Last night on Kudlow, three out of five of the people talking wanted the Fed to raise interest rates to head off inflation and defend the dollar. On these very pages, I am regularly castigated for saying any of the following:
- A "strong dollar" doesn't mean a strong economy
- The government is going to have to pay for its debt, not inflate it away
- Some inflation is all right
- The gold standard won't solve any of the problems people think it will
For someone who hangs out with libertarians a lot, announcing that you're okay with a small amount of inflation to relieve the sticky price problem, and that you think the Fed mostly does a good job, is akin to announcing that you've decided to take up human sacrifice to fill those lonely weekend hours. Nonetheless, I stand by the sentiment: inflation is not the big worry that our economy faces.
To explain why I think the risks lie in the directions of fiscal crisis rather than hyperinflation, I turn to Tyler Cowen, who elegantly summed up the government's problem at a conference I attended last week: inflation only works on stocks of debts, not flows. You can inflate away the value of debt you've already issued, but especially in these modern times, bondholders will rapidly ratchet up the interest rate they charge you. They will increase it by more than the rate of inflation, to compensate them for future inflation risk. Losing your credibility is costly.
To this, my excellent former co-blogger Winterspeak responds that since the United States has a monopoly on the currency it issues, it can't default--it can just keep running the printing presses. There are a couple of problems with this. In some sense, I think it confuses cause with effect: the government gets to borrow in a currency over which it has a monopoly, because it has been a fairly credible steward of that currency. If it inflated away its debts on the scale of, say, a Latin American nation, it would not be able to borrow in dollars any more. And since it needs real goods and services, not little green pieces of paper, that matters.
For his point to be right, you need to believe that the government can end up financing the entire debt by seignorage, which is the revenue that the government gets from issuing currency. In the case of the United States, that revenue is currently pretty considerable, because so many people overseas use dollars as their emergency bank account; essentially, they give us goods and services in exchange for dollars, and then stash those dollars under their mattresses. But that's because we don't inflate the currency.
In the case of a hyperinflation, essentially, the government gets a slight discount, based on the fact that it knows how much money is in existence, and you don't. It prints the dollars, and uses them to buy goods, and then the oversupply of dollars pushes up prices still further. But the discount is actually pretty small, and hyperinflationary seignorage turns out to be a very inefficient way of generating tax revenue, especially in a world where there are modern financial markets monitoring government behavior. The much maligned Laffer Curve is actually a pretty effective model at describing hyperinflation; it's very easy to get on the wrong side, where inflationary expectations and deadweight loss start killing the revenue you can raise. It is possible to end up in a place where, as with the Zimbabwean dollar, your monopoly right to print your currency becomes worthless, because the demand for that currency is essentially zero--no one will give you goods and services in exchange for your paper.