Russ Roberts intereviewed Steve Fazzari, who is, apparently an actual Keynesian. What this means, of course, is that he is very wrong, but it is interesting to listen to because this particular insanity contains its own strange logic. There are two things in particular I found interesting.
Keynesianism ultimately rests on a 'sanctification by group,' assumption, because, while what Fazzari suggests may sound reasonable on it's surface when he is talking about a large number of people, it is immediately obvious it is nonsense if one were to talk about one person. When one person saves it can be trying and painful because saving involves self-discipline and self-denial, but as one person saves that one person builds up those savings until at some point he has enough and he can alter his savings versus consumption ratio. Keynesians have turned the old phrase completely on it's head, and can't see the trees for the forest. In order to even think of increasing demand by government spending you have to assume the economy is some sort of discreet entity and that you can ignore the humans who had incentive to stop demanding so much in the first place.
The second glaring bit of obviousness is the massive limitation of falsifiability. We are, again, in the middle of a massive and painful experiment WHICH WILL FAIL, primarily because there are too many variables. Since we simple cannot hold all but one constant in order to do our tests, inevitably there will be some idiot who insists the reason it didn't work the last time was for some other reason than that it doesn't actually work. The logical outcome of an extreme commitment to falsifiability is suicide. Just analyze any pre-existing assumptions you have (which you must have if you were going to do any testing in the first place) and regress back into a massive refusal to 'know' absolutely anything, test everything- at which point there are many fatal activities just begging to be tested.