Ryan Murphy has been helping me think through this one: is there anything that guarantees that a system of free banking is capable of achieving full employment? He's been thinking of it in terms of guaranteeing stable nominal GDP, which I suppose is comparable.
He pointed me to one of Selgin's articles in JMCB, titled "In-Concert Overexpansion and the Precautionary Demand for Bank Reserves" (sorry, no ungated link). I got the impression I was more comfortable with the article's conclusions than Ryan was. Selgin basically concluded that co-ordinated overexpansion wasn't possible because while more conservative banks would win out if any one bank tried to over-expand, a concerted over-expansion would result in increased precautionary demand for reserves in the entire system, which is contractionary.
This is related to the initial concern about free banking that I posed to Ryan, which was also related to the fact that free banks are not immune to the precautionary demand for money. I had initially posed the point to him in a simple IS-LM framework, which has proven to be a useful framework to think about in the current crisis. All free banking seems to do is change a vertical money supply curve to an upward sloping money supply curve. It doesn't seem to change anything else, so it doesn't seem to solve any of the problems that we (or at least I) worry about. Indeed, as Selgin points out banks demand reserves for precautionary reasons as well. You can see this right now with banks sitting on reserves. IOR is a big part of that story, but do we really think that's the only thing holding them back? Probably not. How can we expect a banking system as prone to precautionary savings as anyone else to guarantee full employment? How does an upward sloping money supply curve in an IS-LM model fix anything? These aren't accusations so much as genuine questions because I'm not expert on free banking and I'm trying to figure out why people think its so great.