In this blogpost, I will explore a fairly recent school of economic thought: complexity economics.
Complexity economics is the study of economic systems as complex systems. Complex systems are systems which consist of interacting individuals that change their actions and strategies in response to the outcome they mutually create. Complexity economists study the emergence of structures and the unfolding of patterns in the economy. Since the 2008 financial crisis, when mainstream equilibrium models had little to add in terms of policy direction, there has been increasing interest in using ideas from complexity theory.
Complexity economists view the economy as a complex system which consists of, belongs to, and overlaps with other complex systems.
In this system, economic patterns such as economic growth and inflation are classified as emergent phenomena because they emerge out of the interactions of heterogeneous agents with heterogeneous expectations. A broader definition of an emergent phenomenon is that it is a “new” pattern which arises as a result of the interactions of a number of elements between which simple relations exist but which cannot be reduced to the particular properties of each of these elements.
Complexity economists such as argue that non-equilibrium is the natural state of the economy. The economy is always in a state of flux, constantly evolving and changing. There are two main reasons for this. One is fundamental uncertainty, the other is technological innovation.
The concept of (fundamental) uncertainty contrasts to that of risk. In the case of risk, all possible future events or consequences of an action or decision are known. Therefore, we can calculate the probability that this event will actually materialize. However, there are many situations in which we do not know all possible outcomes. In these situations of uncertainty, probability calculus has no sound foundation.
To deal with uncertainty, economic agents try to make sense of problems by surmising, making guesses, using past knowledge and experience (Arthur 2013). As a consequence, agents continually update their internal decision making model, which means they constantly adapt, discard and replace the actions or strategies based on their experience as they explore. Such dynamics are also known as ‘evolutionary dynamics.’ Evolutionary models usually do not end up in a stable Nash equilibrium in the presence of noise. If these results can be extrapolated to the broader economy, it should be permanently in disruptive motion as agents explore, learn, and adapt.
Technological innovation is the other important contributor to the economic system’s state of permanent flux. The nature of innovation is such that technological development enables further technological development. It follows that a novel technology is not just a one-time disruption to equilibrium. Instead, it is a permanent ongoing generator and demander of further technologies that themselves generate and demand still further technologies. Thus, technological innovation also contributes to the state of flux, be it somewhat slower than uncertainty.
That being said, the state of flux is often relatively stable and can therefore be approximated by an equilibrium or steady state. In fact, complex systems are often characterized by multiple equilibria, especially in the presence of positive feedback or increasing returns.
The steady state a complex systems ends up with depends on the path towards that steady state. In other words, it is path-dependent. What is more, tiny changes in initial conditions might cause the system to end up in a radically different steady state. If this is the case the system can be classified as chaotic.
Once a system ends up in a steady state, it might not be straightforward to move to another steady state. It might be so resilient to changes that it takes considerable shocks for it to move to another regime. This is also known as a lock-in. On the other hand, if a system’s resilience is decreasing, it might reach a tipping point and suddenly change behavior or move to another regime. Financial markets and economies have historically exhibited sudden and largely unforeseen collapses, at a systemic scale. Such phase transitions may in some cases have been triggered by unpredictable stochastic events. More often, there have been endogenous underlying processes at work.
Summarizing, complexity economists view the economy as a complex system. Aggregate economic phenomena are viewed as patters emerging from the interactions between heterogeneous agents. While economic systems can be in relatively steady states which can be approximated by an equilibrium, the presence of uncertainty and technological innovation ensures that all economic system are in a constant state of flux. What is more, any relatively stable states which do emerge are often not unique, path dependent and, sometimes, even chaotic. On top of that, economic systems regularly go through phase transitions to end up in a different state.
Complexity economics was partially developed to contrast the prevailing neo-classical economics paradigm. It is primarily different from mainstream economists in its ontology and methodology.
In thinking about the economy, the ontology, mainstream economists and have come to rely heavily on the concept of equilibrium. The main difference between complexity economics and the mainstream is the focus on equilibria–static patterns that call for no further behavioral adjustments. Complexity economics portrays the economy not as deterministic, predictable, and mechanistic, but as process dependent, organic, and always evolving. Still, equilibrium economics is not discarded in its entirety. From the perspective of complexity economics, equilibrium economics is a special case of non-equilibrium and hence complexity economics.
Furthermore, complexity economists have been vocal critics of the mainstream rational expectations hypothesis, which assumes that economic agents know the model of the economy and on average take its predictions as valid. Complexity economists believe this is so unrealistic that it potentially invalidates mainstream model outcomes. In some cases, expectations are self-referential: economic outcomes depend on the expectations of agents today. For example if agents want to determine whether or not to go out to a bar. Their decision to go depends on their expectations about how crowded the bar is. If they expect it to be crowded they will stay home and vice versa. Agents learn about the actual crowdedness of the bar the day after – even if they stayed home. In this case, no equilibrium of bar attendance will ever emerge. Instead, it will fluctuate because of the negative relationship between expectations and attendance. Not only does assuming equilibrium not hold in this case, it would fail to predict the fluctuating bar attendance. This is the famous El-Farol bar model.
Given this difference in ontological focus on equilibrium, it is not surprising that, complexity economists tend to use non-equilibrium modelling methods. Furthermore, the methodological approach of complexity economists in more inductive. Formal models are almost always made to explain a set of observed phenomena or stylized facts, which can then in turn be used to inspire further empirical work. While shifting, this is still not always the case in mainstream economics.
This, in a longer form, was first published by Exploring Economics