[tolmasky]

I think you're right without realizing it: the actual criticism is that the onus of proof is on the one proposing the program in question. If there is a random government program X (and I'm not referring to welfare, it can be any program), it should not be allowed to continue without sufficient proof that it is making a positive difference. The reality is that most programs are not backed by much evidence since such evidence is really hard to get or even impossible. We can't really scientifically test how things would happen without program X (and all other things being equal) in either direction. I completely agree that I do not know how much program x actually subtly hurts the economy/society/whatever in the same way you cannot know how much it really helps. Everyone's personal intuition guides them to feel that certain perceived positive consequences outweigh certain unperceived negative consequences, and thus justifies their belief that in that case correlation is causation, but there is no proof. As such, the position presented often is that "if you're going to use my money, it is up to you to prove that it will actually be used well, not up to me to prove that it won't be used well," in the same way that if you present a new theory of physics it is up to you to prove it, not the rest of the community to disprove it.

The problem is thus precisely as you described it: this is an area endemically (and perhaps fundamentally) lacking empiricism, and arguably one where you can't practically have empiricism. I don't think it's a stretch to argue that most policy positions are supported on blind faith, emotion, and pseudo-science.

A good example of this, to hopefully gain some common ground with you and move away from an emotionally charged discussion on welfare, is patents. There is really absolutely no proof that patents "encourage innovation". It's not even clear how to measure that. There may be tons of studies done, but they are useless as we have no baseline to compare to. We don't know what's not being invented because of patents. Had patents come about naturally as common agreements between corporations, then it really wouldn't be my business to opine, but since instead it is a government policy, supported through my own tax dollars, courts, and "implict agreement" to not break said patents under penalty of law, it is very much justified that I should demand they go away without myself needing to prove much of anything -- on the contrary it is those wishing to continue the patent system that need to offer proofs.

[barrkel]

My gf is a healthcare economist working for LSE. She analyses and evaluates state (UK) interventions based on the data as reported. There is science being done, and often the data are fairly clear. But we live in a democracy, and there are chunks of the population who believe that state charity is a soft touch. Many times it is democracy and politics itself that forces inefficient outcomes.