Rational Choice Robots Are Real (and they are us)

One of my pet peeves is when characters in movies don’t behave like real people.  I don’t mean when people teleport, or fly, or grab onto the landing gear of a moving plane – I love sci-fi, action and fantasy.  These things don’t make sense in our world, but can be perfectly consistent with the rules of the world in the film.  What gets me is when the storytellers try to convince us that the characters are normal humans, just like you an me, regardless of whether gravity is the same, yet they behave completely irrationally.

I don’t mean irrationally as in crazy, or emotional, or too short-sighted; real people do that.  I mean in ways that do not benefit them in any way, even according to what we are supposed to believe is their own subjective valuation.  Movie villains are notorious for this.  How often are they portrayed as hell-bent on a particular goal, only to stall just before realizing the goal to deliver some one-liner.  If the character is supposed to crave attention and the recognition of being witty above all else, this might make sense, but that’s not often what we’re supposed to believe about them.  How many movies portray business tycoons as so greedy for more money that they do things that will guarantee them less money?

I have a hunch that this unrealistic portrayal of humans in extraordinary circumstances stems from an unrealistic understanding of humans in ordinary circumstances.  There seems to be a failure among writers and the general public to appreciate the power of rational choice theory in explaining human behavior.  Too many people do not see everyday actions as the result of rational self-interest.  If you don’t think people pick a brand of black beans based on rational choice, then why would they pick a method of murdering their nemesis any more rationally?

Critics of rational choice theory correctly point out that real people are not robots that calculate their projected “utility” with every decision.  This critique is correct insofar is the calculation is not often conscious nor does it involve anything called a “util”.  But calculations of self-interest do occur all the time, and in dizzying complexity.  To take a well-documented example, when seat-belt laws are enacted, injury from accidents tends to increase slightly.

Economist use a rational choice model to explain that humans comfortable with a certain level of risk will maintain that level and if you make them wear seat-belts  they may compensate by driving more dangerously.  I doubt anyone consciously gets into their car after a seat-belt law is passed and says, “I’m comfortable with a 0.7% chance of death every time I drive, and this damn seat-belt lowered it to 0.5%.  I’m going to speed up!”  Still, we make subtle, marginal calculations like this all the time without realizing it.  If you crave a candy bar you may be willing to get in your car and drive at night through the rain to get one – an act that dramatically increases your risk of injury.  But if it’s snow mixed with rain, you may deem it not worth the risk.  A very minor change in risk alters your decision.

It is impossible for an economist to get inside anyone else’s head and understand their subjective preferences or what creates them.  It’s hard enough to understand what generates our own.  Absent some telepathic ability, we are left with few tools to analyze the decisions people make and the norms and institutions that result.  We could assume that people behave in completely irrational, unpredictable ways and have no expectation of generating a desired result when they take any action.  But then, why would someone act if they did not believe, whether rightly or wrongly, that acting would yield some benefit (as defined by them, not some objective standard)?  An assumption of rationally self-interested individuals is the only tool available to us, and it turns out it’s amazingly powerful.

Since value (in the economic sense anyway) is subjective, calling someone a rational chooser does not mean they have no emotions or impulses.  These things are part of what form their judgements of what is valuable.  Neither does it mean people will always choose the most effective means of getting what they desire.  No one is perfectly informed and we often have false ideas about cause and effect.  What we value also changes over time and with circumstance.  All it means is that, given the knowledge and preferences we have we will choose the actions that we perceive as yielding the greatest benefit (subjectively defined) at the time of choosing.  So simple it’s often dismissed as an unhelpful truism.  But how many times do we forget this and assume all kinds of far-fetched things about people’s choices instead?

When we apply this model, even some of the most bizarre behaviors can start to make at least some sense, given the constraints (real or perceived) and context.  (Check out some of the work of Peter Leeson).  This can help us see the root causes of behaviors and social phenomena and, if we so choose, attempt to change the cost/benefit equation and alter the situation.  Without this model we are left making wild and arbitrary guesses and claims about what motivates actions and we pursue remedies that do nothing to change the underlying incentives.  “They hate us because we’re free”, for example, might make one feel better, but it doesn’t offer any valuable insight into why someone rational enough to plan for months would conduct a seemingly irrational act like a suicide bombing.

I love a tall tale at the movies, but it would be a lot better if writers kept their characters actions consistent with the motives and values they are supposed to have.  Understanding real human behavior is the first step in creating pretend worlds that really capture the imagination of real world viewers.

%d bloggers like this: