Paul Krugman reads Tony Yates pointing out that John Taylor is, politely, incoherent in his advocacy that his version of the Taylor Rule be legislated in stone to command the Federal Reserve. Either monetary policy is much more powerful than the arguments for Taylor's version of his rule presumes--in which case it is a very bad idea--or the world is much more risky than the arguments for Taylor's version of his rule presumes--in which case it is a very bad idea. There is simply no coherent way to get from the macroeconomic history of the 2000s to the conclusion that John Taylor's version of his rule is a good idea:
...than [John] Taylor-rule enthusiasts imagined, so why impose a rule devised, we know now, by economists who completely misjudged the risks?... Taylor himself... claims that the whole financial crisis thing was because the Fed departed slightly from his version of the rule in the pre-crisis 2000s. But, as [Tony] Yates points out, this assigns an importance to monetary policy that is wildly at odds with the kind of modeling used to justify the rule in the first place... as Yates does not point out... the distinct whiff of someone inventing ever-more bizarre stories to avoid admitting having been wrong about something. This is not the kind of argument on which to base rules that permanently constrain policy.