It’s been a month since we shared the Introduction to eBTC and we’re excited to share the first of many builder focused progress updates with the Badger community.
Badgers are relentless, and as such, have dedicated a significant amount of time during this bear market to focus on fundamentals and building technology that will truly stand the test of time.
The goal of eBTC is to bring forth a financial primitive that is both immutable and unstoppable, securing the future of Bitcoin within the Ethereum ecosystem.
The high-profile issues related to centralized services in the past year have made that vision all the more important and timely.
In this post, we’ll discuss the steps taken to this point, provide insight into some of the decisions around architecture, and give an overview of what lies ahead as we work towards launching eBTC in the coming months.
Building on Proven Concepts
Given that eBTC shares similar mechanisms with existing CDP Systems, we decided to perform a data-driven study on the most prominent projects to understand what mechanisms and architectures would work best for achieving our goal of providing capital efficient stablecoin loans with a strong peg and censorship resistance.
Details of this research will be shared in a future article, but the conclusion was that the most effective systems in terms of peg stability compromised on decentralization and censorship resistance.
Liquity’s LUSD architecture stood out as an exception, excelling in providing censorship-resistant borrowing with a stable floor peg. Designed with a focus on censorship resistance and decentralization with only ETH as collateral, it was similar to our vision for eBTC. We decided to borrow from Liquity’s design as the foundation for eBTC’s architecture on the grounds of their stable performance, quality of research, and commitment to immutability.
Refining eBTC
We initially challenged ourselves to use the same ETH-only, no governance approach. ETH is the “pristine collateral” of the Ethereum ecosystem, native to the blockchain and therefore inheriting the same decentralization and security properties of the underlying network.
Could we have a bitcoin credit facility that truly inherits the security properties of the underlying consensus layer?
Through our product ideation and modeling efforts we realized that zero-interest borrowing, built on top of staked Ethereum presented a much stronger product market fit than one that charged an interest rate built on top of native ETH.
With eBTC, loans will be denominated in terms of BTC and collateralized solely by one flavor of staked Ethereum or Liquid Staking Derivative (LSD).
This requires some compromises in decentralization and censorship resistance in the short term, as the staking infrastructure is still very much in development and not at the level of native ETH in terms of the aforementioned properties. We also noted that centralization of oracles is a fundamental limitation for the near future. We intend to work through these obstacles over time with a clear roadmap toward further decentralization.
For eBTC, we looked at every aspect of a CDP system from the ground up, using the benefits of hindsight from additional years of protocol development in DeFi to create what we believe to be the most modern CDP design that exists today.
Some notable differences between eBTC in its current state of development and other models include:
- No Stability Pool is used
- Liquidations are open to the free market in order to improve capital efficiency and enhance user experience.
- We envisioned ways to reduce negative debt redistribution impacts while maintaining insolvency management strategies.
- It leverages protocol yield share on the LSD collateral, eliminating the borrowing fee.
We will be sharing a deeper dive on some of the fundamental aspects of the eBTC architecture in an upcoming series of articles.
Modeling: Crafting the Optimal eBTC
With this new model we felt it was imperative to assess the efficiency and safety of this design from an economic standpoint. As a result,we conducted several statistical analysis and modeling exercises to explore the price volatility implications of using a more correlated pair like LSD/BTC. Fortunately we found it to be less volatile than the original ETH/USD pair, minimizing the risk of sudden price fluctuations and, therefore, liquidations. Additionally, we developed several simulations to evaluate the economic implications of the design deviations, although we sought expert input to further validate our findings.
In order to evaluate the economic implications of these fundamental design changes, We partnered with RiskDAO, a service DAO that specializes in offering risk research and assessments to DeFi lending protocols. Over the past two months, Badger has been working closely with RiskDAO to develop a comprehensive model that exposes our design parameters to various market conditions. This ongoing engagement has established a feedback loop for our design process, resulting in the identification of critical stability mechanisms and parameters for the system, as well as confirmation of their security and efficiency.
Furthermore, our collaboration has produced innovative ideas that complement the core design of eBTC and we intend to share more details about this process and the research results soon.
Where are we now?
We are deep in the development process for eBTC. For each aspect of the protocol, we run through a process involving all the disciplines of the team, with feedback from all stakeholders on each stage.
- Product: Ideation around mechanics, utility to our users, and the product’s place in the market.
- Simulations & Modeling: Assessing these system dynamics and risk parameters using a combination of brute-force mathematical sims, volatility backtesting, liquidation models, and now, agent-based simulations
- Solidity Development: Implementing these into safe and efficient smart contracts, and testing them.
- Full-stack Development: Taking these smart contracts and building out the stack for a seamless user experience, including the necessary backend, frontend, and design components.
Despite the challenge of building software while the spec is still in development, we have been able to strategically allocate work to ensure we are able to keep busy across the stack. This speeds up the development process as opposed to a waterfall approach, where each of the stages above needs to occur in a more serialized fashion.
From this process, decisions on the following parameters and features have been made and their corresponding architectural changes have been designed and implemented into the solidity codebase:
- Liquidation mechanics and incentives, determined in conjunction with RiskDAO.
- Minimum Collateral Ratio and Critical Collateral Ratio numbers, determined in conjunction with RiskDAO.
- Implementation details of having an LSD as collateral, and the efficient processing of the associated protocol yield share.
- Minimum CDP size, balancing between safety and accessibility.
- Multiple CDPs per user enabled as a user-experience optimization (versus one active CDP per account in Liquity)
- Oracle Protocol integrations to ensure the fastest and most robust oracle solution that is achievable right now, heavily borrowing from Liquity’s well-considered oracle fallback design.
- Minimized governance features to ensure the product can adapt in key ways while maintaining a non-custodial guarantee and stability for borrowers.
- Flash Loan mechanics for debt and collateral assets.
- More flexible redemption mechanics that can respond to the changes we may introduce as modeling completes.
What’s Next?
Finish Core Definitions
The eBTC protocol is designed to stand alone forever as bitcoin on ethereum, with its intrinsic value proposition to users intact. A design goal for Badger was to ensure that the health of the protocol only requires free market actors taking incentivized actions. We call this the “core” protocol, with the premise that it can outlive its creators or any protocols taking specific guardianship of eBTC.
We are continuing our work with RiskDAO to further model the best redemptions mechanic and the yield fee, with an eye towards how these affect safety and user behavior.
Robust Testing and Auditing Process
We are committed to ensuring the highest level of security for eBTC. As we develop the code, we expand the testing quality in parallel. This involves using the cutting edge tools for integration testing, fuzzing, and static analysis on top of the strong test coverage that is table stakes for developing a safe blockchain protocol.
Our audit process is similarly strong and continuous. We work with external auditors who review our emerging codebase on a feature-by-feature basis to highlight potential issues, show opportunities to optimize, and comment on architecture. This continuous feedback during the development phase helps prepare us for the formal audits to come when we are code-complete.
Expand the eBTC ecosystem
While eBTC is designed to stand alone, we as a DAO are compelled to use our resources to provide all the external support we can to encourage stability and growth. We’re exploring how best to encourage deep liquidity, how to use external stability-pool like mechanics for fast and reliable liquidations, and alternate methods to maintain peg stability, especially in regard to upwards deviation.
As part of the open ethos of eBTC, we also are developing open-source monitoring and analytics tools for the protocol, including liquidation and redemption bots. Development of this, while ongoing, will be at its heaviest when we are running on testnet.
Building Together
Badger DAO has always adhered to the philosophy of building in public with the community and eBTC is no exception. Contributors have spent almost a year researching and designing eBTC and as we progress towards launch we intend to break down important components of the protocol for community conversation, insight and feedback. Stay tuned for future updates!