Re: Business & Media Institute – 2/2/2009 – “’Bulls & Bears’ Guest Can’t Find Example of Stimulus Success” – Lauren O’Reilly

One of the most striking features of the social sciences is that there is never an example of any action that can be shown to have definitely worked or failed to work in solving a problem. The reason for this is that science requires controlled observation of a statistically significant number of solutions applied to essentially the same problem. Time cannot be rolled back to try alternate solutions to the same problem and not enough essentially identical situations can be found to constitute a statistically significant collection for inference.

FDR’s actions did or did not help us through the depression. One can argue that it did (the depression would have been even worse without FDR’s intervention), it didn’t (the depression would have ended on its own without FDR, whose actions actually prolonged the misery), or any host of alternative conclusions (my favorite being that FDR’s policies didn’t work because they weren’t large enough in scope to cure us of depression --- only WW-II was a large enough government action to end the depression). None of these theories has any real evidence to back them up.

Not only the social sciences lack true scientific testability. The theories of the origin of the universe also belong to this class of testless science --- while the basic theories underlying our beliefs about the origin of the universe are testable, there is no way to prove (or disprove) that the universe was actually created according to the predictions of those theories.

In cases like these, belief is based on faith, not testable scientific conclusions. Some political economists see government action as the cause of all good, others see government action as the cause of all bad. Most realize that government has both good and bad influences on the economy. A few realize that there is not a single example in modern (and probably ancient) history where either government or non-government actions were absent, so the best science has to offer is general guidelines on where more or less government action seems correlated with good or bad results --- these are scientifically weak results hardly rising to the level of pronouncements of success or failure in any particular case.