The short rates have to compete with the Central Bank’s overnight policy rate (fed funds in the US). Since very short maturity treasuries are nearly perfect substitutes for overnight lending between banks, short rates track the overnight rate. And that is set by the Central Bank, plain and simple.
Longer rates on Treasury debt are determined largely by expected future short term rates–which will be set by the Central Bank. Krugman is correct that it is a bit more complicated (there’s risk of capital loss if short rates rise; plus there are “habitat” preferences and possibly some liquidity preferences) but that’s close enough. If some Bond Vigilantes go all crazy on us and run out of Treasuries after a credit rating agency downgrade, there will be plenty of noncrazy investors who will eat them for lunch. Look, there have been crazy investors who year after year after year place bets on impending Japanese default. Ain’t going to happen. You should take the other side of their bets against Japan–its free money. They want to pay you for their stupidity.
Ditto US and UK, as Krugman is arguing. Bet on the sovereign’s side.
Here’s a key passage from Krugman:
It’s very hard to come up with any reason why either the US or the UK might default, since they can simply print money if they need cash. And given the absence of real default risk, long-term interest rates should be more or less equal to an average of expected future short-term rates (not exactly, because of maturity risk, but that’s a fairly minor detail). So if you expect the US and UK economies to be depressed for a long time, with the central bank keeping rates low, long rates will be low too — end of story.
Krugman goes on to argue that the inflation mongers (including gold bugs) are also wrong. There’s no significant inflation on the horizon.
I’d go a bit farther than him, however. Even if inflation should rear its ugly head, there’s not much an investor can do about it. You always go for the highest nominal return you can get, given your taste for default, liquidity, and interest rate risk and other possible nasty events. Expected inflation doesn’t change that–since inflation is by definition a nominal problem.
Other than inflation-indexed bonds (which the US Treasury happily sells, and which come moderately close to an inflation hedge) you cannot truly hedge against inflation. People have thought up complicated commodity indexes but what you really need is a price hedge for a personalized basket of goods and services that you will want to buy in the future.
No one is selling that hedge. Gold ain’t it unless you consume only gold. King Midas found that to be a curse–as you no doubt remember.