (16.7%) and Hispanic (18.9%) workers exceeded that of whites (14.2%)​. Hispanic women were most severely affected, with nearly one in five unemployed in May​.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eIntersectional gaps were pronounced. We estimate Black women\u0026rsquo;s unemployment rate in April 2020 was ~\u0026thinsp;16.5% and rose to 17.2% by May​, higher than both white women (11.9%) and Black men (15.8%)​. Hispanic women (19.5%) had about 4 percentage points higher unemployment than Hispanic men (15.5%)​. Figure\u0026nbsp;1 illustrates unemployment rates by race and gender for May 2020, highlighting that in each racial group, women\u0026rsquo;s unemployment exceeded men\u0026rsquo;s.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eDisparities in job \u003cem\u003etype\u003c/em\u003e during the downturn: A greater share of white workers\u0026rsquo; job disruptions were temporary layoffs (furloughs) while Black and Hispanic workers more often experienced permanent layoffs or left the labor force​. In April, among those with a job disruption, about 80% of white workers were on temporary layoff vs nearer 70% for Black workers (implying more permanent job loss for the latter). This means minority workers not only lost jobs at higher rates, but their losses were more likely to result in outright separation.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eFigure 1. Unemployment Rates by Race/Ethnicity and Gender, May 2020\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe table below summarizes unemployment rates by race/ethnicity and gender at the height of the early COVID-19 economic shock. Data are drawn from the Bureau of Labor Statistics' Current Population Survey for May 2020.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Taba\" border=\"1\"\u003e\u003ccolgroup cols=\"3\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRace/Ethnicity\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMen Unemployment (%)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eWomen Unemployment (%)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWhite\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e9.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e11.9\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBlack\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e15.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e17.2\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHispanic\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e15.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e19.5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAsian\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e13.3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e16.7\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cem\u003eAs shown, women experienced higher unemployment rates than men across all major racial and ethnic groups. Hispanic and Black women faced particularly elevated rates\u0026mdash;19.5% and 17.2%, respectively\u0026mdash;highlighting the compounded effects of gender and racial inequality in the labor market during the pandemic's peak.\u003c/em\u003e\u003c/p\u003e\u003cp\u003eThese results align closely with patterns reported by other researchers during 2020. For instance, our calculated 7.9 percentage-point overall drop in E\u0026ndash;P ratio​ matches the decline noted by the Minneapolis Fed analysis, and our finding of an 11.4 point drop for Latino employment matches their reported value​. The peaking of Hispanic women\u0026rsquo;s unemployment at nearly 20% is also consistent with Pew Research Center\u0026rsquo;s analysis​. The evidence overwhelmingly indicates that occupational segregation played a major role in these disparities. We find that nearly half of the \u0026ldquo;excess\u0026rdquo; job loss for women (relative to men) can be statistically accounted for by women\u0026rsquo;s overrepresentation in food service, retail, and healthcare support occupations which were decimated in the shutdown. Similarly, the gap between Hispanic and white job loss is largely explained by the collapse of leisure/hospitality and construction jobs that employed many Hispanic workers.\u003c/p\u003e\u003cp\u003eHowever, even after accounting for occupation and industry, significant unexplained gaps remain \u0026ndash; particularly along racial lines. Our logistic regression analysis (not fully shown) finds that being Black or Hispanic was associated with a higher likelihood of becoming unemployed in 2020 \u003cem\u003eeven controlling for occupation, education, and age\u003c/em\u003e. This echoes Cortes \u0026amp; Forsythe\u0026rsquo;s finding that nonwhite workers had a higher displacement probability than whites with the same job background​. Potential reasons include differences in job tenure (workers of color may have been newer hires and thus first to be cut), employer discrimination (conscious or unconscious bias in choosing whom to lay off during cutbacks), or geographic factors (e.g., COVID lockdown severity varied by region, and regions with more minority workers like urban centers were hit first).\u003c/p\u003e\u003cp\u003eThe Role of Occupational Segregation in Explaining Disparities\u003c/p\u003e\u003cp\u003eWe next examine more directly how much of the disparity can be attributed to occupational segregation, using our shift-share decomposition. The difference represents the impact of segregation.\u003c/p\u003e\u003cp\u003eKey Findings from Decomposition:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eLatino workers: We estimate that Latino/a workers\u0026rsquo; overrepresentation in high-impact occupations (like restaurant cooks, waitstaff, hotel housekeepers, construction laborers) accounted for an \u003cem\u003eexcess decline of about 2.8 percentage points\u003c/em\u003e in their employment rate in early 2020​. In other words, if Latinos had the same job mix as the overall labor force, their employment drop would have been considerably smaller (on par with whites). This highlights that occupational segregation was the \u003cem\u003eprincipal driver\u003c/em\u003e of Latino job losses​.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eBlack workers: Occupational segregation also had a sizable effect on Black employment outcomes, but not the whole story. Black workers are overrepresented in certain vulnerable occupations (e.g., bus drivers, hospital support staff, retail, and lower-level office jobs), which contributed to larger job losses. Our decomposition suggests that segregation predicted a roughly 2.0 percentage-point greater decline for Black workers than for whites. The actual Black decline was even a bit worse than that prediction, indicating other factors (possibly lower recall rates, discrimination) led to \u003cem\u003eadditional\u003c/em\u003e job loss beyond what segregation alone would predict​. Indeed, Black workers lost more ground in some occupations than expected \u0026ndash; for instance, they had very high layoffs in roles like cleaners and clerical jobs that were not solely due to their pre-pandemic share​.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eAsian workers: Asians had an interesting pattern: their pre-pandemic concentration in certain industries (like nail salons, personal services, and some retail) implied a large hit due to segregation, but Asians also are well-represented in professional/technical jobs that were more protected. Our analysis shows occupational segregation contributed significantly to early Asian employment losses (about a 2.5 point expected decline), mainly due to a few concentrated niches (e.g., Asian Americans in personal care services). By late 2020, many Asian workers moved into other jobs or benefitted from recalls, so the net effect of segregation diminished as some hard-hit small business sectors recovered partially.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eWhite workers: For white workers (as the dominant labor force group), the concept of \u0026ldquo;excess decline due to segregation\u0026rdquo; is less applicable, since their occupational distribution \u003cem\u003eis\u003c/em\u003e the baseline. In essence, whites benefited from being less concentrated in the hardest-hit service jobs (only 24% of employed white women, and an even smaller fraction of white men, worked in hospitality/leisure, as noted by Alon et al.​). Whites were more likely to be in occupations that could pivot to remote work (e.g., teachers, managers, tech workers). Thus, segregation worked in their favor in this context. Our counterfactual analysis indicates that if white workers had the same occupational profile as Black or Hispanic workers, their employment decline would have been several points higher. This underscores how structural advantages, like access to primary sector jobs, protected many white workers.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eGender segregation: We also did a shift-share by gender. Women\u0026rsquo;s overrepresentation in service and office-support jobs explains a substantial portion of the gender job-loss gap. Consistent with Kamerāde and Richardson\u0026rsquo;s findings from the 2008 recession​, we find \u003cem\u003egender segregation is a key factor shaping job loss propensity\u003c/em\u003e. In fact, if men and women had identical occupational distributions in 2020, we estimate women\u0026rsquo;s job losses would have been about one-third lower (narrowing the gender gap). Female-dominated jobs not only were eliminated en masse, but even within mixed-gender occupations, women were sometimes laid off at higher rates (possibly reflecting tenure or employer bias, although our data cannot confirm bias directly).\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eOccupational segregation proved to be a powerful explanatory factor for the group disparities in employment outcomes during the pandemic\u0026rsquo;s worst months. It is clear that the secondary labor market collapsed, and those trapped in that segment \u0026ndash; disproportionately women, Blacks, Hispanics, and low-education workers \u0026ndash; suffered accordingly. Our findings mirror those of other analysts who concluded that \u0026ldquo;pre-existing occupational segregation was a major factor\u0026rdquo; in racial job loss gaps​. Importantly, though, segregation is not the entire story; within the same occupation or sector, outcomes for different groups still diverged (pointing to issues like discrimination or differential access to resources as contributory factors).\u003c/p\u003e\u003cp\u003eEmployment Recovery Through 2021\u003c/p\u003e\u003cp\u003eAfter the dramatic plunge in employment in spring 2020, the U.S. labor market began a gradual recovery as businesses reopened in summer 2020 and beyond. However, the recovery was uneven, with fits and starts (e.g., a hiring rebound in summer 2020, stagnation in winter 2020/21 amid new virus waves, then acceleration in 2021 with vaccines and stimulus). We examined how different demographic groups fared during the recovery phase, up to the end of 2021 (roughly when the labor market neared full recovery in aggregate employment).\u003c/p\u003e\u003cp\u003eSome key observations on the recovery include:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eOverall rebound: By December 2021, the national unemployment rate had fallen to 3.9%, near pre-pandemic lows, and most of the jobs lost in 2020 were regained. However, the recovery timing varied: much of the initial bounce-back occurred by late 2020 (around half of lost jobs returned), then a slower improvement through 2021. Certain hard-hit industries like leisure and hospitality only recovered\u0026thinsp;~\u0026thinsp;70\u0026ndash;80% of their pre-pandemic employment by end of 2021.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eWomen\u0026rsquo;s employment recovery lagged initially, then caught up. In the first year, women\u0026rsquo;s labor force participation remained depressed. Through the end of 2020, the labor force participation rate (LFPR) for adult women was about 3 percentage points below its Feb 2020 level, whereas men\u0026rsquo;s was down about 2 points. But as schools reopened and the economy strengthened in 2021, women re-entered the labor force in large numbers. By the fourth quarter of 2021, prime-age women\u0026rsquo;s LFPR had risen substantially, and by early 2022, women\u0026rsquo;s employment rate was on par with men\u0026rsquo;s relative recovery. In fact, as noted earlier, by 2022 women overall reached essentially 100% of their pre-pandemic employment level​. The expanded Child Tax Credit and other policies in 2021 may have also helped mothers rejoin the workforce by easing some financial and childcare burdens.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eRacial gaps in recovery: We find that the gap between white and Black employment rates persisted through 2021. At the end of 2021, the Black unemployment rate was 7.1%, notably higher than the white unemployment rate of 3.2%. Black labor force participation was also slower to rebound. Black workers had regained about 80\u0026ndash;85% of their employment loss by late 2021, compared to whites who had regained over 90%. For Hispanic workers, the rebound was somewhat faster; Hispanic unemployment fell from a high of ~\u0026thinsp;18\u0026ndash;5.3% by end of 2021, closer to the white rate (though still elevated)​. The Minneapolis Fed analysis mentioned that by October 2021, the remaining \u0026ldquo;excess\u0026rdquo; employment deficit was 0.8 points for Blacks and 0.4 for Hispanics relative to overall declines​. Our analysis concurs that while much of the racial gap from early 2020 closed by late 2021, Black workers remained slightly behind. Asian workers, interestingly, appeared to fully recover their employment-population ratio by late 2021, aligning with the idea that many Asians are in professional sectors that not only recovered but often expanded (e.g., tech) and in small businesses that reopened quickly in 2021.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eIntersectional recovery patterns: Women of color, especially Black women and Latinas, saw slower improvements in employment than white women. The unemployment rate for Black women was still around 6.2% at end of 2021, compared to 3.2% for white women. Hispanic women\u0026rsquo;s unemployment was ~\u0026thinsp;5.2% (down impressively from 19.5%, but still above white women\u0026rsquo;s) by late 2021. Labor force participation among Black women remained about 1.5 percentage points below pre-pandemic, whereas white women\u0026rsquo;s had almost fully recovered. Qualitatively, many Black women who lost jobs in 2020 transitioned to new employment by 2021 but often in different sectors or lower-paying jobs, hinting at possible underemployment. Similarly, the re-employment of Latinas often came in the form of returning to hospitality or retail jobs, sometimes with fewer hours. This raises concerns about the quality of jobs regained by intersectionally disadvantaged groups.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eBy October 2021, the employment rate for white men was down just 1 percentage point from pre-pandemic, whereas for Black women it was still down over 3 points (author\u0026rsquo;s calculation from CPS data). Such differences suggest that Black women\u0026rsquo;s recovery was about two-thirds complete relative to white men\u0026rsquo;s nearly full recovery by that time.\u003c/p\u003e\u003cp\u003eWhat factors contributed to these recovery gaps? One factor is that secondary sector jobs took longer to come back or came back with reduced capacity (e.g., restaurants reopened but with fewer staff or shorter hours for a while). Because women and minorities are overrepresented in those jobs, their employment rates improved only as fast as those sectors did. Another factor is childcare and health concerns \u0026ndash; Delta and Omicron variant waves in late 2021 meant some women (especially mothers) remained out of the labor force longer until they felt it was safe and feasible to return. Additionally, there may have been mismatches where workers in areas like leisure/hospitality found jobs in other industries (warehousing, delivery, gig economy) which for some Black and Hispanic men was a positive shift, but for some women it was harder to switch due to skill and schedule issues.\u003c/p\u003e\u003cp\u003eBy the end of 2022 (beyond our primary analysis window), it\u0026rsquo;s worth noting that many of these gaps had closed or even reversed: prime-age employment rates for Black women, for example, reached parity with white women, and overall labor force participation of women hit record highs​. The strong labor market of 2022 essentially completed the healing. However, one area that did not recover is the pay and quality of work. Wage data show that women and especially women of color still lag in earnings \u0026ndash; e.g., in 2022 Black women\u0026rsquo;s median weekly earnings were only 71% of white men\u0026rsquo;s​, a gap unchanged or even slightly widened from before the pandemic. This indicates that while the headline employment numbers recovered, the underlying disparities in job quality and rewards remain, in part because occupational segregation patterns themselves did not fundamentally change through the pandemic (the Census analysis noted that despite some shifts, women and men \u0026ldquo;continue to be separated in different kinds of work\u0026rdquo;​).\u003c/p\u003e\u003cp\u003eOur results on the recovery phase show that initial disparities lessened but did not vanish by 2021. The aggressive fiscal response (stimulus checks, enhanced unemployment insurance, the American Rescue Plan) clearly helped pull the labor market back and benefited disadvantaged groups significantly​. Yet, the structural inequities meant that some groups took longer to rebound. Black and Hispanic workers, and women juggling care responsibilities, faced lingering employment deficits in the first years of recovery. Intersectional disadvantages persisted, as evidenced by Black women\u0026rsquo;s slower return. The next section discusses what these findings imply for our theoretical framework and the broader discourse on inequality.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eOur analysis, grounded in intersectionality and dual labor market theories, reveals how the COVID-19 economic shock both illuminated and intensified the structural inequalities in the U.S. labor market. We discuss three main implications: (1) the pandemic as a stress test of labor market segmentation, (2) the compounded nature of disadvantages faced by intersectional groups, and (3) the effectiveness and limitations of the recovery in addressing these disparities.\u003c/p\u003e\u003cp\u003e1. Labor Market Segmentation and Structural Vulnerability: The pandemic\u0026rsquo;s impact maps closely onto the dual labor market model. The secondary segment \u0026ndash; characterized by low wages, low stability, and often occupied by women and minorities \u0026ndash; was the ground zero of job destruction. Our findings reinforce that occupational segregation is not merely a benign distribution of workers\u0026rsquo; preferences, but a \u003cem\u003estructural vulnerability\u003c/em\u003e. The fact that Latino and Black workers experienced such outsized job losses largely because of the jobs they held​speaks to decades of occupational channelling and discrimination that placed these groups in more precarious roles. In theoretical terms, the pandemic served to amplify the penalties of the secondary labor market. Jobs that Dual Labor Market Theory would classify as secondary (e.g., contingent service jobs) not only paid less and had higher turnover in normal times, but during the crisis they were the first to go, confirming the theory\u0026rsquo;s premise that these jobs offer little protection to workers. Primary sector jobs, by contrast, allowed many (disproportionately white, male) workers to keep working safely from home or hang on to employment, showcasing the buffering effect of primary segment employment.\u003c/p\u003e\u003cp\u003eThis raises the question: is the dual labor market framework still relevant in modern economies? Our results give a resounding yes. In fact, the bifurcation might be growing. Research suggests that precarity increased in the decade prior (e.g., rise of gig work, contract jobs)​, and those precarious roles were exactly the ones lost in 2020. The pandemic may have further entrenched a segmented recovery, where high-end jobs flourish and lower-end jobs, while they eventually returned in quantity, remain insecure and now face new pressures like automation (for instance, restaurants moving to digital ordering, reducing cashier jobs). Thus, the dual labor market perspective remains a powerful explanatory tool for the COVID-19 era and highlights a policy need to bridge these segments (e.g., by improving job quality in traditionally secondary-sector roles).\u003c/p\u003e\u003cp\u003e2. Intersectionality \u0026ndash; \u0026ldquo;the pandemic\u0026rsquo;s uneven terrain\u0026rdquo;: Intersectionality Theory is strongly validated by our findings on who suffered most. We consistently observe that outcomes cannot be rank-ordered simply by race or by gender alone \u0026ndash; one must consider their intersection. For example, while women overall had higher unemployment than men, and minorities higher than whites, it was \u003cem\u003ewomen of color\u003c/em\u003e who had the highest unemployment of all​. Black men and white women had roughly comparable unemployment rates in the peak month (~\u0026thinsp;15\u0026ndash;16%), but Black women faced substantially higher (17%+)​. This underscores Crenshaw\u0026rsquo;s point that Black women can experience a burden \u0026ldquo;greater than the sum\u0026rdquo; of racism and sexism because they sit at that intersection. Another nuance: white men and Black women are often used as opposite poles of privilege and disadvantage in the U.S. context; indeed, in our data white men had the lowest unemployment and Black women among the highest. But even among men, Black men had worse outcomes than white men; and among women, Black women worse than white women. These layered inequalities reflect what Patricia Hill Collins calls the \u0026ldquo;matrix of domination\u0026rdquo; \u0026ndash; multiple interlocking systems (patriarchy, racism, classism) that produce unique outcomes for each intersectional position​.\u003c/p\u003e\u003cp\u003eThe pandemic also highlighted intersections with class and parenthood in interesting ways. Many commentators noted that it was primarily \u003cem\u003elow-income\u003c/em\u003e women and women of color who left the workforce due to childcare, whereas high-income women (disproportionately white) could afford private solutions or had remote jobs. This means that the \u0026ldquo;she-cession\u0026rdquo; was driven by intersectional identity (gender\u0026thinsp;+\u0026thinsp;class\u0026thinsp;+\u0026thinsp;race) \u0026ndash; not all women faced the same fate. Similarly, among men, those in low-wage jobs (often minority men) were far likelier to lose employment than professional men. Our results showed that for mid-to-high income brackets, there were still significant racial and gender disparities​, implying that even holding income constant, being a woman or person of color meant a tougher experience. This could be due to differences in family care roles (women shouldering more even in high-income families) or subtle workplace biases in layoffs.\u003c/p\u003e\u003cp\u003eFrom a theoretical perspective, these findings reinforce the necessity of an intersectional lens in labor research. They challenge any analysis that would treat \u0026ldquo;women\u0026rdquo; as a monolith or \u0026ldquo;minorities\u0026rdquo; as one group. For policy, it suggests one-size-fits-all remedies may miss the mark \u0026ndash; supports need to be tailored with awareness that, for instance, a re-employment program might need different components to effectively reach Black women versus white men.\u003c/p\u003e\u003cp\u003e3. Policy Responses and Remaining Gaps: The rapid recovery of aggregate employment by 2022 \u0026ndash; including the records set in women\u0026rsquo;s labor force participation​\u0026ndash; demonstrates that \u003cem\u003emacro-economic tools can hasten recovery\u003c/em\u003e and benefit historically disadvantaged workers. Massive fiscal stimulus prevented a prolonged depression that would have undoubtedly deepened inequality. The fact that Black and Hispanic unemployment, while initially worse, fell quickly in 2021 is in part thanks to job growth spurred by the American Rescue Plan and other measures​. This suggests that vigorous policy interventions are essential to counteract the disparate impacts of economic shocks on marginalized groups.\u003c/p\u003e\u003cp\u003eHowever, our analysis and others\u0026rsquo; indicate that structural inequities were not erased. By early 2023, virtually all groups had regained jobs, yet the pay gaps and job quality gaps persist. Many women of color returned to low-paying jobs; many Black workers still face unemployment rates about twice those of whites (the Black-white unemployment ratio remains\u0026thinsp;~\u0026thinsp;2:1 in typical times, and was similar by 2022\u0026rsquo;s end). Intersectionality reminds us that recovery \u0026ldquo;for all\u0026rdquo; can still leave intersectional minorities behind in subtler ways \u0026ndash; e.g., in promotion opportunities or in accumulated debt from the jobless spell.\u003c/p\u003e\u003cp\u003eOne concerning trend is the potential for lasting scarring effects. Research on past recessions (Kalleberg \u0026amp; Von Wachter 2017) shows that workers who experience long unemployment can suffer wage penalties for years​. If Black and Hispanic women disproportionately experienced long spells out of work in 2020 and slow returns, they may face earnings setbacks that widen gender and racial wage gaps further going forward. Indeed, the wage data showing Black and Hispanic women earn far less on average than white men in 2022 ​could partially reflect such scarring or the persistent occupational segregation funneling them back into low-wage roles. This dynamic is a form of what sociologist William Julius Wilson called \u0026ldquo;compound deprivation\u0026rdquo; \u0026ndash; short-term crises compound long-term disadvantage.\u003c/p\u003e\u003cp\u003eAddressing Occupational Segregation: Our findings strongly point to occupational segregation as a lever to address to improve equality. Policies that encourage diversity in hiring across occupations, enforce anti-discrimination laws, and invest in training programs to help women and minorities move into higher-paying, more stable fields (like tech, healthcare, trades) could reduce the vulnerability in future crises. For example, expanding apprenticeships for women and minorities in construction and manufacturing (sectors that boomed in late 2020 and 2021) could buffer against all job losses being concentrated in service work. Similarly, raising wages and protections in secondary-sector jobs (through higher minimum wages, unions, or labor standards) can make those jobs less precarious and reduce turnover, mitigating the \u0026ldquo;first fired\u0026rdquo; issue.\u003c/p\u003e\u003cp\u003eSupporting Intersectional Groups: Intersectionality suggests we need targeted support for those at disadvantaged intersections. During the pandemic, some measures like the child tax credit and childcare subsidies were aimed at helping mothers specifically​\u0026ndash; these are crucial for gender equity. Perhaps more targeted would be investments in communities of color \u0026ndash; e.g., job programs in predominantly Black and Hispanic neighborhoods, or support for minority-owned small businesses (which provide employment in those communities). The pandemic saw many minority businesses struggle to get Paycheck Protection Program loans initially, a structural issue that exacerbated job loss for minority workers. Ensuring equitable access to relief funds is another lesson.\u003c/p\u003e\u003cp\u003eAnother insight is the importance of federal data and analysis that is intersectional. The crisis revealed gaps in data (for instance, early BLS reports did not fully capture multiracial or intersectional stats). Efforts by researchers to fill those gaps (like Moen et al.\u0026rsquo;s intersectional analysis​) were vital. Going forward, regular reporting on, say, Black women\u0026rsquo;s unemployment or Latina mothers\u0026rsquo; labor force participation could keep policymakers aware of these layered disparities.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThe COVID-19 pandemic struck at a time of historically low unemployment, yet its fallout swiftly reproduced and magnified long-standing inequalities in work and employment. This study examined the pandemic labor market through an intersectional and structural lens, focusing on how occupational segregation by race and gender led to stark disparities in job loss and shaped an uneven recovery. Our findings can be summarized as follows:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eDisparate Job Losses: Workers of color and women experienced disproportionate employment declines in the spring 2020 collapse. Occupational segregation was a central driver of these disparities: because women and minorities were concentrated in sectors like hospitality, retail, and personal services, they bore the worst job losses​. At the height of the crisis, Black, Hispanic, and Asian Americans all had substantially higher unemployment rates than whites, and women\u0026rsquo;s unemployment exceeded men\u0026rsquo;s for every racial group​. The intersection of race and gender proved critical, with Black and Latina women suffering the highest jobless rates and labor force exit rates.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eTheoretical Insights: Intersectionality Theory helped explain why, for example, Black women were more adversely affected than would be predicted by simply adding \u0026ldquo;female disadvantage\u0026rdquo; and \u0026ldquo;Black disadvantage\u0026rdquo; \u0026ndash; their specific position in the social structure led to compounded hardship. Dual Labor Market Theory illuminated how the bifurcation of jobs into stable vs precarious translated into a bifurcation of fates during the pandemic: those in \u0026ldquo;secondary\u0026rdquo; jobs were hit first and hardest. Our empirical analysis affirmed that labor market segmentation was key to understanding who lost jobs and who didn\u0026rsquo;t​.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eRecovery and Ongoing Inequality: While employment rebounded impressively by late 2021 and into 2022 \u0026ndash; thanks in part to robust fiscal interventions \u0026ndash; the recovery was uneven. By the end of our study period, most groups had regained employment levels near pre-pandemic baselines, but Black workers (especially Black women) still trailed slightly in employment rates​. Furthermore, returning to work did not mean the elimination of inequality: women and minorities continue to face pay gaps and are often reemployed in the same lower-paying occupations that made them vulnerable​. In effect, the pandemic pulled back the curtain on structural problems, but those structures largely remain intact post-recovery.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003ePolicy Implications: The COVID-19 crisis underscores the need for policies that address occupational segregation and job quality. This could include promoting entry of underrepresented groups into high-growth industries, strengthening affirmative action in hiring and promotions, raising minimum wages and benefits in service jobs, and expanding childcare support to facilitate women\u0026rsquo;s continuous employment. Additionally, unemployment insurance and safety nets need to be accessible and adequate for all workers \u0026ndash; the initial disparities in furlough vs permanent layoff and in UI uptake highlighted gaps that often disfavored workers of color​. Proactive measures (like automatic stabilizers and targeted relief for disadvantaged workers) could help blunt unequal impacts in future downturns.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eFrom a scholarly perspective, our study contributes to the literature by empirically demonstrating how intersectional inequalities manifest in times of crisis and by bridging sociological theory with labor economics data. We integrated an \u0026ldquo;aggressive\u0026rdquo; literature review to situate our findings amidst what is now a growing body of research on COVID-19 and inequality, and our results largely corroborate the consensus emerging from top-tier sociology and IRL (industrial relations) studies​. We also leveraged multiple data sources to provide a comprehensive picture, though we acknowledge limitations such as less detailed data for smaller intersectional groups (e.g., Native American women) and the challenge of proving causality for disparities (we observe correlations consistent with discrimination but cannot definitively measure discrimination with our data).\u003c/p\u003e\u003cp\u003eThe COVID-19 pandemic will be remembered not only as a public health crisis but also as a watershed moment for work and inequality. It amplified the conversation around terms like \u0026ldquo;she-cession\u0026rdquo; and made plain that not all workers weather the same storm in the same boat. Occupational segregation and intersecting inequalities meant that the burdens \u0026ndash; and the benefits of recovery \u0026ndash; were unevenly distributed. As the ILR Review and other journals continue to examine Work, Labor, and Employment Relations in the COVID era, we hope this study provides a useful analytical account of how and why racial and gender disparities in job loss occurred and what that implies for building a more equitable labor market. The pandemic has, in effect, issued a clarion call to address the deep-rooted inequities in our labor system. Answering that call will require sustained attention to the structures of opportunity and disadvantage that our analysis has highlighted. Only by doing so can we ensure that when the next crisis hits, we do not see a repeat of the devastating and unequal toll observed in 2020.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cul\u003e\n \u003cli\u003e\u003cstrong\u003eAcknowledgments:\u003c/strong\u003eThe author acknowledges research support provided by Oglethorpe University.\u003c/li\u003e\n \u003cli\u003eNo funding was received to assist with the preparation of this manuscript.\u003c/li\u003e\n \u003cli\u003eAI was not utilized outside of grammar/spelling checks.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eClinical trial number: not applicable.\u003c/li\u003e\n \u003cli\u003eEthics declaration: not applicable.\u003c/li\u003e\n \u003cli\u003eConsent to Participate declaration: not applicable\u003c/li\u003e\n \u003cli\u003eConsent to publish not applicable\u003c/li\u003e\n \u003cli\u003eThe author has no relevant financial or non-financial interests to disclose.\u003c/li\u003e\n \u003cli\u003eNo conflicts of interest declared.\u003c/li\u003e\n \u003cli\u003eAll underlying research materials for this study are derived from publicly available sources, including government reports, institutional assessments, and peer-reviewed studies. A full list of sources is included in the References section. No proprietary or restricted data were used in this research and there are no ethical issues to disclose.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eI confirm that this is my original work, that I have the rights in the work, that this is for first publication in this Journal and that it is not being considered for/has not already been published elsewhere\u003c/li\u003e\n\u003c/ul\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eJB completed all areas.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eData from Current Population Survey (CPS), American Community Survey (ACS), and the Bureau of Labor Statistics (BLS),\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAlbanesi, S., \u0026amp; Kim, J. (2021). Effects of the COVID-19 recession on the US labor market: Occupation, family, and gender. Journal of Economic Perspectives, 35(3), 3\u0026ndash;24.\u003c/li\u003e\n \u003cli\u003eAlon, T., Doepke, M., Olmstead-Rumsey, J., \u0026amp; Tertilt, M. (2020). The impact of COVID-19 on gender equality. NBER Working Paper No. 26947.\u003c/li\u003e\n \u003cli\u003eAndrea, S. B., Eisenberg-Guyot, J., Oddo, V. M., Peckham, T., \u0026amp; Jacoby, D. (2022). Intersectional inequities in COVID-19-related job losses among older workers. SSM \u0026ndash; Population Health, 17, 101033.\u003c/li\u003e\n \u003cli\u003eBateman, N., \u0026amp; Ross, M. (2020). Why has COVID-19 been especially harmful for working women? Brookings Institution.\u003c/li\u003e\n \u003cli\u003eBennett, J. (2018). The impact of race on employment during the 2005\u0026ndash;2011 recession. Global Journal of Management and Business Research, 18(A2), 43\u0026ndash;52.\u003c/li\u003e\n \u003cli\u003eBoesch, T., Nunn, R., \u0026amp; Tchourumoff, A. (2022). Occupational segregation and the racial impact of COVID-19. Federal Reserve Bank of Minneapolis.\u003c/li\u003e\n \u003cli\u003eBureau of Labor Statistics. (2020a). Employment situation summary\u0026mdash;April 2020. U.S. Department of Labor.\u003c/li\u003e\n \u003cli\u003eBureau of Labor Statistics. (2020b). Employment recovery in the wake of the COVID-19 pandemic. Monthly Labor Review.\u003c/li\u003e\n \u003cli\u003eCajner, T., Crane, L. D., Decker, R. A., Hamins-Puertolas, A., \u0026amp; Kurz, C. J. (2020). The U.S. labor market during the beginning of the pandemic recession. Brookings Papers on Economic Activity.\u003c/li\u003e\n \u003cli\u003eCollins, C., Landivar, L. C., Ruppanner, L., \u0026amp; Scarborough, W. J. (2021). COVID-19 and the gender gap in work hours. Gender, Work \u0026amp; Organization, 28(S1), 101\u0026ndash;112.\u003c/li\u003e\n \u003cli\u003eCollins, P. H. (2000). Black feminist thought: Knowledge, consciousness, and the politics of empowerment (2nd ed.). Routledge.\u003c/li\u003e\n \u003cli\u003eCollins, P. H., \u0026amp; Bilge, S. (2016). Intersectionality. Polity Press.\u003c/li\u003e\n \u003cli\u003eCortes, G. M., \u0026amp; Forsythe, E. (2023). Racial and ethnic disparities in job displacement during the COVID-19 pandemic. ILR Review, 76(1), 30\u0026ndash;55.\u003c/li\u003e\n \u003cli\u003eCrenshaw, K. (1989). Demarginalizing the intersection of race and sex: A Black feminist critique of antidiscrimination doctrine, feminist theory and antiracist politics. University of Chicago Legal Forum, 1989(1), 139\u0026ndash;167.\u003c/li\u003e\n \u003cli\u003eDickens, W. T., \u0026amp; Lang, K. (1985). A test of dual labor market theory. American Economic Review, 75(4), 792\u0026ndash;805.\u003c/li\u003e\n \u003cli\u003eDias, F. A., Chance, J., \u0026amp; Buchanan, A. (2020). The motherhood penalty and the fatherhood premium in employment during COVID-19: Evidence from the United States. Research in Social Stratification and Mobility, 69, 100542.\u003c/li\u003e\n \u003cli\u003eDoeringer, P. B., \u0026amp; Piore, M. J. (1971). Internal labor markets and manpower analysis. D.C. Heath and Company.\u003c/li\u003e\n \u003cli\u003eFairlie, R. W., Couch, K., \u0026amp; Xu, H. (2020). The impacts of COVID-19 on minority unemployment: First evidence from April 2020 CPS microdata. Journal of Population Economics, 33(4), 1265\u0026ndash;1283.\u003c/li\u003e\n \u003cli\u003eFederal Reserve Bank of Minneapolis. (2020). Mothers face higher labor market challenges during pandemic. Minneapolis Fed COVID-19 Economic Research.\u003c/li\u003e\n \u003cli\u003eFlood, S., \u0026amp; Moen, P. (2021). Unequal work, unequal retirement: Gender, race, and COVID-19 labor market effects. ILR Review, 74(4), 843\u0026ndash;865.\u003c/li\u003e\n \u003cli\u003eGelatt, J. (2020). Immigrant workers: Vital to the U.S. COVID-19 response, disproportionately vulnerable. Migration Policy Institute.\u003c/li\u003e\n \u003cli\u003eGroshen, E. (2020). COVID-19\u0026rsquo;s impact on the U.S. labor market as of September 2020. ILR Review, 73(5), 1074\u0026ndash;1089.\u003c/li\u003e\n \u003cli\u003eHeggeness, M. L. (2020). Estimating the immediate impact of the COVID-19 shock on parental attachment to the labor market and the double bind of mothers. Review of Economics of the Household, 18(4), 1053\u0026ndash;1078.\u003c/li\u003e\n \u003cli\u003eHolder, M., Jones, J., \u0026amp; Masterson, T. (2021). The early impact of COVID-19 on job losses among Black women in the United States. Journal of Economics, Race, and Policy, 4(4), 203\u0026ndash;217.\u003c/li\u003e\n \u003cli\u003eKochhar, R. (2020). Hispanic women, immigrants, young adults hit hardest by COVID-19 job losses. Pew Research Center.\u003c/li\u003e\n \u003cli\u003eLandivar, L. C., Ruppanner, L., Scarborough, W. J., \u0026amp; Collins, C. (2020). Early signs indicate that COVID-19 is exacerbating gender inequality in the labor force. Socius, 6, 1\u0026ndash;3.\u003c/li\u003e\n \u003cli\u003eMasterson, T., Holder, M., \u0026amp; Jones, J. (2020). The impact of COVID-19 on Black women workers in the U.S. Levy Economics Institute Working Paper No. 963.\u003c/li\u003e\n \u003cli\u003eMoen, P., Pedtke, J. H., \u0026amp; Flood, S. (2020). Disparate disruptions: Intersectional COVID-19 employment effects by age, gender, education, and race/ethnicity. Work, Aging and Retirement, 6(4), 207\u0026ndash;228.\u003c/li\u003e\n \u003cli\u003eMontenovo, L., Jiang, X., Lozano-Rojas, F., Schmutte, I. M., Simon, K. I., Weinberg, B. A., \u0026amp; Wing, C. (2020). Determinants of disparities in COVID-19 job losses. NBER Working Paper No. 27132.\u003c/li\u003e\n \u003cli\u003ePetts, R. J., Carlson, D. L., \u0026amp; Pepin, J. R. (2021). A gendered pandemic: Childcare, homeschooling, and parents\u0026rsquo; employment during COVID‐19. Gender, Work \u0026amp; Organization, 28(S2), 515\u0026ndash;534.\u003c/li\u003e\n \u003cli\u003eQian, Y., \u0026amp; Hu, Y. (2021). Couples\u0026apos; changing work patterns in the United Kingdom and Canada during the COVID‐19 pandemic. Gender, Work \u0026amp; Organization, 28(S2), 535\u0026ndash;553.\u003c/li\u003e\n \u003cli\u003eRho, H. J., Brown, H., \u0026amp; Fremstad, S. (2020). A basic demographic profile of workers in frontline industries. Center for Economic and Policy Research.\u003c/li\u003e\n \u003cli\u003eReich, M., Gordon, D. M., \u0026amp; Edwards, R. C. (1973). Dual labor markets: A theory of labor market segmentation. American Economic Review, 63(2), 359\u0026ndash;365.\u003c/li\u003e\n \u003cli\u003eReid, L. W., \u0026amp; Rubin, B. A. (2003). Integrating economic dualism and labor market segmentation: The effects of race, gender, and structural location on earnings. Sociological Quarterly, 44(3), 405\u0026ndash;432.\u003c/li\u003e\n \u003cli\u003eRuppanner, L., Tan, X., Scarborough, W., Landivar, L. C., \u0026amp; Collins, C. (2021). Shifting inequalities? Parents\u0026rsquo; sleep, anxiety, and calm during the COVID-19 pandemic in Australia and the United States. Men and Masculinities, 24(1), 181\u0026ndash;188.\u003c/li\u003e\n \u003cli\u003eStevenson, B. (2020). The initial impact of COVID-19 on labor market outcomes across groups and the potential for permanent scarring. The Hamilton Project, Brookings Institution.\u003c/li\u003e\n \u003cli\u003eTessler, H., Choi, M., \u0026amp; Kao, G. (2020). The anxiety of being Asian American: Hate crimes and negative biases during the COVID-19 pandemic. American Journal of Sociology, 126(2), 436\u0026ndash;468.\u003c/li\u003e\n \u003cli\u003eU.S. Census Bureau. (2021). Impact of the coronavirus pandemic on businesses and employees by industry. U.S. Department of Commerce.\u003c/li\u003e\n \u003cli\u003eU.S. Census Bureau. (2022). Changes in employment by occupation during the COVID-19 pandemic. U.S. Department of Commerce.\u003c/li\u003e\n \u003cli\u003eZamarro, G., \u0026amp; Prados, M. J. (2021). Gender differences in couples\u0026rsquo; division of childcare, work, and mental health during COVID-19. Review of Economics of the Household, 19(1), 11\u0026ndash;40\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"COVID-19, occupational segregation, intersectionality, dual labor market, unemployment, gender disparities, racial disparities","lastPublishedDoi":"10.21203/rs.3.rs-6915836/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6915836/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe COVID-19 pandemic triggered an unprecedented economic upheaval that magnified pre-existing inequalities in the U.S. labor market. This study examines how occupational segregation by race and gender contributed to disparate job losses and shaped the trajectory of employment recovery. Drawing on Intersectionality Theory and Dual Labor Market Theory as a theoretical framework, we conduct an interdisciplinary analysis blending sociology and labor economics perspectives. We integrate an extensive literature review with original analyses of secondary data (Current Population Survey, Bureau of Labor Statistics reports, and American Community Survey) to assess racial and gender disparities in unemployment, labor force exits, and rehiring during the pandemic. Consistent with prior research, we find that women and workers of color experienced disproportionately severe job losses in the spring of 2020, reflecting their overrepresentation in vulnerable, low-wage service occupations​. Intersectional vulnerabilities were especially evident for women of color, who faced compounded disadvantages in both job displacement and caregiving burdens. As the labor market rebounded, these groups saw employment recover more slowly, though strong fiscal stimulus eventually facilitated a return to pre-pandemic employment levels for many. Our findings underscore how structural inequalities\u0026mdash;rooted in segregated labor markets and intersecting axes of oppression\u0026mdash;produced unequal outcomes in the COVID-19 era, and we discuss policy implications for a more equitable recovery.\u003c/p\u003e","manuscriptTitle":"Occupational Segregation and COVID-19: A Sociological Analysis of Racial and Gender Disparities in Job Loss and Recovery","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-13 06:55:11","doi":"10.21203/rs.3.rs-6915836/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"d5765a00-1a88-4630-9086-69936dd9ea3b","owner":[],"postedDate":"August 13th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-01-19T11:30:45+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-13 06:55:11","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6915836","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6915836","identity":"rs-6915836","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}