Episode 38: Evolution, Complexity, and the Remaking of Economics

The Mutiny Team

The Mutiny Team

We have a special podcast for you today. We’re getting the Mutiny team together.

Taylor and I are doing a podcast together for the first time in a long time.

We discuss Complexity Economics, Sugarscapes, Adaptive Walks vs Random Jumps, Bullwhip Effects, Spherical Cow Problem and more! 

Links mentioned in this show: 

I hope you enjoy this conversation with Taylor as much as I did…

 

Listening options:

 

 

Have comments about the show, or ideas for things you’d like Taylor and Jason to discuss in future episodes? We’d love to hear from you at info@mutinyfund.com.

 

__CONFIG_leads_shortcode__{“id”:”63″}__CONFIG_leads_shortcode__

Transcript Episode 38:

 

Taylor Pearson:

Hello and welcome. This is the Mutiny Investing Podcast. This podcast features long-form conversations on topics relating to investing, markets, risk, volatility, and complex systems.

Disclaimer:

This podcast is provided for informational purposes only, and should not be relied upon as legal, business, investment, or tax advice. All opinions expressed by podcast participants are solely their own opinions and do not necessarily reflect the opinions of Mutiny Fund, their affiliates, or companies featured. Due to industry regulations, participants on this podcast are instructed to not make specific trade recommendations, nor reference best or potential profits. Listeners are reminded that managed features, commodity trading, ForEx trading, and other alternative investments are complex and carry a risk of substantial losses. As such, they’re not suitable for all investors, and you should not rely on any of the information as a substitute for the exercise of your own skill and judgment in making a decision on the appropriateness of such investments. Visit mutinyfund.com/disclaimer for more information.

Jason Buck:

We have a special podcast for you today. We’re getting the Mutiny team together. Taylor and I are doing a podcast together for the first time in a long time, but one of the ideas Taylor had was to maybe talk about the book Origin of Wealth by Eric Beinhocker. I believe it came out 2007-ish, I think, if I’m correct. I think I read it at the time, but I don’t remember it well, so functionally, let’s just assume I haven’t read it. It was great to go over your notes, and I think that gives us a jumping off point to start off with.

Jason Buck:

But starting with the Origin of Wealth, it comes out of the idea that classical economics is incorrect and that complexity economics is maybe a better way to look at the world. So, maybe let’s start with, what is complexity economics from a really simplified version? Is it evolutionary economics? Is it complexity economics? How do you think about what’s the difference between complexity economics and traditional economics?

Taylor Pearson:

Yes. A bunch of people recommended this book to me. Being good, I really enjoyed it. I think there’s a lot of really cool examples. I would say it’s for people that are familiar with the Santa Fe Institute, or people who aren’t, it’s a research institute. I want to say it was founded in the late ’80s or something, and it branched itself as the home of complexity, which this idea of this interdisciplinary field. They have physicists, biologists, economists, all these types of people. The general idea of complexity as a field of study is that when you have environments, or you have systems with many different parts and those different parts are interacting, so we can get into how this is in economics. But a lot of mainstream economics, neoclassical economics, traditional economics, whatever one call it, grew out of physics.

Taylor Pearson:

And if you look at Brownian motion, it’s the most common one I’ve heard of. Right? If you look at Brownian motion, so if you measure all the temperature of all the different particles in the room, if the room is 70 degrees, then some particles are going to be 71 degrees and some particles are going to be 69 degrees, but average that to 70, and those things interact randomly. One particle being higher doesn’t lead to… I’m probably explaining this wrong, but I think you get the idea.

Taylor Pearson:

Everything affecting everything else, whereas in a complex system you do, if one species in an ecosystem goes extinct, the mosquitoes go extinct, or I think the example I was given, I think in Borneo, they had like this mosquito extinction campaign or something, right? Lots of people were dying, I think like malaria, the mosquitoes carrying malaria. So they kill the mosquitoes and it turned out the mosquitoes were this critical part of the ecosystem. And it ended up having all these knock on effects. I think the rat population got out of control because there was some chain linking them to the mosquitoes. But this idea that we have these types of systems with many different variables, all of those are interacting, and how do we think about those, right?

Taylor Pearson:

So you can think about biology, ecosystems, economics. And so, the complexity economics is taking that idea and applying it to economics, instead of looking at something super simplistic, like a supply demand curve. It’s like, “Well, wait, what if there’s actually 42 different things that are contributing this or thousands of different things contributing this and how do we start to think of it?” And then, I think the other way we could talk about that people have heard us talk about, or ergodicity before, I’m sure this idea of ergodicity economics and complexity economics idea is very similar, but this idea of measuring how a system behaves over time and factoring the time element into it. So anyway, that’s my initial feel.

Jason Buck:

Yeah. We’ll definitely get to ergodicity. In another way, think about that, and we will explain it too, is in the sense of you have to factor in time and path dependency. I love the mosquitoes in Borneo. I hadn’t heard that one. As you know, I love a narrative, anecdote, or example. So you always have great ones that I usually haven’t heard of. So that’s another one that the fancy word for it is like atrophic cascades, right? Like whether it was in Yellowstone, I think it was where you get rid of wolves that affects the beavers that affects the elk. You have those cascade of consequences that we don’t realize.

Jason Buck:

But one of the things, I think from your notes, I want to touch on real quick is like, complexity economics may be really related to biology and evolution. And one of your notes was evolution is an algorithm for innovation searching the fitness landscape of a given system. This could be an ecosystem like biology evolution of the economy. The environment creates a design space and then selection, natural or otherwise, tests all the configuration that design space over time. So it’s like a iterative process just the way biology is. But that was one of the things that jumped out at me. But then the other one that jumped out at me that Eric Beinhocker always talks about is the idea that’s after the industrial revolution, you had this Cambrian explosion in SKU diversity, what the hell does skew diversity mean?

Taylor Pearson:

Yeah, right. The SKU is, I actually don’t even know what it stands for. Right? It stands for like an individual item, your kick, if you have an e-commerce score and you have 20 different products, you see 20 different SKUs. I forget what it stands for, honestly, which is embarrassing. Yeah, the way he talks about this idea of as technology progress specifically actually the industrial revolution, that you got this explosion complexity the number of SKUs. And kind of the way, if you look at diversity of species or the fossil record and stuff, you tend to have these big periods of expansion and contraction in terms of ecosystem diversity, like the number of species in ecosystem, that kind of thing. But yeah, I think that the notion, and we could talk about it more later, of a fitness landscape is really interesting, right?

Taylor Pearson:

In biology, there are certain traits which are more or less fit to a given environment the same way, like in a given economic setting, you could think about this, that a company that was really well adapted to the 1950s, and America may be very poorly adapted to the 2020s in China or whatever. So, I like that analogy a lot.

Taylor Pearson:

One of my favorite evolutionary facts, or I guess this is the working through that I’ve heard, I think the eye was like a classic example that intelligent design critics would give us, like how could the eye possibly have evolved because it’s so incredibly complex. And there’s a bunch of theories on how that evolved. But based on the structure of the eye, they think eyes evolved at least twice. So I think octopus eyes and human eyes have no crossover. They both started from some very small spot that had some very basic light sensitivity and evolved independently. But to say certain things are so useful and so helpful that even though the path to getting there so incredibly complex, it’s so pro-adaptive that if anything can ever get there, it becomes super useful. So those evolutionary analogies I think are really apt and really interesting.

Jason Buck:

So by the way, SKU is stock keeping unit.

Taylor Pearson:

Okay.

Jason Buck:

So it’s so funny how we don’t remember those things. We just use it and forget, or never even [inaudible 00:08:48].

Taylor Pearson:

I don’t think I ever knew that, to be honest. I think I just always called them SKUs.

Jason Buck:

It makes sense now though, right?

Taylor Pearson:

Yeah.

Jason Buck:

A stock keeping unit. That’s why you have SKU barcode. But when you’re talking about the eyes, if I want to break my brain though, is I start thinking about the complexity of the eye. And then I realize that it’s only even adapted to see things in 2D, right? And then, our brain forms the 3D image. And so then, that’s how it breaks my brain, trying to think about those kinds of things sometimes if I’m bored.

Jason Buck:

But so, you’re saying there has this evolutionary process to it. So traditional economics, or even neoclassical economics, always deals with how wealth is created and allocated, but it focuses more on the allocation side. And part of it is homo economicus is the idea of we’re all rational creatures and we’re just trying to maximize our individual goals, but that it doesn’t bring in the social aspect. And I think that’s what he’s trying to point out is the social aspect missing. And a lot of the ideas that come from classical economics come from like Léon Walras. By the way, do you know if it’s Walras or Walra? I’ve heard both.

Taylor Pearson:

I don’t know. I don’t know how to pronounce. I’ve only read it.

Jason Buck:

Yeah. But he’s the one that came up with maybe equilibrium dynamics, which is, you referenced earlier, as a form of almost like physics MB. So maybe, that’s part of the problem is if everything is operating on equilibrium, then you’re not really taking accounts the inputs or outputs, and that’s one of the falsifies is that Eric’s trying to make up for. And maybe a good example is this, is you had a example from Henry Ford, when he changed the wage to $5 a day, so that way, his actual workers could buy the cars and everybody thinks it was a great thing, but this is where maybe we get into the social aspects, and we start talking about feedback loops, and then eventually, we can transition to, I guess, talking more deeply about feedback loops.

Taylor Pearson:

Yeah. So Beinhocker, I mean, he makes his arbitrary distinction, but I think it’s a helpful arbitrary distinction of three forces, physical technologies, social technologies and business models. So, maybe that’s kind of the analogy or the metaphor he’s talking about. Those are the three forces that are all co-evolving with one another, and you get these feedback loops. Yeah. So the example I thought of, as I was reading is this Henry Ford example, right?

Taylor Pearson:

So this is the famous story. Henry Ford offered $5 a day of wage to anyone that would work for him. There was this great thing. What I heard about, the reason he did this because the turnover was so high because the work sucked so much. You know what I mean?

Jason Buck:

Yeah.

Taylor Pearson:

He’s paid people a ton because otherwise they quit. So, there was some of that. But then, he also had all these social stipulations on it, like it was available to married men, men under 22, or women who were supporting dependence. I don’t know if it was explicit or just culturally spoken, but there was really a post like buying debt. The only thing you should ever use installment plans for, would be like buying a horse or a car, but really discourage that sort of consumer spending. And then, they interviewed the families for drinking and savings [inaudible 00:11:57], so they were basically profiling people based on, were they drinking too much? Were they saving their money? Or whatever. And the type of people he was screening for, he wanted to mold the culture in a way that fit the automobile industry, who can afford to buy cars, right? These people that they count a lot of debt, they only get in their car. They sit, they’re diligently not saving. They’re not alcoholics. And then, you can draw that out.

Taylor Pearson:

Well, Sinclair Lewis, the book Babbitt, which I think is a 1920s, 1930s, that’s the classic fiction work of… I forget the guy. Whyte, he called it, [inaudible 00:12:37]. He was other for Fortune or a writer for Fortune. He had a book called the Organization Man, and Babbitt in the organization, they were for paying this profile of this, what you think of as like the classical American corporate worker, white dude wearing a suit, waking up at seven o’clock in the morning and saying, “Honey, I’m going to work.” And stopping by the country club on the way home. And that’s sort of her persona. Right? But that persona was intimately tied in with the sort of technological stuff that was going on. Right? Like you had a car and cars meant you could have suburbs because now, you could commute 10 miles to work every day and you could make the drive-in 15 minutes. And so, that was a viable commute.

Taylor Pearson:

So I think that’s just like when you think about automobiles in particular are probably a really good example of just, you look at cities that were designed before and after automobiles, and it’s a totally different city, right? The whole logic of how you structure your life changes when you have a car. And so, those are his primitive examples. Then as you said, you have this time feedback loop where more people start to get cars and that influences cities and that influences who buys cars, and all these things are coming back and forth, interacting with each other.

Jason Buck:

I’m going to try to take us down a tangent here, but maybe segue back to even talking more about this time aspect. I’m curious, when Henry Ford raises that wage so people can buy cars, I think about now when we talk about minimum wage issues, it’s like if we raise the minimum wage for a shorter period of time, that should be highly beneficial to minimum wage employees, right? Because their salaries increased, so they should be increasing their purchasing power. But you would think over time, that cascades through the system to where the goods and services, even minimum wage workers buy are going to go up commensurately with the rising in minimum wage because it’s almost like it’s circular in that nature of the prices at Walmart or Burger King or McDonald’s are likely to go up if you go to a $15 an hour wage.

Jason Buck:

And/or I guess, the other complexity part of that is then maybe the employers eventually figure out robotization to limit the amount of employees.

Taylor Pearson:

Right.

Jason Buck:

So then maybe they can keep the cost down through it. So those are the competing forces of raising minimum wage time. And then, the countervailing force of technology reducing the actual amount of employees. I’m just curious if you have any thoughts on that are minimum wage ideas because it’s hard to think that through because a lot of the classical economics are essentially consequential as philosophers and they think they can look at all those causal chains and that’s the part of complexity is it just gets so weirdly really quickly.

Taylor Pearson:

Yeah. The thing I would think about the minimum wage thing recently too, is actually a different thing, which is, what is the cultural externality, right? So it’s like, there’s all this research about people want things to be fair or whatever. So, in a city, I think Austin just raised its minimum wage $20 an hour or something. Yeah, right there. So you can make the whole economic line of argument that drives everything else, or it drives unemployment up because you can employ those people, blah, blah, blah.

Taylor Pearson:

But then is it culturally? So like the fact that everyone knows everyone is being paid more here that the culture is more cohesive and more productive and that enables things or something. But I think yeah, it’s those thinking about those consequences beyond just, let’s derive some super obvious unemployment thing.

Taylor Pearson:

Or going back, actually another example I heard recently that I thought was really cool in terms of wages and unemployment was, I guess when the USSR came into Serbia after World War II and occupied Serbia, a lot of the Serbians left and moved to Greece, fleeing the communist regimes and stuff and moved their whole families to Greece. And then, after the whole Greek debt crisis early 2010s, a lot of those people moved back to Serbia.

Taylor Pearson:

So it was one year, I want to say it was 2013 or something where 5% of the population of Serbia had just immigrated from Greece. So, it was all these people that were Serbian or Serbian ancestry had moved back from Greece. So everyone’s freaking out that like, “Oh, either wages are going to go down or unemployment’s going to go up or what’s going to happen.” And I think initially, there was some of that, but then I think they had looked at a year or two later, and wages had gone up, and unemployment had gotten down, and they were like, “Whoa, that’s so weird. What’s going on here?” And all the people that had been to Greece had learned new work skills that people in Serbia didn’t know.

Taylor Pearson:

And they came back to Serbia, and they started small businesses and sold these things to other countries and hired other Serbians to work for them. And so, they’d reduce unemployment and whatever body. So I think, you have to start thinking about it. I mean, if you’re just looking at it in simple terms, the way you typically look is more workers, ceteris paribus, either unemployment goes up or wages go down, that’s the neoclassical view.

Taylor Pearson:

And so, I think it’s like, just as we’re talking about this, you can even tell, it is all a bit finger in the air, right? The ideas you’re going to predict exactly what these outcomes are going to be. But I think it’s important both from a policy, and we can get sort of investing things to at least have some sense that it’s that complex, right? I think it gives you a bit of humility about making these decisions and all the different factors that go into them.

Jason Buck:

Because we keep hinting at it, let’s talk about time a little bit. And that’s really what separates complexity economics or ergodicity economics from classical or neoclassical economics, like you said, it’s like classical or neoclassical is ceteris paribus, let’s take a snapshot and supply-demand curves meet, but then they don’t really move through time in any dynamic way. So part of time is path dependencies. And so, maybe we could talk about time a little bit of the ideas of, there’s that classical one about a $100 bill on the street wouldn’t exist. Otherwise, somebody would’ve picked it up. But there’s time delays from the $100 bill on the street. And that’s what maybe classical economics doesn’t pick up on, that complexity economics tries to at least think through is like, what is the time dynamic between an arbitrage opportunity? Because I think, correct me if I’m wrong, classical economics, we think arbitrage happens right away, so there’s no time between an arbitrage, where in the real world, there’s always time between an arbitrage and arbitrage can last from a few seconds to a few decades sometimes.

Taylor Pearson:

Yeah. No, no, but we’re like caricaturing economics a little bit here like most economists understand of, like this is oversimplified [inaudible 00:19:23]. We are well aware that this is the caricature, but just to draw the comparison. Yeah. The classic $100 bill thing. Go ahead. What are you going to say?

Jason Buck:

My favorite actually, you had it in your notes, and I don’t know if this is from the book, or you knew this one, that is the spherical cow problem or that is the most erudite way I think of making fun of classical economics. Do you want to illuminate that or do you want me to? It’s pretty short one.

Taylor Pearson:

But yeah, the version of it, I noticed, a cow is like this, the shape is really hard to deal with mathematically. And so, it’s like, well, if you just assume the cow is a sphere, then managing the cow mathematically in your models is super easy. Is there more to it than that?

Jason Buck:

Yeah. The original, I think, story goes something like, a dairy farmer was having a problem, he wanted to increase production. So they brought a bunch of cross-disciplinary research experts out to his farm and it was led by a physicist. And the idea was then they were going to figure out how to help the farmer. So for two weeks, they go and crunch the numbers, they really work on it. And then they come back and the physicist goes to present and he is like, “Here’s the solution. I can figure out the exact modifications if we just assume that they’re all spherical cows in a vacuum.”

Taylor Pearson:

Right. I don’t know if that’s really true, but it’s a better telling of it. Right? But yeah, it’s a telling story, even if it’s not true.

Jason Buck:

So, part of time, and we start to then incorporate aspects of time, path dependency, then goes into emergence and everything. But I think the best example, and this one’s going to be maybe hard to talk about over audio versus this, it’s always easier to show these ideas, but you had this idea of the Sugarscape, and maybe if you talk about or try to define, or as best you can without using visual additives, and I think you can describe what the Sugarscape is and how the Sugarscape evolves over time.

Taylor Pearson:

Yeah. So this book made me think a little bit differently about entrepreneurship, that you could think of entrepreneurship as you’re arbitraging the state that the economy is in now relative to where it’s going to be. Right? So, the example I thought of, as I was reading was like PayPal, right? Yeah. And someone was going to create a payment processor for the internet, right? That was such an obvious thing, but the way in which it happened, and there’s the whole PayPal story. Musk had his X.com company, and the way it merged, and they started with eBay sellers, and all those sorts of things. Right? They were basically arbitraging the past and the future in a sense. Right?

Taylor Pearson:

So all of a sudden, you had this commerce, and the path dependency you took to get there was important. Right? So PayPal famously figured out eBay power sellers, that was their initial core market. Those were the people that were doing tons of payments volume on the internet. And they could leverage and get a foothold.

Jason Buck:

By the way, do you think that FDX is now the modern version of Musk’s X.com? So the idea of X.com originally is he wanted to be the bank of everything for the whole world and all digitally. And you would have all your asset classes in one portal. And it seems like with FDX, trying to get all of the approvals across all the different asset classes, it sounds like they’re trying to put them all together, which to me is like the fruition of X.com two decades later.

Taylor Pearson:

Yeah. I haven’t thought about it. That’s interesting. I mean, that guy SPF that could do this, wicked smart. So, I don’t know. It seems like he might be able to do it. We’ll see.

Jason Buck:

Yeah. So do you think you can describe the Sugarscape without charts and graphs?

Taylor Pearson:

Yeah. No, I think it’s a fairly straightforward thing. So I think one actually interesting thing about this whole complexity economics that goes back to what we were talking about with social technologies and physical technologies all interacting is like, as the world became more computationally complex, you needed computation to deal with it more effectively. Right? So, in a weird way, the complexity economic stuff makes the problem worse as it’s solving it, right? The world gets more complex, you need more computation to try and deal with it.

Taylor Pearson:

But one of the things they do that’s unique, if you take an economics textbook, if you take a complexity economics course, but there used to be all these simulations, right? You run these things over time where you say, “Okay. T=1, this is what’s going on. Then T=2, T=3.” And you can watch these dynamics play out over time, which is different. I took economics in college and there’s [inaudible 00:24:01], right? Like you’re solving for, “Okay. What is the supply-demand curve given these variables at this point in time?”

Taylor Pearson:

So Sugarscape is an experimented design that’s just a super simple version of this. And it basically starts, you can imagine a terrain with two big piles of sugar on opposite edges of the terrain. And then, you start and you take, say a hundred random agents. Agents is their word for people or actors. And you place them randomly throughout this environment. Right?

Taylor Pearson:

So some of them start in the middle of the densest part of the Sugarscape where there’s all this sugar, and they can eat to their heart’s content. And some of them start in the desert, where they’re far away and then curve in between in various places. And so, you start out with this random distribution and kind of the way the rules of the game work because if you’re close to the sugar, you get wealthier, right? You’re harvesting the sugar that’s the source of wealth or money in the experiment. And over time, you can either stay in the spot you are, or you can move one way or the other.

Taylor Pearson:

And so, at first, everyone’s randomly distributed and then people start moving, and they start moving towards the sugar. Right? Once you find a little bit of sugar, you keep moving in that direction and you find more sugar. Maybe you find more sugar, and eventually, you end up where the populations are clustered around the big mountains of sugar. People are trying to get as close to the center of the mountain of sugar where they can harvest the most sugar possible.

Taylor Pearson:

And so, you get some interesting emerging effects from this. One is when you start out, your wealth distribution is basically equal, right? Everyone just randomly got plopped down with whatever they had, but the path dependence managed a lot. So if you had two people that started next to each other on the board, they both move in a random direction. On the first turn, the first iteration, one finds, goes towards the sugar. One goes away from the sugar. The consequences of just finding the sugar one turn earlier where the other person might wander around for five or six turns, all the while you’re harvesting your sugar and multiplying whatever, ends up being super consequential.

Taylor Pearson:

So, from this random distribution that you start with of everyone with [inaudible 00:26:34], you end up with basically a power law distribution of wealth outcomes, where you have I think those simulation. The examples, there’s 300 different agents. And three agents end up with 40% of the total sugar wealth or whatever, right? It’s the kind of classic 80-20 power law things-

Jason Buck:

Greater distribution. Exactly, yeah.

Taylor Pearson:

… where the path dependence mattered a lot.

Jason Buck:

It’s…

Taylor Pearson:

And it’s like mattered a lot.

Jason Buck:

It’s actually scary when you think about it. You and I love this kind of stuff, but if you actually think about like, what is it, cumulative advantage, right? If you just start off on the lucky, and on a lucky time streak, and you’re able to cumulate more than somebody else? You could sustain that longer, so it’s a matter of just kind of lucky coin flips. Then we think about that wealth distribution that way, and, like you were saying, you related it back to entrepreneurship and whether it’s PayPal, et cetera, it starts to make you a little nervous to like, “Did I just get lucky, and I was able to amass some cumulate advantage?”

Jason Buck:

Maybe we’ll get into, how do you set yourself up to take advantage of that, through probing and testing and all, and trying lots of bets. Then we’ll even maybe talk about Farmers Fable. What do we do about that as a society later on? That Sugarscape example does throw it into a stark reality that you may just be getting lucky, and maybe you just want to try to open yourself up to as much luck as possible.

Taylor Pearson:

Yeah, it’s right. The thing that it reminded me of is if you run Kelly Criterion simulations, right? The Fortune’s Formula, which is another book we both like, has such a good example of this. But the idea of if you know the odds of a game you’re playing, the idea of the Kelly Criterion is the long wealth maximizing strategy, right? Over the long run, it is the thing that will compound wealth the best.

Taylor Pearson:

You can compare it to a bunch of those strategies, but the… I’m trying to… There’s some example or whatever… But it takes a really long time. The long run can be really long, right? And you think about this in investment performance, you have one person running the Kelly optimal strategy and one person doing something that is substantially less optimal. The Kelly optimal strategy can still underperform for like a decade, right?

Jason Buck:

Yeah, man.

Taylor Pearson:

The long run really matters and the long run can be super long.

Jason Buck:

Yeah. I think my [inaudible 00:28:54] for Kelly is you don’t normally see the out-performance until at least 500 iterations. Like you’re saying, so if, for example, somebody’s… Depending on their strategy, if we’re talking about investing, somebody that trades hundreds of thousands of options contracts a day, versus somebody who is like a value investor that may be only buying a few stocks a year? It’s impossible to know that, if throughout their lifetime of being a value investor, if they have less than a hundred iterations? That’s just luck, it’s impossible to really know.

Taylor Pearson:

Yeah.

Jason Buck:

Even cumulative advantage can win there in lucky coin flips.

Taylor Pearson:

Right. Yeah. I talked to private equity people as well, right? Like you’re doing more acquisitions a year, three acquisitions a year or something like that. There’s not enough at bats. There’s so few at bats, it’s really hard to separate luck from skill.

Jason Buck:

So then going from Sugarscape, [inaudible 00:29:42] provides a number of mental models for how complexity economics is different from normal economics. Those four are stocks and flows, behavior irregularities and bounded rationality. Three is system structure and bull effects and four is fitness landscapes. So let’s start with stocks and flows, and what the hell are stock and flows? What would be an example of a stock and a flow?

Taylor Pearson:

Yeah. My guess is a lot of people listening to this podcast are familiar with this one, but the idea is you look at a system, like a bathtub is the classic example. You have the water flowing into the bathtub, that’s the flow. You have the water sitting in the bathtub, that’s the stock, right? And so you can measure, over time, the stocks and flows. If the drain is open and it’s draining out one liter of water per minute, and this spigot is on, and two liters of water are coming in per minute, you know every minute the stock in the bathtub is going up by one liter, because you’re gaining two liters and losing one kind of thing.

Taylor Pearson:

This was kind of like John Maynard Keynes’ General Theory of Unemployment. He was the first one to propose that idea. That was his idea, right? That you have a drop in consumer confidence, that leads to decreased spending, that leads to decreased production, that leads to unemployment which leads to lower consumer confidence. Further drop in spending, right? And you get this feedback loop and that was kind of like his diagnosis of the great depression, right? You were stuck in this feedback loop that you needed some exogenous thing to get out of.

Jason Buck:

Part of that, too, is sometimes people forget it’s positive feedback loops will extend the trend where negative feedback loops will help mean reversion. Is a way to kind of think about it, right?

Taylor Pearson:

Right.

Jason Buck:

Right. Like if there’s an amplitude to positive feedback loops in either direction?

Taylor Pearson:

Yeah. So the negative feedback loop, my best example is the thermostat, right? If you have the temperature set to 70 in your house and it goes to 71, the air conditioning kicks on. If it goes to 69, the heater kicks on and it’s a negative feedback loop that it’s adjusting back to some equilibrium point. But I guess the complexity economics point is a lot of things don’t work that way, right? A lot of things have positive feedback loops, which is where you get these far out of equilibrium outcomes.

Jason Buck:

The way I always thought about it in, especially when you add those feedback loops and time delays of feedback loops, it’s like homeostatic is just a flat line, right? Homeostatic is you have a little bit of oscillations around the median, which would be like a thermostat, right? You’re trying to maintain 72 degrees, you actually might be 70 or 74 and then it just goes right back. That’s like a homeostatic bands.

Jason Buck:

And then when you have these big bands, those are called allostatic bands, basically. So that’s what you see with these feedback loops and dynamic systems, like an economy, is you have these, you’re going far and away from the median or mean. It goes back and crosses through there, but it’s actually never sticking close to there like a homostatic or homeostatic band would. But it’s interesting then, why do we even talk about averages then? I think this would be maybe Eric’s argument, if you have these huge allostatic swings, due to feedback loops and time references, is there ever actually an average, even though some people look at that chart and say, “This is the median or the mean.”, but is that really an average? I mean, because we are moving so far away from whatever that arbitrary line would be.

Taylor Pearson:

Yeah. There’s a good XKCD comic, that I won’t be able to describe here, too. You probably want to talk about it. It’s like it has four different data sets and the mean median and mode of all the data sets is identical, but they look just like completely different?

Jason Buck:

Yeah.

Taylor Pearson:

In fact, it’s like how you represent the data really matters and right, like averages can be deceiving.

Jason Buck:

And then, part of that, is bullwhip effects, but I want to hold off on that for a minute until we get to… Let’s go next to behavior irregularities and bounded rationality. For example, for behavior irregularities, what we call cognitive biases, what are some of your favorite examples of cognitive biases?

Taylor Pearson:

Yeah. So the example you give is the classic example that gets used by the complexity economist is the El Farol Bar Problem. El Farol Bar, it’s a bar in Santa Fe, New Mexico. I was in Santa Fe two weeks ago and I went to the El Farol Bar and took a picture which no one else seemed to get the joke.

Jason Buck:

So jelly…

Taylor Pearson:

I thought it was pretty cool. Yeah, so I had a margarita at the El Farol Bar a couple weeks ago. It’s delicious. The front porch is excellent. I highly recommended if you’re in Santa Fe, but the nature of the problem is, basically, there’s this cool bar and people want to go, there’s live music on Thursday nights, but people only want to go if there’s less than 60 other people there. Because if there’s more than 60 people, it gets too crowded and it’s not fun. They’ll be cramped. They won’t be able to hear the music. They can’t get good service, whatever.

Taylor Pearson:

So the question is how do they sort of decide? And if you look at those different ways and models, but you look at chart, it never settles on 60 people, right? It oscillates around a lot so if you went one week and there were 70 people, you’d be more inclined to think, “Well, the next week, I don’t want to go because there were 70 people last week.”, but then you go, “Well, if everyone else thinks that everyone else is not going to go…”, right? So you end up with this like complex modeling point. You get something that looks kind of, as you said, like allostatic bands where it’s 60-ish people are going, but there’s no real discernible pattern, right? It doesn’t reach some equilibrium point at 60. It’s oscillating around this equilibrium in a way that looks… It goes 59, 61, 52, 78, up and down, et cetera.

Jason Buck:

The part about that… That’s the bounded rationality problem, we don’t know what we don’t know, but it almost has a negative feedback loop to it in a way, right? Because it self-corrects, like the more people are there one week, the less people go the next week and then less people, then the more people will show up because you’re just going on your almost… Speaking of biases, your recency bias of what the last two weeks have kind of looked like and that kind of brings it back to that average, right?

Taylor Pearson:

Right, and I guess it’s kind of like a classic Keynesian beauty contest thing, right?

Jason Buck:

Right.

Taylor Pearson:

You’re predicting what everyone else is. You know, you could just, if you can regress, infinitely, right? Like I’m going to predict what everyone else is going to predict that everyone else is going to predict that everyone else is going to predict-

Jason Buck:

[crosstalk 00:36:05] Right.

Taylor Pearson:

… And, yeah, you end up in this like infinite beauty pageant kind of thing.

Jason Buck:

Part of it is that it creates a random volatility, basically, right? And then D, I’m curious though, on the behavioral side is usually the cognitive biases that we all have, but, here I’m going to throw kind of a curve ball at you. Usually people on the side of talking about the… I think, what is it? Munger says that he’d list at least 30 plus, maybe, cognitive biases that we have or behavioral biases so if you’re in the Charlie Munger or Ben Franklin camp, they think “These are the list of cognitive biases and this is how I avoid them.”

Jason Buck:

But then, if I think about Danny Kahneman, who spent, he says… Spent his whole life working on cognitive biases, but he says he’s just as fallible for every cognitive bias as everybody else. So my question for you, the curve ball, is if all we are is an amalgamation of all these dozens of cognitive biases, is anybody not cognitively biased? What do we kind of talk about? Like all we are is fallible. Like there’s nothing actually we do what we do. We do well at reasoning from rationality, especially because of bounded rationality.

Taylor Pearson:

Yeah. I’m trying to remember when Thinking, Fast and Slow came out, the day I kind of like… I was very taken with all that when it came out and I think I’ve gone almost completely 180. Like I think almost all the behavioral economics, the rationality stuff, is BS, and there’s actually a really interesting experiment. I think it’s called the Copenhagen Experiment. If you Google it, we can try and find the link and put it in the show notes. But it was done in Copenhagen, as you’d expect given the name, and this gets back to our [inaudible 00:37:40] thing. But basically if you expect people… People intuitively understand log wealth better than expected value, and so you can have all these scenarios in which people are making what looks like a suboptimal expected value decision, but it’s, in fact, a log wealth optimizing position.

Taylor Pearson:

And, once you account for that, a lot… I think there’s a lot of these things where it’s just like the way in which rationale… The problem is the way in which rationality is defined, is bad or not helpful? Clearly, some of these are legitimate so that makes some of these do make sense and probably will hold up, but I feel like a lot of this stuff probably goes away in the next 20 years, but, in any case, as it relates to economics and bounded rationality, you’re just acknowledging… That’s not really behavioral bounded rationality, to me, it’s just people don’t know everything, right? It’s like you have this idea that the person is standing in the grocery store about to buy peanut butter and they’re computing their expectations for inflation, relative to the storage cost of the peanut butter in their apartment, you know what I mean? Like no-one’s doing that, right?

Jason Buck:

Yeah [crosstalk 00:38:55].

Taylor Pearson:

[crosstalk 00:38:55] You’re either in the mood for the peanut butter or you’re not and keep going with your life.

Jason Buck:

Man, I was actually hoping you’d push back more on this so we get more of an argument, but I figure you’re right. I think if we would’ve talked about this more than like five years ago, you’d have a very different opinion about the behavioral economic side and cognitive biases and the reason to study those or always have those on hand to make sure. It is interesting, though, I think, that if you are in a group setting, like we do, even at our firm, and, even while we’re making decisions as a group, we do bring up cognitive biases just to double check ourselves. That’s as best you can do, but we’re just fallible. At the end of the day, we don’t know what we don’t know.

Taylor Pearson:

Yeah. I think I can see… But recency biased return chasing like that?

Jason Buck:

[crosstalk 00:39:33] Right.

Taylor Pearson:

[crosstalk 00:39:33] That probably exists, right?

Jason Buck:

[crosstalk 00:39:35] It does, yeah.

Taylor Pearson:

Like you can see that in DAG, if you look at investor flows and I think Vanguards done a bunch of studies on this, right? But people chase returns because it did well recently and they don’t like things that’ve underperformed for long periods of time. So you get these flows in and out and that kind of stuff so I think a lot of that stuff is valid, but it probably won’t hold up as well as I thought it would 10 years ago.

Jason Buck:

All right. You’re getting in now to one of my favorite things is the bullwhip effect and we’re both huge fans of the Beergame. While I’ll make sure we put that in the show notes, so people can find how they play the Beergame, maybe you can describe the bullwhip effect and how it affects supply chains, especially with what we’ve been going through since the Pandy… This is such a great thing to consider and think about is bullwhip effects and how that affects both upstream and downstream on supply chains.

Taylor Pearson:

I didn’t know we were calling the pandemic the Pandy at this point, but I’m going to [crosstalk 00:40:27].

Jason Buck:

[crosstalk 00:40:27] I was trying not to say that because you can get a strike from social media where they shadow ban your stuff for talking about the C word or the P word. So I’m just trying to [crosstalk 00:40:37].

Taylor Pearson:

[crosstalk 00:40:37] I didn’t know all the algorithms were up onto that. I guess we’ll call it the Pandy then. Yeah, I think, and this is-

Jason Buck:

Pandy wandy.

Taylor Pearson:

We mentioned this earlier if you haven’t ever played with this before, if you Google the Beergame simulators, there’re a bunch of free ones online and it takes five or 10 minutes to like play with it? But playing the simulation is actually way more instructive than trying to listen to me, describe it. But the basic idea is you have a supply chain, they use the example of beer. So at the top of the supply chain, you have the plant, then you have the distributor, then you have the retailer and then you have the customer. So you have feedback loops and time delays in between this thing, right? So the retailer places an order with the distribution center once a week, the distribution center places an order with the plant supplier once a week and then the plant supplier will fill out the next week to a distributor and then the next week that case would be or whatever goes to the retailer.

Taylor Pearson:

So if you have a fairly… Let’s see, you know you’re selling 10 cases of beer every week as the retailer, you order 10 cases from the distributor, the distributors 10 cases from the supplier, the same thing happened three weeks ago. Everything’s very smooth, right? But as soon as you start getting like some perturbans. That’s what the idea of the bullwhip effect is. It is basically it like ricochets or it reverberates up and down the change of you’re used to selling 10 cases of beer in a week and you have demand for 20, like, “Oh, no.” Like “I’m the retailer. I could have sold all this beer. I didn’t have enough inventory in stock.” Instead of “I’m going to place, not just the order for 20… Because there might be more sales, I’m going to place for 30 because I got to fulfill the 20 I already sold and I want to have some extra on stock in case it comes.”, right? Then you go to the distributor and it’s 30 and he goes to the plant and orders 30 and it takes however long to get back down to you.

Taylor Pearson:

But then, things get low and now you’re only selling two cases of beer. Now you have 30 cases of beer, and you’re only selling two a week, like “I’m not going to order anything else at all.”, right? So you can end up in this bullwhip where you get this reverberation up and down the supply chain and, of course, the Pandy as we’re calling, was like, right. That’s a lot of what we’re seeing in supply chains all over the place right now. I think we’re recording this end of June.

Taylor Pearson:

Now I think we’re hitting some… There’s some supply guts. I think I saw, summer, some of the consumer discretionary stuff got massively over purchased, there’s like a big supply go of consumer discretionary. And so Williams and Sonoma is running their biggest sale ever because-

Jason Buck:

Yeah.

Taylor Pearson:

… You know… Right. They couldn’t keep up with demand in 2021. [inaudible 00:43:29] their shares through the roof and it dropped off a cliff, but they’d made the orders nine months ago, for the stuff that they have in inventory now. So now they’re having to run massive sales and like try and liquid everything, because they can’t afford to hold all this inventory. Six months, you go there and the exact opposite position, everything’s out of stock. You couldn’t order dinner plates or whatever because everything was backed up.

Jason Buck:

Yeah. So part of that, too… Obviously the supply chain worked for you getting dog toys so if you guys hear that in background on Taylor’s mic-

Taylor Pearson:

That’s how-

Jason Buck:

We’re keeping the dog entertained while we’re on a podcast.

Taylor Pearson:

We have a five month old puppy, which my wife is watching, but yes, he’s a noisy fellow.

Jason Buck:

But what I loved about the Beergame is, as you went through so eloquently, is that’s a simplified version, right?

Taylor Pearson:

Right.

Jason Buck:

Of a one complex system, if we’re going from plant all the way to retailer and just the beer through the distributors and the wholesalers and you just see, when you play the game, how far out of whack it can get and how difficult these things are. Then you’re not even taking account the other complex systems of providing all the supplies to the plant, like the hops farmers and everything and how that can cascade through their system-

Taylor Pearson:

Right.

Jason Buck:

… Or even the transportation system between those systems and then the end consumer, how are they operating and what’s their income like and everything.

Jason Buck:

So it’s like you have all these integrated complex systems that make this bullwhip effect so hard and that’s why, like we said, when this thing started and you shut down the world? I was like, “This is going to be the biggest bullwhip effect we’ve ever seen.”, right. And then it’s likely to last years of those reverberations, because like you said, we didn’t know when China was come back online. Then the retailers over-order, now they’re stuck with all this inventory and then China’s shutting back down again, are they going to over-order? How are they timing that with seasonal?

Taylor Pearson:

Yeah.

Jason Buck:

Especially around Christmas time… I don’t envy anybody in any of these positions and it’s always been shocking to me when people are just vociferously angry about these bullwhip effects to the supply chain. Everybody’s pointing out all the ships in the LA port and all this stuff and I’m like, “What did you think was going to happen when you just press pause on the world economy? When that restarts, those bullwhip effects are just going to be massive.” It’s kind of shocking to me and maybe we can talk about the ideas around maybe inflation and everybody got so mad about transitory. But, to me, what’s the interesting thing about that is I got to learn how what everybody thinks transitory is. Some people thought it was a week. Some people think it’s five years and everybody’s in between.

Taylor Pearson:

Yeah.

Jason Buck:

So when you use a subjective word about transitory and you have all these bullwhip effects, I’m like, “This is like an impossible problem.” Yeah.

Taylor Pearson:

Yeah. That’s the air conditioning tagline. On a long enough timeline, everything’s transitory. Something like that. Yeah. No, I remember, a friend of mine told me this story about he was meeting with a guy that worked in Apple’s supply chain. This was like pre pandemic. This was like 2017 or 2018.

Jason Buck:

Yeah.

Taylor Pearson:

And so these big companies, they have these ERP, enterprise resource planning, right? Like the iPhone, compared to beer, with just all the inputs? You got to have the copper from Chile and the micro chips from Taiwan, you know? This incredible sort of thing and he was coming to the stage where he was like, “Oh, this is going to be amazing. Like it’s apple, it’s one of the largest companies in the world. It’s going to be this sophisticated thing.” and the guy was like, “Yeah, it’s basically like people sending spreadsheets back and forth and screaming at each other on the phone to hit deadlines and stuff.”

Taylor Pearson:

Because it’s just like two cup life. You can’t put this nice, simple model of “All these things arrive at this time” and I’m sure they have techniques and software or whatever, but it was just I thought that was really interesting because if you think of anyone had figured it out, right? At the time, Apple and it’s still one of the largest companies in the world, if not the largest company in the world and it was just like a mess, you know what I mean? The whole supply chain was just like hanging together by threads, in order to get the iPhones made.

Jason Buck:

Well, that’s why I love three things… We have a mutual love for Flexport and everything they’re trying to do. It’s just fascinating. Them trying to bring that whole world online instead of like physical paper and actually be able to track your shipment as it goes across the ocean that you’ve never been able to do before is so fascinating to me. The second one is, and now actually, my mind’s kind of going completely blank. Oh, there was one… I think you sent me the video years ago. The idea of if you could put a whole, one specific supply chain, on a blockchain, so you could almost kind of limit at least capital side or the ordering side of some of those effects and I think they use coffee as an example. From the farmer, all the way to the end user, buying that cup. Through the roasters, through the distributors, through the wholesalers is, at least, by having everybody on that same blockchain and then when that actual cup’s paid for, the money actually cascades through. Because it’s not only the supply chain that add bullwhip effects, it’s the capital base too. So if the end, like Starbucks, has a cost to capital of 1%, but the farmer has a cost to capital of 50%, like on their like loan shark loans in the Dominican Republic? That was the most interesting thing I ever saw in blockchain, was actually putting that supply chain on a blockchain. I’m not sure if it solves the problem, but it’s a much more interesting way of looking at the problem.

Taylor Pearson:

Yeah. I think it was a guy I went to in New York, like 2017, 2018 and he’s just got a interesting point around the global financial crisis, like how much of it was a bad loans issue and how much of it was a transparency issue. A lot of the issue was no one knew if these loans were worth anything, right? Because it was a thousand page paper document that all these lawyers had put together with the transfers of all these mortgages and stuff and so everyone was unsure if their counterparty was good, like how much collateral their counterparty had. Like the DeFi stuff in crypto’s industry because it’s crazy leverage and it’s a casino and you know, all that stuff and whatever, but it’s like a transparent casino, right?

Taylor Pearson:

You can measure exactly how much leverage this protocol has over here and what their exposures are and all that kind of stuff. So yeah, I think that’s super interesting, like just not even making it, initially, more efficient or something, but just if you have that level of transparency, whereas they get you a hink. I remember that article that you’re talking about. I think it’s like once you get off chain assets, it’s like how do you get your end of your Oracle problem?

Jason Buck:

Right.

Taylor Pearson:

Like how do we agree that there, you have 10 tons of steel in your warehouse, you can put that on the blockchain, but, right, what’s the Oracle by which we can assess that kind of thing? It gets really messy there, but I think something is going to come out of that, right? Like there’s a there there, and it’s not going to solve everything everywhere all at once, but I think that’s super cool.

Jason Buck:

Well, and I know the one thing that bothers both of us is how the large corporations can put pressure on everybody else through the chain, speaking of time and [inaudible 00:50:24] through to like net terms, right? If they can go net 90, they can take advantage of their cost to capital and then they’re pushing all those costs onto the smaller players and every little guy kind of always gets screwed in those scenarios in a way. So that’s one way of looking at it. I remember the third thing was the SA iPencil and I’ll make sure I’ll put it in the show notes.

Taylor Pearson:

Yeah.

Jason Buck:

When you were talking about the Apple supply chain, that one’s just great. It’s like all of the inputs that go into making a pencil and nobody has actual tacit knowledge how to make a pencil, but everything around the world that has to come together to make it something as simple as a pencil.

Jason Buck:

I think I’ve told you this one, maybe not, but speaking of back burner ideas for us long term, is like, I would always love to create a documentary that was about iPencil or the way I thought about it was, maybe play on words, but like iPhone or I iPhone is-

Taylor Pearson:

Oh, that’s good.

Jason Buck:

… How does an actual iPhone get made, like through all the supply chain, but maybe shot on smartphone? So maybe it’s called I smartphone so it’s not Apple specific, but you have footage of people digging out coal tan by their hands and like in the Congo to even the Norwegian ship capita and having coffee from Brazil as they’re moving the raw assets across the planet and how all of these things come together. I think it comes from a GK Chesterton quote. It was something like “We don’t lack of wonder, we lack of want for wonders.”? It’s, basically, people are just not thinking about the things we use on a daily basis and the absolute magic that has to come together for these things to exist.

Taylor Pearson:

Yeah. That is a great… Yes. In an unpublished book, I have sitting somewhere in my hard drive, I have like a… I think I tried to make a funny version of iPencil. I think I used some… I tried to remember think I tried to use some like clever quirky item or something, but to just tell the same principle, right? Like you got yeah, in order to get the steel to, or whatever. I can’t remember… To get the wood to make the thing, the foreman at the dock that shipped the wood had to have his cup of coffee before he got on the boat or whatever.

Jason Buck:

Okay. Moving on to the fourth thing with complex economics is fitness landscapes. When I was reading your notes for the beginning part of fitness landscapes, I was having a little hard time following based on those notes but when you use the example of Derek Sivers and CD Baby, that helped me like understand the fitness landscapes a little bit better. Maybe you can tell Derek’s kind of story of how CD Baby started and then we can kind of pick off, from there, all the different pieces to that.

Taylor Pearson:

Yeah. A fitness landscape, you can imagine just like a… Like a topographical map is kind of what it looks like, right? You have hills and valleys and rivers and whatever and the idea… I think the landscape thing is like… I don’t know… Grizzly bears can only live in part of the fitness landscape where there’s like lots of salmon and it’s a certain temperature and it’s whatever… And, this kind of bacteria can live in this other sort of, where it’s hot and, whatever. The bacterias that live in volcano, blah, blah, blah. And so, applying this to the business world, the example I thought of was Derek Sivers.

Taylor Pearson:

So Derek had a company called CD Baby. He started it in the late 90s. He was a musician at the time and he thought it would be cool. I think he had a little bit of a technical background and he was like, “Oh, this internet thing is cool. I wonder if I could sell my band’s CDs on the internet?”, right? “I’m going to set up this internet thing.”, and I think it took him like six months to figure out how to setup the payment processing and build a website. Like everything that takes 10 minutes now?

Jason Buck:

Yeah.

Taylor Pearson:

To upload to your Etsy store or whatever. It took him hundreds of hours over like six months but he figured it out, right? He got something set up and he could post his CDs there and people could click “Buy now” and pay for the CD and they would enter their address and ship it off to them and.

Taylor Pearson:

Now and pay for the CD, and it would go to their address and ship it off to them, and do the whole thing. And so, he started doing this, and other musicians he knew. Like, “Hey, this is cool. Could you add my CD to the website, so my fans can go and buy it too?” And it kind of spiraled, and eventually I think the company got acquired by Amazon, in the early or mid 2000s or something.

Taylor Pearson:

But the idea of the fitness landscape is… And Amazon is maybe the more classic example, was because you had… Kind of the big shift with the internet was zero marginal cost, right? The difference between Amazon and Walmart is Amazon carries a million SKUs, which I now know what SKU stands for, and Walmart carries 20,000 SKUs or whatever.

Taylor Pearson:

And of the 980,000 SKUs that Amazon carries way more than a million SKUs. I don’t know how many they carry, but of the massive amount of SKUs, they’re only selling a few of each one, but the marginal cost to create a different page on your website is functionally zero. Whereas Walmart has real marginal costs. They have real estate costs, right? If you want to carry more items, or more inventory of a specific item, you got to build a bigger store. And then, how big of a store do you build? People have to be able to walk across it. Blah, blah, blah, blah, blah.

Taylor Pearson:

So, you had these big changes. So, if you wanted to sell your CD in a record store, it was going to cost you $3,000 to get all this stuff set up, and sales rates, and blah, blah. I think and CD Baby was charging $30 bucks, because what did it cost them? They just had to put up a page on their website. It wasn’t super material. They could pay you out right away. Right? The payment process, you clear the payment. It went to CD Baby’s bank account. You got paid in a week, as opposed to, it used to take six, nine months.

Taylor Pearson:

I think if you sold on CD Baby, you got to own the customer’s contact information, which is a huge thing, right? Someone’s buying your band’s CDs. You want to be able to tell that person when you’re going to play in their town, or if you have a new CD out, or whatever.

Taylor Pearson:

So, you had this situation where, going back to the sort of physical technology, social technology business model is… You had this new, physical technology, the internet. That changed the fitness landscape, where all of a sudden… CD Baby doesn’t work pre-internet, right? Because it’s selling a bunch of relatively obscure CDs.

Taylor Pearson:

Actually, I’d interviewed Derek at one point. The example he gave me was he had someone that sold songs for sailors. So, there was this woman. She lived on her sailboat, and she loves sailing, and she would write these like sea shanties about sailing or whatever. And she sold 10,000 CDs a year.

Taylor Pearson:

And so, there’s no physical location. What physical location are you going to go to that 10,000 people that are into sailing would go to and buy a CD in? It doesn’t make sense, right? They’re distributed all over the world.

Taylor Pearson:

But it totally worked on CD Baby, right? Because across the entire world, there’s 10,000 people that every year want to buy a new CD about sea shanties, because they want to listen to it on their boating holiday, or whatever. And so, you start to get this feedback loop of that for a period of time made independent musicians a lot more viable in a way they weren’t before.

Taylor Pearson:

And then I don’t know the history of the music industry that well, right? But when streaming came around, right? And that’s changed things a lot more. That drove a lot of the revenue down. I think it went much more in live events.

Taylor Pearson:

But this idea of, you have these different physical technologies, the internet. Streaming was a physical technology, because bandwidth got high enough. The bandwidth good enough that you could stream music. You couldn’t stream music in the late ’90s, because the internet connection wasn’t stable enough. It was always cutting in and out. So, what did you do? You had to download the MP3, or you better get the CD, or then you had to download the MP3 and put it on your iPod, and that whole thing.

Taylor Pearson:

So, as you have these physical technologies changing, you also have these business models changing. Right? So, post streaming, the business model for musicians came way more about live shows. That was the scarce thing. That was where they were making their money, and CD Baby, it became some proliferation, these sort of smaller bands that could sell things online.

Taylor Pearson:

[inaudible 00:58:42] And you can apply this… The idea is that you can kind of apply that to everything, right? You have this interacting effect of… We talked about the automobile earlier, of the physical technology that changes the fitness landscape, right? Internet came along. All of a sudden this thing with zero marginal cost as possible. That has all these downstream effects. [inaudible 00:59:02] And I guess it’s interesting now. We’re talking about everyone sort of hates social media now, right? You have these knock on cultural effects and political effects, and all these sorts of things down the line. So, I really liked that. Ever since I read that, I’ve thought about that metaphor a lot.

Jason Buck:

No, it just made me think about how even the topography of the fitness landscape can change with the technology of like… It can take us, like when you’re talking about music, from a Gaussian bell curve, where you have your normal standard deviations, to when the internet comes around, and streaming music, et cetera. Then you change to a leptokurtic kind of curve, which is basically higher heads and fatter tails.

Jason Buck:

So, it was kind of winner take all effects in the middle, and then you have this long tail distribution. Like you’re saying, you can write sea shanties and make a living off of it. And how, when those technologies come along, you’re basically changing those curves around, and that’s really almost changing the topography of the fitness landscape.

Jason Buck:

And I want to talk about then how we navigate the fitness landscapes. How do we… What’s the pragmatic use case for that? And this is also, by the way, why, as you know, I hate DCFs. Because if a technology comes along and the fitness landscape changes, that’s not in your DCF model. Even if you’re in an innovative company, you’re not factoring in to the time value of innovation, because you don’t know what that innovation’s going to be, coming next year, or how it’s going to change.

Jason Buck:

And we’re talking about automobiles and Henry Ford. I think I’ve told you this one before, it’s like, most people don’t know Kingsford charcoal comes from… The Ford in Kingsford comes from Henry Ford, because when they’re making model tees, they were using a lot of wood to make model tees. And they figured out they could take that wood and turn it into charcoal briquettes. And so that waste product initially became a whole line of business for them. And that would’ve never been in your DCF model, and that changes the fitness landscape.

Jason Buck:

So, when we’re talking about… Part of that was a random jump, so to speak. So, let’s talk about their… This is really getting us into the weeds, and the really good part of like, how do we navigate a fitness landscape? And you use two models of adaptive walks and random jumps. And this one’s, I think, going to be hard for you to explain without visual aids, but try to explain to me, what is an adaptive walk, and then what’s a random jump?

Taylor Pearson:

Yeah. So that… The other thing I thought of with the fitness landscapes. This is the only podcast I can say this on, is looking at option surfaces is kind of an interesting analogy, right? If you’ve seen a visual of an option surface, right? You have… it looks like a fitness land… You have this sort undulating, shifting landscape, that over time, right, you know you have, ball moves more at the money, or out of the money, or this theta moves here, there. There’s sort of like, the time stuff.

Taylor Pearson:

So, I’m trying to think of good ways to describe fitness landscape, and that’s what came to mind for me. But it only works for four people that have looked at ball surfaces.

Jason Buck:

Yeah. Looking at your visual aids and your notes on fitness landscapes, that’s all I was thinking about, was ball surfaces, and Mandelbrotian fractals, and how he created the CGI stuff to be able to create mountain scapes. It’s very similar to fitness landscapes, right?

Taylor Pearson:

Yeah. But so, to your point about the adaptive walk, random jumps, the adaptive walk is basically, you’re standing at some point on the fitness landscape. It’s in sort of the fitness landscape idea, the idea is the taller you are, the more fit it is. Right? So, you want to be on the tallest mountain top, because that’s the most fit place, in business terms. That’s the company that’s the most defensible, with the most future earnings, blah, blah, blah, kind of thing.

Taylor Pearson:

And so, one way you can approach navigating the fitness landscape is you just always go up, right? So, you’re standing at one part, and you look around, and you see what part of the fitness landscape is higher, and you walk up that part. And then you get there, and you look around, and you see what part of the fitness landscape is higher. And you walk up that part.

Taylor Pearson:

The problem with that, of course, is you tend to get constrained by local maxima, right? So, I think there’s a computer science problem. I think it’s called the hill climbing problem. But it’s dealing with algorithms, trying to solve this problem. It’s like, if you just keep going up, whatever hill you… If there’s 1000 mountains in this fitness landscape, once you get to the top of whatever mountain you happen to start closest to, it’s never in your interest in the next turn to descend down the mountain. Right?

Taylor Pearson:

And so, there’s career analogies, right? You went to an Ivy League school, and you got a job in investment banking, and you hate it, and you want to do something different or whatever. But you’re at the top of a mountain, and the only way to get to the other mountain is you have to go down and do some crappy stuff. It looks way worse than what you’re doing right now.

Taylor Pearson:

The random jump is you imagine you have a giant pogo stick, and you can’t see so much around you. Right? Let’s say, your fog of war. There’s fog around you, but you can jump on this pogo stick, and it’ll take you somewhere. So, it could take you… You could jump off a local maxima to a new hill that could be even higher, even better. Or, you could jump off and end up in the valley of death. You’re on a hill, and you jump off and you end up in…

Jason Buck:

Floor is lava.

Taylor Pearson:

… the quick sand. Yeah. Floor is lava. But the general idea is you want some combination of random jumps and adaptive walks. And I thought of this while I was reading Clayton Christensen, disruption theory. His idea of, he calls it incremental innovation and radical innovation. I forget the terms he uses. But his idea is, he looked specifically at the history of computing. You would have, it’s like, “Well, I’m going to make your computer a little faster, a little faster, a little faster.”

Taylor Pearson:

And then you’d have laptops came in, and at first laptops looked dumb, and they were like a toy, you couldn’t do anything important on a laptop. And it requires you redoing your whole supply chains, so all the big companies say, “This [inaudible 01:05:23] is stupid. Why are we’re going to pivot into this?”

Taylor Pearson:

And that allows room for a new entrant to come in. Right? So you end up having to take this big pogo stick jump from, wow, we’re doing really well. Now we have to go sort of back down to doing this lowly startup-y thing, in order to get to the next global maxima.

Jason Buck:

How do you think about… And I think you’re referencing it, and this is what we’re getting to, is that barbell approach, right? Or Pareto’s 80/20. And most of the time, like you said, you don’t want to be either pure adaptive walk, or pure random jump. You need a combination of both, but most of the time you probably want to be in random walk.

Jason Buck:

I also like about random walk is if you can’t go up, you’re going back down to where you started, and back around. There’s a lot of blind alleys and going off on tangents that don’t work, even under adaptive walk. But when you’re structuring a company, say like ours or something, most of the time you want to be doing a lot of these random walks. Sorry, adaptive walks. But then you want to have some of those moonshots with random jumps.

Jason Buck:

So, how do you think about the combination when you’re building a business, of sticking to kind of the walking path, and then having a few jumps here and there, and how do you combine the two?

Taylor Pearson:

Yeah. Taleb’s classic barbell thing is like, you spend 80%. His idea with investing was like, you put 80% of your money in super safe stuff, like short term T bills, and you put 20% in crazy, Moonshot, Crypto, Angel, 100 to 1 potential kind of things, and that actually makes more sense than having sort of all this stuff in the middle.

Taylor Pearson:

Yeah. I think I’ve always just… On a marketing budget, right? Spend 80% of your marketing budget on something that, it’s probably an incremental return, right? Maybe it’s a 20% return, but you have a high confidence it’s going to work. And then take 20% of your marketing budget, and go do a bunch of crazy, weird stuff.

Taylor Pearson:

And actually, there’s a good example in this book that I hadn’t heard before. It was about when Microsoft was developing Windows from MS-DOS. And I forget all the exact details of the story, but basically the point was it got portrayed in the media as this brilliant, innovative… This foresight that Bill Gates had to abandon MS-DOS, which at the time was a big thing, and move to this new system, Windows. What a risk he had taken.

Taylor Pearson:

And the actual story was basically like, they sat down. They were like, “There’s six different possible ways the industry could go. One is we stick with MS-DOS.” They figured out what the strategy was for each of the six different paths, and they started working on all of them. And over time, it became clear that the Windows was the one that was working the best, and it was getting the most traction. It was promising, and they doubled down on Windows, and that became the new OS, and all that kind of stuff.

Taylor Pearson:

But that checks out more with my real world experience. Rarely do I see, in practice, entrepreneurs take these crazy big bets where they roll all the dice. Sometimes it happens, but usually it’s much more like, “Okay. There’s four possible outcomes here. I’m going to kind of hedge, and do something good for each outcome, and see what gets the most traction, and then double down on that.” Right? And in retrospect it looks like, oh, what a brilliant thing. But really, it was some sort of barbell-y type strategy.

Jason Buck:

Well, yeah. And I like that one. It was like the option of these small bets. I think though, too, like you said, they got lauded in the end. But I think in the beginning, when they had all those bets on, they were getting roasted by the press, right? Like they didn’t have a concrete plan.

Taylor Pearson:

Yes.

Jason Buck:

And that was part of it too, was that… Go ahead.

Taylor Pearson:

No, someone shared a thing on Twitter. It was a Time magazine clipping or something from the early 2000s, and it was comparing Amazon and eBay. And it was like a stock analysis kind of thing, like, “Which of these is a better company?” And pluses and minuses, that eBay is good at this, and Amazon is bad at this. Whatever.

Taylor Pearson:

And one of the big ones was like, Amazon is unfocused, and eBay was focused. You know what I mean? So, it was a negative on Amazon. They were like, “Oh, that’s a down vote. They’re not focused. They don’t know what they’re doing.” Whereas eBay, eBay really knows what they’re doing. They’re laser focused or whatever. And obviously we know how that turned out in terms of outcomes there.

Jason Buck:

There was another good example you had in there I hadn’t seen before, is a lot of times, though, some businesses, as they become mature, become too sclerotic, right? They avoid any sense of chaos, even at the fringes, for these optionality bets. And so, they just keep doing the same old things, and they don’t innovate, and they just run themselves in the ground.

Jason Buck:

Your example was Blockbuster versus Netflix. And this is such a good one, I hadn’t really thought about or heard, is like, Blockbuster at the time was making 50% of its revenue from late fees. And then when Netflix showed up and didn’t charge late fees, Blockbuster was too scared to imitate them, because it would’ve tanked their revenue. That’s so interesting to think about, that it was more about the late fees than anything else.

Taylor Pearson:

Yeah. I heard that example. It’s not mine. I got it from somewhere. I can’t remember now. But the idea is, sometimes you have that, where the… And again, this comes back to just the physical technology, business model, social technology all interacting. Right? As you had… Blockbuster’s business model was so predicated on late fees that… I mean, can you imagine Kevin being in that board meeting or whatever? Like, “You got to give up 50% of your revenue to make this shift or whatever.”

Taylor Pearson:

And I think that the Andy Grove Intel story is… The Andy Grove’s Intel, and I forget the exact details, but basically made this massive pivot. Laid off a third of the company, changed business lines or whatever. That story’s so famous, because it just never happens.

Taylor Pearson:

But that’s the level of… At a certain point, if you’re in a declining business, you have to make some big move, right? And I’m sure it was terrible for their profits for many quarters or whatever, but that was the right move for the long run.

Jason Buck:

And then you had this long section that we might actually have to do a separate podcast on, because I want to get to several other things. But one of the things you were talking about is how to use this kind of thinking for hierarchies within the business? What’s the right amount of hierarchies?

Jason Buck:

And so, we can just talk about it a little quickly, is like, one of the issues with one is the density of nodes in a business, and then two, increasing the predictability of decisions. And that’s kind of like what you were showing already with the Microsoft model, is increasing predictability of decisions through lots of maybe small bets. But then the density of node problems, there’re ideas with GM and the Amazon pizza pie kind of model of… How do you think about density of nodes, with everything we’ve talked about?

Taylor Pearson:

Yeah. There’s a clever term for it that’s escaping me now, but there’s an idea in computer science. If you have a software project that’s running late, the way to make it later is to add another engineer to the project. Right? Every engineer you add to the project makes the project even later, which is obviously tongue in cheek, but it’s funny because it’s like kind of true.

Taylor Pearson:

But unless you have some communication overhead, right? So it’s like, as soon as you go from one software engineer to two software engineers, you don’t double the productivity. Maybe it goes up by 50%, right? Because now the one person that was spending 100% of their time building the product is now spending 75% of their time building the product, and 25% of their time talking to the other person building the product to make sure they’re on the same page of how the things is doing.

Taylor Pearson:

So if you have descriptions, even like job descriptions or roles, right? Sometimes, in order to make a decision, you need to have competencies. Scott Adams was the one that kind of talked about this all the time, right? Like, being top 25% in three different things is sometimes more valuable than being top 1% in one thing, right?

Taylor Pearson:

So it’s like, if you understand a little about software engineering, and a little bit about product, and a little bit about marketing, there’s a certain role that’s maybe more useful for you, because you’re able to do enough of both those things to handle. And that’s kind of Amazon. There’s a really good… I’m sure you’ve read it, but if people haven’t read it… What was it called? It’s like, Steve Yegge’s platform rant. It’s on GitHub, but it was a guy, he worked at Amazon late ’90s, and then transitioned to Google. And it was a leaked internal email he was sending to like one of his colleagues at Google. Maybe he accidentally posted it publicly. I can’t remember what the story was.

Taylor Pearson:

But basically he was talking about… Comparing the two, right? He’d been very early on at Amazon. He’d seen it gone. He was like, “Amazon sucks. Everyone there’s an asshole. They’re not nice people. Blah, blah, blah.”

Taylor Pearson:

But he’s like, “They’ve got this one… Jeff Bezos sent out this memo in 2001, and basically said like, ‘No one talks to each other. You make it so that everyone can interface with your team via API, and if you don’t do that, I will fire you.’”.

Taylor Pearson:

And again, that’s the whole memo or whatever. And a lot of people sort of… And I think this makes sense, right? Amazon’s been able to scale for that reason, right? You have these fairly autonomous teams where you don’t rely on this super dense connection, and then that’s famously how AWS got started.

Taylor Pearson:

Right? Which is like, they basically built this thing internally, but the rules for how they built it internally was like, they couldn’t just build it so it was only useful internally. They had to build it so it had an API that other people could plug into, and sort of turned that into a business. All of a sudden, was a lot easier.

Taylor Pearson:

So, I thought that was a really cool example. And then the second thing you mentioned about increasing predictability decision making, to me, that’s like standard operating procedures. And that’s one of the big things you get about that. Right? It’s like, when I know that this process is getting done this way every time, it’s a lot easier for me. I’m removing one variable from the calculation I need to do about how a certain system is going to behave or something.

Taylor Pearson:

I think in that [inaudible 01:15:27], but also just general cultural stuff, right? If you work for a boss or someone, it’s like, seems really erratic, and it’s impossible to predict their decisions, it’s really hard to adjust to that. They’re like a loose cannon that could go off for any reason, anytime. Whereas if they’re pretty consistent, they kind of have the same MO, they’re doing the same things in the same ways, it’s a lot easier to scale.

Jason Buck:

Yeah. So, part of that, when we’re talking about business or investing, given all these things that we’re discussing on this podcast, is like, how do we prepare, not predict for the future? So, kind of knowing all of the things we’ve discussed is like, what is kind of preparation we could put in if we don’t know what the future’s going to hold?

Jason Buck:

And so, we’ve talked about the optionality of small bets. What we haven’t kind of talked about is tinkering. That’s almost like to me in the adaptive walk, or even in the random jumps. It’s like, by having all these little experiments, you never know what you’re going to discover that might lead to your next product. Almost like I was saying with like the Kingsford charcoal. As you’re going along, you can find things along the way.

Jason Buck:

But then the other one we haven’t talked about is redundancy via access capacity. You and I also talk about, it’s like slack in a system, and how important that is, especially if we have bounded rationality, which we talked about at the beginning, is like, you just never know what you don’t know. So, you always want a bit of slack in the system, so your system doesn’t collapse because you’re running it way too tight, because you’re trying to optimize every parameter.

Jason Buck:

So, I’m actually leading you into, actually, I know, which is one of your favorite examples in the world, about trying to optimize versus having some slack in the system. And that’s the German forestry service in the 1700s. So, I’ll let you talk about that example, because I know how much you love that one.

Taylor Pearson:

Yeah. I mean, it goes back to the idea of ergodicity, and like we’re talking about Kelly criterion. The long run optimal strategy isn’t necessarily short run optimal strategy. It’s like usually isn’t the short run optimal strategy.

Taylor Pearson:

But this is from a book called Seeing Like a State, which I think has gotten a lot more popular in recent years. But the guy… I think he’s a professor of early agriculture at Yale or something. He has this very sort of niche thing. But he has this great example in the book of German forestry.

Taylor Pearson:

And so, I think it’s late 18th century, late 17th, 18th or 19th century. Anyway. Sort of early modern German, I think. I want to say it’s late 1700s, late 18th century. The German government comes up. Things are modernizing. I think this is sort of like Biz Mark-y, getting into that era, and they want to have a predictable amount of lumber, because lumber’s a super important input for all the stuff that the government wants should do. They need to be able to build ships, and houses, and people need to burn to heat their homes, blah, blah, blah, blah.

Taylor Pearson:

And so, they come up with this idea that they’re going to basically build these scientific forests. They chopped down all these old forests and they plant them. And basically the visual looks like a field of corn, right? There’s just these perfectly straight… Plant all the trees at the same time, in this perfectly straight row, and you’re maximizing the number of lumber per unit of land. Right?

Taylor Pearson:

Because you don’t have any under brush. These other plants, they’re taking up nutrients out of the soil. It’s super easy to cut these things down. Right? If you’re like cutting lumber in an old forest, and you have an oak tree, and then an ash tree, and then a pine tree, maybe you need different equipment for each of those. It’s a different skill set. You can’t use the specialized workers. Blah, blah, blah.

Taylor Pearson:

So, they plant this perfectly scientific forest, and I think the most interesting part about this story that doesn’t get [inaudible 01:19:24] much is how long it worked better for. So, I think for 60 years or something, it was working phenomenal. The timber yield in Germany was up like 200 or 300%, and it was super predictable, and they could forecast exactly how much timber they needed, and how much they were able to do. And they could allocate it super well, and blah, blah, blah.

Taylor Pearson:

And then all of a sudden, this feedback loop, where it turned out that the underbrush that they had cleared was actually super important, because it allowed certain animals to live there. They put nutrients back in the soil, they allowed the trees to grow. And so, you had, it was, I think, Waldsterben is the German name, but basically it was like, mass forest death.

Taylor Pearson:

So, all these forests just started dying everywhere, and, I mean, the sort of finance implications here are, as you said, instead of having moderate volatility all the time, you had this very similar period of very low volatility, right? The timber yields are super predictable, and then you have this massive volatility event, where your timber yields go down by 60% in a given year.

Taylor Pearson:

And over the long run, it’s worse, even though it looked like, for the first 50, 60 years or whatever it is, it was going to be better. So, we talk about that a bunch, both in terms of A, how you structure a business. Right? I think it’s like, that’s one of the lesson’s everyone’s sort of talking about now with post pand-y, as it were. The supply chain, and all this lean stuff, and just in time. It’s like, oh, actually it can be kind of nice to have some access inventory. Lets you capitalize something.

Taylor Pearson:

Have some excess inventory, let you capitalize on things when everything is out. And then, same from a business perspective. One of my rules of thumb is I try to kind of work at 80% capacity most of the time, is sort of my internet right, because it’s like, and then like twice a year something crazy happens and I got to go to 110% of my capacity. But if I’m already running at 98% capacity just at a normal time, then when the 110% thing hits, I’m screwed, because I don’t have any sort of reserves built up. I’m I’m already at that point. So I think that there’s tons of interesting, I think just in terms of how we think about managing the business or managing portfolios, is that redundancy and stuff that looks inefficient in the short run, just like the old forest look inefficient in the short run actually can often be the most efficient in the long run.

Jason Buck:

Yeah. That made me think of so many things. One, I think there was a recent, I think it was Morgan Housel wrote that recent article of relating investing to almost like marathon training in, like you’re saying, is for marathon trainings to build your aerobic base, you want to be operating 80% of your training runs at 60% of your actual aerobic capacity. And that’s how you actually build up the base. Right? So like you’re saying, you’re building in that slack and the redundancy there instead of going all out every time.

Jason Buck:

Also, I was also laughing because you said you were explaining what the German forestry service looked like, then you said it was under that Bismarcky system. And all I was thinking was the rapper Biz Markie.

Taylor Pearson:

Right, no. Other Bismarck.

Jason Buck:

Yeah, other Bismarck.

Jason Buck:

The other thing that, and then so as people know how much we nerd out, we love talking about even permaculture, because like you said, the latency effect of permaculture, it’s so interesting to think about that as you’re destroying the actual permaculture that you’re growing crops in, through added additives of NPK and tilling the soil, like you said, you actually increase production. So you think you’re doing the right thing, and then decades later you have the soil that has just no organic material left in it. You can’t grow shit in it. So yeah, it’s an interesting trade off, like you said there.

Jason Buck:

And then the other one though that I tend to push back a little bit on, is that the whole just-in-time versus having inventory on hand, it’s not one or the other. Right? I think you would agree. It’s a combination. It makes me so mad during this Pandy Randy, where everybody’s upset about just-in-time, but they don’t realize their entire life has just had flourishing because of just-in-time manufacturing. And what was even shocking to me more than the pandy was how quickly entrepreneurs adjusted. And we didn’t all die off because they were able to adjust this just-in-time.

Jason Buck:

But there’s certain yes, in hospital supplies and everything, you’re going to want more slack in the systems and redundancies. That also is a function of maybe hospitals coming more for profit, et cetera, so there’s issues there. So maybe parts of the chain, you do want some robustness, but other parts of the team chain, the advent adjust-in-time inventory has made our lives so good. I’m just curious your take on that. Yeah.

Taylor Pearson:

Yeah. I think with the pandy, it’s hard because how much did the PPP and all those pro…. You know what I mean? I know a number of like small businesses that definitely would’ve gone under, if not for those sorts of… Right?

Jason Buck:

Yeah.

Taylor Pearson:

They basically got bailed out. Right?

Jason Buck:

Right.

Taylor Pearson:

And so it’s hard. I don’t know, right? Is that the majority of them? I don’t know what that is. Maybe it’s hard to use that as an example because there’s basically this huge external bailout stimulus, whatever you want to call it, that made that viable. But if you read the original Toyota production system stuff, like Shigeo Shingo and Tai Chiono and all that, the way just-in-time gets taught in MBA courses, I’ve never taken an MBA course, but this is my impression from the way people with MBAs talk.

Jason Buck:

We’re equal opportunity. We should have got everybody now. It’s physicists to engineers to MBAs. Keep going, keep going.

Taylor Pearson:

No offense to MBAs, but it got talked about in just strictly financial terms, like, well, if you can reduce the amount of inventory you capture or you’re holding and improve your cashflow, and it just gets talked about in terms of all this. And the original idea, the reason Toyota did it, was it just lets you see inefficiencies, right? So if you had, going back to our stocks and flows analogies, if you have all these extra stocks, these extra buffers built up at every part in the supply chain, you don’t see inefficiencies because the buffers can just cover for the inefficiencies. Right? So some part of the production line’s not working or whatever, but you don’t see that it’s not working for a long time because there’s enough buffer built in that, it can kind of keep eating it up.

Taylor Pearson:

So I think to me, you have to separate… Those are two distinct things. And just-in-time as a structural way to identify inefficiencies or bottlenecks in a system and improve them is invaluable and every business should do it, and it’s super useful and there’s basically no downside. But it’s not mutual because you can also have a buffer. Right? You can also have an excess. So I have the question is, are you relying on it all the time?

Taylor Pearson:

An example would be, if you’re manufacturing something and you have a bunch of excess parts in the warehouse, and the delivery for the parts is late, you might not notice because you’re like, “Oh, well we’ll just go to the warehouse and we’ll just get the extra stuff that’s sitting over there and we’ll use that and we’ll know when the delivery shows up.” And I don’t know, the delivery driver’s drunk or whatever, right? And you don’t identify that problem for a long time because you have this buffer that he can sort of keep showing up late to the delivery or whatever.

Taylor Pearson:

So you want to have the delivery truck come straight to the front of production line. And if it’s not there at 1:00 PM exactly, when it’s supposed to be, you have a process problem. “What happened? Why is the truck there? Blah, blah, blah, blah.” But you still have the excess stock. So that’s my hot take on just-in-time, is used correctly, super valuable.

Jason Buck:

Right.

Taylor Pearson:

Used incorrectly, not valuable.

Jason Buck:

Well, as per everything of pungentry, people are always strawmanning just-in-time, where you just gave a nuance steel man of just-in-time. Yeah, they just want to on just-in-time because it’s a great hot take on a bullet point, like if you’re on the news for 90 seconds. So that’s why I think why you’re seeing a lot of that recently.

Jason Buck:

By the way, I just on a tangent, I came up with a lot of things. So when we were thinking about your German forestry example, and then I was thinking about when, you got it on your notes, it was all about the allocation of wealth is kind of like what we’re talking about, but what is what actually drives wealth? What’s the driver of wealth?

Jason Buck:

And I think maybe that Eric Beinhocker’s going to be talking about this in his new book or his latest book. But I was started thinking about what kind of drives wealth. And he even talks about it in the beginning, correct me if I’m wrong, is the idea of a lot of the physics envy within Walrus and everything with the equilibrium is the idea of it’s almost like a closed loop system. Right? And it’s a static picture, like you said, ceteris paribus, but they’re not thinking about the inputs and outputs. And when I think Eric Beinhocker talks about, it’s an open thermodynamic system.

Jason Buck:

And so I was kind of starting to think about and research some of the inputs in the thermodynamic system. And it’s interesting to think about the history of energy and how that creates the capital, that we can then think about the allocation of that capital over time. So just some little anecdotes. For most of human history, we used humans, animals and wood as energy sources, right? And there’s this idea around energy return per energy invested, and allegedly using human animal and wood, it was a five to one return. So even with that return, you couldn’t really amass capital. You couldn’t be lazy. You’d still have to get up and go to work the next day because the input output was fairly at equilibrium.

Jason Buck:

And then, like you were saying, the coal was actually, I think, originally discovered in the 1300s, but they didn’t really start using it until the 1660s in London, and then you had all this externalities of all the soot and smog and everything from coal. But the reason they transitioned to coal is because they had basically chopped down all the wood, and then getting to the point where the wood was costing so much to ship, it was much cheaper to use coal. So the energy transition was just basically on the cheapness.

Jason Buck:

And when they switched to coal, now they’re getting a 15 to one output on the energy return for energy invested. Now it’s closer to 30 to one, but that 15 to one is then what explodes that economy, and almost the skews, he said, during the industrial revolution is because you had such a access capacity from the energy you were using. But there were some interesting things, thinking about the wood prior to the 1660s is… And an anecdote in, I think it was 1667-ish, the Globe theater moves across the Thames river, right? And Shakespeare and the brothers that own the Globe theater were moving it.

Jason Buck:

And the landlord said, “If you want to keep the theater here, I’m going to turn it into apartments.” They said, “No, we’ll take the wood and build a bigger Globe theater.” And the point they wanted to take the wood, because it was so expensive to get new wood, they wanted to use the old wood and actually transport and go through the hard work of dismantling the current theater, transfer it across the river and rebuilding the theater because just the cost of wood was so high.

Jason Buck:

And then, if you think about the British Navy and now how prosperous the British Navy was at the time is they had what they were called mass poles. So they would look for these old growth, 150-year-old trees that were perfectly straight for mass poles for all the sailing vessels. And so for a long time, they were able to get those in the north of Britain. But then once they chopped them all down, they needed more mass poles, so that’s why they were harvesting from North America, from the new colony in America and all the New England forests. But eventually we had obviously the revolutionary war and now they didn’t have access to mass poles. So this is interesting how a national security issue back in the 16/1700s was mass poles, and then they had to move to Scandinavia to try to find their mass poles.

Jason Buck:

But those were some interesting anecdotes. And then, like you were talking about earlier, we were talking about the idea of when you change the fitness landscape to technology, is oil was discovered in Pennsylvania in the 1860s. But for the first 30 years, it was mainly used for just lighting in homes. We didn’t know the actual use case for it until cars were invented. So I think about like how much that changed the landscape. And now we’re using all parts of the crude oil experiment, from the distillates all the way up to the actual gasoline, so you had you all these production forms that you could use, but how much that dramatically changed the fitness landscape.

Jason Buck:

And going back to this energy return to output, when we found fossil fuels, it was a 30 to one. So now we had stepped up from five to one, to 15 to one with coal, to 30 to one with fossil fuels. The interesting thing is then when we discover nuclear, it’s 100 to one. So it’s just the input output’s insane. And then when we think about renewables that I know Beinhocker’s talking a lot now with Doyne Farmer and everything, is well, wind is maybe at five to 10 to one, and solar is at one to five to one. Once you started adding in redundancy, storage, battery capacity, transmission, et cetera. So are we stepping backwards, right? Should we really be…

Jason Buck:

And I think we’re probably both in more of the Talebian camp. You want to diversify your diversifers with this. Just that the way we think about things is, so you don’t want to be 100% nuclear, but you have to take advantage of that 100 to one energy input to output. That was I just threw a bunch of just random at you, but I don’t know if you had anything you wanted to pull on from that.

Taylor Pearson:

No, I was actually going to use sailboats. That was going to be my example. There’s a good, Energy and Civilization. It’s a Vaclav Smil book. I think Bill Gates was a big fan of his work, but he’s, I guess, energy historian research or something.

Jason Buck:

Yeah.

Taylor Pearson:

But yeah, it’s a really cool… He starts with, and he has all these sorts of calculations, right? It’s like one human being eating 2000 kilo calories a day can generate this much output of energy on a farm. Right? And then, once you have a plow, you have a horse, then that changes it to this much energy you can output once you have a plow, this much energy.

Taylor Pearson:

But one of just the extent to which energy’s input has really made me appreciate that a lot more, but one of the best examples is was sail power and like how much energy, if you think about the total solution harness, why did this little island in the North Atlantic become a global superpower for a 100, 150 years? And his sort of take on it, or his angle, was they basically had this huge energy surge. There was all this wind that was going around, which no one else could effectively harness. And because of they were able to use it, they had the mass poles and blah, blah, blah, that was this massive source of energy which allowed them to do more industrialization. And coal was started in Britain as well initially. Right? So another big energy input or whatever, but yeah-

Jason Buck:

Cumulative advantage.

Taylor Pearson:

-on this, but yeah.

Jason Buck:

No, a cumulated advantage because it all kind of inter relates because throughout this, we were talking about the allocation of wealth, but how does wealth generated? And you have to kind of look at almost the energy inputs to think about how we’re able to even amass capital or wealth to even have the allocation ability to distribute wealth.

Taylor Pearson:

Yeah. And it comes back to sugar scape. Right?

Jason Buck:

Right.

Taylor Pearson:

To what extent was the fact that all the coal was easily mineable in the UK, how much did that matter? This is Jared Diamond Guns, Germs and Steel, right? This was a big part of his sort of thesis, right? That certain parts of the world had certain natural advantages that basically gave them a head start and that became cumulative advantage. And that explains a lot of the way wealth is apportioned across the world today.

Jason Buck:

And then you had a big section obviously on how all this ties into investing. And I think we’ve laid enough breadcrumbs where everybody can see how it ties into investing, or we can do a whole separate podcast on ties investing. And for people that know what we do, all of these principles tie into the way we build portfolios. But part of the things I want to maybe touch on is, I have a few other ideas, but do you have any sort of conclusion ideas or do you want to talk about investing at all? My favorite phrase comes from Hyman Minsky, is stability breeds instability. So when we were talking about all these things and these large alostatic bands and things get way out of whack before they can come back, those are the kinds of things we love talking about.

Taylor Pearson:

Yeah. So there’s a good simulation in this book about stability breach and stability. Do you want to try and explain it?

Jason Buck:

Ah…

Taylor Pearson:

Yeah, we were talking about it before the call, so I’m putting you on the spot, but I think you got it.

Jason Buck:

Yeah. It was basically Doyne Farmer had done an example where he was having a lot of traders having if then an experiment, where each individual agent was given a bit of money to start with. And then as the successful agents became more successful, they got more profits and then they’d have larger trading size. And it was really interesting, is it started off as a broad distribution of kind of returns. And then over time, as they get through more and more iterations, it narrowed and narrowed the cone of returns to look as an absolutely almost flat stable environment.

Jason Buck:

But in that environment, as we actually see today, it’s interesting, this is almost like vol targeting funds, as volatility gets reduced, and you have a smaller distribution, people are going to take larger position sizes because the reduction of that volatility, that stability, gets them to take on larger and larger positions. And so when that happens, as soon as that perturbation happens with outside that range, it just explodes the volatility.

Jason Buck:

So I don’t know if I explained it as well as you probably would, but if you take just random oscillations over time and traders and agents acting upon those over time, you’re going to go from what looks like a broad distribution to very narrow distribution, where you’re reducing volatility until something happens where now they’re taking on such a large size that now it explodes in volatility, and then those distribution of bands go erratically wild after it. So you go from broad to narrow to broad again, it’s kind of the best way to think about it without showing a visual representation.

Taylor Pearson:

Yeah. And I think that there were sort of three types of agents. There was a market maker. There was a fundamental investor. And if the price is higher than the value itself, the price was lower than the value. And then he had technical traders that were using different sort of price input of trend following or something would be an example of that, different sort of things. And yeah, that was the idea. These technical traders started to make lots of money. So started making bigger trades, and all the inefficiencies got dampened out and it looked as if the market was approaching perfect efficiency. And then one little thing goes off and you just get these huge, huge gyrations.

Jason Buck:

In conclusion, thinking about originally he started picking on classical economics saying they had physics envy, and that’s why a lot of the problems with classical economics comes out of they wanted to have a physical like system, and it just doesn’t work when we’re talking about economics. Do you think that complexity economics or ergodicity economics, do you think they have biology envy?

Taylor Pearson:

Yeah, probably.

Jason Buck:

Right?

Taylor Pearson:

I don’t know.

Jason Buck:

Their bias is like…

Taylor Pearson:

Biology envy seems more directionally correct than… And I don’t know. I think you always analogize to things you understand or use metaphors. That’s the whole George Lakof metaphors. We live by this whole idea of that’s how we as humans make sense of things. Right? So pre-computers, there was ideas like, the mind is like a gear machine, an industrial machine, right? The gears are clicking in your mind and that’s how stuff is happening or whatever. And then, once computers came on, we’re like, “Oh no, the minds were like computers,” blah, blah, blah.

Taylor Pearson:

And so I think it is tricky because to some extent, a mind definitely is like a computer, right? That analogy is true and useful in lots of ways. No one really understands how the brain works. Right? But there’s something there of how computers work and you have neurons distributing electrical signals, and just like a computer and that kind of thing. There’s something to that analogy or that metaphor that makes sense. I think there’s something to this and it’s directionally correct in some interesting ways, is my complexity economics review. So I think it’s cool in that sense without taking it too far.

Taylor Pearson:

And I think the people that do this are aware of it in the same way we sort of caricature [inaudible 01:39:39]. And these simulations you’re talking about, these are so simplistic, right? The sugar scape thing, or you were talking like with Doyne Farmer;s thing, right? That’s a massively simplified version of how markets work. Right? And so it’s not really that… Not that this is predictive, but maybe it gives you intuition another way.

Taylor Pearson:

I think one interesting things on the example you gave of stability brings instability is just gets into it, which endogenous factors. Typically thinks, oh, it’s some big thing happens. There’s a crazy earnings report, COVID you have some exogenous factor, but it gets into it just endogenous factors within a market or within sort of an ecosystem can trigger this cascade. And one of the other examples of this book was, I think of the 10 mass extinction events in the fossil record, the current thing is nine of them are basically endogenous, right? Maybe dinosaur one might have been caused by an asteroid that landed in Mexico or whatever. But most of the mass extinction events are just some… Well, there seems to be some dynamics of the internal system that triggered this cascade of events that resulted in a mass extinction without some sort of… You don’t need some exogenous explanation of this came in.

Taylor Pearson:

So we tend to ignore financial news and the problem when it’s like, “Oh, such and such declared earnings today, and that’s why the market tanked,” or, “That’s why the market went up such and such percent.” And my intuition, I think our intuition, is that it’s the endogenous factors are probably underappreciated.

Jason Buck:

As my lighting shut down, I was thinking about throughout this talk, you and I are sitting here pulling out all of these business anecdotes, scientific theories, referencing all the books we’ve read. I think you told me years ago that investing is actually the highest form of liberal arts. Everybody’s like, “Why do you want to get a liberal arts education?” And I think maybe we’re pointing out that when you’re studying things like complexity, it’s so interdisciplinary. So do you think, especially what we’ve learned now, I think about that often, what you said, that especially now in the last few years, do you think that we’ve learned even more, that this a proper portfolio construction is the apogee of a liberal arts education?

Taylor Pearson:

Yeah. I don’t know. I think that’s directionally correct. Right? You have to think about things in a super multidisciplinary way. But I don’t know, actually I’ve been thinking a lot too lately about just how much it even… Someone gave a good example. I’m trying to remember what it was. But there were a lot of people that were sort of bearish in 2008, and I think the fed changed some seemingly obscure rule about bank deposit regulation. I don’t know. This is around Lehman or whatever. Right? But these systems all bleed over, right? You can’t just look at the market. The market bleeds into politics, politics bleed the market, the environment bleed out… It’s all super interrelated. And so, yeah, the more I learned about this, the more I just have like an increasing humility about the extent to which anyone can figure any of this stuff out.

Jason Buck:

Yeah. I think what we’re both just angling for SFI fellowships. That’s all we really want at the end of the day, is to hang out at the Santa Fe Institute all day. This is great. Is there anything else you want to touch on before we go? Are you good?

Taylor Pearson:

No, this was fun. I hope everyone enjoys it. And yeah, let us know feedback, if we could do something like this again or try another format.

Jason Buck:

That was a great idea. Yeah. Look forward to you and I being able to do these more often. This is a great way of doing it for us to share our thoughts and everything, and hopefully people enjoy it. All right. We’ll talk soon.

Taylor Pearson:

Thanks for listening. If you enjoyed today’s show, we’d appreciate it if you would share this show with friends and leave us a review on iTunes as it helps more listeners find the show and join our amazing community. To those of you who already shared or left a review, thank you very sincerely. It does mean a lot to us. If you’d like more information about Mutiny Fund, you can go to mutinyfund.com. For any thoughts on how we can improve this show or questions about anything we’ve talked about here on the podcast today, drop us a message via email. I’m taylor@mutinyfund.com and Jason is jason@mutinyfund.com. Or you can reach us on Twitter. I’m @TaylorPearsonMe, me and Jason is at @JasonMutiny. To hear about new episodes or get our monthly newsletter with reading recommendations, sign up at mutinyfund.com/newsletter.

 

Want to get our best research delivered straight to your inbox?

Join thousands of sophisticated investors and get our best insights on portfolio construction and diversification delivered directly to your inbox.
Subscribe