Equilibrium infrastructure: Roles and Voting Mechanisms
EOSDT is the first decentralized asset-backed stablecoin built on top of the EOSIO smart contract architecture, which boasts low fees, low transaction latency, and a powerful infrastructure designed for scalability. This series of articles provides a first description and analysis of the Equilibrium framework’s objectives and the mechanisms it uses to achieve them.
Let’s start with the main principles. Equilibrium’s objective is to maintain EOSDT’s price parity to the US dollar. It does this by:
- Ensuring the pool of generated EOSDT remains overcollateralized at all times via a collateralization threshold
- Driving supply and demand of the EOSDT token with an Equilibrium fee that directs behavior toward generating or liquidating EOSDT tokens
- Ensuring good governance of the EOSDT ecosystem through mechanisms that require participants to engage the community to accept their proposals for changes to Equilibrium’s structure and parameters
- Ensuring efficient action and transaction processing by producing EOSDT blocks within EOSIO’s architecture, under its delegated proof of stake (DPoS) rules.
The Equilibrium infrastructure
Equilibrium users share the objective of maintaining USD parity for EOSDT, and the ecosystem operates with the Native Utility token (NUT) as the system’s utility currency. All actions, transactions, services, and risks within the framework have an associated cost payable in NUT.
Each of the participants has a specific role in this process.
The position owner
These persons are the EOSDT generators. They create the position where they will be able to deposit collateral. The account is governed by the eosdt.ctract smart contract, which incorporates Equilibrium’s risk parameters. These include the minimum collateral to debt ratio (set at 170%) and the admin fee (set to an annual 1%). Once the collateral position credited, a position in EOSDT can be generated for a minimum collateral to EOSDT ratio of 170% (valued at parity 1 EOSDT = 1 US dollar).
The third-party data provider (oracle)
Collateral value is computed every five minutes with data provided through a trustline by Oraclize (now Provable). There are reasonable checks in place to capture any delays in broadcasting data. A critical five-hour delay could cause system failures. A level of arbitration in price feed will have to be introduced in time, increasing the number of data sources and voting for the most reliable ones (see the role of the NUT holders, described next).
The Native Utility (NUT) holders
The NUT token is the cross-chain-compatible utility currency accepted by Equilibrium, and it plays a role in every step of governance and management processes:
- Admin fees are paid in NUT: An accrued NUT admin fee must be paid before positions can be unwound. It is paid in proportion to the unwound amount.
- In a liquidation scenario, NUT is used to buy the collateral asset from the liquidator contract at a favorable discount. The paid NUT gets burned and the supply of NUT is reduced, contributing to NUT price inflation.
- A NUT holder has the power to vote for a change in the system’s governance, including a change in its parameters. The proposal and voting system is democratic — a 55% vote counts as a majority, and at least 51% of the participants must vote for a proposal to be accepted.
- As we will see later, NUT tokens can also be used to stake a portion of the collateral, allocating more resources or votes for the most favorable block producers (BPs). This approach presumes leveraging any delegated proof of stake (DPoS) architecture, including EOSIO.
For all these reasons, NUT holders play a critical role in Equilibrium’s governance.
Decentralized Governance: BP Voting within EOSIO infrastructure
Because the collateral position is governed by a smart contract, custody risk is very low. But compromised blockchain integrity can put collateral at risk by miscalculating parameters. Voting (or inducing a vote) for an unreasonable critical collateral-to-EOSDT ratio (whether on a standalone basis or in conjunction with other parameters) can have unexpected negative effects.
Imagine a low collateral-to-EOSDT ratio with a higher staking portion of the total collateral held. Since unstaking an EOS position takes three days, the system’s liquidity could be compromised, especially in potential periods of high collateral liquidation (whether voluntary or auto-liquidation).
In order to mitigate these marginal effects, an engine that simulates accounts, positions, prices and participants behavior is under development to perform thorough crash risk scenarios to various parameter inputs, leaving very little to chance.
EOSIO’s governance depends on an ownership-based voting system. It’s the best insurance for continued EOSDT block production. It uses the delegated proof of stake (DPoS) consensus, in which a participant’s votes are proportional to their stake in EOS tokens. They let participants vote every round to select 21 EOS block producers for staking.
The eosnetworkmonitor.io website lists all 21 block producers, and the BP account named eoslaomaocom holds the first place at the time of this writing. What are BP’s powers and obligations?
- Every BP can produce blocks and process transactions against token payments.
A BP gets operational resources like bandwidth/log storage, computational power, and the most precious of all: RAM. These resources are proportional to the amount of tokens staked for three consecutive days, locked by a smart contract in an eosio.stake account. The BP capacity is public knowledge: eoslaomaocom’s current usage is now at 7.6% of its RAM capacity (170.07 KB).
Block producers get to produce six blocks until the next election. EOSIO lets you produce a block every 0.5 seconds at a minimum. Within a round, the BP can produce up to six blocks over six seconds.
The total reward for production is paid in EOS, and it represents 0.25% per year of the total EOS supply. It is determined as a proportion of the number of total blocks produced. For a supply above a billion EOS, the total reward is EOS 2,500,000. That’s $10.8 million to split among producers each year. A 0.75% per year reward is paid per year in proportion to the number of producer candidate votes. A producer candidate outside the 21 selected BPs can therefore benefit from substantial reward.
- Block producers decide whether to reject any transaction or block on the basis that it requires unreasonable computational power.
The next two producers have the power to reject the block after at least one second.
- A block producer can freeze an account with a 15-21 vote of elected producers.
A block producer can initiate a vote on a new constitution change first. If the first vote is successful, the BP can initiate a second vote for new code to be inserted. Implementation needs a 15-21 vote approval to be maintained over 60 days.
- A block producer has the duty to produce and safeguard the blockchain’s integrity.
The DPoS protocol helps safeguard EOSIO integrity, motivating the elected accounts to produce blocks at a substantial rate, and to act in the blockchains best interest. Token holders can force a BP’s removal and replacement if it fails to produce, creates unacceptable processing delays, or freezes an account incorrectly.
There are safeguards in place to prevent the simultaneous production of both a blockchain’s block and a hard-forked block — a BP has to choose. A blockchain fork happens if a majority of producers adopt the same consensus, in which case the consensus will shift by construction on the blockchain fork with the higher producing rate.
For all these reasons and more, using EOS as collateral can have a substantial opportunity cost when it comes to resource staking and election votes. Your voting right can elect the block producer that will operate on the EOSDT blockchain. It can also reward a substantial amount of EOS to a producer candidate.
The collateral held against the generated EOSDT can have a substantial voting power that could be used to safeguard the integrity of the EOSDT blockchain, eventually protecting all its participants. The following is a simple case study to determine an upper bound for an admissible staking portion of the total EOS collateral.
A Reasonable Threshold for Staked EOS Collateral
With a collateral-to-debt ratio CR, an EOSDT generator would be able to leverage his collateral position by a ratio of 1/(1-1/CR) greater than 1. By reinvesting the generated EOSDT into the same collateral asset (EOS), a user can credit the owned position with more EOS tokens, generating more EOSDT, and repeating these steps indefinitely. In practice, a NUT fee for staking the collateral would act as a deterrent.
An EOS holder can stake a total of 1/(1-1/CR) for every EOS s/he owns this way. To avoid using the leverage process to get more votes, Equilibrium holds a ratio noted rr of the collateral in reserve, and it should usually remain unstaked. The remainder can be staked, but the total quantity of EOS available to stake should not exceed the initial quantity of EOS posted as collateral. This requirement sets an upper bound for the staking portion: (1-rr) < 1-1/CR. Let’s call it the fractional reserve bound.
Because of the three-day minimum period to unstake a token, staking a significant proportion of the collateral in a period of high liquidation (like voluntary retrieval of the collateral post debt payment or auto-liquidation) can furthermore lead to a dangerous state of cash flow insolvency. Our adjusted upper bound should be set as the minimum of the fractional reserve bound, and an insolvency-proof bound to be determined.
The risk we are trying to measure is related to the probability of a crash over three days or less, and the intensity of that crash. A significant crash would consume the collateral for a value that exceeds the haircut, such that the parity of EOSDT to the US dollar cannot hold.
But the exciting in-depth analysis of the crash risk will be left for another day. Stay tuned.
Yves Renno, CFA
Lead analyst at Equilibrium
I am a statistician economist from the ENSAE in Paris with a MSc in statistics and financial modelling from Paris VII and La Sorbonne in Paris and a MSc. in financial mathematics from the University of Chicago. I’m a CFA charter holder and a member of the French Association of Actuaries. Principal/Owner at Y&M Advisors Ltd.
I started my career as a trader on a single-stock exotic derivatives trading portfolio, then headed the derivatives sales trading team at Commerzbank. I was also a portfolio manager on a volatility-driven forex-, equity-, and commodities-based global macro fund.