The misinformation that actually matters

People need to know more about current policies

Did you know most Americans believe that a majority of the national debt is owned by China? I learned this in my friend Emily Thorson’s recent book, “The Invented State: Policy Misperceptions in the American Public,” about the factually incorrect beliefs people hold about our government.

This, of course, isn’t always what people mean when they talk about misinformation.

Over the last few years, a cottage industry of people hand-wringing about “misinformation” has sprung up. Most of that misinformation discourse is a kind of lame cope from progressives who don’t want to admit that their side also believes stuff that isn’t true or from mainstream media outlets that want to blame social media for their own failings. Expressions of concern about misinformation have also grown increasingly related to potentially dangerous fantasies of information control and censorship — basically the idea that we could fix politics if we could stop people from saying untrue, right-wing things.

The power of Thorson’s point about misperceptions, though, is that addressing these kinds of latent misperceptions about aspects of the status quo — China holds most of our debt, undocumented immigrants are eligible for SNAP benefits — does not require some new Arbiter of Truth. And somewhat optimistically, her research suggests that, unlike trying to correct misperceptions about candidates or policy proposals, giving people better information about the status quo changes their minds.

Information that doesn’t matter

Back in the 2022 cycle, I was at an IHOP in Ohio and overheard a couple of blue collar guys spouting a lot of insane conspiracy theories about Joe Biden and Kamala Harris. But their conversation eventually meandered into how the CIA had assassinated JFK to prolong the war in Vietnam, so I asked them about that and mentioned that it was a left-wing theory that I was used to hearing from Democrats. They said they used to be Democrats — voted for John Kerry and for Obama in 2008 — and I asked why they switched. And they said fracking had been good for economic development in southern Ohio (and for them personally, because they worked in the fracking supply chain), but Democrats had become very hostile to fossil fuel extraction.

I think that’s an overstated characterization of the situation (domestic oil and gas production are at record highs under Biden), but it’s definitely not a deranged conspiracy theory. It’s just true that, especially in their rhetoric but to an extent in their policymaking, Democrats have become more skeptical of fossil fuel production and also that arguments about climate change have become more high-profile in political debate.

The upshot, I think, is that it would be naive to view these conspiracy theories —Biden is a senile puppet controlled by larger forces just waiting to slot Michelle Obama into the White House — as the reason these guys don’t like Biden. They developed a more negative view of Democrats because of energy issues and a more positive view of Republicans because Trump stopped advocating Social Security cuts and adopted Democrats’ critique of NAFTA. Once they became Republicans, they started believing unflattering conspiracy theories about Biden. And my guess is that these guys believed anti-Bush conspiracy theories in 2004. The partisanship is causing the misinformation, not vice versa.

A lot of the “misinformation” hype that isn’t focused on censorship fantasies seems to me to be wasted effort on this kind of thing.

In principle, of course, people should have accurate rather than inaccurate information. But it’s also natural that people don’t spend a ton of time interrogating exactly which mean things people say about politicians they dislike are true and which are false. This kind of misinformation strikes me as important primarily because elites focus on it in a way that leads them to make bad decisions. It’s easy to listen to a couple of guys who believe crazy shit and think “there’s no way we could ever reach them, they’re crazy,” when in practice, a lot of Obama-Trump swing voters had low social trust and a lot of crazy beliefs the whole time. It doesn’t follow from the fact that people who support your opponent have false beliefs that those false beliefs are the reason they won’t vote for you.

What’s interesting about the policy status quo information is that it actually makes a difference.

Background information makes a difference

Probably the cleanest example in Thorson’s book is that many people believe undocumented immigrants are eligible for SNAP benefits. This is not true, and indeed in many cases, it’s challenging for legal immigrants to get SNAP benefits.

But people don’t know this. I found a funny Reddit thread where someone asks how illegal immigrants get food stamps given that when he got SNAP several years back, he needed to provide ID and proof of citizenship. In the thread, someone “explains” that he’s just wrong and illegal immigrants can’t get SNAP (true) but that “children of undocumented immigrants can apply for DACA” (not true), which “gives them quasi legal status” (more or less) and could allow then to access benefits (not true). Somebody else says that “applying for DACA opens up food stamps, Medicare, and FAFSA.” This is true for FAFSA, but not for SNAP or Medicare. I’m also pretty sure this person meant Medicaid (for the poor), not Medicare (for the elderly), but that is also not true. However, due to a policy change announced last week, DACA recipients will be eligible for Affordable Care Act exchange plans, which was not previously the case.

This stuff is all somewhat hard to keep straight. I had to look it up to check that I remembered the rules correctly, and it turned out there had been this ACA policy announcement that I missed, even though I got a press release from the White House about it.

When the Biden administration and House Republicans reached a deal on a deficit reduction package last year, one of the provisions involved work requirements for SNAP. That generated a bunch of news stories that described the new policy to some degree. But none of them just say “oh by the way, illegal immigrants aren’t eligible — even DACA recipients.” It would be a little odd, mechanically, to insert something like that into a news story. It’s not new, after all, and it wasn’t what the politicians were arguing about.

But Thorson reviews experiments that have looked at whether giving people new information changes their policy attitudes. She finds that in most studies, explicitly trying to correct people’s views about the outcomes of their policy preferences (like when I try to convince people that mass deportations would make inflation worse) doesn’t change attitudes. But when people learn what share of the budget actually goes to foreign aid, they become less supportive of cutting foreign aid. Similarly, learning the average salary of an American K-12 teacher seems to make people less supportive of raising teacher pay.

People believe a lot of weird stuff

One piece of misinformation I sometimes come across is the idea that trying to correct misconceptions is fruitless, because people engage in motivated reasoning and will only double-down on false beliefs. In fact, that kind of backfire effect is quite rare (see Alexander Coppock’s book), and giving people good information typically causes them to update.

What’s challenging is getting people to change their own political commitments.