The Turnout Myth, Part 2

Awhile back I wrote a widely-read post on "The Turnout Myth". In light of the Sanders' surge and the case he is making on his electability, which relies heavily on the alleged tsunami of turnout his candidacy would generate, I will revisit and extend some of these arguments.

1. Evidence from 2016: Sanders is currently the preferred candidate of young (18-29) black voters. Therefore, the theory runs, these voters would bail out if Sanders was not the candidate, resulting in black voter underperformace. Data from the States of Change project, however, indicate that black voter turnout in 2016--when Clinton was preferred over Sanders as the party's candidate--declined more among all older age groups than it did among young black voters. This indicates that the black turnout problem in 2016 was concentrated among non-young black voters, precisely where Sanders is weak. Note also that young black voters will, at best, be around 2 and a 1/2 percent of voters.

The same pattern applied to young voters as a whole: they increased their turnout more than all older age groups in 2016. They also slightly increased their margin of support for the Democratic candidate over the 2012 Obama election. What really killed the Democrats in age terms was a sharp falloff in support among voters 45-64.

2. Evidence from 2018: you want high turnout, 2018 had it. Midterm turnout skyrocketed among young, black and Hispanic voters. Just what Sanders claims he can do, right? But, as has been widely noted, Democrats in 2018--especially the successful ones--did not run on particularly radical programs but rather on opposition to Trump and unpopular GOP programs in health care and other areas. Perhaps these groups just didn't --and don't--like Trump and what the Republicans stand for?

It's also worth noting that, despite this stellar turnout performance, the overwhelming majority of the Democrats' improved performance came not from less Presidential dropoff and more midterm surge but rather from voters who voted in both elections and switched their votes from Republican in 2016 to Democratic in 2018. The Democratic big data firm Catalist-- whose data on 2018 are the best available--estimates that 89 percent of the Democrats' improved performance came from persuasion--from vote-switchers--not turnout.

3. Recent evidence: Recent polls do indicate that Sanders tends to do a bit better among the youngest voters than Biden when matched against Trump.....but he more than loses that advantage when the comparison shifts to older age groups, where he uniformly runs behind Biden. Interestingly, in one of the few recent polls of black voters, the Washington Post poll found Biden and Sanders performing identically against Trump among young (under 35) black voters, while Biden considerably outperforms Sanders among all older groups of black voters.

But perhaps Sanders will make up for this net loss by supercharging youth turnout? Maybe but see the arguments above. Also, a recent CNN poll does not indicate much difference between younger and older Democratic-leaning voters on their likelihood of not voting/going third party if their preferred candidate does not get the nomination. Perhaps voters, young and old, just really want to beat Trump.

4. General evidence: As Brendan Nyhan notes:

"When ideologically extreme candidates narrowly defeat moderates for a party nomination, the political scientists Andrew Hall and Daniel Thompson find, they perform more poorly in the general election, in part because they inspire the other party’s base more than their own."

This is consistent with general political science research. The turnout equation does not necessarily return positive results for a candidates like Sanders. The reverse is more likely. As I've said before it is magical thinking to believe only your side gets to increase turnout in a polarized situation. The other side gets to vote too--and they just might in even larger numbers.