Joe: So let's just dig right in. Michael, thank you for taking the time to talk with me and our audience today. You not only focus on decision-making professionally in the field of finance and teach about it at Columbia University, but you've literally written books about it. I was trying to think back to when we first met and I don't remember. I remember reading The Success Equation before we met just out of interest. I think it came up in a conversation with Annie Duke one time who then said, “Oh, he's a friend of mine.”
Michael: You know, it's interesting that the actual story on The Success Equation and this book about skill and luck is: I have a section in there where I talk about how to think about skill and I actually took something from one of Annie's testimonies to Congress, so that was one of the ways to test skills. So I had this line that you know there's skill in the activity, if you can* lose on purpose. (* Editor's note - transcription here corrects audio version, in which guest misspoke)
Michael: So everyone gives me attribution for this little turn, and I documented it both in the book and I also anytime people say it, I say, “No, I got that from Annie Duke.” This idea that when you're in a probabilistic field, it's very difficult to not think about obviously luck and skill in terms of the outcome. I've learned so much from Annie and I also think that the mission of the Alliance is just, as you point out in your intro, so incredibly important for all of us as citizens.
Joe: I taught at an all boys school. The boys had certain days where they'd walk into the classroom and there would just be a projector up front with the fireplace going and light classical music going and they knew that was a reading day. And they'd walk in and there were bookshelves that were just math and science books, just interesting books, and The Success Equation was one of the ones that was very popular to the point where I had to buy some more copies. The guys would just go and pick a book that they found interesting and read it for the majority of the class period and then the last part of the class was just to talk with the other guys in the class about: what did they learn? What did they find interesting? What questions did it prompt? Your book spread around like wildfire so much so I didn't even know about the previous work, Think Twice.
I don't know how many of our audience members are familiar with either of these books yet. Could you talk a little about Think Twice and what you were trying to accomplish with that?
Michael: Yeah and just to take one step back, Joe, you know, you mentioned my involvement at Columbia Business School, which I've been teaching there since the early 1990s. After about five or six years into it and in finance and investing there obviously sort of some sets of technical skills that you need, but the realization for me was that a lot of what led to great investors versus good investors had nothing to do with those technical skills.
Those were prerequisites to get into the game, but not the keys. It was really how they made decisions and especially the decisions under stress. So that led me to a lot of this psychological literature and thinking about under what conditions do we do well and what don't we do well.
Through that, I had been writing a little bit about this and my editor at Harvard Business Review Press called me and said we're doing a series of books that are meant for executives that are relatively short about various topics. We'd love to have you write about decision-making. The idea of Think Twice was really almost a complete homage to the work of Daniel Kahneman and what we now call System 1 versus System 2 thinking. System 1 is your automatic system. It's fast. It's very difficult to train, but in many cases it's actually quite useful. System 2 is your analytical system. It's slower. It's more purposeful. It's more deliberate. It's more costly. What Think Twice attempts to do is run through a number of different conditions where your mind is going to take you down one path automatically, so your System 1 is going to be activated where you should stop and engage your System 2 to better think about that problem. Hence the title Think Twice means that under certain conditions, you need to think twice. This book came out before Thinking Fast and Slow, and Kahneman had written a ton about this, but this was really this idea of codifying situations where you should appeal to System 2 versus System 1 thinking. It turns out that this is like a little inside baseball. The last chapter of that book is actually about luck and skill. Not surprisingly. I originally had it as the second chapter because I thought it was so cool and so interesting and my editor was like, “Yeah, I don’t know, this luck skill thing seemed a little bit, you know, statistics and so she's said, why don't you put that in the back of the book? You should keep it, but put it in the back. I was getting these emails from friends and again, you have to discount this, and they’re saying, “Oh, your book, really enjoyed it, but that chapter about luck and skill, that was really cool. So I thought, this is interesting, the things people are picking up on.
That encouraged me to sort of do a deeper dive on that one specific topic. In a sense, The Success Equation is a spinoff from the last chapter of Think Twice and tries to develop that idea of skill and luck in more depth. I was inspired by many other writers, including Nassim Taleb, who wrote a book called Fooled by Randomness in 2001. I think Taleb's point is, which is correct, that we tend to underestimate the role of randomness or luck in the outcomes that we observe. But he, I think, came up short in actually giving people frameworks to think about the actual contribution of luck. These are pulling threads together.
I've been very influenced by Moneyball. That encouraged me to start to read a little bit about sports analytics. And of course, all [sports analysts] do is try to understand really what effects can we attribute to skill and what effects can we attribute to luck in order to predict future performance.
And so when you sort of brought those communities together, for me that that's a little bit of where The Success Equation came from. A lot of these ideas, even if you're even if you can't be pinpoint precise in terms of the disaggregation of luck and skill, just the very effort itself can be extraordinarily illuminating to get you to think about the various topics.
Joe: I'm glad you bring up Moneyball because we had a great time reading that with juniors and seniors in high school to the point where they all wanted to read [Moneyball] and Freakonomics around the same time. And then we turned around and used it to revise our admissions process to the private school, having the boys start to design a Moneyball approach to: Who else should be part of this community? Who are we missing? What do you want to add? How are you going to discern that's what you're getting and are these interviews tricking us? Are these essays that are being written maybe by them, maybe by their parents, are they fooling the admissions committee? It was fascinating. And then of course that led them into thinking about their own college admissions work or hiring practices. The other one that really resonated in a similar way was Against the Gods.
Michael: Oh yeah, Peter Bernstein's book Against the Gods. Yeah, it's a beautiful book.
Joe: Oh, it's fascinating. I appreciate Chapter 8 of Think Twice where you started on the skill luck thing, but it was actually the earlier parts where I was noticing a trend that you were focusing on: what are the mistakes? What are the things to avoid in decision-making? I wondered if you could spend a little time talking about this. When you think about someone who's a skillful decision-maker, when you think about people who are managing and improving their decision-making quality, what are the mistakes that they're avoiding? What are the things you're seeing them not do or manage the risk to do?
Michael: You know, Joe, it's a great question. This, that by the way is not only individuals, but also when you think about teams or organizations, what kinds of mistakes do they commonly make?
If I had to put this under one big banner, the big banner I would probably select is they fail to consider alternatives. Let's give a really concrete example. There's this idea that Daniel Kahneman popularized called the inside versus the outside view.
So the inside view says, when I pose a problem to you, the natural way we all solve it, and by the way, we all do this, this is our natural way of doing: we gather information. We combine it with our own input and experience. So we're melding this information we've gathered. And by the way, we may not be gathering all the information we need, but we gather information, we combine it with our own experience, and then we project into the future. The outside view says, I'm going to think about my problem as an instance of a larger reference class. I'm going to ask a broader question, which is what happened when other people were in this situation before?
That's a very unnatural way to think for two reasons. One is you have to leave aside what you've done, your effort, and your own experience. You almost have to discount your own experience and your own input. Second is you have to find and appeal to the base rate, which is the outside view, which may not be at your fingertips.
But what I think Kahneman and many other psychologists have demonstrated beyond the shadow of a doubt is that the combination of this inside and outside view is the best way to think about things. So what happens is we don't consider lots of other things that have happened before lots of other alternatives.
So if you're in a team meeting or even when you're thinking about something in yourself, we tend to sort of narrow in on an answer too quickly without considering and surfacing alternatives. You mention this idea of identifying alternatives and sort of homing in on what seems to be the most sensible one. I'm infinitely fascinated by things like social insects. One of the great stories is a beautiful work by Thomas Seeley at Cornell University on honeybees and how honeybees find new homes. They obviously can test this. They actually take them to an Island in Maine and they can actually mark every bee and see exactly what they're doing.
Their task is to find a new home. So they're out in a swarm. They're very vulnerable. Obviously they need to find a home quickly. Bees have certain ideal homes, south-facing, at a certain height and a certain size. What Sealy does is sets up various homes out there, right in the environment.
And then the bees, the scout bees go out and try to find it and they're remarkably good at it. But the answer is they test everything and they come back and then the bees do these waggle dances to tell their sisters first of all, where this place is and how good it is.
They're saying not only is this place good, you go check it out.That will activate another scout bee to go. And then they do this thing called quorum sensing. So it's an incredible thing that nature has figured out. This importance of identifying as many alternatives as possible and then winnowing down to the one that makes the most sense given what we're trying to do.
So to me, that would be the big mistake is that we tend to not entertain alternatives. We tend when we think about scenarios, we tend not to think about enough scenarios. You know, Don Moore just wrote this wonderful new book called Perfectly Confident and Don talks about [that] we often will just not consider various scenarios. We project out ranges of outcomes that are simply too narrow. One of the manifestations of overconfidence.
Joe: This is part of what I love about talking to you is that there are five different directions the conversation could go.
So I'm going to loop back in here and do my little waggle dance to get your attention back to some things. I don't think that all of our listeners will know what a base rate is. So could you say a little more about that?
Michael: Yeah, thanks for slowing that down a little bit. So, as I mentioned, rather than thinking about your problem as unique to you, what you want to ask is, what happened when other people were in this situation before? So, you're gonna now draw from a distribution of experiences versus your own experience. So let me give you an example. I'll do one from my world that's actually quite simple to articulate.
Let's say you're analyzing a company, and I'll just make this up, say it has $10 billion in revenues. And you're going to say, how fast will this company grow its revenues over the next three, five years or something like that? Well, you can either, obviously you sort of analyze the company and think about the industry and its market share and all those kinds of things, and then make a projection.
And alternatively, and I guess what you'd want to also do to come up with a robust forecast, is you might look at the history of all companies of that size. And simply ask what is the distribution of growth rates for companies of that size? And you'll see, it's not a perfectly normal distribution, but you'll see something like a normal distribution.
And that will allow you to calibrate where your forecast resides within a historical distribution. Now, the one thing I should say about base rates is the application varies across various domains. So some base rate distributions are fairly well behaved, things like sales growth rates for companies and the heights of people. There is another part where it's vastly more complicated where you have these distributions that are much more skewed, much more power law-like, or where there are lots of small observations or very few large observations.
So things that are socially driven, book sales, music, sales, and movies, are much more difficult to do. But in general, this is a wildly underutilized framework. It's interesting, I guest lectured for a course last night at a college and the students said to me, if you could go back in time and give yourself one framework that would make you more informed going forward, what would it be?
And I said, it would be this idea of base rates. I just think it's an incredible concept that carries an enormous amount of intellectual freight once you understand it. It's not natural because you've done the work and you have your own experience and hence we think we check the boxes. And second is that base rates are not always at your fingertips.
If I move from here to Philadelphia, I've never done that before, so I don't know what's going on, but lots of other people have done that before. So there are base rates of results. I just am not aware of what's going on. So a base rate is basically a distribution of past experiences given the same set of initial circumstances. Just understanding how things have unfolded for other people can be very helpful to understand how things might unfold for you or your object of interest.
Joe: How do you go about figuring out what's the right category or proxy that you should use for drawing your base? Like you mentioned, the company size, and I'm thinking a lot of our listeners are working in business, a lot of them are in education, and a lot of them are just thinking about this for their personal lives.
So you come across a decision in your personal life, should I buy or rent? Is the thing that's pressing me as the question a safe choice for the category or do I need to do some work to figure out what's the appropriate base rate I should be looking at?
Michael: Yeah, I mean that case, and Joe there's a little bit of a devil lurking in the details here, because there's this idea of reference class. What's the appropriate reference class? Here’s the tension you're trying to balance. One is to have a large enough reference class that there's some robustness to it. So it’s statistically significant, but it's also narrow enough that it actually aligns with what you're specifically looking at.
So that's the challenge, Daniel Kahneman wrote an article in, I think it was with Dan Lovallo, in the Harvard Business Review and they talk about this, and it's actually in Think Twice. There's a four step process for going through how to think about applying base rates or reference classes. One is, trying to get the right balance between specificity and generality. Then the other one is just looking at the nature of that distribution itself. So as I said, there's a continuum from things that are very clean and simple to things that are very skewed and power-law-like, and that's going to be a lot more difficult to do.
Whereas the application of base rates or reference classes is not perfect, I think it's a vastly underutilized tool. There's a ton of application that is not being used today. So without getting into the advanced trickier states, let's start with that.
The other thing I mentioned is, and you might know this, you might be closer to this even than I am, but you know, I've spoken with Phil Tetlock and Barbara Mellers a fair bit over the years about some of what they do with Good Judgment and this idea of these people who are making really good forecasts, so-called superforecasters and they were trying to measure the impact of training.
And so they would give some people training and other people no training for control. And what they found was the training did improve people's accuracy of their forecasts and it turned out, they found the thing that was most important in that training was basically this: this idea of teaching them about base rates and reference classes and just introducing people to this way of thinking.
And some of the basic tools improved their forecasting acuity. So again, without getting into the advanced, trickier aspects of this, there's just a ton that all of us can use. Business, sports, and investing, are all very fertile areas for this, but like you said, even other domains, there's going to be a lot there that we can work with that's useful. So, I just think it's a very big idea.
Joe: I absolutely agree there's a lot to mine there. I'm conscious of your time. I said that you had kicked off four or five things in my head. So another one that you mentioned that was hoping you could say more about is this idea of exploring versus exploiting.
So these bees are going out and they're looking at all these alternatives. You could just keep going out, looking at more and more alternatives. How do you think about when to shift from an explorer to an exploit strategy? It could be in your work or in your advising clients or however you want to go at that.
Michael: So, Joe you're reading my mind here.Because I have a list of topics that I want to write more about and the explore/exploit thing is one of them. There's literature on this in computer science, the multi-armed bandit problem, but let's not go into that.
Okay. So now I'm going to talk about social insects again. Tell you another great story. This is about ant foraging. We'll use ants as an example. You have a nest of ants and there's a food source.
Exploitation would be when the ants go out, find the food source, and then they just basically focus their attention on getting as much of that food into the nest as possible. The way the ants do this, they basically go out randomly. But they're laying pheromone trails as their means of communication.
Once an ant finds the food, they come back. That trail gets a little stronger. The ant comes out, senses that and so that reinforces it. As they travel the same path that becomes a very strong pheromone trail, and then that's how they figure out how to do that. So that's what ants do. That's really cool.
If you actually watch ants in your backyard and the summertime, they're kind of doing crazy stuff all the time. The scientists were studying them and they would find out that some ants would just peel off of the pheromone trails that lead to the food source. Then they got much more scientific about studying that rate of peeling off and it turns out there's a mathematical probability.
Here's the thing. That's absolutely beautiful, Joe. This is the thing that's so cool. It turns out that the probability of peeling off the path is a fact function of the rate of change in the environment.
So when the environment does not change rapidly, the ants go for mostly exploitation and very little exploration.
When the rate of change is rapid, they know that that food source may be exhausted quickly and there may be other food sources that are popping up around them. They're going to allocate more resources towards exploration. Measuring your rate of change in your environment is not easy.
In things like business, we try to think a lot about this: is this industry stable? Is this industry changing? Is there disruption? But that to me the guiding force is to think about, how do we think about the rate of change in the environment?
And so zero changing environment.
And by the way, if you go to ecosystems and in our world today, there are some environments that are extremely stable and you'll see almost all those species are all about exploitation. They do very little exploration. You'll find others, other ecosystems that are rapidly changing. You'll see the species there that adapt are very big explorers.
They do very little exploitation per se. Then there's everything in between. So that's a part of what I'm challenged with working on is, can we create a framework beyond that qualitative statement?, which I think makes sense. Can we create a framework to think about what is sort of the best mix between those two things?
And I'll just say the other thing is that there's some other really interesting dimensions. One is there's a huge issue with age. When you're young, exploration is much more interesting than exploitation. So you want to try out the new restaurant and meet new people.
As you get older, where your payoffs are much more finite, and there are some days I feel that very much, then you say, I'm just going to exploit. I'm just going to just go to the restaurants I like and hang out with the people I like and I'm not going to bother to bother with this exploration because the payoff may not be sufficient.
There's another really interesting temporal component, which is where you are in your lifecycle. And again, that could be not just humans, but also it could be the life cycle of the ant colony, which obviously is much longer than that of any particular ant.
Joe: I love two by twos for almost anything, a quick little grid. The one that jumps to my mind is personal versus environmental on the one axis and constant versus high change on the other. If you're relatively young, everything seems like it's changing to you. It's not a surprise that you're going to want to explore a lot of things, dating's the easiest example. I even think about when I travel somewhere, the first couple of nights there, I'll try a different restaurant every time. But if I'm not coming back for a year, my last night in town, I'm eating at the restaurant that I liked the best so far.
I'm not going to go exploring again. I don't know that that's the optimal rate, but it does make for a pleasant evening last night in town. I think that'd be a fascinating topic for you to dig into for people. I'm thinking of the book Swarm. I don't know if you saw that one yet and the author is not coming to mind so quickly. It's about how humans basically are social insects in the way that we behave. You probably have it on your shelves there. So for those who can't see, Michael's going to this enormous set of bookshelves behind him and digging it up.
Michael: No, I don't have it. The one I was thinking about was this one. Yours is just Swarm. This is The Perfect Swarm, so that's different.
Joe: No, that's different, that's about complexity, which is also a lot of fun and a whole different topic. Going back to the things you kicked off before. The beekeeping [point] cracked me up because, I don't think I mentioned this to you, but I started keeping bees during this pandemic. I'm home more and we decided let's go ahead and try it, but we lost all of our first colony. Have you ever tried?
Michael: I never tried them. I'm very allergic to bees. That would not be an ideal, but I'm completely fascinated.
And I would just say for any viewers or listeners who are interested, I mentioned Thomas Seeley at Cornell University, of whom I'm just a huge fan, and he wrote a beautiful book. And Joe, if you don't have this, you'll really like this, it's called Honeybee Democracy.
When I say a beautiful book, Tom is not only a very careful and thoughtful scientist. He's a beautiful communicator, both visually and in written and verbal communication. So it's a wonderful, gorgeous book and you'll really enjoy it especially with your beekeeping.
Joe: I appreciate that. So you pivoted a little bit by taking what was at the end of that book and turning it into a whole new rich area of inquiry around skill versus luck. When you think specifically about decision-making, base rates are obviously one [skill], what are some other things that are in the skill category for you? What are things that people can do that are relatively light lifts that would likely improve their decision-making quality?
Michael: I first had the opportunity to meet Daniel Kahneman, which was of course a thrill in 2005. I asked him that exact question. I don't think I can do better than telling you what Daniel Kahneman told me, which was: keep track of your decisions. And so when you make an important decision, not what did you eat for lunch yesterday, but an important decision, write down what you decided, why you decided, what you expect to happen, and if there are specific components that you can express with probabilities, write down those probabilities.
We're not talking about here writing War and Peace. We're talking about [writing] paragraph by paragraph, in your own hand, about a particular decision and then keep track of those decisions. I think, Joe, to your point, and just talking to Annie Duke over the years, that this is something that is obviously completely intimately understood by poker players. When you are in a probabilistic field where the outcomes are determined by both a skill component and a luck component, you have to focus on process.
If it's a pure-skill field, [like] chess, I could play Magnus Carlsen a hundred times in a row and he's going to beat me a hundred times in a row. Those outcomes are all we need to know that his process is better than mine. But in other probabilistic fields, it's conceivable that I could beat Annie Duke in one hand of poker.
It's not conceivable that I can beat her over time, but it's conceivable that I could get lucky and beat her once. The beauty of documenting the quality of your decisions is it allows you to focus on the component that you can control, which is that skill component. That making the right decision and recognizing the outcomes in the short run will not be necessarily perfectly correlated with the quality of your process. But over the long haul, those outcomes will be correlated with the quality of your process. So to me, the biggest thing is to document your decision making process and try to improve that.
Now there's a spinoff component, which is this idea of calibration. So usually when you have a view of the world that's different from others, you have a probability assessment of something.
It could be a magnitude of an outcome, but usually a probability assessment that's different from others. You could imagine just an XY chart where on the x-axis would be the subjective probability, what probably you've assigned, and the y-axis would be the actual outcome of that class, so that's the objective outcome.
And then when you’ve gathered a lot of your forecasts, you can start to plot where your particular decisions lie on that graph. The ideal is to be on the 45 degree angle line, so that when you say something's an 80% probability, it actually happens 80% of the time and a 30% probability happens 30% of the time. What's fascinating to me is that when people keep track of their forecasts and they do it in a way that's probabilistic, as I just described, and they get feedback on those outcomes, it turns out people get better at their calibration.
They get better at this. It’s very famously studied with weather forecasters, where they're getting immediate, timely, and accurate feedback. It's obviously more difficult in other domains. So keeping track of the decisions and then with an eye specifically toward improving your calibration, so that you're more accurate probabilistic forecasts. Again, it sounds like it's super onerous, but it's actually not that bad. You have a notebook. Go to a local drug store, buy a notebook for $3 and just document your decisions, write down those probabilities. Then every few weeks, gather up your probabilistic forecasts, see how the outcomes turned out, and then build up a little database. You can do this in Excel very quickly. You can start to measure calibration. Just the exercise of doing all that, I think, will make people much better without again a ton of additional apparatus, no fancy software, nothing. You know, this is stuff you can do with a notebook and maybe a simple spreadsheet and you're going to be in business.
Joe: So my sense would be that the majority of the benefit for most people who are new to trying to improve the decision quality is the intentionality of it and shifting over to System 2, like you were talking about before. They're going to be reflecting and actually taking into consideration things like base rate.
It's not so much that they're getting rich feedback from the environment on lots of repeated and similar decisions, and therefore improving their calibration from, [thinking the probability] was 30%, [when] it's really 35%.” It's the fact that they're even in the right area with regard to the 50%. They're actually even considering it.
Michael: I'll jump in on that because you're exactly right about that. I'll just give you a very concrete example of that. When I was in an investment firm and we had an investment committee, we would have people come and make presentations on specific kinds of investments they want to make. At the end they would say, how's this going to turn out? They would have to say, we expect these things to happen. Historically they would just do this generally. Like we expect them to pay a dividend or we expect them to sell the business. We said, well, we want you to give a probability for those things.
We want to give it more specificity. It was just remarkable just asking people to assign a probability, just that app, as you point out, compelled a much more thoughtful process, which then led to a much more rich discussion around the issues. So you're right. Probably the calibration is sort of level two.
What level one is when you pose the question, it compels [people] to think about it in a way that is much richer than they would have otherwise. Just mostly slowing down and quantifying things that they're saying that hadn't quantified before.
Joe: I very much appreciate the time you've given us so far. Michael, I'm wondering if I can finish with one last question for you, if we are successful with the Alliance, if we get Decision Education to be something that every young person is not only introduced to, but actually develops some skills in through during their K-12 experience, I say “if,” I think it's “when,” what would you expect to see different about our society or about individual lives, a generation or two from now, if everybody knew about base rates and keeping track of their decisions, and focusing on the process as opposed to outcome, and the other things that we talked about today, or that are related to critical thinking that we've been exploring with our Fellows program. So you're a visitor to the world. It's 50 years from now. A whole generation has been raised in the United States with Decision Education. What looks different to you?
Michael: I think there would be a lot of very good things. By the way, even in the business world and certainly a political domain, I think people will be more scientifically literate. That would be very helpful. They're able to understand and parse evidence and make the appropriate decisions as a consequence. I think people individually hopefully could make better personal decisions for themselves. The COVID experience was very interesting.
We saw some people get a little bit of money often just from the government, go into the stock market, and trade options and stocks. That movie hasn't ended yet, but we've seen this movie many times and that movie tends not to end well. Hopefully some of that would be offset to some degree.
I'll tell you there's one aspect of this that I struggle with, which is interesting, on base rates. I was at an academic conference. They sometimes let outsiders who are clueless into these things just to hang out. It was actually on behavioral strategy. These are professors who study competitive strategy for companies and they have an extra wrinkle, like what are the behavioral elements of that? It's a fascinating topic in general. They invited me to give a talk and I talked about base rates and luck and skill.
It turns out that there was a special guest that day there and that was Daniel Kahneman. We had this very interesting and spirited discussion where Daniel went into a big company and he said, “Here's this thing on base rates and most new products fail, and you guys have to be aware of that.” The head guy from the company said, professor with all due respect, if we actually believed everything you just said, we would never launch a new product and then we would never have these five new awesome products that we've launched. There is this interesting balance just in general, between optimism, and this idea that we all have little psychological bubbles around our heads, and we think we're a little bit better than we actually are. What's good about that from a motivation point of view, it gets you out of bed in the morning. If you're an entrepreneur and you actually know the base rates of failures for entrepreneurs, you might never try in the first place.
There's this really interesting tension between [the fact that] we want that optimism at the global level. We want people to try out stuff at the global level, but if you're making an investment, you have to be more measured so you know that aggregates aren't going to work out. I'll tell you one very quick story on this. My oldest son, when he was in university, was tempted to apply for this thing called the Thiel Fellowship. Peter Thiel has this fellowship like how college is not for everybody and if you have a really cool idea, we'll give you, I think, two years and a hundred thousand dollars and you dropout of college and you work on your project to change the world.
If it doesn't work out, you go back to school. It's not a big deal. They have about 20 Thiel fellows per year. My son said, Dad, I've got an idea, maybe I'd like to apply for this Thiel Fellowship, drop out of college. I was thinking to myself, at the high level, this idea for 20 people to do this, and probably two of them will succeed, is completely awesome. I think that's great. There have been great outcomes, but I'm not sure I want my kid to be part of it. He's most likely going to be the 18 of the 20 that don't get anything done. I had this sort of funny tension. I get the aggregate.
I think the world will be a better place because [it would] allow people to make better decisions. I think ultimately too that it allows us to understand others better. You know, I think it would contribute a lot to empathy. Even today you can see this. There's a wonderful book called Scarcity. People make decisions under a lot more duress than most of us go through every single day and understanding if they're better equipped to make better decisions, that's good.
It will also allow us to understand the decisions of others more effectively. I think that's also a huge positive just societally. I think the world would be a much better place. I just think the mission of the Alliance is so extraordinarily important.
There's a challenge of just a little bit of crowding out, which is [that] a lot of the incentives may encourage educators to achieve certain objectives that are short term and important in some ways, but these broader life skills are important to being a global citizen. They can get crowded out and that's a challenge. How do we introduce these really incredibly important ideas to allow young people from the get-go to make better personal and professional decisions. I'm sure the world would be a better place.
Joe: Well, so thanks for that. We have to come back to this topic at another time about the idea of what's good for the individual versus what's good for society. I had in mind Stanovich's The Robots Rebellion. Have you read that yet?
Michael: Yeah, of course.
Joe: What we're doing may be very good for our genes and it may be very good, even for the species, but it might not necessarily be in the best interest of the individual. I am curious about the question of how much optimism should someone have so that their lives are good lives, versus, [having] a better calibration with reality in the expected outcomes of things. I think it's a fascinating topic.
I'd love to chat with you more about it or what you're working on when we connect next. Michael, thank you so much for this. This is great for our audience to be able to hear how you're thinking about all these things and how you're continuing to apply them to the real world. Much appreciated.
Michael: My pleasure!