What's dividing voters in the 2020 Democratic primary?
Demographics, sexism, racism and—yes—ideology
I never imagined that my feelings toward Cicis Pizza would become an issue in the 2020 primary:
The Takeaway: Demographics is density—except that there are a bunch of political attitudes that correlate with vote share but are not entirely explained by demography. In survey data, Joe Biden does better with sexists and racists than other candidates, and Elizabeth Warren does better with voters who want to abolish ICE and oppose deporting illegal immigrants. I mean, duh—this isn’t rocket science y’all. My point? Sometimes it’s just nice to see that our hypotheses about demographic characteristics and political attitudes are correct.
Editor’s note:
Thanks for reading my thoughts on this subject. And thanks for subscribing! Your membership adds up and makes all this newslettering possible (reminder: I do all this work independently). Please consider sharing online or with a friend; the more readers, the merrier. Remember that apart from getting special articles subscribers can also comment below each post and participate in exclusive threads.
As always, send me your tips about what you’d like to read about next. Or what you don’t want to read. Or your feedback otherwise. Also, cat pictures are nice, so please send them to me! I’m elliott@thecrosstab.com, or @gelliottmorris on Twitter.
Thanks all,
—Elliott
My good friend Alexander Agadjanian has some 2020 primary wisdom I’d like to share with you all. He shared his analysis of primary voting preferences in this series of tweets. Check out these figures:
The first graph shows that voters’ primary preferences are largely driven by their demographic characteristics. This is to be expected. We know that preferences in the 2016 Democratic presidential primary were largely driven by age and education. Because (a) Bernie Sanders is running again in 2020 and (b) primary voting patterns don’t vanish within four years (duh), most demographic cutting lines should persist this time around. Age continues to be a key demographic correlate this year.
Notable is that ideology and party identity are relatively weak predictors of voting behavior. Again, perhaps this makes sense when we are concerned with intra-party affairs. Nate Cohn presented this hypothesis in a recent article for The Upshot and it is good to see some regression coefficients backing it up.
The second figure presents findings that are more pertinent to our understanding of politics in 2019 and 2020, I think. Looking at the bottom half of the right-hand panel, we see first that voters that score higher on a scale measuring hostile sexism are more likely to support male candidates, especially Joe Biden, than female ones. (“Sexists are sexist”—who knew!) We see next that voters that harbor more negative views toward non-whites—what the social scientists are calling “racial resentment” these days, though this is different than racism—are more predisposed to voting for Joe Biden and those who are least “racially resentful” are more likely to vote for Elizabeth Warren and Kamala Harris. (Interestingly, resentment is more negatively correlated with voting for Warren than Harris, a black candidate. This might suggest that there’s something to in-group divisions among whites that is driving the variable’s importance.)
Now combine the two figures. Joe Biden is winning disproportionately both among black Americans and “racially resentful” ones. What does this mean for the general election? As I said in a tweet yesterday, the former vice president is currently poised to repeat Obama’s success in bringing together African American and racially resentful white voters under the same Big Tent. This was Obama’s (successful) strategy in 2008 and 2012—though it should be noted that it’s not guaranteed to be successful in 2019 when many racially resentful whites have fled the Democratic Party and likely will be barred by their partisanship for voting for someone other than Donald Trump. That being said, there is still a non-zero number of racist white swing voters that Biden could appeal to that that Warren might not.
(I also think that there are differences in which Democrats are appealing to whites with high racial resentment vs whites with high white identity. Elizabeth Warren seems to realize that playing on the latter might be less controversial and more beneficial to the party. More on this… sometime in the future???)
I should also note that in The Economist’s latest polling with YouGov, Biden is beating Warren on the question of who can beat Trump in November, but the gap is not as large as you would infer based on Warren’s (supposed) weakness in recreating the Obama coalition:
…I don’t have any concluding remarks to this piece. I mostly just thought the data were interesting. Perhaps I’ll say “Um, Actually, Demography Is Destiny”—except when it’s not.