Issues didn’t drive the 2016 election
After 500 days on the campaign and an emotionally-dulled seven weeks of the aftermath, I’m finally ready to start thinking about politics again. I’m less interested in “what happened”, which has both been thoroughly-explored and poorly-understood. Instead I’m going to talk about an oddly under-explored question, which is “what exactly is a political opinion”? We have a tremendous amount of issue-based polling on the 2016 election on questions like trade and immigration, and deciding how to think about it demands answering this question. To first illustrate this, an example:
There was a fun finding during the early days of Russia’s invasion of Ukraine in 2014: Americans favoring a more aggressive US response were least likely to be able to locate Ukraine on a map. Leaving the substance aside, this shows that you can’t think of all of these opinions as really being the same. If you’re a scholar of the USSR, your opinion to the question “what should America do in Ukraine?” likely reflects your pre-existing beliefs on the US-Russia relationship, Russian nationalism, the regime in Kiev, etc. If the first time you’ve ever heard of Ukraine is in this survey, your answer reflects…very little, frankly, other than whatever loose associations came into your mind. Less-informed voters’ attitudes could change on a dime – and I’d bet that if we did the same survey in late 2016, the correlations would run the opposite direction due to the pro-Russian swing in GOP attitudes.
All political views are like this – arrayed somewhere on a spectrum between a tightly-held belief informed by ideological commitments and pure verbal flatulence. Each voter might have widely varying bundles of them, informed by their interests – e.g., a male Toledo small business owner might have really strong beliefs on OSHA regulation but pure verbal flatulence on abortion access issues, while a young woman working as a lawyer in Cleveland might be the opposite. We would reasonably expect voters to have much more attachment on issues which they encounter on a daily basis, and to be much less attached to positions on issues which are abstract. It is on these latter issues where we would expect more cue-following, where voters’ beliefs follow from what influential leaders are saying rather than driving the positions leaders take.
Russia-US relations remain an excellent example, demonstrated over and over this election season. Historically, Republican voters were much more anti-Russia than Democratic voters and more likely to name it as a prominent threat. But GOP voters’ attitudes on Russia and Putin swung dramatically more positive during this election season, driven by the vocal and frequent praise of both by party leader Donald Trump. The causal effect here is fairly clear: elite rhetoric drove opinion on an issue where most voters lack firm attitudes. However, we can reasonably expect the causal effect to go the opposite direction on issues like “cutting Social Security and Medicare” where the issue is immediate to many voters, and familiar and salient to virtually every voter. Voters will have opinions on this issue that drive their reactions to politicians, rather than the other way around.
Voter behavior in 2016 is more comprehensible if you believe most issues are more like Russia than Social Security. I believe this is particularly true in trade policy, where most voters have little to no awareness of the facts on the ground nor the relevant policy issues. In this light, polls showing a strong Trump performance among anti-trade Midwestern Democrats doesn’t show the issue benefited him, but just the opposite, which is that he benefited the issue among demographics favorable to him. Trump’s trade fixation never functioned as a policy plank. but provided a thinly-veiled mechanism to blame the ills of the nation on shadowy foreigners and exploit in-group defensive solidarity. Incidentally, this suggests to me Democrats have precisely zero electoral ground to gain by adopting anti-trade policies, as they would adopt the policy without gaining the in-group messaging benefit.
Be extremely skeptical of interpretations of politics putting public views on issues in the driver’s seat. This doesn’t matter just for considering 2016, but a great deal for considering how Democrats should look to the years ahead.