Enjoyed the episode? Want to listen later? Subscribe by searching 80,000 Hours wherever you get your podcasts, or click one of the buttons below:

Workaday politicians running for office are probably better informed about the state of public opinion than you are.

And if you are finding yourself baffled as to why someone won’t say something or embrace something that you think they should, it’s probably because their surveys indicate that it’s not that popular, and maybe try to be a little less mad.

Matthew Yglesias

If you read polls saying that the public supports a carbon tax, should you believe them? According to today’s guest — journalist and blogger Matthew Yglesias — it’s complicated, but probably not.

Interpreting opinion polls about specific policies can be a challenge, and it’s easy to trick yourself into believing what you want to believe. Matthew invented a term for a particular type of self-delusion called the ‘pundit’s fallacy’: “the belief that what a politician needs to do to improve his or her political standing is do what the pundit wants substantively.”

If we want to advocate not just for ideas that would be good if implemented, but ideas that have a real shot at getting implemented, we should do our best to understand public opinion as it really is.

The least trustworthy polls are published by think tanks and advocacy campaigns that would love to make their preferred policy seem popular. These surveys can be designed to nudge respondents toward the desired result — for example, by tinkering with question wording and order or shifting how participants are sampled. And if a poll produces the ‘wrong answer’, there’s no need to publish it at all, so the ‘publication bias’ with these sorts of surveys is large.

Matthew says polling run by firms or researchers without any particular desired outcome can be taken more seriously. But the results that we ought to give by far the most weight are those from professional political campaigns trying to win votes and get their candidate elected because they have both the expertise to do polling properly, and a very strong incentive to understand what the public really thinks.

First and foremost, that means representing issues as they would be in a hotly contested campaign. If someone says that they sure like the idea of taxing carbon, how much do they still like it when they find out it means their electricity bills would be $100 higher, and gas will cost 20 cents more a gallon? And do they still like it when they know one of the candidates is against it and says it will cost local jobs? This sort of progressive ‘stress testing’ is more work, but can lead researchers to very different conclusions than just asking people favour ‘policy X’.

The problem is, campaigns run these expensive surveys because they think that having exclusive access to reliable information will give them a competitive advantage. As a result, they often don’t publish the findings, and instead use them to shape what their candidate says and does.

Journalists like Matthew can call up their contacts within campaigns and get a summary from people they trust. But being unable to publish the polling itself, they’re unlikely to be able to persuade sceptics.

That’s a pain and a legitimately hard problem to get around. But when assessing what ideas are winners, one thing Matthew would like everyone to keep in mind is that politics is competitive, and politicians aren’t (all) stupid. If advocating for your pet idea were a great way to win elections, someone would try it and win, and others would copy. If none of the pros are talking about your hobby horse, it might be because they know something you don’t.

One other thing to check that’s more reliable than polling is real-world experience. For example, voters may say they like a carbon tax on the phone — but the very liberal Washington State roundly rejected one in ballot initiatives in 2016 and 2018.

Of course you may want to advocate for what you think is best, even if it wouldn’t pass a popular vote in the face of organised opposition. The public’s ideas can shift, sometimes dramatically and unexpectedly. But at least you’ll be going into the debate with your eyes wide open.

In this extensive conversation, host Rob Wiblin and Matthew also cover:

  • How should a humanitarian think about US military interventions overseas?
  • From an ‘effective altruist’ perspective, was the US wrong to withdraw from Afghanistan?
  • Has NATO ultimately screwed over Ukrainians by misrepresenting the extent of its commitment to their independence?
  • What philosopher does Matthew think is underrated?
  • How big a risk is ubiquitous surveillance?
  • What does Matthew think about wild animal suffering, anti-ageing research, and autonomous weapons?
  • And much more

Get this episode by subscribing to our podcast on the world’s most pressing problems and how to solve them: type 80,000 Hours into your podcasting app. Or read the transcript below.

Producer: Keiran Harris
Audio mastering: Ben Cordell
Transcriptions: Katy Moore

Highlights

How Matt thinks about public opinion research

Matthew Yglesias: Something that I have found is that the world of public opinion research is kind of bifurcated — or trifurcated, if that’s a word. On the one hand, you have people doing polls to go to the media, for public consumption. Then you have people doing polls to issue advocates, who want polls that say their positions are popular. And then you have people doing polls because they’re trying to win campaigns. Survey methods are the same across these things, but what your incentives are and what you actually care about are pretty different.

Matthew Yglesias: Unfortunately, for the mass public, to get a good read on public opinion — like a rigorous look at how things are without putting a lot of English on it with question wording — you do have to talk to the people who are doing private work for political campaigns. They are the ones who have the strongest incentive to get the question right — not in terms of the survey sample, but in terms of what questions you ask people.

Matthew Yglesias: I know there’s a couple guys on Twitter who are very angry that I say carbon taxes are unpopular. They’ve got their polls that say it’s popular, but we know in America that when this has been put to ballot initiative in Washington State, it’s lost badly, twice. We also know that in Europe, where the macropolitics is greener than in the United States, the European elected officials act as if stiff carbon pricing is going to be very unpopular. In Australia, when the government tried to put a carbon tax in place, there was a big backlash to it.

Matthew Yglesias: So people say that if you do the best kind of issue surveys, where you do partisan frames, you say, “Some Democrats are proposing a blah, blah, blah, price on carbon, which they say will reduce climate change in the most cost-effective way possible. Republicans say it’ll raise the price of gas by…” — then you give them a real number, not a crazy lie: “It’ll raise the price of gas by so many cents, raise the price of electricity by so much.” It becomes incredibly unpopular in that framework — like, less popular than commercial legalization of heroin.

Matthew Yglesias: When the price of gasoline spiked in the United States a few months ago, it was a huge deal politically. People were losing their shit about it. Probably the best media poll on this is a Reuters survey, where they asked people, “Should we take drastic action to stop climate change?” Most people said yes. Then they asked, “Would you be willing to pay $100 a year more in taxes to stop climate change?” People said no. A hundred dollars a year is not that drastic, you know? I spend $100 a year on things that I don’t think are important at all.

Matthew Yglesias: So I think that it’s tough. As a professional journalist, the kind of thing that I can do is ask people to tell me things that they cannot say in public, and then I can try to triangulate the things that they have told me against publicly available information — like this Reuters thing, like the Washington polls, like the fact that practical politicians don’t campaign heavily on green tax shifts — and try to make people sort of see the truth. But it’s hard. I never know, because I’m not a huge scoops guy, and if you want to believe that I’m lying to you or that my sources are lying to me, it’s challenging to prove otherwise.

Matthew Yglesias: I have stumbled, I would say, a little bit ass-backwards into being a guy who writes a lot about polls and public opinion surveys. And it was not my intention, exactly — I don’t know that it’s super important for the typical person to be deeply, deeply invested in this. The main thing that I would like the person on the street to take away from it is that workaday politicians running for office are probably better informed about the state of public opinion than you are. And if you are finding yourself baffled as to why someone won’t say something or embrace something that you think they should, it’s probably because their surveys indicate that it’s not that popular, and maybe try to be a little less mad.

Matthew Yglesias: Now, what should you advocate for? Whether it’s you, or the listener, or whoever else, you probably should advocate for the right thing to do. And you should probably advocate for the right thing to do as if you were trying to be persuasive, which I think people oftentimes don’t do on the internet. I would say heavy consumers of political punditry spend a lot of time pounding the table on behalf of what they think is the right thing to do, being very impatient that other people in positions of greater responsibility aren’t saying exactly what they want them to say, and searching for bias-confirming information that indicates that they are right about everything.

Loss aversion and long-term interests

Matthew Yglesias: People dwell on the downside and loss: people are more worried about losing what they have than about gaining something that they don’t currently have. So it’s hard for people on the left to persuade people that some new program is going to be amazing and it’s going to be worth paying the taxes, but it’s also hard for people on the right to win arguments that are like, “Well, without this program, we can cut taxes and that’s going to create better incentives to save and invest, and that’s going to increase growth by 0.002%.” When you compound that out for X number of years, people get a lot of like, “Eh, I don’t know about that.”

Matthew Yglesias: People are also just not great at thinking about the long term. If you ask somebody who’s 40, “Do you care about what’s going to happen to you when you’re 70?” They’d be like, “Of course. I’m not a crazy person.” Now you ask them, do they care what’s going to happen hundreds of generations in the future, they might say no. We might say, “OK, you discount more than you should.” But even when they say that they’re not discounting the future of their own life, I think they just pretty clearly are when they’re making policy decisions.

Matthew Yglesias: People are not in the habit of really thinking about their long-term interests, and we see that in all kinds of behaviors. It’s why there’s people who smoke. It’s why lots of people struggle with all kinds of personal health issues. And in their political thinking, it’s even worse. To really get people to think, “How’s this going to play out?” — it’s tough.

Rob Wiblin: Yeah. People talk a lot about means testing making policies more or less popular. So whether you allow everyone to access a program, or only people who are below some income threshold.

Matthew Yglesias: Yeah. People argue about this a lot. I actually think if you pay attention to what everyone’s saying, they are not in that much of a tension really. If a program exists and everybody is using it, that makes it much, much harder to get rid of, versus a very narrow type of program — so there is a political durability benefit to some universalistic designs. At the same time, the broader you make the program, the more tax revenue you need to get.

Matthew Yglesias: And so, if we do comparative welfare state design, typically the really big welfare states have a broad tax base: they use a lot of payroll tax, they use a lot of value-added tax. Those taxes tend to be unpopular, so it’s harder to create a really broad universal program, but also harder to get rid of a really broad universal program.

Matthew Yglesias: What should you actually try to do? I don’t know that there is a general answer to that kind of question. I think the most important thing about means testing is how it actually makes the programs function — and that making everyone fill out 17 forms to verify their income is itself quite costly.

How military intervention looks as a humanitarian intervention

Matthew Yglesias: This was my original entry point into the world of effective altruism, because I was very taken in the late ‘90s by the idea of humanitarian military intervention, by the idea that we’re doing incredible good in the world in Kosovo. And that this was a big thing, that post–Cold War, the American military could end genocide around the world, and do all kinds of good things.

Matthew Yglesias: I think in Iraq, that line of thinking turned out poorly on a sort of first-order practical basis. But that also got me thinking about the question of, “How do you help people around the world?” If you want to take a non-national interest view of the world, and say, “Who do we help?”, it turns out that there are radically cheaper ways of helping people in need around the world. If you want to do something that is selfless, there’s so much that can be done. We’re not close to exhausting the possibilities for doing that. I try to do some of it with my personal giving. I try to encourage others to do it. I try to encourage the government to do it.

Matthew Yglesias: But it began to seem to me that another area where we had to be more honest with ourselves about, “What are we actually doing here? Is the military apparatus that exists really a tool for human betterment?” Now, it can be. When there was the tsunami in Indonesia, the fact that we have a navy. The navy is full of boats — they’re really good boats, they have really skilled sailors. They were able to deliver a lot of humanitarian relief aid. It was great that they did that.

Matthew Yglesias: I’m not gonna go full paranoia, like it was just a propaganda game to help the CIA dominate the universe. But it is not really helpful to think about military force as a primary tool for humanitarianism. The cost benefit is not there. The incentives aren’t aligned. It’s not what the people are trained to do. And you see how quickly it turns. As long as we were fighting in Afghanistan, we would say we were doing all these good humanitarian things. The moment the Taliban takes over Afghanistan, we have them under severe sanctions that are causing a famine over there.

Matthew Yglesias: So I think we should not do that. Famine is very bad. But it’s a warmaking tool: their mission is to beat the Taliban. When beating the Taliban involves doing a certain amount of humanitarian propaganda — which involves a certain amount of humanitarian real stuff — the military goes and does it. But the moment the way to beat the Taliban is to starve the country into submission, that’s what that exact same kind of apparatus flips to.

Matthew Yglesias: It’s challenging because the policymakers don’t speak honestly about their own commitment to humanitarian values. So we are very eager to back opposition figures against authoritarian regimes sometimes, but not against Kuwait. There’s not going to be a colour revolution there. We can say casually, “Well, that has something to do with oil” or something.

Matthew Yglesias: But nobody is going to write down on a piece of paper, actually, what are the criteria in which the United States or other Western governments will back democratic opposition? When will we stop people from massacring? In part because we don’t want to give a clear green light to authoritarian regimes — we always want the Saudis to be guessing like, “How much can we get away with?” But it makes it a very hazy landscape, and I think it’s not closely comparable to global public health interventions — if we could convince people to care a little bit more about distributing insecticide-treated bed nets, we could stop malaria cases.

What governments should do about comets and supervolcanoes

Matthew Yglesias: NASA poked around on the supervolcano issue a few years ago, and they seem to have the idea that you could try to cool down the magma underneath Yellowstone with essentially injecting water. That’s related to the idea of advanced geothermal as an electricity-generating concept.

Matthew Yglesias: So there are these regulatory sensitivities around geothermal drilling on federal lands — a ton of sensitivity about doing it in an actual national park, which is where Yellowstone is. But I think that sacrificing a portion of the park in order to not have literally the entire planet explode would probably be a win-win. And we could offset that with some more parkland someplace else. Again, as you would expect, it’s not totally clear that that would work, but it’s worth investing some money in the exploration of whether or not it works. There’s a few other supervolcanoes that are out there.

Matthew Yglesias: As I understand it, after Deep Impact and Armageddon came out, we actually got the government to track asteroids better. And comets are just a little bit harder to track: they come in at a sharper angle and they go further away. But we should put some more telescopes up there and try to find them.

Matthew Yglesias: There’s a lot of nostalgia for the kind of heroic age of NASA and space exploration, which was very motivated in its heyday as a kind of national defense imperative against the Soviet Union. Today I don’t think that you can very plausibly claim that we need to go to Mars to stick it to China. But there’s a very clear national defense rationale for comprehensively tracking objects in outer space as far away as we can.

Matthew Yglesias: The amount of money that’s spent on defense programs is very, very, very large, and tracking those comets is good — especially as we have more private sector interest in some of the sort of sexy low-hanging fruit of, “Let’s get some human beings and have them go around the Earth in a circle.” You know, good for the entrepreneurs. And there’s not a huge ROI financially in tracking comets, but socially it’s very valuable.

How Matt thinks about advances in AI

Rob Wiblin: Do you think we should do anything more now to prepare domestic society or the international order for the possibility of major advances in AI over the coming decades?

Matthew Yglesias: We definitely should. The question is always what, right? As the person who asks the questions on podcasts, I’ve had people from the effective altruism and longtermist world on the show. Or just talked to them informally, to be like, “What should I do? Say I decide that I take AI risk extremely seriously. What do you want me to say about this?” And their answers always strike me as fairly fuzzy, so I’m left at a bit of an impasse.

Matthew Yglesias: Now again, to the young people out there, if you are technically literate and you’re in your early 20s or late teens and wondering what to do with your life, there is a lot of demand for good ideas about incentive-compatible artificial intelligence — how to make this maximally beneficial rather than threatening to humanity, on a policy level. I’m in the market for hot takes.

Rob Wiblin: Yeah, I think that there are policy ideas coming up through the academic conversation, but it does still feel like the conversation is pretty preliminary. And they seem fairly reluctant to go out and publicly advocate for anything in particular, which is interesting. I suppose they just worry that their ideas could do more harm than good a lot of the time.

Matthew Yglesias: That’s definitely a risk that’s out there. And I worry sometimes about how you have communities, or you have schools of thought — and that’s good. But sometimes something can become a scene. It’s like to be “in” in part of the tribe, you’d be like, “Oh, I’m really worried about AI existential risk.” OK, but “being worried” about things doesn’t accomplish anything in life.

Matthew Yglesias: And you can just kind of dismiss other problems. Say, like, “Well, don’t worry about that — worry about the AI risk.” But is there a tradeoff? Is there something that a typical person — a typical elected official even — should be doing differently? I’m not sure. I’m not saying no, but I haven’t been super convinced, in the way that I have on pandemics or even supervolcanoes. There are these kind of fiscal tradeoffs in the NASA budget, in NIH — what they care about that really addresses existential risk on those topics.

Matt’s views on the effective altruism community

Matthew Yglesias: I’m for it, I think. I want to be effective; I want to be altruistic. I think it’s good — I think that these lines of thought are very good. I should have said this when we were talking about foreign policy before, but actually one of my EA origins is when I was in college, I saw Peter Singer lecture about Iraq. And I was an Iraq War–supporting student at that time, and I thought that what he had to say was very impressive and challenging to my preconceptions. Part of why I went to see him in the first place is that I had no idea what he thought about foreign policy, but I’d just read some of his general books about normative ethics and stuff like that — so I was bought in and therefore open to it.

Matthew Yglesias: So I think it’s great. Something I worry about is that as you start to have a community, all communities can become a little bit of an extreme version of themselves — when that isn’t necessarily where intellectual movements actually do good in the world at the margin.

Rob Wiblin: So your concern would be that if people are all talking to one another — or they’re trying to impress other people who agree with their broad worldview — then you tend to get this extremizing effect over time, and people maybe solidifying views that have gone too far in some direction and that’s not the most accurate?

Matthew Yglesias: The views can be wrong, but also they can become a… OK, what is “effective altruism”? Effective altruism is the idea that we should be more critical and more rigorous about our altruistic impulses, right? And that we should push people to ask the questions, “Is this really a good way to help people? Are you too focused on things that are close to you in your community, too focused on things that get praised by other people? Or should we cast a wider net? Should we deploy more rigorous empirical methods when we evaluate our programs?” And that’s good, those are a good set of questions, a good set of pressures in the world. So we say, “Is military intervention humanitarian? Should we be clapping for the guy giving money to the art museum? Here’s this great study about bed nets and malaria.”

Matthew Yglesias: Then you have a community that buys those premises, and they start talking to each other a lot. They eventually develop the idea that we all should just think about the long term and we shouldn’t discount at all. And if you don’t discount at all, the only thing that really matters is human extinction. And a bunch of smart people have thought about this, and the biggest threat to human extinction is rogue artificial intelligence.

Matthew Yglesias: So now we’ve all read that book, and we’ve all put it around. And now what it means to be an effective altruist is, somebody is like, “What’s on your mind?” and you’re just like, “The threat of human extinction due to artificial intelligence.” And a large share of people are going to hear that and they’re going to be like, “What? That’s weird.” And an even larger share of people are going to be like, “Well, what does that even have to do with me?” Right? So now you’re not actually asking people to be more self-critical about their charitable giving or to give more money at the margin. You’re not pushing —

Matthew Yglesias: You’re asking people to take this Kierkegaard-like leap and join the community — as opposed to being in your community and living your life and caring about roughly the things you care about, but shifting your orientation a little bit.

Matthew Yglesias: I think that you become less effective, in a meta sense, when you become this set of doctrines that are quite odd and esoteric to most people — and that also don’t have a lot to do with people’s lives and the decisions that they’re actually making — versus one that’s trying to say to people, “For your end-of-year charitable contributions, consider the GiveWell Maximum Impact Fund as something to do. Think about giving a little bit more than you usually do. Think more critically about the doing/allowing distinction in your life.”

Matthew Yglesias: These are really important real-world messages, I think. As I’ve seen this evolve over the years, I both am myself somewhat bought in and try to get people to think more about x-risk and other stuff like that. But I worry also that the most broadly relevant stuff can get lost a little bit.

The impact of Matt’s work

Matthew Yglesias: When I’m trying to be self-conscious and reflective about what I do — this is what I try to do: I take ideas that exist in the academic-y realm and that are very neglected in politics, and bring them to more people’s attention, until the point where someone has done enough with it that it becomes a conventional political conflict that people then cover as, “Well, the YIMBYs are fighting against the NIMBYs.” And at that point, I don’t think that political punditry actually does a lot. That’s the most common kind of punditry: “I’m going to write an article about why the Freedom to Vote Act is good.” That strikes me as incredibly low efficacy.

Rob Wiblin: Because it’s so saturated?

Matthew Yglesias: It’s incredibly saturated, but also it’s on the docket, right?

Rob Wiblin: I see. It’s already on the agenda. People are going to think about it regardless.

Matthew Yglesias: Right. For actual members of the United States Senate to be taking a vote on an issue is so far down the chain, and the idea that Kyrsten Sinema is going to read my blog posts —

Rob Wiblin: Crack open the newspaper.

Matthew Yglesias: — and be like, “Shit, I’ve got this all wrong.” That’s an insane level of hubris, and yet it seems to be what most people think is the humble way to do columnist work. I think it makes so much more sense to just be like, “Hey, here’s a thing,” and then maybe one person somewhere is like, “Oh, maybe I could work on that.” Or one guy who’s rich is like, “Sure, I could cut a check to that,” and you see where it goes.

Matthew Yglesias: It’s not that hard to believe that you can convince one of the world’s many, many, many wealthy people to, one time, deliver some financial support to something that nobody is doing. That, to me, is a much more realistic aspiration in life. And yet I think in my field, it sounds more egomaniacal to be trying to take credit for some early-stage grants than to be like, “I’m waging the war of ideas. I’m going to convince people that Donald Trump is bad.” Like, how would I do that?

Articles, books, and other media discussed in the show

Matthew’s work:

Polling, public opinion, and voting:

Philosophical views:

Other 80,000 Hours Podcast episodes:

Everything else:

Transcript

Table of Contents

Rob’s intro [00:00:00]

Rob Wiblin: Hi listeners, this is The 80,000 Hours Podcast, where we have unusually in-depth conversations about the world’s most pressing problems, what you can do to solve them, and why writing a blog post probably won’t convince Mr Putin to reassess his Ukraine policy. I’m Rob Wiblin, Head of Research at 80,000 Hours.

Many of you have likely heard of today’s guest — journalist Matthew Yglesias — and I give him a thorough intro in just a second, so all I’ll say is that I really enjoyed listening back over this interview, even though I was down and out with COVID when Keiran sent it to me.

Before that though, a reminder that we’ve got two job opportunities here at 80,000 Hours which you might want to think about applying for.

First off, we’re hiring career advisors to join our one-on-one team. Our advisors do video calls with hundreds talented and altruistic people in order to help them find the highest-impact career they can.

We’ve found that people from a surprising range of professional backgrounds can be good at this. If you want to hear more about what it’s like you can get a taste in episode 75 — Michelle Hutchinson on what people most often ask 80,000 Hours.

It’s a London-based role with a starting salary around £65,000, and applications are closing very soon on 20 February so we won’t get a chance to remind you again.

We’re also hiring a new Head of Job Board to improve and expand our job board. You can check out what it looks like now at 80000hours.org/jobs.

It’s already a very popular service but the new job board lead will be responsible for figuring out how the board can help even more people find jobs where they have a huge impact, and manage the team that works on it.

That’s also a London-based role, and a typical starting salary for someone with five years of relevant experience would be something like £72,000 per year. You’ve got an extra week for that one as it’s closing on the 27 February.

If you’d like to learn more about either role or what it’s like working at 80,000 Hours in general, head to 80000hours.org/latest and scroll down a bit to the blog posts announcing each position, or use the links in the show notes. If you know someone who would be perfect for the role, do let them know ASAP so they can put in an application.

All right, without further ado, I bring you Matthew Yglesias.

The interview begins [00:02:05]

Rob Wiblin: Today, I’m speaking with journalist Matthew Yglesias. Matthew originally studied philosophy at Harvard, and started blogging before it was cool — back in 2002, at the age of 21. That got him a bunch of attention, and he went on to write columns at various US magazines, including The American Prospect, The Atlantic, and Slate. In 2014, he joined Ezra Klein and Melissa Bell, among others, to cofound the news website Vox, which has since grown to become a very well-known and widely read online publication.

Rob Wiblin: Matt initially took a management role at Vox, but ultimately found he preferred to work on the creative side, hosting the US policy podcast called The Weeds, as well as writing regular columns. In 2020, he left his job at Vox to write his own personal subscription newsletter called Slow Boring, using a service called Substack. This quickly attracted many paying subscribers, both increasing his income substantially and setting the example that subscription newsletters could be a viable model for independent journalists.

Rob Wiblin: On the book front, back in 2012, Matt wrote The Rent Is Too Damn High, which argued that single-family residential zoning was causing massive harm in the United States. This popularization of academic research helped mainstream the “Yes, in my back yard” movement — the YIMBY movement — which has since gained many supporters and achieved some major legislative victories in recent years.

Rob Wiblin: Then last year, in 2021, Matthew also wrote One Billion Americans: The Case for Thinking Bigger, in which he argued that the US should aim to triple its population by making it more practical and appealing for people to have children, as well as dramatically increasing both skilled and unskilled immigration.

Rob Wiblin: Thanks so much for coming on the podcast, Matt.

Matthew Yglesias: Thanks for having me.

Rob Wiblin: I hope we’re going to get to talk about your views on foreign policy and how to use opinion polling. But first, as we ask everyone, what are you working on at the moment, and why do you think it’s important?

Matthew Yglesias: I work on a wide variety of subjects simultaneously, trying to publish five or more columns a week. But something that has been a theme of mine that I’m working on a piece on right now — one I think is going to continue to be an important subject and that’s very aligned with your audience — is trying to get people to think outside of the day-to-day hurly-burly of COVID-19 policy, and think more about pandemic prevention and biodefense issues as a theme that we should be grappling with on a forward-looking basis.

Matthew Yglesias: Because there will be more variants. There will be more viruses. Things could be worse in the future than what we’ve just been through over the past two years. And yelling about who needs to wear a mask when is ultimately less significant than, “Can we improve our readiness to prevent the recurrence of this kind of problem?”

Rob Wiblin: I’m going to have a few questions about that later on in the interview. And also a few listeners are curious to know what kind of voodoo magic, or what deal with the devil you’ve made to be able to produce so much content, as well as being so active on Twitter almost every day.

Autonomous weapons [00:04:42]

Rob Wiblin: But first off, you’re a really widely read guy who has the specialty of being a public policy generalist. You also tend to be pretty game to do your best to answer just any sort of random, hard questions. So I basically wanted to rapid-fire you a bunch of issues where I don’t already know your views, despite the fact that you have published and said a lot on other interview shows. And though you’re not an expert on these topics, I think you do have plenty of relevant knowledge, so I’m curious to know what you think. Does that sound good?

Matthew Yglesias: Yeah, let’s do it. Lightning round.

Rob Wiblin: Cool. Feel free to skip if you don’t have a solution to some of these quite bedeviling problems. What, if anything, would you like to see governments do about the growing capabilities and use of autonomous weapons?

Matthew Yglesias: This is not a super-deep suggestion, but it seems like you need to have some kind of international summiteering about it — in the way that the US and Soviet Union tried to do these arms reduction treaties about the stockpile of nuclear weapons — because countries are not going to unilaterally want to disavow autonomous weapons systems. But you can imagine an arms race in that area is even more dangerous than the arms race with nuclear ballistic missiles.

Rob Wiblin: The interesting thing about autonomous weapons is it seems to be actually somewhat leveling the playing field between richer countries and poorer countries — it’s allowing countries that don’t have as impressive militaries to potentially create a lot of trouble for more powerful nations. And it might be quite challenging to get them to give up that ability, especially as even the powerful nations don’t seem super keen to give up the option of autonomous weapons either.

Matthew Yglesias: No. This is a challenging problem. But that being said, there was a way world history could have gone after 1946, in which you would’ve said something similar: that just a small stockpile of atomic weapons gives medium-sized countries a great deal of protection against great power–bullying. It’s a very sympathetic-to-the-regime way of looking at it, but this is how the Iranians would say their nuclear experiments have gone. But there was a successful move to really sort of stigmatize that, and have a small number of countries monopolize the nuclear weapons, then have them negotiate with each other to try to reduce the amount that’s in existence. So I don’t think it’s in principle impossible to do, but obviously quite hard.

Rob Wiblin: The same is true actually of chemical and biological weapons, in the sense that the technological barriers there are even lower. And yet we have successfully managed to massively reduce their use, relative to some counterfactual history where they were totally normalized.

Matthew Yglesias: Yes. We don’t think that much about biological weapons these days, because I think we believe we’ve had a lot of success there, but it’s also something worth worrying about more. Obviously, the risk of biological warfare spilling over into the rest of the world is extremely high.

India and the US [00:07:25]

Rob Wiblin: Yeah. OK, which country do you think the United States should be doing more to cultivate a closer alliance with? Maybe one that it is currently neglecting a little bit?

Matthew Yglesias: I think this is a pretty obvious answer, but India is kind of the one that’s on the table. The US has historically been aligned with Pakistan against India in South Asian security crises. Now that we are not present in Afghanistan, we do not have the same kind of dependencies on Pakistan as we used to. And there’s a lot of alignment on values with India and strategic reasons for that as well.

Rob Wiblin: Do you have a take on why it is? I think the US is pursuing a friendlier, closer relationship with India, but it seems like it could be doing more. Do you have an idea of maybe why it’s not?

Matthew Yglesias: I do think that the involvement in Afghanistan has been a complicating factor in that for a long time. The Bush and Obama administrations, and certainly Trump, all made efforts to sort of draw closer with India. But India is very invested in its relationship with its neighbors. Pakistan is very invested in its relationship with India. As long as the United States had thousands of soldiers on the ground in Afghanistan, we needed to prioritize their safety and their security, and that really limited how far we could go with India.

Evidence-backed interventions for reducing the harm done by racial prejudices [00:08:38]

Rob Wiblin: That makes sense. What are some likely impactful or evidence-backed interventions for reducing the harm done by racial and other prejudices in the workplace? A listener submitted this one because you’ve recently written a bit about this topic, pointing out that there are some interventions that have been tested and found not to work. But are there any that you are more positive about?

Matthew Yglesias: One thing that we see is that actually some of the things that are done in the name of racial tolerance training are directly counterproductive. So not doing that would be good. And changing the legal liability standard that has generated those programs is one that we could do. Right now, at least in the US — I assume different countries have different frameworks for this — the understanding of businesses is that if you do something that is called “diversity training,” that that gives you a lot of protection from anti-discrimination lawsuits. So that reduces both the incentive to actually address racial discrimination among your staff, and also to put forward training programs that generate backlash and backfiring.

Matthew Yglesias: So, to develop a legal doctrine in which people are held accountable for discriminatory practices, and are not immunized by saying, “We did some training” — that creates a situation in which big businesses that have a lot of money to spend and have big legal teams have to say, “Look, we need to invest in research. Is there something we can do that will actually improve the situation?” Rather than what they’re doing right now.

Rob Wiblin: I guess you could go further, and say that it’s not enough to just put on programs that say that they’re anti-racism or anti-prejudice. They actually need to put on programs that have been demonstrated to have positive impacts.

Matthew Yglesias: Well, I think you measure by outcomes rather than by inputs. So if you’re going to be penalized for discriminatory practices, then it’s in your interest to come up with a way to have less of them happen, rather than being rewarded for having the program.

Factory farming [00:10:44]

Rob Wiblin: That makes sense. What should we do about the problem of animals suffering in factory farms? Other than the natural answer of supporting the development of alternative protein sources or meat alternatives?

Matthew Yglesias: I mean, the good news — if you want to call it “good news” — about factory farms is that they’re so bad that there’s a lot of room for regulatory improvements at the margin. And relatively small-scale changes — I think we’ve seen some states outlaw the worst kind of egg-laying crates, the worst kind of gestation stuff for pigs — so that’s an obvious approach. I think it’s the one animal rights groups are taking, but it seems really worth supporting to me.

Rob Wiblin: I spoke to Leah Garcés last year, and she pointed out that any regulations that you put in place to improve animal wellbeing on farms — that then increase the price of the products — actually causes some people to shift over to using the often more expensive alternative protein sources, which then builds that industry and helps it to reach a large scale. So potentially, there’s these flow-through R&D impacts that could be quite large.

Matthew Yglesias: It’s potentially a virtuous circle in which you are producing animals in more humane conditions. It’s more expensive to do it that way. There’s more incentive to switch. There’s more value in researching the other things, and also potentially more value on the animal husbandry side in finding cost-effective ways to raise animals more humanely.

Rob Wiblin: Feel free to skip this, but are there any specific animal welfare changes to farms? Or is that maybe getting too specific to be in your area?

Matthew Yglesias: It’s deeper into the weeds than I know. Something I’m sure is familiar to your listeners is just the crude fact that chickens are, by the numbers, the overwhelmingly predominant farm animals. So, even very small improvements in the welfare of chickens has an incredible aggregate impact. The person on the street has not thought that through in that exact way, but it’s actually very, very important if we can make chickens’ lives slightly better.

Wild animal suffering [00:12:41]

Rob Wiblin: I’m not surprised that I haven’t heard you asked this one on other podcasts, at least as far as I know. What should we do about animals dying of hunger, cold, heat, or thirst in the wild?

Matthew Yglesias: This is something that just got on my radar recently as even a topic that exists, and it kind of blew my mind. I may be wrong — and this foundational work may have been done someplace else — but from what I could see looking around in it, I feel like just the very basics, research-wise, need to be done here. Because it seems like wild animals are living in a fairly Malthusian kind of condition, and so you worry.

Matthew Yglesias: I saw one proposal to say that we should actually be vaccinating wild animals, so that they suffer less from disease. Which, that’s interesting — that sounded like an interesting thing to say. But then I started worrying about if you’re going to increase the number but make their welfare even worse by doing that? I honestly don’t know.

Matthew Yglesias: We have a lot of people who study wildlife; it’s a field that exists. But I think that the welfare of animals is not something that we have thought about a lot as a society versus just preserving their habitats. And really just to ask scientists to try to help us understand better what the consequences would be of less disease out there. Unfortunately, we can’t, as far as I can tell, ask predators to use humane slaughter methods. That would be a clear win, but I have no idea how you’d achieve that.

Rob Wiblin: I guess the thing that jumps out that you could try to do for wild animals, at least the larger mammals, is the same thing that has happened to humans: make the resources that they need more abundant, but at the same time have some level of restraint on reproduction — whether imposed or self-generated — so you don’t just get all of the gains eaten up by higher population. But how you do that, I’m not quite sure.

Matthew Yglesias: Right. I mean, I know there’s efforts — not for humanitarian reasons — to use contraceptives to control rat population sizes in urban areas. And that does not seem to have succeeded all that well. If it was working, if this was brilliantly reducing wild rat populations in Washington, DC, I would say, “Wait a minute. We should apply this with the wild animals.” But in the general subject of 80,000 Hours, this feels to me like an area in which we could use more smart people, not just working on the policy question, but actually, “What can we do here? Do we understand this issue? Do the technologies work?” And if they did, I’d be happy to say, “Let’s go out. Let’s go use them.”

Rob Wiblin: At least, “Let’s see what happens when we try it on a small scale.”

Vaccine development [00:15:20]

Rob Wiblin: What’s an area of science that deserves more funding than it currently gets? Even if that funding had to come at the expense of other government-funded science?

Matthew Yglesias: We were just talking about one that I think is interesting here. We have seen that vaccine development is a very promising area. And I have been a little taken aback to learn how little real effort goes into that on a day-to-day basis. There are scientists who believe you could develop vaccines that target whole virus families. And they have some money — they have jobs and things — but the advance market commitments that were made to help spur the mRNA vaccines for COVID, that seems very useful. To tell people, “Look, if you make this, we will buy seven billion doses of it.” That might involve spending no money, but it might involve spending a very large amount of money. And it could help sort of drive private-sector dollars and innovation in a very useful way.

Rob Wiblin: Yeah, we’ll stick up a link to the Wikipedia entry on advance market commitments, and why they’re potentially really useful for spurring innovation at the commercialization level.

Anti-aging research [00:16:27]

Rob Wiblin: Should we direct more biomedical research to slowing down the rate at which people age?

Matthew Yglesias: I’m not against that. That’s one where I have trouble seeing why the private market wouldn’t deliver on there. I feel like there would be a lot of money in an anti-aging pill — you know, targeting the developer.

Rob Wiblin: That’s a good point.

Matthew Yglesias: Again, if somebody tells me, “I’m working on anti-aging research,” I’m not going to be like, “What the hell, man? That’s horrible.” But I think the basic biomedical issue tends to be that we have a lot of private R&D for problems that are common and that impact rich people — so baldness, aging in general, things like that — whereas malaria is a big deal, but the people it affects are mostly poor. And so we could really use more public sector or philanthropic money there. So I don’t know. I mean, I will change my view on that if someone tells me there’s a huge market failure there, but it seems like something that commerce should address.

Rob Wiblin: It’s a good point. My guess is that the anti-aging folks might say something like, “We really are able to make massive progress on this in our lifetime, and might be able to get to a product that’s broadly useful, but it is decades away and so anything that we research now or patent now would have expired by the time you can actually make a product.” So you kind of have this gap between the basic research and the application, where we need external funding in order to get past that.

Matthew Yglesias: A broader question is our regulatory framework for clinical research in general, which is something that I’ve gotten quite concerned about. There’s a good book by a guy named Alan Wertheimer, called Rethinking the Ethics of Clinical Research. This is a philosopher who hadn’t been a bioethics guy, and then, at the end of his career, he did a fellowship at the National Institutes of Health, and he wrote this book. And it’s basically like, “What the hell, guys? Why aren’t you applying the normal ethical concepts of consent to these research things? Why is there this much of a higher standard in the bioethical world?” He’s not even saying you should be the most crass version of a consequentialist you could imagine, that’s like, “Maybe we can do secret experiments, but it’s for the greater good,” but it’s like, “Why can’t you accept the normal standard of ‘voluntary’?”

Rob Wiblin: That we use almost everywhere.

Matthew Yglesias: Right. If you want to get somebody to go work on a fishing boat, you’ve got to pay them more than for other blue-collar jobs, because it’s more dangerous. But if you want to do it for money, that’s a perfectly legitimate career option. But that’s not true for clinical studies. And there’s a lot, separate from the funding side. I do think that we could be making faster progress on all the biomedical fronts if we relaxed the standards for what constituted “voluntary.”

Should the US develop a semiconductor industry? [00:19:13]

Rob Wiblin: Yeah. Should the US do what it takes to develop a domestic semiconductor industry, so it isn’t so dependent on places like Taiwan or the PRC?

Matthew Yglesias: Yeah. I think the expected value there is positive. I’ve read this article in The Economist that was really dumping on the ideas of US and EU semiconductor production. They were like, “It’s going to lead to wasted money and overcapacity” — but how bad is that? Say they’re probably right. Probably it’s fine. Probably it’s a waste of money. That doesn’t sound so bad, versus how a minor disruption to semiconductor production has been a big problem over the course of this pandemic. You could imagine that getting way worse. I mean, an actual shooting war between China and Taiwan is not that unlikely in the scheme of things, and preparing for that in different ways seems easily worth doing.

Rob Wiblin: It sounds like that article might be slightly missing the point, which isn’t that doing this is economically efficient in the median case — it’s a variance-reducing play, which is expensive typically, but you’re buying insurance against a bad outcome.

Matthew Yglesias: I like The Economist. I appreciate free markets — I sometimes scold progressives for not doing enough to appreciate free markets. But I think we know that financial markets do not adequately insure against low-probability, high-severity risks. That’s true of truly existential risks, but also just of any kind of low-probability events. You can’t get insurance against massive disruption of the semiconductor industry. So then there’s no investment in the hedge against that, and we need governments to reduce downside tail risk. It is worth some short-term efficiency loss to do that.

Rob Wiblin: I actually think maybe the best argument against developing the ability to make these kinds of products locally is that the dependence of the US on China is actually really good, because it reduces the risk of war — that we should want the US to back down, because it would be so bad for the two countries to go to war. In fact, it’s perversely good that the US can’t stand on its own two feet. But I mean, it’s an interesting and slightly counterintuitive argument — it sounds a bit too clever by half.

Matthew Yglesias: Well, that was Norman Angell’s argument before World War I. He said we have so much economic interdependency that the costs of European great power conflict would be so high that it can’t possibly happen, because it would be totally irrational. I mean, that was correct: World War I was very costly — everybody ended up much worse off as a result of that. But it didn’t mean it didn’t happen. So I hesitate to put too many eggs in that basket.

What we should do about various existential risks [00:21:58]

Rob Wiblin: Let’s partially move on and talk about existential risk–related policy issues specifically for a little bit. In a recent review of the film, Don’t Look Up, which is on Netflix, you had this great bit that just made my eyes light up. It was about major threats to humanity’s survival. If you’ll forgive me, I think it’s worth reading it for listeners, because I just love it so much.

Rob Wiblin: “I don’t want to tell you that there has never been a story in the mainstream press about supervolcanoes, but there really aren’t very many. There’s no mainstream constituency at all to fund a larger scientific effort to understand supervolcano risk and how to mitigate it. And a candidate for office who goes on ‘Meet The Press’ to say one of his top priorities in Congress is improving US efforts to tackle supervolcanoes would be roundly mocked.

Rob Wiblin: “I don’t think this is because journalists are bad people who want us all to die in a spectacular accident. But we’re a bunch of apes who’ve evolved to be very attuned to the machinations of high-status apes so that we can navigate the factional landscape effectively. The reporters and the editors and producers they report to are apes, programming for an audience of apes. So you get a lot of stories about who is fighting whom and through what means. […]

Rob Wiblin: “By contrast, on something like Covid-19’s origins, we’ve had a decent amount of coverage of the lab leak controversy but essentially no coverage of what is being done to prevent future lab leaks (basically nothing) or to prevent future zoonotic crossover events (again, nothing). […] But again, either way, we’re not doing anything to counter either route for transmission, and that (shocking! alarming! insane!) fact gets way less attention than the latest round of ‘who’s yelling at whom about masks?’”

Rob Wiblin: So normally I dedicate a lot of the interview to kind of fleshing out exactly that issue about public rationality. Except we’ve already covered it a fair bit in 2021 — with guests like Carl Shulman, and Ezra Klein, and Andrew Yang — so that might feel like a bit of a repeat for listeners. But is there anything you wanted to add to or highlight from that extract?

Matthew Yglesias: No, you’ve had a lot of people on the show talking about this. I think it’s great that there’s a community that is talking about this, and the only real question is how do we carry that message forward to more people? And it’s why I liked that movie, despite some of its problems as a climate analogy. I want to encourage more people to do pop culture about existential risks. I think that’s a meaningful way of increasing engagement.

What governments should do to stop the next pandemic [00:24:00]

Rob Wiblin: So in lieu of discussing why politicians don’t make x-risks part of their policy platform, which I guess people probably do understand, I wanted to give you a chance to mention any specific things you think might actually be worth doing about these various huge risks. Obviously this is super challenging, and I’m asking because often I don’t have a clue — so if you don’t have any ideas, feel free to pass.

Rob Wiblin: What would you like to see governments do to stop the next pandemic getting started at its source, whether that is from animals or from labs?

Matthew Yglesias: We talked a little bit about vaccines and advance market commitments, but I think that we should be trying at least to develop these family-wide vaccines. That can allow us to get ahead of viruses that literally don’t exist right now, because we know something about the parameters in which they might emerge. We also do have to look at not the lab leak per se, but I think the prevalence of gain-of-function research is very dangerous. We should be pursuing efforts to sort of tamp down on doing that.

Matthew Yglesias: There’s also mid-range stuff on surveillance and detection that could be useful. John Hickenlooper was talking in a Senate hearing recently about what we would need to do to scale up genomic surveillance — basically look at the wastewater of different American cities and be kind of constantly scanning to see if there is anything new popping up in here. The technology to do that exists. It costs some money. It’s the kind of thing where it would be wasteful on one level: you’d be running a lot of sequencing of people’s poop and not finding anything in it. But it’s good to put that infrastructure in place before you actually need to use it, so it could be available in somewhat of a timely manner.

Rob Wiblin: Yeah, I’ve only vaguely followed this story, so maybe you can correct my understanding here, but I’ve seen Dr. Fauci giving congressional testimony, claiming that the US doesn’t fund gain-of-function research. But it seems to me like it’s kind of a semantic game, where he’s claiming — based on some very narrow technical scientific definition of exactly what gain-of-function research is and what it isn’t — that they aren’t.

Rob Wiblin: But the thing that we’re actually worried about is making viruses that could cause a massive pandemic. And they are doing that — or potentially they’re funding research like that — even though it’s not technically, by their definition, called “gain-of-function research.” Is that your understanding?

Matthew Yglesias: Yeah, there’s definitions splitting on both “What does it mean to be funding something?” and “What does it mean for something to be gain-of-function research?” But I think that there is a philosophy that trying to make more dangerous pathogens is a very important way to develop defenses against the pathogens. There’s always been this ambiguity in the biowarfare space between you’re not allowed to try to develop offensive weaponry, but you are allowed to try to develop defenses. So there’s a strong military incentive to essentially smuggle offensive weapons development under the heading of, “Well, we need this for defense.”

Matthew Yglesias: I think that we should be looking at that telescope through the opposite end of the lens — really trying to clamp down on what we’re doing there, and then develop other ways of thinking about defensive research. If you could develop defensive technologies that don’t depend on you understanding the most small-grained details of the virus’s protein arrangement, then you don’t have the excuse that we need to be constantly fiddling with them in order to see if this stuff works.

Rob Wiblin: Yeah. I think it’s interesting that in March 2020, people were really up in arms about this issue that everyone was assuming that COVID had come from a wet market in Wuhan — that it was because of people having all of these live animals in a market, and then people eating them, having close exposure. People were incensed by this, understandably, and talking about how we could get rid of it. As far as I know, nothing’s been done and it’s completely disappeared from the agenda. Like, how much would we have to pay China to shut down the wet markets? How many billions of dollars would it cost? And wouldn’t that be worth it?

Matthew Yglesias: There’s been this extreme literalism of this debate, where people felt sure that it had come from a poorly supervised wet market and were really fired up about shutting them down. But the reason they thought it came from a poorly supervised wet market is that ex ante that seemed very risky.

Matthew Yglesias: Now subsequent analysis has cast doubt on the idea that it actually did come that way. But the whole premise that that’s a risky thing to be doing, that’s still true, right? So now the Chinese are in a blame-shifting exercise — they have some frozen food theory, and I think that that’s wrong from what I’ve heard from people with technical capabilities. But as a forward-looking question, that doesn’t matter, right? Clearly, having a very wide array of live animals in very close proximity to each other and close proximity to urban areas — with creatures being brought in from all different parts of the world — that is dangerous. That is biologically dangerous.

Matthew Yglesias: It’s not that high-value to China either. There are things that it’s hard to get cooperation on because they’re so important, but the Chinese government has fitfully tried to marginalize these things over the years for different reasons. And then it’s culturally sensitive issues, so they get pushback, et cetera, et cetera. But it’s frustrating that you can’t get a real move there.

Rob Wiblin: I was just thinking, what could actually be done? I wonder, could the US just say, “Look, give us a list of everyone who works in every wet market in China, and if they shut down their thing and don’t open up a new one, we’ll give each one of them a million dollars.” Seems like it would be an absolute bargain. I don’t know, maybe a million is too much. At that price, it wouldn’t be worth it. But what about $100,000? Would they be willing to go get a different job for that?

Matthew Yglesias: Right, exactly.

Rob Wiblin: I guess this is the kind of thing that economists think is very sensible, but then you just know in your gut that this could never happen for some reason.

Matthew Yglesias: Unfortunately, it’s not a super cooperative environment between the US and China right now. This is again one of these things that’s understandable but unfortunate, that there is a lot of psychology about not looking weak happening in different places. So the question of who is making a concession to whom feels very significant to everyone, and that makes it hard to take any kind of concrete action — even though really the stakes here are not that high, and people’s interests seem fairly aligned.

Matthew Yglesias: I do think that the most likely way to get Chinese wet markets cracked down on is for the Chinese to do it on their own — unilaterally, not under pressure from the United States — just because they decide they would like to. For them to look like they were backing down is not something they would want to do. And for Western countries to seem like they were bribing China is not something we would want to do. So I hope Xi Jinping listens to your show, and is just going to do the right thing here.

Rob Wiblin: Yeah. It is interesting that perversely neglecting and not talking about this issue might be the best way to solve the problem.

Matthew Yglesias: You never know, right? As a person who likes to talk about things, I would always like it to be the case that shining more attention on a topic is the solution. And it is for some problems, but unfortunately not for everything.

Comets and supervolcanoes [00:31:30]

Rob Wiblin: Yeah. What, if anything, would you like to see the government do about comets and/or supervolcanoes?

Matthew Yglesias: NASA poked around on the supervolcano issue a few years ago, and they seem to have the idea that you could try to cool down the magma underneath Yellowstone with essentially injecting water. That’s related to the idea of advanced geothermal as an electricity-generating concept.

Matthew Yglesias: So there are these regulatory sensitivities around geothermal drilling on federal lands — a ton of sensitivity about doing it in an actual national park, which is where Yellowstone is. But I think that sacrificing a portion of the park in order to not have literally the entire planet explode would probably be a win-win. And we could offset that with some more parkland someplace else. Again, as you would expect, it’s not totally clear that that would work, but it’s worth investing some money in the exploration of whether or not it works. There’s a few other supervolcanoes that are out there.

Matthew Yglesias: As I understand it, after Deep Impact and Armageddon came out, we actually got the government to track asteroids better. And comets are just a little bit harder to track: they come in at a sharper angle and they go further away. But we should put some more telescopes up there and try to find them.

Matthew Yglesias: There’s a lot of nostalgia for the kind of heroic age of NASA and space exploration, which was very motivated in its heyday as a kind of national defense imperative against the Soviet Union. Today I don’t think that you can very plausibly claim that we need to go to Mars to stick it to China. But there’s a very clear national defense rationale for comprehensively tracking objects in outer space as far away as we can.

Matthew Yglesias: The amount of money that’s spent on defense programs is very, very, very large, and tracking those comets is good — especially as we have more private sector interest in some of the sort of sexy low-hanging fruit of, “Let’s get some human beings and have them go around the Earth in a circle.” You know, good for the entrepreneurs. And there’s not a huge ROI financially in tracking comets, but socially it’s very valuable.

Rob Wiblin: Just a few weeks ago, we saw the James Webb telescope go up, and obviously it has an unbelievable ability to detect objects that are extremely far away — in this case, billions of lightyears away. So I don’t understand the technology, but I figure there must be some way to turn money into the ability to see comets that are further away than what we currently do, if we really set our mind to it.

Matthew Yglesias: I think that this is just a question of how much of the field of view are we actually looking at, at any given time. You need more than one telescope and they’re expensive.

Nuclear weapons [00:34:25]

Rob Wiblin: Any ideas for how to reduce the risk of a war with nuclear weapons? Or reduce the number of weapons that are used if one does occur?

Matthew Yglesias: Not really. [laughs] I mean, we got to do our best. We’re probably going to talk about my sometimes-unpopular opinions about Russia out there, but you’ve got to ask yourself with any kind of international conflict: is the juice worth the squeeze? And conflict between the US and Russia is very dangerous. It’s worth asking yourself what the stakes really are, and what the upside to some of these things is.

Matthew Yglesias: US and China is the more hot geopolitical topic. China does not have that many nuclear weapons right now. We should be keeping in mind: how can we keep it that way? And anytime the US and Russia can agree to reduce our nuclear stockpiles, that has a safety benefit — in terms of that bilateral relationship, but also in terms of reassuring the Chinese that they do not need to embark on a crash program to drastically increase the number of nuclear weapons they have. Theoretically, the major nuclear powers committed to disarmament some time ago. They’re not actually doing that, but anything that’s done that counts as baby steps in that direction is useful.

Advances in AI [00:35:46]

Rob Wiblin: Do you think we should do anything more now to prepare domestic society or the international order for the possibility of major advances in AI over the coming decades?

Matthew Yglesias: We definitely should. The question is always what, right? As the person who asks the questions on podcasts, I’ve had people from the effective altruism and longtermist world on the show. Or just talked to them informally, to be like, “What should I do? Say I decide that I take AI risk extremely seriously. What do you want me to say about this?” And their answers always strike me as fairly fuzzy, so I’m left at a bit of an impasse.

Matthew Yglesias: Now again, to the young people out there, if you are technically literate and you’re in your early 20s or late teens and wondering what to do with your life, there is a lot of demand for good ideas about incentive-compatible artificial intelligence — how to make this maximally beneficial rather than threatening to humanity, on a policy level. I’m in the market for hot takes.

Rob Wiblin: Yeah, I think that there are policy ideas coming up through the academic conversation, but it does still feel like the conversation is pretty preliminary. And they seem fairly reluctant to go out and publicly advocate for anything in particular, which is interesting. I suppose they just worry that their ideas could do more harm than good a lot of the time.

Matthew Yglesias: That’s definitely a risk that’s out there. And I worry sometimes about how you have communities, or you have schools of thought — and that’s good. But sometimes something can become a scene. It’s like to be “in” in part of the tribe, you’d be like, “Oh, I’m really worried about AI existential risk.” OK, but “being worried” about things doesn’t accomplish anything in life.

Matthew Yglesias: And you can just kind of dismiss other problems. Say, like, “Well, don’t worry about that — worry about the AI risk.” But is there a tradeoff? Is there something that a typical person — a typical elected official even — should be doing differently? I’m not sure. I’m not saying no, but I haven’t been super convinced, in the way that I have on pandemics or even supervolcanoes. There are these kind of fiscal tradeoffs in the NASA budget, in NIH — what they care about that really addresses existential risk on those topics.

Rob Wiblin: Yeah. 80,000 Hours to some extent exists to try to solve this problem. I guess we have good ideas, as you were suggesting, for what you should do if you’re early in your career, or you’re willing to fully change your career to go and become a domain expert in a particular policy issue or some particular technical problem. But it has been a lot harder to come up with one like, “You’re a 40-year-old generalist who has a lot of Twitter followers. What should you do?” We haven’t quite nailed that one yet.

Matthew Yglesias: It’s a tough one. I’m pretty useless. I don’t know.

Surveillance systems [00:38:45]

Rob Wiblin: [laughs] What would you like to see governments do about the threat that surveillance systems or technologies could be abused for oppressive or authoritarian purposes in the future? Including in countries where it’s not obvious that they definitely will be used that way.

Matthew Yglesias: Is that the future or is that the present? I’m not really sure. What should governments do about it? I don’t know. I tend to think that we sometimes — in the United States at least — over-index on “Let’s not use this technology for good, because in principle it could be used for evil.”

Matthew Yglesias: I see what people are saying about that. So in the UK you have a lot of CCTV cameras around everywhere, and the police can use that footage to try to catch murderers. And that’s very stigmatized in the United States, because it’s seen as a privacy risk and an oppressive government could use that technology to do really bad things. I never find that all that persuasive. If the oppressive government takes over in the future, I don’t think it’ll be that hard for them to install the cameras.

Matthew Yglesias: I don’t want to say it’s an overrated problem, because it’s obviously quite a serious problem. I think the PRC has developed a level of oppressiveness that is beyond the capabilities of 20th century states. We know 20th century totalitarianism was worse than the worst governments of the 16th century because they had more capacity. So it’s a big issue, but I don’t know that having liberal states refuse to develop state capacity is a good answer to that. On some levels, we need to be as capable as we can: we want to show that liberal states can succeed and can deliver good governance in an effective way and not just lose out.

Rob Wiblin: I can see what you’re saying about not deploying the cameras — it doesn’t super help, because a malicious government can just buy some cameras and stick them up. But it seems like it’s worth imagining in like 10 years’, 20 years’, 30 years’ time that the US government, the executive, was trying to use the things that we imagine they’re going to have to oppress and suppress their political enemies.

Rob Wiblin: What other things could we do to make it more difficult? Like legally, potentially, to slow this down? You’d have to wait for the judges to die and be replaced in order to be able to get something through. Or if you don’t control Congress — Congress has passed these laws in the past — making it very difficult to do X, Y, and Z, in order to buy time. That potentially seems worth doing.

Matthew Yglesias: Yes, I’m for civil liberties, like normal people have and do. Americans are very invested in the idea that a certain kind of proceduralism is safeguarding our liberties. And if you look cross-sectionally at the UK, at Canada, at Ireland, at Australia, New Zealand — I don’t actually find that very compelling, that all the countries that have this kind of Anglophone cultural heritage have sort of similar social values, sort of similar outcomes. We actually have wildly different baseline political institutions.

Matthew Yglesias: The US and Britain diverged incredibly sharply in the late 18th century in our concept of what to do about the king oppressing you. Britain went all-in on parliament — “We should democratize the parliamentary elections. We should empower the House of Commons” — and you now have this hyper-sovereign unicameral legislature. We went the other way: we’re all-in on counterbalance theory, and our version of the House of Lords is super important, we have all this federalism, et cetera. I don’t know that that stuff has given us any more practical liberties or a real safeguard against tyranny on a practical level.

Rob Wiblin: Maybe the case is clearer in the UK. I guess you’re saying that our protection is actually the values of people in the society: that an individual person who wants to abuse technology this way won’t be able to, because people will stop them because they don’t like it. And that makes a lot of sense.

Rob Wiblin: But what would you do if you wanted to add more safeguards? In the UK, it seems like Parliament does just have a lot of authority to do pretty crazy stuff if you could get a majority to go with it. I might feel happier if there were more safeguards to stop Parliament really going off the rails. I think Parliament probably won’t go off the rails, just because of the culture of the UK and of the UK political class, as you say. But I don’t know whether I want to bank everything on that.

Matthew Yglesias: Yeah, we’ll see.

How Matt thinks about public opinion research [00:43:22]

Rob Wiblin: Yeah. OK, let’s move on and talk about how you approach using polling and other indications of what voters think to shape the ideas that you advocate for. This is another topic where a few of my followers thought that you had the wrong idea, or at least the wrong idea in specific cases. When you’re researching a problem, like climate emissions or childhood poverty in the US, how do you personally use indications of public opinion?

Matthew Yglesias: Oh yeah. This is one of these things where you wind up sometimes in life getting backed into lines of work that you didn’t expect you would be in. And I should maybe reevaluate. But something that I have found is that the world of public opinion research is kind of bifurcated — or trifurcated, if that’s a word.

Matthew Yglesias: On the one hand, you have people doing polls to go to the media, for public consumption. Then you have people doing polls to issue advocates, who want polls that say their positions are popular. And then you have people doing polls because they’re trying to win campaigns. Survey methods are the same across these things, but what your incentives are and what you actually care about are pretty different.

Matthew Yglesias: Unfortunately, for the mass public, to get a good read on public opinion — like a rigorous look at how things are without putting a lot of English on it with question wording — you do have to talk to the people who are doing private work for political campaigns. They are the ones who have the strongest incentive to get the question right — not in terms of the survey sample, but in terms of what questions you ask people.

Matthew Yglesias: I know there’s a couple guys on Twitter who are very angry that I say carbon taxes are unpopular.

Rob Wiblin: Yeah. That was one of the people who put this on Twitter.

Matthew Yglesias: They’ve got their polls that say it’s popular, but we know in America that when this has been put to ballot initiative in Washington State, it’s lost badly, twice. We also know that in Europe, where the macropolitics is greener than in the United States, the European elected officials act as if stiff carbon pricing is going to be very unpopular.

Rob Wiblin: In what way? They don’t talk about it?

Matthew Yglesias: And they don’t do it, right? Because they don’t —

Rob Wiblin: Isn’t it that the EU carbon market is… The price per ton of carbon isn’t that low now, right?

Matthew Yglesias: It’s pretty low. And they do all this other kind of green stuff, right? And their opposition parties aren’t winning elections on the basis of these things. In Australia, when the government tried to put a carbon tax in place, there was a big backlash to it.

Matthew Yglesias: So people say that if you do the best kind of issue surveys, where you do partisan frames, you say, “Some Democrats are proposing a blah, blah, blah, price on carbon, which they say will reduce climate change in the most cost-effective way possible. Republicans say it’ll raise the price of gas by…” — then you give them a real number, not a crazy lie: “It’ll raise the price of gas by so many cents, raise the price of electricity by so much.” It becomes incredibly unpopular in that framework — like, less popular than commercial legalization of heroin.

Matthew Yglesias: When the price of gasoline spiked in the United States a few months ago, it was a huge deal politically. People were losing their shit about it. Probably the best media poll on this is a Reuters survey where they asked people, “Should we take drastic action to stop climate change?” Most people said yes. Then they asked, “Would you be willing to pay $100 a year more in taxes to stop climate change?” People said no. A hundred dollars a year is not that drastic, you know? I spend $100 a year on things that I don’t think are important at all.

Rob Wiblin: Yeah. It’s 0.2% of the average household income or something like that.

Matthew Yglesias: Right. I don’t want to say it’s nothing, but I mean, I’ve given $100 to charitable causes that I don’t even think are very good causes, just because somebody asked me to and I had a relationship with them.

Matthew Yglesias: So I think that it’s tough. As a professional journalist, the kind of thing that I can do is ask people to tell me things that they cannot say in public, and then I can try to triangulate the things that they have told me against publicly available information — like this Reuters thing, like the Washington polls, like the fact that practical politicians don’t campaign heavily on green tax shifts — and try to make people sort of see the truth. But it’s hard. I never know, because I’m not a huge scoops guy, and if you want to believe that I’m lying to you or that my sources are lying to me, it’s challenging to prove otherwise.

Rob Wiblin: I see. So the issue is that you’re getting what you think are the most reliable indications of public opinion, and also where public opinion would go if this became a live political issue, from people who are working on campaigns. And the polling data is private or the research is private, so you can’t share it. Sometimes it conflicts with perhaps advocacy polls that have been put out, or perhaps less high-quality public polling from Gallup, where maybe the question’s not quite right. Or they haven’t really tried to stress test how long will people really believe this, if it started costing money?

Matthew Yglesias: It’s the stress testing in particular that’s important. There’s a phenomenon called acquiescence bias, where if you ask people questions, they’re inclined to say yes. Researchers used to think that this had something to do with talking to people on the phone, and they were very bullish when internet polling first started. They thought they were going to get rid of this, but they didn’t. Which seems a little odd, because you could imagine that you don’t want to fight with the guy on the phone — you want to agree. But it’s quite common, and unless you give people the argument/counterargument, you can show that all kinds of things are popular, and contradictory things.

Matthew Yglesias: This is not breaking new ground — this is Converse’s classic work on public opinion. But people don’t have really deep and consistent views, which sometimes gets taken to mean that you could get away with anything in politics. But that’s also not true: there’s very consistent patterns of public backlash and resistance to things that raise middle-class people’s taxes — in particular to high-salience taxes, to increases in the cost of energy. You see that not just in the US, but really around the world.

Matthew Yglesias: I think a very underrated thing for American progressives looking at politics, is that we’ll look across the ocean and say, “Oh, these Europeans, they have these great healthcare systems. Why can’t we have one like that?” The answer in a lot of cases is that the healthcare systems were created decades ago, when the cost of providing them was much, much lower, and so it was done with small taxes. And once it exists, people become very attached to it. If the quality of the healthcare services degraded, they are in fact willing to put more funds into it for the sake of better services.

Matthew Yglesias: But because we didn’t adopt a universal healthcare system when Truman was president, or when JFK was president, we are now stuck with the present-day cost structure, and it’s just a much heavier lift to get people to embrace it. The 2022 cost of the National Health Service was not presented to the British public in 1948 — they got the 1940s cost, and it’s good for Clement Attlee.

Issues with trusting public opinion polls [00:51:18]

Rob Wiblin: Yeah. OK, I’ve got a couple of uncertainties about how this works in practice. I think one objection will be to say, you’ve got these different categories of polling. You read any individual poll and it’s really hard to tell what the motivation was of the person who originated it. Is the question quite right? What other followup questions were there? Could a different group have said a different thing? And you’re having to refer to these private conversations with people who won’t talk to most other folks, and then they might disagree as well.

Rob Wiblin: It just seems like this is a real mess to analyze. And that’s demonstrated by the fact that different people who have similar policy preferences ultimately, actually disagree about what they think the public might favor if push came to shove. Maybe this enterprise is a bit doomed, and instead we should just be advocating for what we think is substantively best, because at least we can do that.

Matthew Yglesias: I do think it’s a question of who “we” are. I have stumbled, I would say, a little bit ass-backwards into being a guy who writes a lot about polls and public opinion surveys. And it was not my intention, exactly — I don’t know that it’s super important for the typical person to be deeply, deeply invested in this. The main thing that I would like the person on the street to take away from it is that workaday politicians running for office are probably better informed about the state of public opinion than you are. And if you are finding yourself baffled as to why someone won’t say something or embrace something that you think they should, it’s probably because their surveys indicate that it’s not that popular, and maybe try to be a little less mad.

Matthew Yglesias: Now, what should you advocate for? Whether it’s you, or the listener, or whoever else, you probably should advocate for the right thing to do. And you should probably advocate for the right thing to do as if you were trying to be persuasive, which I think people oftentimes don’t do on the internet. I would say heavy consumers of political punditry spend a lot of time pounding the table on behalf of what they think is the right thing to do, being very impatient that other people in positions of greater responsibility aren’t saying exactly what they want them to say, and searching for bias-confirming information that indicates that they are right about everything.

Matthew Yglesias: Now, I don’t think any 80,000 Hours people are actually in that headspace.

Rob Wiblin: Oh, I mean, come on. I’m in that headspace. We were pounding the table earlier about the Chinese wet markets.

Matthew Yglesias: I mean, everybody is sometimes, but like…

Rob Wiblin: Yeah. We’re not the worst offenders there.

Matthew Yglesias: I don’t think it’s huge breaking news out here to tell people who listen to your show that people spend a lot of time on pseudo-arguments and confirmation bias. I try to be more of a rationalist than the average political pundit, more of an effective altruist than the average political pundit. But I really am a political pundit, whose life is mostly responding to other people in the punditry space — and there are a lot of bad arguments, wrong ideas, fallacious information going around on these subjects.

Matthew Yglesias: I’ve gotten into some pushing back on it, some level of audience interest in these kinds of ideas. But most of all, something that absolutely is just relevant to your main topics here, is that there are lots of ways of making political change that do not involve leaders of political parties pronouncing the right thing in the heat of a political campaign, in a high-profile media way. Right?

Matthew Yglesias: That is a thing that happens in politics: people are like, “Our campaign is going to be waged on X, Y, and Z, and we’re going to elevate that to the top of the discourse.” It’s an important part of life. It’s an important part of policy change, but it’s absolutely not the only way that policy changes. I think it’s only a good way to change policy when you are really, really convinced that public opinion is behind you.

Matthew Yglesias: Like we were talking about before, I wish the government would do something about supervolcanoes. I think it would be crazy for Joe Biden to give a State of the Union address where he is like, “My fellow Americans, there’s a 1 in 10,000 chance that a large magma deposit under Yellowstone National Park…” People would think that was bizarre, and he shouldn’t do it.

Rob Wiblin: I’m not sure whether that would help or hurt. I have no idea what effect that would have.

Matthew Yglesias: I don’t know either. I’ll put it this way. If the pros tell me, they’re like, “Matt, we did you a favor: we looked at that, and it sucks,” I think that’s fine. What I am asking them to do is see if they can’t have a conversation, staffer-to-staffer, in the Interior Department appropriations, where they get some money to say like, “Does this Yellowstone cooling idea work?”

Matthew Yglesias: And you want to find bipartisan cover: you want a Republican — ideally one who represents Wyoming — to be like, “Yes.” (I think Yellowstone’s in Wyoming, apologies if it’s someplace else.) But it’s a big problem, and someone should do something about it. Which does not mean it should be the centerpiece of a national political campaign.

Rob Wiblin: Yeah. Another general concern I have in this area is the idea that just a random person who you might phone up has opinions about the child tax credit or all of these other policy issues that someone somewhere cares about, and wants to know what people think. They have no opinion on so many things, just as I have no opinion on the great majority of topics, because I’ve never thought about them and I don’t really care about them.

Rob Wiblin: One of the reasons why it’s so hard to suss out what people think is that sometimes they just don’t think anything about it, and so we don’t know — until, I guess, it becomes a partisan conversation and we see how people react to news stories and so on. But maybe the whole thing of, “What is public opinion about X or Y” is kind of malformed in a sense, and it’s super unstable, and there is no real answer.

Matthew Yglesias: I think that there’s a level on which that’s true and a level on which it’s not true. One thing you can see is that we have referendums and initiatives pretty frequently in the United States, so you can relate how those referendum outcomes look compared to early polling on the thing. Academics who study this say that it tends to fall back to baseline — that really popular referendums underperform, and really unpopular ones overperform. So the argument kind of pushes people a little bit back toward the center, but only modestly so.

Matthew Yglesias: But if you do a good survey, that proposes that there will be an argument and says that there are two sides to the issue… It’s true that most people have not really given much thought to most issues, but their gut reaction to a pro and con argument is pretty predictive about where they’re going to be at the end of the day — precisely because they’re not deep reflectors on it.

Matthew Yglesias: Now, what would be harder to say is, “What would people think about this subject if they studied it intensively for months?” But they’re not going to study it intensively for months. What they’re going to see is some pro and con ads on television occasionally, that they do not engage with very deeply. Just as I’m sure there’s lots of things you don’t have an opinion on, but somebody who knows you well can develop a pretty good mental model of you. Or they don’t even need to know you that well, to be like, “OK, this is a longtermist, this is an effective altruist, this is a cosmopolitan — so how is he going to come out on this question that I am factually well informed about once he becomes better informed?” Because we can make these kinds of predictions.

Matthew Yglesias: You asked me at the top of this what my opinion was on a bunch of stuff that you didn’t know what my view was. But I doubt I really shocked you on any of those answers, right?

Rob Wiblin: Yeah. That’s true.

Matthew Yglesias: Because you have some information about the general argumentative landscape.

Rob Wiblin: You’re a somewhat unusual person though, in that you think about these things all the time, so you have a more structured framework on how you approach these questions. Whereas many people’s interests are not primarily public policy, so they’re perhaps more likely to form somewhat random opinions based on what they heard.

Matthew Yglesias: Well, I think what you see people default to is selfishness and short-term thinking, right? That’s the sort of core underlying constraint of public opinion in a democracy. A very valuable thing we can do in the world is push people culturally, personally, to be more broadminded and less shortsighted in their thinking.

Matthew Yglesias: This is why climate change is a difficult problem, because you’re asking people to make sacrifices for the long-term benefit, largely of foreigners. It’s not that hard to convince people that greenhouse gases are real or that it would be desirable to have climate change not happen — but it’s very hard to push people to make changes in their personal lives or to embrace policies that would force them to change. But if you can obscure the costs or do things that seem like win-wins, like we’re going to develop carbon capture technology. We’re like, “Oh, yeah, that’s great. It sounds good. Let’s go do that.”

Matthew Yglesias: And it’s exactly why it’s so much easier to get me to agree that factory farmed meat is a moral problem than to get me to stop eating it — because, I don’t know, I’m a bad person, you know? I try to be better. I eat less than I used to and I espouse a lot of correct positions. I like to think that I could bring myself through to vote yes on a ballot referendum that would make this more costly for me — that I am at that level of reflectiveness. But we are weak, pitiful creatures.

Rob Wiblin: This is reminding me, I have this kind of cached belief that middle-class people don’t want to hear about how good it is to raise prices on goods that they buy all the time. They don’t really want to hear about how they’re going to have to pay more taxes. Just for the same reason that all of us don’t want to have our income lowered. We don’t like to take a pay cut. We don’t like to take a tax increase.

Rob Wiblin: But I’ve also heard this other line of research that the selfish voter hypothesis is kind of wrong, because voters very often take positions that are largely expressive about communicating their values, because their views and how they vote makes very little difference to the actual outcome. So if you enjoy saying, “It would be good for taxes to be higher,” then that can massively outweigh the downside of the tiny possibility that you saying that could actually cause taxes on you to be increased.

Rob Wiblin: You can get people to take, in politics or in voting, very idealistic positions that maybe if they’re actually implemented, they’ll be quite pissed off. But they’re willing to express views that would be very costly to them in some sense. Have you read that literature?

Matthew Yglesias: Yeah. I think it’s always interesting how much things tend to flip when it becomes more concrete in a lot of ways. The evolution of political polarization in most Western countries has come to be that educated people are on the left politically, even though oftentimes they’re reasonably affluent, which is just contrary to how politics has traditionally been organized. So they will espouse lots of tax-increasing ideas that are not necessarily in the interests of urban, young, cosmopolitan, professional-minded people. And yet it has been challenging to actually get those kinds of increases enacted.

Matthew Yglesias: One of the ways that the Democratic Party in the US has evolved over the years is that Barack Obama promised to raise taxes only on people earning over $250,000 a year. Then to get a measure through Congress, he had to raise that ceiling to $450,000 a year. Joe Biden took over the $450,000 baseline, but he has struggled to achieve even that, and is looking now at a very, very, very narrow kind of tax base. I do think that that’s because the influx of more affluent people into the Democratic Party has made it more challenging to actually raise taxes, even though there is a lot of support for this kind of symbolic egalitarianism, and in some ways more support than ever for being mean to billionaires and things like that.

Matthew Yglesias: I’m trying to think of some other examples that you have of this. It’s common to meet people who are very worried about the environment and sustainability, but then if you tell them that gas stoves are bad — this is common in America, at least, to have gas stoves — they’ll freak out, and be like, “That’s terrible,” because it’s so concrete. People here are not used to the high-quality electric induction stoves and the old-fashioned electric coil stoves, but of course they work and you can cook food. They’re not as good, and it’s a big yuppy lifestyle signifier to have a gas stove — it shows that you’ve made it. When you actually propose taking it away from people, they get very leery.

Matthew Yglesias: So New York just made a rule that new construction is not going to be allowed to have the gas hookups. But they made sure that if you have existing construction, it’s not only that you don’t have to give up your gas stove, but you’ll be able to replace it with another gas stove. They’re trying to thread the needle there. People want to take action, and they came up with a way so that the action will not accrue to the people who are actually voting right now. Politics is tough in that respect.

Rob Wiblin: Yeah. The gas stove one is interesting, because it’s not only a global environmental concern, it’s also a health concern for the people in the house. It tends to cause a bit of respiratory problems for people to have gas stoves. It’s not so large that people tend to notice, but folks who have studied this have realized that it’s just bad for people’s health.

Matthew Yglesias: That may be the future — making a deterrent — because you could flip the script on that from “This is classy” to “You are indifferent to the lives of your children.”

Rob Wiblin: “Quit smoking.”

Matthew Yglesias: Yeah, exactly.

The influence of prior beliefs [01:05:53]

Rob Wiblin: I think some of what’s going on here is that I suspect that quite a lot of your reading of polling, and mine as well — how we interpret it — is very colored by priors that we have about what things people are likely to support and what things they probably won’t. When someone comes to me with polling saying that actually, a great majority of people support a carbon tax, I’m like, “No way, no way,” because I grew up in the debates about the carbon tax in Australia. I was in favor of it, but I could just see that the attack ads were really successful. People did not want to have their gas prices go up, and so I had this very strong prior that it’s going to be really hard in a real political campaign to get most people to vote in favor of a carbon tax that is going to substantially raise costs for them.

Rob Wiblin: Maybe I could be wrong about that, but I have all these perceptions about what can get through the political process and what can’t. I suppose people who have different priors about that could look at the same polling and come away with quite a different conclusion than what you or I might, because there’s just so many polls to choose from.

Matthew Yglesias: Yeah. I do think that that’s true. The other thing I will say is, this is something where I have seen a rapid evolution of conventional wisdom to positions that I think are wrong. That 15 years ago, everybody knew that raising the gas tax was unpopular, but also that it might be a good thing to do, and I could just say that in a story and people were like, “Yeah. No, that’s fine.”

Matthew Yglesias: Certain very specific progressive advocacy groups invested a fair amount of money in trying to convince people that their entire Democratic Party policy agenda is super popular, and it took real work to dislodge people from longstanding conventional wisdom. Because one group of people gets impatient with me because they’re like, “You’re being wrong, blah, blah, blah.” Another group of people gets impatient with me because they’re like, “Wait, everybody knows this.” And I’m much more on the side of “everybody knows this.”

Matthew Yglesias: I think that there was a fairly recent US-specific effort to create a bit of conventional wisdom that the American electorate really yearns for transformation into a European-style welfare state. That is not true, and most people have mostly been aware that it’s not true, and are sort of coming back to reality. I kind of hope to not spend the rest of my life writing and thinking about this subject, because I actually think it’s not that interesting.

Matthew Yglesias: You can see — if you do surveys on very abstract values about individualism, authority, religion, et cetera — that American public opinion is just somewhat to the right of European public opinion. Americans are more individualistic, they’re more religious, they’re more hierarchical. Not wildly so, there’s plenty of overlap, but discernibly. I think that’s what people have mostly thought from history. Maybe this is too wildly speculative, but I think there was a self-selection of more individualism-minded people to immigrate into the colonies and all of the settler states had lower taxes and stuff than the continent. I mean, I don’t know.

Rob Wiblin: Yeah, yeah, yeah. I do sometimes say to Europeans that it’s not surprising that Americans are more skeptical about government programs, because the government seems worse at delivering programs. Then sometimes people respond like, “Yeah, well that’s a deliberate scheme by people who don’t want the government to be doing more things, is to make it bad at doing things.” There’s people who are like that.

Matthew Yglesias: I mean, there’s definitely something to that, right?

Rob Wiblin: There’s some truth to that, yeah.

Matthew Yglesias: That’s an area that I find much more rewarding to think about and focus on: how can we improve programs, and make people like them better, and do something useful in life.

Rob Wiblin: Yeah. I don’t know this group that’s been trying to persuade people that the progressive policy agenda was super popular all along. Although many of them probably do sincerely believe that, and they would probably respond that there has been polling to show that depending on how you frame things — if you ask the question the right way, if you present these policies with the benefits in mind — then it’s easier to persuade people to support them than you might have thought. There’s a lot of latent potential support that this polling is bringing up.

Matthew Yglesias: Yes. Though I think the best version of the argument is almost the opposite: when you put things in place, they tend to be fairly durable. So there’s something to be said for a YOLO attitude to this kind of thing. Barack Obama’s healthcare plan was not very popular when Congress enacted it, there was a significant backlash to it in 2010. But then, when there was an effort to remove it, there was a backlash to that.

Matthew Yglesias: So there’s a tension in US politics right now between that line of thinking: you know, Biden and the Democrats should get done what they can, while they can, and who cares what happens next. And this kind of, “An authoritarian Republican Party is poised to extinguish democracy” rhetoric. I think that a lot of people on the left have an unresolved tension in their views of what they’re saying. Because saying, “The opposition is really, really evil” — that’s a leftist stance. But saying, “We need to be incredibly bold with what we do” — that’s a leftist stance too. And then you could square the circle by saying, “Well, being really, really bold is going to be overwhelmingly popular,” and so that’s very attractive, right?

Matthew Yglesias: It just relieves dissonance. And I think, as is the case with a lot of attractive dissonance-reducing beliefs, it’s not very well supported empirically. And there’s a tough judgment call to be made about how risky do you really want policymakers to be? I was a kid when Bill Clinton was president; I was a teenager, so I was aware of what happened. I’ve always been interested in politics, and he was super duper popular — and he also didn’t have that much in the way of great achievements.

Matthew Yglesias: It seemed to me at the time that that was a pretty disappointing way to live your life and way to be doing politics, but it’s hard to argue with the results, right? People really loved that. That’s a style of politics that’s pretty attractive.

Loss aversion [01:12:19]

Rob Wiblin: Yeah. We talked about some general preconceptions that we have, or beliefs that we have about what kind of policies are popular and which ones aren’t. One is that people don’t like to pay more money if that’s really salient. I guess another one is just that there’s a certain fraction of people who just don’t like changing stuff. Are there any other design principles that you might want to always keep in mind when designing a policy about what kinds of things end up being popular and not popular?

Matthew Yglesias: I mean, I do think that it’s not just the status quo, but it’s risk aversion. People dwell on the downside and loss. I shouldn’t say risk aversion — it’s loss aversion: people are more worried about losing what they have than about gaining something that they don’t currently have. So it’s hard for people on the left to persuade people that some new program is going to be amazing and it’s going to be worth paying the taxes, but it’s also hard for people on the right to win arguments that are like, “Well, without this program, we can cut taxes and that’s going to create better incentives to save and invest, and that’s going to increase growth by 0.002%.” When you compound that out for X number of years, people get a lot of like, “Eh, I don’t know about that.”

Matthew Yglesias: People are also just not great at thinking about the long term. If you ask somebody who’s 40, “Do you care about what’s going to happen to you when you’re 70?” They’d be like, “Of course. I’m not a crazy person.” Now you ask them, do they care what’s going to happen hundreds of generations in the future, they might say no. We might say, “OK, you discount more than you should.” But even when they say that they’re not discounting the future of their own life, I think they just pretty clearly are when they’re making policy decisions.

Matthew Yglesias: People are not in the habit of really thinking about their long-term interests, and we see that in all kinds of behaviors. It’s why there’s people who smoke. It’s why lots of people struggle with all kinds of personal health issues. And in their political thinking, it’s even worse. To really get people to think, “How’s this going to play out?” — it’s tough.

Rob Wiblin: Yeah. People talk a lot about means testing making policies more or less popular. So for people overseas, “means testing” I think is the term for whether you allow everyone to access a program, or only people who are below some income threshold.

Matthew Yglesias: Yeah. People argue about this a lot. I actually think if you pay attention to what everyone’s saying, they are not in that much of a tension really. If a program exists and everybody is using it, that makes it much, much harder to get rid of, versus a very narrow type of program — so there is a political durability benefit to some universalistic designs. At the same time, the broader you make the program, the more tax revenue you need to get.

Matthew Yglesias: And so, if we do comparative welfare state design, typically the really big welfare states have a broad tax base: they use a lot of payroll tax, they use a lot of value-added tax. Those taxes tend to be unpopular, so it’s harder to create a really broad universal program, but also harder to get rid of a really broad universal program.

Matthew Yglesias: What should you actually try to do? I don’t know that there is a general answer to that kind of question. I think the most important thing about means testing is how it actually makes the programs function — and that making everyone fill out 17 forms to verify their income is itself quite costly.

Rob Wiblin: To try to wrap up this section, the main trap that I worry about falling into myself — and that other people I think should worry about falling into — is I think the Pundit’s Fallacy. I’m not sure whether you came up with this or whether someone else came up with it, but it’s the delusion that the policy that you think is best on the merits of its welfare outcomes is also the most popular politically — the thing that’s going to win elections.

Rob Wiblin: People really want to believe that, but it’s not always true. In fact, it’s very often horribly not true. If someone’s worried about falling into that trap, but they do believe that a policy that they think is really good is going to be popular, what should they do in order to double-check whether they’re deluding themselves or not?

Matthew Yglesias: I think that the best thing to do is to try to look at real-world political battles that have happened — maybe in the past, maybe even in other countries. You were saying you lived through the carbon tax fight in Australia. I saw that happen. I paid some attention to it as it played out, and I think it would be hard to have watched that and then be like, “Oh, but in America, the situation’s wildly different.” I mean, the situation in America is different, but I can’t think of any relevant way in which it’s different, and it would just check you.

Matthew Yglesias: I’m glad that the Labour Party tried that, because I think it was the right thing to do. I also think it gave us useful information. It’s good, actually, for smaller countries to be more experimental in their political outcomes, because it helps the bigger countries understand what’s going on. And the decisions we make are ultimately more consequential in the world, just because we’re larger.

Matthew Yglesias: I think that that’s a good place to look for relevant evidence, and then just be skeptical. Just know that, of course, you want to believe that what’s compelling to you would be compelling to everyone — but try to be reflective. Now, if you happen to be someone who is a big-time swing voter, who has a robust record of —

Rob Wiblin: Real middle-of-the-ground views, maybe.

Matthew Yglesias: Or just feels cross pressure, for whatever reason. If you can say honestly that you have voted for different parties multiple times and change your mind a lot about which party to support, then introspecting probably does tell you something realistic. But if you’re not, then trying to make up an idea of what you think someone who is like that would like to hear, that’s pretty unreliable.

Rob Wiblin: It’s challenging.

Matthew Yglesias: I think that that’s really where people get into this fallacy territory: they have trouble empathizing with people who genuinely flip between the parties, because that seems so odd. In America at least, the vast majority of people aren’t like that. It’s something like 90% of the population are on one side or the other, hardcore partisans, and it’s a tiny number of flip-floppers who are determining the elections.

Matt’s take on military adventurism [01:18:54]

Rob Wiblin: Let’s push on and talk about your general view on what I’ll call “military adventurism.” You’re not a foreign policy specialist in particular, but you did take an interest in it, as most of us do. And in 2008, you wrote this book Heads in the Sand: How the Republicans Screw Up Foreign Policy and Foreign Policy Screws Up the Democrats.

Rob Wiblin: Last year, you wrote pretty forcefully in favor of the US withdrawing from Afghanistan, and you continued to support it even after it became clear that the result would be the Taliban taking over the country very, very quickly. More recently, you said you think the US should promise not to enlarge NATO or deploy troops or military equipment in Eastern Europe, if that keeps Russia happy and less likely to attack countries like Ukraine.

Rob Wiblin: I think it’s fair to say that when it comes to overseas military intervention, you’re pretty skeptical — though, like I do, you favor coordinating and collaborating with other countries in lots of peaceful ways. I basically asked my Twitter followers, “People who agree with Matthew Yglesias most of the time, what do you disagree with him about?” And this unwillingness to use military force was a top pick, with a whole lot of people saying that they were a lot more likely to be keen to have a go at using military force to do good than you do. So I’d like to talk about why you take the stances you do on foreign policy, and where those could conceivably go wrong.

Matthew Yglesias: Sure. The one thing I do want to say is that I feel like there is a tendency in the more hawkish school of thought in the United States to just completely want to ignore tradeoffs of kind of aggressive behavior, and to just focus much too much on aligning yourself with being right on certain kinds of things.

Matthew Yglesias: So right now, we are worried about Russia going into Ukraine. Previously, we were arguing about whether the US should try to sustain its military position in Afghanistan. A few years before that, we were worried about whether Russia was being too aggressive in going after Ukraine. After Russia invaded Crimea, the United States put some very severe sanctions on Russia to punish them for having done that. Russia reacted to that by closing off the northern supply lines — through which NATO was moving military supplies and equipment into Afghanistan — through Russia, which is north of Afghanistan.

Matthew Yglesias: Russia has long-standing strong interests in Afghanistan, which they perceive as being part of their neighborhood. At the time, nobody took that particularly seriously in the American policy community. It was seen as, we were doing the right thing — we were standing up for Ukraine. But by closing that northern supply line — which supplied “only about 15 to 20%” of NATO supplies into Afghanistan — it made us completely dependent on Pakistan for sustaining our bases there.

Matthew Yglesias: But if you go all the way back to George W. Bush’s memoirs and Barack Obama’s memoirs, American policymakers had always, always said that the Pakistani government was playing a kind of double game in Afghanistan — in which they were both our key allies, but they were also supporting the Taliban. So by deciding that we were so upset about Russia seizing Crimea, I think years ago we made a decision to make winning in Afghanistan, in effect, impossible.

Matthew Yglesias: By the time Joe Biden is president, he’s faced with this question: do we have thousands of troops indefinitely in Afghanistan? Not to beat the Taliban, not to stabilize the country, but to prolong this war. He said no. I agreed with him no. But the people who would be on the other side of me on that, they were also on the other side of me in the Crimea question. And they’re also going to say we need to do a lot to back Taiwan, say — where I’m actually more sympathetic to the hawkish point of view.

Matthew Yglesias: But to me, the view that we should cut Russia off from the swift payment system — there’s a cost to that in terms of our ability to build a coalition to contain China. You have to have some sense of limits and focus in a world where the United States is not able to successfully dictate to every country around the world what it is they actually want to do. I think that the American national security community tends to practice a kind of logroll — where the Afghanistan hawks and the Russia hawks and the China hawks and the people who are really mad about Venezuela all kind of support each other. But that doesn’t make sense. That’s not how the world works.

Rob Wiblin: Because there’s limited resources, basically. If you want to spend more on Venezuela, then you’ve got to spend less on something else.

Matthew Yglesias: Well, particularly because the tendency is to half-ass these things. Now, you could have said, “Kicking Russia out of Crimea is incredibly important, and we should go all-in on that.” I might have disagreed, but it might have worked. But what we chose to do on the Crimea issue was apply a level of sanction that nobody had a good faith belief would actually cause Russia to back down, but which induced further cost to our position in Afghanistan. The reason we did it was because nobody has wanted to concede a Russian sphere of influence in Ukraine. But have we achieved a lot for Ukrainian independence over the years?

Matthew Yglesias: I don’t want to cast this as the sort of, “Oh, woe is Russia, Putin is right about everything” initiative, but we have done a lot of things over the years. We have said that the door is open to NATO membership for Ukraine. When there was a protest movement in Ukraine against an authoritarian government there that was Russia-aligned, we backed the protestors — with words, with material resources, we took their side. The European Union, not the United States, tried to do this trade agreement with Ukraine that the Russians were very opposed to. We sanctioned Russia after they took over Crimea — when factually, I think everyone agrees the vast majority of people living in Crimea are Russian speakers, they wanted to be part of Russia. So we sanctioned Russia there.

Matthew Yglesias: These are all things that we have done for the sake of Ukraine and the Ukrainian people, but have Ukrainians actually been made better off by any of this? Have we been giving them false hope as to how much assistance they can actually expect from the United States? From Britain?

Rob Wiblin: Especially in the long run.

Matthew Yglesias: Right, in the long run. It seems to me, just factually speaking, that in the end, Ukraine is going to end up in the Russian sphere of influence. And if we had been clear from the beginning about the fact that the West is actually not that interested in Ukrainian independence, that Ukrainians would’ve been better off. They had to make decisions.

Matthew Yglesias: There’s this saying: “Poor Mexico, so far from God, so close to the United States.” Mexican politicians have, I think, had to embrace realism about that, that they are close to the United States of America. I think that we have tended to lie to the Ukrainians and made them feel that they are closer to God than they actually are, but they’re close to Russia.

Rob Wiblin: Right. I guess the problem in that specific case is that Russia cares a lot about Ukraine and is willing to do a lot about it. The reality is we care somewhat, but considering the coming decades, we’re just not going to put in the resources that are commensurate with what would be necessary to stop Russia. So it might make more sense to basically concede that and give up at the beginning, rather than do the half-hearted thing and then give up later on anyway. It sounds like that’s kind of your case.

Matthew Yglesias: Right. You need to gut check with what’s really important, particularly for the United States. The flip side of us being so far from many of the other great powers of the world is that it can be very tempting for us to engage in a kind of cheap talk.

Matthew Yglesias: So Americans have become very critical of Germany’s approach to Russia because they have this natural gas. I don’t know 100% all the details. And maybe the Germans are wrong. If German people want to say, “Internal to German politics, we are making a big mistake here. We shouldn’t be turning off these nuclear power plants. We shouldn’t be importing German gas.” That all sounds plausible to me, and I could be convinced. The flip side of that though, is that the Germans actually have skin in the game on both sides.

Matthew Yglesias: Some people feel that, if Russia dominates Ukraine, that’s a huge risk to European security writ large. It’s a classic Munich appeasement kind of argument. I think the Poles feel very strongly about this, as do the Lithuanians, people like that. So Europeans are closer to it. They bear both more costs from Russian aggression and also more costs from really confronting Russia super aggressively.

Matthew Yglesias: If a united Europe was saying, “We really want to back Ukraine and we need some piece of logistical help from the United States of America. There’s some kind of airplane part that we really need.” I’d be like, sure, give it to them. But it’s really unclear to me how true that is, versus Americans engaging in a kind of cheap talk. And nobody cared about this Afghanistan northern supply line thing when it was up there.

Matthew Yglesias: So right now, we’re freaking out about Russia. But six months from now, are we going to be freaking out about Iran, and is Russia selling them something that relates to their nuclear program? Or are we going to be freaking out about China? What do we care about? We need to try to think more rigorously than I think we do.

Rob Wiblin: About these tradeoffs.

Matthew Yglesias: About these tradeoffs. About what do we care about.

How military intervention looks as a humanitarian intervention [01:29:12]

Rob Wiblin: So we’ve gone into these specific cases very quickly. Is it possible to maybe step back to a slightly higher level of abstraction and say, what is your big-picture take on the US and other countries using military force to make the world a better place?

Matthew Yglesias: Absolutely. This was my original entry point into the world of effective altruism, because I was very taken in the late ‘90s by the idea of humanitarian military intervention, by the idea that we’re doing incredible good in the world in Kosovo. And that this was a big thing, that post–Cold War, the American military could end genocide around the world, and do all kinds of good things.

Matthew Yglesias: I think in Iraq, that line of thinking turned out poorly on a sort of first-order practical basis. But that also got me thinking about the question of, “How do you help people around the world?” If you want to take a non-national interest view of the world, and say, “Who do we help?”, it turns out that there are radically cheaper ways of helping people in need around the world. If you want to do something that is selfless, there’s so much that can be done. We’re not close to exhausting the possibilities for doing that. I try to do some of it with my personal giving. I try to encourage others to do it. I try to encourage the government to do it.

Matthew Yglesias: But it began to seem to me that another area where we had to be more honest with ourselves about, “What are we actually doing here? Is the military apparatus that exists really a tool for human betterment?” Now, it can be. When there was the tsunami in Indonesia, the fact that we have a navy. The navy is full of boats — they’re really good boats, they have really skilled sailors. They were able to deliver a lot of humanitarian relief aid. It was great that they did that.

Matthew Yglesias: I’m not gonna go full paranoia, like it was just a propaganda game to help the CIA dominate the universe. But it is not really helpful to think about military force as a primary tool for humanitarianism. The cost benefit is not there. The incentives aren’t aligned. It’s not what the people are trained to do. And you see how quickly it turns. As long as we were fighting in Afghanistan, we would say we were doing all these good humanitarian things. The moment the Taliban takes over Afghanistan, we have them under severe sanctions that are causing a famine over there.

Matthew Yglesias: So I think we should not do that. Famine is very bad. But it’s a warmaking tool: their mission is to beat the Taliban. When beating the Taliban involves doing a certain amount of humanitarian propaganda — which involves a certain amount of humanitarian real stuff — the military goes and does it. But the moment the way to beat the Taliban is to starve the country into submission, that’s what that exact same kind of apparatus flips to.

Rob Wiblin: Yeah. OK, so the basic idea is, we want to evaluate military intervention based on its humanitarian outcomes. How many people did it help? How much did it improve their wellbeing? How much did it cost — in terms of dollars, in terms of equipment, in terms of talent, in terms of people? Basically you just think that if you analyze that rigorously, using the same kind of cost-benefit strategy that we do with everything else, then it looks way worse as a humanitarian approach than many other things that we could do.

Rob Wiblin: One obvious thing is, if you put that money towards fighting disease — say, giving antivirals to people with HIV — there’s no army on the other side trying to give more people HIV, trying to push back and make things very difficult for you. And most actions that you can do in the world are not a matter of military conflict. There’s a huge space of other ways that you could try to help people that don’t involve the army.

Matthew Yglesias: Well, it’s also that the impacts are so uncertain. So NATO intervened in Libya to prevent a specific massacre that they thought was about to occur, and because they thought Gaddafi was bad in general. It was a costly intervention, but it also felt like they were doing a lot of good. But then you look at how that situation has played out over subsequent years — it’s not obvious to me that on net there were any humanitarian benefits there. Now, maybe there were. It’s very hard to say. Or look at that intervention in Libya: did that suggest to opposition forces in Syria that if they could provoke the Syrian government into massacring civilians, that NATO would intervene and help them win the civil war? I think that would’ve been a reasonable inference to draw from what played out in Libya.

Matthew Yglesias: I think that the architects of that intervention maybe even hoped that that inference would play out. The message they wanted to send was, “Hey, regimes of the world, don’t massacre civilians, or you may provoke intervention.” But that also sends to opposition movements around the world, “Hey, if you can bait the regime into massacring civilians, you’ll get an intervention.” But it turned out they didn’t get an intervention in Syria.

Rob Wiblin: They got an intractable civil war.

Matthew Yglesias: They got massacres, they got an intractable civil war. They got a huge humanitarian disaster. I don’t want to say that that definitely wouldn’t have happened if we hadn’t intervened in Syria, but I think we increased the likelihood.

Matthew Yglesias: Again, it’s a question of carelessness about what is the real impact of these kinds of things in the world. Casual proponents don’t study it rigorously. But also the military: there are certain things they look at very closely, because they want to understand what they’re doing and they want to get better at their jobs. And the global humanitarian situation is not on that list. It’s not something that they focus on. It’s not the culture of the armed forces. It’s not what they are for.

Matthew Yglesias: It’s challenging because the policymakers don’t speak honestly about their own commitment to humanitarian values. So we are very eager to back opposition figures against authoritarian regimes sometimes, but not against Kuwait. There’s not going to be a colour revolution there. We can say casually, “Well, that has something to do with oil” or something.

Matthew Yglesias: But nobody is going to write down on a piece of paper, actually, what are the criteria in which the United States or other Western governments will back democratic opposition? When will we stop people from massacring? In part because we don’t want to give a clear green light to authoritarian regimes — we always want the Saudis to be guessing like, “How much can we get away with?” But it makes it a very hazy landscape, and I think it’s not closely comparable to global public health interventions in terms of —

Rob Wiblin: Predictability.

Matthew Yglesias: — we could do some more good. If we could convince people to care a little bit more about distributing insecticide-treated bed nets, we could stop malaria cases.

Rob Wiblin: Yeah. I think even the proponents would concede that trying to forecast exactly what impact a military intervention is going to have is quite difficult — and substantially more difficult than figuring out what the likely consequences are of distributing antiviral drugs to people with HIV. You don’t get the same kind of risk of massive blowback, or on net having negative effects, even though almost all interventions do have some risk of unintended consequences.

Matthew Yglesias: To sign a more hawkish note, if we were talking about China, I am more of a patriot, more of a hawk, more of a “we got to help the allies” kind of guy there.

Rob Wiblin: Why is that?

Matthew Yglesias: One thing is that I think that the stakes are higher: that Taiwan, Korea, Japan are fairly large, stable democracies. We have established alliance relationships with them. I think that backing out of existing defense commitments can be very destabilizing in its own way. I think that the PRC regime is really bad — has proven itself to be really bad in its treatment of Uyghurs, and its treatment of Tibetans, and its general level of aggressiveness. I also think that China is a real… I don’t quite want to say a threat, but a player on the world stage: quite large economy, growing fast, very powerful.

Matthew Yglesias: There is something to the old-fashioned idea of either the US is going to be the kind of hegemonic player or China is. There isn’t some world in which it’s going to be Norway. So for all the flaws and hypocrisy of the American national security community, I think if you ask the people in Korea, in Vietnam, in India, in Japan — they are all more comfortable with the United States being a big player in that region than with us not being a player in that region. There are a lot of questions about what do you do specifically there, but I think we should do something. But to me, that’s a reason not to be distracted by low-probability-of-success ventures in central Asia.

Where Matt does favor military intervention [01:38:27]

Rob Wiblin: Here we’ve been laying out the case against. To help see the limiting principle, what is a case where you favored overseas military intervention? Or a case where maybe we didn’t do it and we should have?

Matthew Yglesias: I think that the first Persian Gulf War, responding to an invasion of Kuwait with a high degree of efficacy, that was good. That worked out well. I don’t know that I would stipulate beyond that a lot of interventions that did happen. The best case for military force is where you have clear and credible commitments.

Matthew Yglesias: One reason in some ways that I think it would be good to kind of forswear NATO involvement in Ukraine is that we’ve gotten very far with NATO — with the idea that the US and Britain and France and Canada and all these other countries will go to war to defend Lithuania. I think a reasonable person could ask, “Is that really true? Would we really go to war for Lithuania?” But we’ve said that we will. We want the Russians to believe that we will. To say that we wouldn’t — to like, kick them out — could be very destabilizing. Very, very dangerous.

Rob Wiblin: I see.

Matthew Yglesias: So to clarify that we are, honest to God, not going to make that commitment to Ukraine. We appreciate that it is very serious, but that this is a vow we have really made to a bunch of countries, and we stand by that, and we are going to focus on that. I think that that’s important in some ways: we have these defense commitments, and we want them to be credible.

Matthew Yglesias: The other big thing that the American military does is participating in some of these global public goods around piracy and sealanes and things like that. We don’t talk about it a lot. It’s not like, “Remember that one time six years ago that a little flotilla scared off some Somalis?” But that’s good. It’s good that we are not constantly talking about the big anti-piracy wars. That means that it’s working, and it’s good to have these facilities out there.

Rob Wiblin: Cases that people sometimes bring up include the Vietnamese invasion of Cambodia to get rid of the Khmer Rouge as a good humanitarian intervention.

Matthew Yglesias: Yes.

Rob Wiblin: Admittedly, the Khmer Rouge was especially atrocious.

Matthew Yglesias: They were bad.

Rob Wiblin: Oddly enough, they were kind of a US ally at the time, or like halfway a US ally in Cold War politics. So that’s one case to be raised. Bosnia, which is a case that I know less about. I guess the West didn’t intervene in Rwanda and sometimes people regret that. Yeah, are there any other historical cases that you think are at least interesting?

Matthew Yglesias: The Rwanda one is interesting. There’s a lot of criticism of our failure to intervene there. What I’ve never quite heard is what exactly the pro-interventionists would’ve liked to have seen happen. The thing that they normally say is that there was this very low-cost stuff that we could have done — you know, bombing certain radio stations — that would’ve had a strong, beneficial impact. That might be true. Obviously it would not be the end of the world to have blown up a couple of radio stations somewhere. But would that have satisfied the desire?

Matthew Yglesias: What I know weighs on presidents, when they actually get in office, is that you say, “OK, I’m going to do X because it’s going to stop Y.” But then if people are still being killed, it’s hard to go out in front of the cameras and be like, “Well, actually you don’t realize this –”

Rob Wiblin: Do the radio stations, but that’s it.

Matthew Yglesias: Well, “Under the counterfactual, were the radio stations there, even more people would have been killed.” Instead, the reporters are like, “They’re still killing people there. What are you going to do about it?” Then you’re sending more people in. You’re directly intervening into a civil war that’s on the other side of the planet, whose stakes are very unclear, that’s not well understood by American government officials. It can get hard.

Matthew Yglesias: A book came out recently arguing that in fact, the whole Western understanding of the origin of that round of fighting is mistaken. Tutsi rebels brought down the prior president of Rwanda’s plane after all. And Paul Kagame was the bad guy in all of this. So I don’t know if that’s true — I don’t want to be too much of a Rwanda revisionist out here. But it seemed to me that certainly the people making these points don’t know that the other thing is true, right?

Matthew Yglesias: There’s a lot that we don’t know about the dynamics of civil wars in foreign countries where we don’t speak the language. I don’t speak the language. You don’t speak the language. I think most of our listeners don’t speak the language. But also the decision-makers at AFRICOM don’t speak the language.

Matthew Yglesias: One thing I can say about American Russia policy is that we do have a lot of Russia experts. There is a lot of subject matter knowledge about Russia that’s been developed in the United States over the years, and that is good. We’re not completely blundering in the dark with no comprehension of what the issues are. But I feel like that’s something we’re often in the position of doing. It’s maybe something we should try to improve at. I think Americans have never been comfortable in a kind of “imperial role” around the world, and have never invested in the idea that maybe we need important people who speak Pashto and can go run our little colony in Afghanistan. But there’s also wisdom in trying to not try that hard.

Matthew Yglesias: The Vietnam Cambodia case is an interesting one, because I think you would say there were clear humanitarian benefits to what the Vietnamese did there. But of course that’s also not the reason that they did it. It’s just like, good for them. I don’t think you can deny that that was a beneficial thing to do, but it’s also clear that this just had to do with their geopolitical conflict with China, not anything else. So I don’t totally know what to say about that.

Why smart people disagree [01:44:24]

Rob Wiblin: Yeah. Yeah. OK, so my sympathies in general are kind of with the position that you are taking. But I do know smart people who are well-meaning who disagree, and often think it is worth going in. So I’m going to try to represent what I think they might say.

Rob Wiblin: In the case of Afghanistan, I think a lot of people saw the Taliban taking over, humanitarian crisis, looks terrible. And people who have a more effective altruist mindset were doing calculations on how much it would have cost to stay in there indefinitely every year. What was the ongoing annual cost of occupying the country in order to prevent the Taliban from taking over? Not having, potentially, the delusion that the Taliban’s ever going to go — it’s kind of a very long-term operation.

Rob Wiblin: I just looked at the average over the last 20 years that the US was in Afghanistan, and it was about $60 billion a year. Then you think, how many people do we get to live under the Western-supported regime there rather than the Taliban? Maybe it’s about half the country? It seemed like it went back and forth, and some people were under this kind of hybrid regime, where it depended on the time of day exactly which government had more control.

Rob Wiblin: But maybe that’s like 20 million people. In which case, you come up with something like $3,000 per person per year to have people living under one government, rather than the other. In a world where it seems like you can save a child’s life for something like that, maybe that doesn’t seem so great, or it doesn’t seem equivalently good to the very best global health interventions that we can find.

Rob Wiblin: But then you might think, “If we stop holding back the Taliban in Afghanistan, is the money going to be redirected to these very best global health interventions?” Maybe that actually isn’t the tradeoff at all, because the American public is just never going to be persuaded to spend $60 billion on global health in that way. So it’s occupy Afghanistan or don’t — and otherwise the money just goes into some pool, and who knows what it’s going to be spent on. Could be something good, could be something bad. And then you could do worse than spending $3,000 a year to protect someone in Kabul from being under Taliban control. So what would you say to that?

Matthew Yglesias: You didn’t find me before this debate was joined in a hot way out there, being like, “We got to get the troops out of Afghanistan” — like, Amerikkka with lots of Ks in it protesting in front of the White House gates — because I did see the view that this is better than sort of not doing it. But I kind of feel, in these foreign intervention cases, there’s a conflict between an act and a rule consequentialist view of how to deal with the world. And my strong conviction is that the military as a humanitarian tool is a bad practice — that its cost effectiveness is really bad compared to the most effective interventions, that in many individual cases it backfires and is net harmful, that the consequences are really unpredictable.

Matthew Yglesias: It would be convenient if I could make the argument that actually, everything that we’ve ever done in the name of humanitarianism is counterproductive and negative. I think that that’s a little bit of an underrated perspective. People doing that kind of math tended to neglect the fact that the war happening had very negative humanitarian consequences: there were all these bombs dropping and different things happening in the areas where the war was fighting. And the people who were most represented in the Western media were people who lived in Kabul — where our security umbrella was at its most useful — and people who were educated, spoke Western languages, et cetera. But there were lots of people living in villages, living already under Taliban control, but not enjoying the fact that they were getting attacked back and forth, et cetera.

Matthew Yglesias: All of that being said, yes, there can be these benefits in specific cases. But I don’t think that it’s a good mentality to cultivate — that this is what we are doing, that we are going to solve global problems that way.

Matthew Yglesias: In the space that you occupy, it is taken for granted that there are all of these highly effective interventions that we could be doing. I think that that is still not a mainstream perspective in American political commentary — that there is this real axis of either you are an isolationist, or you are really enthusiastic about the use of American military force abroad. So either you’re a table-pounding national interest guy, or you want to have lots of wars everywhere. I think it is just so important to bring the message to people that —

Rob Wiblin: There’s this other option.

Matthew Yglesias: Not just another option, but just that it’s not even close. If you actually want to say, “Hey, we should care a little bit more about humanitarian issues in the world,” then the amount of money that’s involved in that is trivial compared to what it costs to keep an aircraft carrier circulating around the world.

Matthew Yglesias: I don’t know how full of BS American elites are when they say they care about the global world, but I don’t think they’re 100% full of shit. You know what I mean? They do care a little, and if they could direct that tiny amount of caring in a reasonable way, they could do an incredible amount of good. And they might even start feeling better about themselves, and get in a virtuous cycle, where we’re actually accomplishing things that do good. And I just don’t think that that’s what is going on in Afghanistan, on a really fundamental level.

Rob Wiblin: Yeah. A few other things one could say on this is that it’s not realistic to think that most of the money saved from leaving Afghanistan, or even very much of it, is going to go towards what we think of as the very best humanitarian interventions. But you’re then not thinking about, is it politically plausible to go to the American public and say, “We want to spend $60 billion a year occupying Afghanistan forever, or for the indefinite future, because of the humanitarian benefits to people in Kabul and other more urban areas of Afghanistan.” The American public would at some point get sick of it, and either withdraw completely or withdraw enough that they weren’t effectively controlling most of the country.

Matthew Yglesias: If that’s plausible, right? If you think you could win a political argument for $60 billion forever in Afghanistan, why not get $5 billion for the bed nets? If the argument is that, “This is going to be sustainable, because even though the cost benefit is so bad, it involves soldiers. So we can lie to people and say that it’s a national security imperative,” then that’s a rickety argument. That reminds me of my six-year-old gluing Popsicle sticks together in a really awkward way.

Matthew Yglesias: I think that we all need to be cautious of the fact that we can construct bad arguments because they serve certain kinds of psychological ends. And I think that a lot of what you have in this national security discourse is people wanting to align themselves with the good guys. Taking the side of giving in to Putin, or giving in to the Taliban — that’s a bad feel. And if you say, “No, we shouldn’t withdraw this year. We don’t have to.” And somebody else is like, “Yeah, but we’re going to withdraw at some point.” So the bad thing is going to happen, but it’s not going to be my idea to withdraw.

Matthew Yglesias: I think that that’s been compelling to a lot of presidents. The Obama administration did not believe that they had a winning strategy in Afghanistan, but they also didn’t want to be the ones who suffered the fall of Kabul. So they punted it to Trump, who did not believe he had a winning strategy in Afghanistan, but didn’t want to be the one who saw the fall of Kabul. So he punted it to Biden. I think that the Biden administration’s experience — just as a pure matter of politics — has 100% vindicated the cynicism of Obama and Trump. He has taken so much more political damage for admitting that he didn’t have a strategy to win the war than the Obama administration took for lying about it.

Rob Wiblin: Well, for putting him in that situation.

Matthew Yglesias: They called it the Afghanistan Papers, but it’s been very well documented. There was all this information coming into DC from generals on the ground that was like, “The stuff that we’re doing isn’t working. All this money we’re spending on the Afghan army, it’s all being wasted and taken away.” But the right thing to do, in a narrow sense, was to kick it down the road.

Matthew Yglesias: I feel fired up about this Afghanistan issue. Just because I don’t normally recommend that political leaders try to do the right thing, regardless of the political cost. It’s easy to say that you think people should do that, but in some ways it doesn’t stand up to that much scrutiny. But Biden actually did it. But it’s very costly — taking a principled stance is risky business.

Rob Wiblin: Yeah, I agree. I do have the reaction that like, right or wrong, blaming Biden for this situation was a bit absurd, rather than blaming the people who’d been creating the situation that we’re presently in for the last 20 years. Clearly a lot of responsibility has to go to the people earlier in the chain.

Matthew Yglesias: You see these guys, like a general who commanded for years there, would come on TV and be like, “Oh, Joe Biden, he really screwed this up.” Well, what about you?

Rob Wiblin: So I’m doing a bad job defending interventionism here, but another thing that I wanted to raise is that sometimes, when doing this cost-benefit analysis, it’s possible to be somewhat starry-eyed — somewhat unrealistic about the actual humanitarian benefit being provided for people on the ground.

Rob Wiblin: In order to evaluate how much we should be willing to pay to keep people under Western government rather than the Taliban, we need to say, how good is this government that we have in Afghanistan? What services is it providing? Then, realistically, to someone who’s in a village in Afghanistan, how much would they hate being under the Taliban? I think it’s very hard, including for me, to get away from thinking about how I would feel about that situation, and how I would like the government that we’re putting in there to be — rather than how it actually is. And also, I’m not super informed, so I’ve got to imagine, “What would a good government do?” rather than, “How does this government actually function, and how corrupt is it?”

Rob Wiblin: Likewise in Ukraine. On a gut level, I feel revolted by the idea of Russia taking over Ukraine, after Ukrainians have suffered under the yoke of Russia throughout history many times. But then, if I try to think on a moment-to-moment basis, a day-to-day basis, how much worse off would Eastern Ukrainians or Crimeans be if they were in Russia rather than in Ukraine? I’m not sure what the wellbeing cost is exactly, or how large it would be in the long term.

Matthew Yglesias: This is where longtermism as a philosophy is very important, but challenging. Even people who are bought into it, we all, myself included, struggle to apply these insights outside of the specialized domain. Where people are like, “As a longtermist, I care a lot about x-risks.”

Matthew Yglesias: But if we think about things people are just arguing about, and we think about them without extreme discounting, how much does it matter to the long-term future of Ukraine, whether it’s governed from Kiev or from Moscow? There’s a way of doing the Russian nationalist perspective on this: “It’s never been its own country. It’s not even a real language.” That’s stupid. That’s crazy. But what we do know is that the political boundary lines of Eastern Europe have varied quite a bit over history, and there’s a lot of contingency into how language families are constructed, how political allegiances and political identities are constructed.

Matthew Yglesias: It’s just not obvious that one way or the other is going to produce superior outcomes for the longer run. There was this incredible, I don’t want to say fad, but there was a sense around the turn of the 19th to 20th centuries that the political organization of Central and Eastern Europe was really bad. It was really bad to have the Habsburg Empire, and that we should reorganize everything along the lines of national boundaries — as they said, like Spain, France and the UK were — but I think actually weren’t. And I don’t think that experiment turned out well at all. At least, in the medium run.

Matthew Yglesias: A lot of bad things happened to try to reorganize Central and Eastern Europe along nationalistic lines. They are now the most homogenous countries around, Poland and Czechia, Slovakia. They really conform to the nation-state ideal in a way that the countries that originated it don’t. But what was the benefit of that, in the end? After all that bloodshed, what occurred there? So I don’t want to defend imperial projects, but why should we be so bought into nationalistic projects either?

The case for NATO taking an active stance in Ukraine [01:57:34]

Rob Wiblin: Yeah. Coming back to Ukraine, someone who wanted to advocate for NATO, taking a fairly active stance there, would say something like, “We’ve managed to reduce the number of cross-country invasions and annexations massively, close to zero. We really want to preserve that norm.” So, when Russia did the first blatant annexation in many, many years, we wanted to make this extremely costly to them, so that other countries would see that and be deterred from invading other countries. The idea would be, we could impose costs on Russia that would seem very costly to Russia, and scare off other countries — at not too high a cost, not too high a risk of causing nuclear war.

Rob Wiblin: We would do the sanctions, we would impose economic irritation. We would send weapons to people they don’t like in Ukraine. Just hassle them, and be like, “Well, you did something we don’t like, so we’re going to be a pain in your ass.” Likewise, they might think today Russia wants to invade Ukraine, but we could deter them for an acceptable cost by threatening to, tit for tat, be even more of an annoyance to Russia if they go ahead and do this thing that we really hate. I suppose that sometimes that kind of logic might be right, but what’s going wrong sometimes with that reasoning?

Matthew Yglesias: I definitely think that that is the best characterization of the American government’s actual policy, which has not been actually to help Ukraine, but to inflict costs on Russia. Some of that has been the sanctioning, but some of it has been the drip, drip, drip, arms and supplies to Ukraine. We have been giving successive Ukrainian governments enough support to induce them to not give up.

Rob Wiblin: But not more.

Matthew Yglesias: Which is costly to Russia. And I think in some ways it is a little bit cruel to Ukrainians. They are being used to establish some kind of normative concept in the abstract, that does not have that much to do with the actual interests of the typical person, living in Kiev or wherever else. It is an important norm. I cited the original Persian Gulf War as something that I thought was worth doing. I think that it would be desirable to think about this norm and its enforcement in a more kind of global and comprehensive route.

Matthew Yglesias: If you go back to Maidan Square, if you could travel back in time and get into meetings with Victoria Nuland, you probably couldn’t have gotten an Oval Office meeting on the subject of US policy toward Ukraine in 2013. But in retrospect, they should have held a meeting, and been like, “Let’s be a little less careless about this.” And say, “OK, there’s this protest movement. Our specialists want to back the protest movement. We want to overturn this autocrat because he’s bad. We also like that the opposition is pro-European and they want to do a trade deal with the EU.”

Matthew Yglesias: But let’s think about all the considerations here. What if Putin loses it and decides he wants to invade and take over the Russian-speaking areas of Ukraine? What are we going to do? We’ve got to do something because we need to uphold the norm that that’s bad. But we’re not going to actually fight the war with Russia. So how good a job are we going to do of upholding the norm? And what are we going to do? Are we going to make a big public statement that’s like, “We don’t care about the Ukrainian opposition, because we’re worried that it’ll lead Putin to call our bluff and undermine global norms.”

Matthew Yglesias: It’s like, no, but what if we just supported them less? What if we just turned all the dials down 30%? Then some would say, “No, that’s bad, because then they’ll lose.” But maybe that would be good. That would be just another sad case of people living under an authoritarian regime — which unfortunately a lot of people do — but we could uphold all of our norms through the private hypocrisy.

Matthew Yglesias: Something that I was struck by is, back in 1955, Austria issued a unilateral [declaration of perpetual neutrality])https://en.wikipedia.org/wiki/Declaration_of_Neutrality), in which they said that they would never join a military alliance with any country, for any reason. The day after they made that unilateral declaration, Soviet occupation forces left the country. And because Soviet occupation forces left the country, so did US, French, and British forces. Today, that unilateral declaration day is like the national holiday in Austria. The Austrian State Treaty — which had been agreed to by the four powers before that — didn’t actually go into effect until Austria unilaterally forswore NATO membership.

Matthew Yglesias: Now of course, for the Western powers to make Austria forswear NATO membership would’ve been totally unacceptable — they’re a sovereign country, a democracy, et cetera. For Austria to say that as a concession to the Soviet Union? Unthinkable — it’s appeasement, blah, blah, blah. But it’s lucky that things worked out in that coincidental way.

Matthew Yglesias: And I think in an era before the internet and before intense media cooperation, it was easier to handle things in a hypocritical way with a lot of winks and nudges — nobody really knows what was said to whom when. The whole Ukraine thing, if you had gone way back and said, “Look, our core interest here is in not having there be an overt invasion,” you could have handled that in so many different ways.

Rob Wiblin: That could have been achievable, yeah. Interesting. When talking about overseas intervention, I find that people who I read — including sometimes you — tend to use the expression, “it’s not in the US national interest.” We would never regard that as a decisive argument in other areas — like whether it’s in the US national interest to prevent kids dying of malaria, we should do it anyway, because it’s the moral thing to do. But this is often raised as a reason to just get out of Afghanistan or whatever else. Why do you make that argument sometimes, even though you wouldn’t probably make it in other areas?

Matthew Yglesias: I think it’s important to try to be clear and rigorous in what it is that we are talking about. I would make the argument that it’s not particularly in the American national interest to invest heavily in antimalarial drugs, if someone was seriously offering the argument that it was. You know what I mean? If we were having a debate about how do we prevent Americans from being killed by terrorists, and someone was like, “What we really ought to do is distribute malaria drugs,” then I’d be like, “No, that’s stupid.”

Rob Wiblin: Make people like us and then they won’t.

Matthew Yglesias: You could do it. I would say, “No, I don’t think that that makes sense.” Now, if somebody wants to say, “Look, this is a really good way to help people,” then that’s great. I’m kind of a Humean in terms of what motivates people. It’s hard to have a coherent argument that’s like, “You should care about the abstract wellbeing of other people.” I think that most of us do care about the abstract wellbeing of other people, at least some of the time. Very few people are like, “No, I don’t care at all,” with no interest whatsoever in helping others. And then it’s good to think about how to do that.

Matthew Yglesias: Then we have other modes in which we’re thinking about American national security, or we’re thinking about geopolitical competition. And it’s good to try to think about what is and isn’t actually useful there, what is and isn’t actually helpful there. You could say some given military intervention is being done for selfless humanitarian reasons, but as we’ve been talking about for a long time, I think that’s a pretty dubious line of argument — just like saying that malaria drugs are how we’re going to fight terrorism. I don’t think that lethal assistance to Ukraine is how we’re going to accomplish humanitarian ends.

Matthew Yglesias: People try to do more than one thing in their life. Certainly governments do, democratic polities do. I don’t think that trying to push everyone to be maximum cosmopolitanism 100% of the time is that constructive of a way to engage with the political system. We’re trying to operate on different margins at different times. I would always encourage everyone to try to think more cosmopolitanly, but I see it as more the opposite — that cosmopolitan rationales get opportunistically invoked in the military context.

Rob Wiblin: What do you think is the best argument that the hawks have? If you came to think in 10 years’ time that your worldview here was kind of misguided, why might that be?

Matthew Yglesias: The case that I disagree with — but I don’t know that I have a knockdown argument against — is the idea that just pushing in the hawkish direction all the time is ultimately the only way to prevent the world being overrun by bad autocrats. I’m saying, “Let’s keep our powder dry. Let’s focus on where we can really accomplish things.” I agree that containing China is important. I agree that supporting the alliance system is important.

Matthew Yglesias: I think the view on the other side is like, “Look, we just do what we can. We are rolling the ball uphill against the selfishness and shortsightedness of the American people. And if we can make a stand in Ukraine — or in Afghanistan, or Taiwan, or Lithuania, or wherever it is — we’re just going to defend all the hills because they are all important.”

Matthew Yglesias: I don’t find that line of thinking attractive in some way. To me, it feels intellectually dishonest. I don’t love Straussian-type political interventions, where I’m saying, “We’ve got to tell people one thing for the sake of this other thing.” But I don’t know that it’s wrong. This is an actual Leo Strauss example: people believing in God might be good, even though I don’t think it’s true. I might be uncomfortable spending a lot of time running around the world, being like, “Stop going to church, man. This is all lies.” Because I think it probably is, on average, good for people to go to church and to believe in God and to do these things. But I don’t think it’s correct.

Rob Wiblin: You wouldn’t go out and advocate for religion on that basis.

Matthew Yglesias: I wouldn’t advocate for religion, but I’m also not super comfortable advocating against it.

One Billion Americans [02:08:02]

Rob Wiblin: Yeah. Someone on Twitter puts you this question: “Realistically, what’s the best shot at liberal democracies retaining the balance of global power in 50 years?” An easy one.

Matthew Yglesias: What do you mean? What’s the best shot? This is my book. This is One Billion Americans.

Rob Wiblin: Yeah. Do you want to make a quick pitch for that?

Matthew Yglesias: I go on at great length about it. You could apply this to other countries too, but particularly the strong demand to migrate to wealthy liberal democracies is a real source of strength for those countries — economically, but also geopolitically. We should be taking that much more seriously, taking much more advantage of it.

Matthew Yglesias: There’s also things about fertility rates and pro-family policies there. But we should have there be lots of people in America, and in Canada, Australia, Germany, all these countries. They are good — people vote with their feet to want to go and live in them. We don’t need to be full-on open borders to be less paranoid about the idea that people are going to come live in the best countries in the world.

Rob Wiblin: Doubling the US population would make its military commitments far more credible, because it would have twice as much GDP and it would make its cultural clout much, much larger. It would also have lots more resources for humanitarian work as well, so would expand on all of these dimensions.

Matthew Yglesias: I also think there’s this fundamental dilemma in our relationship with China right now, which is that it’s very dangerous to have lots of conflict with China. There’s lots of downside risk to that, lots of problems for the world. At the same time, China seems pretty bad. There’s lots of good normative reasons to want to help allies to resist them.

Matthew Yglesias: Right now, if Xi sits down with Joe Biden and he says, “President Biden, I’m always trying to make China richer. We’re trying to have catch-up economic growth. Economic growth in China has done all this good for the world. It’s reduced poverty, and any national leader just wants his people to do well. But if we do well, we’re going to become more powerful than the United States. So I think that you, every day, are trying to undermine us.

Matthew Yglesias: “You come and you say this thing about wet markets, or gain-of-function research, or climate change, or open trade, or freedom of the seas. But I don’t believe you. I think that you are a patriotic American leader who believes in American values, and does not want China to be the most powerful country in the world. So I think that all you do is sit with your advisors at night and scheme about how to undermine China and destroy our prosperity.”

Matthew Yglesias: And then Biden’s going to be like, “That’s not true. That’s a lie. That’s not what we do here. You’re being crazy.” But it’s like, well, what are we doing? What is the plan? What is our message to the world about how we stay on top without taking down everybody else? The One Billion Americans idea is an answer to that question.

Rob Wiblin: Yeah. That’s the alternative.

Matthew Yglesias: That we do not accept the premise that we have to be the third-largest population country in the world. That we are, in terms of different things happening here, going to be as helpful as we can to our friends in India. We are going to be as welcoming as we can to people who think the American way of life is great. We are going to help American parents have children.

Matthew Yglesias: And we are going to wish China well — yes, a little bit skeptically, but we are glad that there’s less poverty there. We are hoping that you guys invent great things and do good stuff for the world. But we’re staying on top, and it’s your problem that people don’t want to move to your country because you’ve created this dystopian surveillance state, and that’s on you.

Matt’s views on the effective altruism community [02:11:46]

Rob Wiblin: Yeah. OK, let’s move on and talk a bit about some intellectual communities that I’m very familiar with, and a lot of listeners will be familiar with, but I’m just curious to hear your general impressions about. And that’s the effective altruism community and the longtermist community. I imagine you’ve largely just read or listened to the content that people like me put out, but you might still have some pretty useful impressions or advice. And I wouldn’t hold back, because listeners to the show — when I meet them or when they’re online — they love flattery, but they also really like constructive criticism. So yeah, what do you make of effective altruism?

Matthew Yglesias: I’m for it, I think. I want to be effective; I want to be altruistic. I think it’s good — I think that these lines of thought are very good. I should have said this when we were talking about foreign policy before, but actually one of my EA origins is when I was in college, I saw Peter Singer lecture about Iraq. And I was an Iraq War–supporting student at that time, and I thought that what he had to say was very impressive and challenging to my preconceptions. Part of why I went to see him in the first place is that I had no idea what he thought about foreign policy, but I’d just read some of his general books about normative ethics and stuff like that — so I was bought in and therefore open to it.

Matthew Yglesias: So I think it’s great. Something I worry about is that as you start to have a community, all communities can become a little bit of an extreme version of themselves — when that isn’t necessarily where intellectual movements actually do good in the world at the margin.

Rob Wiblin: So your concern would be that if people are all talking to one another — or they’re trying to impress other people who agree with their broad worldview — then you tend to get this extremizing effect over time, and people maybe solidifying views that have gone too far in some direction and that’s not the most accurate?

Matthew Yglesias: The views can be wrong, but also they can become a… OK, what is “effective altruism”? Effective altruism is the idea that we should be more critical and more rigorous about our altruistic impulses, right? And that we should push people to ask the questions, “Is this really a good way to help people? Are you too focused on things that are close to you in your community, too focused on things that get praised by other people? Or should we cast a wider net? Should we deploy more rigorous empirical methods when we evaluate our programs?” And that’s good, those are a good set of questions, a good set of pressures in the world. So we say, “Is military intervention humanitarian? Should we be clapping for the guy giving money to the art museum? Here’s this great study about bed nets and malaria.”

Matthew Yglesias: Then you have a community that buys those premises, and they start talking to each other a lot. They eventually develop the idea that we all should just think about the long term and we shouldn’t discount at all. And if you don’t discount at all, the only thing that really matters is human extinction. And a bunch of smart people have thought about this, and the biggest threat to human extinction is rogue artificial intelligence.

Matthew Yglesias: So now we’ve all read that book, and we’ve all put it around. And now what it means to be an effective altruist is, somebody is like, “What’s on your mind?” and you’re just like, “The threat of human extinction due to artificial intelligence.” And a large share of people are going to hear that and they’re going to be like, “What? That’s weird.” And an even larger share of people are going to be like, “Well, what does that even have to do with me?” Right? So now you’re not actually asking people to be more self-critical about their charitable giving or to give more money at the margin. You’re not pushing —

Rob Wiblin: You don’t really have anything to tell people or anything to say that’s actually actionable.

Matthew Yglesias: Right, or it’s not actionable on the margin. You’re asking people to take this Kierkegaard-like leap and join the community — as opposed to being in your community and living your life and caring about roughly the things you care about, but shifting your orientation a little bit.

Matthew Yglesias: I think that you become less effective, in a meta sense, when you become this set of doctrines that are quite odd and esoteric to most people — and that also don’t have a lot to do with people’s lives and the decisions that they’re actually making — versus one that’s trying to say to people, “For your end-of-year charitable contributions, consider the GiveWell Maximum Impact Fund as something to do. Think about giving a little bit more than you usually do. Think more critically about the doing/allowing distinction in your life.”

Matthew Yglesias: These are really important real-world messages, I think. As I’ve seen this evolve over the years, I both am myself somewhat bought in and try to get people to think more about x-risk and other stuff like that. But I worry also that the most broadly relevant stuff can get lost a little bit.

Rob Wiblin: It’s a very interesting challenge that you have as, now, a pretty large group of people. You want to make sure that you preserve the ability to take advantage of this sort of avant-garde altruism or these avant-garde ideas, which might allow you to pioneer some new area that people were neglecting that they shouldn’t have been, and identify new problems and new solutions that are really off the beaten track. But you also don’t want to lose the potential to shift lots of people in a positive direction on the margin of doing stuff that’s more relevant to their lives, ideas that they can connect with.

Rob Wiblin: These things are quite in tension. Especially, I suppose, because the more avant-garde ideas can often be more interesting and provoke people a bit more, and get people into debates and draw in a lot of the conversation — even if for the great majority of people, they’re very hard to action. And they’re also just easier to dispute — there’s a lot more premises that are going into them, so it’s less clear that they’re correct.

Matthew Yglesias: I was interested when I started hearing about people I know going to effective altruism conferences. I thought to myself, that actually doesn’t sound very effective altruism-y. I always associate effective altruism with ideas like, “Maybe don’t do the conference, just give the money — that airfare could do a lot of good in the world.” But also, why do people like going to conferences? Well, it’s good. You meet people, you hang out, you have a good time. So that’s totally understandable — I’m not against people having fun.

Matthew Yglesias: So it’s a movement. I think it disproportionately attracts quantitatively minded people — people who are above average in their analyticness, their interest in abstract ideas. Well, what’s the most fun part of this to think about? And it’s maybe a lot of speculations about artificial intelligence.

Rob Wiblin: Research.

Matthew Yglesias: Yeah. It suits us well. We got a lot of people who are working as computer programmers, and we’re saying that that’s really important. And we’re actually losing the part where we are trying to bring the gospel to the world. Because I do think that this is — not in an insulting way — a kind of a modern-day version of a religious movement, that has a distinct, Parfit-derived metaphysics, in which the ethics relate to the metaphysics and all this kind of stuff. And that’s good, but…

Rob Wiblin: It’s definitely a worldview. It’s a cosmology, maybe.

Matthew Yglesias: Sure. But the goal should be to get people into the worldview — and not just to be in the monastery, studying the higher mysteries.

Matt’s views on the longtermist community [02:19:48]

Rob Wiblin: Yeah. Is there anything you would say differently about longtermists — so people who focus on existential risks — that’s different from that, or any way that they could potentially be more impactful?

Matthew Yglesias: Obviously these are sort of aligned and adjacent kinds of things. The basic longtermist point is that we should not discount the future in the way that we do. And that’s actually a very general point that has a lot of applications to different things in the world. I just would like to see longtermists doing longtermist interventions across the board, and trying to encourage people to do less discounting in the whole dimension of things, rather than zeroing in exclusively on the existential risk problem. Which I don’t want to say is not important, because it obviously is quite important — I think we started this podcast with me trying to say to people who are not regular listeners of yours, that it should be more important.

Matthew Yglesias: But this is just an issue that comes up all the time. There’s this kind of esoteric question in American public policy about whether the Congressional Budget Office should score the cost of federal credit programs by using the market rate of interest rather than the government’s actual rate of interest. And I really firmly believe that they shouldn’t. But there’s a lot of smart people of good faith, and they really believe that the current scoring methodology is wrong — that this is an illusion that the government can save money by tapping its cheap cost of funds, because the market interest rate is the true —

Rob Wiblin: Social opportunity cost.

Matthew Yglesias: — yeah, is the true social discount rate. I would say that that is actually the high-status position, that the majority of smart wonky people want to make this change.

Rob Wiblin: But it’s deliberately making things more short term. Thinking less about the future.

Matthew Yglesias: Right. So the normal construction of this is like: lazy leftists like the CBO scoring methodology, but smart wonks understand that the market interest rate is a better source of the true cost. And I really want to bring in the big picture. I was a philosophy major in college, which I feel is never relevant to anything, but it’s incredibly relevant to this. And I’m yelling at these economists all the time that no, there is no good reason to encourage the political system to engage in more short-term thinking. That’s an insane idea. And the fact that you happen to not like the federal student loan program is not a good reason to endorse this dumb metaphysics about the world.

Matt’s struggle to become more of a rationalist [02:22:42]

Rob Wiblin: Yeah. A listener wrote on Twitter that last year you said on the Rationally Speaking podcast that you were trying to become more of a rationalist. How is that going, and what have been the obstacles?

Matthew Yglesias: It’s an up-and-down struggle. I tried to do an explicit quantified forecasting exercise on my blog. I did terrible at it — it was really painful to look back. And that experience is just a reminder that there’s a genuine pain point in trying to be precise in what you say.

Matthew Yglesias: I have found it difficult. It’s not that it’s hard to do — it’s that it’s professionally and interpersonally challenging to not make predictions that I don’t stand behind, to express uncertainty about things that I sincerely feel pretty uncertain about. I thought that I had tamped down the certainty level of my predictions, but I was actually way overconfident. So I both felt that I’ve been struggling to be less overconfident and I’m still being too overconfident.

Matthew Yglesias: And it’s actually really hard, I think, to come to grips with the scope of uncertainty about things. I was in a conversation with people, and half of them were really sure that if Republicans have a majority in the future, they’re going to change the filibuster rule. And the other half of them were really sure that they wouldn’t. And it was just hard to participate in the conversation as the guy who’s like, “I think you’re both making reasonable points, and it’s hard to say what’s going to happen, and it’s probably going to depend.” Which, in this case, I’m good to go with my forecast that “It’s difficult to say.” But it’s just hard to know how to actually participate in society. How do I pitch that op-ed, that’s like —

Rob Wiblin: “I don’t know what will happen.”

Matthew Yglesias: — that’s like, “Democrats, keep in mind that you don’t really know what’s going to happen in the future.”

Rob Wiblin: Maybe it’s a sufficiently neglected approach that people would be interested in that op-ed.

Matthew Yglesias: Right. Maybe. It’s a tough one because it’s not just done for bad faith or instrumental reasons, but so much of how people in my line of work raise alarm about things is by claiming that they are very likely. It’s challenging rhetorically to be like, “I want you to pay more attention to this improbable thing,” without phrasing it as, “This improbable thing is actually super likely.”

Rob Wiblin: I do actually see newspaper articles saying it’s really hard to say what’s going to happen with X, but I think they’re almost always in the Financial Times and directed at investors. I’m very interested to hear about, like, “It’s really hard to say what China’s going to do about trade.” They need to know.

Matthew Yglesias: It’s a good example of the structural problem with political news as an industry: it’s a cheap talk audience. And the business press is largely aimed at providing actionable information to people who arguably have something at stake — so there is much more interest on the audience side in equivocation and delivering bad news and delving into the details of things that are a little bit tedious than in the political domain.

Megaprojects [02:26:20]

Rob Wiblin: Yeah. Lately, with some kinds of financial assets getting much more valuable, in principle there might be billions of philanthropic dollars available to work on longtermist and existential risk–focused projects, if really high-quality opportunities could be found. Do you have any ideas for what ways the megadonors who generally agree with your take on things should be looking to grant their money?

Matthew Yglesias: Yeah, it’s interesting when you scale. You have to choose between continuing to put marginal money into things that you are really convinced will be good, but where the marginal return may be bad, versus you’re at the better margin, but you’re less sure about the project.

Matthew Yglesias: I think that with the influx of money that seems to be on the offing, that it is time to fund the bad projects — to be a little weird about it. Just because I think a lot of the main areas that have been clearly identified as compelling don’t have super clear funding needs right now. So things like maybe we should try to fund more media projects and try to get a large popular audience for content that has some EA and longtermist themes in it. That could be good. Maybe we need more technical analysis of wildlife population dynamics. Not because we’re convinced that the welfare of wild animals is the most important thing in the world, but because that’s what we —

Rob Wiblin: We got to go down the return curve.

Matthew Yglesias: Yeah. But it’s because we really don’t seem to be anywhere in understanding that problem. And it’s not unimportant and nobody’s worried — I mean, it’s very neglected. But that gets to be more uncomfortable.

Matthew Yglesias: So going back to the question, the initial EA wave was like, “We need to spend money on things that aren’t necessarily validated by mainstream society.” Now, I think the next thing is we need to spend money on things that aren’t necessarily validated by the highest status, biggest influencer people in the EA world. Not because there’s necessarily anything wrong with them or their ideas, but when you’ve succeeded in getting more money, you’ve got to go into the weirder spaces there.

Rob Wiblin: One way I would think of this is that, as you scale, each time you get 10 times more money, you have to diversify across lots of different approaches for thinking about things and finding stuff — because you’re just going to saturate the ones that you have already. So it’s like you have one analytical framework for finding great grant opportunities that’s worked at the hundred-million-dollar scale, but you just won’t be able to find enough stuff that fits really well within that framework and scores really well at the billion- or 10-billion-dollar scale. So now you need to think about different ways that you can get an edge, different ways to get an alpha.

Matthew Yglesias: And there’s more almost abstract research-type questions that start becoming relevant. Like, “Does Hindu nationalist politics have a positive impact on animal welfare?” Because it is preventing middle-class Indian people from increasing their meat consumption as they become wealthier. I don’t know. I met a guy who very strongly made the case to me that that was true. It seemed plausible. It was weird. I was like, really? And then he showed me three statistics, and I was like, “OK, that might be true.” As far as I know, nobody is researching that, because in a classical scholarship sense, if you went to an American academic conference and propose that, people will say, “Well, that’s weird. That’s not an interesting question.”

Rob Wiblin: That’s not what academics do.

Matthew Yglesias: But it actually is an interesting question, and we got to look at that. There’s a lot of odd angles that are worth investigating if you have a lot of resources to deploy — versus if you’re saying, Open Philanthropy spins out of GiveWell, which spins out of a couple management consultants saying, “How can we turn our consultant brain on the question of what should our personal charity look like?” And that’s a super interesting, very important question. But as things scale, you’re asking, “What do we do with plentiful resources?” And you get very different answers.

Rob Wiblin: Yeah, yeah. Speaking of plentiful resources, what do you think of the idea of founding a new university with a kind of longtermist research ethos and focus? That’s the kind of thing that could absorb a hundred million dollars, or potentially much more.

Matthew Yglesias: I think that founding new universities with different approaches is a very good idea. I think that a lot of good has come out of the Western university concept. But there are also clearly a lot of complaints that people have with it, and there is a lot of institutional isomorphism. There’s got to be something better you could do.

Rob Wiblin: It is very interesting to observe how similar universities are, like all of these quirky things that they tend to have in common. And how little of the design space seems to have been explored.

Matthew Yglesias: And the mentality that they have. So at Penn, they’ve just given up on ever being more prestigious than Harvard and Yale. They’re going to be very similar to Harvard and Yale. They’re going to be very prestigious, but a little bit less so. And they’re going to just do the same old thing. And most universities are looking up and they’re like, “How can we be just like Penn, but less prestigious than that?”

Matthew Yglesias: And that’s odd. Just as somebody who knows a lot of academics, intersects with a lot of academics, but has mostly spent my life in private business, normally you’re supposed to differentiate your product and try to defeat the competition. And academia doesn’t really do that. I would love to see someone try to create a new institution that shows them all. Like the way Leland Stanford did a while ago, like, “This is going to be a brand new thing that’s going to be better.” Except now there’s been convergence, so Stanford is just the same as every place else.

The impact of Matt’s work [02:32:28]

Rob Wiblin: Yeah. Moving on from the EA, longtermism, philanthropy stuff, it’d be good to talk for a minute about a few issues related to actually concurrently doing good with your career, and what you’ve learned about that.

Rob Wiblin: The first one comes from an audience member, and it asks: “Is Yglesias actually the popular origin of the ‘Yes, in my back yard’ movement? Because land use policy was personally off my radar completely before his book back in 2012, although some economists had written about it at various points. And then there’s been this boom in YIMBY groups in the last five to seven years.” So what’s the cause of the story here, do you think?

Matthew Yglesias: It’s hard to know, and I don’t want to overstate my role or give myself too much credit in anything. But as far as I know, Sonja Trauss was the first person to do self-conscious YIMBY organizing out in the San Francisco Bay Area, and was given money by someone from the technology industry out there. And I do believe that that early funding came from people who had read stuff that Ryan Avent and I had written.

Matthew Yglesias: We in turn were writing about academic ideas that you can find in Ed Glaeser, you can find in William Fischel, that a guy named David Schleicher (who Ryan and I both know personally) had written about on a scholarly level. So I do like to think that we played a role. Now of course, the other question you can always ask in life is, there’s the direct causal linkage, and there’s the but-for causal question.

Rob Wiblin: Yeah. What would happen otherwise?

Matthew Yglesias: Right. And I’m less certain. There were two of us who I know of doing sort of popular press takes that were very similar in very similar spaces — which makes me think that either of us would have been dispensable, and possibly that other people could have come and filled those spaces. They’re definitely not ideas that I made up.

Matthew Yglesias: But as far as I can tell — when I’m trying to be self-conscious and reflective about what I do — this is what I try to do: I take ideas that exist in the academic-y realm and that are very neglected in politics, and bring them to more people’s attention, until the point where someone has done enough with it that it becomes a conventional political conflict that people then cover as, “Well, the YIMBYs are fighting against the NIMBYs.” And at that point, I don’t think that political punditry actually does a lot. That’s the most common kind of punditry: “I’m going to write an article about why the Freedom to Vote Act is good.” That strikes me as incredibly low efficacy.

Rob Wiblin: Because it’s so saturated?

Matthew Yglesias: It’s incredibly saturated, but also it’s on the docket, right?

Rob Wiblin: I see. It’s already on the agenda. People are going to think about it regardless.

Matthew Yglesias: Right. For actual members of the United States Senate to be taking a vote on an issue is so far down the chain, and the idea that Kyrsten Sinema is going to read my blog posts —

Rob Wiblin: Crack open the newspaper.

Matthew Yglesias: — and be like, “Shit, I’ve got this all wrong.” That’s an insane level of hubris, and yet it seems to be what most people think is the humble way to do columnist work. I think it makes so much more sense to just be like, “Hey, here’s a thing,” and then maybe one person somewhere is like, “Oh, maybe I could work on that.” Or one guy who’s rich is like, “Sure, I could cut a check to that,” and you see where it goes.

Matthew Yglesias: It’s not that hard to believe that you can convince one of the world’s many, many, many wealthy people to, one time, deliver some financial support to something that nobody is doing. That, to me, is a much more realistic aspiration in life. And yet I think in my field, it sounds more egomaniacal to be trying to take credit for some early-stage grants than to be like, “I’m waging the war of ideas. I’m going to convince people that Donald Trump is bad.” Like, how would I do that?

Rob Wiblin: This is a very interesting idea. Basically you’re saying the further something is along in this process, the more people have already thought about it, the more people are already entrenched, the more it’s already on the docket — it’s just so hard to move the needle at that point, because all of the opinions have kind of gotten expressed. But if no one’s talking about something and you’re the first one to raise it, then you can actually get something that counterfactually might not have ever gotten to anyone’s attention, or might not have actually gotten a project started.

Rob Wiblin: And there you really have made a counterfactual difference, potentially. Maybe the NIMBY problem was so severe that eventually we would have ended up with the YIMBY movement as we have it, because people just would have noticed how much damage was getting done. But you writing that book at the right time and other people publishing research when they did might have sped it up by three months, six months, nine months.

Matthew Yglesias: And trying to suggest lines of action. This is pretty deep in the weeds, but Glaeser is an economist. He had this analysis of the costs of NIMBYism. But it’s really David Schleicher, who I mentioned, who had a theory that without changing people’s minds or their souls or their interests, that if you elevated these conflicts into state legislatures, that that would make it possible to make progress. And Brian Hanlon and some others in California took up that idea and ran with it, and I think they started to have real success there. And again, just helping more people hear a suggestion that maybe we should try something that literally nobody was trying.

Matthew Yglesias: So Scott Wiener is a state Senator. He became the champion of this YIMBY legislation. Some of his bills have passed, and some of his bills have not passed. Some of them have gotten support from the powerful California State Senate president, and some of them have not. If she wants to call me up and know what I think about things, I would love to have that conversation with her. But I think that trying to exercise leverage at that level is really hard — but just trying to put in the water the idea that maybe we need to make this a state politics issue rather than a city government issue… Like, why not give it a try? There’s thousands of state senators in America who could give it a try.

Rob Wiblin: Yeah. I guess we’re slightly talking our book here, because both of us do popularization of ideas, trying to make the world a better place. But when I think about this show, for example, one way that we could have an impact is that someone listens to it and makes a grant to something that was kind of off the agenda. Or someone changes their career to study an academic research topic that previously no one was thinking about.

Rob Wiblin: But there’s this other, more diffuse possible impact, which is that we just kind of change how people think in a general way — towards thinking more about cost-benefit analysis and thinking more about actual, concrete welfare benefits. And we don’t know where exactly that will cash out, but hopefully they’ll think about foreign policy, for example, better at some future time, because we got them to think in such a way that they would start to consider, “What are the alternative uses of these resources?”, rather than just going in for the first thing they think of.

Rob Wiblin: But it’s so hard to quantify that second, more diffuse thing. I think that that effect’s probably large, but I can’t demonstrate that at all, so I could be completely wrong. Maybe, in fact, posts, podcasters, and bloggers just don’t have much impact on how people think and how they approach problems in a more general way.

Matthew Yglesias: I think that’s tough. That, on some level, is the dream: to say, “We are shifting mass opinion in some small but really important way” — that there’s this enormous ship, and we are steering it in microscopic but important mechanisms. I don’t know. I’m not sure that I believe that.

Matthew Yglesias: Or I think that if we do do that, it is probably mediated by people who do work in mass culture. So that if you convince people that existential risk is an important topic, you might get more movies that treat it as an issue, and then more people will start to think that this is not an insane thing to think about. And particularly if you get multiple cuts at it — so one is a heavily satirical movie about a comet, but you could also have a really serious drama about a comet. I know that like my — I shouldn’t quite say “thinking”; it’s almost like “pre-thinking” — about COVID was very heavily influenced by Contagion, a Steven Soderbergh movie that I just liked a lot.

Rob Wiblin: Yeah. Mine too.

Matthew Yglesias: So when this started, I had a lot of assumptions that the unfolding of COVID-19 would mirror the events depicted in Contagion. And to some extent, that led me to correct beliefs, like the whole forsythia thing — that exactly happened in the real world. At the same time, Contagion really primed me to think that there would be mass panic.

Rob Wiblin: And that was wrong.

Matthew Yglesias: Which is not the case. If anything, it’s been the opposite. There’s been an elite-led effort to get people to worry more. But a lot of indifference, and nothing on the scale of, “We’re going to quarantine the whole city of Chicago, and people are going to try to flee.” Absolutely nothing like that has happened, and that’s interesting. As far as I know, nobody has ever tried to depict in pop culture a serious pandemic that is somehow not quite serious enough to induce mass panic. It’s a weird idea, right?

Rob Wiblin: Yeah, what a funny movie.

Matthew Yglesias: But it had never been there. And one of the things there is that, MERV 4 or whatever in that movie, or a superflu in The Stand — they’re all much more deadly than COVID. So now let’s tell a story about something that is substantially more deadly than other respiratory viruses, but still not that deadly. I don’t know. Nobody thought of that. Maybe you couldn’t have made it work as drama.

Matthew Yglesias: Station Eleven is another example. It’s now out on HBO, and it’s weird to see now a modern-day production of an apocalyptic respiratory virus story. Because you’re like, “But what about the respiratory virus that kills millions instead of billions?” Could there be a middle-ground story? Could we scare people about, “Well, what if it kills hundreds of millions?” I don’t know. There’s more to work with in the creative landscape than I think we already know.

Rob Wiblin: Yeah. A lot of listeners wanted to ask how you get so much done while also posting so much on Twitter. I’m a regular Twitter follower of yours, so I can see that almost every day, you’re spending time bouncing ideas back and forth, reading stuff, posting it — but you still manage to churn out just so many articles. What’s the approach?

Matthew Yglesias: I don’t know. People like to ask me this question, and I always feel like it’s a bad question to ask people, because I think that what people would really like to know is: do I have productivity tips that will help them at the margin? And I don’t. I write very quickly, so that’s good for me. It’s a lucky break.

Matthew Yglesias: The prestigious thing to do in journalism is to write lovingly crafted long-form narratives. And I don’t do that, ever. I think that there’s a quality time curve for people’s writing, and I go up really fast, and then I hit a lower peak than a lot of the best journalists. But that’s not a technique. I have tried to work for a really long time at things, and it just doesn’t get better. I don’t have any strong sense of visual memory, so I can’t describe what things look like. I can’t paint a compelling scene for people. But I’m good at writing fast, so that’s why I do that.

Rob Wiblin: Do you think that there might be kind of a synergy with Twitter? So you’re on Twitter talking to people all the time. It helps you get ideas. It helps keep you motivated, because you see these people being wrong, and that motivates you to correct them in your articles. Is actually being in the thick of this chaotic conversation maybe part of the secret of how you get a lot done?

Matthew Yglesias: I think within reason. Like everyone, I have allowed myself to be distracted by social media when I should be doing other things. But I find Twitter to be helpful. If you are good at remembering things that you see quickly, it just provides an incredible quantity of information. The downside is that it’s poorly structured, so for some people, that becomes just a waste or a muddle or a source of anxiety. But I feel that I learn an incredible amount on Twitter from responses people give to me, from other things that they see, from the people who I follow. It’s sometimes hard to know what wrong ideas have non-trivial amounts of support in the world, and that kind of helps. So I think it’s useful.

Rob Wiblin: Do you use any tools to organize and recall information, beyond just your own memory and searching for stuff that you saw?

Matthew Yglesias: No. I have a file on the Notes app where I put links, and I kind of jot stuff down, but I don’t have any great tools.

Rob Wiblin: Yeah. I think that kind of thing fell out of fashion, because people just noticed that they were spending so much time organizing their email relative to the value that it was providing. And unfortunately, I suspect for many people who were flipping between current information so quickly, it just takes too long to organize it, relative to the benefit you get. And so you just have to deal with your frail memory.

Matthew Yglesias: It also just depends. Like, what is your actual need? This is why I try to make the point about the many things that people in the writing world would like to do, that I think I don’t do well. I don’t actually meet a lot of people who say that what they would like in life is to be incredibly high-volume, journalist-takes guys. Which is fine. I mean, I love my life. I think I’ve been very lucky, and it’s successful and things like that. But there are lots of things to do in the world. I mostly think people should try to do things that will be impactful and that are important and that are useful. But part of that is that you should try to find things to do that you are good at — both because it’ll be good, but because it’s more sustainable.

Rob Wiblin: So much easier, yeah.

Matthew Yglesias: I mean, I work long hours. It’s become a little unfashionable to say that working a lot is good, because it’s like you’re giving it to the man or something like that. But I like my job. I have a job that I find very easy and fulfilling and rewarding. I think a lot of other people, though, might find it incredibly stressful, because I’m on deadline every single day. And I know people who love reporting or love ideas or journalism, but that would bother them.

Matt’s philosophical views [02:47:58]

Rob Wiblin: Yeah. As you mentioned earlier, you originally studied philosophy, and I think that definitely comes through in your writing, even if you’re not writing about philosophy topics. But I haven’t seen you have a chance to just directly address some philosophy questions lately, so I have a handful for you.

Matthew Yglesias: All right.

Rob Wiblin: First off, what’s your most unusual or most unexpected philosophical position?

Matthew Yglesias: This is, I think, going to be not at all unusual in this audience’s niche, but I buy into Derek Parfit’s criticisms of personal identity. I think that that is correct: that the idea that young you and old you are “the same person” in some way that makes you so special and different from everybody else in the world is really a kind of illusion. And I think that that’s a very unusual view in society, even though probably in this audience, a lot of people have heard about that.

Rob Wiblin: Is there a philosophical thought experiment that you particularly like that people might not have heard of?

Matthew Yglesias: Ooh, that people might not have heard of. No, the good ones are the ones that people have heard of. To slightly twist it, I took Introduction to Philosophy when I was a freshman, which happened to be soon after The Matrix came out. And my professor for that very wisely assigned David Chalmers’s essay about The Matrix, which is really an essay about the brain-in-the-vat thought experiment, called “The Matrix as Metaphysics.”

Matthew Yglesias: He’s basically trying to get you to take the brain-in-the-vat hypothesis differently. Instead of saying, “This is a thought experiment about epistemology, it’s a skeptical thought experiment, maybe everything is wrong,” maybe there’s just a fact about the world. We all, at some point in our lives, find out that what we think of as solid objects are actually made up of atoms, and those atoms are mostly empty space, and the apparently solid world around us is mostly void. And we mostly don’t lose it over that insight. We’re like, “That is interesting. That is an interesting fact about how the world works.” And so it could be that the substrate to quarks is some kind of digital representation — and that’s a fact about the world, not “Everything you think you know is wrong.”

Rob Wiblin: Is there a particularly underrated philosopher, in your view? Other than Parfit, although Parfit’s pretty highly rated now.

Matthew Yglesias: Yeah. See, I don’t know. I was trying to think of an answer to this question, so I was looking up ratings of philosophers. I was going to say I really like Quine. I think Quine’s underrated. But I saw this poll, and he was like number five as the most important philosophers of all time. So I don’t think he’s underrated if he’s at five.

Matthew Yglesias: So I can say, what has been influential to me is, I think, Parfit, Quine. Broadly speaking, I think the David Hume tradition in philosophical thought, that is a little skeptical of the enterprise of philosophy. So I came down eventually to Richard Rorty, who is probably highly rated in general culture because a lot of English professors like him — but I think he’s got a bad reputation among professional academic philosophers — but whose work I think is very important to me personally. And I think it’s a good example of how the things that you learn how to do taking philosophy classes can be useful and important in the world, even if the things that people with philosophy PhDs try to get published in journals are maybe not actually that important.

Rob Wiblin: Yeah, we’ll find something from Rorty to link to, for people who want to check that out. As is apparent through this conversation, you’re reasonably sympathetic to utilitarianism or consequentialism or worrying about maximizing wellbeing. But famously, utilitarianism — at least in hypothetical situations, if not in practical situations that we usually face — could endorse violations of human rights or individual autonomy that most people regard as unacceptable. Do you have any preferred way of reconciling intuitions about the importance of wellbeing with those about the importance of non-interference with others?

Matthew Yglesias: I think the best way to think about this is using prudential guidelines. If I do the thought experiment in which I’m killing you and I’m harvesting your organs and I’m giving them to other people, then it’s stipulated that that really is what I’m doing: that I really am saving all these people’s lives. But if we thought about, “How are we going to set up institutions that are going to function? What is the murder-and-organ-harvesting bureaucracy going to look like?”, then that’s probably a bad road to go down — there are lots and lots of much more common-sensical interventions we could make to the way our organ donor system works that would increase the availability of life-saving donations. And there would be something crazy about going to the “kill innocent people and steal their organs” concept, and that somebody who ran around advocating that is actually not being utilitarian.

Rob Wiblin: There’s something wrong in their head.

Matthew Yglesias: Well, it’s not just there’s something wrong in their head —

Rob Wiblin: They’re not being utilitarian. Right, yeah.

Matthew Yglesias: — but that’s a crazy way to spend your life, advocating in favor of that position or actually doing the murders. Even if the act could be justified, the opportunity cost is tremendous, and it doesn’t make any kind of sense. If we’re thinking about our intuitions, you say in the thought experiment that “I have stipulated all these things.” But we’re picturing a guy who, what he’s doing with his time is murdering people and harvesting their organs. And why are you doing that? That’s not a compelling utilitarian use of time. I know Christine Korsgaard would tell me that that’s a cop-out, and that I’m not taking the problem seriously, and that I’m a bad philosophy student, but I think that that’s right.

Rob Wiblin: Yeah. It’s a great frustration to me that you very often see in philosophy classes, or just education about philosophy, books or people will say, “Utilitarianism would advocate doing X — a terrible thing, X.” And I’m like, I know a lot of people who are utilitarians, and none of them advocate that. All of them think that this is completely misguided, and in fact, the consequences would be very bad. So can’t you find something that utilitarians actually do think that is bad?

Rob Wiblin: To some degree it’s a cop-out, because you could construct the situation differently — change the conditions such that they would endorse something really bad. But it is interesting that in the thought experiments that people make to try to pincer utilitarians, they actually just wouldn’t say that in the real world, that you ought to do the bad thing.

Matthew Yglesias: I always thought that Nietzsche’s critique of utilitarianism, for that reason, is relatively more compelling than the ones you generally hear.

Rob Wiblin: What’s that?

Matthew Yglesias: Because he takes on things that utilitarians actually say, and he says, “Look, this derogation of art and aesthetic greatness is wrong” — that this is a philosophy for a nation of shopkeepers. That if we imagine we are in Zarathustra’s strait of the eternal return, and we’re saying, “What was one great moment that makes it all worthwhile?”, that the utilitarian doesn’t have the answer to that. I don’t know that I agree with those takes, but they, to me, recognizably engage with a real controversy.

Rob Wiblin: With what a real person like me might think.

Matthew Yglesias: Right. And it gives me pause, because I am very comfortable saying that at the current margin, people should give money to global public health rather than to their local art museum. But now, if you ask me, “What if there weren’t any art museums?”, I’d feel kind of bummed about that. That does sound bad to me. And I’m now retreating to the idea that we’re nowhere near that margin — that being more EA, being more consequentialist, being more wellbeing focused would make the world a better place. I’m 100% percent confident of that. But my confidence goes down if I’m like —

Rob Wiblin: The further you go.

Matthew Yglesias: — Well, what if they just sold the Met and they were like, “We’re going to have an apartment building here — you know, YIMBY.” And there just won’t be art in America. I don’t know, and I’m grasping for, like, “Well, that’s not going to happen,” or “When we address all these problems, actually, there’ll be more art than ever.” And, you know, I don’t know.

Rob Wiblin: Maybe deep down, you’re a pluralist.

Matthew Yglesias: I’m more troubled by the argument because it’s real — it’s not asserting that, “Utilitarians would say that we should do racism because it increases people’s utiles.” I’m like, “But are they saying that? Who is actually saying that?” Whereas people really are saying that, like, push-pin is as good as poetry. That’s an actual position in Bentham’s text.

The value of formal education [02:56:59]

Rob Wiblin: Yeah. OK, as we get to the end of the interview, I’ve got one listener question that was extremely popular with the listeners, but I just couldn’t find any other way to fit it in, so I’m just going to do it. Basically someone wrote in: “I noticed Matt strongly advocating for ‘Don’t let kids miss any school, and schooling is super important,’ but I don’t know why he thinks that, given the arguments that formal education is mostly signaling, particularly coming from Bryan Caplan’s argument in The Case Against Education.”

Rob Wiblin: So for listeners who aren’t familiar, Bryan Caplan argued in his book that formal education is much less practically useful than it seems, and that one often barely benefits from the things that one learns in formal education. And the reason educated people do so much better in practice is because education is a way of seeming smart and out-competing other people for jobs, basically. And we talked about that in episode 32 of this show. Is there a simple way to explain why you do think it’s very valuable to go to school, and missing school or doing less school is harmful?

Matthew Yglesias: I always thought that Bryan was applying real research conclusions to out-of-sample situations, and he was going from, “At the existing margins that people operate on, we don’t see education interventions making a huge difference” to “It doesn’t matter if people go to school.” And the interesting thing to me about the pandemic — the reason I found this to be so important — was that for reasons outside of the educational domain, we tried something really dramatic, which was like, “What if small children just missed a whole year of school?” And the impacts there that have been studied seem to have been quite large, and it’s an important extra data point into the overall argument about education.

Matthew Yglesias: Now, Bryan could say, “We know that the impacts were very, very large, but also, those impacts were very, very short-term. And we’re going to have to see how that plays out over the longer term.” So if you have his strong prior, I’m not sure that the one year utterly defeats his argument, but I think it does tend to undermine it. If you had the previous prior that education’s important, this was good evidence for your view.

Matthew Yglesias: I’ve been so frustrated with the teacher union interests, because I think that they have been missing the forest for the trees here — they actually are not aware that there is this Bryan Caplan school of thought that their work is totally useless, and they don’t realize that they actually just got an important piece of information in favor of their view on a really important, big question. And they wanted to deride it because it supported their position on a much less significant, much more short-term question about, “Should they be hurried back to work, even when there’s community spread, blah blah blah blah blah?” So that is what I think about that.

Rob Wiblin: It sounds like you thought that Bryan was barking up the wrong tree before all of this and that he was underestimating how valuable a marginal primary school or high school is. Is there a way of summing up where you think he’s reading the evidence wrong?

Matthew Yglesias: Again, I think that he is too focused on small changes, and that you have to look at the rare examples of bigger changes. So there’s a study of a school strike in Belgium that lasted a very long time — but it only impacted the French schools, not the Flemish ones. But that study found very large effects. So that was why I was very worried about the school closures from the beginning, because I was familiar with some of that literature about drastic disruptions to schooling.

Matthew Yglesias: I think that he was too — I don’t think he’s wrong. I think the typical person with high SAT scores worries maybe a little too much about which university they actually attend and what it is they go on and do there. I think he’s maybe also overestimating the signaling, and it’s just not that important at all what it is you do if you’re really, really smart.

Matthew Yglesias: Being super smart is very valuable in life, and part of the value of being super smart is that you can easily recover from screwing up a lot of stuff, because you’ll impress people wherever you show up. And we have these weird cultural tropes about this, where there’s this “diamond in the rough” kind of thing, and we maybe worry about that. Like, “Oh, it’s unfair that this brilliant person isn’t as recognized as they are.” But someone who’s genuinely brilliant, they’re going to be OK. They might be working a terrible job, but then they’ll become the manager, and they’ll go up.

Matthew Yglesias: But for policy, we need to worry about average and below-average-ish people, and I think that there’s good evidence from schooling disruptions that baseline education makes a lot of difference in people’s ability to do basic reading and math. We know from macrohistory that we went from almost nobody could read to almost everybody can read, and not enough time passed for that to be an evolutionary change in humans’ intellectual capabilities. That was a policy shift and a shift in the availability of books, and that makes a big difference in people’s ability to participate in the modern world.

Worst thing Matt’s ever advocated for [03:02:25]

Rob Wiblin: OK, we’ve gone over the time you agreed. You’ve been very generous. So actually, final question now: What is the wrongest thing you’ve ever advocated for?

Matthew Yglesias: The wrongest thing I’ve ever advocated for?

Rob Wiblin: Yeah. The wrongest or the worst thing you’ve ever accidentally been a strong proponent of.

Matthew Yglesias: Oh, invading Iraq. Absolutely.

Rob Wiblin: Any others?

Matthew Yglesias: Any others? I mean, it’s hard to narrow it down to one super-specific thing. I used to be on the other side of a lot of the debates about polling and public opinion that we were talking about before — I was a big proponent of the “people are yearning for all of my ideas” kind of thing. And I think that I contributed in a meaningful way to creating some of the conventional wisdom that I’m now against.

Rob Wiblin: All right. My guest today has been Matthew Yglesias. Thanks so much for coming on The 80,000 Hours Podcast, Matthew.

Matthew Yglesias: Thank you. This was great.

Rob’s outro [03:03:22]

Rob Wiblin: One thing Matt mentioned is his attempts to eat less meat, and in that spirit I wanted to mention a new fellowship with the Reducetarian Foundation which aims to help people do just that. It’s for undergrads, includes workshops, and a summer internship with one of the Foundation’s many partner organizations, and each fellow will get a $7,500 stipend. Apply at reducetarian.org/fellowship-apply.

Speaking of opportunities for impact, a reminder that 80,000 Hours itself is hiring career advisors and someone new to lead on our job board. You can find out a lot more about both of those roles and what it’s like to be one of my colleagues over at 80000hours.org/latest. Do that soon though because those vacancies close on the 20th and 27th of February.

Alright, the 80,000 Hours Podcast is produced and edited by Keiran Harris.

Audio mastering and technical editing by Ben Cordell.

Full transcripts and an extensive collection of links to learn more are available on our site and put together by Katy Moore.

Thanks for joining, talk to you again soon.

Related episodes

About the show

The 80,000 Hours Podcast features unusually in-depth conversations about the world’s most pressing problems and how you can use your career to solve them. We invite guests pursuing a wide range of career paths - from academics and activists to entrepreneurs and policymakers — to analyse the case for and against working on different issues and which approaches are best for solving them.

The 80,000 Hours Podcast is produced and edited by Keiran Harris. Get in touch with feedback or guest suggestions by emailing [email protected]

What should I listen to first?

We've carefully selected ten episodes we think it could make sense to listen to first, on a separate podcast feed:

Check out 'Effective Altruism: An Introduction'

Subscribe by searching for 80,000 Hours wherever you get podcasts, or click one of the buttons below:

If you're new, see the podcast homepage for ideas on where to start, or browse our full episode archive.