In a series of recent papers, Micheal Dooley, Peter Garber, and David Folkerts-Landau have been arguing against the textbook open economy macro model (and said model's dire predictions about the current set of global financial accounting imbalances). They argue that since Sovereign debt is not really collectible, poor countries must post effective collateral to get financial flows from rich countries. They further argue that past exports as represented by the US current account deficit are precisely this collateral.
In other words, China has given us a ton of stuff in exchange for t-bills. If they expropriate US or other rich country FDI, the US cancels their claim to the t-bills and we get the stuff for free. That is to say, China's huge reserve holding of dollars is just collateral against any appropriation of the FDI being done there.
Here is how they put it in their latest NBER working paper (gated, sorry):
The nature of the social collateral is so obvious it is hard to see. If the center cannot seize goods or services after a default, it has to import the goods and services before the default and create a net liability. If the periphery then defaults on its half of the implicit contract, the center can simply default on its gross liability and keep the collateral. The periphery's current account suplus provides the collateral to support the financial intermediation that is at the heart of development strategies. The interest paid on the net position is nothing more than the usual risk-free interest paid on collateral.
This is really cool.
Finally, a link to other Dooley et al papers on this topic along with criticisms of their approach is here.
UPDATE: Interesting comment over on Marginal Revolution, on this topic....
There's a practical problem with canceling China's dollar assets: The enormous secondary market in T-bills means that China can easily sell them to some third party who could redeem them at face value. I can't think of any way to close this loophole without effectively shutting down all trade in T-bills, which has enormous negative consequences for the U.S.
Sounds right, and raises an interesting problem. Even if sold at a discount, a quantity of t-bills that large might well drive prices down, and therefore raise interest rates a LOT at the next auction. Good point, Ammianus.
UPDATE II: A correction, from Belligerati. So, never mind on the secondary market way out. Angus was right all along.