Trade was an important issue in the recent presidential election, but not in the way the media and many prominent observers have led us to believe. The dominant narrative in the media was that disgruntled manufacturing workers whose jobs had been sent overseas emerged, understandably, as trade’s strong opponents, thus making Trump with his strong anti-trade rhetoric their natural ally.
But the data tell another story. Using panel data collected by GfK Ltd. for the Institute for the Study of Citizens and Politics at the University of Pennsylvania, we interviewed over 1,200 citizens in October 2012, shortly before the Obama-Romney presidential election, and then re-interviewed the same 1,200 people just before the most recent presidential election. What changed between 2012 and 2016 was not the extent to which Americans felt they had been hurt by trade. As shown in Figure 1, in response to a question asking these people whether they had been helped or hurt by trade, there was no difference on average between 2012 and 2016. The average response was for people to say their family had been unaffected one way or the other. It’s possible that the mean could remain unchanged while more people were hurt as well as helped, but that did not appear to be the case either.
Instead, two changes combined to make trade an important issue in the 2016 presidential election. First, as shown in Figure 2, when it comes to people’s personal opinions about trade, they became far more opposed to trade. Figure 2 shows the percentage opposed to trade in response to three different survey items. The percentage of those opposing more trade agreements increased by 10 percentage points during this four year period. Likewise, the percent opposing “more trade agreements like NAFTA” increased by around 15 percentage points, probably because NAFTA was the primary agreement named by both candidates as regrettable (in Clinton’s case) or as “a disaster” in Trump’s more extreme parlance. The percentage believing that increasing trade had hurt the US economy also increased significantly, though by a smaller amount. Clearly trade has taken a beating in public opinion, perhaps because neither presidential candidate in 2016 was willing to champion it.
However, this does not mean there were no differences in people’s perceptions of the candidates. As shown in Figure 3, in 2012, the perceived stance of the Republican and Democratic candidates on this issue was virtually identical. The average opinion in the mass public was also right at the neutral midpoint. By 2016, not only had public attitudes toward trade changed among the mass public, their perceptions of the Republican and Democratic candidates had changed a great deal as well. As shown at the bottom of Figure 3, by 2016 Trump had repositioned the Republicans far to the anti-trade side of Democrats, with the average citizen in between the two, but notably closer to the Republican than the Democratic candidate. In short, the average citizen was closer to the Trump’s views than to Clinton’s.
In Figure 4, a closely related series of questions on attitudes toward China shows a similar pattern. The public’s views of whether China is primarily a threat or an opportunity did not change from 2012 to 2016; it was already slightly to the threat side of this 7-point scale. The perceived positions of the Republican and Democratic candidates were on the “opportunity” side of the spectrum, for both parties. But again, the perceived positions of the Republican and Democratic candidates changed a great deal by 2016. Trump’s position became one in which China was perceived as even more threatening than it was perceived to be by the public. As with general trade attitudes, the Democratic candidate’s position became seen as slightly more positive toward China than in 2012. Whereas in 2012, the Democratic candidate was perceived to be closer to the average American, by 2016 the Republican candidate was clearly much closer to the American people on China.
These findings suggest that the dominant media narrative about who turned against trade and why is largely incorrect. Further, they suggest that while the public did indeed, become more anti-trade, particularly so when the question was framed in the same terms that the candidates themselves were using – namely badmouthing NAFTA. But the far larger shifts were in the major candidates’ perceived positions on these issues. By emphasizing a strong anti-China and anti-trade position, Trump effectively moved his party closer to where many Americans stood on these issues. Whether he himself or a collection of other forces effectively led public opinion in this direction over the last four years remains an open question. But it was not people’s personal experiences that led to this position. Instead, their view was that it was bad for the nation as a whole, but not necessarily them personally.
The public’s usual cues in this policy arena are now confusing, to say the least. The party that formerly championed free markets is now more protectionist, and the party traditionally more opposed to trade is perceived as more favorable toward it. This much change in public perceptions of where the party stands from one election to the next is unusual, and it suggests a public that lacks clear partisan cues on these issues. Whether this change also suggests a highly malleable public remains to be seen.