TBH, I've always found the criticism that economics "can't predict" to be a bit strange. I don't know many modern economists who claim that our models are highly predictive. Most think of them as helping us to understand better, and thus make better predictions. What helps us make better predictions is the better understanding, not that we follow the mathematical slavishly. And even at places like the Fed, that's how it works. Model results are only one input among many, including human insight and experience. Feels like they're attacking a straw man.