Scientists studying U.S. census rolls from the 1940s have discovered a surprising pattern: White men who grew up one door down from a Black neighbor were more likely to register as Democrats 70 years later — even when compared to other whites from the same neighborhood.
The findings, described in the journal Science Advances, suggest that young boys exposed to people of different backgrounds may grow up to be more liberal-minded adults.
"White men who had a Black neighbor in 1940, compared to white men who did not, are more likely to be associated with racially liberal politics, as indicated by their registration with the Democratic Party even as late as 2017," the authors wrote.
It’s an idea supported by converging lines of evidence from economics, political science, psychology and sociology, said Linda Tropp, a social psychologist at the University of Massachusetts Amherst who was not involved in the study.
As diversity rises in the U.S. and other western countries, it has been met with stiff resistance from some quarters, said Ryan Enos, a political scientist at Harvard University and one of paper’s lead authors.
“One thing we know from a lot of academic research — and we know this from just our own two eyes — is that there can often be negative consequences of [diversity] because some people aren't comfortable with it,” Enos said.
Recent history in the U.S. provides a host of examples. Enos pointed to the 2016 election of Donald Trump, a Republican who took office following the nation’s first Black president, Barack Obama, a Democrat.
On the campaign trail, Trump courted voters with rhetoric frequently deemed racist or xenophobic. It was Trump supporters who led the deadly Jan. 6 insurrection at the U.S. Capitol following the former president’s attempts to cast doubt over his reelection loss.
But many researchers argue that the negative reactions to diversity, including discrimination and social and political instability, are short-term effects — and that diverse societies can grow to be very successful in the long run.
Interpersonal relationships are key to that goal, scientists say. These bonds between individuals across different ethnic groups can help reduce prejudice and lead to more social harmony.
This kind of contact may be especially powerful when those bridges between people are built in adolescence, psychologists say.
But it’s been difficult to measure the long-term effects of social contact between different groups. Studies have typically measured the effects of this kind of contact within a matter of days.
For this paper, Enos and his colleagues took a much longer view, combining cutting-edge machine learning techniques with U.S. census data from 80 years ago.
The 1940 census, in its attempt to record every person living in the country that year, likely reached roughly 99% of the population. Workers went door-to-door collecting personal information for residents, including name, age, gender, race, place of birth, years of education, labor earnings and employment status. (These individual records only recently became available: They aren’t made public for 72 years in order to protect the privacy of the people who respond.)
The scientists then used an algorithm to match as many people from the census as they could with voter file data from California in 2005, North Carolina in 2009 and both states plus Nebraska in 2017. These three were the only states where citizens provide their place of birth when registering to vote, making it easier to match files. But in 1940, these same folks were living all over the U.S., providing the scientists with near-blanket coverage of the country.
The scientists used the registered voters’ declared political party as a proxy for political leanings. They focused on men, who typically kept the same name throughout their lives. Women, who often changed their surnames after marriage, were harder to track.
Because census workers went door-to-door, residents’ names would appear right by their next-door neighbors on the list, two names away from their neighbor who lived two homes down, and so on. This meant that the scientists could get an extremely fine-grained look at which individuals had an immediate neighbor of another race.
After examining the records for more than 650,000 white men, the researchers found that those who lived next door to Black neighbors in their youth were 1.5 to 4.2 percentage points more likely be registered Democrats in the 2005/2009 data, and 2.8 to 5.3 percentage points more likely to be Democrats than their peers in the 2017 data. They were also less likely to be registered Republicans by about the same amount.
Given that Democratic affiliation is associated with more racially liberal politics, the researchers said it may reflect a more liberal outlook on the part of white men who grew up with Black neighbors.
Racial attitudes and partisanship are "highly correlated in the U.S. and other countries, with members of left-of-center parties consistently displaying less ethnic and racial prejudice than members of right-of-center parties," the authors wrote. "Indeed, in the United States, anti-Black prejudice was a major driver of the sorting of voters into parties in the mid-Twentieth Century and remains among the strongest predictors of vote choice in recent elections."
The scientists also ran separate analyses to account for other possible confounding factors: whether some progressive white families chose to live near Black neighbors; whether some Democrats were more conservative holdovers from the era of the Southern Democrat, before desegregation and the civil rights era motivated many white Democrats to move to the Republican Party; and whether the effect held in rural areas, where homes were spaced farther apart. In all cases, the pattern held.
While the percentages may seem small, they have the potential to make a significant impact, especially when multiplied by the millions of people who lived in the U.S. at the time of the study, Enos said.
“Ask how many more or less people would have different racial attitudes if we had a more or less segregated society at the time — it adds up to a lot,” he said.
Tropp praised the findings’ ability to look at such detailed population data.
The results highlighted the need to be clear on what’s meant by “contact” with people from different backgrounds, she added.
Previously, “what people understood to be contact between members of different groups was kind of a fuzzy concept,” she said.
Some research has shown that having greater proportions of another group in a community can be associated with more negative intergroup attitudes — or greater perceptions of threat posed by that other group, she said.
Simply living in the same neighborhood, no matter how diverse that neighborhood is, may not qualify as real contact. In fact, it may be counterproductive if it’s not combined with actual person-to-person interaction.
“What I think is really special about this research is that it's focusing not just at the neighborhood level but at the level of a specific neighbor … so that you're really getting closer to people's lived experiences,” Tropp said.
Tyrone Forman, a sociologist at the University of Illinois at Chicago, sounded a note of caution, pointing out that the study assumes that neighbors of different races had some meaningful interaction.
“That’s a big assumption,” he said — particularly considering how segregated much of the United States was at the time.
And although living side-by-side with people of other races or ethnicities may have long-term social benefits, American society may not be integrated enough to take advantage of them, he added.
“We continue to live in a society in which schools are re-segregating,” he pointed out. The urgency to integrate American communities, he added, “is no longer there.”
“At the national scale there's a heck of a lot of resistance to racial integration,” she said. “And I think that's a real problem for our society moving forward.”
This story originally appeared in Los Angeles Times.