I just came up with the idea that all government employees should be nihilistic and that policies should be made exclusively by game-theorists and econometricians.
My arguing goes as follows:
One has to take, the prisoners dilemma (look it up if you don't understand the picture):
As one can imagine in this system it would be best to defect as average gain would be (5+1)/2 =3 whereas if one would cooperate the average gain is (3+0)/2=1.5.
However, while this is greatly effective if the game was played once on average the chances of success would decrease over time because the other player would predict that you would defect and would do so constantly himself as well, leading to a gain of 1 for every game.
Game theory has actually solved this problem in practice. Around 1980 there were several competitions held between mathematicians to write computer programs that would have to play out the prisoners dilemma an indefinite number of times against all the other programs and come out with as much gain as possible. Obviously the programs that would cooperate too much wouldn't gain too much because they would not gain a lot against a program that would defect most of the time eventhough they'd make a fair bit of gain against other programs that would predominantly cooperate. On the other side a program that would defect most of the time would make very little gain if they'd met another program that would defect most of the time. But what program was most effective?
The winner was a surprisingly simple program called 'Tit for Tat'. It would always start with a cooperation. Then every time the other program would cooperate as well the 'Tit for tat' program would reward the other program by cooperating again, but if the other program defected the 'tit for tat program would defect the next time as well as to punish the other program. However, forgiveness was given as soon as the other program would do a round of cooperation again. This way, against a predominantly defecting program, the 'Tit for tat' system would probably perform marginally worse, because of the initial compliance, and against a predominantly cooperating program 'Tit for tat' would perform equally well because both would get a gain of 3 most of the time.
In short, this website presents a list of strategies that were developed. Up until 'Tullock' they are placed in order of effectiveness. In another, bigger study, with more programs 'Tit for tat' came out best once again, despite being one of the simplest programs. Just as surprisingly, 'nasty' programs in general didn't do nearly as well 'nice' programs.
And therefore I plead, Nihilists as government so we get rid of pre-existing ideologies and econometricians and game-theorists as policy makers so that we may find more of these fundamentally fair and highly effective strategies for real life situations.
If you found the talk about 'Tit for tat' interesting, I suggest to you that you watch the 1986 BBC horizon documentary "Nice guys finish first".