Skip to content
Report

The Varied Voice of Labor: Unpacking the Political Engagement of Labor in the 2024 Election

05/06/2025

“The Varied Voice of Labor: Unpacking the Political Engagement of Labor in the 2024 Election,” focuses on unions’ role in political education, communication, and mobilization. In collaboration with Columbia Labor Lab, we explore how unions serve as sources of information about the government, the economy, and specific public policies for their members, as well as how they mobilize members for political action across different levels of government.​​

Below are downloadable figures and tables from this report. To read the full report, download it here.

FIGURES AND TABLES

Figure 1

Download Image | LinkedIn  |   Facebook  |   Twitter  |   BlueSky

Figure 2

Download Image | LinkedIn  |   Facebook  |   Twitter  |   BlueSky

Figure 3

Download Image | LinkedIn  |   Facebook  |   Twitter  |   BlueSky

Figure 4

Download Image | LinkedIn  |   Facebook  |   Twitter  |   BlueSky

Figure 5

Download Image | LinkedIn  |   Facebook  |   Twitter  |   BlueSky

Figure 6

Download Image | LinkedIn  |   Facebook  |   Twitter  |   BlueSky

Figure 7

Download Image | LinkedIn  |   Facebook  |   Twitter  |   BlueSky

Figure 8

Download Image | LinkedIn  |   Facebook  |   Twitter  |   BlueSky

Figure 9

Download Image | LinkedIn  |   Facebook  |   Twitter  |   BlueSky

Figure 10

Download Image | LinkedIn  |   Facebook  |   Twitter  |   BlueSky

TABLE 1: Number of respondents for each union

A significant advantage of our approach was that we were able to construct a large enough sample to include significant subsamples of individual unions. We determined 50 was a minimum number of observations for each union to be included in any analysis. While we want to be cautious about inferences drawn from these subsamples, especially for smaller unions, we can still report preferences and attitudes of these unions’ members. The overall subsample for each union is listed in Table 1, below.

UnionRespondents
SEIU194
NEA157
AFSCME125
AFT98
IBT72
UFCW56
CWA56

Given our large sample of union voters, we were able to compare union members’ vote choice with non-union workers in our survey. We found that 56% of union members reported voting for Kamala Harris, compared to 48.2% of non-union workers; 32.4% of union members reported voting for Donald Trump, compared to 35.9% of non-members, reported in Table 2.

CandidateUnion member (%)Non-union member (%)Difference (pp)
Kamala Harris56% (SE = 1.4)48.2% (SE = 1.4)7.8 pp
Donald Trump32.4% (SE = 1.3)35.9% (SE = 1.3)-3.4 pp
Other3.6% (SE = 0.6)3.6% (SE = 0.5)0 pp
Did not vote7.9% (SE = 0.9)12.2% (SE = 0.9)-4.3 pp

Table 3: 2024 vote choice by union

We were able to explore voting patterns across individual unions, and find substantial differences, reported in Table 3. Among those unions where we had sufficient observations for comparison, Harris’ support ranged from just 32.9% among Teamsters (IBT) members to 77.5% among National Education Association (NEA) members. Unions with significant public sector membership like NEA and AFT generally show stronger Democratic support, while unions with more private sector representation like IBT and UFCW show relatively stronger Republican support.

UnionHarrisTrumpOtherNon-vote
SEIU57.3% (SE = 4.2)27.4% (SE = 3.6)1.7% (SE = 1)13.6% (SE = 3.5)
NEA77.5% (SE = 3.4)14.9% (SE = 2.9)4.5% (SE = 1.7)3.1% (SE = 1.4)
AFSCME52.1% (SE 5.6)31.9% (SE = 4.9)4.7% (SE = 2.1)11.3% (SE = 5.3)
AFT75.2% (SE = 4.6)14.1% (SE = 3.7)4.8% (SE = 2.3)6% (SE = 2.5)
IBT32.9% (SE = 6)49.4% (SE 6.8)7.8% (SE = 6.2)9.9% (SE = 3.9)
CWA65.8% (SE = 8.1)23.3% (SE = 7.8)4.9% (SE = 3.3)6% (SE = 3.5)
UFCW49% (SE = 8.3)39% (SE =  9)0.6% (SE 0.6)11.5% (SE = 4.5)

Table 4: Comparison of union members vs non-union workers on inflation experience

Many news outlets, pundits, and strategists have pointed to inflation as one of the top issues in the 2024 elections. As some commenters saw it, Harris was perceived by voters as representing the incumbent administration, and like many other incumbents around the world, Harris was punished by an electorate that was experiencing a prolonged cost-of-living crisis. One question we explored in this survey was whether there were alternative narratives around inflation that could have been articulated by the campaigns and civil society groups, such as unions. On the whole, the variation between union members and non-union workers was not particularly large, as reported in Table 4.

ExperienceUnion membersnon-union workers
Groceries91% (SE = 0.9)88% (SE = 0.9)
Gas67.1% (SE = 1.3)69.5% (SE = 1.3)
Rent30.1% (SE = 1.4)33.3% (SE = 1.3)
Mortgage19.8% (SE = 1.1)19.8% (SE = 1.1)
None4.1% (SE = 0.6)4.4% (SE = 0.6)

Table 5: Attribution of inflation, union vs non-union workers

As summarized in Table 5, union members were more likely to attribute inflation to corporate greed (48.5% vs 40.5%), while non-union workers were more likely to blame government policies (45.8% vs 38.8%). Given the timing of the survey, just four days after the inauguration of Donald Trump, and that the questions emphasized the 2024 election, we interpret blame for government decisions as directed towards the Biden-Harris Administration. These findings suggest that union members tended to understand the sources of inflation differently from non-union workers.

CauseUnion membersNon-union workers
Corporate greed48.5% (SE = 1.4)40.5% (SE = 1.4)
Government38.8% (SE = 1.4)45.8% (SE = 1.4)
Covid-19 Pandemic9.8% (SE = 0.8)9.4% (SE = 0.8)
Higher wages3.1% (SE = 0.6)4.3% (SE = 0.6)

Table 6: Vote choice by inflation attribution among union members

We found that beliefs about the causes of inflation among union members strongly predicted voting behavior in the 2024 election, as shown in Table 6. Those union members who blamed corporate greed overwhelmingly supported Harris (68.8%), while those blaming government policies predominantly supported Trump (53.6%).

CauseHarrisTrumpOtherDid Not Vote
Corporate Greed68.8% (SE = 1.4)17.6% (SE = 1.2)4.3% (SE = 0.7)9.3% (SE = 1)
Government32.9% (SE = 1.4)53.6% (SE = 1.5)3.2% (SE = 0.5)10.3% (SE = 1)
Pandemic67.7% (SE = 2.9)21.7% (SE = 2.6)2.1% (SE = 0.8)8.5% (SE = 1.8)
Higher wages38.2% (SE = 5.6)42.1% (SE = 5.4)3.4% (SE = 1.8)16.3% (SE = 4)

Table 7: Attribution of inflation by union

The final variation across union members we explore is across unions, and as outlined in Table 7, below, there were large differences across unions in attribution of inflation. In general, more left-leaning unions representing public-sector workers were more likely to attribute inflation to corporate greed over the government or the pandemic, with two important exceptions: SEIU (where a slim plurality of members blamed the government over corporate greed) and UFCW (where corporate greed was neck-and-neck with government as the source for inflation). 

UnionCorporate GreedGovernmentPandemicHigher Wages
SEIU42.8% (SE = 4.3)46.8% (SE = 4.3)7.9% (SE = 2.1)2.5% (SE = 1.3)
NEA65.8% (SE = 3.9)24.3% (SE = 3.6)8.9% (SE = 2.4)1% (SE = 0.7)
AFSCME52.8% (SE = 5.6)34.9% (SE = 5.1)6.8% (SE = 2.3)5.5% (SE = 3.9)
AFT59.4% (SE = 5.1)21.9% (SE = 4.3)16.8% (SE = 3.9)1.9% (SE = 1.4)
IBT40.5% (SE = 7)50.1% (SE = 6.8)7% (SE = 3)2.4% (SE = 1.7)
CWA54.3% (SE = 7.8)31.2% (SE = 7.1)12.7% (SE = 4.5)1.7% (SE = 1.7)
UFCW41.5% (SE = 7.9)46.3% (SE = 8.6)8.7% (SE = 4)3.5% (SE = 2.5)

Table 8: information source about the economy, among union members

We asked respondents, “thinking about how you get information about the economy, what source of information do you trust the most?” As Table 8 shows, there was significant variation in how respondents answer. Notably, unions were not a common source of information about the economy, even for union members.

Source% Selected
TV or radio22.1% (SE = 1.2)
Newspapers17.4% (SE = 1)
Social media or podcasts17.2% (SE = 1)
Union11.8% (SE = 0.9)
Friends or family11.6% (SE = 1)
Political candidates1.9% (SE = 0.4)
None of the above17.9% (SE = 1.1)

Table 9: Local union asked member in 2024 to…

As Table 9 indicates, union members were most frequently asked to share political information in their community (20.8% of all union members reported this ask), and were least likely to be asked to contribute money to political campaigns (just 10.2%). Union requests for political engagement to members tended to be clustered: most union members (64%) who received at least one request from their local union also received at least one other type of mobilization. 

Attend a political event17.4% (SE = 1.1)
Volunteer for political campaign12.9% (SE = 1)
Share political information in community20.8% (SE = 1.1)
Contribute money to political campaign10.2% (SE = 0.9)
Contact local official17.2% (SE = 17.1)
Contact state official16.9% (SE = 1)
Contact federal official12.5% (SE = 1)
No political mobilization56.1% (SE = 1.4)

Table 10: Frequency of political mobilization across unions

Table 10 shows that the frequency of union members reporting no political mobilization varied from under half in the NEA, CWA, and SEIU to over 70% for IBT and UFCW. Different unions clearly vary in the intensity of their political mobilization, with members in unions like the NEA or CWA receiving about two different types of political mobilization requests, while AFSCME, IBT, and UFCW members received fewer than one type of political mobilization request, on average.

UnionNo political mobilizationAverage number of types of political asks
NEA39.8% (SE = 4)1.7 (SE = 0.2)
CWA44.1% (SE = 7.8)1.7 (SE = 0.3)
SEIU43.8% (SE = 4.2)1.2 (SE = 0.1)
AFT44.4% (SE = 5.2)1.2 (SE = .2)
AFSCME57.6% (SE = 5.6)0.9 (SE = 0.2)
IBT67.8% (SE = 6)0.6 (SE = 0.1)
UFCW81.1% (SE = 5.5)0.3 (SE = 0.1)

Table 11: Types of union political mobilization by sector

In Table 11, we review the types of political mobilization requests from unions to their members across different sectors. While public sector union members, and especially those employed in K-12 education, tended to receive the most requests for political engagement overall, there was high intensity political mobilization in other sectors as well, such as construction union members being asked to attend political events or share political information.

All union membersPrivate sector membersPublic sector membersEducation membersManufacturing membersConstruction members
Attend a political event17.4% (SE = 1.1)15.9% (SE = 1.5)18.8% (SE = 1.6)20% (SE = 2.4)13.4% (SE = 4.3)22.8% (SE = 4.8)
Volunteer for political campaign12.9% (SE = 1)11.1% (SE = 1.2)14.9% (SE = 1.5)17.9% (SE = 2.3)9.7% (SE = 3.9)13.9% (SE = 3.8)
Share political information in community20.8% (SE = 1.1)18.3% (SE = 1.5)23.5% (SE = 1.7)26.1% (SE = 2.5)13.1% (SE = 3.8)23% (SE = 4.8)
Contribute money to political campaign10.2% (SE = 0.9)8.9% (SE = 1.2)11.6% (SE = 1.2)11.3% 
(SE = 1.6)
4.8% (SE = 2.4)13.2% (SE = 3.4)
Contact a local official17.2% (SE = 17.1)15.5% (SE = 1.4)19% (SE = 1.4)26.2% (SE = 2.2)9.8% (SE = 3.8)16% (SE = 4)
Contact a state official16.9% (SE = 1)13.9% (SE = 1.4)20% (SE = 1.5)27% (SE = 2.2)4.3% (SE = 2.5)13.6% (SE = 3.7)
Contact a federal official12.5% (SE = 1)9.4% (SE = 1.1)15.8% (SE = 1.4)15.5% (SE = 1.8)2.4% (SE = 1.7)8.3% (SE = 3)

Table 12: Frequency of Union Political Communication about Government

As Table 12 indicates, communication about government was more common than political mobilization requests, and roughly similar across levels of government. Most union members (just under 80%) reported receiving information about the government at least once a year, and the most common frequency was between monthly and a few times a year. A small minority of union members-around 13%–reported very frequent communications on a more regular basis, every week or two. 

FrequencyState / local governmentFederal government
Every week or two13.4 (SE = 1.2)12.9 (SE = 1)
Every month28.9 (SE = 0.7)24.7 (SE = 1.2)
A few times a year29.2 (SE = 1.3)31.1 (SE = 1.3)
Once a year6.6 (SE = 1.3)7 (SE = 0.7)
Never21.9 (SE = 1.2)24.2 (SE = 1.2)

Table 13: Union Members Reporting No Information about Government, by Union

As with political mobilization, there was substantial variation across unions in the extent to which they were sharing communication about different levels of government. We document this variation in Table 13, which shows the proportion of union members reporting no communication from their local union about either state or local government or the federal government.

UnionNo state / local government information sharingNo federal government information sharing
SEIU14.5% (SE = 2.6)14.8% (SE = 2.6)
NEA13% (SE = 2.7)20.3% (SE = 3.2)
AFSCME26.8% (SE = 5.4)38% (SE = 5.5)
AFT12.6% (SE = 3.6)18.4% (SE = 4.1)
IBT26.1% (SE = 5.5)34.4% (SE = 6.9)
CWA24.7% (SE = 7.9)20.4% (SE = 7.7)
UFCW48.6% (SE = 8.3)46.2% (SE = 8.2)

ABOUT THE AUTHORS

Alexander Hertel-Fernandez is an associate professor of International and Public Affairs at Columbia University’s School of International and Public Affairs and co-director of Columbia Labor Lab. He is political scientist who studies the political economy of the United States, with an emphasis on the politics of organized interests and public policy.

Max Kiefel Headshot

Max Kiefel is a Fellow at the Center for Labor and a Just Economy. His research is in political sociology and comparative political economy. At CLJE, Max is leading two research projects: an exploration of how participation in democratic unionism influences the political development of workers; and an analysis of how workers are responding to the energy transition, with a specific focus on the electric vehicle industry.

Alan Yan Headshot

Alan Yan is a PhD candidate at U.C. Berkeley and the director of research at Indigo Engineering. He’s broadly interested in the intersection of class, race, and ideology. Specifically, he is working on how class and racial rhetoric affect the strength and salience of class and racial identities and the types of policy voters will support. Along these lines, he is examining workers’ preferences toward economic democracy and developing a competing conceptualization of ideology.

Our Latest

All Our Latest →