I would say a system which continues to exist, despite predictions it should fail, is more effective than the model which predicts it failing. You and I may think that growing the pie is better than turning the screws, but the system doesn't care. It just does what it does.

Maybe I'm wrong here, but I'm reading many of the comments as: "The goal should be X, managers do Y, so managers should stop" when the goal is not X.