Skip to content
Hello
Policy Report Jobs & Labor

More than Solidarity: How Labor Unions Preserved Latino Jobs

Using Current Population Survey data, this report analyzes the impact of unionization on Latino employment rates and wages between January 2020 and June 2021.
More than Solidarity: How Labor Unions Preserved Latino Jobs

Executive Summary

Unstable employment and stagnant wages characterized the labor market participation of Latino workers throughout the COVID-19 pandemic. These trends reflect long-standing workforce inequities. Historically, Latinos are more likely than non-Latino whites to have jobs with low wages, minimal benefits, and unstable employment. However, Latinos covered by labor union contracts exhibited greater employment stability and earned higher wages than their non-union counterparts over the course of the pandemic.

This report compares national employment and wage trends for unionized Latino workers to those of non-unionized Latino workers during the COVID-19 downturn and recovery. Our analysis indicates that unionized Latino workers experienced more stable employment and maintained higher wages than their non-union counterparts. Union workers of other racial and ethnic groups also experienced greater employment stability early in the pandemic, but the effects of unionization were strongest among Latinos.

Our analysis suggests that unionization—even within the same industry and occupation—preserved employment and wages for workers during the COVID-19 pandemic, accounting for variations in unionization across occupations, industries, and worker characteristics. These findings support recent calls to expand access to unionization and highlight the need to pass federal legislation that makes it easier for workers to access the stable employment, higher wages, and more substantial benefits unions provide.

Our main findings are as follows:

  1. Between April and June 2020, the height of the pandemic’s economic recession, non-unionized Latinos lost seven times as many jobs as unionized Latinos (without controlling for occupation or worker characteristics).
    1. Unionized Latino employment fell 2.5 percent (-66,700 workers).
    2. Non-union Latino employment shrank by 18.5 percent (nearly 4.3 million workers).
  1. Unionized Latino employment grew throughout the entirety of 2020 and only began to decrease in 2021.
  1. In the public sector, unionized Black and Asian American or Pacific Islander (AAPI) workers suffered the heaviest job losses during COVID-19 compared to other racial and ethnic groups. A quarter of jobs held by unionized Black workers in the public sector were lost between March 2020 and June 2021.
  1. After controlling for worker and job characteristics, union coverage was associated with a 1.1 percent lower probability of unemployment during the pandemic.
    1. For Latinos, unionization decreased the probability of unemployment by 3.1 percent.
    2. Black and Latina women faced the highest probabilities of unemployment compared to other groups.
  1. Across all racial and ethnic groups, unionized workers earned more than non-unionized workers during the COVID-19 economic downturn (without controlling for occupation or worker characteristics).
    1. This wage effect was strongest for Latino and Black workers.
    2. Latino and Black workers continued to earn lower wages than Whites, regardless of union status.

Policy Recommendations: 

  1. Pass and enforce federal legislation that supports the right to form a union and engage in collective bargaining.
    1. Reinforce the right to join a union that all workers are entitled to under the National Labor Relations Act (NLRA) of 1935.
    2. Pass the Protecting the Right to Organize (PRO) Act of 2021 which would make unionization more accessible for workers by dismantling right-to-work laws, legalizing solidarity strikes, banning employers from holding anti-union meetings during union elections, and prohibiting employers from taking disciplinary action against employees who seek to unionize.
    3. Impose stronger penalties on employers and corporations that use illegal practices to discourage and crack down on union organizing.
  1. In the short term, support workers not currently represented by unions by passing legislation at the state, local, and federal level that raises their standards of living.
    1. Pass a $15 minimum wage at the federal level.
    2. Pass state and local minimum wage laws that keep up with costs of living.
    3. Expand the capacity and diligence of social safety net programs, such as unemployment benefits and childcare services.
  1. Implement legislation and programs to mitigate wage inequities and employment instability in sectors who experienced heavy job losses during the pandemic and low unionization rates, and which tend to employ Latinos, African Americans, Asian Americans and Pacific Islanders, and women (e.g., service sector work and public employment).
    1. Support public sector employment by expediting the re-employment of workers adversely affected by the pandemic.
    2. Implement a federal jobs guarantee to expand access to public sector jobs that have historically provided financial stability and upward mobility for women and Black workers.
    3. Expand programs that provide childcare or childcare subsidies for women who are employed or desire to be employed.
  1. Implement long-term, systematic solutions that grant workers (especially workers of color) equitable pathways to union jobs that provide stability and job quality.
    1. Invest in labor-management training partnerships—such as California’s High Road Training Partnership1—that generate economic mobility for underserved and low-income workers through vocational training, education, and quality career ladder jobs while improving worker and employer competitiveness in a rapidly-changing, carbon-constrained economy.
    2. Expand access to career apprenticeships and career pathways into union jobs, especially for young workers of color.
    3. Incorporate worker voices and representation on corporate boards and into company decision-making.
    4. Eliminate onerous, costly, and inefficient civil service exams that pose unnecessary and discriminatory barriers to entry to unionized public sector jobs within state and local governments.
LPPI

Introduction: Labor Unions, Latinos, and COVID-19

Traditionally, labor unions—organizations of workers who come together for a common goal—operate as vehicles of collective bargaining. They negotiate contracts, wages, and benefits for workers with employers in nearly every occupation. In the United States, coverage under labor union contracts has been associated with higher wages, more stable employment, and more expansive benefits, including health care coverage and pensions.2 Union membership, however, peaked in 1945 and 1956 when roughly a third of workers were labor union members.3 Since then, unionization in the U.S. workforce has steadily declined. As of 2020, the union membership rate of the U.S. workforce was down to 10.8 percent overall and 6.3 percent in the private sector.4

Despite historically low rates of union membership, the benefits of coverage under a union contract are still apparent across the workforce in the present day. In 2019, the Bureau of Labor Statistics (BLS) reported that unionized workers made nearly 20 percent more in median weekly earnings than non-unionized workers.5 A 2020 study from the Economic Policy Institute (EPI) found that workers covered by a union contract earned 11.2 percent more than a peer with a similar education, occupation, and experience level.6 The BLS also reported that 95 percent of union workers had access to health care compared to 68 percent of non-union workers, a difference of 27 percentage points.7

Labor unions have also helped Latinos earn higher wages, access benefits, and build wealth. In 2019, the BLS reported that weekly average earnings of unionized Latino workers were 28 percent higher than those of non-unionized Latinos.8 After controlling for worker characteristics, coverage under a union contract still raises Latino workers’ wages by 17.6 percent, or $2.60 per hour. Unionized Latinos are also 26 percent more likely to have employer-provided health insurance and 27 percent more likely to have a pension plan,9 and they possess ten times the wealth of non-union Latinos.10 However, just 9.8 percent of Latino workers are unionized, a lower rate than that of most other racial groups.11

However, despite the pandemic’s disproportionate impact on Latinos—over half of Latino workers lost a job or took a pay cut as of August 202012—it is unclear how unionized Latinos fared compared to non-unionized Latinos. If union membership provided Latino workers greater employment stability during the pandemic, strengthening unions and collective worker power could prevent employment losses for unionized workers in the next crisis. It is also important to understand whether any union effects vary by gender, considering that women were more likely to experience unemployment or leave the labor market during the pandemic.13

This report analyzes the role of labor union coverage in preserving employment and wages during the COVID-19 economic downturn and recovery, with a focus on Latino workers. We found that unionized workers generally experienced smaller job losses in comparison to non-unionized workers; however, outcomes varied based on race, gender, occupation, and industry. We incorporate comparisons of trends for Latino workers to other demographic groups where appropriate.

Dataset and Methodology

To understand the impact of union membership on preserving Latino employment during the pandemic, we used the Current Population Survey (CPS), a monthly survey of 60,000 households and the primary source of labor force and unionization statistics in the U.S. Of these 60,000 households, roughly one-fourth are asked additional labor questions related to current pay, usual hours workers, and unionization, among other topics. This subsample, known as the Outgoing Rotation Group (ORG), is the primary source for our data analysis.

Using ORG data from January 2020 through June 2021, we investigated trends in union coverage,14 employment, and wages by race, gender, and occupation.15 In order to overcome sample size limitations, we aggregated the basic monthly CPS samples into three-month periods, or quarters.16 Because all survey respondents with union membership are employed, we looked at changes in total employment by union coverage to understand how union coverage impacted job security throughout the pandemic.

Download the data behind the report.
Latinos in Unions Dataset

Additionally, we analyzed workers’ responses to a series of new questions added to the CPS in May 2020. These new questions were designed to gather data on the effects of the COVID-19 pandemic on job stability, income stability, and ability to work remotely.17 Specifically, the new questions asked whether respondents worked at home at any point in the last four weeks due to the pandemic or were unable to work at any point in the last four weeks because an employer closed or lost business due to the pandemic.

These additional questions also allowed us to explore the specific effects of unionization on unemployment. Following Gezici and Ozay,18 we examined differences in the likelihood of unemployment—defined as an inability to work in the last four weeks due to the pandemic—across different race and gender groups using data from May 2020 through June 2021. Using a probit regression model, we control for worker characteristics, such as education and age, and job characteristics.19 We also report results for the Latino subsample of our dataset.

Lastly, to estimate hourly wages, we followed the Economic Policy Institute’s estimation criteria.20 Our wage analysis is limited to persons in the labor force aged 16 and older employed in the public or private sector with valid time and wage data. We removed outliers from the data but did not enforce smoothing or replace top-coded values. Note that our data and analysis focus on national trends; however, the nationwide labor market may not fully reflect economic realities in every geography or at the local level.

Racial and Intersectional Analysis of Union and Non-Union Employment

Throughout the COVID-19 economic downturn, unionized Latino21 workers were more likely to stay employed compared to their non-union counterparts, especially during the height of the pandemic-induced recession. However, employment status of unionized and non-unionized workers varied across race and ethnicity, occupation, sector, and gender. The following section discusses these differences and describes how unionization impacted employment for various groups throughout the course of the COVID-19 recession and recovery.

Unionized Latino Employment Remained Stable Throughout the Pandemic

In the first quarter of 2020, well over 2.5 million Latino workers were unionized, while 23.1 million Latino workers had no union affiliation. At the height of the pandemic’s economic recession in the second quarter of 2020, unionized Latino employment fell 2.5 percent (Figure 1), losing roughly 66,700 workers. Meanwhile, non-unionized Latino employment shrank by 18.5 percent, shedding nearly 4.3 million workers, a rate of job loss seven times higher than for unionized Latino workers.

Unionized workers endured smaller relative employment losses than non-union workers during the pandemic’s peak (Figure 1). Overall, unionized employment fell 10.2 percent from the first quarter of 2020 to the second quarter while non-union employment fell 12.9 percent. This trend holds true across racial and ethnic groups, except for Asian American and Pacific Islanders; unionized AAPI workers suffered greater losses than non-unionized workers at the beginning of the pandemic.

Figure 1. At the Pandemic’s Peak, Unionized Workers Generally Suffered Fewer Job Losses
Quarter-over-Quarter Employment Change by Union Status and Race (2020Q1 to 2020Q2)

Source: LPPI analysis of Current Population Survey Outgoing Rotation Group (2020-2021), available online.

During the economic recovery, unionized Latinos were employed at or above pre-pandemic levels. In fact, unionized Latino employment grew throughout 2020 and only began to contract in 2021 (Figure 2). The second quarter of 2021 marked a low point in the pandemic for unionized Latinos; employment in this quarter was 3.3 percent below pre-pandemic levels (but remained above non-unionized Latino employment).

For non-unionized Latino workers, the economic recovery has been slow but steady. From the second quarter of 2020 to the second quarter of 2021, the non-unionized Latino workforce grew 17.5 percent, adding over 2.7 million workers. However, as of the second quarter of 2021, the non-unionized Latino workforce was still 4.2 percent smaller than it was before the pandemic.

At the height of the recession, unionized workers experienced fewer job losses than non-unionized workers. However, recent trends are more mixed. As of the second quarter of 2020, unionized Black employment remains 16.6 percent below pre-pandemic levels—more than any other ethnic group—while employment for unionized White and AAPI workers are 8.9 percent and 5.3 percent below pre-pandemic levels, respectively. For all three non-Latino ethnic groups, non-union employment is now closer to recovery than union employment, possibly due to public-sector job losses.

Figure 2. Unionized Latino Workers Largely Kept Their Jobs Throughout the Pandemic
Percent Change in Quarterly Employment by Union Status and Race (vs. 2020Q1)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source: LPPI analysis of Current Population Survey Outgoing Rotation Group (2020-2021), available online.

Variations in Occupation Among Union and Non-Union Latino Workers

Unionized Latinos may have maintained employment at higher rates than non-unionized Latinos through the course of the pandemic in part because of the types of jobs they held. Before the onset of the COVID-19 pandemic, unionized Latinos were 9.1 percentage points more likely than non-union Latinos to work in management, business, science, and arts roles (Figure 3). In contrast, non-union Latino workers were more likely to be employed in service and sales and office occupations.

Figure 3. Pre-Pandemic, Union-Covered Latinos Were More Likely to Work in Professional Roles
Percent of Workers by Unionization, Occupation, and Race (2020Q1)

Source: LPPI analysis of Current Population Survey Outgoing Rotation Group (2020-2021), available online.

Notably, service workers experienced the brunt of pandemic-induced unemployment (Figure 4)—one in four service workers was out of work in April 2020—while unemployment for management and professional roles was lower than for all other occupations. BLS research also suggests that only 7.9 percent of service workers had the ability to work from home compared to 86.6 percent of management workers and 64.4 percent of professional workers.22

In comparison to non-Latino union workers, unionized Latinos were 7.2 percentage points more likely to work in service roles and 15 percentage points less likely to work in management, business, science, or art roles. However, despite the heightened vulnerability of service workers, unionized Latino employment remained close to pre-pandemic levels, which was not the case for union-covered White and Black workers. This may suggest a differential role of unions in preserving employment for Latino workers.

Figure 4. Service Workers Experienced the Highest Rates of Pandemic-Related Job Loss
Unemployment Rate by Occupation (Not Seasonally Adjusted), All Workers
Source: U.S. Bureau of Labor Statistics Labor Force Statistics (2019-2021), available online.

Unionized Public Sector Workers Experienced Disproportionate Job Losses

Differences in the industry of employment may also have played a role in determining whether unionized workers stayed employed during the recession. The public sector, for instance, employs a large share of union members. In 2017, roughly 51.3 percent of unionized workers were employed in the private sector, compared to 48.7 percent in the public sector.23 Public sector workers were also much more likely to be unionized than those in the private sector.24

Unionized government jobs include essential occupations like public school teachers, police, postal workers, firefighters, and social workers. Additionally, workers are often attracted to the public sector for its job stability, livable wages, and greater access to retirement and health care benefits.25 Historically, government employment has also served as an important vehicle for Black workers to reach the middle class, affording them opportunities not available in the private sector.26

Pre-pandemic, unionized workers of all races and ethnicities were more likely to work in the public sector compared to non-union workers (Figure 5). Over half of unionized Black and White workers were employed in the public sector, compared to 13.5 percent and 11.5 percent of non-union workers, respectively. Comparatively few unionized Latinos worked in the public sector (41.3 percent), although unionized Latinos were more likely to work for the government than non-unionized Latinos.

Figure 5. Pre-Pandemic, Unionized Latinos were Less Likely to be Employed in the Public Sector
Workers by Unionization, Sector, and Race (2020Q1)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source: LPPI analysis of Current Population Survey Outgoing Rotation Group (2020-2021), available online.

During the COVID-19 pandemic, employment losses in the public sector were generally less severe than in the private sector; however, the magnitude of these losses varied across racial and ethnic groups.27 Unionized Black and AAPI public sector workers suffered the highest rates of job loss (Figure 6). In fact, a quarter of jobs held by unionized Black workers in the public sector were lost between the first quarter of 2020 and the second quarter of 2021. This could be due to the recession’s impact on state and local governments, whose primary industries employ a higher proportion of black women, as will be discussed in Section D.

Recent research from the Center for Equitable Growth suggests that working for the government used to protect Black workers from job loss, but it has become less protective against layoffs over time.28 Public sector employment was also slower to recover from the 2008 recession.29 These factors could also explain why unionized Black workers were more susceptible to job loss and have experienced less job growth than non-union workers.

In the public sector, non-unionized Latinos suffered greater pandemic-related job losses than unionized Latinos. However, comparatively fewer unionized Latinos are employed in the public sector, meaning that public sector job losses disproportionately impacted Black and AAPI workers.

Figure 6. Public Sector Union Job Losses Disproportionately Affected Black and AAPI Workers During the Pandemic
Change in Employment for Public Sector Workers by Union Status and Race (2021Q2 vs. 2020Q1)
Source: LPPI analysis of Current Population Survey Public Use Microdata (2020-2021). Available online.

Lingering Disparities Among Black and Latina Women

The aggregate figures discussed thus far don’t offer any insight into differences in gender-based economic outcomes. Previous LPPI research highlighted Latinas’ pronounced exit from the workforce during the COVID-19 pandemic, as well as Latinas’ and Black women’s elevated unemployment rates relative to White women. From March 2020 to March 2021, Latina labor force participation fell 2.74 percent, the largest change of any demographic group during that time period.

In part, these relatively high labor force participation drops are due to Latinas being disproportionately responsible for family care obligations when compared to Latino men, a burden exacerbated by school and day care closures.30 Gezici and Ozay also found that Latinas and Black women faced higher probabilities of unemployment—compared to White men—due to the COVID-19 pandemic, even after controlling for worker and job characteristics.31

The findings here reflect similar trends. Although unionized women experienced fewer job losses in the early stages of the pandemic, women’s employment has yet to reach pre-pandemic levels irrespective of union coverage (Figure 7). Outcomes are also worse for Latinas and Black women.

As of the second quarter of 2021, unionized Latina employment was 9.8 percent below pre-pandemic levels, compared to 7.5 percent for non-union Latinas. Similarly, 21.9 percent fewer unionized Black women were working as of the second quarter of 2021, compared to 4.1 percent fewer non-union Black women. White women have fared better; unemployment for unionized White women was only 6.3 percent below pre-pandemic levels as of the second quarter of 2021, while non-union employment was 4.7 percent lower.

The lack of economic recovery for unionized women is likely due to their relatively high levels of employment in the public sector. Women overall comprise over 60 percent of workers in state and local government.32 Nearly half of Black women were employed in the primary industries of state and local government (education, health services, and public administration) in 2019.33 The recession brought intense budget cuts to state and local governments. Consequently, public sector workers–disproportionately Black women–experienced the deepest job losses as states were forced to keep budgets balanced without debt financing.34

Overall, women held roughly two-thirds of the government jobs lost between June and September 2020.35 These heavy job losses likely dragged women’s union employment rates down during the recession. Women’s union employment rates have also been slow to recover.

Figure 7. Regardless of Union Status, Women of Color Face Extended Job Loss in 2021
Percent Change in Quarterly Employment by Union Status and Race (vs. 2020Q1)

 

 

 

 

 

 

 

Source: LPPI analysis of Current Population Survey Outgoing Rotation Group (2020-2021), available online.
Note: We do not disaggregate for unionized AAPI women due to small sample sizes.

Other Job Security Measures

The CPS’ new coronavirus-related questions shed light on other employment characteristics that varied by union status. Unionized Latinos were more likely to report working from home when compared to non-union Latinos. At the height of the COVID-19 pandemic in the second quarter of 2020, unionized Latino workers were nearly 10 percentage points more likely to have teleworked than non-union Latinos (Figure 8). Unionized Latinos consistently reported higher work-from-home rates throughout the entirety of the pandemic.

Unionized Latino workers were also more likely than their non-union counterparts to be able to work at all during the pandemic. At the peak of the pandemic in 2020Q2, unionized Latino workers were 11 percentage points more likely to report working than non-union Latinos. More than one in five non-union Latinos reported not working during the second quarter of 2020, compared to only one in 10 unionized Latinos.

As coronavirus cases have fallen and states have reopened, both union and non-union Latinos have returned to work, though fewer union-covered Latinos missed work at all. Both factors may point to the differences in occupation and sector for unionized Latinos: unionized Latinos are more likely to work in the public sector and in management and professional occupations, both of which can be performed remotely at higher rates than service work.

Figure 8. Unionized Latino Workers’ Employment Circumstances During COVID-19
Percent of Latino Workers who Worked Remotely or did not Work in the Last Four Weeks

 

 

 

 

 

 

 

Source: LPPI analysis of Current Population Survey Outgoing Rotation Group (2020-2021), available online.
Note: Data for 2020Q2 excludes April, as the Coronavirus questions were introduced in May.

Regression Results: Unions Decreased the Probability of Unemployment

The new questions introduced to the CPS in May also provide an opportunity to examine the potential effect of union membership on job stability. Building on Gezici and Ozay,36 we used the ORG dataset to examine whether gender, race, and union coverage were associated with workers’ probability of unemployment—or being unable to work in the last four weeks due to business closures or loss of business—during the pandemic.37

Compared with White men, all racial/ethnic and gender groups had a higher probability of experiencing unemployment from May 2020 through June 2021 (Figure 9).38 This negative effect is strongest for Black women and Latinas, who were 2.2 and 2.1 percent more likely to experience unemployment, corroborating our findings on their extended job loss. Latinos and Black men both experienced a 2 percent greater probability of unemployment over the same time period, while White women experienced the lowest probability of unemployment.

Union representation reduced the probability of being unable to work throughout the pandemic by 1.1 percent (Figure 9). The effect of union coverage was stronger for Latinos. Unionized Latinos had a 3.1 percent lower probability of unemployment compared to non-union Latinos. While unionization offers additional job stability to all workers, our analysis demonstrates that unions protect Latino workers to a greater degree.

Figure 9. Latinos More Exposed to Job Loss, but Unionization Offered Greater Stability During the COVID-19 Pandemic
Average Marginal Effects on the Probability of Unemployment (May 2020-June 2021)

 

 

 

 

 

 

Source: LPPI analysis of Current Population Survey Outgoing Rotation Group (May 2020-June 2021).
Note: All variables included here are statistically significant at the .001 percent level.

Union vs. Non-Union Wages for the Latino and Non-Latino Workforce

Turning to earnings, unionized workers consistently earned more than non-union workers prior to the COVID-19 pandemic. This trend has held true for both the Latino and non-Latino workforce, although the gaps have closed slightly.

In the first quarter of 2020, unionized Latino workers averaged earnings of $23.20 an hour—30.1 percent more than their non-union counterparts. At the peak of the COVID-19 pandemic in the second quarter of 2020, median wages for both unionized and non-union Latino workers increased (Figure 10). However, recent research from the Economic Policy Institute shows that wages grew in 2020 because most jobs lost in 2020 were held by low-wage earners.39 As more workers have re-entered the workforce, union wages have come down to their pre-COVID levels. Non-union wages have remained slightly higher, likely due to the heavy exit of low-wage workers.

Additionally, union-covered workers consistently earn higher wages than non-union workers across all racial and ethnic groups (Figure 10). This difference in pay is most significant for Black and Latino workers. In the second quarter of 2020, for instance, unionized Black and Latino workers earned 21.8 percent and 26.1 percent more than their non-union counterparts, respectively, compared to 9.9 percent for AAPI workers and 17.6 for White workers. While we do not control for worker characteristics here, these findings are in line with the economic literature, which finds the positive wage benefit from unions is greatest for Black and Latino workers (even after controlling for worker characteristics).40

Figure 10. Unionized Workers Consistently Earned Higher Wages than Non-Union Workers During the COVID-19 Pandemic
Percent Difference in Median Hourly Wage (Unionized vs. Non-Unionized) by Race


Source: LPPI analysis of Current Population Survey Outgoing Rotation Group (2020-2021), available online.
Note: Union wage benefit is not regression adjusted.

Despite the additional pay from union representation, however, Latinos consistently earned less than White and AAPI workers (Table 1). As of the second quarter of 2021, unionized Latino workers earned $23.20 an hour, while White union workers earned $27 an hour (a $3.80 per hour difference in pay). This Latino-White pay gap remains larger for non-union workers, with White non-union workers earning roughly $5 more per hour than non-union Latinos. Even with these persistent gaps, our analysis shows that unionized Latinos earned higher wages than non-unionized Latinos on average during COVID-19.

Similarly, unionized women across racial and ethnic groups consistently earned higher wages than their non-union counterparts. Pre-pandemic, unionized Latinas averaged $22.60 per hour, while non-union Latinas earned $15.60 (a difference of 30.9 percent). As of the second quarter of 2021, unionized Latinas earned $5.60 more per hour than non-union Latinas, while unionized Black and White women earned $4.30 and $5.20 more per hour than their respective non-union counterparts. Additionally, resembling workers in the aggregate, White women consistently earned more than both Black and Latina women regardless of union status.

Table 1. Latino and Black Workers Earned Less than White Workers During the COVID-19 Pandemic and Recovery, Regardless of Union Status
Median Hourly Wages by Quarter and Year (2021 Dollars)

Difference vs. White Workers
2020Q1 2020Q2 2021Q2 2020Q1 2020Q2 2021Q2
Unionized
Latino $23.20 $25.40 $23.20 -15.3% -8.0% -14.1%
AAPI $27.90 $32.10 $29.10 1.8% 16.3% 7.8%
Black $22.50 $23.30 $21.70 -17.9% -15.6% -19.6%
White $27.40 $27.60 $27.00
 
Not Unionized
Latino $16.20 $18.00 $17.20 -26.7% -24.4% -22.5%
AAPI $26.30 $30.40 $26.20 19.0% 27.7% 18.0%
Black $16.90 $18.20 $17.00 -23.5% -23.5% -23.4%
White $22.10 $23.80 $22.20

 

Source: LPPI analysis of Current Population Survey Outgoing Rotation Group (2020-2021), available online.
Note: Wages inflated to June 2021 dollars using Bureau of Labor Statistics CPI-U.

Policy Recommendations

As the economic recovery from the COVID-19 pandemic continues, union coverage is helping workers maintain steadier employment and higher wages, regardless of sector, occupation, race, ethnicity, and gender. For Latinos, data indicates that union representation correlates not only with higher levels of employment and the preservation of higher wages, but also ability to telework. However, gaps remain when it comes to preserving employment for Black, AAPI, and women workers, especially women of color.

We recommend the following policy actions to expand access to union coverage and close lingering gaps among marginalized groups, especially Latinos and Latinas:

  1. Pass and enforce federal legislation that supports the right to form a union and engage in collective bargaining.
    1. Reinforce the right to join a union that all workers are entitled to under the National Labor Relations Act (NLRA) of 1935.
    2. Pass the Protecting the Right to Organize (PRO) Act of 2021 which would make unionization more accessible for workers by dismantling right-to-work laws, legalizing solidarity strikes, banning employers from holding anti-union meetings during union elections, and prohibiting employers from taking disciplinary action against employees who seek to unionize.
    3. Impose stronger penalties on employers and corporations that use illegal practices to discourage and crack down on union organizing.
  1. In the short term, support workers not currently represented by unions by passing legislation at the state, local, and federal level that raises their standards of living.
    1. Pass a $15 minimum wage at the federal level.
    2. Pass state and local minimum wage laws that keep up with costs of living.
    3. Expand the capacity and diligence of social safety net programs, such as unemployment benefits and childcare services.
  1. Implement legislation and programs to mitigate wage inequities and employment instability in sectors who experienced heavy job losses during the pandemic and low unionization rates, and which tend to employ Latinos, African Americans, Asian Americans and Pacific Islanders, and women (e.g., service sector work and public employment).
    1. Support public sector employment by expediting the re-employment of workers adversely affected by the pandemic.
    2. Implement a federal jobs guarantee to expand access to public sector jobs that have historically provided financial stability and upward mobility for women and Black workers.
    3. Expand programs that provide childcare or childcare subsidies for women who are employed or desire to be employed.
  1. Implement long-term, systematic solutions that grant workers (especially workers of color) equitable pathways to union jobs that provide stability and job quality.
    1. Invest in workforce development initiatives—such as California’s High Road Training Partnership—that generate economic mobility for underserved and low-income workers through vocational training, education, and quality career ladder jobs while improving worker and employer competitiveness in a rapidly-changing, carbon-constrained economy.
    2. Expand access to career apprenticeships and career pathways into union jobs, especially for young workers of color.
    3. Incorporate worker voices and representation on corporate boards and into company decision-making.
    4. Eliminate onerous, costly, and inefficient civil service exams that pose unnecessary and discriminatory barriers to entry to unionized public sector jobs within state and local governments.

Conclusion

Labor unions offer more than solidarity. They are a means of collective power and economic mobility by which workers can gain greater job stability, improved pay, and benefits. Union representation is critically important in moments of economic shock to ensure that vulnerable workers—especially low-wage workers of color—have the material resources necessary to weather an unpredictable and unprecedented catastrophe. Unions both advocate for and function as a social safety net in and of themselves.

We find evidence that union membership was linked to maintaining employment stability for Latino workers through the COVID-19 pandemic. Unionized Latinos experienced fewer job losses when compared with non-union Latino workers. At the height of the recession, unionized Latino employment fell 2.5 percent, compared to 18.5 percent for non-union Latinos. What’s more, unionized Latino employment remained at pre-pandemic levels through the entirety of 2020 and only fell in 2021. After controlling for worker and job characteristics, we find that union coverage reduced the probability of unemployment by 1.1 percent for all workers and by 3.1 percent for Latino workers. These results are statistically significant at the .001 level.

Worker occupation and industry also matter. Before the pandemic, unionized Latinos were more likely to work in professional and management occupations, which maintained low unemployment rates and high work from home rates. Non-unionized Latinos were more likely to work in service occupations. Unionized Latinos were also more likely to work in the public sector than non-unionized Latinos. For Black workers, public sector employment, a majority of which is unionized, is a historically important means of entering the middle class. However, likely due to pandemic-related budget cuts, unionized Black workers lost a quarter of public sector jobs between the first quarter of 2020 and the second quarter of 2021.

Unionized workers of all races and genders also enjoyed higher wages throughout the course of the pandemic. The union-related pay bump is largest for Latino and Black workers but is a consistent feature of union employment for all workers. While we did not control for worker and job characteristics in our wage analysis, previous economic research confirms a significant wage premium, even after accounting for these variables.

These findings underline the importance of protecting workers’ right to join unions, form unions, and collectively bargain as a matter of federal policy. For non-unionized workers, rebuilding our economic foundations for the most vulnerable will ensure that workers receive just compensation for their efforts and have the resources to meet their full economic potential. This will require re-imagining current support systems—beyond increasing the minimum wage—to fully empower each Latina, woman, and worker of color to participate in a twenty-first century economy.

Endnotes

1 High Road Training Partnerships (California: California Workforce Development Board, June 2018), available online.

2 Celine McNicholas, Lynn Rhinehart, Margaret Poydock, Heidi Shierholz, and Daniel Perez, Why Unions are Good for Workers–Especially in a Crisis like COVID-19 (Washington D.C.: Economic Policy Institute, August 2020), available online.

3 Economic Policy Institute, Working People Have Been Thwarted in their Efforts to Bargain for Better Wages by Attacks on Unions, August 2019, available online.

4 United States Bureau of Labor Statistics, Union Members – 2020, (Washington D.C.: United States Department of Labor, January 2021), available online.

5 United States Bureau of Labor Statistics, Nonunion Workers had Weekly Earnings 81 Percent of Union Members in 2019 (Washington D.C.: United States Department of Labor, February 2020), available online.

6 Celine McNicholas, Lynn Rhinehart, Margaret Poydock, Heidi Shierholz, and Daniel Perez, Why Unions are Good for Workers–Especially in a Crisis like COVID-19 (Washington D.C.: Economic Policy Institute, August 2020), available online.

7 United States Bureau of Labor Statistics, Union Workers More Likely than Nonunion Workers to Have Healthcare Benefits in 2019 (Washington D.C.: United States Department of Labor, October 2019), available online.

8 United States Bureau of Labor Statistics, Nonunion Workers had Weekly Earnings 81 Percent of Union Members in 2019 (Washington D.C.: United States Department of Labor, February 2020), available online.

9 John Schmitt, Unions and Upward Mobility for Latino Workers (Washington D.C.: Center for Economic and Policy Research, September 2008), available online.

10 Christian E. Weller and David Madland, Union Membership Narrows the Racial Wealth Gap for Families of Color (Washington D.C.: Center for American Progress, September 2018), available online.

11 United States Bureau of Labor Statistics, Union Members – 2020, (Washington D.C.: United States Department of Labor, January 2021), available online.

12  Kim Parker, Rachel Minkin, and Jesse Bennett, Economic Fallout From COVID-19 Continues to Hit Lower-Income Americans the Hardest(Washington D.C.: Pew Research Center, September 2020), available online.

13 Armagan Gezici and Ozge Ozay, “An Intersectional Analysis of COVID-19 Unemployment,” Journal of Economics, Race, and Policy (2020) 3: 270-281, available online; and Kassandra Hernández, Diana Garcia, Paula Nazario, Michael Rios, and Rodrigo Domínguez-Villegas, Latinas Exiting the Workforce: How the Pandemic Revealed Historic Disadvantages and Heightened Economic Hardship (Los Angeles: UCLA Latino Policy and Politics Initiative, June 2021), available online.

14 We define workers as covered by a union if they were a) members of a labor union or b) covered by a union but not a member.

15 We use the 2020-21 Current Population Survey Public Use Microdata, accessed via the Integrated Public Use Microdata Series (IPUMS). Sarah Flood, Miriam King, Renae Rodgers, Steven Ruggles and J. Robert Warren, Integrated Public Use Microdata Series, Current Population Survey: Version 8.0, Current Population Survey (Minneapolis, MN: IPUMS, 2020).

16 Sample sizes are provided in Appendix Tables 1 and 2 (see attached report).

17 United States Bureau of Labor Statistics, Measuring the Effects of the Coronavirus (COVID-19) Pandemic Using the Current Population Survey, (Washington D.C.: United States Department of Labor, Updated October 2020), available online.

18 Armagan Gezici and Ozge Ozay, “An Intersectional Analysis of COVID-19 Unemployment,” Journal of Economics, Race, and Policy (2020) 3: 270-281, available online.

19 A direct comparison between unionized and non-unionized workers has limitations due to self-selection into labor unions. The probit regression model in this report addresses observed heterogeneity between unionized and non-unionized workers. However, this model does not address unobserved heterogeneity. See H. Gregg Lewis, “Union Relative Wage Effects” in Handbook of Labor Economics, Volume II (Amsterdam, Netherlands: Elsevier, February 1987).

20 For more details, see Economic Policy Institute, “Methodology for Measuring Wages and Benefits, February 2019, available online.

21 While the data analyzed in this brief uses the U.S. Census Bureau definition and term “Hispanic,” we use Latino and Hispanic interchangeably in this brief. Latinos can be of any race. All other race categories are non-Hispanic.

22 Matthew Dey, Harley Frazis, Mark A. Loewenstein, and Hugette Sun, Ability to Work from Home: Evidence from Two Surveys and Implications for the Labor Market in the COVID-19 Pandemic (Washington, D.C.: Bureau of Labor Statistics, June 2020), available online.

23 Julia Wolfe and John Schmitt, A Profile of Union Workers in State and Local Government (Washington D.C.: Economic Policy Institute, June 2018), available online.

24 Ibid.

25 Kate Bahn and Carmen Sanchez Cumming, The Consequences of Job Displacement for U.S. Workers (Washington, D.C.: Washington Center for Equitable Growth, October 2020), available online.

26 Michael Madowitz, Anne Price, and Christian E. Weller, Public Work Provides Economic Security for Black Families and Communities (Washington D.C.: Center for American Progress, October 2020), available online.

27 Kate Bahn and Carmen Sanchez Cumming, What the Coronavirus Recession Means for Public-Sector Employment (Washington D.C.: Washington Center for Equitable Growth, October 2020), available online.

28 Elizabeth Wrigley-Field and Nathan Seltzer, Unequally Insecure: Rising Black/White Disparities in Job Displacement, 1981-2017, Working Paper, Washington Center for Equitable Growth, February 2020, available online.

29 Edwin Pome and Nicholas Saxon, Effects of Economic Downturn on Private and Public Employment (Washington D.C.: Census Bureau, October 2019), available online.

30 Kassandra Hernández, Diana Garcia, Paula Nazario, Michael Rios, and Rodrigo Domínguez-Villegas, Latinas Exiting the Workforce: How the Pandemic Revealed Historic Disadvantages and Heightened Economic Hardship (Los Angeles: UCLA Latino Policy and Politics Initiative, June 2021), available online.

31 Armagan Gezici and Ozge Ozay, “An Intersectional Analysis of COVID-19 Unemployment,” Journal of Economics, Race, and Policy (2020) 3: 270-281, available online.

32 David Cooper and Julia Wolfe, Cuts to the State and Local Public Sector Will Disproportionately Harm Women and Black Workers (Washington D.C.: Economic Policy Institute, July 2020), available online.

33 Ibid.

34 Kate Bahn and Carmen Sanchez Cumming, What the Coronavirus Recession Means for Public-Sector Employment (Washington D.C.: Washington Center for Equitable Growth, October 2020), available online.

35 Michael Madowitz, Anne Price, and Christian E. Weller, Public Work Provides Economic Security for Black Families and Communities (Washington D.C.: Center for American Progress, October 2020), available online.

36 Armagan Gezici and Ozge Ozay, “An Intersectional Analysis of COVID-19 Unemployment,” Journal of Economics, Race, and Policy (2020) 3: 270-281, available online.

37 The baseline probit model includes eight gender/racial intersections. We also include control variables such as age, educational attainment, industry and occupation of employment, an industry-specific “essential” designation, whether each person worked remotely in the previous four weeks, time period, and state of residence. In our second model, we also include an occupation-specific “essential” designation, workers’ sector of industry (i.e., private vs public sector), and union representation. We also report separate results for the Latino subsample of our data. A full list of variables is included in Appendix Table 3, and full results are in Appendix Tables 4 and 5 of the above-linked report.

38 All findings reported here are statistically significant at the .001 percent level.

39 Elise Gould and Jori Kandra, Wages Grew in 2020 Because the Bottom Fell Out of the Low-Wage Labor Market (Washington D.C.: Economic Policy Institute, February 2021), available online.

See also Rakesh Kochhar and Jesse Bennett, Immigrants in U.S. Experienced Higher Unemployment in the Pandemic but have Closed the Gap(Washington D.C.: Pew Research Center, July 2021), available online.

40 “Union Wage Premium by Demographic Group, 2011,” Table 4.33 in Lawrence Mishel et al., The State of Working America, 12th Edition, an Economic Policy Institute book (Ithaca, N.Y.: Cornell Univ. Press, 2012), available online.

Images courtesy of Pop&Zebra and Jurien Huggins via Unsplash.