For and Against "Toy Models"

I liked a lot of what Scott Sumner had to say in a little rant against Lawrence Ball's argument about NGDP targeting (which is not to say I think Ball is wrong). This part was especially good:

"I make no apologies for ignoring these little toy models, and having my policy analysis incorporate a complex mixture of politics, macroeconomic history, well-established basic economic principles, and logic."

The economy is a complex subject of study and ultimately we bring a lot of strategies to bear on doing economic science, so I like this statement. The only real objection I have is to his point about "ignoring these little toy models". First and foremost, as Bob Murphy points out, most people see Scott Sumner as the pre-eminent toy modeler, theorizing simple relationships between only a handful of variables and building broad conclusions on that. I'd also object because there's no reason "toy models" can't be very important to thinking about the economy. I see theoretical models in economics as clearer explanations of economic mechanisms or processes that we think are important. We really don't have the ability to "model the economy" in a deterministic or totalizing way. But we have a lot of great tools to model processes that are important and that can be useful in thinking about the way the economy works. So I am for all those things that Sumner listed and for a polyglot approach, but against his rejection of so called "toy models".

This was pretty surprising to hear from Sumner too: "Haven’t we seen ultra-low real interest rates for several years? Where is the robust growth?". It was one of those moments where I sat back and wondered if he even keeps up with what other people outside of the NGDP targeting community talk about. In fact, through this whole post he refers to the opposition as "inflation targeters" (as opposed to NGDP targeters)... I suppose that's the big divide he sees in the world, and I guess if you see the world that way those two sentences I quoted might make more sense.