The basic problem with IOUs backed by single issuers is counter-party risk. So for example, an IOU for USD from an exchange is always subject to the viability of that exchange, and whether you will eventually be able to withdraw.
The solution proffered by BitShares to date has been to eliminate counter-party risk altogether, by backing tokens with BTS. But that also has a weakness - investment risk. BTS can be volatile, and its fundamentals can change dramatically, especially for what amounts today to a micro-cap stock (the future might be very different of course!). The wider market is not currently familiar or confident in the stability of BTS. The way to mitigate this is by stringent collateral requirements. But here there is a trade-off - increasing security for longs means making the product less attractive for shorts.
Another avenue that could be explored is one that seeks to reduce both the counter-party risk and the investment risk. This is done by matching the nature of the collateral as closely as possible to the nature of the asset. (One way to think of this is matching assets to liabilities, reducing the relative risk).
If there were enough IOUs for an asset issued in the UIA market, then any of these IOUs could be used as collateral to back a token in the corresponding asset. A haircut on the value of that collateral would be required to allow for the prospect of issuer default, much in the same way as we allow a haircut on BTS for the volatility risk (a haircut is just another way of saying you need more than 100% of the value held as collateral). But we can go further than this. Shorts could choose which IOUs they want to include as collateral, and in what mix, and because they always bear the first loss on any devaluation of that collateral, it is in their interests to ensure they get the proper mix of quality and diversification.
In the future, suppose there were many exchange or gateway based IOUs for USD, issued as UIAs. Lets label these EXA.USD, EXB.USD, EXC.USD, ... , EXZ.USD. Most will trade close to par. Those that fall below par will have increasing collateral requirements, and shorts will act to move away from these to higher quality IOUs. Thus its the shorts that manage the counter-party risk on behalf of the longs.
In this way a USD token is potentially backed by all the exchange members, with the risk profile managed by users heavily incentivised to maintain quality and diversification.
For high quality IOUs, haircuts might be as little as 10% (this can be adjusted higher if the collateral pool is not very diversified, and lowered as diversification increases). Shorts should be able to earn a reward in the form of a fee from longs (even 1% pa would represent a 10% pa return for shorts) and from market-making. Thus their main reward is an income stream, rather than speculation.
Risk cannot be removed, only changed in form. This is just another way of packaging risk for users of pegged assets.
Topic: Idea for the future: Asset tokens backed by decentralised IOUs (Read 4055 times)
