Wherefore art thou, rationality?

I’ve been bombarded recently with people telling me about economists’ perception of rationality and the wonders of behavioural economics. The term homo economicus gets thrown around with gay abandon as a generic criticism of economics. Oliver Woods claimed that rationality means having perfect information and being entirely self-interested. At the other end of the scale, Will Wilkinson extends rationality to include anything that’s “…the best we can do given our numerous limitations.” Tim Harford goes as far, in his book ‘The Logic of Life’, as suggesting that someone can be termed rational if they respond to incentives. So, if even economists don’t agree on what rationality is, how can we complain that economists are silly to speak of humans as rational?

The starting point for a definition of rationality in economics comes from textbooks. Microeconomics lecturers teach us that a person is rational if they act as if they know what they like and try to get more of it (I don’t want to get in over my head with technical details here). That, in itself, seems fairly reasonable (if we ignore satisficing models). Unfortunately, that sort of rationality doesn’t get us very far in explaining anything or constructing models of behaviour.

In order to say something about peoples’ actions we need to make further assumptions. For example, we might assume perfect information, perfect foresight or that people have unlimited and costless computational power. These assumptions were initially made to simplify the modelling process. They then became so ubiquitous in economics that people started to identify them as a part of the idea of rationality. Since the pioneering work of people like Daniel Kahneman and Herbert Simon, economists have started to relax these assumptions and replace them with substitutes that are based upon experimental observations of human behaviour. That is what is often called behavioural economics and we now term that version of rationality bounded rationality.

So rationality, as a concept, means different things to different people. To some economists, it means what the public think of as homo economicus. To some it means only that we do what’s best for us, given that we’re inherently imperfect. What’s important is the particular set of assumptions that one makes when modelling a problem or process. Rather than attacking some straw-man conception of rationality, look a little deeper and attack the assumptions and arguments themselves.