What those county "realignment" charts actually show
No, they don't prove that Trump was the working class candidate.
|Dec 13, 2020|
Joe Biden won poor voters. He even won voters in the middle income bracket. Everything even remotely resembling sound evidence confirms this. If you’re experiencing déjà vu here, its’s because I explained this just a month ago — in a post where I dared pundits who insist that Trump won the working class to put their money where their mouth is. None have been willing to do so — but right-wing figures from pundits to Senators are still lying about it!
As both of these links show, disinformation about this question often relies on data about red versus blue counties. Perhaps the most common version of this move comes from the arch-centrist Brookings Institution. For the past two elections, Brookings has been publishing the same infographic:
Leaning on county-level charts like this, Republicans from the hard right to the Post Left make the same rhetorical move: since the GOP won counties with a smaller cumulative GPD, they are therefore the party of the working class.
To illustrate the problem with this kind of analysis, consider a national population that looks like this:
Here, each figure represents a worker — blue for Democratic voters, red for Republican voters, and grey for non-voters. Already, it’s clear that a plurality of workers have no real political alignment, and that of those who do, the majority are Democrats. Now, imagine that we divide this national population into two counties:
So what does this county chart tell us?
It’s entirely possible that Red County has a smaller GDP than Blue County, perhaps because the former is a rural or post-industrial hinterland while the latter is a developed urban center — but this implies nothing about the workers who populate these areas! You can have the same number of people doing productive work in both counties, or even more people doing productive work in Blue County, with a lower total GDP coming from Red County. You cannot deduce any information about where workers are and who is doing what kind of work simply by looking at GDP totals.
Describing political alignments at the county level actually obscures the alignment of workers because it oversimplifies divisions within each county. Only 30% of the population lives in so-called “landslide” counties: counties that overwhelmingly vote red or blue. Everyone else lives in counties where voters are much more divided. This illustration reflects that fact: though both counties swing one way or the other, each contains a significant number of workers who vote against the local plurality.
Describing political alignments at the county level also obscures the significant number of workers who don’t vote at all. And this point is particularly crucial to pay attention to when we are talking about poor voters, who turn out less than almost any other demographic.
Ultimately, these county-level analyses obscure a much more complicated situation. Republicans win more votes from workers who live in low-GDP counties, but they do not win all of them, or even a majority of them. And meanwhile, Democrats win even more votes from workers who live in high-GDP counties — though again, not all of them or even a majority of them. At the same time, a huge number of workers in both counties — often the plurality of them — don’t vote at all.
It may very well be the case that Republicans will eventually become the party of the working class, but that prediction demands a much more rigorous and sophisticated analysis than we’re getting from these silly county infographics. If one cares about where the working class is right now, the data remains clear: not voting, or voting for Democrats, with only a minority backing the GOP.