The latest polling data on likely Donald Trump voters shows us more clearly what the “Trump coalition” looks like. As Michael Tesler wrote recently in the Monkey Cage, it “unites resentment of minority groups with support for economically progressive policies,” as well as opposition to free trade.

Democrats worry that Trump’s focus on economic protectionism and opposition to trade deals could turn Democrat-leaning Rust Belt states into swing states, giving the Republican nominee an unexpected path to the White House in 2016.

We can learn a little from students of Western European politics here.

As Cas Mudde, another Monkey Cage contributor, points out, the “supporter of Trump … is almost identical to the populist radical right voter in [Western] Europe.”

  • They share a strong opposition to immigration and free trade, as well as support for the social insurance programs they benefit from.
  • They have in common frustration and anger at the “establishment” and at the political “system.”
  • They are both very likely to be from non-urban, blue-collar households, to have received only secondary education, and to live in areas facing economic decline.

Here’s what we know about radical right voters in countries such as France or Britain:

1) Blue-collar European workers who now vote for the radical right didn’t get there from the left side of the political spectrum.

2) Any shifts from the left to the radical right are glacially slow. In Europe, working-class voters first move away from mainstream left parties. A subset of these voters then moves toward the radical right. In other words, working-class voters abandon the left before they shift allegiances to the extreme right.

If we take that and extrapolate to the U.S. elections, we can conclude that:

1) Trump is very unlikely to attract working-class voters who currently self-identify as Democrats, even in Rust Belt states.

2) U.S. working-class voters in Rust Belt states who end up voting for Trump instead of Hillary Clinton in November were almost certainly not going to vote for a Democrat in the first place.

Parallels from France and Britain

Let’s look at these points in more detail. If the French experience with the National Front (FN) tells us anything, it’s that people who harbor a mix of anti-system, anti-immigrant and pro-spending views aren’t especially attracted to the “left.”

French social scientists Florent Gougou and Nonna Mayer document this extensively:

The most inclined to vote for the National Front are the right-wing workers, the “conservative blue collars,” or the “neither-nor-ers,” those who define themselves as “neither right wing nor left-wing,” the politically disaffected. Even in the golden age of the “class vote” there always was a sizeable minority of the working class voting for the Right, either because of the traditions of the region where they lived or for individual reasons. … And if one looks at the workers who declared an intention to vote for [former FN standard-bearer Jean-Marie] Le Pen on the eve of the first round of the 2007 presidential election, 43 percent defined themselves as ‘somewhat right wing,’ 41 percent ‘neither left wing nor right wing’ or refused to answer. Only 16 percent claimed to be ‘somewhat left wing.’

In Britain, Geoffrey Evans and Jon Mellon show that very few Labour voters jumped ship to swim directly over to the far-right U.K. Independence Party. Data from the last general elections confirm this: Between 2010 and 2015, Conservatives lost much more support to UKIP than Labour did. In the short term, left-wing voters seem unlikely to switch to populist and anti-immigration candidates on the extreme right.

Sure, UKIP has supporters with a history of voting Labour. But these voters defected long ago in reaction to the Labour Party’s move to the center right on economic issues after Tony Blair’s election as party leader in 1994. Later, some of these disaffected voters shifted still further right to Nigel Farage’s anti-immigration, anti-Europe party.

The same is true in France. The decline in the share of the working-class vote going to the left did not benefit the radical right. As you can see in the table below, between 1981 and 2007, there was a sharp decline in the share of working-class votes going to left-wing parties. During this period, the share of the working-class votes going to Jean-Marie Le Pen was pretty much stable, at around 18 percent.

Compare 2007 and 2012: When the FN does well, it is the mainstream right-wing party that suffers (and vice versa).

Who do blue-collar workers vote for in France’s elections? 1981-2012

The FN share of the working-class vote has increased sharply since the 1980s. (It was close to 40 percent in the last regional elections.) However, this is not because working-class voters switched en masse from voting for the left to voting for the radical right.

What, then, explains this increase? The bulk of the FN uptick comes from generational replacement. As Gougou explains, the working-class individuals who vote for right-wing parties today are not the same people as the working-class individuals who supported left-wing parties yesterday.

What about U.S. voters?

American survey data indicates that some of the short- and long-term patterns documented in Europe are at play in the United States.

There is little evidence that Trump will succeed in poaching voters from the Democratic camp.

The most inclined to support Trump have no sympathy for the Democratic Party. Individuals who identify as Democrats have little sympathy for Trump.

A pilot study from the American National Election Studies (ANES)  asked respondents, “Regardless of whether you will vote in the Republican primary this year, which Republican candidate do you prefer?” The survey allowed respondents to voice their sympathy for a given candidate, independent of electability or any other strategic considerations.

Four out of five (80 percent) pro-Trump answers to this question came from individuals who identify as either “independent” or “Republican.” That’s true even if you look only at answers from non-college-educated whites.

Pundits have worried that a Clinton vs. Trump matchup could push some voters, especially independents who are attracted to Bernie Sanders, toward Trump.

I find no evidence that pro-Sanders independents are more likely to also express support for Trump. Among independents who like Sanders, 16 percent also like Trump. Among independent who express support for another democratic candidate than Sanders, 28 percent also like Trump. If Trump attracts independent voters, it is unlikely they will be Sanders sympathisers.

Like the radical right in Europe, Trump is most likely benefiting from long-term evolutions among white voters with no college degree in the U.S. working class

Since 1972, the independent NORC General Social Surveys (GSS) have looked at attitudes and trends in American society. Lane Kenworthy and colleagues use longitudinal data from the GSS to examine which party white voters without a college degree identify with.

Their findings are shown in the figure below. In summary:

  • Since 1970, the share of non-college-educated whites calling themselves “Democrats” fell by approximately 20 points, from 60 percent to 40 percent.
  • Most of the decline happened between 1970 and 1991. While identification with the Democratic Party decreased, identification with the Republican Party increased, reaching 40 percent in 1991.
  • Since the early 1990s, working-class support for the Democrats has been stable at around 40 percent.
  • During this period, the share identifying as independent has doubled from 12 to 25 percent. In contrast, the share of college-educated whites who identify as independents has remained stable at around 10 percent.

Readers might wonder whether education (tallied here as the absence of a university degree) is the best way to identify low-skilled workers affected by globalization and free trade. The GSS also provides information on individual respondents’ jobs. I ran the same analysis focusing on whites in traditional blue-collar jobs (instead of using education). I found the same patterns.

Together, the ANES and the GSS results indicate that Trump supporters are unlikely to have been Obama voters. However, once upon a time, they might have been Jimmy Carter voters.

Trump’s rise most likely benefits from the working class slowly, over a generation, leaving the Democratic Party. Trump’s rise hasn’t caused that defection. The fact that he’s benefiting is the result of forces that have been at work for decades.

As in Europe, generational replacement plays an important role. Voters born after 1960 are the ones feeding the post-1991 increase in the share of independents. Again, this is specific to non-college-educated whites: There is no similar trend among younger cohorts of college-educated whites.

Trump’s Rust Belt strategy: The bottom line

Here’s the flip side of the findings, above: In the short term, leaders of the Democratic Party should not be too concerned about defection from white blue-collar Democrats.

Those who will vote for Trump left the Democratic Party a few election cycles ago, without turning these states from blue to red.

White working-class voters who have stuck with the Democratic Party have a strong party identity. They don’t share Trump’s ethnocentric worldviews. There is no reason to expect that they will defect toward Trump.

Trump will need to mobilize working-class voters who have already voted Republican in the past, while also dramatically increasing the turnout among the younger and politically unaffiliated white working class.

Note: Due to changes during editing, an earlier version of this post inaccurately characterized subgroup results from the ANES survey, in a way that did not alter conclusions of the post as a whole. It has been updated to correct those inaccuracies. 

Charlotte Cavaillé is a post-doctoral fellow at the Institute of Advanced Studies in Toulouse, studying comparative politics and the political economy of advanced capitalist countries.