From recession to pandemic: Displacement among workers with disabilities from 2007 through 2021

1.1 Disability and labor market inequality

People with disabilities have lower labor market participation compared to similarly situated people without disabilities and tend to experience greater disadvantages at work (Brown & Moloney, 2019). Throughout the 1990s and 2000s, before and after the Great Recession, labor market participation among disabled Americans declined yearly, despite protections put in place to improve job outcomes (Bruyère, 2016; Maroto & Pettinicchio 2022a; U.S. BLS, 2020). Consequently, about 80% of Americans with disabilities were not in the labor force in 2021. Many are considered discouraged workers –those who stopped looking for work and are not included in unemployment rates (U.S. BLS, 2022a).

Struggles are also evident among those who remain in the labor market. Workers with disabilities tend to earn less than those without disabilities, which is linked to household poverty and income inequality (Jajtner et al., 2020; Maroto, Pettinicchio, & Patterson, 2019). In addition, people with disabilities experience greater earnings volatility, they are more likely to be in precarious and contract-based jobs, and they often are subject to occupational segregation (Jolly & Wagner, 2023; Maroto & Pettinicchio, 2014; Schur & Kruse, 2022). They are also more likely to be paid low-to-subminimum wages (Maroto & Pettinicchio, 2022b).

Given these labor market conditions, people with disabilities earn on average 26% less than similar workers, but this can range from 10 to 37% less based on the type of disability (Maroto and Pettinicchio 2014, 2015). Individuals reporting multiple disabilities, cognitive disabilities, and disabilities related to independent living are also more likely to report lower earnings (Brucker, Houtenville, & Lauer 2016; Jones 2011). Lower earnings, limited benefits, and poorer working conditions place people with disabilities in more precarious labor market situations that further contribute to higher rates of poverty among this group and leave them at risk for job displacement.

1.2 Disparities in job displacement

Drawing from broader frameworks of labor market inequality, scholars point to several interrelated factors that explain job displacement, and, especially, how job displacement disproportionately affects some groups more than others. These range from changing economic contexts and shifts in occupations and industry sectors to changing job demands and demographics to discrimination. For example, Spalter-Roth and Deitch (1999) analyzed the unequal effects of job displacement among different categories of people resulting from major labor market restructuring in the 1990s, focusing on workers in manufacturing and in service sector industries. Even though most displaced workers were able to find subsequent work, many ended up in lower paying jobs, and this was especially true among women and people of color. Their research also pointed to the intersectional aspects of negative post-displacement effects and occupational segmentation based on gender and race. Additionally, immigrant workers are more likely to experience job displacement than native-born workers, with more severe repercussions for labor market and earnings outcomes down the road (Bratsberg, Raaum, & Røed, 2018).

These patterns were found repeatedly in subsequent work on displacement. Wilson’s (2009) analysis of the erosion of wages for less-skilled workers points to how African American workers clustered in low-skilled jobs have been disproportionately affected by displacement. Examining lower-income workers during mass layoffs, Andersson and colleagues (2018) found that the duration of joblessness was especially lengthy for displaced Black, female, and older workers. But, it may not only be lower income workers who feel the effects of displacement. According to Roscigno et al. (2012) and Wilson and Roscigno (2018), some of the highest rates of displacement were among professional and managerial jobs, which also disproportionately affected African Americans, especially older workers.

As a commonly experienced life-course event that triggers potential long-term labor market costs (DiPrete, 2002; DiPrete & McManus, 2000; Gangl, 2004, 2006), job displacement results in a variety of negative consequences for workers. This is especially the case when there are lengthy periods between being displaced and becoming reemployed. Displacement can increase the chances of subsequent job losses (Stevens, 1991). Unemployment spells may have long-lasting, scarring effects on workers’ future wages –effects that are often felt many years after initial unemployment (Shuey & Wilson 2017). Displacement is also often characterized by downward employer mobility (Mouw and Kalleberg 2010). This occurs when a person returns to work with a new employer and lower quality job with lower wages, which further contributes to wage inequality (see also Ruhm, 1991 and Brand, 2006). Finally, displacement also has important health impacts (Black, Devereaux, & Salvanes, 2015), which the pandemic made even more acute, highlighting the importance of considering context when assessing disparities in displacement rates.

1.3 Job displacement and context

Job loss is linked to larger economic and policy contexts, which results in shifting rates and changing disparities in displacement over time. Linked to the mortgage crisis and subprime lending, the Great Recession spurred massive layoffs from the end of 2007 through the beginning of 2009 (Farber, 2017; Kalleberg & von Wachter, 2017). Mitra and Kruse’s (2016) key study of job displacement found that between 2007 and 2013 men and women with disabilities were almost twice as likely to experience job displacement as men and women without disabilities for almost any reason, but especially for their position or shift being abolished. And so, prior to the pandemic, and despite already having lower than average employment levels, disabled people saw greater negative effects on displacement from the Great Recession (Fogg, Harrington, & McMahon, 2010; Livermore & Honeycutt, 2015).

The COVID-19 pandemic further highlighted problematic relationships between the labor market and disability (Maroto, Pettinicchio, & Lukk, 2021; Umucu, 2021). The employment rate for people with disabilities in the United States declined from 19.3% in 2019 to 17.9% in 2020 and increased slightly to 19.1% in 2021. While unemployment rates declined among both disabled and non-disabled workers between 2020 and 2021, unemployment rates were still about twice as high for people with disabilities.

Job losses, however, followed a different pattern during the COVID-19 pandemic than during the 2008 recession. After spiking in the summer of 2020, unemployment rates quickly declined over the following year (Cortes & Forsythe, 2023). Still, disparities in unemployment and displacement by education, race, and gender remained apparent (Montenovo et al., 2022; U.S. BLS, 2022b). Women experienced greater job loss and displacement during the pandemic than men (Mooi-Reci & Risman, 2021; Montenovo et al., 2022), as did people who identify as Indigenous in Canada (St-Denis, 2020). Non-white American workers were more likely to experience job loss because of their over-representation in low-income jobs, which saw a great deal of displacement (Cortes & Forsythe, 2023). They were also less likely to recover from displacement between 2020 and 2021.

In addition to financial insecurity and distress (Maroto et al., 2023; Pettinicchio et al., 2021), those who experienced involuntary job displacement during the pandemic were more likely to experience declining mental health (Grace 2022). They also are more likely to experience additional health issues associated with corresponding losses in healthcare benefits and insurance, as well as increased food insecurity (Bundorf, Gupta, and Kim 2021; Milovanska-Farrington 2023). Job displacement during the pandemic also revealed its gendered nature as women potentially delayed job searches due to expectations about caregiving responsibilities (Collins, et al., 2021). This means that involuntary job loss during the pandemic extends beyond its impacts on re-employment and financial security to include a host of related effects tied to health and family. Consequently, longer term scarring effects can be quite robust disproportionately affecting marginalized communities.

In light of previous work on displacement across groups, including people with disabilities, and the varying employment consequences of the pandemic, we expect that people with disabilities will have higher rates of displacement for survey waves covering the pandemic and these will further vary by disability type. The pandemic acted as an exogenous force on work, fundamentally changing its very nature in both the short-and-long-term. Concerned about the intersection of their health and disability status, workers faced a great deal of uncertainty about whether they would still have jobs throughout and post-pandemic, drawing parallels with involuntary job loss experiences during the Great Recession (Reeves, et al., 2014). Some, especially those in better paying jobs within management, professional, and administrative fields, were able to continue work remotely, benefitting from more flexible work arrangements (Baker, 2020). This had some positive effects on well-being but not necessarily for women who also had to manage caregiving responsibilities (Fan and Moen 2023). For other groups, including people with disabilities, working from home was not feasible in large part because of the jobs they occupy (Dey et al., 2020; Maroto, Pettinicchio, and Lukk 2021). Job losses exacerbated the already precarious labor market positions for members of these groups.

2.1 Data

We study displacement among people with different disabilities using data from the 2010, 2012, 2014, 2018, 2020, and 2022 waves of Current Population Study Displaced Worker Supplement (Flood et al., 2022). The CPS is a monthly labor force survey conducted by the U.S. Census Bureau. The DWS comprises a sample of workers age 20 years or older who involuntary lost or left their jobs due to a plant or company closing, insufficient work, or the completion of a seasonal job in the three calendar years prior to the survey wave. This means that data from the 2010 wave incorporate displaced workers from 2007–2009 and data from the 2022 wave incorporate displaced workers from 2019–2021 with pooled data providing the ability to analyze experiences of displacement from 2007 through 2021.

We begin our analysis with the 2010 DWS wave after which CPS revised their set of disability-related questions to provide better information regarding different types of potential limitations.² To study disparities in displacement rates, we restrict our sample to workers who completed the CPS DWS and who were at risk of job loss during each sample period. Following Farber (2004, 2011) and Mitra and Kruse (2016), this incorporates employed workers and those who reported job loss during the sample period. It does not include workers who were not in the labor force for reasons such as those related to retirement, education, or caregiving. We also restrict analyses to non-military workers with occupational data. Our final sample size is 344,729 individuals for the sample of workers who were displaced or at risk for job displacement at the time of the survey. Finally, we weight all descriptive statistics and model estimates using survey-provided weights that adjust for non-interview status within households and for population characteristics based on age, race, sex, and state of residence (CPS, 2022).

2.2 Measures and analytic models

Our primary outcome variable of job displacement measures whether the respondent experienced job displacement within the three years prior to the survey wave. It is based on responses to the survey question: “During the last 3 calendar years, that is, January [YEAR] through December [YEAR], did you lose a job, or leave one because: your plant or company closed or moved, your position or shift was abolished, insufficient work or another similar reason?” Respondents who answered “Yes” to this question were considered to have been displaced from their jobs and then asked a series of follow-up questions regarding job displacement.

Our primary predictor variable, disability status includes limitations based on six questions used by the CPS to identify the population with disabilities (Livermore et al. 2011). In addition to disability status, we also include five variables to measure the type of disability. These variables indicate whether the respondent reported a cognitive, physical (ambulatory), hearing, vision, or independent living or self-care (IDL) disability.³ Cognitive difficulties include those related to learning, remembering, concentrating, or making decisions. Physical or ambulatory difficulties include anything that limits a respondent in one or more basic physical activities. Vision difficulties indicate whether the respondent was blind or had serious difficulty seeing even with corrective lenses. Hearing difficulties indicate whether the respondent was deaf or had serious difficulty hearing. Independent living difficulties indicate the presence of any condition that makes it difficult or impossible to perform basic activities outside the home alone, and self-care difficulties include personal needs, such as bathing and dressing. As many individuals experience multiple disabilities, these categories are not mutually exclusive.

We also include a series of demographic, education, and work-related control variables across models. Gender is a categorical variable of male (the referent) or female. We indicate race/ethnicity with a categorical variable measured as non-Hispanic white (the referent), non-Hispanic Black, Hispanic, or non-Hispanic other, and we control for citizenship status.⁴ We measure age in years and include a quadratic age-squared term to account for any non-linear relationships. Marital status is a categorical variable that indicates whether the respondent was currently married (the referent), never married, or separated, divorced, or widowed, and any children indicates whether any children under age 18 were present in the respondent’s household. We measure educational attainment with a categorical variable that indicates whether the respondent obtained a high school diploma or equivalent degree or less than that (the referent); attended some college without obtaining a degree; obtained an Associate’s (two-year) degree; completed college with a Bachelor’s (four-year) degree; or obtained a Master’s, professional, or Doctorate degree.

In addition to these covariates, we also control for the respondent’s major industry and occupation of employment, as involuntary job losses are often relegated to certain parts of the labor market. Major occupation includes the following seven categories: management, business, and financial occupations (referent); professional and related; service; sales and related; office and administrative support; natural resource, construction, and maintenance; and production, transportation, and material moving. Major industry includes the following 10 categories: agriculture and natural resources; construction; manufacturing; wholesale and retail trade; transportation and utilities; information and financial activities; professional and business services; educational and health services (referent); leisure and hospitality; other services; and public administration. For displaced workers, we use the occupations and industries of the jobs they were displaced from, and for continually employed workers, we use their current reported industry and occupation. Finally, to better account for changing economic and political environments, we also include indicator variables for the respondent’s state of residence and for the survey year.

Table 1 presents descriptive statistics for displaced workers and continually employed workers in the pooled dataset. Between 2007–2021, approximately 6.6% of workers were displaced. Displaced workers were more likely to have all types of disabilities; 7.0% of displaced workers reported having at least one disability compared to 3.6% of continually employed workers. Displaced workers were also younger and more likely to be male, belong to a racial minority group, and not married. They also tended to have lower levels of education.

We use a series of logistic regression models to examine the relationship between disability, disability type, and displacement. To show how this relationship varies over time, we incorporate interaction terms between disability and survey wave year and estimate models separately by survey wave year. Although we include control variables within our models, our results are primarily descriptive with the goal of documenting relationships over time. We present and discuss most of our results as either average marginal effects or predicted probabilities with results averaged across the population for ease of interpretation.