I don't want to be labeled as the great defender of Keynes. But to be fair, he recommended counter-cyclical policy, not deficit spending bubble pumping that has been enacted in his name since.
That is one of the problems with the dismal science, Keynes was a great and intelligent guy in his own right, and very well might be accurate in everything that he intended to communicate about what he understood about Economics; but then the failures of his students to understand (and therefore implement) his ideas are still Keynes own failures. It's like saying that the Bible is the perfect word of God, and it's humans that screw it up; although it could be true taken alone it makes no difference for the outcomes. Praxeology (and thus Austrian economic theory) views human interactions (including economic ones) to be a social science, and studies it from that perspective/paradigm. It makes few promises about predictions, because humans are too complex to break down their actions into mathmatical algos, there is always going to be data that cannot be represented in such simulations. It's like those 3D computer simulations of flocks of birds. Sure we can model the mathmatics to mimic such behavior in a computer simulation, what we cannot do is use those mathmatics to predict how real flocks of birds would actually act in the real world. The reason for this is that the simulations are simply mimicry of real birds, and must make numerous assumptions about the decision making processes of any given flock of birds. Just as this decision making process is a complex and rapid group communications process that we can't come close to gathering enough data to model, the collective decisions that millions of independent human beings make each day can't be simulated either, for the simple fact that it's impossible for the observer to know all the data that all of the individuals know. Austrians tend to refer to this principle as a 'fatal conceit'.
