Let's say I notice that a screw is loose on a handle of a pot. This often happens with one particular old pot we have that we love. So I go get a Phillips screwdriver and tighten the screw. Would you say that the screw is a tool? Or would you say that the screwdriver is a tool? I say that the screw is the object I'm trying to affect and that the screwdriver is the tool.
What does this have to do with economics? Here's what. On a blog post recently Scott Sumner writes:
Because central banks are used to using short term rates as their primary policy tool, policy may well become sub-optimal once rates hit zero.
But interest rates can't be a tool. They're an objective: they're like the screw on our old pot. They might not be the ultimate goal. But fixing the handle to the pot isn't the ultimate goal either. The ultimate goal is to use the pot to cook something. I see this a lot in the writing of monetary economists: it's a basic category error. It's hard to believe that making such an error won't ever lead to other errors in thinking.