Social Fragmentation And the Trump Win
Totally unregulated alt media is the key threat to federal governance
To understand what happened this week empirically, let’s examine the math behind this week’s election. I will apply the dual lenses of a corporate market researcher AND anthropologist.
Here we go!
Around 65% of eligible adult voters participated, although this percentage appears to have dropped slightly since the 2020 federal election.3 By how much? By 2% of the VEP (i.e., the 265.8M adults). This may seem negligible to readers, but I’ll show why it’s not too tiny to matter in the current cultural era.
OK. So, what happened in the Presidential vote?
Only 84% of ballots had a presidential choice4 (some folks only care about local problems)
Only 28% of the eligible U.S. population voted for Trump to be in power5
In the popular vote, Trump won by only 3% of those who marked a Presidential choice. 8 million individuals provided Trump with the nationwide population margin to win.
In reality, the Presidential voting margin in only 6-8 states really mattered this week.
This is NOT a mandate in any statistical sense of a majority, and it does not qualify as a social science generalization (i.e., “America” voted for Trump. It did not.)
While voter participation has been up in recent Presidential elections, it was much higher in the 19th century, when 90% of the population was white and ‘man-only.’6 We still have a voter participation problem operating at scale.
~8 million voters cast critical votes for Trump allowing him to win the Electoral College
But ~5 million voters did not show up this week vs. 2020 (!)
Had these apathetic voters shown up, Trump’s national margin could have shrunk a lot (or perhaps grown)
Small swings in apathy and emotional pull can and do routinely decide Presidential races in the country today.
We deal with victory margins amenable to large-scale consumer marketing playbooks developed by leading consumer brands (e.g., Tide, Oreo, Bounty).
These marketing campaigns use research to isolate the most potent emotional argument to attract the largest brand-aware consumers to repurchase. Leading brands routinely develop campaigns to win back “lapsed” consumers for a tiny bump in sales volume. These campaigns may be executed in just 1-2 media channels and invisible to the general public. An effective Superbowl ad sometimes fits into this category, moving a tiny group of Americans to do something soon.
This commercial marketing work is similar to Russian interference on social media platforms or right-wing podcasts in terms of appealing to tiny population segments.
These population segments matter to flat billion-dollar consumer brands trying to grow by capturing 50-100 basis points of market share from a competitor. These kinds of segments also matter to today’s Presidential candidates. It is easier than ever to design ten campaigns aimed to inflame emotion in 10 unrelated groups where there is a variable you can manipulate to cause fear.
However, most pundits and political scientists are examining the usual demographic variables to isolate who contributed to Trump’s eight million-dollar margin.
Here’s just a taste of the usual demographic thinking:
9% swing from Catholics to Trump vs. 2020 7
Rural counties in Georgia swung hard to Trump8
Ethnically diverse (i.e., black and immigrant-heavy) counties swung hard in Michigan9
Counties with a majority black population swung to Trump10
Hispanics swing to Trump big in Arizona, and the SW11
Why would some immigrants, Catholics, Arizona Latinos, and African Americans in black-majority counties shift toward Trump’s populism? These folks appear to have nothing in common whatsoever.
The problem with this line of thinking is that demographics are easy to measure. Respondents self-identify readily with these big buckets because we are trained to do it in adulthood and by the US Census forms we fill out every 10 years. However, tiny sub-segments of large demographics can not be explained by generalizations of that demographic itself. This is flawed social science thinking.
All of the tiny voter ‘swing’ groups listed above could have something else in common unrelated to their position in these large demographic groups. Or there could be an intersecting variable (an internal continuum of disagreement) linked to something these groups have in common. We have missing variables we can not account for that most likely explain why sub-segments of blacks, Hispanics, and immigrants swung to Trump.
So, we can create demographic variable descriptions of WHO swung, but this does not explain WHY they did. This is a common fallacy repeatedly encouraged by political scientists, even authors with big audiences on this platform. It is also common among the average, poorly trained consumer insights executive at large consumer brands. It encourages sloppy stereotyping of Hispanics, Blacks, and immigrants, among other things.
The analytical problem we face is much more profound. It lies in our lifestyle-fractured society. I’ve recently written a book about how America’s actual glue as a nation is the desire to maximize personal autonomy (i.e., something only tiny religious communities -Mormons, the Amish, Mennonites, and cults—heavily disparage).
This has led to an explosion of lifestyle diversity, dividing local communities, neighborhoods, and even households.
Parallel to this macro-trend, in the past 100 years, we have become more socially isolated and less involved in community activities. We simply socialize in person far less than we did decades ago.
Our leisure time is primarily spent in individually curated behaviors (recreation and media consumption), not collectively organized events.
Media is a primary disintegrating agent here. Television is so cheap at 75 pennies per hour that it has overtaken our leisure time to an incredibly problematic degree. The spread of internet-based smartphones then accelerated the effective reach of very niche alternative media. Podcasts have become a potent force in the past decade, and many of them deal with alternative political ideologies and theories. Many are spewing total nonsense from the perspective of a well-educated American. Substack is filled with nonsense publications spreading non-science, bullsh*t, and conspiratorial junk thinking. Some routinely hit the Top of the Science Leaderboard.
Substack’s owners exemplify the apotheosis of America’s belief in radical personal autonomy, fueling the spread of niche beliefs and whipping up emotions that activate small, influential population segments.
A campaign no longer needs CNN or Fox News to swing 8 million voters. Increasingly, that’s NOT where the undecided or swing voter will be found and manipulated.
It’s right here on Substack or within Joe Rogan’s listening audience.
Without shared information sources, America steadily disintegrates further.
My best guess is that the demographic swing groups listed above each received messages designed to inflame extreme segments of each group on just one issue.
Pro-Life Extremist Catholics - Since a minority of Catholics are very pro-life, then use ‘candidate’ Randall Terry’s vile ads featuring pictures of fetal and infant corpses to shock them into action the next day.
Resentful Middle-Class Immigrants - We know there is a minority of legal immigrants who are very annoyed by illegal immigrants gaining quick access to jobs when “I struggled hard to do it legally.” This is often a condescending class argument within these groups when you poke around. So, the intersection of middle-class resentment and immigrant identity becomes critical.
When a bad faith actor behaves like a consumer marketer, the goal is to deploy emotion-laden stories to motivate low-think ballot-filling. We know that fear and humor are the primary drivers in consumer marketing, though consumer brands skew towards humor because their objective is to maximize reach in the target segment. Political ads tend to stoke fear as the critical emotional variable because fear aligns with imaginings of external threats to one’s niche lifestyle or beliefs. And fear motivates a smaller group to act immediately. Humor is more memorable than motivating.
Also, voting to protect yourself is vastly more compelling than voting to reform society (an abstraction).
The former has an immediate, albeit largely imagined, payoff. Not the latter.
America’s current political system is polarized, yes, but it is the fear-based manipulation of tiny, extreme population segments that will continue to determine many races at many levels.
This stoking of fear runs in parallel to the high baseline anxiety many of us feel due to the extreme lifestyle fragmentation all around us. If you don’t know what to say to your neighbor because you know they are gay or voluntarily childless or atheist, that mundane awkwardness exposes many of us to fear-based manipulation from bad faith media sources. When the world is genuinely confusing us, it takes little to push some into high anxiety right before election day.
This is why I quietly support the unthinkable - tighter control of content on media platforms, including this one. This could reduce the volume of fear-based misinformation, bad-faith content, and trust-dissolving content fed to tiny adult groups with only one purpose - to direct them to vote for one person (without telling them directly to do this). It could also focus on restricting this content only in the last month before an election to neuter its power.
If we want a less volatile, functional federal government (and I recognize that many Americans do not like this), we must control the last-minute fear-stoking sent to tiny, vulnerable, and extreme audiences via the Internet. We must reign in the unhinged long tail of alt media, where the problem lies.
I do not expect this to happen. Ever.
The cat and genie are both out…
I’m using the % reported by the University of Florida’s Election Lab in a recent Foreign Policy online article and the US Census 18+ population, or the Voter Eligible Population. The latter is the standard for tracking voter participation. https://foreignpolicy.com/projects/2024-us-president-election-live-updates-harris-trump/?article_anchor=us-presidential-election-2024-voter-turnout-comparison#cookie_message_anchor
I used the latest American Community Survey data from the US Census. The monthly Current Population Survey would be slightly more accurate, but not by much.
I’m using the same Voter Eligible Population standard as Professor McDonald does in this chart. However, this is not a definitive result. https://www.electproject.org/national-1789-present
~147M marked a presidential choice according to the latest tally on Google.- https://bit.ly/40EoapC
72.7M Americans voted for Donald Trump, according to the latest online tally housed by Lord Google.
https://www.electproject.org/national-1789-present
According to exit polling collected by Fox News on election night, Catholics across the country swung 9 percentage points in Trump’s favor. 52 M Catholics - * 10% = 5.2M voting adults. It’s not abortion, though, since only 16-22% of Catholics favor pro-life legislation.
https://www.theguardian.com/us-news/ng-interactive/2024/nov/06/the-key-swings-that-handed-trump-the-white-house-a-visual-analysis
Ibid
Ibid
Ibid
Really interesting statistics and perspective! Thank-you.
I also wonder how many angels can dance on the head of a pin.