Paul Krugman is the picador:
...Whenever someone steps up to declare that "Keynesian" economics is logically and empirically flawed... you know what comes next: a series of logical and empirical howlers — crude errors of reasoning, assertions of fact that can be checked and rejected in a minute or two. Levine doesn’t disappoint. Right at the beginning... he says:
Now suppose that the phone guy suddenly decides he doesn’t like tattoos enough to be bothered building a phone.
OK, stop right there. That’s an adverse supply shock, and no Keynesian claims that demand-side policies can cure the economy from the effects of such shocks.
If you have a harvest failure, deficit spending can’t put the crops back in the fields. But that’s not what happened to the world economy in 2008, or in 1930.... So it’s the usual.
Meanwhile, on the empirical side.... Levine [is] actually [an] anti-monetarists too, although [he] may not realize it; [his] whole beef is with the idea that... pumping up demand in any way, monetary or fiscal, can ever be helpful. And they invariably live under a strange delusion: that the empirical evidence supports their position. This was never really true...
And Chris Sims is the matador:
...and also that the Federal Reserve could not, because of delays in data availability, respond within the quarter to changes in output or the general level of prices. The model also included an attempt to identify a money demand equation.... The effects of monetary policy identified this way were quite plausible: a monetary contraction raised interest rates, reduced output and investment, reduced the money stock, and slowly decreased prices. The effects on output of unpredictable disturbances to monetary policy were non-trivial, but accounted for only a modest fraction of overall variability in output....
This pattern of results turned out to be robust in a great deal of subsequent research by others that considered data from other countries and time periods and used a variety of other approaches to SVAR-style minimalist identification. A summary of some of this research appeared in Leeper, Sims, and Zha (1996). It was widely accepted as a reasonable quantitative assessment of how monetary policy changes affect the economy...
Krugman sums up:
By any normal set of intellectual criteria... the evidence that monetary shocks have real effects was and is overwhelming, and it’s very difficult to write down a model in which this is true but in which fiscal policy is never effective at least [at the zero lower bound]. The spectacular success of liquidity-trap predictions these past 6 years is just icing on the cake. To understand why anti-Keynesian delusions persist, then, we need to turn to other social sciences, and try to make sense of the sociological forces that keep these delusions alive.
It is very hard to know how to properly engage with somebody like Levine. His arguments are not coherent. He has not done his homework. And he thinks that reasoning with the mighty power of his own brain, without consulting the theoretical and empirical literature since Say, Ricardo, Malthus, and Mill first clashed about this from 1800 to 1830, he can come up with eternal truths.
One does need to point out that Levine's arguments are not coherent, and that he has not done his homework, and that we are embarrassed that he is in our discipline. I am not sorry that that makes him angry. It is the truth. And making him angry is, after all, the only thing that might induce him to mark his beliefs to market.
What one might seek to do is to construct a smarter David K. Levine--to ask and answer the question: "What would David K. Levine write if he had done his homework and was coherent?" But, as Daniel Davies pointed out on a related occasion back in 2003, to transform the writer's argument to make it smart, coherent, and homework-based is fundamentally false to the writer: sort of along the lines of writing the poems the ancient Romans would have written had they been Scottish.
So the only road we can take is the sociological road. We can try to answer the question: "What is it about the social matrix in which Levine is embedded that would lead him to claim that production and employment are invariant to monetary policy in an economy characterized by a lot of frictions and sluggishnesses of price adjustment?"
What would answers to that question look like?