Thermal Macroeconomics: An axiomatic theory of aggregate economic phenomena

Hello everyone, my name is Navid Hakimi and I'm very glad to be joined here by Professor Chatter and Dr. McKay from Warwick Institute and Mathematics Institute in the UK. They're going to talk about thermal macroeconomics, a thermodynamic approach to economics. So the first part of the talk would be them going through the presentation and then hopefully we're going to have time to ask some questions and ask their opinion about the mapping between their concepts and active inference and also more maybe general questions about macroeconomics. So very glad and very excited for the talk and I give the mic to Dr. McKay. Right, so in fact, Nick here, I'm going to start off with a brief sort of conceptual outline and then Robert is going to give a little more detailed technical summary. And I should say that I wish I was at the Mathematics Institute at Warwick because it's a very grand place to be, but I'm not that mathematical. So I'm in the Behavioural Science group at Warwick Business School and a Behavioural Scientist by background. And in fact, it's probably a good place to start actually because an interesting question for people interested in economics from a behavioural point of view is how do we make sense of economic theory as it's currently practised, which assumes, for theoretical reasons, extremely rational and sophisticated agents often have perfect knowledge and perfect calculating abilities and they do perfect Bayesian inference and they'd maximise some utility, each one. And then we put those agents together and explain aggregate phenomena as a kind of complicated outcome of all those extremely rational agents. And the behavioural spin on that is always, well, yes, but people aren't really like that, are they? And you get all kinds of irrationalities at the individual level and perhaps some of those lead to a group level of rationalities as well. So, for example, booms and busts through herding and so on. But the basic problem for trying to found an economic theory is that if it turns out that the agents that we're working with are much, much more heterogeneous and complicated and not as rational as in standard theory, does that invalidate or at least throw into question standard economic theory? So an interesting strategy to assuage that worry, which is roughly where Robert and I have been going, although there are a variety of motivations you might use for this approach, but this is one of them, is to say, well, hang on, at the aggregate level, maybe economies can work in quite a predictable way, even if the individual elements within them are extremely complicated and heterogeneous. And in fact, we sort of have to believe that because real economists don't really think that players in markets are perfectly informed about all possible outside options and have a perfect understanding of the implications of all their choices. And it's just clearly not the case. And yes, at least on the face of it, basic economic principles seem to be reasonably applicable to the real world. So it seems like you have to really have some belief that there's a way of understanding the aggregate phenomena without understanding micro foundations in any great deal, because they're just too complicated to understand. And then a natural parallel that comes to mind, it's not come to mind to us uniquely, because this has been something that people saw about forever, at least a century, probably more is the thought that well, in physics, you can say all kinds of aggregate things about, for example, the temperatures and pressures and so on, without really understanding micro foundations terribly well, indeed, of course, classical thermodynamics arose before people had any real idea about micro foundations, and I think weren't too sure about whether the atomic theory was correct, and really basic things like this. So it turns out to be possible, at least in physical context, to have an aggregate theory without really understanding the micro…