While recent political debate has often focused on the shift of working-class voters away from the Democratic Party, new analysis provides evidence that another shift is actually driving the modern political divide. Over the past 40 years, high-income, highly educated, white-collar white voters have steadily moved toward the Democratic Party. These findings were recently published in the Proceedings of the National Academy of Sciences.
Researchers Karin Bilbig and Paula Englund conducted the study to examine common theories about voting behavior in the United States. After the 2016 presidential election, many political commentators suggested that Donald Trump won because the white working class had deserted the Democratic Party. Some analysts argued that Democrats lost these voters by failing to offer a strong economic solution to lost manufacturing jobs. Other commentators have suggested that this change is related to racial status threats following the election of President Barack Obama.
Wilbig and England found that many of these past analyzes focused only on recent elections and treated socioeconomic status as a simple binary concept. They wanted to evaluate a longer timeline to see exactly which groups of voters changed their political preferences. To do this, the researchers analyzed data from 1980 to 2020 and mapped out the voting habits of different socio-economic groups over time. They aimed to see whether changes in political inequality are caused by low-status voters moving to the political right or by high-status voters moving to the political left.
To investigate these trends, scientists used data from the General Social Survey. This is a nationally representative survey that tracks the attitudes and behaviors of people living in the United States. The researchers limited their analysis to whites who reported voting in the most recent presidential election, resulting in a sample size of 27,795 voters. They specifically focused on white voters. Because this demographic is experiencing the most remarkable political realignment in recent history.
The researchers measured socioeconomic status using three different variables, including household income, education level, and occupational class. Regarding income, the survey data were adjusted for inflation and household size. Next, we divided voters into deciles. This means that the population has been divided into 10 equally sized groups based on income. Education was divided into five categories, ranging from those with less than a high school diploma to those with a graduate degree or higher.
Occupational classes were divided into three broad categories. The working class included manual laborers such as roofers and lower-level service workers such as cashiers. The middle class included non-professional self-employed workers and supervisors of manual workers. Finally, the white-collar class consisted of professionals, administrators, and senior administrative personnel such as teachers and registered nurses.
When researchers looked at household income, they found little evidence that Democrats were steadily losing low-income white voters. The bottom five income deciles did not show a consistent long-term trend away from Democratic presidential candidates over four decades. Although support for the Democratic Party declined among some middle- and low-income groups in the 2016 election, it often rebounded in 2020. For example, the low-income 10% of voters has remained roughly evenly divided between the two major parties since 2004.
Rather, the most consistent trend has been large-scale changes in top earners. In 1980, only 27 percent of white voters in the top 10 income groups supported Democratic presidential candidates. By the 2020 election, that number had increased to 61%. This upward trend occurred rapidly, with notable increases in approval ratings around the 2008 and 2020 elections.
Analyzes of education level revealed a similar pattern. White voters without a high school diploma have recently begun to turn away from the Democratic Party, dropping from 49 percent support in 2012 to 26 percent in 2020. But this particular group has shrunk as a share of the electorate, making up just 3 percent of all white voters in 2020. White voters with a high school diploma have also shown a decline in recent years, but their support for Democrats in 2020 remained above the decline. 1980s and 1990s.
At the top tier of education, the data shows a significant shift toward Democrats. In 1980, only 34 percent of white voters with advanced graduate degrees supported Democratic candidates. By 2020, this number had doubled to 68%. White voters with a standard four-year college degree also increased significantly, and Democratic support rose from 36% in 2004 to 53% in 2008 and has stabilized at a higher level ever since.
Finally, when assessing occupational class, researchers found that working-class white voters have fluctuated up and down over the decades. Historically, this group has tended to move back and forth between the two major parties depending on elections. Working-class support for Democrats did fall from 47 percent in 2012 to 36 percent in 2020. However, this recent low is very similar to previous historical lows observed in the 1984 and 2000 elections, meaning there has been no consistent downward trend for 40 years.
In contrast, white-collar workers showed a significant and sustained increase in support for the Democratic Party throughout the period studied. In 1980, only 32 percent of white-collar voters supported Democratic candidates. This number steadily increased, reaching 54 percent by the 2020 election. The data provide evidence that changes in the political divide between different social classes are driven primarily by a steady shift to the left among wealthy and highly educated professionals.
The authors point out several potential misconceptions and limitations of their study. People might think that low-status white voters don’t move to the right at all. While recent elections have certainly tilted lower-status voters toward the Republican Party, this shift only began around 2012. Researchers caution against assuming this recent drop is a permanent realignment, as it may just be a temporary response to Donald Trump’s specific political appeals.
Another interesting point is why political analysts got these trends wrong in the first place. Scientists suspect that many people who study elections belong to these high-status groups. These analysts may not have been aware of the change in their demographics, as their own social circles gradually moved leftward. Rather, they may have believed that the growing political divide was solely due to a shift to the right among working-class people.
Future research should continue to monitor future elections to see if the post-2012 shift to the working class becomes a permanent trend. Additionally, scientists need to investigate differences in how people vote and how they identify themselves. Supplementary data to this study suggest that low-status white voters are abandoning the Democratic identity label at a faster rate than they are abandoning Democratic voting habits. Examining the lag between party identification and actual voting behavior could provide deeper insights into contemporary political polarization.
The study, “Elites moved closer to democracy than non-elites moved away: Income, education, and occupational class in U.S. presidential elections from 1980 to 2020,” was authored by Karin Bilbig and Paula Englund.
