A currency is designed to be a fungible and non-callable asset. A pegged Derivative currency, such as BitUSD, is backed by a cryptocurrency held as collateral. The "issuer" is "short" the dollar and extra-long the cryptocurrency. The buyer is simply long the dollar.
BitShares created the first working pegged asset system by allowing anyone to take out a short position by posting collateral and issuing BitUSD at a minimum 1.5:1 collateral:debt ratio. The least collateralized position was forced to provide liquidity for BitUSD holders any time the market price fell more than a couple percent below the dollar (if the BitUSD holder opted to use forced liquidation).
To prevent abuse of the price feed, all forced liquidation was delayed.
In the event of a "black swan" all shorts have their positions liquidated at the price feed and all holders of BitUSD are only promised a fixed redemption rate.
There are several problems with this design:
- There is very poor liquidity in the BitUSD / BitShares market creating large spreads
- The shorts take all the risk and only profit when the price of BitShares rises
- Blackswans are perpetual and very disruptive.
- It is "every short for themselves"
- Due to the risk/reward ratio the supply can be limited
- The collateral requirements limit opportunity for leverage.
We present a new approach to pegged assets where the short-positions cooperate to provide the service of a pegged asset with high liquidity. They make money by encouraging people to trade their pegged asset and earning income from the trading fees rather than seeking heavy leverage in a speculative market. They also generate money by earning interest on personal short positions.
An initial user deposits a Collateral Currency (C) into an smart contract and provides the initial price feed. A new Debt token (D) is issued based upon the price feed and a 1.5:1 C:D ratio and the issued tokens are deposited into the Bancor market maker. At this point in time there is 0 leverage by the market maker because no D have been sold. The initial user is also issued exchange tokens (E) in the market maker.
At this point people can buy E or D and the Bancor algorithm will provide liquidity between C, E, and D. Due to the fees charged by the the market maker the value of E will increase in terms of C.
Collateral currency = Smart Token/reserve of parent currency
Issued tokens = Bounty Tokens (distributed to early holders / community supporters)
Collateral Ratio (C:D) = reciprocal of Loan-to-Value Ratio (LTV)
To maximize the utility of the D token, the market maker needs to maintain a narrow trading range of D vs the Dollar. The more consistant and reliable this trading range is, the more people (arbitrageur) will be willing to hold and trade D. There are several situations that can occur:
-
D is trading above a dollar +5%
a. Maker is fully collateralized
C:D>1.5- issue new D and deposit into maker such that collateral ratio is 1.5:1b. Maker is not fully collateralized
C:D<1.5- adjust the maker weights to lower the redemption prices (defending capital of maker), arbitrageur will probably prevent this reality.Marker Weights = Connector Weights (in Bancor)
Redemption Price: The price at which a bond may be repurchased by the issuer before maturity
-
D is selling for less than a dollar -5%
a. Maker is fully collateralized
C:D>1.5- adjust the maker weights to increase redemption pricesb. Maker is under collateralized
C:D<1.5- stop E -> C and E -> D trades. - offer bonus on C->E and D->E trades. - on D->E conversions take received D out of circulation rather than add to the market maker - on C<->D conversion continue as normal - stop attempting adjusting maker ratio to defend the price feed and let the price rise until above +1%
Value of E = C - D where D == all in circulation, so E->C conversions should always assume all outstanding D was settled at current maker price. The result of such a conversion will raise the collateral ratio, unless they are forced to buy and retire some D at the current ratio. The algorithm must ensure the individual selling E doesn't leave those holding E worse-off from a D/E perspective (doesnot reduce D to a large extent). An individual buying E will create new D to maintain the same D/E ratio.
This implies that when value of all outstanding D is greater than all C that E cannot be sold until the network generates enough in trading fees to recaptialize the market. This is like a company with more debt than equity not allowing buybacks. In fact, E should not be sellable any time the collateral ratio falls below 1.5:1.
BitShares is typical margin call territory, but holders of E have a chance at future liquidity if the situation improves. While E is not sellable, E can be purchased at a 10% discount to its theoretical value, this will dilute existing holders of E but will raise capital and hopefully move E holders closer to eventual liquidity.
The price feed informs the algorithm of significant deviations between the Bancor effective price and the target peg. The price feed is necessarily a lagging indicator and may also factor in natural spreads between different exchanges. Therefore, the price feed shall have no impact unless there is a significant deviation (5%). When such a deviation occurs, the ratio is automatically adjusted to 4%.
In other words, the price feed keeps the maker in the "channel" but does not attempt to set the real-time prices. If there is a sudden change and the price feed differs from maker by 50% then after the adjustment it will still differ by 4%.
Effective Price = Connected Tokens exchanged / Smart Tokens exchanged
Under this model holders of E are short the dollar and make money to recollateralize their positions via market activity.
Anyone selling E must realize the losses as a result of being short.
Anyone buying E can get in to take their place at the current collateral ratio.
The value of E is equal to the value of a margin postion.
Anyone can buy E for a combination C and D equal to the current collateral ratio.
Anyone may sell E for a personal margin position with equal ratio of C and D.
Anyone may buy E with a personal margin position.
If they only have C, then they must use some of C to buy D first (which will move the price).
If they only have D, then they must use some of D to buy C first (which will also move the price).
Anyone can buy and sell E based upon Bancor balances of C and (all D), they must sell their E for a combination of D and C at current ratio, then sell the C or D for the other.
Anytime collateral level falls below 1.5 selling E is blocked and buying of E is given a 10% bonus.
Anyone can convert D<->C using Bancor maker configured to maintain price within +/- 5% of the price feed.