[nl]

That's the Keynesian view where you can move a lever (cause) and measure the changes (effect).

Thats not Keynesian economics at all! Or rather it's a view that people from schools of thought completely opposite to Keynesian also believe.

Does anyone not think this? Eg, Quantitative Easing: does anyone (Keynesian or not) think it does nothing?

[caseysoftware]

Everything does something but it's hard->impossible to directly map cause to effect. The problem is there are numerous causes and effects all playing out at once and they all interact.

Or to put it another way: it's the three-body problem extended to millions->billions of bodies which use psychology+perception as the main force instead of gravity.

[maerF0x0]

also there Nth order effects which are confounding subsequent moves. This is why any idea that a certain party of president leads to good/bad economies is nearly impossible to tell, it could just be a lagging indicator from previous one(s)