My working hypothesis is that “swing voters” are mostly a myth created by pollsters (SORRY) and the media. The idea that there is some large swath of the population who, despite our nation’s immensely partisan tendencies, are compelled to fairly regularly change the party the vote for between D and R from election to election seems too fantastical. Party identity in the US is deeply rooted in values and identity, and not fleeting fancy. Of course we know there are some genuine swing voters out there, but not – in my view – in the numbers popularly conceived.
So when I think about election outcomes I tend to focus less the concept of “swing voters” and instead on turnout. It’s all about turnout, right? For me, election outcomes are simpler than the buffeting opinions of a hypothetical and highly unfaithful group of voters, somehow large enough to impact election outcomes in a country of 300 million, but not large enough to identify and profile consistently. What really matters is who shows up.
Our working turnout models (LINK) suggest that 2016 may show lower turnout than either 2012 or 2008. This is based on simple temporal comparators, looking at what the electorate was telling us at this point in the election cycle 4 years ago, and 4 years before that, etc. While this is by no means conclusive, there is some internal logic to this notion: Obama was a dynamic, unexpected, and exciting figure for young people and minority voters (who together mainly explained the higher turnouts of 2008 and 2012). Clinton, while the first female in her position, is older and very much a ‘known entity’ in the political space, and certainly is already seen as part of The Establishment. The galvanization we saw around Obama may not transfer easily (or at all) to Clinton.
It seems instead that the most galvanizing figure of the 2016 election cycle is Donald Trump, who is exciting and activating a base of white, working class voters to the polls and manages to keep surprising ‘traditional’ Republicans and Democrats alike with his success. What is critical to note here is that this constituency of Trump supporters are already much more likely to show up and vote on Election Day than Obama’s coalition were, based on demographic voting trends alone.
An early signal that this differential is present already can be found by looking at turnout figures so far for the Primaries:
Almost unfailingly so far, with both the Democratic and Republican primaries highly contested, we see turnout has declined for the Democrats (vs. 2008) and increased for the Republicans (vs 2012). To me this suggests that Trump is activating the Republican base, and Clinton is failing to do the same with the Democrats. There are notable exceptions of course, with Michigan being the most prominent example: high turnout in both the Democrat and Republican races yielded a surprise Sanders win on the Democrat side, catching the public and pollsters unaware.
It is critical to note here that there is not an established correlation between Primary Election turnout and General Election turnout, and the plethora of candidates is also a likely driver of turnout. But this does still speak to a current high base level of enthusiasm on the part of the Republicans, presumably driven by polarizing figure Donald Trump, compared to the less engaged Democratic base. If Donald Trump can unite this enthusiastic Republican party – which still remains to be seen! – and if Democrats remain underwhelmed by Clinton’s ‘Establishment’ status, Trump could sail quite readily to victory against Clinton in November.