Re: > I suppose there is some connection between your attempts to study the > tools of complexity analysis and your ability to see the complexity > of the real economy! I hope so Rakesh! > Yet I must ask: why would a devaluation affect the ability to deliver > the dollars since these commitments are probably written in dollars? > It's not like the US is Thailand with debts in dollars and revenue > and money creation powers in another currency (baht)? <snip> Much of why it could be dangerous relates to the wonderful world of derivatives, which I had better add that I don't pretend to completely understand. But a lot of it consists of "swaps", where borrowers in one country with a variable interest rate on their debt swap it with those in another with a fixed rate. The parties exchange debts in return for swapping the repayment liabilities, so it is quite possible that lots of US corporations could lose heaps out of a dollar devaluation. There is also the danger of counterparty risk--one party could hold what looks like a winning position on some speculative derivative, but the losses on the other side are so enormous that the party(ies) there could be unable to deliver. One bankruptcy causes the other, and on it goes. Plus lots of risky positions have supposedly been "insured". Insurance works with Gaussian, non-related risks; it doesn't work with fractal, interdependent ones. A failure somewhere in the system that meant one insurance contract couldn't be delivered on could cascade through the rest of the chain. These are all massive unknowns. The level of derivatives that exists now has no parallel in the past, and no-one knows what it likely to happen if the chain unravels. I think it's the uncertainty of it all that paralyses the decision makers, and also makes them inclined to continue one of the imbalances that has led to the proliferation of derivatives--the US credit bubble--even if other sectors such as manufacturing can make an objective case as to how a lower dollar might help them. Cheers, Steve
This archive was generated by hypermail 2b30 : Sun Jun 02 2002 - 00:00:07 EDT