Initial Network State
Our initial goal is not to find a stable price. This may seem antithetical to our currency aspirations, but we ensure you it is not. GIZA can be tuned to optimize for different things. The main tradeoff is volatility and profitability versus stability and consistency. With volatility and profit comes growth; this is what we want early on.
With tight policy and scale, GIZA should function well as a stable asset. Upward and downward pressures should stabilize at some non-intrinsic value. With loose policy, regardless of scale, Giza has the potential to act as a wealth creation machine. The market premium of the token measures the positive sum of the game; all extrinsic value is new wealth created.
Alpha State
The initial network features a one-way treasury (money goes in, none comes out), the bonding contract (through which supply increases and profits are produced), and the staking contract (where profits are distributed).
The following are the initial policy states:
BCV
BCV varies based on bond types. It is tuned regularly by the Policy team to meet the protocol goals. For example, if the protocol wants to accumulate more liquidity into its treasury, it can lower the BCV for liquidity bonds to increase their bond capacity.
Bond vesting term
It is set to Fantom blocks or approximately five days for all bond types.
GIZA distribution
Every time someone purchases a bond, the proceed will go to the Giza treasury. A corresponding amount of GIZA will be minted and distributed to three parties:
Bonder
The bond purchaser will receive the quoted amount of GIZA linearly over the vesting term.
DAO
The DAO receives the same amount of GIZA as the bonder. This represents the DAO profit.
Stakers
After accounting for the GIZA distributed to the bonder and the DAO, the rest will be distributed among all stakers in the protocol.
Last updated