actually, the black vote is very stable and predictable
Over at the New Yorker, Collier Myerson has an article arguing that the black vote is not monolithic:
Many believe that Bernie Sanders will lose the Democratic primary to Hillary Clinton in part because he cannot galvanize “the black vote.” Writing for The Nation on February 28th, Joan Walsh declared, “When the history of the 2016 presidential primary is written, if Hillary Clinton is the party’s nominee, it will show that Senator Bernie Sanders’s campaign effectively ended in South Carolina.” Why? Because Clinton learned, from 2008, to treat the state “as a proxy for the black Democratic primary vote.” That year, after she lost the state badly to Barack Obama, “her campaign hemorrhaged African-American support” and never recovered. This year, Walsh posits, that is more or less Sanders’s problem.
Collier then argues with this view. He makes some sensible points. For example, race is not the only variable that strongly predicts vote. Age and income also predict voting. Also, he argues that it is a mistake to think that the black vote somehow obscures the “real Democratic vote,” as if Black voters don’t matter. I agree. That is an absurd, not to mention racist, view.
However, Joan Walsh’ original point is actually correct. As far as the Democratic party goes, the Black vote is a proxy for the party. You lose that vote and you lose the whole party. In fact, I would say that if you aren’t the establishment candidate, you’ve lost the election because Black voters didn’t approve you already. Since 1976, Black voters have voted for the insider or incumbent 6 out of 9 times when there has been a contested election, often clocking in at 80% for the insider. Every nominee has gotten the Black vote, except Dukakis, who lost the Black vote to Jesse Jackson.
So is there any difference this year? Let’s check the CBS & CNN exit polls:
- Iowa: Not enough data (3% black voters)
- New Hampshire: Not enough data (2% Black voters).
- Nevada: 75% Clinton.
- South Carolina: 86% Clinton.
Sample Super Tuesday States:
Sample post-super Tuesday states:
- Michigan: Clinton 68%.
- Illinois: Clinton 70% .
- Missouri: Clinton 67%.
- Ohio: Clinton 71%.
- Wisconsin: 69% Clinton.
- North Carolina: 80% Clinton.
Well, we have data over time, from early in the season till recently, and from every region of the country that has enough Black voters in the sample. What you find is that Sanders loses the Black vote by 35% to 60% (!) and a lot of that seems due to geography, where Southern black voters give Clinton about 15% to 20% more. I think this supports Walsh’s view that the black is large in size and crucial in the primary. I’d also add that it is very sticky and has a tipping point. Thus, if you start losing voters in this category, it is likely that you will see a flip of the whole bloc and you will likely loses.
The only mistake that Walsh makes in her analysis is that she says Sanders had the same problem as Clinton 2008. Wrong. Clinton is the only candidate in modern Democratic history to lose the Black vote after having a lead. Sanders never made a dent in the first place, which is normal for a challenger in a Democratic primary election.