Intro. [Recording date: October 27, 2022.]
Russ Roberts: Today is October 27th, 2022, and my guest is author Annie Duke. Her latest book and the subject of today's conversation is, Quit: The Power of Knowing When to Walk Away. Annie, welcome to EconTalk.
Annie Duke: Well, I'm happy to be here, Russ.
Russ Roberts: Now, I couldn't help but notice that your book has a one-syllable title, which is ideal. It's a fabulous thing. But, it also happens to rhyme with Grit, a book with an apparently different perspective, but which is Angela Duckworth's book, which we've talked about on this program. But, talk about what is the difference between 'quit' and 'grit.' They sound like they're opposite.
Annie Duke: Yeah. So, yeah, the fact that it's called Quit and rhymes with Grit is not accidental. That is by design. So, let me just first say that I really don't have any quibble with the book Grit. I think everybody should go and read it.
I do have a quibble with some of the takeaways that people take from it. Which is not anything on Angela Duckworth's part, because these are not the takeaways she would wish that people took from it.
Here's the issue: is that grit and quit--those two decisions--are the same decision, and we don't think of them that way. We think of them as polar opposites.
But, I mean, if you think about it logically, any day that I choose to stay in my job is a day I'm choosing not to quit. Any day that I quit my job is a day I'm choosing not to stay.
And so, at any moment, given that we've started something, we have a choice whether to stick with it or to go and shift and do something else.
And, where we get into trouble is with the calibration issue. Right? Like, when is the right time to quit? When is the right time to stick to things?
And, my quibble with the takeaways about grit in general is that grit is good. Grit is a virtue. The people who persevere are the heroes of our stories. If at first you don't succeed, try, try again. Quitters never win. Winners never quit.
Like, Russ, if I called you a quitter, would I be complimenting you?
Russ Roberts: No.
Annie Duke: No, I'd be insulting you.
And, in fact, if you look up 'quitter' in a thesaurus, you'll see that one of the synonyms is coward.
And, that's where I really--that's where I kind of get mad. Right? And, it's a little bit why the title of the book is so in-your-face, because I am kind of mad about that. Because, I think that it does incredible damage to people in terms of their ability to actually achieve their goals, because people are getting stuck in things that just really aren't worthwhile. It isn't worth them sticking to them for fear that somehow, like, if they quit, they're a loser, or a failure, or people are going to judge them harshly for it. A variety of reasons that they won't do it.
And, the opportunity costs associated with that are so great, separate and apart from the ground that you're losing just by sticking to a loser, anyway. Right?
And, I think it's tragic. And we need to start saying: Quitting is a skill, and it's one that you should get good at. Because, unlike the idea that if you stick to things, you'll be successful. No, if you stick to the stuff that's worthwhile, you'll be successful, but you got to quit the rest.
Russ Roberts: You have a line in the book which is just quite profound: "The opposite of a great virtue is also a great virtue." And, that seems--I think most people would say, 'Well, that can't be true,' or worse, 'That's a lie. That's just ridiculous.' What do you mean by that? And, I think one of the reasons I love it is that it's memorable and it might help you make a decision that you would otherwise miss if you didn't remember that. So, talk about what you mean.
Annie Duke: Yeah. Okay. So, let me just give credit where credit is due. When I started working on the book, it was during the pandemic. And so, I asked a whole bunch of people that I know to get on Zooms with me. And, one of them was Phil Tetlock, author of Superforecasting--really brilliant man. And, they all knew that I wanted to talk about this concept of quitting. And, I got on the call with him and he said, 'I've been thinking about this in relation to grit. And, I think it's wonderful because the opposite of a great virtue is also a great virtue.'
So, he was making a play on: The opposite of a great truth is also a great truth. I think that what we need to understand is that everything has upsides and downsides. So, grit is a virtue when you're sticking to things through the hard times because the goal that you're trying to reach is worth it.
And, that is indeed a virtue because we don't--when your kid goes out on the soccer field and just has one really bad game and storms off the field and says, 'I want to quit', you don't want them to do that. Overall, if they enjoy soccer, if you think that it's something that they're getting great benefit out of, you want to teach them that it's a virtue to be able to take the downs in order to achieve the ups. Right?
That being said, quitting is also a virtue because if they get a concussion on the field, you don't want them to continue the game. And, that's what we have to remember, is that in circumstances where the world has given us new information that tells us that what we're doing is no longer worthwhile, it is virtuous to quit.
And, in fact, I would say that there's certain cases where it becomes a moral imperative to quit. Yeah. So, I'll give you just a brief example of moral imperative to quit.
So, let's say--let me come at it from two different ways. So, there's a wonderful story of quitting that occurs on the top of Mount Everest. It, in fact, opens the chapter that says the opposite of a great virtue is also a great virtue. And, I think that we think of people who climb Everest as sort of the epitome of grittiness--that these are the stories that you're telling about grittiness. But there's a wonderful story about quittiness that's there. So, this story is about Dr. Stuart Hutchinson, John Taske, and Lou Kasischke, and they're climbing up Everest. They're part of one of those climbing expeditions in the 1990s that were very popular. There's eight climbers, three climbing sherpas, and an expedition leader. And, on Summit Day where you leave from Camp Four--so you've already done quite a bit of climbing up to Camp Four from base camp--you leave at midnight.
And, the expedition leader has set a turnaround time. So, what's a turnaround time? It's no matter where you are on the mountain, if you're not at the summit by 1:00 PM, you must turn around. Pretty simple. The reason why the turnaround time is 1:00 PM is because they don't want people to descend what's called the southeast ridge in darkness. It's a very narrow part of the mountain. It's very easy to slip if you can't see what you're doing. And, if you fall, you're either going to fall to your death into Nepal or fall to your death into Tibet. Take your pick. Neither of them, I assume, would you like to do.
All right: so our three climbers, Hutchinson, Taske, and Kasischke, are climbing. And, this was at a time when the mountain was starting to get crowded and they got basically literal traffic jams on the mountain trying to get up to the summit because so many people were trying to go at once.
So, it's very slow going on this day. And, their expedition leader comes up behind them and Hutchinson says to the expedition leader, 'Hey, what time do you think it's going to be? How long do you think it's going to be until we get to the summit?' And, the expedition leader says 'Three hours.' Goes on ahead to sort of try to make up some ground and get to the summit himself. Hutchinson holds Taske and Kasischke back and says, 'We have a problem. If it's going to be three hours to the summit, it's already 11:30 AM. Seems to me we're not going to get to the summit until 2:30. That's well past the turnaround time. So, it appears we have butted up against that and we have to turn around now.' So, they did. And they lived.
Now, Russ, I'm sure it's obvious to you why you've never heard this story. Like, where's the drama, right?
I mean, three climbers followed the rules. They turned around, they lived. Like, nobody's making a movie out of it. Except they did. They were part of the climbing expedition chronicled in John Krakauer's Into Thin Air. Rob Hall was their expedition leader, in fact, the one who told them that it was three hours to the summit. Rob Hall, I think we all know, went to the top of the mountain, got there at two, an hour past the turnaround time, waited for Doug Hansen to get there until four, and they both perished atop the mountain. They never made it a bit down. They were on top of the summit.
So, you might say, 'Okay, well if it was in the book and also in the movies, maybe they just didn't talk about them because what a boring story.' But, they did. They said they were the best climbers on the mountain.
And so, first of all, the thing number one is: Why don't we even know who they are? In all of this drama, here are these people who quit beautifully and turned around and lived and yet we don't even remember them. So, I think that's important because even people who persevere in conditions that are bad, past the turnaround time that he himself had set and perish, we still admire them. We still consider them the heroes of our story.
But, this is where I think we get into a moral imperative to be good quitters, which is: Hutchinson, Taske, and Kasischke all had families. Two of them were doctors; they had patients. And don't they have a moral imperative to turn around in that situation? They know that they should, the probability of death is too high. And, now they have people that they can go back to and continue on with their lives and make those people's lives richer for their presence in them.
And so that, I think, gets a little bit at this idea of moral imperative.
I think the other place where you can see a moral imperative is quite common in, for example, in startup culture, where a startup will be clearly failing; someone will say, 'Hey, it seems like it's not going well. You're not hitting any of your benchmarks. You're missing all your targets. You haven't achieved product market fit--whatever--it seems like you should shut it down.'
And, people will say, 'But I owe it to my employees.' So, they're using the language of duty here: I have a duty to my employees to keep it going. But, if we think about it, they actually have a duty to quit. Why? Because once they've determined that the equity isn't worthwhile--and startup employees are generally working for very low cash comp when compared to what they could get on the market, but they're working for equity that they deem to be possibly life-changing. Once the founder has determined that equity is not worth it, they have a duty to the employee to allow them to go, so that they can go get paid what they deserve. Whether that's at a new startup, where they're going to be working for equity that has more value, or whether it's in an enterprise where they're just going to get salaried at their market rate.
So, I think that we turn that on its head,right? We say: 'I have a duty to stick it out because I've convinced these employees to come work for me for no money and equity. And so, I got to keep trying.' Except that the minute that you've determined that equity isn't worthwhile, the duty is actually the opposite. Just to shut it down and let them go free.
Russ Roberts: Those are incredible stories. Obviously the Everest one, slightly more incredible than the founding employee story, but they're both powerful because they illuminate a moral issue that on the surface doesn't seem like a moral issue.
And, I think your insight about character is very à propos. We often admire those people who don't quit because they, quote, "persevered," when in fact it was irrational or immoral. The story I like to tell of Fred Smith when he started FedEx and he ran out of money; and he went to Chicago to the bankers from Memphis and they turned him down. They said no. And, he was going to get back on the plane and fly back to Memphis and tell his employees that he was sorry that he couldn't make payroll. This was not a tough decision because the cash register was empty, the bank account was empty.
But, instead he went to Reno. He saw Reno on the board of departures, put all that he had--I think he'd taken money from his sister's, their shared trust fund. And he got sued for this, too, by the way. And, he goes--I don't know whether this is the money he took or he was taking it all along--I can't remember. But, he ends up in Reno and he puts whatever money he has on red or 17 or whatever it is and makes just enough to go back and make payroll. And the rest is history. And, I love that story because it's about gumption and guts and not quitting and persevering and believing in your dream.
The problem is that's the story we hear. The ones that we don't hear the ones where it was a bad dream, wasn't going to make it, and the hubris and ego of the founder--other people paid the price for that. Now in his case, he made it; I have a lot of respect for Fred Smith. Tremendous amount.
But he was a visionary. Most visionaries have a very different quitting compass. That's very bad mixed metaphor. But, they struggle to make those decisions--for ego and for just delusion. And, we celebrate the ones who make it and we don't chronicle the people who don't make it. And, that is--there are pluses and minuses to that. But, I think your observation is fantastic.
The other point I want to make is that: I just want to come back to this mantra of Phil Tetlock--past Econtalk guest, God bless him--"The opposite of a great virtue is also great virtue." One of the ones I love that is: You have to learn how to say no. And, that's a very powerful truth. It's really true.
Annie Duke: But, you also have to learn how to say yes.
Russ Roberts: Correct. Even sometimes saying yes to things that don't look promising, lead to extraordinary changes in your life. And, so--
Annie Duke: That's right.
Russ Roberts: All these things are a question of nuance, I think, and balance. Before we--
Annie Duke: Actually, speaking of Phil Tetlock, during the pandemic when I was somewhat busy, he reached out to me and said, 'We're having trouble creating good training for novice forecasters and these counterfactual forecasting problems. You kind of teach this stuff and consult on it. So, maybe you would be able to put it into terms or a voice that would actually create a good training and you could maybe think about the things that actually work with your clients and apply that to this training.' So, I said, 'Yes.' Why? Because I love Phil--and Barb, by the way, his wife--and I was willing to make time for that.
And, that turned into four very large-scale studies that were incredibly fruitful. So, I completely agree with you, right? I'm trying to work on both, right? Being more careful about saying no to things that I'm predicting are not going to be worth my time. And, saying yes to stuff that looks kind of wild and crazy, but wouldn't that be cool? And, I might learn something super new about myself or something super new about the world. So, I love that example because that's a good case of, like, the yin and yang, right?
Russ Roberts: The reason I like it is, there's this other piece to it for me, which is: you might make a human connection that you otherwise wouldn't make, that's not going to make you more money, and it's not going to lead to all those other studies, not going to help you understand something. You're just going to have a human experience that's precious. And I love that. It's very powerful.
I think in the case of the yes/no, what we're saying is you have to make room in your life for serendipity. There are things that are going to come along you can't predict, can't imagine. And, if you always say no, you'll be comforted by the fact that you had more time for other things. But, you'll never see the things you didn't get. And, you write about that a lot in the book actually. Very thoughtfully.
Annie Duke: Yeah, so actually I'd like to--in relation to that, I'd like to bring up a little fact about ants, because I think this goes really well with that. So, you know the song "The Ants Go Marching One by One, Hurrah, Hurrah," right? So, we know that we have that image. If you've seen any cartoon or you've actually watched ants on a nature show, they're marching in a line, right?
Russ Roberts: Yeah, they're really good at that.
Annie Duke: So, those ants are forager ants. They're a part of the colony that's meant to go out and find food, basically. And so, if you watch these forager ants approach, like, a new territory, you'll see that they're all kind of scattered around. So, they're not marching in a line, yet. And then, one of them will find food and they'll take the food and they'll be carrying it back to the colony.
And, on the way back they lay down a pheromone trail. So, it's just a chemical trail that the other ants are going to detect.
So, they're only doing it on the way back: once they've found food, they're laying down this trail.
So, at first it's pretty faint because it's only one ant. But, now if another ant detects that trail, it will now go along the trail; it will find the food.
And then, when it's bringing that food back, it will also lay a pheromone trail down on top of that. And, you can see how this trail is now getting reinforced, attracting more and more ants to the same trail until they're marching one by one--to whatever, the watermelon that fell on the ground. Right?
So, that's how we think about them. But, actually if you look at the behavior, once there's a strong pheromone trail laid down, what you'll see is about 10 to 15% of the ants don't actually get with the program. They're just kind of wandering around.
So, what's the deal with those ants, right? Are they ant-anarchists? Are they malingerers? Like, what's the deal with these malingering ants?
And, it turns out no, they're not anarchists at all. They're not malingerers. They're actually serving an incredibly important function for the colony, which is that they are continuing to explore. So, you've got the ants that are exploiting the food source that's high-quality, a watermelon or whatever. But, the other ants are continuing to explore. So, they're saying yes in that sense, right? They're like, 'Yeah, sure, I'll keep go looking around.'
And, why is that so incredibly important that they're doing that? Well, first of all, the food source might go away. So, someone might clean the watermelon up. Like, maybe it's on the back deck or something like that and someone comes out with the hose, and then that watermelon is gone. It's really good that this 10 to 15% of the colony is continuing to explore other food sources because it means they have backups.
Russ Roberts: Insurance.
Annie Duke: It's insurance that allows them to sort of cover--to your point, they're increasing the chances for serendipity, for finding something else that's really great.
The other thing--and I think that this is an overlooked point--is that it may be that the food source that they have is totally stable, but the other ants might find a better one. And, that's the issue of opportunity cost--right?--is that once we're exploiting something, whether it's a product that an enterprise is selling, or a hobby that we're pursuing, or a project, a job, whatever it is, once we're doing that, we tend to cease to explore.
So, I think it's funny that a lot of the encouragement is around saying no, because I think we're actually quite good at saying no, because we actually don't even consider the possibility of saying yes or no. And, if you don't consider the possibility of saying yes or no, you're saying no to all of that stuff by default. Right?
So, what the ants are doing is saying, 'Well, this is great. I love that. But maybe there's something better out there.' And, they're continuing to explore it. So, it's serving dual purposes: It's giving them a backup plan, but it's also allowing them to find something that really ought to have been their Plan A.
And, I think this relates exactly to what you're saying, right? And you can see this behavior, this duality in the ants, because they're doing both things at once. They're exploiting the food source that's there, but they're also continuing to explore and basically say yes to all the other places that you could go look and they're more likely to find something.
So, obviously we're not ants. We don't have a big colony. I can't clone myself. But, to your point, I can say yes to stuff. And if I say yes to stuff, maybe I'm going to find something there that's awesome or a good backup plan or better than what I'm already doing.
Russ Roberts: Yeah. I'm going to say something about quitting that I'm curious to get your reaction. It's a personality trait of mine and I've often thought of it as a flaw, but it maybe it's a feature, not a bug; and it's consistent with your point. I tend to get very excited about new projects; and I'm not the best collaborator and I haven't been until I got in this job as President of Shalem College. I kind of picked things where I didn't have to collaborate. Right? When you're a research fellow at the Hoover Institution, it's a deliberately lonely life. It's not lonely--it's just that you're often alone; and you can collaborate with other people in your field if you want, but you're also free to just work on what you love. And it's really lovely.
But, when you have to collaborate, in my experience of my own self, if I get really excited about a project and I need your help--okay?--or we're going to do it together; and I tell you about and I'm all fired up; and you go, 'Well that's really cool. I like it, too. That's fantastic.' And, then nothing--you don't follow up, you don't respond. Maybe you got busy. Maybe you decided you didn't idea as much as I did initially. I lose all my enthusiasm. Right? I have very little self--because I'm going to find another one. I'm like that ant. I'm going to go off, I'll find another thing I'm excited about, and I'll find somebody who does want to do it. Or, I'll get you fired up about the new one. And, I've always wondered whether that's a character flaw that I very quickly give up on what I was so excited about to start with.
And, now you're making me feel better about it. It still could be a character flaw, but I think it's a recognition of the opportunity cost. If you're not enthusiastic and you're not following through with me and I'm going to have to then carry the ball by myself, I deflate. I'm done. I'm going to find a new project. I'm going to wander off, find a different piece of watermelon. And, I think I've never thought about that as a possible good character trait. Maybe it is.
Annie Duke: Yeah. Well, okay. So, here's the thing, though. I mean again, it depends on your values: it depends on what the signals for success might be for you. So, for you, it sounds like you don't feel like you're going to be successful in a project if you don't have a collaborator who is equally enthusiastic. That's what it sounds like.
So, this goes into something that we could call 'kill criteria.' If you want to be softer about it, you could call them 'exit criteria.' I like the term 'kill criteria' for the same reason that I put in very large letters the Quit right on the front of my book because I want people to think about these things this way, in the boldest possible terms.
But, at any rate, so for you as you're thinking about, 'Ooh, I'm kind of interested in this idea; I want to explore this. What are the things that would tell me that this isn't going to be something that's a really valuable use of my time? Well, if I can't get a collaborator to be as successful as I am, then I really ought not to do this. So, I'm going to do very little thinking about it. I'm going to form enough of an idea to be able to communicate it to someone who I would like to collaborate with. And, if they're not as excited and engaged as I am, I already know it's not going to be worth my time.' Right?
So, that's actually a really reasonable way to approach a project. So, I actually approach books that way. When I have an idea for a book, there are a few people that I call. I hardly form the book idea, I'm able--with Quit, it was something like, Quit, the opposite of Grit. I said that. And, I don't mean, like I said, I don't mean, like, the opposite, but I mean the dialogue with Grit: that, I think that people in general think that we quit things too early. I think the science tells us that we quit things too late. And, I would really like to explore this topic. That was about what I had.
And, I wrote, I think the first person I wrote was Michael Mogenson. But then, I think Danny Kahneman followed quickly after that, and Phil Tetlock, because I just wanted to see how did they react to that. And, then they were really excited. They were like, 'Okay, yeah.' So, then I'm like, 'Okay, I think now I should go further.' Right?
So, I'm always sort of pushing to see, like, is this a no or is this a yes? And, I know that if can't get--if Danny Kahneman thinks it's a stupid idea and it's not worth exploring, that's a really good signal for me. So, I shouldn't put a whole lot of work into it until I've got those gut checks from people who are way smarter than me, much deeper into the science than I am, and are going to tell me whether it's something that they think is worth putting on a piece of paper.
And obviously this has to do, in particular, with what I like to write about, which is, to be fair, their science. So, if the people who created the science don't think it's worth writing about, I'm not going to continue with it.
Russ Roberts: The parallel thing with Grit is: But don't you have enough faith in your own idea? What, you're going to rely on other people to decide whether this is a good project for you? And, I wonder sometimes when I get shot down with a creative idea--a not really outside the box idea, something more crazy than just, 'Here's an idea for a book,' but I propose something absurd--I wonder if--and I get shot down--I go, 'No one likes it.' It's very hard for, I think, most of us emotionally to then say, 'But I think I'm still right, especially if the people we're asking we respect and are smarter than we are.' And I think, I worry sometimes that I cast my own decisions in that kind of light that, 'Yeah, it's a rational decision because I needed to. They didn't think it was worth it. They're smarter than I am.'
But, sometimes I wonder if it's just like: I'm a quitter. And, it comes to your point about the cultural baggage that we have--mostly from our parents. You gave the example of the kid on the soccer field. A lot of what we do as parents and a lot of what our parents did to us is to get us to push through pain. Because often--not always, but often--great rewards come from that. And that is hard--for human beings to anticipate those rewards sometimes. Especially when we're young, we have trouble.
Annie Duke: And, let me just emphasize that: Especially when we're young. So, I just want to emphasize that, because I think that, separate and apart from where parents go wrong with that--and they do, they take it too far--but obviously it's a good lesson to take someone who's six and has never gone through the downs to see what the ups might be on the back end of it and teach them you don't need to quit. You can push through it. I agree, especially when they're young.
The problem is that we think that applies to 30-year-olds. That's the problem. And it doesn't, because 30-year-olds aren't walking off the soccer field. That's the problem.
Russ Roberts: Yeah. Well, I want to take another example that you use in the book that will apply to the Everest example. And, I think it's an incredibly poignant and powerful example of it. The example you use is that if you finish a half-marathon, people are impressed. 'Wow, you ran 13.1 miles. That's a lot.' But, if you run a marathon and you stop halfway, you're a quitter--and you did the exact same thing.
Annie Duke: That's right.
Russ Roberts: And, I think about the absurdity, the utter tragic absurdity, of being 300 meters from the top of Mount Everest--
Annie Duke: And, you're a loser--
Russ Roberts: and, it's one o'clock, and you're supposed to turn back. And you say, 'I'm not going to stop short of the summit. I can see it from here.' And of course, the answer, one answer you should give yourself is: If I can see the summit and I'm 300 meters away, didn't I kind of do what I wanted to do? Can't I--
Annie Duke: To that point, you know, there was something interesting because these are all cognitive phenomena. Right?
So, one of the things that I want to be clear about is that what we're talking about is the cognitive state of being in the losses. So, when you think about your balance sheet--right?--like, 'in the losses' means that you're losing from whatever a mark was. Right?
So, if you buy a stock, the mark is going to be the price that you bought it at. And, if you're below that, you're 'in the losses.' If you're above it, you're 'in the gains.'
So, that would be on an actual ledger, right? On an actual balance sheet. But, we have this mental accounting that occurs, which gets distorted. So, sometimes it overlaps. If I buy a stock at 50 and it's trading at 40, both on my physical ledger and my cognitive ledger--my mental accounting--I'm in the losses in both.
But, if I buy a stock at 50; it goes up to 75 and is now trading at 60, on my actual physical ledger, I'm in the gains $10. But, in my mental account, I'm in the losses $15 because I'm 15 short of 75 now. Right? Okay? So it doesn't matter that I was up 10.
So, when we take, like, a marathon--and this really interesting thing about a half-marathon versus a full marathon or where we are in comparison to Everest--if it's a half-marathon, the goal, the end point, is 13.1 miles. So, if I complete that, I am now no longer in the losses in comparison to that goal. But, if I only complete 13.1 miles in the context of a marathon, I am short 13.1 miles now: I'm in the losses. No matter that, if I created a physical ledger, I would be in the gains 13.1 miles.
In other words, physical ledgers measure from the starting line. But, mental ledgers measure from the finish line.
So, this is the problem we have with Everest, right? I'm 300 feet from the summit. Never mind that I just climbed 29,000 feet in the air. I'm a loser if I turn around because I'm closing that mental account in the losses.
So, if you wonder why does somebody continue past the turnaround time or even get to the summit at 4:00 PM--which is what Doug Hansen did, even though the turnaround time was 1:00 PM--it's because he was in the losses--
Russ Roberts: In his head--
Annie Duke: And, as Richard Thaler--in his head--so, Richard Thaler points out, we do not like to close mental accounts in the losses.
So, anyway, Richard sent me something hilarious. It was like--it's probably about a year ago and it was a little bit complex. It didn't end up in the book. But, basically there's some sort of argument now, among mountaineers, that if you look at, like, the popular peaks that people climb, there's some argument about what the peak actually is.
Russ Roberts: Oh, love it.
Annie Duke: So, now, all of a sudden, they're saying that a bunch of people who say they've done, like, the seven peaks or whatever, the seven summits, maybe they didn't actually summit them because there is now an argument about what exactly is the top of Everest or what is the top of Tillman Borough[?], which just brings up the absurdity of all of this in the first place. It's completely absurd, but it's the way that we work cognitively.
Russ Roberts: Right.
Russ Roberts: But, I'm going to push back a little bit because I do think there's a powerful reason that we struggle with this mental accounting. Right?
And, anybody who has run--who has been a runner or done the equivalent of running in a project, meaning a long, arduous trek--I think understands this.
And, I just want to say, and I used to keep this quiet because I thought it was too, it wasn't sufficiently humble. I ran a full marathon when I was younger and finished in the blazingly fast time of four hours and 20 minutes. But, the fact is I am very proud of that. And, I'm proud of the fact that I finished. The fact that for the week after I couldn't climb stairs without a great deal of pain--let's put that to the side and let's ignore the fact I could have really done some long run damage to my body.
But, the reason I finished--and it was painful. I wasn't spitting up blood or anything and a bone wasn't sticking out of my leg. But, the reason--it was hard--the reason I finished is partly because of my dad. My dad said 'Don't quit, finish what you plan.' This whole idea of this mental accounting.
And the reason that's useful--the flip side I think of your argument is--if you start off to climb Everest or run a marathon and say, 'Well, I'll just get as far as I'm comfortable and I'll try to get far and whatever it is will be gravy. If it's five miles, great. If it's 13.1, I'll be proud. Twenty would be wonderful. And, if by some chance I finish, 'Oh that's nice.'
You don't get very far. Often, that we feel--at least, maybe it's wrong--but we feel that if we take that approach, we're going to cheat ourselves. We're going to quit too soon.
So, instead we go to the other extreme. Which is insane. Which is: 'Got to finish, got to finish, otherwise I'm a loser.'
And, we use that as a way to push past short run pain for long run benefit. It's why we go to grad school. It's why we invest in a startup. It's why we run marathons.
And, a lot of it by the way, of course, is self-esteem. I mean, we didn't talk about this, but when I read Into Thin Air, for me--and it's a great read; if you haven't read it, it's an extraordinary read. I finished that book thinking: This is insane. This is--to what purpose did this person lose half his nose? To what purpose did these people die? They didn't achieve anything. And of course, their answer would have been, 'No, I tested myself and was not found wanting.'
And, there's something deep inside us that needs that. Whether it's hit the approval of our parents--often no longer alive, we don't care--we still push through. There's value to it. And, it's also--in many ways what you're saying is it's like a sickness, almost. And, it is a little complicated that way.
Annie Duke: So, this is what I would say. The opposite of a great virtue is also a great virtue. And, that's true when it comes to goals that we set for ourselves. Goals are motivators: as you said, they get us to push toward the finish line even when things are hard. And, that is not a bad thing.
Now, I would argue that if you enter a marathon saying 'I'll run as far as I feel comfortable,' that you're still going to run toward the finish line, because, no matter whether you say that to yourself or not, there's a finish line. You're not going to want to quit before you get to it. So, I have very little concern about people saying, 'But if I start, if I'm not feeling good,' whatever. I think that creating an all-or-nothing situation around the finish line is probably not helpful since we already do that in our heads.
But, it's true. Right? Like, if I'm on the last two miles and my legs are cramping, it'll make me finish. And, I'm going to feel pretty good about that. I'm going to look back on that and feel pretty proud of having pushed through. It's probably a good thing that I did, assuming that I wasn't in medical danger. Right?
But here is the problem. This is where we get to the 'opposite of a great virtue is also a great virtue.' Let's take Siobhan O'Keefe. Siobhan O'Keefe entered the 2019 marathon. On mile four, she started experiencing pain in her leg. On mile eight, her fibula bone snapped. She broke her leg. Now obviously the medical tent was like, 'Yo! Stop running.' But, she did not and she finished the race.
Now, this is where we get into trouble with this, right? Because the great thing about goals is that it sets a finish line and it gets you to continue to run toward it even when it's hard. The bad thing about goals is that it sets a finish line and it gets you to run toward it no matter what. Even when your leg is broken.
And, if we take, it creates--in some ways, it creates a short-termism, right? Grit is really meant to help you with the long view: 'I know it's bad right now, but it's going to be worth it in the long run.' But, weirdly, when we set these goals, it creates a short-termism, because the goal itself becomes the object of our grit. Whatever that short term finish line is. Because I assume for Siobhan O'Keefe, the goal was, 'I love running marathons and I would like to run many of them in my life.' This was not her first marathon. And, by continuing to run, she was risking grave injury that might have prevented her from ever running another one.
So, she was actually causing herself to lose ground toward what she herself had declared would make her happy. And, that's where we get into real trouble.
And, this really goes under another thing that Richard Thaler talks about: is that goals are really graded pass/fail. And, not only does it mean that you're going to head toward them no matter what, but he points out that it can stop you from starting things that are worthwhile. Because as he said, if the only thing that is success is getting a gold medal in gymnastics, why would you ever take your first lesson?
Russ Roberts: Yeah.
Russ Roberts: You were a very successful poker player. One of the things that's fun about poker, it's a game everybody can play. Many people have played it casually, and most of us who played it casually had no idea what the real serious poker is until--and then it became a TV phenomenon. And, I think a lot of people got access to it, and it created great prizes, and so on. But, I think it's another wonderful, simple mantra for people to think about outside of poker, which is: Knowing when to fold 'em. And, talk about your own experiences, what you learned from poker. First, talk about what you achieved in poker. You showed some grit: you pushed through; I'm sure you pushed through a few losses.
Annie Duke: I played for 18 years. I did quit too late--which isn't surprising. One of the hardest things to quit is who you are. And, when you're on television known as a poker player, gosh knows, that becomes your identity. And, if you walk away, what does that mean for you? I think that's very hard.
So, here's the thing about poker. So, first of all, obviously the power of knowing when to walk away, as a nod to Kenny Rogers: When I was playing poker, I would get very annoyed because any time that I went on the radio or was going on a television show, they would usually play that song: 'Got to know when to hold 'em, know when to fold 'em, know when to walk away and know when to run,' as I was coming on--as the intro to me coming on. And so, I grew to hate that song.
Russ Roberts: Imagine.
Annie Duke: Eventually they switched it to Poker Face, by Lady Gaga. So, that was a little bit better.
But I decided I was going to reclaim the song so that I could love it again. And, here's the thing--and I think Kenny Rogers actually says something very insightful here about the game of poker. You got to know when to hold 'em--that's about sticking. Know when to fold them--that's about quitting. Know when to walk away--that's about quitting. Know when to run--also about quitting. So, 75% of the refrain is about quitting, not sticking. And, that's actually very much true to poker.
So, I think that when people think about what makes a poker player amazing and they were going to list off, It's a real amazing ability to read the other player's hand, super-aggressive. Like, they're bold and courageous and pushing all their chips into the pot.' Well, first of all, let me just say: Great poker players try to avoid putting all their chips in the pot. They're very picky about when they do that. They actually play something called small ball, more. But, that's a whole other story.
But, regardless of that, here's the thing: if you really want to know what separates great players from amateurs, it's quitting. So, folding is quitting. So, quitting is just cutting your losses. That's all it is, right? And so, in game theory, all it means is stopping something that you've already started.
Russ Roberts: And, it means finalizing that loss in the ledger, right?
Annie Duke: Right, exactly.
Russ Roberts: It means accepting it and closing that part of the ledger in the negative, in the red, when there's always a chance you might get that inside straight.
Annie Duke: Right. Exactly. That's something that's been well-documented--originally really in 1979 from Kahneman/Tversky, that when we have those losses on the books and we have to now quit and turn those into a sure loss, that we'll become risk-seekers. In other words, we want to keep the gamble on, actually a little bit to the FedEx guy. Right? He became a risk-seeker and went and gambled this money, which is nuts.
Russ Roberts: Literally. He went--
Russ Roberts: But, it was--either way it was a gamble, right?
Annie Duke: Yeah. But, it was nuts to go. And, I guess I would've sued him, too: 'What are you doing?'
But, regardless, we become risk-seeking when we're in the losses. Now in order to keep risk on, that means you can't fold, because fold is risk off. All of it, right? It's like I'm going to take all the risk off. So, that's why we think about loss cutting.
So, amateur players are terrible at this. So, when you look at amateur players, when you look at their first two card starting combinations that they get dealt in the game of Texas Hold 'Em in a full-handed game, they'll play over 50% of the two-card starting combinations.
Now, I just want to remind you that there's nine people at the table. So, if you think about sort of what's their fair share, it would be one ninth. Now obviously you don't just play your fair share, particularly if you're good. But let's agree that an amateur should not be playing over 50%, given that there's nine players at the table. That's probably not a winning idea.
Russ Roberts: My mom plays a hundred percent, and if she could play 120, she would, in our family poker games. 'I want to see the cards.' 'I get it, Mom.'
Annie Duke: Yes. Yes. Okay. So, professionals will play between 15 and 25% of the two-card combinations that they're dealt.
Now think about this: The relationship between how many hands you play and how good you are is correlated. So, the better you are, the more hands you're allowed to play. And, the reason is that you can get more edge over the other people at the table in terms of the choices that you make later that would allow you to play more than your fair share of hands, right? Because, like, 25% is more than your fair share at a nine-handed table, except that you're going to be better than everybody else at playing them. And so, therefore you're allowed to. Right? But, amateurs are playing over 50%. This is part of the reason why a pro can play 25%, because they're playing all these really bad hands. All right? So, that's Number One.
Now, to the point of your mother: What are the reasons that somebody won't fold there? There are a variety of them. But right at the start, part of it is: It's so painful to fold a hand where you then see the rest of the cards and you realize you would have made something good, that people refuse to do it. There's a saying in poker: Any two cards can win. And what professionals add onto the end of it very quietly is: But not enough of the time to be profitable.
So, but, it's like the number of times someone leans, 'Oh, I folded a seven/deuce and look, now there's a seven, seven, and a two on the board.' It's like, okay? Right? I mean there's a few people who had continued up Everest and they lived--
Russ Roberts: Yeah, exactly--
Annie Duke: I mean, they got frostbite and stuff, but they lived. Right? Or the FedEx guy--I mean he lived, right? But, would we know the story if he went and gambled, and then he lost on that bet? Not in a million years.
And that's almost always going to happen. Particularly, by the way, if you're betting 17--right?--which is one number. So, mostly you're going to lose all your dough and nobody's ever going to hear about you. 'But, what if?' Right? So, these what-ifs, these counterfactuals, are incredibly hard for us.
And so, what ends up happening, and part of the reason why people play these hands, is that once you've started something, the only way to know for sure how it would turn out is to keep going. And, what that means is that we're going to butt up against the certainty that there is nothing else we could do but fold or quit before we're willing to do so.
So, your mom plays the hand because she wants to see the next cards to be guaranteed that they have no relationship whatever to her hand. So, she has no regrets about folding because she understands at that point it's a certainty that she could not have won. But, that is long after the point that it's correct to walk away from things, because you're already fallen into the crevasse. You're already at the top of Everest and it's 2:00 PM.
Russ Roberts: But, you don't know my mom. She's really lucky. So, it turns out it's a good strategy for her.
Annie Duke: Well there you go. Good strategy for her.
Now, as we get back to Kenny Rogers, the other thing amateurs do poorly is fold after they've already entered the pot. Okay, so once you decide to play the hand, you now are putting money in the pot. Okay?
So, what you'll hear people say--they literally say this out loud--is, 'Well, I couldn't fold because I had too much money in the pot already. I had to protect my chips.' And, it's like: Not your chips anymore. That's money already in the pot. What matters is: Is the next dollar you're going to put in the pot worthwhile?
And, this is just very, very classic sunk-cost effect. We take into account the resources that we've already spent in deciding whether to continue and spend more. Those resources have nothing to do with it. The only thing that matters is: if I bet a dollar here, am I getting a positive return on that dollar? Doesn't matter that I already put money in the pot. I shouldn't care about that. But, boy, amateurs really, really, really, really care about it.
Russ Roberts: The flip side of that is, 'Well, I didn't lose the money, because it wasn't mine. It was the house's.' Well, once it's in your pile, it's yours and you gave it away.
Annie Duke: Oh my gosh. Right? Right. That's so true.
Russ Roberts: I want to come back to sunk cost. But, I'd like you to reflect, if you can--and I don't know if it's useful--but I'd like you to reflect on how that experience of 18 years--you know, looking at amateurs and saying, 'Hmm. That's a mistake. I will not make that one.' Or 'I made a mistake and I won't do it again.' Did that spill over into the rest of your life or were you just a really good poker player?
Annie Duke: No, it spilled over into the rest of my life. Most of the things that I learned from poker spilled over into the rest of my life.
Now, I will tell you that there is evidence that people do get better at this stuff with experience. In other words, when you see a lot of iterations of it, you get better.
So, an example of that would be--so there was a very large scale study that was done by Colin Camerer, along with a few collaborators, including Richard Thaler, where they looked at trip sheets from cab drivers in the 1980s. This was before Uber, obviously. And, back then what you would do is you would rent a cab, because most people didn't own the medallion. You'd rent a cab for a 12-hour shift and then it would be up to you, Russ, to decide when you wanted to drive during that.
So, what they wanted to understand is--because they had the trip sheets--were cab drivers driving a lot when there were lots of fares and driving very little when there were very few, which would be a rational strategy.
In the same way in poker, I want to maximize the time I'm playing well and minimize the time I'm playing poorly. I want to maximize my time in good games, meaning where the other players are quite bad, and I want to minimize my time in games where the other players are quite good. So, this is just generally what we want to do in life.
And so, they wanted to know if that was what those cab drivers were doing.
And, they found something very surprising, which was that when there were lots of fares, the cab drivers would quit really fast. And when there were very few fares, they would go forever.
So, this is going to bring us back to Mount Everest and the marathoners.
In fact, this, by the way--they had this strategy was so bad that compared to a rational strategy, they would've made 15% more money if they had actually followed what a rational actor would do. And, in fact, if they had just said, 'I'm getting the cab; I'm going to drive for six hours no matter what,' they would've made 8% more than they did.
So, the question is: Why, when there are no fares, are they driving for, like, the whole 12 hours? And, why when there are lots of fares, are they getting out of their cab in, like, an hour and a half?
And, it's because they had a goal: they set a finish line. So, they had an earnings goal for the day, say $300. And, when they hit the earnings goal, they quit. Because they're done. They crossed--in a marathon, nobody keeps running past the--like, 'Oh, I feel pretty good today: I'm just going to keep going.'
Russ Roberts: That's a great point.
Annie Duke: Right? It's why people who run half-marathons don't just randomly finish a marathon because they feel in fine fettle. Right? Like, you finish once you're done. So, anyway, so they would finish once they hit their earnings goal, but if they hadn't hit their earnings goal, they would keep going forever. Okay. So, we know that that's actually quite bad behavior.
Well, somebody followed up with a study--and I'm just blanking on the name, it's in the book, so please go look in the book because I'm blanking.
Russ Roberts: We'll put a link up to it in the show notes.
Annie Duke: Yeah. Followed up. And what they did find was that people did get better with experience. So, they still weren't perfect, but the cab drivers who are real veterans were much better than the ones who were not, in terms of this behavior.
Russ Roberts: Yeah. I'm always skeptical of that kind of study because--you don't know: it could be they really need to make $300 every day because they won't make their rent. And, you could say, rationally, 'Well, make the $300 tomorrow at the time when the fares are easier to get.' But, maybe it's not so predictable when the fares are good and when they're not so good. And so, maybe as you get experience, you get better at maybe predicting that and you can smooth it a little bit. But, I have a feeling--
Annie Duke: Yeah, except the difference is that it's that the experienced drivers, they don't stop at their earnings goal. They just drive when the driving is good.
Russ Roberts: But, the goal of life is to make as much money as possible, maybe they want more leisure--
Annie Duke: Well, except the goal of life, the goal in life, it's also not to spend--
Russ Roberts: 12 hours driving around?
Annie Duke: Right, exactly. Like, that's not fun.
So, essentially, if you're going to spend some of your time driving around in a cab for 12 hours, let's make sure that you're making, like, $1200 bucks doing it, and then quit the day that it's not--Number One.
And, then Number Two, the other thing is that there are predictable times when there's going to be lots of fares, like when there's concerts around. And, if we think about an inefficient market, this is why when there's a big concert or it's rush hour, it's impossible to find a cab--because all of these cabs made their earnings and they got off the street. Right? So, it actually creates kind of an inefficient market as well.
So, yeah, I don't think it's just, like, 'I got to make $300 a day,' because obviously they have to make $300 on average, because what the veteran drivers are doing is they're driving past the $300 mark.
Because they understand, like, 'No, as long as this concert's around, I'm just going to,' or 'it's a busy time at the airport, I'm going to drop the thing off, come right back to the airport, get the fare, drop the thing off, come right back to the airport. And, I'm going to do that as long as it's busy.'
Russ Roberts: Yeah. If I'd been the referee on that paper, ask him to show that the--I'd check into how predictable, how easy it is to figure out when there are a lot fares. Certainly, yeah, when there's a concert. Yeah. But, day-to-day, I don't know--
Annie Duke: I mean, you have morning rush hour, afternoon rush hour, Saturday nights--like, it's New York City, right?
Russ Roberts: But, there could be more variability than we expect.
Russ Roberts: But, the interesting point also for me is that--and this is true of Uber also, and we had John List talking about that in a recent episode--it's newcomers, amateurs, whether they're poker players, cab drivers, you name it. They make lots of mistakes. And, the best ones learn and get better at it and narrow that gap. And, the others just go, 'I'm not good at this. Bye.' And, they drop out the pool after a while. So, there's a steady inflow, often--
Annie Duke: Not in poker.
Russ Roberts: Well--
Annie Duke: I mean it depends on how much money you have.
Russ Roberts: As Nassim Taleb would point out--
Annie Duke: And, this is where I would, this where the problem is in poker: is that--and I think this is generally the problem with quitting--that the more uncertain the system that you're deciding in, the more that it becomes a petri dish for cognitive bias.
Okay? So, the question is why would people continue trying to butt up against the dead certainty that you have to turn around? Well, because the objectively correct moment to quit is a decision made under uncertainty.
There's this particular irony to this, which is that when we start things, most things that we start are made under uncertainty. There'll be an influence of luck on the outcome, like a pandemic hitting, or a recession, or something like that. But then, there's also just we know very little in comparison to all there is to be known. So, we can think about somebody who's investing in a startup for example, that maybe is early product market fit or pre-product market fit. This is very high uncertainty. You have three people on the team. You don't really know how that person's going to be as a CEO [Chief Executive Officer]. Is the product really going to work out? Are customers going to buy it? All these questions.
Think about taking a job. What do you know about the company that you're going and taking a job with? Like, nothing.
So, we know that we have to make those starting decisions under conditions of uncertainty. Lucky for us, we have this really valuable option, which is the option to quit, to cut our losses.
The problem is, and the irony is, that that that decision is also made under uncertainty. That the objectively correct moment to quit, it's nothing particularly bad that's going to happen. It's 11:30 AM, you're on the mountain, you have plenty of oxygen, there's no blizzard, nothing is really bad that's happening right then. But, it is the objectively correct moment to quit. But what that means is that if you quit that moment, there's still some chance. There's still some chance--probably too low, but certainly too low--that you could still make it to the summit and make it back down alive. Some people did. I mean under great duress, right?
So, all those biases that have to do with sunk cost fallacy, with not wanting to quit accounts in their life, with over-optimism--optimism bias, for example--dig into those environments to make us keep going, even when we're not particularly good at it.
And, this is particularly true in poker for this reason. If I look back on why I won or lost, I have two things I could think about: skill and luck. Right? Now it's usually going to be some sort of combination of the two, but we're binary in the way that we think. So, as I'm looking back on that, something called 'self-serving bias' is going to dig in and self-serving bias is just this: I won because I played awesome and I lost because I got unlucky. And, you hear people say this all the time, and they will continue well past the time that it is obvious. I mean, I'm talking years and years that it is obvious that they have no business sitting at a table.
Russ Roberts: But, the thing I love about the Everest example, which we haven't talked about and I didn't think about until this conversation when I was reading the book: That one o'clock turnaround time, that's a veteran insight.
Russ Roberts: If you were on that mountain and you looked ahead and you thought, '300 meters. I'll be there by, I'll make this easily; and okay, if I get there by two o'clock, how long could it take me to get downhill?' The fact that that was understood and accepted as the turnaround time and they still didn't do it is so fascinating because in real life you rarely know when to cut your losses. You might in poker because poker's a very constrained game--
Annie Duke: Oh, no, no, not--
Russ Roberts: You could.
Annie Duke: Not in the middle of a hand.
Russ Roberts: No, but you could. You could if you're a good player, you could figure out because the odds are very constrained.
Annie Duke: Yes, you can get close.
Russ Roberts: But, in real life the finish line is usually not clear. The turnaround point is not clear. And so, you're always going to--often because of that, for lots of reasons you've been talking about--you're going to push beyond one o'clock.
But, in this case, the tragic case of Mount Everest, that one o'clock wasn't just like, 'Eh, I think one o'clock would be a good time to turn around' or 'I think it's going to take me a while.'
Annie Duke: No, it was--yeah. So, first of all, let me just say in poker, the problem is that it actually is very hard to figure out for sure because you can't see the down cards.
Russ Roberts: Good point.
Annie Duke: So, everybody's cards are face down. So, trying to figure out when the odds go against you is actually quite difficult.
Russ Roberts: Fair enough.
Annie Duke: And, not only that, even when you've decided that maybe your hand isn't going to win, you have the option to bluff.
Russ Roberts: True.
Annie Duke: So, this is part of the reason why quitting is so darn hard. Whereas, in chess it's a much easier decision, for the reasons that you say. It's pretty obvious that you're about to get checkmated, right?
Russ Roberts: When you're down a castle, you're down a rook--it's over. For the good player.
Annie Duke: Exactly. So you know, you really know what your position is there.
So, this is part of the reason why people can make a lot of money at poker because the better I am at figuring out the appropriate moment to turn around--what the turnaround time actually is--the better off I'm going to be in comparison to other people.
And, it's actually an incredibly hard problem. It's actually very much like in life. I don't know--I don't know all the information. If I knew it, I could calculate the odds. If I had the information in front of me, I could surely do some quant work on it and figure out what my odds are. It'd be pretty simple quant work, as a matter of fact. I could create a game theory optimal table, and I would know exactly how often I should bluff and exactly how often I should fold, and whatnot.
But, I got to figure out what the other person's holding. So, this becomes actually quite a big problem.
But, yes, this is the thing: it's that we can think about these quitting problems twofold. One is that it requires you to be able to forecast into the future. So, that's that idea of I've got a startup and I've missed--I've got money in the bank--but I've missed some targets. I'm looking at what month-over-month new user growth is. You know, whatever. And, these signs are pointing to things are kind of bad. But, I've got money in the bank. So, what I have to be able to do is foresee that the signs that I'm seeing right now that have to do with the exploding costs of acquiring a customer, whatnot, that those signals are adverse in a way that are going to tell me that I'm not going to get to a venture-scale business here and I ought to quit.
So, that's a forecasting problem. I got to get myself into the future.
Now, what you're describing is that there's also a different type of time traveling that occurs, which is either the whole of your past experience helping you to figure these things out, or people who have done it before you. So, on the case of Everest, it's people who have done it before you who say: 1:00 PM, that is the time; things that happen after that are really bad, and don't keep going after that. And, we need to pay attention to those people because when we think about life, it's very rare that we get to run a Monte Carlo. But, if you've had lots and lots and lots of people go up that mountain, they've run a Monte Carlo for you.
Russ Roberts: A 'Monte Carlo' being a simulation of many, many trials. So, you can get an idea of the risk.
Annie Duke: Right. Exactly. Sorry. Yes, exactly. The geek in me comes out occasionally.
So, now you've got your Monte Carlo simulation, right? Because you have all these people who've done it before you. And, man, if you're not paying attention to them, what a waste, because that's a real gift that the world is going to give you, because it's so rare that we actually can run those kinds of simulations in order to get some guidance on what the appropriate sticking or quitting situation is.
Russ Roberts: Yeah. In real life, the mountain you're climbing is not the same as the one that the other people climbed and therefore you can convince yourself--etc., etc.
Russ Roberts: I want to close--I want to close with another wonderful image in your book, which I think is very powerful and very useful and often goes against our psychological grain, which is: The monkeys and the pedestals. Talk about that image and how to use it for one's own purposes.
Annie Duke: Yeah. Okay. So, we've talked about a lot of the impediments to quitting. Right? Finish lines--they make it hard to quit. Those very clear goals: good side to those and bad side to those. Issues that have to do with identity, who you are.
But, there's all these issues that have to do with resources that we put into things. So, let's just call that, broadly, capital. And, capital isn't just money. It can be your time and attention and effort.
So, we know that once we start to invest capital in something, that that capital itself is going to make it hard for us to walk away. Partly because we as humans tend to think about waste as a backward-looking problem when it's really a forward-looking problem. Right? 'I don't want to quit now because I'll have wasted my time.'
Like, for example, if you worked really hard to get a Ph.D., it's very hard to walk away from academics if that was your plan, because then, 'I'll have wasted--why was I in the Ph.D. program? I'll have wasted my time.' So on, so forth.
But, what we really care about is if your goal is to be happy and fulfilled: Is the path forward the way to do that? But, we don't think that way.
So, monkeys and pedestals is really trying to get you to reduce the debris that you're bringing into any decision to quit. In other words, to minimize the capital that you spend before you figure out whether something is worthwhile in continuing or not.
So, that's the point of this mental model. It comes from Astro Teller [Eric "Astro" Teller], who is the CEO--otherwise known as Captain of Moonshots, over at X, which is Google's in-house innovation hub.
So, they're trying to take world-changing ideas from idea to commercialization in 5-10 years. So, obviously this is very uncertain. Right? Like, I mean, they're really working in places of high uncertainty. We're not talking about incremental change where you kind of know exactly what the outcome is going to be. They're delving into the unknowns.
And so, they use a mental model and try to help them to figure out how to approach these projects so they when they find out--so that they can find out what they need to know as quickly as possible in order to be able to, as they think about all these options that they consider, that they can quit all the options that aren't worth pursuing, that aren't going to get them to where they want to go and concentrate the capital on the options that are worth pursuing. So that is their goal, just like the cab driver who is--right? Okay.
So, monkeys and pedestals goes like this. Let's imagine that you've decided you're going to make a bunch of money by training a monkey to juggle flaming torches while standing on a pedestal. What part of the project should you approach first? Building the pedestal? Or, seeing if you can train the monkey to juggle those flaming torches?
And, Astro Teller's insight is: You better train the monkey to juggle the flaming torches, because otherwise what's the point? Right?
So, he's saying the hard part of the problem, the thing you don't know if you can do, is whether you can teach that monkey to juggle flaming torches. That's the unknown. And so, there's no point in doing any other part of the project if you haven't solved for that unknown.
That's Insight Number One.
Insight Number Two is that if you do build the pedestal, you will feel like you've made progress, but you will have made no progress at all, because you already know you can do it. So, therefore you've learned nothing. It doesn't actually advance you toward your goal in really any way because you're doing something--I mean, you can turn a milk crate upside down--you're doing something you already know you can do. So, that's Insight Number Two.
And, Insight Number Three is that in creating that false progress by building the pedestal, you have now created sunk costs. You have now created ownership over the pedestal, endowment to the pedestal. Your identity is now more deeply entwined in the thing that you're trying to do.
And so, now, when you find out that the monkey is super-hard to train, you're much less likely to quit--because you have built the pedestal.
All right. So, I'll give you an example of monkeys and pedestals.
I assume you've heard of the Hyperloop. So, trying to sort of vacuum tube passengers from New York to LA in two and a half hours, like, in these vacuum tubes. So--
Russ Roberts: It's an Elon Musk project, yeah.
Annie Duke: So--well, it's also a Virgin--Virgin is doing it as well.
Russ Roberts: Great, competition.
Annie Duke: So, yeah. So, Astro Teller--so Google X was approached about this project. They were pitched it. So, after they pitched it, the team at X did a monkeys and pedestals exercise, which is what they always do when they're thinking about a project. And, they said, 'Okay, what are the monkeys here?' Well, the monkey that obviously might come to mind is: Can you actually build something like that that will work, that will actually cause something to go through a vacuum tube? And, that technology was already proven. So, that wasn't a question. You could definitely move things through vacuum tubes. So, that was not actually a monkey.
But they did identify two other monkeys. One was a regulatory issue. When you're building that, you got to go through how many different townships--I don't even know. Each of them is going to have different regulations. I mean, my gosh, it's a beast. The regulatory issue is a beast. And, as Astro Teller said, 'We're Peter Pans with Ph.D.s: I don't know if we can solve that one.' So, they felt that that was a pretty big one.
Russ Roberts: Say that again?
Annie Duke: They're Peter Pans with Ph.D.s. What are they going to do with the regulatory issues?
Russ Roberts: You mean, they're not going to be able to make any progress on it.
Annie Duke: Yeah, that's what they felt like. They felt like it was a pretty hard one.
And, then the other thing--actually, the other monkey, which is the more significant one here was, 'Well gosh, if this thing is moving so fast that you can get from LA [Los Angeles] to San Francisco in two and a half hours, can we actually stop it and not kill everybody on board?' So, that's a pretty big monkey. I don't know if you can teach that monkey to juggle. So, they thought about that and they said, 'Well what would we have to do in order to be able to know that we could stop it safely?' And they realized--they kind of did some math--and they realized we're going to have to build almost the whole system to figure that out.
And so, we're going to build this humongous pedestal before we ever know whether we can solve the monkey. This isn't for us. So, they rejected it. Now they rejected it in 15 minutes. Now let's--really, that's how long it took them. That's why this mental model is so powerful.
So, let's now flash forward many years to Virgin, that is doing it. You can look it up. There's a recent New York Times article and Virgin's Hyperloop project has run into two humongous problems. I think they've raised, like, I think over a hundred million dollars. So, they spent millions and millions and millions and millions and millions of dollars. So much money, so much time. And, they've run into two problems. Can you guess what they are?
Russ Roberts: Hmmm. Could it be the regulatory problem and the stopping problem?
Annie Duke: Yes. So, the regulatory problem they don't think they can solve. It seems to be really thorny. The stopping problem, they've managed to build a sixth--they've built enough of it to get it up to a sixth of the speed. That's it. So, they haven't really ever done a real safety test. And, that's as far as they've gotten.
And so, now, remember: remember, Astro Teller also has this insight that once you've built all these pedestals--which they've done, right, because they're building the system, which is a big pedestal because you already know you can do it--that, when you butt up against those intractable monkeys, you won't stop.
Russ Roberts: That's a great--
Annie Duke: And, that's exactly what's happening, because they're not stopping. Instead they're pivoting to say, 'Well, we're not actually going to bring people on this thing. We're just going to transport cargo.' It's, like: Cargo? What? We already know how to get cargo from one end of the country to the other.
Russ Roberts: It would be better to have it be a little bit faster, but it's a really expensive way to get a little bit faster.
Annie Duke: So, they were trying to create world-changing innovative change and now they're creating an incremental improvement over the current way that we transport cargo--because they won't give it up, which would be probably the correct answer at this point.
Russ Roberts: Yeah. I just have to say that--I want to reference the Mary Hirschfeld episode we did where she defended the rationality of sunk costs. And, as an economist who for years said it was the quintessential example of irrationality, she did make me think about that there are some virtues to ignoring sunk costs--even though in most situations irrational, they are a commitment device. And so, buying the gym membership is a way to--and if you ignore the sunk cost, it's a way to actually maybe get you to the gym. So, we'll put a link up to that episode. You can see whether you're convinced by it.
Annie Duke: That's true, although we know that people don't use them.
Russ Roberts: Yeah. Well, it's not the best example.
Annie Duke: Here's a good example. Here's a good example of an incredibly perverse use of that. So, the California Bullet Train, which is supposed to connect LA [Los Angeles] to San Francisco to the north: in 2010, they floated a bond for $9 billion on a $33 billion projected budget for high speed rails. It's going to be a bullet train, like you see in Japan.
The first section of track that they approved was between Madera and Fresno. So, this is on flat land in the Central Valley. Let's call that a pedestal, because we know we can build track on flat land. So, that was what they approved.
They broke ground around 2015. In 2018, someone says, 'Ooh, we have a problem. We just happened to notice'--literally, I mean this is basically how it went--'we just happened to notice that there's two mountain ranges, the Tehachapi Mountains that are to the north of LA and the Diablo Range, even bigger, to the south of San Francisco. And, just want to let you know, don't know if we can blast tunnels through mountains that are in seismically active areas, nor operate a train safely through those things.'
Okay, well, I mean: Come on. Those are monkeys. And, there's also a monkey, which is NIMBY [Not In My Back Yard] issues, which is kind of similar to the regulatory issue.
Russ Roberts: 'Not in my backyard.' Yeah.
Annie Duke: Yeah, yeah. So, anyway, once they figured out, 'Oh, there's this problem with the monkeys,' the projected budget now exploded to $80 billion. It went to Governor Newsom to decide what to do. And, he said, 'Oh, okay. Well, what we'll do is we'll build track--the next section of track will be between Bakersfield and Merced,' also in the Central Valley on flat land--
Russ Roberts: Big pedestal, yeah--
Annie Duke: Yes. 'And, then we'll build track from San Francisco to Silicon Valley,'--also on flat land. So, he literally approved two more pedestals.
Now, there's been some articles now talking about what a disaster this is. The budget is now well over a hundred billion. And, they still have not done an engineering feasibility study on the mountains yet. It was supposed to be completed in 2021. That's now re-forecasted to 2033; but who knows?
Anyway, I saw someone on Twitter defend it saying, 'Well, it's better to do it in Central Valley for two reasons. One is it's easier.' Okay, but who cares? Like, what are you going to do? You're going to connect Bakersfield to Fresno? Because that seems like about what you're going to do. Yeah, okay. It's easier. So, that's a big pedestal. But, then they said a really--what I think is malign thing--which is: 'If we spend the taxpayer money, it's more likely the NIMBYs will give up.'
And, what I say to that--it's a--right?--is, 'Oh. Sunk cost is a cudgel. So, now you spent $9 billion of taxpayer money on a budget that's gone from $33 billion to, I think the last projection was, like, $115 or so. And so, now you're telling me that you're trying to beat the NIMBYs over the head with the sunk cost when you haven't even addressed the mountains. So, you're going to continue building something that you don't know if you can build in a seismically active area and you're going to spend a hundred billion more of taxpayer dollars?'
Russ Roberts: Insane.
Annie Duke: I mean it's nuts.
And so, I think that's one of the worst things about, in a lot of ways, what politics does with sunk costs is they end up using it as a cudgel. And what a waste of taxpayer money. See, that's the thing: is that in service of 'I don't want to waste $9 billion in taxpayer money,' they're now looking to waste way more. Which, if you wanted to address climate issues in California or better serve the Central Valley, it seems to me if you had a hundred billion dollars to create economic prosperity in the Central Valley or address climate issues in California, it could be done a lot easier than building track on flat land that goes from nowhere to nowhere.
Russ Roberts: I will recommend the episode, EconTalk episode--I hope I pronounce his name correctly, I've now forgotten--but, it's Bent Flyvbjerg on Megaprojects. If you Google econtalk megaprojects, you'll find it; and we will put a link up to it.
When you said that Virgin has spent a hundred million, that's a lot of money; I'm thinking, 'Not really.' Not compared to what they're going to have to spend to solve some of these other problems.
Annie Duke: Well, that's the thing.
Russ Roberts: It's still a lot of money.
Russ Roberts: You almost got a Ph.D. and you almost pursued an academic career. How did that inform your thinking about this book and your own experience there?
Annie Duke: Well, I just would like to say that this is a very good lesson, which is: We think about quitting as closing a door and sealing it shut. But, for many things that we quit, if we stop thinking about it as a decision that we can't reverse, it'll make it a lot easier for us.
Because, I am now an academic. I do research with Phil Tetlock. I do research with Maurice Schweitzer. I do research with Jay Van Bavel up at NYU [New York University]. Maurice is at the Wharton School. Phil is in psychology and the Wharton School at University of Pennsylvania.
I currently teach at the University of Pennsylvania. And, I am just now enrolled as a graduate student with Phil Tetlock as my advisor, because those studies that I ended up doing that I said yes to were large-scale enough and enough work that he said, 'Just write them up. You should just re-enroll, because then you can finish your Ph.D.' Because I had done enough work for a dissertation at that point.
So, there you go. So, I circled back. Not everybody circles back, but you can. For many things that you choose to quit, you can go back. And it's something that you should think about.
I will say that my quitting then really has informed the rest of my life, which is: I became a much bigger quitter. Because I realized--I just realized--that that fear of what's on the other side should really go away, right? Because there's stuff on the other side.
And, I want to be clear: I know that there are people who have circumstances where they can't just go quit their job. I totally get that. And, I'm not suggesting that they do if they need, they have to make rent. So, people have more limited ability to--some people have more of an ability to quit than other people do--that simply have to do with their circumstances.
But, what I did learn from that is that: no matter what your circumstances are, if you can create one more option for yourself, you can be a little bit more ant-like. Your life will be better because you'll be more rational. It will help you to be more rational about whether you want to stay with what you're doing or quit. So, if you can just create one more option for yourself, you're better off.
And, then that comes up with, like, what the really sad part is, is that there's people who have lots and lots of options and yet they stay stuck in things. Right? And, think about all the people who don't have those options, who are stuck by circumstances, who would love to have that optionality to be able to go switch. Right? So, I think that we really do need to be thinking about it this way.
And, for me, because I quit academics, what happened--and I was forced to do it, I was sick, so I had to take time off--I discovered poker. And what I realized after that point is that there's usually something on the other side and that something might be really cool.
Russ Roberts: My guest today has been Annie Duke. Annie, thanks for being part of EconTalk.
Annie Duke: Well, thank you. This was a lovely conversation.