Evolutionary economics and game theory

Like the agents they study, evolutionary economics is highly heterogeneous. Models are ad-hoc and serve as heuristic guides to specific problems. This is similar to theoretical biology, where evolutionary models are independent of each other. Even the general theory of inclusive fitness does not provide a non-controversial unifying framework. Although there is no single framework, evolutionary economists are united by four main assumptions about the world:

  1. The world is constantly changing. Qualitative change is common, and fundamentally different from the
    quantifiable gradual change that can be studied with standard equilibrium approaches
  2. The generation of novelty is an important agent of economic change. The generation of this novelty is fundamentally unpredictable.
  3. Economic systems are complex systems; emergent properties, and non-linear and chaotic interactions put fundamental limits of prediction. Generation of novelty and complexity make evolutionary change irreversible.
  4. Human institutions and social arrangements emerge through self-organization and undesigned order. However, there is no agreement on if market or the emergence of state are more fundamental.

Nelson and Winter, the founders of modern evolutionary economics, define a theory as “a tool of inquiry” and distinguish between two types of theories: appreciative and formal. An appreciative theory is characterized by a broad process of analysis and understanding, with a ‘focus on the endeavour in which the theoretical tools are applied’, including engagement with empirical data. In contrast, a formal theory focuses on “improving or extending or corroborating the tool itself”.

For me, this definition of theory is not consist with what I consider the usual use of the word. In common language, or in the precise sense of Popper, a theory is something that can be invalidated or falsified. A tool cannot be falsified, it can only be bad or inconvenient for a task. Thus, in my distinction between theory and framework (or maybe in Kuhn’s words: paradigm), Nelson and Winter’s definition would fall under framework. However, it does not fully specify a framework, but only the process by which the framework is used. Hence, I would refer to what they call ‘theory’ as a process paradigm. We then have two categories of process paradigms: the appreciative process and the formal process.

If I understand Nelson and Winter correctly, then a scientist practicing the appreciative process is primarily concerned with describing a specific phenomena, or answering a specific question. She pursues this goal with any tools at her disposal only with the constraint of matching or fitting whatever is considered as empirical fact in her paradigm. This pragmatic approach reminds me of two very distinct groups: people that provide verbal descriptive theories (say much of psychology) on the one hand, and engineers on the other hand. I find it curious that such different fields would fall under one roof, and I think a further distinction of the appreciative process paradigm is needed: descriptive versus predictive.

A scientist with the formal process paradigm is concerned with building, connecting, refining, and testing theories. I find myself usually in this camp; I am primarily concerned with theories for the sake of theories, how they relate to each other, and if or how they can be best tested. Of course, most frameworks combine an element of both appreciative and formal process paradigms.

For Hodgson and Huang (2012), evolutionary economists primarily focus on the appreciative process and evolutionary game theorists on the formal. This marks some fundamental incompatibilities between the two approaches, but the authors believe them to be reconcilable. They suggest that EGT models can be used within EE as heuristic guides.

I think that Hodgson and Huang’s (2012) main fault is not acknowledge the power of abstraction. The primary skill of a computer scientist is knowing how to look at a problem and figure out what central concepts need to be preserved in an abstraction and which can be ignored. Once an abstract theory is present then whatever results are necessities in that theory will translate down into any concrete models. This allows you to study “simple” abstract models and establish truths about any model that has to embed the abstraction as a part of it. It gives you a way to study possible theories.

Unfortunately, little of EGT is done by computer scientists. Most models only relate to others through a verbal theory, and as such general abstract statements are difficult. However, I think there are families of models (say games on graphs) that could lend themselves to abstract analysis. If EGT pursues this avenue then it will be an extremely valuable tool for EE; it will allow them to make general abstract statements and prove theorems of the sort that are usually purvey of neoclassical economics.

ResearchBlogging.orgHodgson, G., & Huang, K. (2010). Evolutionary game theory and evolutionary economics: are they different species? Journal of Evolutionary Economics, 22 (2), 345-366 DOI: 10.1007/s00191-010-0203-3

About Artem Kaznatcheev
From the Department of Computer Science at Oxford University and Department of Translational Hematology & Oncology Research at Cleveland Clinic, I marvel at the world through algorithmic lenses. My mind is drawn to evolutionary dynamics, theoretical computer science, mathematical oncology, computational learning theory, and philosophy of science. Previously I was at the Department of Integrated Mathematical Oncology at Moffitt Cancer Center, and the School of Computer Science and Department of Psychology at McGill University. In a past life, I worried about quantum queries at the Institute for Quantum Computing and Department of Combinatorics & Optimization at University of Waterloo and as a visitor to the Centre for Quantum Technologies at National University of Singapore. Meander with me on Google+ and Twitter.

3 Responses to Evolutionary economics and game theory

  1. Pingback: How teachers help us learn deterministic finite automata | Theory, Evolution, and Games Group

  2. Pingback: Learning and evolution are different dynamics | Theory, Evolution, and Games Group

  3. Pingback: Cataloging a year of blogging: from behavior to society and mind | Theory, Evolution, and Games Group

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s