Is complexity economics the answer?

Now, in our last episode, you heard all about complexity economics from Don Farmer. Now, in this episode, we're going to stay with the theme of economics, but we're going to tackle it in a different way. We are joined by Nicholas Grun, CEO of Lateral Economics, patron of the Australian Digital Alliance, and visiting professor at Kings College London. Now, Nicholas is going to talk about more traditional economics, and he's going to talk about the very real issues that we have in that discipline. He's going to talk about some of the potential solutions to those issues, and he really poses the question to us, is complexity economics the solution to these issues? Does it actually add anything to solving these problems? Now, you shouldn't view this episode with Nicholas as some sort of response to Don's views in our previous episode, because this episode with Nicholas was actually recorded earlier than Don's. So, with that caveat, let's jump into economics and let's join Nicholas to talk about why there's a real danger associated with jumping from one paradigm straight into another. This is Simplifying Complexity, a podcast where we explore the underlying principles of complex systems, systems that seem to defy our rational view of the world. Like economies, ecologies, or even you or me. I'm forensic engineer Sean Brady, and I'll be your host. Nicholas, thank you very much for being on the show. Hi Sean. So, what's your background in economics? Well, my background in economics is quite an interesting one, which is that I became an economist long before I studied any economics, and that's because my father was an economist, and somebody else who did that very effectively was Australia's former Prime Minister Paul Keating, who didn't have an economic education, didn't study economics, but by talking about it, he got the hang of it. And I remember, I used to work, I'll probably offend my old boss a bit, saying this, but I used to work for John Dawkins, who was the treasurer, who had an economic training. And Paul Keating's understanding of the economic way of thinking was better than John Dawkins. And likewise, the first bit of professional economics I did was working for John Button, who was the Minister for Industry, and being given a naughty policy problem, which is that we had an incredibly complex protection racket, a protection regime under which Australian cars were manufactured. And we had to reform it, and there were two factions, and one was protectionists, and the other was free traders, and they didn't really speak to each other, and they inhabited different universes. And I was completely sympathetic with the objectives of the free traders, but the free traders didn't actually even listen to the protectionists. The protectionists, unsurprisingly, came from the industry, and they understood the economics of the industry much better than the economists, because economies of scale are incredibly important in the industry, and they're invisible to economists. Now, someone could pick me up and say there was actually a big study of economies of scale by the economists, but they didn't incorporate it into their thinking, they did a whole lot of analysis on it. Anyway, that was realising that the economists were using their discipline, they weren't using their discipline as the beginning of thinking in a practical way about what the options were and what the best political compromise was, and they actually weren't even addressing the weaknesses of their own position. Their own position was that either for reasons of political necessity or because it comes out of the textbook, we would gradually reduce protection. Then the question becomes, if you're going to gradually reduce protection over, say, 10 years from a very high level to a very low level, then the next question becomes, well, that's a highly expensive, all this protection we're giving the industry is a very expensive thing to do. How do we do it in such a way that we maximise the chance that at the end of this process, the industry's had enough support to invest in its future rather than clinging onto its past? That would mean that we would want to make sure that it was given as much assistance for exports as for protecting its own domestic market. That was actually something that the free traders objected to. I encountered all this and I thought, look, these people, they know more economics, but they're not using it to solve problems and pose themselves difficult problems so that they can see where the weaknesses in their own position lies. That's a template experience, a formative experience. In many ways, I could tell a very similar story about all of the things that I've done in economics. I guess the bottom line is you use these often very simple ideas out of economics, but you don't think that economics gives you the answer. It gives you the tools and then you have to use them. You have to try to inhabit the problem that you're trying to solve and you use these as tools and you have to use judgment and common sense and then you have to talk to people and listen to what they say and not just say, "Oh, well, they would say that, wouldn't they?" Yes, they would. They've got incentives too, but there's information in there. That's my way of doing economics. And is that just really what their position is? The dogma, so to speak, was more important than the reality of its application and that specific system. Is that the fundamental issue there? Yes, but I like to use words that are not pejorative. So you could say they're disciplined in a very simple way to think and one that I've already mentioned to you in previous discussions in which I use when I talk to young economists. And I say, "You've been turned out of university thinking that you know 80% of the answer and then the other 20% will change as you go from situation to situation. I'm here to tell you that in my view, you know 10, 15, 20% of the answer and almost all of it will." And that's a set of hints and that's a set of tools and a set of hints about how you might like to think. And then you have to try to solve a problem. And people aren't trained to solve problems. They're trained to learn curriculum. So it's everywhere. Education system does not teach how to solve problems. It stuffs you full of a curriculum and it tests you on. So go to complexity science. You're not a fan necessarily of it. But can we start with the criticisms that the complexity economists, if you can use that term, have of traditional or neoclassical economics. How legitimate are they? How accurate are they? I think they're completely legitimate. And you will have seen some of the stuff I've written. I'm just despairing about the state of economic science because it is one curriculum that everyone's learned. And by definition, in terms of the way I'm thinking and speaking now, by definition, it will be a tiny part of the answer. So if you give yourself another place to look from, which complexity science does to some extent, you can see how bizarrely pro-crusty and how bizarre the orthodoxy is because the orthodoxy is just a set of hints about how the economy might be. It's a kind of highly stylized way to pin down all the moving parts. So at least you can get a bit of an idea. It's a wrong idea, but it's better than no idea about the kind of thing that's happening. And I think that in that sense, orthodox economics says we should be grateful for how much it helps us to think about things. But it's still wrong and it can be wrong in very big ways. So going up to the global financial crisis, we have macroeconomic models all around the world and they don't have a financial sector in them. Well, you know, that's the essence of the problem. So there's this and some complexity models help you see the kinds of things that lead to financial blow-ups and so on. So it's just bleedingly obvious that we're being captured by our tools and everyone sits around as if they're helpless about it. Now, my gripe with complexity economics is what I call it's a paradigm war. So you critique one paradigm and then you say you need a new paradigm. But if that was the case, then complexity science would have sorted a lot of things out by now. It hasn't. It has shown us the kind of analysis that captures the phenomena that we want to capture. That's it. And most of it will still be about common sense. It won't be about the curriculum. That's a small part of the answer. It's not a big part of the answer. But because it's a major part of people's egos, because it's a major part of people's identity, it gets far too much attention. As a consultant, I can tell you it's also the major part of a consultancy, which is some tools, some tricks, some system that you can sell clients and then you can do it again and again and again and again. And that you make much more money that way than saying to your clients the sort of thing that Warren Buffett says, which is try and think about this stuff as it comes at you and the late Charlie Munger. Charlie Munger has 100 tools in his toolbox, maybe 10 or so important ones, perhaps a few more than that. And he uses them to try to help him identify value. That's not how we teach anything. But there you are. That's what works. And what are the big issues that are wrong or cause problems? And we've touched on these in the series before, but I think it's useful to sort of get your thoughts on Nicholas. Well, I would say that macro, we sort of haven't turned up and we still haven't turned up the kinds of things we need in macro. And this is models of the macro economy. So it's models of the economic cycle models, which try and answer questions like how should we set interest rates or that sort of stuff. What leads us into trouble is the financial sector. And we need models of the financial sector that are more thoughtfully constructed than the ones that we have and the ones that we use, even those ones that have been built since the financial crisis. Because the financial sector is so important, increasingly important, and because if you go and get a mortgage, they will lend you 80% of the value of the home. And if you think about that for a little while, if prices are much higher, they'll lend much more. And if they lend much more, prices will go higher. So we are in you engineers know this is positive feedback. So we have a macro economy that has all this positive feedback in it. And yet our micro economic practices assume that macro prices give us information that is kind of stable and reliable. So that's a simple way to explain it. There are other ways to explain it. Lots of interesting ways to explain it. Of course, Keynes explained it in more complicated terms. If you like to get a bit philosophical about it, it's a kind of recursive reference of meaning. So that's a complicated thought. And I'm not sure I'll even myself be able to work it out as I try and articulate it. But I'm basing my idea of the value of this thing on other people's idea of the value of the thing. And other people are doing the same thing. And we're betting money on this. And so this produces the capacity for large swings upward and downward. I mean, another thing that's important about finance is financialization. So in this world, the best analogy here is that if you watch a very exciting piece of football at the MCG, you're tempted to stand up so you can see better. And everyone's tempted to stand up so they can see better. But of course, once everyone stood up, everyone's standing up, they'd all rather be sitting down, but they're all standing up. And we would all rather house prices were three or four times annual earnings as they were in the 1960s. But a major reason they're eight and nine times earnings is that we can all go out and buy and bid against our neighbor to buy the house that we want. And so the stock of housing becomes financialized. And this is an expensive service. It's provided in a pretty outrageous way by state backed firms that are allowed to privatize the gains. And then if they fall over, we hand the money to prop them back up again. So the financial sector is poisoning our economy in many ways. And that's something that central banks should be worried about. The only central bank that I've ever known to be worried about it really is the Bank of England when Andy Haldane was there. And he published just eye popping calculations of the implicit subsidy to large banks during the financial crisis because they were perceived as a safe place to put your money, not because they were run conservatively, but because people knew that if they fell over, the government would just write them a check. And what about some of the basic things that you hear criticized from a complex economics perspective? So what's your thoughts on the whole equilibrium thing? Is that the concept that neoclassical economics relies on equilibrium? Is that okay in some places? Is it not okay in others? Is it okay? Or is it a disaster? I think all of these things are fine. If you understand how they came about and then you behave with some fidelity to how they came about, then it's fine. So the idea of equilibrium basically turns up, I presume, in the 1870s when the marginal revolution was underway. So we had this discursive text based way of thinking about economies as different classes of people. This is classical economics and it was a kind of deductive process, but it was all written out in words, almost all of it. And then in the 1870s, people got the idea that they could ape classical mechanics. And then, you know, you'd have a supply curve and a demand curve. And then, of course, they don't tell you anything unless you say how do they relate to each other. And it's a very reasonable thing to say that, well, if we're going to use this as an intellectual construct, we'll talk about where they intersect and we'll call that an equilibrium. Unfortunately, let a few generations go and put this through our educational system. And then we teach economists, and I remember Brian Arthur talking about this on your program, we teach economists that they do equilibrium. You know, and so Brian Arthur does some analysis of disequilibrium and they say that's not economics. Well, more for them. So Brian Arthur is doing work that is entirely consistent with the animating spirit of the original architects of neoclassical economics of what is now orthodox economics. But unfortunately, it turned into its own form of scholasticism. And, you know, I watched this all the time. Let me give you a really simple example from the case that I mentioned to you, which was working in the car industry. And I mentioned to you that the what was then called the Industries Assistance Commission and in Australia is called the Productivity Commission. Now, it actually did a special study of the economies of scale in engine production and panel production, a range of other products. And so if I said they've got the wrong paradigm, they should have done something else. They can say and their paradigm didn't handle economies of scale. They can say, well, pigs ass, it did. Here's the studies. But then when you look at the studies, here's one of the things they did. They defined minimum economic scale, which is a thing you kind of get out of the textbook. They defined minimum economic scale as that level of production at which a doubling production would not lower unit costs by more than 2%. Now, that number is obviously just an arbitrary number. It's like saying statistical significance is 99 or 95 or whatever. It's someone made it up. More than that, they made it up in the United States. And the United States is a very big market. And it didn't occur to these people who are trained in economics that this thing that they just pulled out of an American textbook, there are other textbooks that say that minimum economic scale is that volume at which a doubling will not reduce costs by more than 5%. So it's different as you would expect. Somebody just made it up. But was there a process in the industry's Assistance Commission saying, should we use that textbook or that textbook or better still, just use that as a start and say, we think the numbers in Australia in this industry is probably 7% or 10%. And here's BMW, which has still got lower volumes than Australian Holdens. And it's doing fine. And Sub's got half the volumes that Australian Holdens have. It's okay. That's how the textbook needs to be related to the world. But it didn't happen then. And it's not happening now. And that's got nothing to do with complexity theory or chaos theory or anything else. So one big takeout for me is pay attention and use these things, adapt them to help you think. That's where most of the gains are. And complexity theory can sit there in a way that is quite like the way I would like to see neoclassical theory sit there as a set of prompts, a set of lenses to check things out through. You've said that complexity economics doesn't quite live up to the promise it had. What do you mean by that? Yeah. So once you have children, you have to let them out into the world and their joy and a revelation. But then they're not quite what you had in mind. And that's true of paradigms too. Predicting when the next recession will happen is really, really hard. And you can have a paradigm shift and it'll turn out, I mean, you know this expression that politicians campaign in poetry and govern in prose. So another paradigm is a great thing to campaign in poetry with. It's a great way to point out how really dumb the existing paradigm is because every paradigm is dumb. Every paradigm is a set of pretty crude, pretty simple approximations. And then you say people should understand complexity. One of the terms that turns up constantly is, you know, the way people are thinking is linear, whatever the hell that means. I talk about that a lot too. Yeah. Silly. I mean, economic models aren't linear. There are bits of them that are linear. There are bits that are not linear. But we like to say that. And then we think, well, obviously something that can do curves around reality is going to work better. Well, you try it. So you can build these things and I think it's great to build them. But they won't allow you to turn something that was really, really hard into something that's even remotely easy. So if you're trying to do something hard, which is apprehend reality and maybe predict it and then maybe influence it, that's hard. And so there's another takeout, Socratic Wisdom, which if we talk about forecasting at some stage, I'll have plenty more to say. But whatever you end up with will be very unsatisfactory. And the answer to that is that that was supposed to be the beginning of the process, not 80% of the work. But that's unfortunately the way it often is taken to be. Yeah. So you see complex economics, neoclassical economics is a set of toolkits that you can take and start to use and think sensibly, presumably about the specific problem you're trying to solve and not in a generalistic, because that wouldn't make it a paradigm, presumably, you know, when you're trying to make it too general, you're trying to make it a law or a rule. Well, no, I'm happy with the idea that it is a paradigm. It's just that the word paradigm first got cooked up in physics and was the Thomas Kuhn's idea of the structure of scientific revolutions. And you those two possible models that the earth goes around the sun or the sun goes around the earth. And once you sort that out, a lot of things come into focus here. A paradigm can tell you many things, it can give you many instincts about a very large phenomenon. That's fine. But then it's just the very first impression and then your task changes. And it's partly to adapt it. It's partly to throw it away where it doesn't work. It's partly to find other tools. And it's mostly about exercising your judgment to try and work out what kind of risks you're taking. What are the upsides? What are the downsides? Who do you have to persuade? Who's involved in this? Whose agency do you want to try to empower within this system? And economics just gives you any of these paradigms will give you tiny little beginnings of thoughts. That's all it will give you. If you go to a neoclassical tool or concept or an idea or a starting place and you go the same in complexity science, what are the one thing from each of them that maybe you use the most or is your favorite or is the most curious and most interesting? But if you were to take a tool from each of the tool kits or an idea from each of the tool kits, does anything jump to mind? I'll let you in on a guilty secret. Not only do I not do much modeling, but modeling doesn't actually achieve a lot in the world. Let me give you a shocking example. When I was actually on the Productivity Commission, it was then called the Industry Commission and we were doing an inquiry into the car industry. And we had a guy who got a lot of publicity a long time ago in the 80s and early 90s called Michael Pusey from the University of New South Wales and he was a sociologist and he said economists were mesmerized, they were caught up in their own models. Now that sounds like a very similar thing to what I'm saying. But I'm saying something slightly different, which is that their models are not some mathematical model that you crank through and read out an answer. They're models in our head, they're connections between things. So the Industry Commission produced a draft report on what we should do to tariffs and we all knew what the answer was. We should cut tariffs. Everyone knew that. That was because of the model in their head. But we were also getting some modeling done of our various options of what impact that have on the industry and the rest of the economy. And they were being done in Monash University in a thing called COPS. It was called COPS as I recall and it was at Monash and it's now moved to Victoria University run by the fine man Peter Dixon and his daughter now, I think. And COPS were late in providing their modeling and they were so late that we couldn't get it in the draft report. We didn't know what the modeling said when we issued our draft report and nobody was embarrassed by that. So the modeling was a kind of an embellishment. The modeling was not, the sort of technical modeling was not part of the process of using what we thought was our common sense and most of which comes from the discipline and comes from, and also the sort of neoliberal assumption which doesn't really come from neoclassical economics, which is markets are better than governments. That's sort of the crude way people thought then. What neoclassical economics does is it gives you lists of checklists of things where if you want to ask, will the market perform better than the government, then you can ask, are there externalities? Is it a competitive market? Is there asymmetric information? We didn't do any of that. Although you could say we kind of did a bit of it in our hits. So most of these kinds of things that I'm familiar with, the modeling is usually the sort of number crunching modeling is a very small part of the process. And in fact, if you think about it, if you were waiting on numbers to come out of a black box, that's a pretty scary way to be making important decisions. You want to have more of an intuition for what's going on. So to get back to your original question, if we have a neoclassical economics of these questions, you have an industry, you shock it with lower tariffs, it gets smaller, the rest of the economy adjusts. I'm not sure that complexity economics can tell you very much. But then if you're wondering about, let's say, you're wondering about, is there a critical mass, which if the industry falls below a certain critical mass, will it get wiped out? Well, the neoclassical tools will be terrible at that. And so you might try to build some kind of understanding using other techniques to give yourself some purchase on that question. But again, I'm a skeptic as to how much of I'm a skeptic that you could get that far with it. But at least you're not using completely the wrong tool to answer a question. But maybe the main insight is we don't know that. And then the other thing is to then go and look at empirical examples. And you can see that in Germany, Germany had quite a small car industry. Anyway, the cars we think of as luxury cars, particularly BMW, was quite a smaller manufacturer than Holden was until 1978 or something like that. That tells you quite a lot, much more than some model would. And why, in that particular example you give, why is neoclassical tools absolutely the wrong tool to use? You might be able to rig something up in neoclassical economics, but because neoclassical economics assumes the problem away. Neoclassical economics will assume that these markets are competitive. So it won't give you these continuities that something which takes those kinds of things more seriously would. Again, I don't know whether you'd call it complexity, but there are so many things going on. There's learning, learning by doing, learning with R&D. That's not in neoclassical models. There's some stuff in neoclassical economics that has some things to say about it. I don't know how useful they are, but in the computable general equilibrium model, the modeling that was being done to answer this question, if we shock the industry by cutting tariffs, how much smaller will it get? How many jobs will go? Where will they go? Who will it advantage? Who will it disadvantage? A CG model which makes neoclassical assumptions generally, that'll give you a reasonable cut at the answer. At least it gives you an answer. One of the nice things about this is that if you work it out in a model, at least it gives you an answer where you aren't fooling yourself by double counting something, which is very easy to do in the back of your mind. Whereas if you work it all out in a model, at least you know that this adds up and we haven't double counted anything and we haven't left anything out. What do you mean by double counting in that example? Well, you can say things like as the industry gets bigger, it will get more economies of scale and then you think of those economies of scale also as learning economies. Now that's become dynamic economies of scale and you're now talking about two things, not one. A model will force you to define these kinds of terms. Whereas a general discussion will allow you to lump them all in in a kind of commonsensical bundle and that's not all bad. Being absolutely clear about everything, you can't do that. But the strength of it is that if you go through the modeling exercise, it's one of the things I think is if you knew it's Bismarck's comment or supposed to be Bismarck's comment about government, you know, it's like sausage is its best if you don't see them being made. And this is true of modeling as well. I think it would have been fine to come up with some rules of thumb and say there's this many people in the industry. If you cut tariffs, we think it'll get about 10% smaller. That's this many jobs and tourism will do well because it's in the traded sector and it's got no exposure to this kind of thing and its costs have gone down a bit. So that's a kind of commonsensical way to solve it rather than putting it through an extremely complex model. You don't really know at the end of that process whether the extremely complex model has given you a better answer than a more broad brushstroke answer. But it's good to have both of them because sometimes the complex modeling will turn something up that surprises you. And it's always interesting to say, well, why did the model tell us that? And then you investigate the mechanism and you think, yeah, that's quite a good point that I would have missed. So when you're talking about modeling like that, Nicholas, you're not really saying that you're using the model to essentially do scenario planning. So you're not using it necessarily. You're advocating not using it for a predictive role, but a scenario planning role. So at least you can think out the possible things that could happen. Yeah. So I think that there's some value in that and the answer in a way is yes. The only thing I would say is that scenario planning, you've always got to start getting suspicious when something becomes very, very much flavor of the month. And scenario planning came in, I think in the '80s, Shell came out of the oil shocks of the mid and late '70s better than its competitors. And there was a, shall we call him, an entrepreneurial senior executive at Shell, and he said it was all due to our scenario planning. Now, it might have been due to their scenario planning, and I don't for a minute want to say that their scenario planning wasn't good scenario planning. But then scenario planning became a thing. And once it becomes a thing, you get an army of consultants and they provide these three or four or five scenarios and they've got hokey names like, you know what I mean, you know, we solve our problems and we live in peace and harmony. And then another one is social decay and at least, and the answer to these kinds of questions is really the same as the answer I gave in a debate I had with Paul Krugman when he said, well, I, Nicholas, you know, seems very against the use of mathematical modeling. And the answer is I'm not against mathematical modeling. If you can do it well, if you can do it in a way that contributes, that's good. And there are lots of ways to do it in a way that is dumb. And I don't know any mathematical modeling that can possibly be add inside if it's not embedded in thoughtful, discursive analysis, thoughtful analysis that you talk about and you put in words and so on. And I'd say exactly the same thing about scenario planning. It's not the framework that you shoehorn it into. It's thought. It's human thought. It's a little bit creative. Although, of course, we've now sort of romanticized and sentimentalized that word. It's a little bit creative. It calls on rigorous thinking. If there will be times when you need to sharpen differences in order to really work out what you're saying, there'll be other times when that has become a kind of reflex action and it becomes a kind of cleverness competition. So these things are quite delicate things. And this is why they're very hard to systematize. And we're in a society which has these kind of stovepipes everywhere that sell you systematized things, whether they're universities and you're a student or whether they're consultancies or whether they're government departments. And what loses out is the ability to take a bit of that and then to say, right, well, where does that leave us? And what other stuff? What doesn't it help us with? And so on and so forth. So it sounds like almost an evasive answer. But the answer is pretty much nothing that you can do, which if you do it well, will make a contribution. If you do it in a well-judged way, it'll make a contribution. And if you don't, it won't. And there's nothing you can pick off the shelf, like a cigarette packet and say, "I smoke alpine." And then say, "This is going to get me the answer." That, I think, is the enemy, which makes me think I should come up with a name for it. But I haven't got one right now. It's interesting. I completely agree in a lot of the work I do. I see exactly the same thing that not only do consultants want to sell an off-the-shelf idea, but you can systematize it and leave it and do all those sorts of things. But it's amazing that many clients want to buy it because they want to buy the certainty that this solution purports to give. And that's part of the big issue. Absolutely. Yes, it looks certain. It looks, and again, it gives you someone else to blame if it doesn't work out too well. And consultants always flatter upwards. And flattering upwards is the entire decision-making structure of our entire civilization. It could lead me to lots of digressions on ancient Athens and Venice and other places, but maybe we'll leave that to another time. Just to finish, Nicholas, what would your top three tips be to someone trying to solve an economics problem in the real world or whatever sort of environment you want to put it in? Of all we've talked about, how do you bring all that together into three things? So the first thing is to know what Socrates came to know, which is that the more he learned about the world, the less he knew that he understood more and more what a tiny fragment of knowledge that he had. And that's true in most areas. It's not really true of some disciplines. It's not that true, say, of accounting. What the numbers refer to, it might be true, but the numbers are the numbers and accounting is accounting. And if I tell you that you've got to the sort of things I've been saying, I would be absurd to say to an accountant, and they'd be absurd to say to an engineer if their task was to build a bridge, and it was a certain kind of bridge, and there would be a great deal of established discipline. There was also adaptation and the process of getting to that point, maybe the sort of advice I'm providing would be more sort of be prepared to chance my arm to say that I might have something to say there. But there are areas where disciplines have been developed, which are pretty strong, and you learn them and you apply them and you try and keep your wits about you. But it's not what I mentioned at the beginning, which is kind of 80% your judgment and common sense and 20% the discipline. But that would be my first rule, understand how little you know. I think the other thing I would say is rather than fall in love with a new paradigm, by all means have different run different paradigms against each other to see what kinds of insights they deliver to you. But if you know a lot about a paradigm, try and make it work harder for you. And I can give you an example, and maybe we can talk about this in more detail in another discussion. But in my discussion with Paul Krugman, Paul Krugman says that trade economists forgot about scale economies in the 1950s and 60s. Now, that's not quite true. A certain type of trade economists did. And the reason they did it is that it worked out better for their maths to simply assume perfect competition, which is sort of a very intense competition of many, many players. You're going to have to explain what a trade economist is. Well, and a trade economist is an economist who's a specialised area of economics, and it says how does trade work? So it'll look at Australia's trade with China, America's trade with China and Russia and whoever else is trading with. It won't look at the economics of infrastructure in domestic infrastructure. It's an area. And in the 50s and 60s, it simply became what you did in this field to assume perfect competition. And then low and behold, scale economies don't exist in a world of perfect competition. They're logically inconsistent with a perfect competition. I think even if your viewers don't know the causal chain by which that works, they can get the idea that you're making certain assumptions and those assumptions themselves simply rule out certain phenomena. Now, there were plenty of people writing on the edges of this discipline about scale economies. They were thinking creatively and effectively about them. And Korea and Taiwan were building their entire economic policy and their trade policy around scale economies. Scale economies are those things we talked about earlier, which when you make one car, so the first car that was made in Australia cost a million pounds. The next car cost 20,000 pounds and got down to well below a thousand pounds. This is in 1948. So scale economies, the more you make the cheaper it is to make these things or to do this service and everybody will be familiar with that. And that was simply ruled out by assumption in all these models of trade. Now, when the 1980s came along and people were trying to understand why Korea and Taiwan were doing as well as they could, those trade models, those highly articulated mathematical models, could only give you the wrong answer. Because the thing that was of the essence, which was that Korea was specializing in small cars and exporting small cars and transmissions and iron castings and importing various other things, it was doing that because of a scale economies. And Paul Krugman was a major part of inventing this new area called new trade theory, strategic trade theory. And what he did was he put scale economies in there. That's all fine. Now, in doing it, he worked out, I would argue, why they weren't there in the first place, which is if you put them in there, they got taken out for a reason. The reason is that if you have them in there, you can't get the maths out. Therefore, you can either forget about them and keep doing what you're doing and you say, here's the maths and isn't it lovely? Or you say, well, we'll see what we can do as far as fitting this in the model, but we'll just have to do most of our thinking the way they're doing it. The trade officials in Korea and Taiwan are doing it. We'll do it without common sense. We'll make special ad hoc models to try and measure things and talk about these things. So on a number of occasions now, Krugman says, well, you know, it's easy to say they should have had these things in the models, but nobody knew about these things until people like me, Paul Krugman, put them into the models. That's simply not true. There are lots of publications, but the publications treating those matters were not in the style that was approved of. And so they didn't get published in top journals. And I just regard that as a disgrace that the commanding heights, the cleverest people in a discipline all around the world, just are complicit in a situation where an extremely important phenomenon is simply ignored. But that was the case. And that you mentioned Brian Arthur, and that's what happened to him as well. And that's exactly what happened to Brian Arthur. That's exactly, people said completely stupid things like, well, that's not what we do here. That's not economics. So that was somehow the second, my second response, which was to, you know, there was nothing wrong with this paradigm. You just had to understand its limitations. And then when the particular phenomenon called for it, when you can see that this thing is an important thing, then you put those models to one side that you keep them because they can help you get a general intuition for how trade flows work. And then you realize that nevertheless, it's wrong. And we'll just have to use our common sense. We'll have to use some ad hoc models, some little models that can do in a spreadsheet or something to get to help us do the best we can. So that's another piece of advice, which is to not think the answer is in another paradigm. Other paradigms, by all means, take an interest in them. But the real value is delivered in doing it well, doing whatever you do, whether it's mathematical analysis or just trying to get common sense. There are ways to do that insightfully. There are an infinite number of ways to do that in a dumb way. And if we have silly debates about whether mathematics belongs in analysis, rather than good examples of it and bad examples in it, we just sort of, again, we're in dialogue about the wrong thing. And that sort of leads me, I suppose, to another really big thing. And again, it's from Charlie Munger, who says, "You'd be amazed how much money Warren and I have made." Warren Buffett, who was the richest man in the world for a while there, simply by avoiding or trying to avoid obvious errors. And there are obvious errors all over the place. I mentioned COVID. It was an obvious error once we found out that it wasn't a flu to keep doing four or five months of planning on the basis that it was a flu, because that's what we had in the protocol. There are obvious errors made all the time. And by keeping our wits about us, by understanding that the heart of this enterprise is critical thinking, self-aware critical thought, and dialogue with others, especially people who don't see it the way we do. If we don't understand that that's the core of it, then we just set ourselves up for one mistake after another. I think, for me, one of the things that you clearly have no issue with the criticism complexity science have, you're just simply saying, throwing out one paradigm and mindlessly embracing another. That's not the solution either. No. And we've got to get beyond campaigning and poetry. Since I came up with that during this conversation, I'm quite proud of it. I quite like that as well, yes. Yeah. And the poetry is fine. The poetry hits the mark. And now we're back to where human beings have always been, which is that we're the smartest being on the planet and we're pretty dumb when you compare us with what the hell is going on. And we just got to make the best of it. And that's a rich meal. And something we didn't really talk about. Maybe we can, in a subsequent discussion, that is a rich ethical meal. It's not just a cognitive meal. And we almost completely forget about that. And I think that's an incredibly important thing. Perfect. Thanks for listening to Simplifying Complexity, where we look at the key concepts of complexity science with expert minds from across the world. Concepts like emergence, self-organization, adaptation, networks, scaling, tipping points, and much more. This podcast was produced by Brady Heywood and Wavelength Creative. To make sure you don't miss an episode, be sure to subscribe to or follow the show in your podcast app. I'm Sean Brady, and I'll see you in our next episode. you