Last year’s midterm election results were hardly unusual for a party holding the presidency. Similar electoral setbacks had occurred during the presidencies of Bill Clinton and Barack Obama. But this one was portrayed as if it were somehow unique — an explicit rejection of President Trump’s nationalist and anti-immigration policies.

For some, the electoral losses in Orange County, California were particularly galling. “You want to see the future? Look no further than the demographic death spiral in the place once considered a cornerstone of the party,” wrote one GOP strategist.

In a state that had once launched the careers of Ronald Reagan and Richard Nixon, Republicans had fought the rising tide of demographic change and were crushed, they said. Now the GOP was repeating the same mistake on the national stage.

Such arguments are not new. They have long been a staple of establishment Republicans who support the corporate open borders agenda. They also represent a fundamental misunderstanding what is happening in the American Southwest.

California, New Mexico, and the region’s other states are not trending left solely (or even primarily) because of Republican intransigence on immigration. They are trending left because of larger socioeconomic trends and migratory patterns that may lead to America’s eventual dissolution.

The Southwest Paradox

For any close observer of race relations, the politics of California and the Southwest must be puzzling. Extensive research on the 2016 election found close links between White attitudes toward race and immigration and support for Donald Trump. Other research has found a similar link between these attitudes and greater awareness of demographic change, with close physical proximity to Latinos playing an important contributing role.

Given the breadth of this evidence, recent general election results in America’s Southwest seem incomprehensible. These states — defined for our purposes as including California, Nevada, Arizona, New Mexico, Colorado, and Texas — all have large and growing Latino populations, but their White populations have responded not by shifting right, but to the left.

Some observers, such as Ron Unz of The Unz Review, have noted this unexpected trend in race relations and concluded that those who argue that increased diversity will eventually tear the country apart are simply wrong. According to Unz, the Dissident Right has erred by treating White-Latino relations as if they are the same as White-Black relations. There is ample evidence that proximity to Blacks has produced a significant backlash among Whites in places like the old South, but there appears to be less evidence of a similar backlash to Latinos. Unz attributes this difference, at least in part, to lower Latino crime rates and greater mutual understanding once Whites get to know their Latino neighbors better.

“With such a large fraction of our immigrant population living in states displaying such negligible levels of nativist rancor,” he wrote, “the likelihood that today’s immigration controversy at the national level will produce any long-lasting negative consequences seems very low to me.”

Is Unz right? Will America’s Latino population simply follow in the footsteps of previous generations of immigrants by assimilating and contributing to America’s culture and growth? Are the Dissident Right’s fears irrational and unfounded, as the left and corporate elite keep assuring us?

The answer is no. The extensive research on this subject is not wrong. The Southwest Paradox is merely an artifact of larger socioeconomic forces.

Solving the Paradox

To understand why, first consider a related paradox. If one were to closely examine White voting patterns across the United States, it would be natural to assume — consistent with the experimental research — that Whites who live in highly diverse neighborhoods would be more likely to react negatively and become more conservative. But this is not true. Whites who live in diverse neighborhoods are not more conservative than other Whites, they are usually more liberal. The primary reason for this is uncomplicated: White flight.

The research on White flight has shown a common recurring pattern. When minorities first move into a White neighborhood, the reaction among Whites is only mildly negative at first, but after diversity rises above a certain tipping point — believed to be around 25 percent for Latinos — White flight begins in earnest. In general, the Whites who move first are the most ethnocentric and/or most likely to be adversely affected (often families with children). Their departure causes the neighborhood to become less White, which in turn causes more Whites to leave (and others to avoid moving in). This process produces a cascading effect that usually transforms the neighborhood within a few years.

After this process has played out, such neighborhoods will often retain a small White population, but it is usually one that is more tolerant of diversity or more able to protect itself through higher housing prices, gated communities, and private schools. The pattern is similar for Whites in gentrifying urban neighborhoods. In each case, the demographic profile of such Whites is fairly consistent — they tend to be disproportionately liberal, single, and childless. Depending on the neighborhood, they often have higher incomes and are more likely to have a college degree. These are the Whites who are responsible for the seemingly paradoxical result of Whites living in more diverse neighborhoods being more liberal.

The political effects of White flight and gentrification are reasonably well understood, but it is becoming increasingly clear that interstate migration is playing a similar role. This phenomenon was first noted in the popular press by Bill Bishop, author of an influential book on the subject called The Big Sort, which attributed much the nation’s growing political divide to differences in where we choose to live. Although Bishop’s methodology was criticized, his conclusions were substantially confirmed by other academic research. The only real disagreements were not over whether it was happening, but why.

Some, like Richard Florida, have focused on the migratory patterns of college-educated Whites — specifically what he calls the “creative class” — who are disproportionately moving to a select number of cosmopolitan regions and states for economic reasons. Others have cited the departure of more conservative working-class Whites from these same areas, often because of rising costs of living. Still others have highlighted more explicitly political reasons or other lifestyle choices that produce the same net effect.

Whatever the reasons (probably a combination of the above), the resulting demographics look a lot like those produced by White flight. Just like the Whites who live in more diverse neighborhoods, the Whites who live in more cosmopolitan cities and states tend to be more liberal, better educated, less religious, and disproportionately unmarried and childless. Nearly every state in America’s Southwest exhibits these same traits.

These demographic changes have helped nudge southwestern states to the left, but the trend has also been reinforced by another recent political development. The “Great Awokening,” a sharp left turn in the racial attitudes of college-educated White liberals over the past few years, has further accelerated the leftward drift of Whites living in the nation’s more cosmopolitan regions.

Given this increase in White wokeness, a final contributor is noteworthy for its implied hypocrisy. Despite the Southwest’s purported reputation for benign White-Latino relations, these states rank among the most segregated in the country. Racial segregation is growing not just in more conservative places like suburban Dallas, but also liberal cities like Los Angeles and the San Francisco Bay Area. According to a study by UCLA’s Civil Rights Project, the two states where Latinos are least likely to attend a majority White public school are liberal New Mexico and California respectively.

The Impact of Latinos on White Voting

Taken together, these trends suggest that the liberalism of America’s Southwest is not due to more amicable relations between its White and Latino populations. Instead, they are the accidental byproduct of larger social factors that have offset and concealed the negative effects.

To confirm this hypothesis, we turn to a large, publicly available survey data set housed at Harvard called the Cooperative Congressional Election Study (CCES). This survey was administered to over 60,000 individuals in 2016, over half of whom voted and were White. Each survey respondent was also geo-coded, which allows the incorporation of state, county, and local (zip code) variables such as local levels of diversity and other demographics from the Census Bureau.

The full multivariate regression results can be found at the bottom of this article, but the top-line results are straightforward. In general, after controlling for a wide variety of other factors such as gender, marriage, religion, and education, the analysis finds that Whites who live in more diverse states were more likely to vote for Donald Trump, with proximity to Latinos having roughly half the impact of proximity to Blacks. (The effects of living close to Asians and Native Americans were statistically insignificant).

These effects are not uniform, however.  As suggested by similar studies, Whites who live in heavily diverse zip codes (Black or Latino) tend to be more liberal and were thus more likely to vote against Trump. By contrast, Whites who lived outside of heavily Latino neighborhoods, either elsewhere in the same county or the same state, were more conservative and more likely to vote for Trump. These results demonstrate the variable effects of White flight.

Altogether, the combined effects — state, county, and zip code — shifted the White vote toward Trump by about one percent for every 6 percentage points of Latinos in a state’s population. In California, for example, where Latinos comprised 38% of the population in 2016, the model estimates that White Californians shifted right by about 6 percent from where they otherwise would have been based on their education and other demographic factors.

Importantly, however, these are average effects. A more detailed state-level analysis shows that in the Whitest states there were no county or state-level effects. The impact was strictly local, with growing local Latino populations causing Whites to become more conservative, a common pre-White flight result.

At the other end of the demographic spectrum in heavily Latino states, state level pro-Trump effects do not appear until a state’s Latino population approaches 20 percent. They peak at 30 percent (Arizona), and begin to decline after that (Texas, California, and New Mexico). This suggests that states with the largest Latino populations are starting to experience the same liberalizing White flight effects that are found in the nation’s most diverse zip codes.

The analysis also sheds light on why the politics of America’s Southwest are so different from the similarly diverse South. Whites in the South are conservative in part because of the presence of large Black populations, but they are also more conservative because their White populations are much more religiously conservative.

The migration patterns that helped make California and the Southwest more liberal are also having the reverse effect in the South. States like Alabama and Mississippi draw relatively few college-educated White liberals from elsewhere in the country. Unsurprisingly, the few exceptions to this rule (places like Atlanta or North Carolina’s Research Triangle) have politics that more closely resemble the Southwest.

Two Americas

If our analysis stopped here, the conclusions would be only mildly interesting. Yes, the seemingly benign White-Latino race relations in the Southwest are largely illusory, the incidental byproduct of larger interstate migration patterns, but so what? The politics of these states are still trending left, no matter what the cause. How does this change the conclusion that Trumpism is a losing political proposition in the long run?

The answer can be found by stepping back from a narrow examination of trends in the Southwest and instead looking at the nation as a whole. The Whites who flee or avoid moving to these states have not disappeared into the ether. They have simply chosen to live elsewhere and, in the process, made the rest of the nation more conservative.

The following map shows White voting trends from 2000 to 2016, two comparable election years when the GOP won the presidency but narrowly lost the popular vote. As expected, the map shows Whites in the Southwest and on the West Coast trending left over this period. But it also shows Whites in much of the rest of the country shifting to the right. This rightward trend includes the midwestern states that helped elect Donald Trump. It also shows a substantial rightward shift even in the liberal Northeast, where Whites in states like New Jersey, New York, and much of New England have also been moving sharply to the right.

This is a story that one almost never hears from the mainstream media. There are countless articles concern trolling the GOP for its losses in more diverse states like California, but there is almost nothing written about the rightward drift of the rest of White America.

These trends are two sides of the same coin and they point to a very different conclusion. This is not the story of Republicans or the Dissident Right waging a losing demographic battle. It is the story of a nation that is slowly, but inexorably, becoming more divided along racial and geographic lines. (See The Racial Realignment of American Politics).

To anyone even vaguely familiar with the larger literature on ethnic conflict, this pattern is completely predictable. The fact that there is not even a hint of the dangers in the mainstream media despite obvious lessons from conflicts in Yugoslavia, Rwanda, and Darfur only demonstrates the establishment’s control over the media narrative. The warning lights are flashing red all around us — and we are flying directly into the coming storm.

Patrick McDermott is a political analyst in Washington, DC.