Donald Trump’s victory in the presidential election highlighted that when it comes to identity politics, the largest and most powerful identity group has a natural advantage. White identity politics help, at least in part, to explain Trump’s success. And insofar as racial identity plays a chief role in political polarization, candidates who promote white identity political movements like “All Lives Matter” and xenophobia-imbued protectionism will continue to wield a numerical edge. In contrast to identity-centered organizing, affinity-based organizing hinges on shared aims and values. Affinity-based organizing should be employed by political parties to avoid the numerical advantage of a candidate just because they belong to the largest identity group.

In 2015, Census data projected that by 2044, more than half of Americans will belong to a minority group. But historically, “white” has remained the largest identity category in the U.S. because its membership expands to include new identities over time. Therefore, it is unlikely that whiteness will be displaced as a plurality identity.

As an example of this expansion, consider the Irish. In How the Irish Became White, historian Noel Ignatiev documented the complex ways in which the Irish population in the U.S. began self-identifying as white and being perceived as white by American society as they became more materially and socially successful. Similarly, historians James Barrett and David Roediger explore how eastern and southern European immigrants who came to America during the late 19th century and early 20th century were originally considered non-white. For example, Barrett and Roediger note that the term guinea, which referred to African slaves sold off the northwest coast, was at one point applied to Italians, indicating that they were not considered part of the white majority. Likewise, historians found that in the early 20th-century, Greeks were sometimes discriminated against under Jim Crow laws. Today, these groups are generally considered to be white.

Currently, there are hints from the Census that Hispanics are now “becoming white.” In 2014, the Pew Research Center reported that 2.5 million Americans who said they were of Hispanic origin — the Census does not include Hispanic origin as a race — and “some other race” in 2000 marked that they were of Hispanic origin and “white” 10 years later. Of course, this is just a tick box on a census form. But since there is historical precedent, it’s not unreasonable to think we are seeing whiteness expand again. As long as whiteness continues to expand, it has a good chance of remaining the largest racial identity group. Thus, when race becomes a salient political marker, politicians like Trump who affirm white identity will continue to have an advantage.

Trump’s victory cannot be attributed to identity politics alone, but policy specifics were certainly not his calling card. An article in the journal Trends in Cognitive Science recently proposed a relevant voter choice model. They suggested that on every issue, voters get some policy utility (“Will my taxes be lower?”) and some identity utility (“Does this boost my social identity?”). Trump’s proposed policies throughout the election cycle oscillated between vacuous and contradictory. So his policy utility — insofar as his policies can be decoded — was probably not nearly as high as the identity utility people got from him.

It’s simplistic to argue over whether Trump was elected by racists or people with legitimate economic grievances. This assumes the electorate is sectioned into heroes and villains. Trump’s campaign resonated with a lot of people who have a shared identity of whiteness and are also feeling downtrodden. This isn’t new. His campaign echoed those of self-described “paleoconservative” Pat Buchanan in 1992 and 1996 and populist William Jennings Bryan in 1896 — two other American politicians who resonated with down-and-out whites.

The Jan. 2 Economist article, “Pitchfork Politics.” opened with the claim “Before Donald Trump, there was Pat Buchanan.” Socially, Mr. Buchanan was panned as too bigoted for mainstream conservatives. Economically, like Trump, he mixed protectionism and xenophobia, leveraging the familiar idea that immigration and free trade acts are harming America’s working class. The Economist profile goes on to summarize his grievances: “Immigrants have reached even small communities, factory jobs have vanished and interventionist wars launched by George W. Bush left Americans ‘with ashes in our mouths.’”

On the similarities between Bryan and Trump, David Frum wrote in The Wall Street Journal, “Both men championed constituencies that formerly occupied a position of cultural and political dominance: small farmers in Bryan’s case, the white working class in Trump’s.” Exchange “gold standard” for “globalization” and “Southern farmers” for “Rust Belt blue collar workers,” and the similarities become clear. Just as Trump bemoans the decline of manufacturing work, Jennings warned of the dire consequences of decentering agrarian farmers: “Burn down your cities and leave our farms, and your cities will spring up again as if by magic; but destroy our farms and the grass will grow in the streets of every city in the country.”

In short, the trope of white identity politics for the disaffected is nothing new in America.

What can be done? Basing politics around affinity — shared goals and values — rather than identity could take away the inherent numerical advantage of white demagogues in identity politics. Trump isn’t known for his ideological consistency, and many of his pull factors were identity-based. Pew reported that “attitudes about immigration, Islam and racial diversity” were a bigger predictor of support for Trump than “other political values — including opinions about whether the U.S. economic system is unfair.” At the same time, both low life satisfaction rates and poor personal finances were correlated with Trump support before the election. It is too reductive to call these voters wholly racist or ignorant. It’s likely that people were directing real economic grievances into white identity politics.

Racism and economic distress often interact. For example, there’s a well-documented link between cotton prices and the number of lynchings in the South during the late 19th and early 20th century. Economist Cornelius Christian says that when cotton prices were low, “lynchings cause[d] blacks to migrate away, lowering labour supply and increasing wages for white labourers.”

Then and now, there is an imperative for a viable labor party — a party based on affinity, not identity. There should be avenues to channel economic dissatisfaction and affect real change in a healthy, non-reactionary way.

Of course, there isn’t always a clear divide between affinity and identity organizing. A member of the ACLU might support Black Lives Matter because African Americans tend to be most deprived of civil liberties. And within affinity groups, there is no reason that identity should not be acknowledged. Even with groups that are explicitly identity-based, it’s common to attract broader support the more intersectional the group is. Though on purely pragmatic grounds, if politics prize identity, the group representing a plurality already has a leg up; for a winning coalition, they don’t need any supporters but their own. An identity group that is not in the majority starts from behind. There was a stronger correlation between white resentment and Republican candidate support this election than in the last. With race as a salient political marker, Trump’s white identity politics had a numerical advantage.