The whole notion of “misinformation” as conventionally construed has taken some blows lately, including scandals in the misinformation research field and, more importantly, some great work from Brendan Nyhan, Emily Thorson, and co-authors showing that “in our review of behavioural science research on online misinformation, we document a pattern of low exposure to false and inflammatory content that is concentrated among a narrow fringe with strong motivations to seek out such information.”

From where I sit, that’s all to the good — I’ve been complaining for years about the problems with this framework as an explanation for political outcomes. It is true that there is fringe content circulating on the internet and also that some of your political enemies probably believe some of it, but there’s little reason to believe that such content exerts an important causal influence on American politics.

That said, one complaint I still have about the “misinformation” paradigm is that even in its debunking forms, it uses a generic term in a highly specific and somewhat peculiar way.

People have a lot of erroneous beliefs about the policy status quo in the United States, and that seems to matter. These beliefs are normally not formed via exposure to some kind of social media misinformation; they’re just about things that aren’t in the news very much and that people misunderstand. Which is to say that “people having information that is not correct” is absolutely a huge deal in politics… it’s just not necessarily “misinformation” in the sense that the misinformation police intend. In Dylan Matthews’ profile of the State Department’s small but very successful intelligence bureau, for example, one thing that comes through is that the bulk of American intelligence agencies genuinely believed that Iraq had an advanced nuclear weapons program. This erroneous information had a huge impact on the media, on the mass public understanding of political debates 2002-2003, in decision-making in Washington, and on the broad trajectory of American politics.

And I think erroneous ideas that are perpetrated by mainstream institutions — what I’m going to call “elite misinformation” — are a really big deal in an underrated way.

I don’t want to rehash this in detail, because it’s been well covered recently, but a good example of this sort of misinformation is the narrative about a huge rise in maternal mortality in the United States. Because as a growing chorus of critics has been pointing out, this increase was largely the mechanical result of a change in counting methods, not in the public health situation. That’s bad, but what’s really shocking, as I learned from Jerusalem Demsas, is that key actors are totally unapologetic about sowing confusion:

Christopher M. Zahn, the interim CEO of the American College of Obstetricians and Gynecologists, wrote a lengthy statement in response, arguing that “reducing the U.S. maternal mortality crisis to ‘overestimation’” is “irresponsible and minimizes the many lives lost and the families that have been deeply affected.” Why? Because it “would be an unfortunate setback to see all the hard work of health care professionals, policy makers, patient advocates, and other stakeholders be undermined.” Rather than pointing out any major methodological flaw in the paper, Zahn’s statement expresses the concern that it could undermine the laudable goal of improving maternal health.

This strikes me as a shortsighted and pernicious way to think about the purpose of communicating with the public. And yet, people are out here saying it in public!

The case of the mystery fossil fuel subsidy

Years ago, when I didn’t cover climate and energy topics at all, I floated the idea of a piece making the point that not only were carbon tax proposals politically toxic, but all these various schemes to subsidize clean energy seemed like a relatively inefficient way to reduce fossil fuel use. Wouldn’t it be better to just get global governments to cut their massive subsidies for fossil fuel use first?

After all, I’d seen lots of stories with headlines like this:

• “Fossil Fuel Firms Are Subsidized at a Rate of $11 Million Per Minute” — Mother Jones

• “Why Are Taxpayers Propping Up the Fossil Fuel Industry?” — NYT

• “Why Are Governments Still Subsidizing Fossil Fuels?” —Washington Post

• “Explainer: Global fossil fuel subsidies on the rise despite calls for phase-out” —Reuters

But even though these scandalously high subsidies were widely covered in the mainstream press, nobody really seemed to follow the basic logic of them. If fossil fuels are receiving massive subsidies, then by far the best and most growth-friendly way to decarbonize is to cut the subsidies. On the one hand, yes, this information made me feel more strongly than ever that the world should act decisively on climate change. At the same time, though, it made the actual policies being pushed by mainstream environmental groups seem kind of weird. Why not tackle the subsidies?

Well, it turns out that these stories about subsidies, though widespread, are incredibly misleading.

They’re based on an annual report put out by the International Monetary Fund, whose most recent edition is headlined “Fossil Fuel Subsidies Surged to Record $7 Trillion.” That’s a lot! But check out the fine print on the chart.

The vast majority of the “subsidies” are “implicit subsidies,” which include “undercharging for environmental costs.” In other words, they are characterizing governments’ failure to impose a carbon tax as a “subsidy” for fossil fuel use.

As a thought experiment in a textbook chapter about externalities, I think that’s a perfectly sensible formulation. And to some extent, I think the point the IMF is making here is dead on: If world governments came together to get each state to locally impose a carbon tax equal to the global social cost of carbon, that would raise approximately $7 trillion in revenue. And that revenue could be used to reduce outstanding public sector debt burdens or reduce broad taxes on consumption and labor. In the aggregate, the world would end up with cleaner air and a more stable climate, and the economic cost of doing this would be minimal, or arguably even negative. As I’ve written before, a carbon tax is, in fact, a good idea, but the big challenge of climate policy is it involves difficult coordination problems, because for any one country, setting the tax at the global social cost of carbon would be too high.

But instead of explaining itself clearly, the IMF is putting out misleading headlines, which are then picked up and bounced around by mainstream media outlets.

In the context of misinformation studies, it probably wouldn’t count as misinformation if someone believed, because they read a New York Times write-up of an IMF report, that “Around the world, taxpayers are helping to support fossil fuels through subsidies when their money could be funding green energy transitions instead.” And, in fact, to the credit of the New York Times article in question, the ninth paragraph of the story does explain what’s going on. But the whole beginning of the article makes it seem like we’re talking about direct subsidies. And because there are nonzero direct subsidies, the rest of the article discusses why those are politically difficult to scrap in the countries that have them. Similarly, the IMF explains what they mean right there in the fine print of the chart. They’re not lying, exactly. But it’s hard to understand why the IMF would do this, unless they thought it was a good idea to trick people.

There’s lots of this going around

A lot of elite misinformation takes the form of catastrophism.

The expiration of Covid-era child care subsidies was going to create all kinds of systemic disasters for the American economy, but those disasters never materialized. What actually happened is that some people ended up needing to pay more for child care, which can be a genuine hardship. Exaggerating the problem here was, I assume, supposed to inspire action. Or remember net neutrality? When Obama created those rules in 2015, it was supposedly a big deal, and then when Trump scrapped them, all kinds of “sky is falling” rhetoric emerged. But the actual results of the change were, at most, quite mild. Under Biden, net neutrality is back, and I don’t think anyone noticed.

The Republican version of this that I see most frequently are claims that tax cuts will magically pay for themselves.

I don’t know Republican Party politics well enough to understand the extent to which rank-and-file cadres of the conservative movement genuinely believe this. But I do know Democratic Party politics well, and my experience is that just as I, personally, was misled for years by the IMF fossil fuel subsidies claims, lots of people who work professionally in liberal politics sincerely believed that child care cliffs or net neutrality disasters were looming.

Which is why I think all this sloppy work and misleading rhetoric is both more impactful than a lot of people realize, and also a lot less tactically savvy than those doing it think. The problem is that it’s about a million times easier to persuade a highly engaged member of your team of something than it is to persuade a swing voter (who is probably skeptical, cynical, and not that engaged with politics) or a member of the opposition (who instinctively assumes you’re lying about everything). So when you put something out there that’s false or misleading, you’re about a million times more likely to confuse people who are friendly to your cause than to actually persuade anyone worth persuading.

Lying to people is bad

I think we’ve probably all been tempted at some point to either propagate something untrue or else quietly tolerate someone else doing it in pursuit of a cause we believe in. There’s an old tweet I wrote over a decade ago defending a lowball projection of high-speed rail construction costs on this kind of means-ends grounds. People sometimes throw that one back in my face, and fair enough — it was a bad take. But I think I’ve done my penance over this, and my work over the past 10 years has really emphasized how challenging it is to actually build complicated infrastructure projects in a bad epistemic environment.

You can dupe people into a one-off, like the invasion of Iraq, but to have a sustained program of policy reform, you need to do things that work. And the political system is too large — hundreds of legislators, thousands of staffers, innumerable bureaucrats and interest groups, along with media, influencers, and other stakeholders — to operate on a conspiratorial basis.

Getting people worked up about incorrect maternal mortality numbers is a good way to drive some attention to the issue. But what’s supposed to happen next? A governor or senator who gets interested in the problem of skyrocketing maternal mortality is going to ask what’s driving this trend, and unless you have a persuasive answer, you’re not going to be able to get anything done. And in this case, since the trend is basically fake, it’s unlikely you will have a persuasive answer! What you might have are some convincing ideas for cost-effective interventions to save pregnant women’s lives. But if you have that, why not just pitch them without the fake trend?

The climate version of this is, if anything, worse. Convincing people that there are trillions in fossil fuel subsidies makes the case for regulatory action, new investments in green energy, or really any kind of climate action look weaker, not stronger. By characterizing “failure to enact a carbon tax” as a form of subsidy, you’re confusing loosely attentive people who care about climate change, making politicians look more gutless than they are and existing policy look dumber than it is, and suggesting that there’s an easy solution at hand when there isn’t.