One of our three commentators has asked me to engage in the following thought experiment.
A section of the country secedes, apparently peacefully and I am placed in charge of determining how the new country will pay for things.
The task is to address these questions (or if you think they are the wrong questions explain why);
What will be your currency? How will it be created? Will it have an exchange rate fixed to a commodity (gold?) or be free floating? What will you do with your new citizens "old" US$ ? How will you control inflation?
First, the country doesn't pay for anything. And I don't mean this in a semantic sense - I mean that I (as Treasury secretary) would push for anarcho-capitalism and the dismantling of any national state. (This is why libertarians can't get into politics - the first thing we want to do is eliminate the state!) But, since that is probably a cop-out, I'll move forward with the questions.
There wouldn't be an official currency, and the state would not then be able to be the currency monopolist. Banking, including note issuance, would be entirely free. The empirical experience here is Scotland in the 18th century. I imagine that people would develop some matrix of exchange rates among the various currencies, much like exists today among national currencies. The currency, then, would be created by individual banks in a competitive market.
The commodity backing is an interesting question. Scottish banks has a gold backing, but they only held a 2% reserve. The only modern free banking experiment was Hong Kong (ended pretty much in the 1930s, but officially in 1965), but those banks were required to hold 100% foreign currency reserves. So it's not quite the same free banking situation as Scotland. That said, I don't see why the choice of commodity backing shouldn't also be competitive. Gold is obviously a good choice, but any precious metal would do, really. Some banks might try to back their currency fully with the lending on real assets. The problem with the latter is portability (I can't withdraw Jim's land, for example). So I think it would end up being some sort of portable, storable, commodity. Could be gold, silver, platinum.... And, the rate of redemption (one note = 0.01 oz of gold or something) would also be set by the individual banks.
There would be no, repeat no, central banking authority.
Old $ US would still be useful as long as the remaining U.S. had open trade with our new little operation. The rate of exchange to the new currencies issued by banks would, again, be set by the market. Some citizens might even want to be like Panama, and trade only in $US. Again, if the former U.S. maintained open trade with our new operation, people could work in the old US and get paid in $US. Ain't no thang.
Finally, the inflation question. Price inflation is a result of monetary inflation, so the root question is, how would you control money supply. This was the original concern with regard to free banking - that a bank could make money by overissuing its own currency. I won't present the whole refutation here (one can consult Vera Smith, or George Selgin, or Larry White on this issue). But the crux is the clearing process. If one bank overissues relative to its asset backing, as other banks clear more of the notes and the commodity (e.g. gold) is transferred from one bank's reserves to the other's, the overissuing bank will find its reserves becoming depleted. The value of the currency will decrease, as will trust in the bank's ability to refund commodity for currency. People will avoid using the bank's currency and it will lose market share and profit. So general money supply increases will be curtailed.
Note that there will remain one important source of price inflation, and that is an increase in the supply of the commodity that backs the currency. If a new gold mine, for example, is discovered, the value of current gold holdings must decrease if demand stays the same. This inflation can create problems, of course. This was the root cause of the Amsterdam tulip bubble (see Doug French's book at the Mises Institute on this issue.)
I'm sure there'll be questions, but that's my story.