Politics 1
6. Why are rational game theory models so poor at explaining political behaviour?
Rational game theory models make three assumptions about the individuals they model:
1) Omniscience - people know pretty much everything about the decision being modelled.
2) Rationality - people are completely rational in their decisions.
3) Self-Interest - people seek to optimize the outcome with respect to themselves (this is not universal but common enough to be discussed here).
The problem is these assumptions are commonly violated when it comes to political behaviour.
First: omniscience, as I have argued in previous posts social sciences are not sciences. They cannot be predictive since simply by making the prediction they may cause it to come about. For instance the flawed economics paper which suggested when the debt:GDP ratio of a nation exceeded 90% economic growth would stall caused many politicians and business leaders to lose confidence in nations that exceeded this threshold resulting it them implementing austerity policies and refusing to invest in growing their business both of which retard economic growth. In addition the social sciences can perform experiments to support/refute theories only in small-scale artificial circumstances which are often not close representations of the system they are trying to test. A recent example here was a study comparing the behaviour of university undergraduates (the typical experimental subjects for many social science experiments) to that of prisoners in the classic prisoner's dilemma, which found prisoners were more trusting in the single-trial situation but less cooperative in the multiple-trial game than undergraduates. So when it comes to predicting the outcome of public policies even the academics don't know much for certain. Furthermore most people spend relatively little time studying politics and policies before making their decision, fewer than 1 in 5 people watched the leader debates in the last UK election. Add to that the propensity of political leaders to lie, misdirect, and deliberately obscure facts (which has lead to election-ads being exempt from truth-in-advertising laws in several countries) and the typical individual is knowledge-poor when it comes to politics.
Second: rationality, we as human beings like to pride ourselves in being rational and logical. A pride that is terribly misplaced. Humans are frequently illogical, and suffer from hundreds of detected cognitive biases. Recent work done here in Oxford is even suggesting that rewards can actually change perception.
In general the more important, rarer, and more costly the decision the more rational humans will be. Political decisions are often quite frequent as politics is a common topic of conversation. They are often perceived to be of not much importance since particularly in first-past-the-post electoral systems a single vote often has little effect on the outcome. And finally in the current era of cynicism it seems that politicians/policies have little effect on one's daily life thus political decisions come at relatively low cost. Stacking the deck against rationality. Instead, individuals typically acquire a political stance from their parents/community which is then subject to confirmation bias which emphasizes agreeing information and viewpoints and minimizing those that disagree. Confirmation bias is particularly efficient in the realm of politics because of the great uncertainty and lack of irrefutable evidence on most points as discussed above; leaving nearly equal evidence on all sides of various debates.
Third: self-interest, humans do not act in the narrow self interest proscribed by most game-theory models. Parents will risk their lives to save their children, family members will defend each others' interests, and soldiers will volunteer to die to protect their communities. These are all examples of narrow-sense altruism which is in contradiction to rational game theory models. Evolution demands that humans are in a broad-sense self-interested, that is our genes have survived to make us by acting selfishly to maximize their chances to be passed on to the next generation. Helping one's family members either directly or indirectly by protecting the community benefits the individual's genes. Politics happens at the community level where broad-sense self-interest often causes altruistic behaviours: providing for the poor could benefit the children of the rich in the future if they suffer misfortune. Going to war could prevent future generations suffering under a genocidal tyrant. Loyalty to a party today could bring rewards when that party gets into power.
Thus rational game theory models don't work and are unlikely to ever work for political behaviour. Humans are not selfish rational beings even when the knowledge exists to make an informed decision. When that knowledge is lacking or deliberately obscured, as is the case for politics, there is little chance people will behave in even the vaguest semblance of rationality.
