If the total value of the backing drops below the total nominal value of the issued currency, then why anyone would ever bother to buy this one over a freshly created copy? The latter always begins its life with full backing, and thus has a relative advantage over the former. It seems to me that once this threshold is breached, the old currency will have to sell at a discount to compete with the new one, thus depleting the escrow fund until it's wiped out completely. Notice that the old currency remains technically insolvent no matter how much buying the escrow fund does, and it can never eliminate the advantage that the new currency has over it. Its demise is a certainty at this point.
All GoldCoins are backed by the same escrow fund, to the same degree. GoldCoins are fungible, and there wouldn't be one that was more funded than another.
That sounds almost impossible, given that these currencies are pegged to arbitrary tickers. The protocol has no knowledge of whether or not ticker 254326626778378 is the price according to its publisher of the same thing that ticker 7883689934799834 is the price of, surely?
Over time the protocol can observe that both tickers report the same price each day over some span of time, but the two publishers might not even agree every day that the price of gold or oil or whatever they are trying to publish the price of is the same wherever they each are located, so maybe even though both are doing their honest best to publish the price of the same commodity that commodity might legitimately not cost the same amount at the spot-sales shop on the block they live on.
How can the code even begin to even try to decide whether the stuff i am trying to publish "the" (or a) price of on my ticker is or is not the same stuff that someone else is trying to publish a (or "the") price of on their ticker?
-MarkM-