Thomas Edsall trolls the Democratic Party (Part 1 of 2)

Thomas Edsall is a capable reporter who has ironically done some important stories on the far right. I say "ironically," because at least since his 1992 book Chain Reaction: The Impact of Race, Rights, and Taxes on American Politics, he comes back again and again to trolling liberals for seeking the votes of people of color.

In Part 2, I look at the position Edsall's essay takes on hot-button race-related issues.

In Democrats Can't Just Give the People What They Want New York Times 10/13/2021, he does a bit more trolling. I don't know if he wrote the headline, because it doesn't exactly fit the essay. But it's a wonderful summary of establishment liberal politics!

(Actually, Democrats can give a majority of the people a lot of the most important public policies they want. But between corporate donations, neoliberal ideology, Manchinema, and the Senate filibuster, they are not delivering on most of it.)

Chain Reaction basically argued that whichever of the two major parties was perceived as the most hostile to minorities and especially African-Americans would dominate US politics indefinitely. It was kind of embarrassing for him when Bill Clinton won the Presidency in 1992. Edsall took to the pages of the New York Review of Books to try to explain what happened (What Clinton Won 12/03//1992)

The "Democrats Can't..." piece basically drags similar tired tropes into 2021. The basic idea is that Democrats are losers because too many white voters are upset that the Democrats are too popular among voters of color. There's a new twist here, which is handwringing about increased percentages of black and Latino voters won by Trump in 2020.

This is a real phenomenon and is worth analyzing carefully. But there are two problems in Edsall's current election analysis. One is a confusion of math and simple algebra. Every vote counts equally, at least within states. A Presidential candidate wins a state by winning a numerical majority of the votes. If 100 people from group X votes in Election #1 and 5% of them vote for the Republican, the Republican gets five votes and the Democrat gets 95. If 1000 people from group X votes in Election #2 and 10% of them vote for the Republican, the Republican gets 100 votes and the Democrat gets 900 votes.

Obviously, a 900-100 Democrats numerical margin is preferable for them to a 95-5 numerical margin.

It's important to watch such trends among identifiable voting segments. But the second problem is the lack of a specific consideration of the particularities of the 2020 election, the high level of interest and intense partisan polarization drove high turnout. The change in the ethnic percentages could mainly be a function of that. It requires more analysis than looking at the percentages. In the case of Latino voters, issues like Trump's clownish regime-change effort in Venezuela and his renewed drastic sanctions of Cuba may have played a role in the turnout and the percentages. In the heavily Latino area of the Rio Grande Valley, there was a real problem with Democratic party operations, where the party organization that has to turn out the vote seems to have more-or-less fallen apart.

Edsall also rolls out a current favorite media consideration, the trend of "college-educated" voters gravitating to Democrats while "non-college-educated" ones lean more Republican. The media and too many political analysts use this difference as a proxy for class, the non-college-educated assumed to be "working class" and the college-educated being middle-class to wealthy.

Political polls tend to have much more complete data on education levels than on income or type of work, so education level is an obvious proxy. But a very flawed one. But some of the biggest and most effective labor unions in the US right now are teachers, nurses, and public employees. Those jobs have a large percent of members with college degrees. (At one time, I was an SEIU union steward in a public agency in California where most of the members had college degrees even back then.) And there are lots of businesspeople without college degrees who are loyal Republicans because ”free enterprise“, or something.

In a recent analysis, Paul Heideman examined data from the General Social Survey (GSS) "which has been asking consistent questions of a representative sample of Americans for almost fifty years" and provides information on type of employment as well as partisan alignment. ("Behind the Republican Party Crack-up" Catalyst 5:2, 2021). Using his employment-based definition of working-class, which treats nurses as workers, for instance, he found that there has been a long-term reduction in working-class voters' party alignments in the context of a hollowing-out of traditional party structures in the US, but nothing that could reasonably be called a major party realignment among workers toward the Republican Party:

Far from becoming more working class, the Republican coalition has become less working class over time. Again, the overall trend is class dealignment. Where the Democratic Party was once far more class-polarized than the Republican Party, both parties have become less working class over time, such that the degree of class polarization in both parties is approaching equal. Independents, meanwhile, remain highly class-polarized, with little change over the past half-century.

However the Republican Party has changed since the 1980s, the driving force has plainly not been the rise of working-class Republicanism. The Democrats, it is true, have experienced near-catastrophic levels of working-class exit. But the Republicans have not, in the main, reaped the gains of this. Partisan polarization within the working class has diminished, with the result being that no party commands a clear majority of working-class support. Similarly, within the Republican Party, the share of the party made up of workers has actually diminished over the last few decades. [my emphasis]

But for popular consumption, the idea of the Republicans as the "working-class party" is useful for the party's populist posturing.