Mises did indeed attest that the empirical method (positivism, or the scientific method) is not the correct methodology for reaching reliable conclusions on matters of economics. However, he did not reject the scientific method. He merely pointed out that the empirical method is ideal for matters of the natural sciences because when it comes to inanimate objects results are predictable. This is because the natural world is deterministic. Copper always melts at 1,085 °C. That which does not melt at 1,085 °C is not copper. Human beings, however, have differing values, tastes and information all of which strongly influence their decision making. Will a person buy a hamburger at half price? Certainly many more will than at full price, but for example I will not because I am a vegetarian. Some people will buy twice as many. It all depends on their individual ideas, expectations and preferences.
When it comes to matters of economics there is no way to set up an experiment and control for all factors as we can in a laboratory. We cannot, say, increase the minimum wage in one region to measure the effects on overall employment and compare it to the effects in another region where it is not increased, firstly because the implementation (or lack of it) in one region will also affect the behaviour of individuals in the other (some may move to take advantage of higher wages, others may stay to take advantage of cheaper labour costs.) Secondly because an economy is complicated, and thousands of other factors would also be at play in both regions which cannot possibly be controlled for. We cannot say (for example) that just because employment rose in one region after a minimum wage increase that this change was due to the minimum wage increase, only that it happened. Because we have no counter-factual there is no way to tell if employment would have increased even more if the minimum wage increase had not been enacted.
Thus, Mises attested, a more suitable method of approaching economic questions would be necessary than the empirical one, based upon the foundation of certain axioms about human nature which could be reached by introspection (or thymology as Mises labelled it to distinguish from different uses of the term introspection) In the interest of brevity and simplicity we only need to consider three of these; and I will explain why each should be uncontroversial.
The first is that humans act. This claim is hard to argue against because the very act of doing so would be to make a performative contradiction. The second is that when people act, they do so in order to substitute a state of affairs which they consider preferable for one they consider less preferable according to their own values. Again, I see no argument against it; if one is completely satisfied with their situation as it is they do not but stop. Even the attempt at meditation or mindful acceptance of the present moment is based on the impulse to substitute a pleasant state of surrender to circumstances outside of ones control for a less pleasant state of resisting them. Finally, I add, that people respond to incentives. This axiomatically true because what people do not respond to can hardly be considered an incentive – by the very definition of the word.
(To anticipate a potential objection I draw an analogy; some people might scold a dog for poor behaviour in the hope of stopping that behaviour, without realising they are reinforcing the behaviour because the dog may enjoy negative attention better than no attention whatsoever. This does not mean that the master is failing to incentivise the dog; it just so happens that he is not incentivising the dog to do what he thinks he is.)
An acceptance of these three axioms (for which their is unlikely to be any disproof) reveals that it is incentives which guide economic decision making among individuals. This we all accept, if not in our conscious philosophy, than by our actions. To illustrate with our earlier example of the half-price hamburger, the business owner offers this deal in the hope of selling more burgers by creating knowledge of his product. (One might argue that not everyone will respond to incentives, illustrated by the fact that I will not buy the burger because I’m a vegetarian – but in fact in my case the incentive is simply not great enough to suit my values; were the price of the burger was -$10 I might well take the money and give the burger away to a homeless person.) Even our policy-makers are led by their unconscious acceptance of these axioms, for example, they may put a tax on alcohol do discourage its consumption, while placing a subsidy on a solar panel to encourage its use. These are not policies guided by the empirical method which is better suited to the natural science, but by the aprioristic understanding that humans act, and in their actions they respond to incentives. That is the method, named praxeology, which Ludwig von Mises advocated.
Source: Seeing the Unseen