by Stephen Fisher.

This question can be addressed from various angles and at different levels. I will start with the basic features of the result, then discuss the patterns of change since 2012, all before trying to address the bewilderment of many who find it hard to understand how anyone could vote for him.

Trump won the presidency by holding on to states that Romney won and winning some states that Obama previously won, including Iowa, Florida, Ohio, Pennsylvania and Wisconsin. He did this by a combination of a national vote swing of around 1.5 points since 2012, plus above average swings in some key battleground states, such as Pennsylvania 3 points, Michigan 5 points and Iowa 10 points.*

At the same time, Trump lost the national popular vote but won the presidency by narrowly winning Michigan, Wisconsin, Pennsylvania and Florida with leads of less that 1.5 points each. So, had the swing to Trump in these states been smaller by just 1 point then Hilary Clinton would have won outright.  This was a close election.

Why were there bigger than average swings in the key battleground states? The geography of the result shows that Trump made large gains in rural counties, especially in the industrial Midwest where whites without a college education are the majority (see also here). The battleground states have a disproportionate number of such people and so, apart from Florida, all the states mentioned above were affected by this trend. This pattern is the main reason that the dominant narrative at the moment is that Trump won the election by winning over and mobilizing the white working class.

One problem with making the narrative about class, and so about economic issues, is that there was not much a pattern of voting by income. The exit polls results below show that the rich were more likely to vote for Trump than the poor so why blame the working class? The white working class may have swung more heavily towards Trump, but they were far from being the most Trump supporting group.

Also, the white working class narrative seems to be based on the widening education divide, and that might be as much or more about social values as about economics. Higher education is typically associated with having a liberal, internationalist and pro-immigration outlook and this election was as much if not more about those kinds of issues than economic ones.

A further related problem with a narrative about class and economics is explaining why it applies to whites only when the economic difficulties faced by non-whites are much greater. The figures above show that the education gap does not extend to minorities who are overwhelmingly Democrat regardless of education.

In truth, exit poll figures for minorities are contested. Polling estimates for the Latino vote suggest a big movement to Clinton relative to 2012, but the exit poll found the opposite due to issues of sample design. But these are arguments over just how overwhelmingly Democrat the Black and Latino votes were. Even if it turns out that the Black vote was a little less strong for Clinton than it was for Obama it will remain the case that Trump’s victory was overwhelmingly the product of decisions by white voters. The result has been called a “whitelash against a black president”, but all Republican victories are delivered by whites and the swing this time was much smaller than the historical average after a two-term presidency. This does not seem to have been an especially anti-Obama protest even though white-ethnocentrism probably played a part.

This election outcome represents incremental change on a relatively stable pattern of voting in US elections. Most Trump voters were Romney voters in 2012 and most Clinton voters formally supported Obama (see here). This feature and the relatively modest scale of overall change since 2012 means that all narratives based on fixating on the features of this election in isolation are problematic. This includes the idea that this was all about globalisation and even that it was mostly about Trump versus Clinton.

The coverage of this election in the media has been so dominated by discussion of the candidates and there have been so many claims about how completely different this election is from previous ones that it is hard to think about the election as not being all about Trump versus Clinton. But to understand the outcome we have to think about the extent to which the election was just another US presidential election between a Republican and Democrat. Viewed this way it is not hard to understand why the Republican candidate was elected. The Democrats have had control of the White House for two terms and it is rare for a party to win three in a row. The economy has only been growing slowly and real wage growth has been poor. These factors, while they did not make a Republican victory anywhere near inevitable, did make this a relatively hard contest for the Democrats to win. The fact that Clinton looks likely to have won the popular vote means that Trump underperformed compared to what you might expect a generic Republican to have done. See here for an estimate of the so-called Trump tax.

But for many, and perhaps most observers outside the US, this kind of explanation goes nowhere near to addressing their sense of bewilderment with the result. Many feel that Trump was so outrageous and awful that Republican voters should not simply have carried on voting Republican as usual. For some the idea of voting Trump is such an anathema they find it hard to understand how anyone could to it. Why didn’t his candidacy produce a fundamental change from past patterns of voting behaviour? There may be nothing anyone can say that constitutes a satisfactory explanation, but here’s a start.

What is objectionable to many about Trump is a mixture of his pronouncements and policy positions on the one hand, and his personal qualities, especially his moral failings on the other. I shall start with the former.

Trump’s policies appear to have been reasonably popular and seemed to motivate voters. According to the exit poll 41% support “building a wall along the entire US border with Mexico” and the supporters of this policy overwhelmingly (86%) voted for Trump. Although a minority, as many as 25% of voters though that illegal immigrants working in the US should be deported, and they overwhelmingly voted for Trump. Only slightly more voters, 42% to 38%, said that the effect of trade with other countries takes away jobs than said that trade creates more jobs. Attitudes to this issue are pretty strongly related to vote choice, but not so much as for immigration. Anti-immigrant and anti-globalization views, often linked as parochial nationalist, are very widespread and strongly linked to voting Republican and seem to account for a big part of Trump’s success.

Trump also took a clear pro-life stance which appeals especially to the Christian right. Of the 26% of voters who are white and describe themselves as evangelical or born again, some 81% voted for Trump.

Although his detailed policy offerings were sometimes vague and fluctuating he pretty consistently sent signals in line with the traditional values of the right, including on guns, climate change, and tax cutting (e.g. here).

The appeal of Trump’s populist anti-Washington elite message is not hard to understand. It is one that has been tried before; aspects of Obama’s appeal were also for political reform. What people on the left find particularly puzzling is the popularity of socially conservative, authoritarian and anti-immigrant policies. Jonathan Haidt writes excellently about how liberals and social conservatives find it hard to understand each other. In the context of this election his discussion of the appeal parochial nationalism as a negative reaction to globalism is also useful even if he overstates the importance of this issue.

But even acknowledging that many and or even most voters in the US are socially conservative, hostile to illegal immigration, financially stagnant or in difficulty, and fed up with what they see as corrupt government by rich globalist liberals for rich globalist liberals, there is still a particular puzzle as to why they voted for someone who infringed their traditional values, most notably with his behaviour towards women (e.g. here). How could so many people vote for someone of such dubious moral standing and so manifestly unfit to be president?

The explanation for this seems to be a mixture of denial, perceptual bias and deliberately excusing him for political gain. There is plenty of research showing that people have perceptual biases that mean they see the world differently depending on which political side they are on. Unless the economy is obviously terrible, supporters of the government will tend to say it is doing well or ok while those against the government say it is doing badly. That kind of partisan bias was clearly going on still at this election and it extended to serious matters of conduct. Those inclined to vote for Clinton then typically thought that she did nothing illegal in setting up a personal email address and server for her work as Secretary of State, but Trump supporters thought she was a criminal.

But Trump violated far more basic moral norms than those Clinton has been accused of. The video of him boasting about violating women without their consent makes this hard to deny. Nonetheless, the New York Times/CBS polls found 41% saying that “the allegations that Donald Trump made unwanted sexual advances against women” were “mostly not true.” This seems to be a striking level of denial, but there are also important questions sabout just how serious a problem people think such behaviour is. Some 69%, said that the allegations did not affect their chances of voting for Trump.

In addition to denial there was also making excuses. There is a long tradition in Christianity of hating the sin and not the sinner. Prominent figures on the Christian right acknowledged Trump was not the perfect candidate but argued that no one is without sin and compared Trump as a “flawed leader” with King David and Moses (see here and here). How much such views resonated with Christian voters I’m not sure, but this does seem to indicate how personal failings can be overlooked in favour of political motivations even by those you might think ought to be most critical of those failings.  Not only was the white evangelical vote for Trump phenomenally high at 81%, the exit poll suggests it was fully 8 points higher than it was for Romney in 2012.

If deeply religious and socially conservative people were prepared to simply turn a blind eye to Trump’s moral failings, it is much less surprising that they and many others were unconcerned about issues of his competence and qualifications. The exit poll suggests as many as 27% of Trump voters were willing to admit that they thought that Trump did not have the temperament to serve effectively as president. A similar number of Trump voters did not think he was honest and trustworthy. As one life-long Republican put it, “Yeah, Trump, he’s so awful I don’t know how I’m going to live with myself when I vote for him.”

Hopefully future research will be able to shed more light on how many Trump voters held their noses, how many were partly in denial, and how many were totally happy with who he was and what he represented. Also much more work is needed on the relative roles and contributions of the various factors above and others not mentioned, not least sexism, gender, populism and charisma. There are also very important questions about why the voters had to choose between Clinton and Trump in the first place.

To those who feel that voting for Trump was reprehensible, I should say that my attempt to explain the choice is not an attempt to justify it. His voters are responsible for his election, and so somewhat also for the consequences of his presidency. Elections can have profound effects, which makes voting a serious responsibility for citizens. This also means it is worthwhile trying to understand voting behaviour.

Notes:

*Here I use the term swing in the British sense, i.e. to mean the average of the changes in the Republican and Democrat vote shares so that a 1 point swing is equivalent to 1% of the voters switching from one to the other and widening/narrowing the lead by 2 points. Figures are approximate at the time of writing and taken from the New York Times projection.)