Loading summary
Transcript112 lines
- [00:00]
A
The following is with the debut Italian on this podcast, the one and only Luca Delana. Luca is an extremely successful author of nine books, and he writes these alongside his day job as an independent consultant advising businesses across the world at the intersection of risk and behavioral psychology. This makes Luca one of the best communicators of the subject for today's conversation, which funnily enough, is the very same title of his most recent book, Ergodicity. Now, we'll get into Ergo to see in this podcast, but I'm sure you'll all also recognize that it is a common theme throughout the inserto throughout Nassim to Lebs brilliant inserto. And so I'm bringing it to this podcast feed just because it's been up on the Curious Worldview podcast feed for a few weeks now. I just want to bring it to this feed to remind you all of the potential overlap in our interests. Let's assume we have more interest in common than simply the inserto. Risk, randomness and chance. All incredible themes for life. But I want this to be an example of the type of show which is every week happening at a Curious Worldview podcast. It is the top link in this description. Go over there, pump your good juice into that algorithm. Let's assume we've more interesting common Take a risk. In true Talebian fashion, step outside your comfort zone and check out my other podcast. And with absolutely no further ado, enjoy Yoga Dictity with Luca Delana. Okay, Luca, welcome, sir. You make ergodicity so clearly understandable in the book through examples. But when then I try to go and explain it to somebody, I come very, very short of words. So how do you describe ergodicity without examples?
- [01:43]
B
So for me, ergodicity is the study of the effect of time horizons on decisions and strategies. So what we see is that in the real world, there is no such thing as the optimal strategy. It's always the optimal strategy for a given time horizon and the study of how of which strategy is optimal for a time horizon or how a given strategy changes over different time horizons. That's a good. Easy.
- [02:19]
A
Okay, but before we think of any examples, can you. Is it possible, and I know it's a very difficult thing to do, but is it possible to tighten that up even a little bit more?
- [02:29]
B
So good. This is the difference between the outcomes of doing an action once and doing it many times.
- [02:38]
A
Beautiful. And why does it matter?
- [02:42]
B
Well, it matters because you might take an. You might compute the outcome of doing an action once and think that if you repeat it X times, you will get X times that amount, which in the real world is just wrong. Very often you will get less than X times that amount. And the study of ergodicity will tell you which actions you need to take to make sure that if you take an action X times, you get as close as possible to getting X times the returns of doing it once.
- [03:17]
A
Okay, so in the real world, talk about some of the domains that ergodicity applies to, or at least is most applicable and relevant to.
- [03:28]
B
The usual domains that come to mind are investing and gambling. We see a lot of examples in which if you have to do a gamble once, you would evaluate some gambles as positive, such as in poker. Sometimes it pays on a single gamble to take some risks. But if you have to repeat the gamble, then you must also think about survival. Because in poker, for example, if your aim is to win a tournament, you cannot take the same amount of risks as as if your aim were to win a single hand to maximize the amount of money in a single hand. The same applies to investments. Some investments might be a good bet if you're taking the investment once, but they become bad bets if you have to take the same amount of investment year over year for 10 or 20 years. And these are, for example, the investments in which there is a chance that you go bankrupt. And the reason is because if you're only considering one year going, like losing the investment means that you lose the money. And so if you have an investment that has a 50% chance of returning triple the money and 50% chance of going bankrupt, it's a good investment if you take it once, because the average is that you make 150, that you make a 50% return. But if you're taking that investment 10 times, you will think what you get is 50% compared to 10 years. It goes to, I don't know how much like triple or something like that. But in reality, if you take that investment 10 times at some point, you are almost sure that before the 10 years ends, you will end up bankrupt. And this difference is why it's so important to play your hand, your investments and your life differently. If you have a long time horizon.
- [05:32]
A
Hence the importance of the different time horizons.
- [05:35]
B
Exactly.
- [05:36]
A
Opening definition. But outside of games of chance, which you could definitely include investing in those games of chance, just the regular person listening who has no exposure in the markets and simply is just going about their lives, why does ergodicity matter to them?
- [05:55]
B
Well, for example, if you have a job, if you are a regular 9 to 5 employee. You might think that the best way to optimize your career during this year is to work as hard as possible. And that's probably true. Because if you work as hard as possible in only one year, there aren't many chances that you will burn out or that it will endanger your marriage or your health. But if you repeat the choice over your career and you work for as hard as possible every year, year after year, it's almost guaranteed that you will end up with a burnout or a ruined marriage or something like that. So if your time horizon is to maximize your career over 10, 20, 30 years, then necessarily the optimal solution is not to work as hard as you can, but as hard as you can without endangering your marriage, your health, your mental health and so on. Another example is again about your life. If you make any sport or if you just work out at the gym, you might think, I want to be ambitious, I want to maximize my health gains, so I will train as hard as I can, I will try to lift as much weight as I can this session. That's a bad recipe because maybe you will maximize your gains today, but it's almost guaranteed that in a few months you will injure yourself and that will create a very big setback. And if you want to maximize your gains over, let's say two years, then you need to work out not as hard as you can, but, but a little bit less. And this is not about not being ambitious, because this is about being very ambitious. If you want to be ambitious in the long term, you must think about survival and you must think about avoiding setbacks.
- [08:00]
A
Applying this example to bodybuilding is very appropriate at the moment. I'm not sure if you've seen, but over the last several years, and presumably for the history of bodybuilding, a lot of peak athletes, top performance have died in their 30s precisely because of this point. You know, they were maximizing the short term games gains in both sense, but gains into the, to the point of optimizing their body in the most elite condition possible in the shortest amount of time at the cost of long term survival. They're playing a very risky game and unfortunately some people have died because of it. But it, it kind of reminds me of the threshold to disorder in the Simta labs antifragile, which sort of gets overlooked a lot of the time. It's like it's not simply the case that what doesn't kill me makes me stronger. You know, it's more the case over a long time horizon. That which doesn't cripple me completely does slowly iterate to a stronger body.
- [09:09]
B
Exactly. And I love this example. So two reasons. One is because you make this point that even the antifragile, like our body, our body is antifragile. But even the antifragile, it has a threshold after which more stress causes fragility. And we should acknowledge this. And especially the more often, the more repeated the exposure to that stress that might cause fragility, the more likely that we will break down. And then the other reason why I love your example is that you talked about a domain in which people try to be number one. And my advice for everyone, very ambitious, is always don't aim for number one, aim for top 1%. And the reason is because there are reproducible strategies to get a top 1%. Well, maybe in some domains, maybe top 5%, top 1%, but there is a clear roadmap. And if you're willing to put the work, you have very good chances of getting there, top 1%, top 5%, but very, very good. And without exposing yourself to much of the negative effects. Conversely, if you aim to be number one, you will have to take very big risks. Either risks of injury, or you will have to spend so much time working out that maybe you will not be able to have a good marriage or something like that. And then even if you do everything right to become number one, there is the chances that there is someone else who did it just as right as you. Maybe he had some advantage, maybe he knew someone or he had a better genetics than you, and you will still not get number one. Whereas instead, if you aim for top 1% or top 5%, depending on the domains, if you do everything right, there is a very, very good chance that you get there. So it's not being less ambitious to be number one, to be top 1% compared to number one. But if I have ambition for you means realizing your full potential. You have better chances of realizing your full potential if you aim to top 1% than if you aim to top 2. Number one.
- [11:25]
A
This point is exactly said in a Taleb quote which you feature in the back half of your book, which is that solid financial success is largely due to skills, hard work and wisdom, but wild success is more likely to be the result of reckless betting, extreme luck and folly. I believe in the book you were using Elon Musk as an example to make that point.
- [11:51]
B
Yeah, exactly like Elon Musk, very skilled entrepreneur. So people think because Elon Musk is very skilled, it must mean that most of his wealth is due to skill. And I disagree. And the reason is you can make a thought experiment. You can think about 10 parallel worlds in which Elon Musk has the same upbringing, same skills, and then he goes to fund his first company. And then the ten parallel worlds start to diverge. And probably because he's very skilled in all of these 10 parallel worlds, he will be successful.
- [12:33]
A
He'll be top 1%.
- [12:34]
B
In all he will be top 1%. Probably. Maybe not, because he takes very, very big risks, but still in most of.
- [12:42]
A
Those, just because he's exceptional.
- [12:44]
B
Yes, in most of these worlds he will be top 1%. And maybe in quite a few worlds he will be a billionaire, but maybe he will have 1 billion, 5 billion, 10 billion. But in this world he has 230 billion. If the average wealth over his 10 parallel worlds is 20 billion and his current wealth right now is more than 200 billion, it means that 90% of his wealth, the difference is due to luck. And that doesn't take away anything from Elon Musk's skills. And this is very important. We need to recognize that it's not either you're lucky or you're skilled, but. But even if you're skilled, lot of your outcome will depend on luck. Especially the more you use high variant strategies which are necessary to get to the number one. And so the point I make in the chapter is two points. Point number one, even if you are the most skilled person in the world and you work the hardest, there must still be someone else who comes out ahead of you because they took more risks and they got luckier. And then point number two, if you increase your risks to get a better chance of Getting to number one, you will decrease your average outcome, which means over 10 parallel worlds you will increase your outcome in one of those 10 parallel worlds, maybe, but you will decrease your outcomes in the other nine. And most people, what they really want is not a chance to get number one if it means that they will be miserable if the chance doesn't go true. What most people want is to be quite successful with high certainty. And if that's what you want, then you shouldn't aim to number one, but you should aim for top 1% or top 5%.
- [14:56]
A
And then although this group of people in the top 1% would largely all be similarly skilled in their exceptionalism, it's so non linear. The difference in outcomes from being in the top 1% to then being Musk, Bezos, Arno, etc in the top 0.0001%. It's, it's. Maybe you could comment on just how dramatically different it is, say between the top 50%, the top 1%, and then the top 1% and the top 0.1%.
- [15:34]
B
Exactly. The difference is dramatic because of these fat tails. But let's also not look only at the positive side of the fat tail, but let's also look at the negative side. If you are Elon Musk or Jeff Bezos, all the negatives of being super wealthy are exceptionally increased. All your eyes are on you. You get old people trying to scam you, all the kind of things. If you are number one, you don't get a chance of having the good things of a normal life. Whereas if you get top 1%, you will still get a lot of the benefits of being of an exceptional success and you will still get very few of the negatives of extreme success. If you are the wealthiest 1% in your town, probably you will not even have reporters under your doorstep and paparazzi and all kind of negative things. So that's another reason why I advocate for going for one top 1% and not top number one.
- [16:43]
A
But let's move away then from how wealth might be distributed to a different domain, say podcasting, for example. Now, this podcast is in the top 1% of downloaded, but it is a fractional, is not even fractional. It's a rounding error for one of the top 0.1% downloaded podcasts. So again, if you could double down on this point of the dramatic nonlinear difference in outcomes, the further away you get from the mean.
- [17:14]
B
Yeah. So first of all, congratulations for being in the top 1% of podcasts.
- [17:18]
A
It's not a lot, I'm telling you.
- [17:21]
B
And yeah, so first I want to comment on this it's not a lot thing. Like, so of course it's a lot of hard work and skills. And it's also that you've been doing that for a long time. And I remember a tweet from Morgan Housel where he said, if you get average market returns, but you manage to get the average market returns for 20 years straight, you will probably get in the top 10 or 5% of investors. And that is both true. And it just shows how much survival plus not having setbacks that allow compounding can bring you. So this is already one thing. And then what you're saying is completely true. Like, even if you are top 1%, just because the distribution is so much fat tail, you still probably have a lot of downloads, but very, very fewer than the top podcast. And on this like part of it, sadly, it's in the rule of the game. Because a lot of people, one of the criteria, for example, for choosing which podcast is just, oh, maybe I will listen to the most popular podcast. It's popular, must be good, something like these, or it's popular so I can talk about it with my friends. And so sadly that's, that's in, that's part of the, part of the game. But the other thing I've noticed about these non linearities is that the same thing applies to my readers and I'm sure that it applies to your listeners. There are most of your listeners and most of my readers who have a very little audience. And there are a few listeners and readers which have a large audience. And one recommendation from one of those power listeners or power users can grow your following base or subscriptions by a lot. And so for example on this, I've read this concept, I think it was from someone called Brian Dino or something similar where he said, I write my content to be useful for everyone because that's what I'm trying to do, like help a lot of people. But the way I present it, the copy that I write when I market my books and my products, I write it for the people who have the power to then distribute it to a large audience. And I think that that makes really lot of sense and it's something that I'm trying to, to also implement to some form. So it's not like you compromise the content because the risk is that then you compromise the content, you try to do engagement bait and whatnot. So it's absolutely not that you don't compromise your content, but then when you're talking about it, to move it out there, you try to appeal to those power users and that's another way of like instead to leverage those nonlinearities that exist out there.
- [20:36]
A
Are you applying your education and ergodicity to the various domains within which you operate?
- [20:44]
B
Yeah, it definitely applies to consulting, meaning that I make a very big effort not to work as hard as I can, but leave plenty of time to my health, to my family to rest and all these other kind of things because I want to be able to keep doing this job at top level for like until I'm 60 or 70 or something, not just because I really love it. And for books, for example, what I'm trying to do is that I never try to maximize short term attention. I never try to write the sensational tweet or the sensational headline or anything like that, because I know that those are things that bring attention in the short term, but they break trust with my oldest reader. And if I were here only for six months for any reasons, then it would make sense for me to go for the exceptional headline. But. But if I'm here for the next 30 years, which is what I'm trying to do, then it really doesn't make sense for me to compromise the trust of my readers in any way. And that's why, for example, I always try to never dumb things down. I try to write as simple as possible. For example, I wrote a book on ergodicity without a single mathematical formula. But I never try to dumb it down because I never want to compromise the trust in my readers and especially in the part of my readers which is the most, the most engaged one, the one that wants, that sees the most value in what I read. Not like the occasional reader.
- [22:33]
A
How do you wrestle with Taleb's commentary on consultants in Skin in the Game?
- [22:39]
B
Well, I think that Taleb has a good point in general because as consultants we don't really have effectively skin in the game in the companies that we work with. And that's a problem that I tried to solve. Like in the past I tried to set contracts in which I would get paid years afterwards based on the results. But I discovered that my good intention simply went possible. Meaning that apart from a lot of tax issues, like not, not that it's disadvantages, but I mean like really sometimes it's not possible. Like government is saying why are you not getting paid for the work? So other than that, also like companies like they are okay to have for some type of engagement, they are okay to have longer term things like this, but not really and especially not at the point in which it will be economically sensitive. Because if I know that I would do a good job, so if you're going to pay me on results, then I will get a big, a big chunk of it. And you will think that for companies it would make sense to put just a fee and proportional. But I've noticed that it's not really how it works. Like you really don't get many, many companies that are interested in doing that to the level. So as much as it sounds good in practice, like it's not really easy to do that. Now how you can solve the skin in the game problem. So first of all, you can try to look for consultants which do have skin in the game. And I think that as an independent consultant, I have, for example, more skin in the game that a consultant that works for a large company, because the consultant that works for a large company, they will just move to another consulting company. Maybe they are not really touched personally on their reputation. Whereas myself, if I screw up on a project, the R will be done. And so that's for me a very good incentive already not to try to optimize, not to play some of the games that Taleb describes in the book from some content. And then the last thing is, and this is something that I wrote like. Like it's a bit of my pet peeve is in a consulting, when you are interviewing a consultant, like during the first call, when the company is saying what they need and the consultant is giving a high level view of what it could be done, you can already see whether that's a consultant, whether the engagement will go well or not, or at least whether it's an engagement that's structured in a way that it's not one of these Rubin traits that are in the books. And you can see it from both parts. So first thing is the company, like some companies, they come to consultants and they want something. I just want you to come for a three hours workshop and that's it. I will talk your fee, we will never see each other again, and that's it. And that's already like something which already has very big limitations. And I give you two reasons. One is, even if the person is the best person at giving workshops in the world, the chances are that that will remain a workshop because there is no follow up. There is no talk even about how the follow up will be. For example, the managers that attend the workshop will learn something. Is there something that will make sure that they apply it in practice? If it's not there, then already the engagement is structured in a way in which you already are taking some risks. And then you see it from the point of view of the consultant. A good consultant will ask about what's the follow up? How we make sure that this thing bring results, how we make sure that people will not just gain the same. And if you don't see the consultant on the other side that is asking these questions, then you know that you have a big risk. And if you see the consultant that is asking those difficult questions, then of course you don't have the certainty. But let's say that there is a chance that you are in a much better situation.
- [27:28]
A
Nice. I just asked because, I mean, I agree there is a lot of consultants out there that do very good and necessary work. But it's just funny, it's in very classical Taleb fashion in his writing style, that it is all or nothing. There's, there's a, there's no in between and the way he'll just disparage, you know, an entire field of people.
- [27:51]
B
And if I, if I can add one more thing. The thing is that even if you get like the point, the point is that sometimes companies, they don't think about the skin in the game problem. And you must realize that in basically all cases the responsibility of skinning the game stays within the company. Which means of course you want to hire someone which will give you good recommendations. And of course as a consultant you have the ethical obligation to give good recommendations and hopefully you do and you think for the longer term of the future of the client and, and surely I do. That said, the company has a responsibility to never take what the consultant says at face value and always think whether it makes sense for them and whether it makes sense for the long term, because ultimately they are the ones who have the authority, who have the knowledge of the risks and so on. So I just wanted to do a comment like operationally on Taleb. He's completely right that for example, the governor of the bank doesn't have the correct incentives and that's a big problem. And he touches it very correctly and fully in his book. One thing I will add is that every organization should recognize that most individuals, unless they are the founders, they don't have and inherent and they don't have to the same degree the skin in the game that the company itself does. And if they care about the skin in the game of the company, they should put some processes and some care into making sure that those people, that even if there is a person with not fully aligned incentives, which is almost impossible to do, still you will not get decisions that put the company at risk.
- [29:49]
A
I just want to quickly go back to the fat tail distributions that we were talking about earlier. There was wealth and then podcast downloads. What does, what does, what does? The lessons there in the non linear difference in outcomes, the further away you get from the mean. What is the lesson for ergodicity within that?
- [30:14]
B
Well, the lesson is that in those games you need, if you care about getting like in the top, like if you care about making an impact. So for example, having your podcast really listened by millions, something like this, you need to consider in your strategies, like in how you market your book and so on, you need to consider that you need to have strategies which have the potential to create the variance that you need to get there. So this is one aspect that's one Phase of the medal. So whenever you ask, like if your objective is to get, I don't know, 1 million downloads, you need to ask yourself, if I do this, is there the potential that I get to 1 million download? And if the answer is not, if I do this, probably I will get, I don't know, 10,000 downloads, then you know that it's not enough and you need to do something else that has this variance, this risk, this positive risk, and upside that it gets there. That's the face of the metal. And then the second phase of the medal is that you must take those risks only in a way that do not compromise yourself like that. Do not endanger yourself. And by this I mean maybe you make some statement which is a bit hedgy, but it should never be edgy to the point that you lose your reputation. So you need to find this balance. Edgy, but still correct. Edgy, but still useful. Like, never had you so exaggerated that it actually breaks trust that it's not useful anymore. Something like this, you should work hard, and probably it means that you should find ways to work harder than most other people at the podcast, but you should also do it in a way that don't compromise your marriage, your health and so on. Like, these are the two phases that you should always consider, and if you consider them, at some point, you will find something that works for you.
- [32:28]
A
Who is Ole Peters?
- [32:32]
B
Ole Peters is a researcher that wrote probably the paper that explained ergodicity the best from a formal way with and explained some questions such as why does insurance exist? And if you are interested in the technicalities of ergodicity, he's definitely the person to look for. And he has a PDF called Ergodicity Economics that's freely available on the Internet that gives you a good technical overview of ergodic. However, he's extremely technical and you need to be able to follow advanced math for. And just, just on this, I want just to touch the point on insurance, which, which I think is very interesting. Like Ole Peters was bringing this problem, which is if you have a house, for example, that's 1 million of value and it has a 1% chance of burning. You would want. The maximum you want to pay for insurance is $10,000, 100% of the value, right? It doesn't make sense for you to pay more than that. However, for the insurance company, it doesn't make sense to get clients that pay less than €10,000. And if you think only about expected values, about the expected value of the insurance, there is no Overlap in which both the insurance and the insurer, sorry, the person insured and the insurer, they think that it makes sense to shake hands. However, ergodicity explains why there is such overlap. And the reason is because the person, the individual cannot survive their house burning down because then they're bankrupt and homeless and what do they do? So they are willing to pay a bit more than €10,000, whereas the insurance, if an house burns down, they don't have their survival at home because they are insuring a thousand houses. And because of that you have an overlap between, between the two prices and hands get shaken. And that's just one of the things that you can read in his paper and.
- [35:09]
A
Yeah, is that an original discovery from Odler?
- [35:15]
B
I haven't read it anywhere else, but I cannot comment on whether it's his discovery or not.
- [35:23]
A
I'm, I'm. This sounds so nerdy and, and kind of silly, but I am fascinated by how insurance actually works because it underlies everything. And I'm fascinated by how much failure there is to understand the difference between the projected risk of ruin, the projected fire. You know, you say it's a 10 chance. It's like, well, how many unknown variables did you fail to compute to get to that 10 chance than the actual chance of ruin? And just how in there, in the difference between those two numbers so much of the world economy is sort of hinged on. I just think it's super fascinating.
- [36:15]
B
Yeah, I think, I think so too. And if you think about it like a lot of it is about insuring things that have very much not large tails and reinsuring themselves so that everything stays not fat tailed. But I'm always thinking like if there are really, really like for other domains, indeed, it's a miracle that is working somehow.
- [36:46]
A
Okay, so to this Ola Peters fella again to go to take us both back to square one where we started trying to define ergodicity without examples. Ole writes in writing my book Ergodicity Economics, I keep running into the problem that, that ergodicity economics resolves some big issues in economics elegantly, like some of these examples you've already given. But it's very hard to explain to the layperson why there was an issue at all in the first place.
- [37:19]
B
Yeah, Maya, like my two, like I would advise only to follow the two principles that I followed in my book, which is, number one, don't mention the word ergodicity even until you are halfway through the book. Because if you try to start with defining ergodicity you get into all these kind of awkward things and you need some building blocks in the reader and just don't do it. Start with examples. Build an understanding of what the problems are with examples. And only after, only after a large point you can try. Start using the word ergodicity. And then the second thing is, well, of course, like, we are appealing to different audiences. Like, I suppose that Olli is appealing to a. More. To an audience with a larger background in mathematics. Whereas the objective of my book was to let those people let this concept be understood by everyone. And so my choice was I won't use a single mathematical formula in the whole book. And that was almost like a creative constraint that forced me to find good examples that would resonate, that people would immediately pick up, and enabled me to give language to readers of the book to use in their everyday life. Like the. The most common comment that I get about the book is these are concepts that I somehow intuitively understood, but I didn't have a language to talk about them. And this language cannot be a mathematical language. It has to be a non mathematical language.
- [39:12]
A
That must have been amazing feedback to receive. Really says that you're onto something.
- [39:17]
B
Yes, yes, thank you.
- [39:19]
A
And I have the exact same feeling from Talebs in Certo. It's all kind of instinctual knowledge. We kind of get it to some degree just by experiencing life. But then to explain it is a whole nother thing. And then it's beautiful. You can start seeing it and applying it in all these other places. Like, for instance, your description of your cousin, the professional skier, and the reason why he's not now a professional skier. That simple example of ergodicity. I've been trying to. Well, not trying to, but I'm thinking about how it could apply to a game that's very close to my heart called cricket. So maybe I could explain to you why I think it applies to cricket. But first, if you wouldn't mind explaining to the audience the anecdote of your professional skiing cousin.
- [40:13]
B
Yes. The example I make is that my cousin was extremely good at skiing since a very, very young age. I think he must have started when he was 3 or 4 years old. And he was very good. He made it to the world championship for his age bracket and then he had an injury after the other, and before he's 18 years old, he had to quit skiing competitively. And the lesson I got from him is that it's not the fastest skier that win the race, but the fastest skier of those who finish the race. And the broader point is that survival, sorry, performance is subordinate to survival, which means, of course, performance determines who comes ahead, but only between those who survived. And that's why survival is even more important. And then I get like one point that I get from some readers at that moment is, yes, Luca, but at the end, a race is so short and you can just give it all in that race. You can just take some risks and that is what will bring you to win the race. And yes, but that's only one race. And if you want, yeah, if you want to be a professional skier, you cannot just run one race. You will have to run thousands of races, thousands of races in the competition, plus so many other ski practice runs. And if your way to get faster and to get ahead is to take exceptional risks, you will get yourself out of your job very quickly.
- [42:08]
A
So even though your cousin may have been in the top 0.1% of all potential skis in the world, on the time horizon of say, a 20 year career, he wasn't around to even play and therefore not, you know, not realized, even in the top 1% of skis. I think it's such a compelling way to look at it, to reinforce that time horizon point from the beginning. But then as well, just how avoiding ruin. Avoiding zero is the highest possible, most important goal on a long time horizon. And it's amazing as well some of the economic examples that are given how people can accumulate enormous wealth over 15, 20 year careers and, and lose it in an afternoon. It doesn't matter on the next day what their career was before that because they've got nothing to show for it. It's just like avoiding risk, avoiding ruin.
- [43:08]
B
Exactly. Ryan. I love this point. And back to the very beginning of the post of this podcast. Something that people misunderstand is that they think that if you want to be top 1%, then you need to be top 1% every single day. And that's not true because there will be people who drop out. And because there are people who drop out or people who have a much briefer career than you, it's very possible that if you are top 5% every day, you will end up in the top 1%. So it's okay to carve some slack into your everyday life. It's okay to, to play it safer than, than it looks like at the.
- [43:56]
A
Risk of beating the same horse to death. And maybe the audience won't want to hear it, but at the risk of that nonetheless. Can you explain what you mean by being in the top 5% every day might mean that you're in the top 1% in the long run.
- [44:12]
B
Well, let's imagine that I want to be in the top 1% people on Twitter. I might be tempted to look at the top 1% people by engagement today, and I might see that they post, I don't know, I'm saying a random number, 30 tweets a day. And I might think, ah, to be in the top 1%, I must also post 30 tweets a day. And then maybe because of that, I either burn out or I write terrible tweets just because I need to get the quota and I break the trust of my readers and I end up nowhere. But the thing is, if you look about people who are in the top 1% of Twitters, not like the top 0001%, but people who are in the top 1%, probably you will see that they don't necessarily tweet so much, but they very consistently tweet things that make sense. Or you might see that they tweet a lot, because a lot of times when you tweet, you don't know whether the tweet will be very successful or not. But my point is, whatever they do, they do it in a way that it's sustainable to keep doing it for 10 years. And that's the point. You should not imitate anyone who you cannot imitate in a way that's sustainable. This is the principle.
- [45:36]
A
And there are so many examples of the benefits of that accrue to you if you play the long game. Does compounding exist in all domains? Therefore, that's why it makes sense to hang around for the long term, why a long time horizon is important, why being in the top 5% over a long time beats being in the top 0.1% in the short time. Because these benefits come from compounding.
- [46:06]
B
I do think that compounding more or less applies to all domains, of course, to very different degrees. There are some domains in which there are bounds to how fast you can compound or to how much you can compound until the end. But I do think that in principle, compounding applies almost everywhere, if not even just in your own compounding. In, for example, the fact that if you get stronger today, then you can lift stronger weights tomorrow. In the fact that if I become a better writer today, that I can write things that I couldn't have possibly been writing before, and so on, and.
- [46:53]
A
Those marginal gains are only observable over a long time horizon as well.
- [47:00]
B
Correct.
- [47:02]
A
I might end up cutting this out, but I just want to Try and explain why I think this the you can explain ergoticity in cricket and why a batsman might benefit from taking these lessons. Are you familiar at all with the sport?
- [47:15]
B
What happens very briefly, like I think I've watched one game once.
- [47:21]
A
Okay, so I'll just say that one half of the game is the batsman and the way that they score points is by hitting runs. But every time they hit a run, they very they expose themselves to all the types of way to get out, all the types of ruin, but at the, not even at the margins, but just by hanging around for a long time, you'll inadvertently face enough balls and play enough low risk shots that you will slowly accumulate runs anyway. And therefore, although it would be a very boring spectator way to look at it, a batsman who simply only prioritizes not getting out when they go in will over the course of a career likely score more runs than a super talented person who likes to try and score runs off every single ball.
- [48:15]
B
Thank you. Wonderful example.
- [48:19]
A
Hopefully, hopefully it can make sense. One day we can watch this cricket game together and you'll see exactly what I'm talking about. This is from the book Behavioral change is non ergotic. The distribution of efforts matter. So I would like to ask you to explain that more, please.
- [48:42]
B
Yeah. In my work as a management consultant, I see so many managers will repeat maybe like a core value once a month or they ask their people to do something differently once a month. And those managers, they might keep asking it once a month for their work career and not achieve any change. And the reason is, for example, imagine that it's a warehouse manager and he's asking people to store the components correctly. And on the first day of the month he has to store the things correctly. And then on the second day the people do it. But then on the third day the people put the things incorrectly and the manager doesn't notice or doesn't tell them, they will keep doing it incorrectly, they will lose the good habit. And then after one month again for one day, the manager will remember people out of good habit and then they will forget again, no change whatsoever. Instead, managers that remind their people of the good habit every day for one month they achieve lasting change. And then they might go the rest of their careers, okay, maybe not, but almost the rest of their careers without repeating it again. And the habit will still stick because it was ingrained during those 30 days. And that's what I mean by behavioral changes. Non ergodic. Doing one thing many times is like doing 100 times in one month. And doing it 100 times in 30 years yields very different results.
- [50:27]
A
And it's similar reasoning, but not behavioral change. There was something from the book that stood out to me as well. In the subject of customer acquisition, you're far better off in terms of measuring how effective you are at acquiring customers to reach out to say the same 10 customers 10 times than 100 customers one time, which I think is replicable in reality and as well. So it's like nice to actually confirm it as well.
- [51:05]
B
Yeah, exactly. Like you might make 100 sales, but if those 100 sales, they come from 100 different people or 10 times the same 10 people, like 10, 10 people getting the product 10 times each. You get very different profiles with the very different long term outcomes and very different strategies which are optimal. So you cannot just average it out.
- [51:34]
A
There was a really interesting chapter on the Kelly criterion and as well I would love to hear you explain what are examples of the Kelly criterion in nature.
- [51:47]
B
Yes. So the Kelly criterion is a way that gamblers can use to decide how much to bet on each gamble. And the idea is you never go all in. You bet a small part of your wealth. How much you bet depends by what is your hedge. But you never go 100%, it's always a small fraction. And the reason is because even for example, if you have a 70% chance of winning, and like you double your money if you win, and you lose everything if you lose, so it looks like a good gamble, but if you go in at some point, you will lose your money. And if you play 30% of your wealth on each gamble, still at some point there will be three losses in a row and you will lose everything. So you need to play much smaller amounts to ensure that you can keep sustainably playing the game so that the law of large number will apply and you are almost guaranteed to win what your hedge would say that you win. So this is the Kelly criterion and it has a formula with which you can calibrate how much you bet. Now, there are parallels in natural life and they come from moods. And in the book I make this the example of Imagine two hunter gatherers that eat by collecting berries. And one hunter gatherer has no moods, they don't feel moods. And the other hunter gatherer is very moody. They get excited and then they get demoralized and so on. And the question is, which one will collect more berries? And the answer is it's going to be the one which can fill moods. And the reason is because berries in nature, they are not homogeneously distributed over the environment. You will have a few bushes clustered together full of berries, and then you will have a few bushes without any berries. And if you don't feel any mood, once you get to a bush that doesn't have any berries, you will check it, and then you will check the one next to it, and you will check the one next to it and you will lose a lot of time. Conversely, the gatherer which has mood, they will sample one bush and if they find berries, they will be very excited and they will keep looking at the nearby bushes, which are more likely to have berries. And then maybe they hit on a bush that doesn't have any berries and they will lose motivation. And instead of checking the bush which is nearby, which has a low chance of containing berries, they will maybe work for a little while. And that's a much more efficient strategy. And so this is a form of Kelly criterion, which is decide how much time you dedicate to a task based on your observed payoff, recent payoff of the task. That's a way to adapt to non homogeneous environments, and that's a good adaptation that nature gave us.
- [55:18]
A
Can you give that same example in a modern context?
- [55:23]
B
Yes. For example, myself, when I decide which books to write, if I meticulously wrote on each single idea for books that I have, I will spend a lot of time writing books that maybe are sound but people don't really want, or maybe that they present an angle that doesn't make sense to the reader. Instead, what happens is that as I start writing a book, I maybe publish a few tweets about it, and if people respond well to those tweets, I get excited, I think I'm on something, and I spend more time writing that book, which is good because that book is more likely to have a good payoff. And then some other times, instead, I would start writing a book and then I publish a few tweets about it and I will see that people don't really respond to them, and then I will lose motivation and maybe I will put the book on the side and work on something else. And that's good. It prevents me from spending too much time on things which have a low payoff. Then of course, excesses are always bad, but within reason, that's usually a good adaptation.
- [56:36]
A
I can't help but think of survivorship bias when you give that example. Similarly, earlier, when you were talking about emulating the top 1% of people on Twitter as a strategy, I'd love to hear you reflect on the role that survivorship bias has in your overall, overall ergotic worldview.
- [56:58]
B
Yes. In the book I make this example of mimetic societies in which I explain that the human tendency to imitate others is a positive adaptation. And that's because. Imagine that you get catapulted to the past. You are in prehistoric times and you don't know how to survive. Your best bet is to imitate what people do around you, even if you don't understand, because the behaviors that they do are likely to be behaviors that are good for survival. And the reason is because if you can imitate those people, it's because they are still around. Probably in that tribe there were a few other people which had behaviors which were not good for survival, but they already died. They are not there to be imitated anymore. So this is the rationale for imitation and that's why it leverages survivorship bias and it's why we are prone to survivorship bias and it's why we imitate people. Now the problem is that that worked excellently in past environments where people had a lot of skin in the game. It works much less today where people have less skin in the game. And it might happen that someone which displays behaviors which are not good for you to imitate, they are still around to imitate. And this is particularly true on places such as Twitter and so on where you might optimize for short term engagement and you might get short term visibility, but that visibility make you prone to being imitated. And that's why generally, as a rule of thumb, you shouldn't imitate someone who hasn't been around for a long time.
- [58:51]
A
The Lindy effect.
- [58:52]
B
Yes, that's. That's one of the things about Lindy effect, like one, like we probably know, like most of us know that India fact as the longer someone has been around, the longer it's likely to stay around. The rationale behind that is that age is an inverse estimate for hazard rate. Of course there are bounds because we know that once someone gets to 90 years old, his hazard rate actually is higher than one who was 50 years old because of biology and stuff. But before you get to the boundary, this is correct. Age is an inverse estimate for hazard rate. The chances that you will disappear soon. And this is why Lind is a good proxy for what to imitate. Because the longer someone is around, the lower the chance that the behaviors they exhibit have a high hazard rate.
- [60:02]
A
What do you think is Linda Lindy about podcasting?
- [60:07]
B
Well, I don't listen sadly to enough podcasts to be able to say, but My guess would be it would be very interesting like to look at what are the characteristics, the similar characteristics for podcasts that have been along for a long time. Then of course you want to like remove the fact that maybe some podcast is successful just for the fact that they, they were the first one ever, you know, like this kind of effect.
- [60:36]
A
But who do you have in mind there?
- [60:39]
B
I, I don't know. I don't, I don't listen to enough, I don't, I, I don't listen to enough podcasts to be knowledgeable about this. But, but that would be my, that would be my reasoning. Like try to understand about, look for, look for some and especially look at something that people who haven't been around for long are doing, but not people who have been around for long. I would be very skeptical on this. And if you think about it, there is a ton of engagement bait that goes in this, that goes in this direction. Tons of trust breaking moves and so on.
- [61:18]
A
Yeah, it's at times demoralizing to think that you're competing with certain things and then at times really uplifting because you feel like you, you know, you're doing sort of the right thing, content with going slow, not trying to win any short term gains with the, with the overall point in mind that just time spent after a certain threshold is a pretty good indicator of quality. But yeah, it can be demoralizing sometimes and you know, throw your hands in the air and say, woe is me. It's all unfair.
- [62:02]
B
Yeah. And by the way, I heard on this point that you made, I heard someone a couple of years ago, I cannot remember which was. But they made a point similar to if you have a business, rather than aiming to optimize growth or something like that, just think what has to happen, what has to be done for your business to be still relevant in 30 years. And if you manage to do that, probably everything else which you need to do will descend from there. And I tend to very much agree at least to have some fundaments and then on those fundamentals elements you can, you can build something on top of this.
- [62:46]
A
That what you just said reminded me of this point from the book, which I just want to, if nothing else, complimentary one for it describing so clearly before we finish the chat. And it was the point on how natural selection will either be acted upon you or you act upon it when it comes to the longevity of a business. And you gave the point in reference to whether you fire underperforming employees or whether you keep them around because you don't have the stomach to fire them. And it was such a point, it was such a brilliantly well made point because ultimately the takeaway was natural selections coming for you either way.
- [63:28]
B
Yeah, yeah. No, this is a point that. So there is this course that I run which is regularly, which is called antifragile organizations. And the main point that I'm making is change is inevitable. It might happen on you or within you. Like it might happen on your company if your company doesn't adapt. Or change might happen within your company so that your company adapts and it will not get excluded from the ecosystem. The same for the employee. Change might happen on the employee, which means circumstances, changes and the employee might get four fire. Or change might happen within the employee so that he adapts and he becomes fit when the environment changes. And anti fragility, lot of anti fragility is about pulling changes forward in time. So adapting before there is the need, overcompensating. They are all forms of pulling change forward in time and then pushing forward change within you so that it doesn't happen on you. And like common mistake that I hear about from some readers of Antifragile Taleb's book. I get like I hear them asking the question how do I become antifragile? And for me that's the wrong approach. That's the wrong question because you are all identified, fragile. The question is how do you become more antifragile? And that's a world different answer. Which, which, which is much more actionable. And becoming antifragile, for example means pulling change forward in time. What does it mean pulling change forward in time? It means you are. You listen to problem and you are changed to problems in accordance to problems before they hurt you. One practical example, apply problem solving to near misses. Like every one of us, both companies, companies, they make incident investigation. There is an incident, someone gets hurt and they make an investigation and then they change something. We do the same. We crash with our car, for example. We change our behavior as a result. That's necessary, that's not sufficient. Because if you only change in response to what hurt you, you guarantee that you will be hurt. Instead you want to change before something hurts you. And how do you do that? You change in response to near misses, which is if I cross a red light because I was distracted, even if I don't get hurt, I must take it as a signal that some change is needed because maybe the next time I would be hurt. If you are in a company and an accident happens, I don't know something falls but no one gets hurt, you should still adapt to that because the difference between being hurt and not hurt was just luck. The near miss highlighted a problem that you need to adapt, otherwise you will get hurt to and that's an example of what pulling change forward in time means.
- [66:49]
A
Trying to identify risks you avoided, even though you didn't have to experience the downside of the risk could be a way to think about it in your daily life as well as you're going about things.
- [67:03]
B
Exactly. And that's why for example, when you do risk management analysis, one mistake of people do is to look only at the individual or company at hand. Instead you want to look at all the other individuals or companies in the same situation and look what they suffered from. Because even if it did never happen to you, it's likely that it might happen in the future. And so, for example, one good exercise if you have a business is not to just to think why can my business fail? But you also want to ask yourself why did the other businesses fail? And you should probably adapt to this even if you feel like it doesn't h it doesn't apply to you because the chances are that actually that it will apply to you.
- [67:54]
A
So Luca Nissim Taleb, he features prominently throughout the book. He is surely the largest influence on my own world view. You know the five part series the Incerto. Have you met the man?
- [68:14]
B
So I unfortunately I never met Nasim Taleb. There were a couple of occasions in which we were in the same town but unfortunately didn't never manage to meet. I would love to meet meet him. He's definitely been the largest influence on my work.
- [68:31]
A
And have you had correspondence with him?
- [68:35]
B
Not really correspondence. Like did comment on each other's tweets a few times, but not which I would properly call correspondence.
- [68:47]
A
Is he familiar with your book?
- [68:51]
B
He mentioned that he bought one of my books, the the World Traumatifying Glass. But I don't know what, what was his reaction? He never mentioned.
- [69:02]
A
Okay, you've got to get your mate Russ Roberts to put in a good word for you.
- [69:08]
B
Yes, totally.
- [69:11]
A
Okay. But maybe if you could reflect a little bit on how he's influenced your own worldview.
- [69:19]
B
Yeah. So first of all, I never really thought like so many of the problems that he mentioned in his book were things that I either never considered before or I vastly underestimated their importance. And just that is really invaluable and that's why they're always my first recommendation when anyone asks me about books. And secondly, he Gave a very good language and examples like if you pay attention when you read, you really, really understand some problems. And I don't think that I will be so good at my job now if I hadn't read him, meaning that probably I would have been quite good on technical short term side, but I would have forgotten about a lot of things which are the ones which make the difference over the long term and over the sustainability. And so really unbelievable.
- [70:26]
A
For example, talk about how he's tangibly influenced your worldview.
- [70:39]
B
Well, one example is the fact that you shouldn't really focus on what's frequent, but you should focus on what's important. Like this idea which you get already from the black swan. I think it's quite important and it's an influence. Like in my job I try to prioritize the actions I do. Not based on apparent relevance, on apparent frequency, but on impact. Same thing like when I do risk analysis and so on. So this is definitely one thing. Another thing is the importance of skinning the game. Again, not as incentive because I already knew about the importance of aligning incentives, but the importance of removing the people who are wrong from the position of creating harm. Again, that's the point which I came to appreciate much more thanks to him. And then once you see it, you don't unseat anymore. You notice, for example, that some companies are very, very good at doing this internally at removing people who created harm in one way or another from the position of creating this harm. And that's something that companies that are not good, some fail to do in some way. Now these are some of the things then of course fooled by randomness was invaluable in collecting data. Like in, in my approach to analysis.
- [72:23]
A
Analysing data, collecting anecdotes and stories and stuff.
- [72:27]
B
Correct. That said, that was something that I think I was already doing before. Like one thing that I tend to do quite differently from a lot of other consultants is that I don't collect data because I think, well, one reason is that if a data is likely to, to a dysfunction, that dysfunction is very much likely to have. So if I'm collecting data because I think there might be this function that dysfunction is probably tainting the data and there is no point in collecting the data anyway. That's already one reason. And then the second reason is that in companies, people, you get some people which are expert at changing the data, manipulating the data, making the letter data look good and so on, you get some people whose career is centered on that. So you cannot rely on data or if you collect data at scale, because you're collecting it at scale, you will be distant from the people measuring it. And because of that you will lose so much nuance and information. So I never do that. And instead I try to talk to people. I try to talk to people at home, all kind, all levels at the organization. I talk to talk with, I try to talk with two line workers, two middle managers, two senior leaders. You usually get such a better picture of the organization and of their problems and of their opportunities.
- [73:55]
A
I'm speaking to Scott Patterson next Tuesday about his book Chaos Kings, which profiles universa. To Leb Spitznagel Bill Ackman's Covid trade. I wanted to ask you what question you suggest. I ask Scott to cut to the central message, which is the value of taking daily losses over time just so you remain available to the extremely rare, unforeseen circumstance.
- [74:24]
B
Cool. That's great. I'm like halfway through his book, so I didn't get the to the end yet. So maybe what I'm saying is not correct, but I think that couple of good questions will be number one, like what principles allow them to keep taking the losses over time. So I'm not meaning like, I don't mean like the rationale which is quite clear, but I mean in the face of losing money every day, and maybe people questioning you about it, like what are the systems that other people should apply? Like, imagine that you have a person that is convinced that this is the right approach, but they need to make sure that they will be able to follow it despite all the pressures that might come. I think that that could be a possible question.
- [75:25]
A
I love that. That's a fantastic question because yeah, he sort of does explain it, doesn't he? Just in how Spitznagels are extremely confident, born to be traitor and doesn't really give a what anyone else thinks. And. But still, I mean to, to face up every single day with only losses to show for yourself. Yeah. What principles does it take to, to maintain that? And then now they look back and they're laughing. You know, I think they made $1.5 billion or something their, their hedge fund over the last two, three years. So Luca, these are three questions I try to ask every single guest. The first of which though, I'm really excited to ask you about because it lies hand in hand in the same bed as ergodicity. And that is the role of serendipity. So I wanted to ask you about the role that serendipity has played in your Life. And then as well, if you could broadly reflect on the role that serendipity, or at least what the relationship between serendipity and ergodicity is.
- [76:35]
B
Well, I think that serendipity is extremely important, especially for some professions. There are some professions in which serendipity doesn't play an important role. There are some professions in which instead it plays an extremely important role. You don't know before you start exactly what has to be done. So you need to try a few things. You need to let space for serendipity and the link to egodicity is that you need to have strategies that allow the space for serendipity, that allow the number of repetition that it takes for serendipity to bring its fruits, and you need to have strategies that allow for that.
- [77:20]
A
And in your own life, what role is serendipity played?
- [77:24]
B
Well, so in my own life, definitely, like, there are books, for example, that I definitely didn't like, plan on writing. Like, with like a very rational approach. Like, which book would work? Like, nothing like that. Like, it just happened to me, maybe that I write a tweet or two on the topic and then I got good response, and then that's an inspiration or something. Something. Something like that. Now the link with ergodicity is that you cannot rely on inspiration to come, like to strike a destination right moment. Even the people who say I write like inspiration strike every morning at 6:00am like, you know, you know, the anecdote, that's inspiration, like for good enough, that's the inspiration to write a good enough 20 pages. That's not the inspiration, like to write them. The one thing, for example, that will make it a bestseller, that might take more time. You cannot have it every day. And so you need to have a strategy that allows you to take this time. And for me, for example, it was the fact that when I quit my job and for different reason, I had some time to dedicate to writing books, I made sure that I actually had the time. So, for example, I structured my life in a way in which I would have very little expenses so that I could afford the time to write one book, two book, three books, until I wrote one that would really resonate and so on. So my advice would be allow space for serendipity, allow the number of repetitions, have a strategy that allows you to stay long enough in the game so that at some point serendipity will strike.
- [79:15]
A
And perhaps optimize for behaviors that also maximize serendipitous outcomes for Example, you publishing a book is but one of the.
- [79:25]
B
Most.
- [79:27]
A
Inviting, the most possible serendipity into your life.
- [79:30]
B
Oh yeah.
- [79:31]
A
Ideas out there. And you have no Idea who the 101st reader will be and what influence they might end up having on you. You know, maybe you, I don't know if you're married, but maybe you meet your partner, they're a fan of the book. You know, like pewdiepie met his wife through. She was a fan of his channel. It's serendipity right there. Every, I mean, every person's life is riddled with serendipity, whether they realize it or not. But it is. The most dramatic changes, of course, are not planned. They come through a sweep of randomness. And if you're lucky, it's because you've optimized forward and it's some sort of positive serendipity. Okay, look, Luca, if I'm honest, I mean, I could really just keep talking to you and talking to you. I think though, we should wrap up the podcast here and allow for maybe potentially new episodes down the line. But I'll just finish with these final two questions. The first being what is a country that you're particularly bullish on?
- [80:40]
B
So I am quite bullish on Singapore because they have a very good governance and they both the governance and the population, they tend to care about things that are needed for long term success more than other countries, more than other populations. So I'm bullish on them. I'm relatively, and I'm really, I would say that I'm relatively bullish on China, but I know China much less than Singapore. Like my wife is Singaporean, I spend some time there every year, so I know Singapore much better. China, I only know it second or third hand. So I would be bullish on them. I'm bullish on quite a bit of Southeast Asia, to be fair. In general, I'm quite bullish on some countries. In Eastern Europe, I'm seeing people of real talent, a focus on building what's needed for long term success in the country there. I'm much bearish on Western Europe, though. It's. Its potential remains high. But it really needs to get straight the politicians itself and it really needs to get politicians that start caring about what's matter for in the long term, which they definitely don't have right now.
- [82:17]
A
And yeah, finally, Luca, a conversation between any two people of history, dead or alive, no language barrier. So if you were to listen to a podcast, who are you listening to?
- [82:29]
B
So on one side there will definitely be nasimta lab. And on the other side, let's think about someone maybe that's not alive.
- [82:42]
A
Should Feyman.
- [82:45]
B
I'm thinking. So if it were for it, if it were for concrete if it were for concrete applications, I would love to have a conversation between Nassim Taleb and Lee Kuan Yew. Lee Kuan Yew was the former prime minister of, of Singapore. I think that they would. Yeah, I think that it would be a very constructive conversation.
- [83:12]
A
That is one of my most favorite answers ever. About 50 of the answers is Jesus and Buddha. So I think that would be Taleb, Lee Kuan Yu. Absolutely. That'd be fantastic. Luca, thank you so much for being generous with your time. And as well, thank you for writing that book. It's amazing.
- [83:31]
B
Thank you. Ryan, thank you for having me here.