Here’s an excerpt of Sarah Kliff’s interview with Debbie Mills, a Trump voter:

Are you surprised how much Republicans are talking about repeal [of Obamacare]?

No.

Did you expect — do you think they’ll do it, or do you think it’ll be too hard?

I’m hoping that they don’t, ’cause, I mean, what would they do then? Would this go away?

Yes, possibly.

The insurance?

It will go, if they repeal it. I mean, that’s what they promised to do in so many elections.

Right … so … I don’t know. … [snip]

Our interview began to make her a bit nervous.

You’re scaring me now on the insurance part … I’m afraid now that the insurance is going to go away and we’re going to be up a creek.

How could this happen? Why would vulnerable people vote against their own essential interests when they are aware of the stakes and have no altruistic or principled objections to the policy that’s at risk?

Kliff proposes that it was reasonable to doubt that Obamacare really would be repealed, since many political experts also predicted that it would become untouchable, like Social Security. Even to this day, there is a chance that Republicans will leave it alone. But, if we assume that Debbie Mills should have voted for Clinton over Trump to preserve Obamacare, then here are two familiar explanations for her choice:

1. Trust. Nobody really knows that political leaders will do in the future. Nobody even knows what any given policy will accomplish. We all rely on information, interpretation, predictions, and promises from sources that we trust. I think many Trump voters did not trust Trump to do anything specific that he said, but they did trust his general competence and alignment with their interests. Meanwhile, they distrusted Clinton’s motivations. I believe they were wrong in these judgments, but the difference is not my superior rationality. Rather, we made different assessments of trustworthiness.
2. Salience: Joseph Schumpeter observed in 1942 that “the typical citizen drops down to a lower level of mental performance as soon as he enters the political field. He argues and analyzes in a way which he would readily recognize as infantile within the sphere of his real interests. He becomes a primitive again. His thinking becomes associative and affective.” This was true, Schumpeter said, of “educated people and of people who are successfully active in non-political walks of life.” The reason was basically that each vote hardly counted, so it was irrational to spend a lot of time sorting through the “masses of information” that were already available in 1942–not to mention the vast masses today–to make careful judgments. In the absence of “immediate responsibility, ignorance will persist.” And voting, although a responsibility, is too small to compel much attention. Schumpeter originated this way of thinking about politics, but much subsequent psychological research has reinforced it.

I would add a third explanation, and this one is significant because it is likely to change.

3. Divided government. For 28 of the past 36 years, the elected branches of the federal government have been divided between Democrats and Republicans. Even during the remaining eight years (two each under Clinton and Obama and four under G.W. Bush), Senate filibusters, opposing state governments, and courts have checked the majority’s power. This is one reason that net government spending–federal plus state plus local–rose by nearly 50% under a conservative-sounding president (Bush) and leveled off under under a liberal-sounding president (Obama). As long as the ideology of our most prominent leader is largely unrelated to the actual policies in place, voters get poor feedback from their choices at the ballot box. That makes them unlikely to learn.

I am not saying that it’s a Good Thing that we now face unified Republican government in most of the country. Mills will probably lose her health insurance, which could shorten or wreck her life. Many others will pay a severe price as well. It’s relatively easy for me to see the bright side, since I am not nearly as vulnerable.

But there is a bright side. If you believe in electoral democracy at all, you must acknowledge that voters will make mistakes severe enough to cost lives. The argument for electoral democracy is that voters will learn from such mistakes. But we have frustrated such learning for more than a generation. The political system has performed very poorly at times–killing half a million Iraqis, incarcerating 2 million Americans, allowing our industrial cities to whither wither away–but few citizens have had to rethink their prior assumptions about which ideology is better for them. The signal has been lost in the noise.

The signal is now about to be heard pretty clearly. Democrats and other progressives should amplify it by constantly drawing connections between the reigning ideology and its outcomes, and by refusing to mitigate the short-term damage. Then Trump’s 2016 victory will be Pyrrhic.