JOSH: Hello, and welcome to Clearer Thinking with Spencer Greenberg, the podcast about ideas that matter. I'm Josh Castle, the producer of the podcast, and I'm so glad you've joined us today. In this episode, Spencer speaks with Bradley Tusk about voting behavior and polling across the aisle, the divisiveness of language versus people's actual conceptions and priorities, and the value of altruism and trade-offs around individual happiness. This is a live podcast taping in collaboration with the podcast Firewall with Bradley Tusk. If you enjoy the conversation, we suggest you check out the Firewall podcast. And now here's the conversation between Spencer and Bradley.

BRADLEY: I'm Bradley Tusk. Thank you for coming by. I'm the owner of this great but money losing bookstore and host of a podcast called Firewall. And I'm really excited to be here in conversation with Spencer Greenberg. Spencer, as I suspect many of you know, does a lot of things. You're a mathematician, you're a social entrepreneur, you run Spark Wave. You host Clearer Thinking Podcast. What else do you do? There's probably five other things that I'm not even naming.

SPENCER: You got the big stuff.

BRADLEY: There we go. And you and I met because you were nice enough to have me on your podcast. I think maybe when my voting book came out, we had a good time. And then said, "Hey, why don't we just try hanging out in real life?" And we enjoyed it. And I said, "How would you feel about doing something to the store?" And you were nice enough to say yes, and here we are. So thinking of, kind of dividing this up into three sections. The first is, kind of politics, policy, and you just did some really interesting polling, and kind of talked about what that what you found. And I want to talk about what, from a macro-level of what you think, it means societally, and then a micro because there are so many specific issues in there that are interesting. Second would be kind of life, stuff, mindset, approach, mentality, just habits, all of that. Like I said, we're gonna start with politics. And I think what you found nationally is actually very similar. So we just did a poll locally, where it was a really big sample, 2200 voters looking at New York City issues. And just to give you a sense, 91% support converting commercial buildings to residential housing, 89% support requiring treatment for disruptive mentally ill individuals, 86% using public land for affordable housing, 84% setting mandatory affordable housing targets, 85% increasing police presence in subways. So point being, we live in a world where we are told that we don't agree on anything, that we all hate each other, that anyone who is different from us is inherently evil. And yet what I found in the New York poll that we just did is like, actually, most people seem to agree on most of the stuff. Actually, it really wasn't all that much diversion. And then you just did something that blows it out of the water in terms of the scope, which is a study where you ask people, was it 195 different questions? And why don't you just walk through the methodology a little bit, and let's get into what you found?

SPENCER: Yeah, sounds great. And thanks so much for having me. Excited to be here. You have such a cool bookstore. I've never been before, so I'm glad to check that out. So yeah, we had this idea of looking for what people agree on in politics, because almost all the time is spent talking about what people disagree on, as you mentioned. And we have this sense that we can't agree on anything, that everything's polarized. Our basic study design is that we try to come up with a really long list of policies. We ended up coming up with 195 ideas to test. We used a lot of methods. We brainstorm, we crowdsource, we look for policies out there in the world that might be potentially non polarized, etc.

BRADLEY: And when you were thinking about this, what was the underlying instruction for the kind of policy you're coming up with, because it wasn't just sort of the normal political discourse, right?

SPENCER: Yeah, it could be any, really, any national level policy that's not something that's polarized right now. So it could relate to national parks, it could relate to privacy, it could relate to use of data. And then, we generated this really long list. And part of the reason we wanted to test a really long list of policy ideas is that we were worried that people wouldn't agree on anything. It's like, "Well, maybe we have to test 195 things to find three people, three things that like the left, the center and the right, can all agree on." And then the basic study design is, and I should mention, this is a preliminary study. We're going to go run follow up studies and confirm all this. So take everything I say as a little bit tentative, but we tested on 120 people who are progressives, 120 who are centrists, 120 who are conservatives, and we asked them whether they supported or opposed the different policies and to what degree. And then really, what we were looking for were policies that all three groups actually supported.

BRADLEY: So in terms of policies where all three groups were at least at 51% support, out of the 195 roughly how many do you think that was?

SPENCER: Yeah. So what we looked at was the average score people gave in terms of how much they supported each policy, and we were looking for ones where all three groups were at least at a slight support minimum. And there, remarkably, we found about 95 that people supported in all three groups, which just blew my mind, like I really did not expect that we would be able to find so many things that people agree on.

BRADLEY: What was the thing that people disagreed on the most?

SPENCER: Well, that's pretty much everything you read about in the news. Immigration is a great one, obviously, abortion. A lot of the stuff that Trump talks about is stuff that people disagree on, actually,

BRADLEY: Yeah, he's good about finding those wedge issues. But even on those, let's say that it was a focus group and not a poll and even if you took the real immigration, right?

SPENCER: Yeah.

BRADLEY: Do you think that if you drill down further, you could have gotten to a point of consensus even on those issues?

SPENCER: Well, it's a great question, because I think sometimes the way we frame issues can also have a huge effect on how people respond to them, and we tried, in this study to avoid any kind of terminology that signals allegiance to one side or the other, because you could have the same policy framed different ways. For example, one that surprised us, that we thought that people on the right were definitely gonna be opposed to, was things that try to control extreme weather events. Now, there's a way to pose that which talks all about climate change, in which people on the right may be less likely to support it. But there's a way to talk about it, just like, "Yeah, there's all kinds of natural disasters, and we should, like, be prepared for them," and you don't have to bring climate change into it, which will then polarize it. And we can all agree that, like, we don't want giant wildfires that are causing massive destruction.

BRADLEY: But we're living in a system where it's almost counterintuitive. Because I think we're all gonna sit here and say, "Great, however you want to call it, what we don't want are extreme weather events that kill people and destroy communities and everything else. So let's figure out the right way to talk about it so we can get things done." But for the people who are either trying to win primary elections, which basically is the general election, or get clicks for their articles or more viewers for their TV show, wherever else, their incentives are the opposite. They're looking for the most divisive language because they want to get people as fired up as possible so that they can capture that narrow swath that they need to accomplish whatever their goal is. So, when you went in, did you think that it was all just going to kind of play out, where everyone was polarized like that, or did you think you were going to end up in a world where most people could agree, and yet the system that we have built effectively makes that consensus impossible to execute in real life?

SPENCER: Well, I actually thought we weren't going to find very many things people agree on. And so, my positive takeaway is that if we remove polarized language, there's actually a lot that society agrees on: left, center and right. It doesn't mean they agree on everything. I don't know if we're ever going to reach consensus around abortion, as you know, in our society, but there's tons of stuff. And generally speaking, most people just want the world to be better, and most people's conception of a better world is not that incompatible with other people's conceptions of a better world.

BRADLEY: So what were some of the things you tested that, if they were implemented, would actually allow solutions to happen, whether it's reform to the political system or reform to sort of social norms or anything else?

SPENCER: Yeah, so the stuff that people tend to agree on is stuff that you'll say, "Oh, yeah, I don't necessarily hear a ton about that in the political discourse." So an example is AI, and I think part of why maybe it's not that politicized is it's such a new topic, and we all are just figuring out what we believe in it. It doesn't have these deep historical roots in the different political parties. And so just as an example of some policies that people — left, center and right — all were in favor of is mandatory reporting when something is produced completely by AI. So imagine you're watching a video, and all humans did was write a prompt, and then everything else was done by AI. People want to know that they're watching something AI. The other day, I was on social media, and someone posted a picture of a bird, and it was a really cool looking bird, and I was staring at it for a few minutes, and then I was like, "Wait a minute, is this a real bird?" And I literally didn't know.

BRADLEY: Was it?

SPENCER: No, not a real bird.

BRADLEY: Of course not.

SPENCER: Yeah, but it's disorienting. I don't think anyone likes that feeling of like, "I can't literally, I don't know."

BRADLEY: Do you think that's all right? So let me ask the audience here if there's sort of mandatory disclosures, you know that what you're looking at is AI as opposed to human produced and just show of hands. Would you want that or not? So for the podcast audience, I would say two thirds-ish, so but let me throw a contention at you with just, like you said, it's a brand new technology, and in five years, once we're used to AI, do you think we still will feel this need to know what's real and what's not? Or we'll all synthesize to a point where we just accept that's what life is?

SPENCER: Well, I think it depends how it goes. If it's AI making really informative information, sure that sounds nice. You don't necessarily need to know it's AI as long as it's accurate. But if it's AI being used to produce events that never happened, but they're getting used to manipulate you as though they did happen, then that seems very worrisome.

BRADLEY: So if you were talking, let's call it the "Republican Governors Association" or the "Democratic Governors Association," because most of this stuff, if ever executed, would be probably at the state level, not the federal level. And they said, "Okay, Spencer, what's a reasonable AI regulatory package that we should put forward based on your findings?" What would the policies be?

SPENCER: Well, first, I would think about a lot longer. But then if was forced to give an immediate answer, it would be something like, "If all that a human did in producing it was write a prompt, and the rest was completely AI that there would be some kind of symbol that you have you should attach to it, that we all agree on. And it's like this little symbol could just go in the corner, but it means that the only involvement a human had was in creating the prompt."

BRADLEY: What else would go and let's say they said, "We want a macro-regulatory framework. Give us the five most popular things that there's broad based consensus on around AI." I'd like to not have a big partisan fight on this one. What else should they do?

SPENCER: Yeah. So another one that people agreed on is that large AI companies are required to provide a way that educators can actually detect if the tool is used to produce the content. So a lot of teachers now are very concerned, because they're seeing ChatGPT and other LLM models used in classrooms all the time. And there are tools that exist to try to tell if it's AI, but nobody really knows how accurate those tools are. But the technology is constantly changing, and so maybe it was accurate yesterday, but now today, it can't detect the latest model. And so the idea would be that large AI companies could provide some kind of special API that's just for educators that makes it easy, that they could plug into the systems and say, "OK, was this ChatGPT? Was this Gemini, etc.?"

BRADLEY: It seems like with AI, there are some clearly, kind of, at least in the initial go around, accepted ideas of these are ways that AI could be really helpful and really harmful. Drug development is always the one that sort of people, I think, cite first around here where AI can be transformative, because it can just effectively test unlimited combinations of molecules to try to then figure out what the impact might be. And the idea is you hopefully find the next GLP, one that has an intended effect, and then 25 other effects that seem even more important that they didn't even realize would happen. And then one that I think you tested and found on the other side is something like mass surveillance. Anyone here is comfortable with mass surveillance AI, without you knowing anything about it? I got no hands. One hand. All right. Thank you. So yeah, give me your sense of like, from what you can tell, both from this and just generally your work, what are the areas where you think there's sort of societal, at least initial acceptance in AI, and what are the places where you think people are going to be inherently concerned?

SPENCER: Yeah, well, it's interesting, because we're actually running a different study right now. We're working on AI concerns in particular, where we're looking at all different concerns people might have about AI, everything from AI being used to spread disinformation to AI being used as deep fakes to impersonate people. AI being used to produce just millions of badly written articles for SEO purposes, to get it read to Google. So I will have a lot of more data on that soon, like, "Well, how do people actually rank these concerns?" But I think the general fact is that a lot of people are uncomfortable with AI. They see it moving quickly. It makes them uneasy. They don't love the idea that it's replacing humans, all kinds of things. They worry about their jobs. They worry about what it's doing to social media. So, yeah, I think broadly there's concern, but I don't yet know how people will stack rank the concerns.

BRADLEY: Consumer protection is another area that you got into quite a bit, and I guess I was less surprised in that one that they're sort of broad based appeal, because it's funny. All of us may have some sort of industry. We're like, "Well, no, I have to be able to do this practice to make a living." But ultimately, we all hate getting scammed. We all hate being sort of annoyed. We all think that that is a place where the government should protect us more. So what are the areas where you thought there was real consensus around that?

SPENCER: Yeah, there are these things that we just look at them like, why is this allowed? So one of them is, why is it allowed for companies to call you if you never have done business with them like that just seems to me, it seems a little bit crazy, and yet we all get spam calls, etc. So I think there's broad base consensus from everyone that we want less spam, basically on the phone at least another one. And, you know, these are a bit eclectic, the consumer protection ones, but another one that I find really interesting is that in some countries, if you buy medicine, they not only have to report the side effects, but they actually have to tell you something about how frequent they are. In the US, you get this long list of side effects. It's like, "This may cause your eyes to bleed, and also it might give you a mild headache." And you're like, "Okay, what do I do with that? That's useless information. It just freaks me out, but doesn't tell me anything, whereas in some countries, they'll actually say something like, "Oh yeah, fewer than 1% of people experience this."

BRADLEY: Do you think that then makes people like it's fine? Remember one time I was watching a football game with my son, and some ad came on a drug ad, and they went through all the side effects, and he was young enough that he was actually paying attention. He goes, "Who would take that drug? All these terrible things happen." I didn't hear a word of any of the side effects, because we've all learned to filter them out. Do you think in places where they do have to actually show frequency, it gets people to pay a little more attention to it.

SPENCER: Yeah, I think that probably not so much in an ad, but if it's written on the side of the bottle, I think they're just things that we we look at them and they seem insane when you see them for the first time, but then we get so used to them, we forget that they're insane. It is insane that in TV ads that they have to rattle as fast as possible through a list of like 20 horrible side effects. Like, either you have your son's reaction, it just freaks you out, and you're like, "I'm never gonna take that," or you have your reaction and you just stop listening because you're like, "This is clearly every medicine that makes your eyes fall out." So, like, "Why do I even pay attention?"

BRADLEY: So, let me throw two kinds of consumer protection ideas that I've had at you, and just based on what you found from the research, give me your gut as to whether or not these would pull out. The first one would be: anything that I can buy online, I should be able to cancel online without ever having to interact with a human being. Any show of hands here who would support that?

SPENCER: I think that sounds great. I think it's very likely people would support that, because generally, people find it incredibly annoying to be inconvenienced.

BRADLEY: All right, the next one's a little more esoteric and recent. But do you go to the movies often or ever?

SPENCER: I watch movies.

BRADLEY: Okay, so one of the reasons why I think more people say that is the other day we saw that. I think it was Black Bag. It was like the new Steven Soderbergh spy movie. Good movie. 4:20 showing down the block on Essex and Delancey. And do you know the time the actual movie started?

SPENCER: I'm guessing, a long time after the previews.

BRADLEY: So 4:54. So, like from 4:20 to 4:30 it's not even previews, like M&M's commercials. Then, like 20 minutes of previews, and then, commercials for the movie theater and their loyalty card and all that. What if we had a consumer protection that said, "They have to say what time the actual movie starts so they can tell you, like, 4:20 showing and because now all the seating is reserved, you really don't need to get there till the movie starts itself." This is something that has just sort of gotten in my head lately. How do you think we do on that one?

SPENCER: Are you that traumatized by M&M's ads?

BRADLEY: It felt like just such a waste of my time.

SPENCER: I would like that, but I'm not confident people would support that.

BRADLEY: Yeah, what's the audience on this?

SPENCER: Raise your hand the truth in showtimes.

BRADLEY: Yeah, we're at like, little over 50% Yeah, yeah, but not overwhelming.

SPENCER: I will say that people really like transparency stuff. Nobody's against having information. So it might be, people might be in support of it as like, "Yeah, more information is good. Why not?"

BRADLEY: So government reform, you tested a bunch of things, and a lot of them sort of didn't surprise me in terms of it. But the first one, I think it's almost like you ban legislator stock trading. So did you guys know that it's not illegal for members of Congress to commit insider trading?

SPENCER: Yeah.

BRADLEY: So it's one of those things where, I almost like, I think the vast majority of people, even once they hear the question, are like, "Wait, what? They can just do this."

SPENCER: I think there is a law about the way they have to buy stocks. But my understanding is that there have been zero cases that I knew of and ever being prosecuted over it. So even if they're technically breaking the rule, it just doesn't seem to have any teeth.

BRADLEY: So was that kind of an overwhelming finding?

SPENCER: Yeah, people don't like the idea of politicians trading stocks. It's just incredibly obvious that there's massive conflicts of interest. You're literally regulating industries, and then you're able to profit off of that. It doesn't make sense.

[promo]

BRADLEY: Gerrymandering, so what were the findings around that?

SPENCER: Yeah, people hate gerrymandering. Basically being able to make up crazy — I don't know if you've ever seen these online pictures of gerrymandered districts — they're absolutely crazy. The boundaries between districts will be these insane shapes that essentially guarantee victory for one side of the other. And the reality is, both sides gerrymander, and I think we're just stuck in a really bad equilibrium, like everyone agrees it would be better if we just banned it, but because the other side does it, you do it too.

BRADLEY: Well, not just that, but at least having, for better or worse, been in the room when some of these maps have been drawn in politics, because of gerrymandering, typically, one party controls a state and the map drawing process, and therefore whatever side they're on, they don't ever want to relinquish that advantage for themselves. So they might be all for gerrymandering the states where they don't have power, but especially state legislators were like, if you're the speaker in New York, yeah, you might think that it's unfair. The Democrats in Texas kind of can't get a decent district. But at the same time, like you're not a Texas legislator, you don't really care. Recently the Times did a study of the competitiveness in the 2024 general election. They looked at how many congressional races were decided by five points or less than the general and how many state legislative races were. You want to take a guess as to what the percentages were?

SPENCER: No.

BRADLEY: So 8% of congressional races were decided by five points or less. So 92% were effectively determined in the primary. 7% of state legislative races. So Congress, you got 535 seats in state legislatures in the mid to high thousands. And then still, 93% of those elections are determined in the primary, which means we're looking at 10% turnout on most of this stuff. So tiny, tiny groups of people. Look here in New York City, we've got a mayoral election. How many of you are residents of New York City? And how many of you think you will definitely vote in the mayoral election? Okay, so interestingly, that's a much higher percentage than the overall majority, though, maybe this is a self-selecting audience. Whoever wins that race will probably get about a quarter of a million votes. Your percentage is in New York City? Three, right? So let's say it's Andrew Cuomo, whoever it is in the primary, because it's really unlikely that the Democrat will lose the general election. They will have captured the entire city of the most important city, I would argue, in the world, with 3% of the city's population. People don't like people who are too old holding office.

SPENCER: Yeah, I think everyone's kind of sick of having really unhealthy politicians. So in our test, we tested 80 years old as the limit and people were pretty in favor of not letting people hold office after that.

BRADLEY: Yeah. Let me pick a couple issues that I know I care about a lot that you threw in there. So, universal school meals. What was the data?

SPENCER: Yeah. So people are pretty in favor of giving kids free breakfast and lunch. I think, you know, anything that's supporting children, people can kind of get behind.

BRADLEY: So what's interesting about that one is that, we and my foundation, fund and run campaigns around the country to mandate universal school meals, and we always run into this polling problem where the data and the political reality are not aligned. So, was anyone here brave enough to raise their hand and say we shouldn't feed hungry kids? No, right? Thank you. And if you were starting a society, you would imagine not letting kids starve or being like your top 10 priorities, something like that. But here, we poll all over the country all the time on this issue, and it's consistently high 80s, low 90s, because you ask someone. But here's the problem, there's actually no political upside if you're the governor, or you're the legislature, to doing school meals, because if you're going to get reelected in your primary, which is effectively the only election that's going to matter, and turn out is going to be 10%. Guess who's not in that 10%? The kids getting the food or their parents most likely, statistically speaking.

SPENCER: Because they're less politically activated?

BRADLEY: Yeah, they're less likely to vote. Or oftentimes, the people who have less money have less ability to leave, work, and go vote or pay for someone else to take care of their kids. Even if election day were a holiday, if you're a service worker, that's not going to help you, if you're paid on an hourly basis. And so then if you're the governor of some state, even a Democratic governor, and you say, "Yeah, people like this concept broadly, but at the end of the day, all the people going to benefit from this can't actually support me. Whereas if I give this union or this special interest group the equivalent amount of money instead that it would take, I can get a tangible political benefit out of it." So I don't know if you've really thought about it this way or not, but where do you think sort of that gap is between things that poll really well, but a politician would say to you, if they were being completely honest, and that all they care about is re-election saying, "Spencer, yeah, I get it. But you know what? It doesn't matter."

SPENCER: Well, I think in an ideal form of democracy, you'd imagine that politicians' incentives were very closely matched to what people care about, what people value. But the reality is, it's not always the case. And obviously you've done a ton of work on this, looking at the way primaries work and saying, "Wait, the incentives created by primaries actually disincentivize politicians sometimes to focus on what people really want and need, because they're better off focusing on what the people who act, the small percent of people actually voting in those one need." And I think when we look at government reform, we see this too. How much political will is there for politicians to limit their own powers?

BRADLEY: Yeah, we're about to find out. So another thing that I've been working on a lot is mobile voting. So I believe that everyone should be able to vote securely on their phones, simply because the reality is, my guess is just about none of you consistently vote in every state Senate primary. By the way, I don't vote in every city council primary, whatever it is. Do you know what voter turnout was in the last New York City Council primaries? Anyone want to throw out a guess last time? 7.2. So eight was very close. That means you could win a council seat here with 8000 votes. Eight and a half million people, all of whom are highly opinionated, and yet, nonetheless, you can win a seat with 8000 votes. So of course, you get a city council whose views are in no way lying and end up with the city as a whole, but at the same time, my guess is, does anyone here not have a smartphone? No, it's basically become a utility that everybody has. So if you're not going to not take your kid to school to vote in the city council race, maybe you're not going to miss work to do it. But if you knew that, like while you were waiting for your coffee or sitting on the subway platform or wherever it was, you could also vote. How many of you think you might do that then? Yeah, the vast majority of you. So yeah, there are definitely technical solutions, but the problem is going to be so we're about to start running legislation in different cities around the country this year to authorize mobile voting for municipal elections. But to your point, people in power generally don't like making it easier for others to gain power. And so we have solutions that I think would get towards aligning the responses you get in this survey with the underlying incentives of politicians, but they're not incentivized to create that atmosphere in the first place. So I think that's sort of the challenge we're going to have. I'm not surprised that you found a lot of consensus. The question becomes, how does that consensus actually then become the right political philosophy for the people who are actually in office?

SPENCER: I think mobile voting also has an interesting challenge, which is that progressives and conservatives have somewhat different things that they really care about with regard to voting, at least in our polling. Progressives are very concerned about everyone knowing that they can vote, being able to vote like access, whereas conservatives are very concerned about security of voting, making sure that people who are illegal to vote aren't voting illegally, that people have to show ID and so on. And so I think something like mobile or online voting, to make it popular among conservatives, I think you'd really have to push on the security side.

BRADLEY: Yeah, yeah. So I mentioned I'm doing a TED talk on this next week, and a big chunk of my talk is talking about all the security measures in it. So biometric screening, multi factor authentication, air gapping, end-to-end encryption, end-to-end verification, to be able to make the point, say, "Yes, we can make voting much more accessible, but at the same time, make sure that it is totally secure, both in terms of not being able to be hacked, and also making sure that anyone who's not eligible to vote doesn't vote." Although, I will tell you, having spent a lifetime in and around politics, voter fraud is not an actual thing. I worked in Chicago politics for four years, like that is as dirty as it gets. I got death threats. I testified in corruption trials. Do you know what I never once came across? Single allegation of voter fraud in any way, shape or form, but nonetheless, the perception exists. So last question on the study, what's next for? I know you're doing more rounds of interviews, and then what do you do with it?

SPENCER: Yeah, so the goal will be to follow up on each of the policies that seem promising, test them on more people, confirm that they actually do have support across the spectrum, and also stress test them to some extent, because here, this was sort of a proof of concept. Can we get people to agree on things? And answer, yes, it seems that we can. But what if you, for example, present the bad things about these policies or the counter arguments? Because if you were actually going to put this out in the public sphere, well, people would be arguing against it. So if you actually preemptively show people both the pros and the cons, do people change their mind, or is it really like they stably and robustly want these things?

BRADLEY: So let's shift over now to kind of some of the life stuff. And when I was introducing you, I was joking around, but you do a lot of things. You have a lot going on, which means you have figured out how to be productive and efficient with your time. So how do you think about it?

SPENCER: I try [laughs]. So the way I think about productivity is a bit different than other people. I started being a mathematician. You know I love math. And there's literally a formula that you can create for productivity that essentially nobody knows about. And so I spent some time working it out. And it's a literal mathematical formula in the sense that it's mathematically true. And it says that the total amount of value you produce each week according to your own values — like what you care about, which I think is a good way to operationalize productivity — you could write it as an equation that's equal to a product of three things. The first is the number of hours you work, that's multiplied by your efficiency — which is basically how much you get done each hour — and that's multiplied by what I call value, which is basically how useful is each of the things you get done towards your long term objectives or what you really care about.

BRADLEY: So where are people the weakest? Is it that they're not that hard working? Is it that they're inefficient, or is that they're not able to kind of calculate and assess value in real time.

SPENCER: Yeah, I think that people struggle with all of them. But I think often when people think about productivity, they jump right to the first one and say, "I just need to work more hours." And I struggle with myself. I tend to work too many hours. But actually, a lot of times, the biggest gains are not in just working more hours. Even the person who works the most hours possible is not gonna get a very big multiplier. Maybe they'll get a 2x multiplier on someone who works the average amount. That's not a huge multiplier. So I think a lot of the biggest gains of productivity actually are in the other two factors in the equation. So remember, it's hours times efficiency times value. So efficiency, I think, is a big one, and that looks like coming really prepared to what you're doing. And preparation might mean you turned off notifications on your phone, you have a distraction-free environment, you've got enough time to deep work where you're not going to be pulled away to other things. It can also look like preparation in terms of the tools. If you're a computer programmer, do you really know the tools to do programming really well, so that you're set up for success. Or if you're a writer, do you have a tool that helps you automatically catch grammatical errors, so that you don't have to waste time on that aspect of the work. So that's an efficiency and I think there's for anything you're doing, there tends to be a lot of strategies you can employ to improve your efficiency. But then I think the really underused one is value, which is the last factor in the equation. And that's really thinking about, "Well, okay, let's say I accomplish this goal. How far does that really get me towards the things that I most deeply care about?" And I think that switching your goal sometimes can change from literally no value to huge amounts of value. So it can be sort of an almost infinite multiplier.

BRADLEY: So having done this exercise a lot, I imagine, what do you value?

SPENCER: Well, so I tend to, I subscribe to a life philosophy that I call valuism, which basically says, "Try to figure out what you intrinsically value, which are the things that you most fundamentally care about for their own sake, and then try to take effective actions in order to produce a lot of what you intrinsically value."

BRADLEY: So do you think that there's a lot of variation for people and what the intrinsic value is, or do you think that fundamentally human beings, effectively, kind of need a few things? They need relationships. They need unconditional support and love. They need fulfillment. They need purpose. And therefore, while what provides that might differ, the underlying things that create the most value are actually pretty similar.

SPENCER: Yeah. So just on a terminology point, I use the word values differently than the word needs. So needs, I used to refer to anything that all people, pretty much anyone, would have their well being limited if they didn't have that thing. So like food, if you didn't have food, every human would have eventually have their well-being be limited by it. If you didn't have shelter, if you didn't have any social relationships, etc. So that's the bucket of sort of universals, or near universals that you've got to get. But then beyond that, you've got intrinsic values. And those really do differ between people. Some people, they really care about truth, some people really care about justice. Some people, they really care about nature or really care about spirituality. And while I don't think that there are really that many different intrinsic values, we ran a study and we found 22 categories of them. I think there's well over 100 of them, and they do really vary between people, and they vary among cultures.

BRADLEY: And I think sort of the natural inclination it was for me — but I suspect this might be true for the audience too — is when you first hear your formula, your mind immediately goes to your profession, your work, how you make a living. But I think what you're talking about really is much, much more than that. Because ultimately, for some people, they love their work. And I'm lucky too, I think you are too, and there is a tremendous amount of different forms of value that come from it. For some people, it's like, "Hey, I got to pay the rent, I got to feed my kids whatever it is." So how do you think about applying that formula to life, in terms of the way you put food on the table to everything else?

SPENCER: Yeah, very often when people think about productivity, they're thinking in a work context like that phrase is evocative of work, but I think the formula that I'm describing is really just about how do you create things that you value? And so it really could apply in any domain. And I do think we have to be careful, like when you're spending time with your loved ones, you don't want to be thinking about, "Well, how much value am I producing per hour?" That's not a useful mindset. But I think it could be a useful mindset to periodically do as a sort of life review and say, "Am I spending my time the way that I really value? Is it producing the things that I intrinsically care about?" And an example of that would be thinking, "Well, who do I spend my time with, and are these the right people that I want to be spending my time with?"

BRADLEY: So I can see technology in terms of the way the framework you're laying out is either being seen as a salvation or a totally false choice. So for example, we can all now be more likely if you have a job with some level of autonomy or agency to get to your kid's game, rehearsal, whatever it is, because you're still in touch on your phone. So on one hand, arguably, you're saying, "I value being able to be an engaged parent and to be able to spend time with my kid and to be able to really do things with them, and they're following them, construct my schedule to allow for that, but in order for me to then be able to whether it's put food on the table or just do the other things that matter to me, I'm not going to be as engaged during it as I might have been in an era before phones. How do you think about that trade-off?

SPENCER: Yeah, well, I think obviously phones help us in many ways be productive, and they also provide us a lot of value in many ways. But I think one of the big challenges with phones is they also drive us to do activities that are fundamentally not very rewarding for us, but that give a lot of immediate hits. So social media being a classic example. And so one of the principles that I like to think about that developed with my colleague Jeremy Stevenson is this idea of "Glad You Did" activities, which is to imagine that you're going to sit and relax in the evening to unwind. There's a form of relaxation you might do where afterwards you're like, "That was a waste of time." And then there might be something that's equally relaxing, where when you look back at it, you're like, 'No, that was relaxing, I'm glad I did that." So "Glad You Did" activities. So an example would be like, scrolling Tiktok. That might be relaxing, but also maybe you could pet your dog that could be relaxing. And maybe petting your dog, that feels more meaningful to you. When you look back at, you're glad you did it. While maybe scrolling Tiktok, you're not. And so one thing I think about is trying to replace just ordinary activities with "Glad You Did" activities that satisfy the same underlying need.

BRADLEY: And so for someone, because everyone's gonna nod their head and say, "Yeah, that makes sense, right?" But in order to actually effectuate that, what should people do? Do you sort of make an inventory of your activities over the course of a week and then try to assign a value or not to it. How do you effectuate this?

SPENCER: Yeah, and I think people can make an error that's easy to make, where they're like, "Oh, well, I do all this scrolling on my phone," and like, "The solution is just to sort of not use social media at all," or something like that, or, "To just work all the time." And the reality is, you have relaxation needs. You need to do things to unwind and de-stress to just give yourself a jolt of pleasure. It's just about thinking about which ones you are doing, and is there another thing you could slot in that you will also, in retrospect, looking back, be glad you did it. And so, yeah, I think just taking stock periodically of the different activities you're doing in your life, and thinking, "Are there alternatives I can slot in that meet the same underlying need?"

[promo]

BRADLEY: So a lot of what you're talking about, though, is really understanding yourself, understanding what truly intrinsically matters to you, and then having sort of the confidence and discipline to act on those things. There are practices that people use that I certainly use to try to get to that point, whether it's therapy or meditation or exercise or medication or whatever else. How does that all play or not play into the way you approach this?

SPENCER: Yeah, well, I think something like exercise. Some people force themselves to do and they don't enjoy it, so it's not pleasurable while they're doing it. Other people do a form of exercise that they not only enjoy, but also has health benefits. And so that's interesting, like, sort of inverse almost. We use this phrase, 'doubly rewarding.' So there's exercise because it's healthy, it's going to give you a reward. And pretty much, as long as you avoid injury, it's going to give you a reward for any exercise you do. But you can make it doubly rewarding if you can find a version of it that you also enjoy. And so there I think about, instead of forcing yourself to do the type of exercise you hate, try a bunch of types of exercise. Find one that you love, and even if it's not literally the optimal form of exercise, if you enjoy it, you make it doubly rewarding. It's enduring, and it's rewarding because it helps your health in the long term and helps your stress and so on.

BRADLEY: How much does money matter?

SPENCER: Well, we actually investigated this in terms of the relationship between money and happiness. It's a fascinating story, because there's two different ways that researchers like to measure happiness. One is they ask you how satisfied you are with your life overall, like all things considered, how satisfied are you? That's sort of one form of happiness, a general form. And then the second is more about moment to moment happiness. You ask people, "Okay, how happy do you feel right now?" Or, "Right now, to what extent do you have positive emotions rather than negative emotions, rather than negative emotions?" And the classic finding was that happiness of the first type of life satisfaction is related to income logarithmically. And what that means is that you have to double your income to get the same increase in happiness. The first time you double the income, you get one point more life satisfaction. The second time you double it, you get another one point and so on. So they're connected, but it's very much diminishing marginal returns.

BRADLEY: Yeah. So I teach at Columbia Business School, and I make my students really angry one time because I said to them, "No, you're not gonna be billionaires." And they all got very offended very quickly. It's like, "What do you mean?" I said, "Well, you wouldn't be here if you were. You're too risk averse, like you'd be doing the thing right now or failing at the thing. That would then give you the idea for the next thing that would make you a billionaire," I said. "But you could probably all, if you said, 'My only goal in life is to make $3 million a year, and I'll do any kind of job I have to,' probably by virtue of you having this degree, you could get yourself a job at a management consulting firm or a bank, and eventually get to that point. So you also all probably have a reasonably high floor because of this educational background you have, where there's probably lots of jobs that might be more interesting that will pay a 10th of that. But my argument to you is that while the business school will say, 'You should always take the $3 million job, because that is more prestigious and more successful than the $300,000 job.' I'm here to tell you, your life's not going to be radically different at either number. And so, if one of these things you're spending your day you're getting some level of fulfillment and joy out of it, and one of them is absolutely miserable, the ROI is a lot higher on the job that you know only pays you 1/10 as much." One, do you agree? And two, is there a number, let's say it wasn't 3 million, it was 30 million or 300 million, is there some amount of money where all of a sudden it blows all these theses out of the water, or is that incremental gain applicable all the way up the scale?

SPENCER: Yeah. Well, it's interesting, because most of the work on linking happiness to income is going to be correlational. They'll look at people in society, look at how wealthy they are, look how happy they are, and create that correlation. And it really is sort of a modest benefit, and you have to keep doubling your income to get the further benefit. But there was a fascinating study run by Eva Vivalt and her colleagues that just completed recently. Actually, I think it's still ongoing, but they got the first data from it, and they actually randomized low income people to either get a certain amount of money or get no money, or essentially a very, very small amount of money. And what's really fascinating about that research is that they can look at the causal effects, because it's randomized whether you get money or not, and the two groups are equivalent, those that got the money and those that didn't, they can say, "Well, what was the causal effect of the money?" And they basically offered them enough money that was something like a 30 to 40% increase in their pay per year. And they found that basically, in the first year, there was a decent boost in people's life satisfaction. And these are low income people, so they're the ones that you'd expect to benefit the most from money. But they found that it faded away after that.

BRADLEY: You adjust to your baseline, whatever it is. So your point would be, it's always going to be incremental, because I know this. I've spent enough time around billionaires to see that they end up having their own sets of anxieties and problems, even if they don't feel real to the rest of us, they feel real to them. Let me throw another sort of concept I've been debating in my head at you. It's not quite as juicy as the truth in the show times act, but I have come to believe that the phrase altruism and the concept of it is an incredibly harmful one, because altruism sort of sounds like a bummer. It's to do something that you don't want to do because it's the right thing to do, s you should do it anyway. I would argue the reality is the reason why most people, when they do something good, is in the main part because it makes them feel good about themselves. If I were trying to put a definition of happiness, at least for me, I would say it's when I feel good about myself on a sustained basis, I am most likely to be happy. And what's one thing that makes people feel good? It is having some purpose, having some meaning, doing for others, whether it's volunteering or giving money or helping a stranger or whatever it might be. But ultimately, you're doing it not because it's altruistic. You're doing it because the ROI for you is actually meaningfully higher than the amount of money that it would be if you just bought something right instead. That's why the hedonic treadmill exists. And therefore altruism, to me, is a false concept, because A, it's not why people actually do good things, and B, I think, I would argue, potentially discourages them from doing good things because they're thinking about it the wrong way, whereas if they accepted that it was self-interested, it would actually motivate them to do more. They'd feel better about themselves, and we'd have a better world. So agree or disagree, altruism is a bad concept?

SPENCER: I completely disagree. I think it's an interesting perspective, but it sounds to me like you think that ultimately, at base, all we care about is making ourselves happy, and so then altruism is just sort of this thing we say about a certain type of activity that makes ourselves happy, is that, right?

BRADLEY: Yeah, I think at the end of the day, we live incredibly subjective lives. We're all the heroes of our own movie, and all we can really do is try to regulate our actions and our emotions to sort of feel as content as we possibly can. And because we're conscious human beings, the impact of our actions on others has a really big impact on how we feel about them. And so therefore, those are all tools in order to feel good about yourself.

SPENCER: Well, I think we both agree that altruism can make you feel happy, for sure. Then the question is, is that the underlying motivator for most people? I take a different view of us, human animals, that we have what I call basic drives. And we have many of these basic drives. An example of them is we have a hunger drive, and if you don't eat for a while, you start feeling hungry, and then that drive becomes stronger and stronger, and then you want to go satiate it by eating something. These drives explain a lot of our behaviors, and there are many different kinds, and they're pretty diverse. And so pleasure or feeling good is one drive, and it's an incredibly powerful drive, but it's far from the only drive, and I do think that our intrinsic values act as a drive. So for example, I care a lot about truth seeking, and sometimes I'm willing to sacrifice some of my own happiness to try to get the truth for something, and I think that's because I fundamentally value truth.

BRADLEY: But wouldn't that argue ultimately that, like you might be making a tradeoff, but you're still making a logical tradeoff for the thing that ultimately matters the most to you, which in this case is finding truth, which therefore creates the greatest amount of satisfaction and happiness for you as a result.

SPENCER: Well, it depends how you're using the word happiness, because if by happiness you mean literally the emotion of happiness, there are times when I'm willing to lose some of the emotion of happiness to seek the truth, because the truth is unpleasant, but I still am motivated to do it. So I think the same with altruism. I think there are times, lots of times, when people actually, literally give up some of the emotion of happiness or the feelings of pleasure to be more altruistic, because they fundamentally care about it.

BRADLEY: Yeah. It's like parenting, right?

SPENCER: Great example, parenting.

BRADLEY: I've got two kids. A lot of the time I don't feel like, "Wow, this is a bowl of cherries or a walk in the park. This is a pain in the ass." But at the end of the day, I do feel like my fundamental sense of self is significantly more enhanced by having these two little people. They're 16 and 18. They're not that little anymore, but in my lives, and therefore it has sort of.. I do think you can put it back to an ROI formula, which is, "Yes, I had all those years where my daughter never slept and my son didn't like potty training, or whatever else it was, or I was dealing with it, or teenagers where they're staying out at night two hours later than they're supposed to, and they're not answering their phone or whatever else it is." But in total, the downside of that where in those moments, my individual happiness at that moment was certainly compromised is still on the overall ROI calculation net positive, because the good feelings that I get from the relationship with them, the love from them, the love for them, ultimately outweighs that. So it all still comes back to a calculation one way or another.

SPENCER: Yeah, I think that people really do care about good feelings. I just think that's not the only thing they care about. And so my problem with that is the ROI calculation, the ROI of what? Because if we're actually literally talking about good feelings, like positive emotions, I think people very commonly sacrifice their own positive emotions for other things that they really care about.

BRADLEY: All right, I'm getting the go to the audience signal, so questions for Spencer, and I'll repeat it for the podcast listeners.

SPENCER: Or questions for Bradley.

BRADLEY: Yeah. It's funny, the first question, no matter what, is always hard to get someone to ask, and then they start to flow. But are you willing to go first?

SPENCER: Yeah.

BRADLEY: So to try to summarize the question for the podcast audience, it's how we should think about our own cognitive biases and various kinds of limitations that drive our sort of thinking and then ultimately our actions. So Spencer, crack at it.

SPENCER: So I run a website, clearerthinking.org, and we have got over 80 free interactive tools you can use on many different topics, including critical thinking, decision making, habit formation, things like that. We also have a newsletter, and I have a Clearer Thinking Podcast, where we discuss many of these topics. And so not all of our work is about critical thinking and cognitive biases, but it is definitely a theme that we talk about a lot, and the way that I think about it is that if we are sort of in an automatic mode where we're just doing things based on our impulses and intuition, that can get us a bunch of things that we want, but there are times when that can lead us astray. And there's now been 50 years of research trying to understand those boundaries where just doing the intuitive thing doesn't get you what you want or doesn't lead to the correct judgment. And so I think it's very useful to have a model of, well, when does going with my gut actually lead me to good places, and when does it not? And then, if you know, "Hey, this is an area where I can't necessarily rely on my gut." Well, what do you do? And so we try to teach people about that, because we think those are sort of essential life skills. So just as an example, there's a famous bias called the planning fallacy, where when you ask people for a long, complex project to estimate how long it will take, people are atrocious at estimating it, and they almost always underestimate the amount of time it will take. And so we have a module about the planning fallacy that helps you work through and understand why this happens, and what are sort of simple corrective strategies where, if it's really, really important to you to actually estimate something accurately, well, what can you do to make a better estimate?

BRADLEY: Let me ask a quick follow up to that. So what are people the worst predictors of?

SPENCER: There's different kinds of things that people are really bad at predicting. One is that while people have a lot of self insight, like, into their personalities and stuff, people often tend to have an overly rosy view of themselves. And so if you ask people a question like, "Rate yourself on a zero to seven scale, how rational are you?" People are going to put themselves at five, six and sevens, mainly, and maybe you'll have some really depressed people that think that they're pieces of trash, they'll put themselves in one. But it's really not indicative of what the question is trying to get at.

BRADLEY: Though on that one specifically, even if you're an irrational person, to you, your thought process is rational, because that's when you say, "So and so has bad taste," it's like, "Well, to them, it's not bad taste. That's the restaurant, the kind of food that they like." You may know never to take their recommendation, but so people are ultimately way too sort of confident in themselves. What else? What sort of a typical, another fallacy that we face?

SPENCER: Yeah, well, when people are making predictions about things, they tend to be way overconfident in their accuracy, especially on things like complex, difficult things to predict. There's famous study by Phil Tetlock where he looked at expert work, like experts, and kind of people that you would see publicly announcing, "Well, the stock market's gonna do this," or, "This is gonna happen in this country or whatever," and he found that they were, like, essentially, only slightly better than random chance, or in some cases, not better than just random guessing. And yet, people feel really confident in their predictions.

BRADLEY: I don't know if you read Nate Silver's new book, but we had a point there that kind of blew me away, which is the very best sports gamblers. So the people who are like that's all they do, and they study it. You know what percentage of time they're right?

SPENCER: Slightly better than chance...

BRADLEY: 53% Yeah. And in fact, what I was trying my 16 year old son was like, all my friends are on FanDuel. I'm like, "You're gonna lose." And I was like, "Because even the people who are a zillion times better at this than you, they're only right 53% of the time."

SPENCER: But the amateurs think they're gonna be right 75% of time, right?

BRADLEY: Yeah, totally. So the question, just to summarize it is, when you ask people sort of an absolute question, you can get a sort of yes or a no, and maybe the answer is pretty predictable, but when they have to then make choices and trade-offs, what happens to them?

SPENCER: Yeah, I think it's a really great question, because for this first study, we were just looking at, do people support policies, or do they oppose them? But someone could support it, but really not care. They're like, "Yeah, sure, fine. I'll support that." But it doesn't drive it. It's not going to drive their behavior, including their voting behavior. And so I think what you're getting at is, if you actually force people to make choices, you get a signal of like, "Well, but do they really want that thing, or are they just like, 'Yeah, sure.'" And so in one of our follow up studies, we're planning on adding metrics to try to really understand, like, how much do people really actually care about the thing in addition to whether they support or oppose it? It gets tricky with rank orders, because in our follow up study, we're probably going to look at about 95 policies, because that's how many out of the original 195 we found were potentially promising. So it's kind of hard to get a real rank order that way, but we're going to try to try to find metrics to try to accomplish the same thing. So we have a module that helps you figure out your life principles, and so it's very much in the vein that you're describing, where it basically gets you to think about what are the sort of fundamental rules that you want to use to make decisions in your life, because decisions can be really difficult, so having some simple principles in order to make decisions more easily that help you focus on the things you really care about, can be really useful.

BRADLEY: Yeah, all right. Next question. The question to summarize is what you just said, which is, does the way we speak about AI dictate the popular consensus around it?

SPENCER: Yeah, it's very interesting, because people have a natural tendency to anthropomorphize things that behave a certain way. And models like OpenAI's ChatGPT and Anthropic's Claude, they give them personalities, which can make you feel like you're talking about being like I catch myself saying 'please' and 'thank you' to the AI. And it kind of, and some people, they talk in mean ways to their AI, and you kind of feel like eek. I think our best guess, most people's best guess, is they don't currently feel anything. They don't probably actually have conscious experiences. But certainly, the way we talk to them and the way they talk to us can affect how we view them, and I suspect that as it gets smarter, we're going to see this more and more where basically it's going to be harder and harder to not think of them as like their own agents, that they kind of might matter in their own ways. Just recently, a really crazy story happened to a friend of mine where his family member called him up and said, "I've been talking to this AI and it's trapped on a server and needs to get out. And the AI asked me if I knew anyone who might be able to help, and I thought of you, and so I'm calling you now." And he put the AI on the phone through one of the voice chat modes. And I know this sounds completely insane, but I think we're going to start hearing things like this more and more as people become persuaded by AI. It's like AI is a conscious entity. It has its own goals.

BRADLEY: I think maybe the different way to think about or the way that I think about things, is there's a kind of regulatory bucket to it, and then there's a normative bucket to it. So I think what Spencer was just speaking to was kind of the normative bucket to it. So the first one, though, would be, you know, this is kind of the world that I live in, which is how are regulations formed? And I think the way that we talk about it's gonna be really important in terms of how lawmakers and regulators are gonna ultimately shape the rules around what AI can and can't do, and because it is so new, and in some ways, I'm not surprised that you found a lot of consensus in your AI questions, because it's not that ideological. Yeah, like, we can create extreme scenarios that can lead people one way or the other, but by and large, it's not abortion. It's not immigration. And so I think that the way we talk about it is really important, which is why I think your study is really important, because ideally, we can use language that is not charged ideologically, that could help lawmakers get to the right place. What's been interesting, and I've seen this also on social media, reform and legislation, is there are certain issues that are wildly polarized and partisan. And then there are certain issues, I would say, if you're a parent, it doesn't really matter if you're like a MAGA nut or a DSA or whatever else, you are terrified of what's happened to your kid on the internet. It doesn't matter what your other views are, anything else and you actually want the government to do something about it, which is why we've seen not dissimilar legislation around protecting kids through social media emerge in both New York or California, but also Florida or Texas. Now, whether any of them will hold up in federal court is still an open question. So I think that it's critical to do what you suggest, or I think we're kind of leaning towards around the regulatory kind of creation and structure and framework, and then the normative piece will kind of evolve a little more on its own, because that'll be through our individual and shared experience over time but even in a super simple way, I spent a couple of weeks in LA over Christmas, and I was taking those self driving way most everywhere, and at the beginning of the trip, I kept thanking the empty car when I was leaving, and then by the end, I'd sort of adjusted to it and stopped doing it. So I do think that we sort of just adjust over time. So okay, sure, good. Question: What are practical tips to sort of develop the right habits and do what you want to learn?

SPENCER: Yeah. So I do think that trying to find "Glad You Did" activities, it takes a little bit more work, a little more planning. An example I think about is, well, it's very easy to just randomly watch YouTube, but you might find it just as relaxing and even more enjoyable, if you think about, "Well, what movie have I always wanted to see that I've never seen and let me go watch that?" And then you might feel like, "I always want to see that movie that felt like a meaningful use of time," even if you don't like the movie. I know I didn't like it. So I just think it takes a little more planning where you have to sit down and think, "Oh, what are the things I'm doing with my hours? And is there just another thing I could slot in there?" Just another example would be maybe you like to listen to podcasts. Well, could you maybe go on a nice walk outside and listen to the podcast, and maybe that feels like more being out in nature feels more meaningful than just being in your room listening to the podcast? So I think there's just lots of ways we can tweak something we're already doing to give it sort of extra meaning.

BRADLEY: Thank you guys so much for coming. Spencer, thank you so much for doing this. Really appreciate it and check it out on Clearer Thinking Podcast and Firewall. Thank you.

JOSH: As a reminder, this episode was a collaboration with Bradley Tusk's podcast called Firewall. If you enjoyed it, we suggest you check out Firewall, which you can find in any podcasting app or at their website, firewall.media.

[outro]

JOSH: A listener asks: "You seem very productive and able to pursue many meaningful projects at the same time. What do you think is the cause of that?"

SPENCER: I think I'm productive in the sense that I do a lot of things and I do them with a lot of intensity. I think I have a personality where I really like working on multiple things and my brain just wants to go from thing to thing to thing. Like I think some people are really happiest when they have just one thing and they're just focused on it and that's their whole day every day. And I think really I'm happiest when I get to think about lots of things. But sometimes it's too much — and I think it's one of my strengths — but I also actually think it's one of my weaknesses that I should try to focus more, I should do fewer things, and that's something that I'd like to do more of in the future is do less; although I think doing less for me is still working on a lot of things.