Hello, everybody. Today we have Brian Arthur for our guest spot. Brian is one of the founders of what is now called complexity economics, that is viewing the economy as a complex adaptive system. Brian has spent many years as a professor at Stanford, and then as a faculty member -- research professor at the Santa Fe Institute. We overlapped by several years in the early days. And Brian was the very first director of the SFI Economics Program He’s currently a visiting researcher at the Palo Alto Research Center in California and he’s an external professor at the Santa Fe Institute. Brian has written many papers and books on economics and complexity. I’ve put a link to one of his recent papers on complexity economics on the course materials page so you can download it. Brian’s most recent book is called The Nature of Technology -- What it is, and how it evolves. So welcome, Brian. Thank you, I’m delighted to be here. Brian -- why do you think the economy is a complex adaptive system? Well, I think it’s a very old way, actually to look at the economy In economics we tend to think of the various actors or agents in the economy, the players in the economy, if you like, as together creating some outcome that they react to, so could be investors in the stock market, they’re buying and selling a particular stock and that creates how much is bought and sold and how the price adjusts, and then they react to that new prize. So I tend to see complex systems as systems of multiple elements creating some pattern or outcome that those elements react to, and economics is certainly very highly interactive system of that sort. Also it’s true that economics has seen itself that way for a long, long time. When Adam Smith was writing economics in the 1770s he pointed out that what we would now call the agents in the economy, he mentioned butcher and baker and candlestick maker they’re all creating something that they’re reacting to so at least 200 or 230 years ago, 240 years ago, people were looking at the economy as a complex system. A little bit after that, a hundred or so years after that, things changed. How did they change? Well, there was much more tendency after about 1870 the idea was to mathemetize the economy and to make sure that we could analyze that with equations and do that. Things got quite a bit simplified. So economists in the Victorian times in England anyway began and in Switzerland, places like that, began to look at the economy as more like a machine something like a perfectly ordered machine with all these interacting parts and all these parts were working together to create something that was consistent or in balance. So the emphasis was on being in balance. Everything was in a sort of stasis, a bit like not that it was static and not going anywhere, but more like the surface of a lake reaches a certain level, occasionally there’s ripples and it comes back to that level again. So for at least the last 100 or more years, 120 or 30 years we’ve been looking at the economy as an equilibrium, perfectly designed or perfectly equilibrated machine in equilibrium with everything in balance, moving but everything very finely balanced in some sort of perfection Now we’re getting away from that again. People are looking at the economy in a much more organic sort of -- almost biological way. That’s right We’re starting to see the economy again, I think the way people did two or three -- two and a half hundred years ago. We’re starting to see that people -- that there are new opportunities coming along all the time. It’s a bit like the biosphere, new species arrive all the time, and that changes the outcome so the difference actually between how we look at the economy now and under as a complex adaptive system versus equilibrium economics, is a little bit like the difference in possibly in biology. If you choose to look at species, you know the number of wolves and the number of spruce trees and the number of sheep in the economy, you can see maybe that they’re in some sort of overall equilibrium, and have an equilibrium biology based on that. But if you look over the long term, that’s not the way it works at all. You see new species in fact, whole new orders of species coming in all the way from primitive times, slime molds building up into some primitive types of plants and fish and so on, and radiating into very different things, and then we get dinosaurs and reptiles and mammals bringing in different eras and so we’re starting to look at the economy this way again, that there’s new things coming along those are creating new structures, the players in the economy are constantly reacting, and they're themselves bringing -- those reactions bring along yet fresh opportunities, so the economy is evolving as behaving very much like an ecology. It’s interesting that Darwin used ideas from economics to understand biology, and now people are going the opposite direction, right? It’s amazing, Darwin was reading Malthus in I guess the 1830s and so was Wallace, by the way, they came up with their theory of evolution by reading economics, and now I think the traffic’s in both directions very much. What we’re looking at now is that players in the economy, the agents in the economy people in firms and so on, are exploring, always exploring new opportunities and exploring new strategies, new ways to forecast new ways to think, new types of actions, in some sort of situation that is created by all those actions and strategies and forecasts, so we’re tending to think of people trying out different strategies in an ecology that consists of other strategies. The view of economics is very ecological, and so we’re having conversations with ecologists, people like Robert May many, many others as well. And ecologists, I hope are learning from economists as well, because we’ve been looking at this sort of thing for a couple of hundred years This point of view, what I want to emphasize is that the point of view harks back 200 years ago, viewing the economy as evolving, changing always changing, always adapting but we have a lot of new tools that we’re bringing on and those are helping redefine the economy as a complex adaptive system. That was actually my next question, because you’ve coined this term complexity economics which is now getting a lot of press and attention and I was wondering -- maybe you’ve already answered this -- but what characterizes complexity economics versus, you know, economics? Versus standard economics? Is it the tools of complex systems modeling, do you think? No, I think it’s -- complexity economics means taking the interactions in the economy seriously, and we’re not seeing the economy as always static or as always in equilibrium The economy is full of opportunities and the players or the agents in the economy, be they banks or firms or consumers are always adapting and so when you see people or players in the economy always adapting to the situation they create that may or may not settle down into an equilibrium, so we’re beginning to see very much in many, many parts of the economy finance, international trade, that an equilibrium isn’t reached. It’s one thing leads to another. I think it was Churchill, he was asked about history and how would he define history, and he said, history? History is one damned thing after another. And in one sense the economy is like that, too. We’re looking at the economy as being open, as creating opportunities, the agents are always reacting to that and in doing so they’re changing the situation again If you want to look at everything that’s happening as being in a balanced equilibrium, you get standard economics and you can do a lot of analysis that way It’s a simplification, and it has served its purpose very well for about a hundred years, but if you want to generalize and say it’s always open and changing other things adapt in that particular part of the economy you’re looking at then you’ve complexity economics. It generalizes standard economics. And the reason we can do that now, by the way, is because we have stochastic nonlinear stochastic process theory, i.e. fancy probability theory, we have nonlinear dynamics, and we’ve computation. We can do a lot by closely specifying these problems with models and then looking at those models in the way you’d look at a lab experiment. Creating a model where there is some sort of interaction that may not necessarily lead to an equilibrium and then watching carefully what happens and the interesting thing, and this is what connects everything with complexity, the interesting thing is you start to see phenomena that you see in other fields. You get multiple possible outcomes, so if you play the experiment once you might head over this way, everybody’s driving on the left side of the road, if it’s left up to chance, or play history again, small events, you might wind up with all cars drive on the right side of the road just by chance. We’re speaking English, or at least I hope so, but we could equally have been speaking French or something if small events had steered us into different paths a couple of hundred years ago. So you see phenomena that you see in physics, or you might see in immunology different patterns coming out, small events steering into different ones certain patterns that repeat, you can’t perfectly predict, but what you can do is study the structures and patterns that come out. And you see plenty of phase transitions, and I’m sure that -- your course takers, I’m sure are very familiar with that, in some regimes the system can go chaotic, in other regimes it settles into some unique attractor, and in the middle, quite wonderful complex things can happen with long correlation lengths and things like that. So not surprisingly in economics we’re seeing a lot of phenomena that you’re seeing in other fields.