Increasingly, I am hearing people claim that turnout was disastrously low during this election because everyone hated both candidates and that this is the only reason that Trump won. As usual, the MSM is too incompetent to assess the validity of this claim.
There are two basic ways of calculating voter turnout. On the one hand, you can talk about the proportion of people who were eligible to vote that voted. On the other, you can talk about the number of people over the age of 18 who voted. These groups are not the same.
The “voting age” population includes felons, noncitizens, and other groups who cannot legally vote. When the MSM claims that this year had abnormally low turnout they are typically talking about turnout in the voting age population.
The problem with this is that variation in this form of turnout can be due to changes in the proportion of the population which is of voting age but which cannot vote. Because of this, such data cannot be used to test claims about how candidate favorability impact the election. If you are ineligible to vote, whether or not you like a candidate doesn’t impact whether or not you vote.
Here is what happens when we look at turnout among people who can actually legally vote to begin with:
(All turnout data in this article came from the United States Elections Project.)
Across this 52 year period, an average of 57.2% of eligible voters voted for president. This is practically identical to the 57.8% of eligible voters who voted this year. When the MSM bother to look at eligible voter turnout, they often still miss this by only comparing 2016 to 2012. A lot of people voted in the 2004-2012 elections. These years were unusual. 2016 was not.
In fact, voter turnout has been declining since 2008. 2016 simply continued to follow a trend which began with 2012.
Moreover, during this 52 year period, the median change in voter turnout was -.60%. The change between 2012 and 2016 was -.80%. So not only was the absolute turnout level nothing out of the ordinary, the change in turnout relative to the last election was pretty normal too.
Still, you might think that the decrease in voter turnout helped Trump. This is not true. As it turns out, the more that voter turnout fell in a state relative to 2012 the less well Trump tended to do.
This relationship is especially striking if we compare the average margin of victory for Trump in states where turnout decreased relative to 2012 to his average margin of victory in states in which turnout increased.
As can be seen, the data we have rather strongly suggests that the fall in turnout hurt, rather than helped, Donald Trump.
As for the other part of this story, it is no doubt true that Trump’s campaign was made easier by the fact that a lot of people hate Hilary Clinton. However, the idea that this explains why he won seems implausible given that Trump’s net favorability ratings were consistently lower than Clinton’s this whole election. Explaining Trump’s victory by appealing to people generally not liking Clinton seems especially ridiculous in light of the fact that Trump lost the popular vote.
If you want to explain why Trump won you have to look at racial demographics. Trump’s net favorability was worse than Clinton’s among the total US population. However, Trump’s net favorability was better than Clinton’s among Whites and especially among non-college educated Whites. In may, Gallup found that 49% of Whites without a college degree and 41% of Whites generally had a positive view of Donald Trump. By contrast, only 26% of noncollege grad Whites and 30% of Whites in general had a favorable view of Hilary Clinton.
This is favorability data that can make sense of the election outcome. As I documented previously, the important swing states that Trump won had abnormally high numbers of Whites and Whites without college degrees.
Thus, if you want to understand why Trump won they need to answer this question: why was Trump liked and Clinton hated among Whites and especially among the White working class?