Rationality is a modelling assumption, not a diagnosis.

Listen, sunshine. It does not matter whether or not someone is making a rational choice, because rationality has nothing to do with rationality whatsoever.

Yet another book jacket blurb advises that ‘choice architecture’ or some other form of ‘market compliant paternalism’ is needed because people do not make ‘rational’ choices. Unfortunately, the next adoring quote on the jacket was not Hume saying “Commit this book to the flames because it contains nothing but sophistry and illusion.” That’s a shame, as no economic concept is more misused than the economic concept of rationality. So here it is straight:

Listen up, people! Folk take decisions on all kinds of bases and in all kinds of ways. They’re romantic, impulsive, ornery, all kinds of states that might mean that they don’t make the ‘best’ choice available. It actually doesn’t matter how they take the decision, as long as they are able to take the decision in the first place (i.e. they have agency and choice). That’s when we come in – we can see a decision and want to explore its implications. But because we aren’t carrying around psychiatrists or brain scanners or whatever, we don’t have any way to reconstruct their decisions for the sake of modelling them.

So what do we do? We use a set of axioms for modelling preference that you can take together as being called ‘rationality’ in the economic sense. Because they are used for modelling they are stated mathematically – and no, I don’t think there’s a single one of us who actually believe that human decisions are actually best viewed as mathematical constructs. Those things ensure consistency when you are doing modelling so that you can at least infer something about the decision. They do not say a single thing about how the decision was -or ought to be – actually arrived at.

The problem is that rational has another meaning, to do with some psychobabble stuff allegedly. This rational is actually not necessarily economically rational. A person who always made rational economic decisions would actually be fairly irrational to a normal person who’d find it odd that someone could be so creepy about the way they choose – we’d be talking about Spock or Lecter, not a normal human here. (And in fact the overwhelming evidence is that people take decisions only when there is some emotional component to the decision, and that people who don’t associate emotion with decision actually become indecisive!)

But the two rationals are not at all alike. One, psychological if you like, is a diagnosis based on the body of literature created by psychologists to determine who is aberrant to better enable their social punishment, and the other is a construction used to get a sensible answer to difficult questions involving understanding decisions whose dimensions you don’t fully understand. Eliding the meaning of the two, or applying the latter to normative questions about what people should choose (as opposed to what they do choose) is not even wrong, it’s creepy (You don’t make good decisions, so I’ll take your control over them away. That’s actually how abusers operate, sheeple.)

So in short, it does not matter to an economist how a person takes a decision, and nothing in the construction of the decision is of any here or there importance to us. Only the results are observable, and the mathsy bit is there to help analyse that. Rationality is not a justification for overruling a person’s decision, just an assumption that allows any modelling to be done at all.

Thank you. You can go about your business.

(The author is the author of Business classics “What Fooking Bastard Ate My Cheese, I’ll have the Twat” and “Achieving a New Reality: Leadership Lessons from Phillip K. Dick”)