TBV Research
The below report is available in document format here
Introduction
On Jun 3, 2024, cupojoseph proposed a pilot stage for funding R&D for the implementation and execution of operating a treasury-backed vault (TBV). He suggests that leveraging a TBV and issuing a stablecoin against ARB (and other treasury assets) in the vault could be a more effective funding mechanism for grants and other initiatives–noting leveraging ARB to issue stablecoins could relieve ARB selling pressure.
After back and forth discourse amongst the community, both in favor and against a DAO-owned TBV, members of the DAO (Entropy and L2Beat) expressed that funding R&D for this proposal is redundant given the ARDC. Hence, Blockworks Research and Chaos Labs were assigned the task of outlining the operational structure and performing a risk analysis of operating a TBV.
Summary
In collaboration with Chaos Labs, we highlight operational requirements and suggest R&D, an oversight committee, and a risk management committee are essential pieces for effectively operating a CDP, or TBV. We also consider the operational setup, and compare the high level trade-offs between a simple framework and a more complex one.
Alongside the operational setup, we provide a risk analysis on leveraging ARB in a DAO-owned CDP, offer recommendations for how to properly set the relevant parameters and manage the risk, and share our recommendation and interpretations.
We ultimately conclude operating and managing a CDP or TBV using ARB is highly capital inefficient (especially taking into consideration the DAO’s past spending) and is not worth the risk/reward. This is due to the highly volatile nature of ARB, lack of available liquidity, and the sheer amount of capital that is required to adequately support the DAO’s historical spending behaviors. However, it is worth revisiting leveraging ARB in a debt position once market conditions and ARB liquidity have considerably matured.
That said, this preliminary work is sufficient for the DAO to gauge interest in this direction, and with an updated perspective on the nature of operating a CDP, we encourage members of the community and delegates to carefully review and open further discussions. If necessary, further research can, for example, outline the risks associated with managing a TBV, specifically, at different scales.
Overview of CDPs and TBVs
For educational purposes, here is an overview of existing protocols that the DAO could utilize. While some of these protocols are not active in the Arbitrum ecosystem, or are active in select parts, we recommend reviewing these CDP mechanisms to gain a more thorough understanding.
Maker DAO and DAI
Maker’s main product is DAI, a stablecoin that’s pegged to the U.S. Dollar. DAI’s supply curve moves through a credit manager set of smart contracts on Ethereum. This creates a mechanism by which actors are incentivized to adjust the supply curve to keep the peg price at $1 as market conditions and DAI demand fluctuate. Additionally, Maker allows users to mint DAI through the Maker Vaults, where users borrow DAI and lock up their assets as collateral, repaying the accrued fees on a later date.
Maker uses several risk parameters for each collateral type:
- Liquidation ratio: the minimum ratio of collateral value to debt per Vault (usually overcollateralized).
- Debt ceiling: the max amount of DAI generated for a collateral type.
- Stability fee: the fee that accrues on debt in a vault on an annual basis (ex: 5% per year)
Maker’s mechanisms to maintain solvency are:
- DAI issuance is influenced by stability fees, DAI savings rate, debt ceiling adjustments, and the Peg Stability Module.
- Peg Stability Module: The Peg Stability Module (PSM) maintains the stability of DAI by enabling users to swap collateral tokens directly for DAI. Users can swap other stables for DAI which helps keep DAI fixed to the value of $1. Importantly, the amount of trades the PSM can offer is limited by the debt ceiling of deposited collateral. Thus, upon crossing the debt threshold, the PSM cannot trade in stables for DAI.
- Liquidation: The closer the collateral value of a vault is to the verge of debt, the more risk the vault takes on. Thus the system liquidates overly risky vaults, and liquidations are performed through a gradual dutch auction.
- MKR Mint/Burn: If bad debt hits a certain threshold, a Flapper/Flopper smart contract burns, mints, and sells MKR in order to recap the MakerDAO protocol in times of insolvency.
All interactions with vaults are done through the DssCdpManager (CDP Manager) contract. Vaults may also be transferred between owners using this same contract.
Curve and crvUSD
While crvUSD is not live on Arbitrum as of now, Curve is a potential option should crvUSD expand to the Arbitrum network.
In the typical hard liquidation, when a borrower’s loan becomes undercollateralized, their position is liquidated and converted. Curve’s lending-liquidating AMM algorithm (LLAMMA) is able to soft liquidate, where it gradually converts the collateral into stablecoins as the value of the collateral decreases, preserving some of the borrower’s original capital.
Curve only offers overcollateralized loans, with the loan-to-value ratio dependent on the risk of the collateral. Thus, depending on the collateral, there exists a set of bands for the liquidation range for which soft liquidations can occur. Unlike other AMMs, the price offered by Curve’s LLAMMA is not reliant on the balance of assets in the pool, rather an external price oracle. As a result, if the oracle price increases/decreases by 1% then the LLAMMA price will increase/decrease by at least 1%.
Curve’s stablecoin, crvUSD, is pegged through a basket of fiat stables, with stability pools and peg keepers to further uphold the price at $1. The stability pool for crvUSD holds the largest fiat-backed stablecoins, USDC, USDT, USDP, and TUSD. The Peg Keepers are similar to the AMO concept used by Frax Finance. They’re smart contracts that perform algorithmic market operations to fix the price of crvUSD to its peg. More specifically, peg keepers are assigned to stability pools and can perform market operations under certain circumstances. For example, if and only if the price of crvUSD is above $1 in a stability pool, will the peg keeper be able to mint and deposit crvUSD into the pool. In this scenario, adding single sided crvUSD when the price is above $1 puts downward pressure on the price, moving it back toward the target. The reverse can be said for when the price of crvUSD is below the $1 threshold.
Finally, the other important detail to understand about Curve’s lending mechanisms is its variable interest rate. Borrowers pay a variable interest rate to the protocol based on their total outstanding debt. The interest rate is dependent on the current crvUSD price and the shape of the interest rate curve using risk parameters, and the activity of peg keepers (debt ratio and target debt).
Aave and GHO
Aave’s GHO stablecoin is a CDP style protocol similar to Curve and crvUSD. It is a system of lending pools where users’ deposits are funneled into a liquidity pool. Borrowers may access these pools when looking for a loan. Facilitators (Aave), can mint and burn GHO tokens, each is assigned a ‘bucket’ with a capacity representing the maximum amount they can generate. The total available supply of GHO is calculated based on the capacities of all Facilitators, ensuring the system remains overcollateralized. New facilitators are approved by Aave DAO to ensure that they operate within parameters.
To maintain its peg and system stability, GHO has stabilization mechanisms controlled by Aave DAO–managing borrow rates, collateral requirements, and other parameters for minting and burning GHO. There is also a discount model that gives discounted borrowing rates to stkAAVE holders.
Furthermore, Aave offers other features for risk management: isolation mode, siloed assets, DAO-elected permissioned roles, and then traditional features like supply and borrow caps. Isolation mode is a feature limiting risky assets. Assets in isolation mode may only receive certain types of stablecoins.
Open Dollar and OD
Open Dollar is a stablecoin CDP protocol designed for non-fungible vaults for users to stake collateral and borrow Open Dollar’s OD stablecoin. As an Arbitrum native protocol, Open Dollar strives to make OD Arbitrum’s native stablecoin.
Similar to Aave and Maker, Open Dollar is also controlled by a governance entity and the ODG token. The OD stablecoin is overcollateralized and uses a similar managed float regime mechanism to Maker. This means that the OD/USD exchange rate is determined by supply and demand, and the protocol tries to restabilize by either devaluation or revaluation. Notably it accepts fewer forms of collateral (only ARB and ETH variants) compared to Maker and AAVE.
Some of the listed CDP protocols have been battle-tested through new deployments, making them worth considering for exploration. Additionally, one of the listed protocols serves as an alternative mentioned in the original proposal.
These are very brief overviews of each protocol, to give the DAO a sense for the type of protocol it would use to leverage ARB in a TBV. Further research work would be required to investigate which protocol is most suitable for the DAO. This would likely include working closely with candidates for the relevant committees outlined below.
In order to effectively manage a CDP or TBV, the DAO would have to fund the relevant committees/roles responsible for tasks such as developing the necessary set of smart contracts for facilitating a debt position and maintaining the communication, health, and security of the DAO’s position. These responsibilities primarily include (but are not limited to) an R&D committee, an oversight committee, and a risk committee.
Relevant Committees
In this section, we’ll outline roles and responsibilities that are necessary for effectively managing a CDP/TBV. While we are not suggesting the DAO needs to create three separate committees, it is important to note the essential parts of executing this initiative, even if it means one entity is capable of serving two roles.
R&D: Oversee the strategic and technical development of the CDP/TBV, and work closely with the DAO, other committees, or protocols to ensure the efficacy of the CDP/TBV’s strategy, management, and implementation. This includes and is not limited to:
- Customizing relevant smart contracts to fit the needs and standards of the DAO.
- Conducting security audits of smart contracts and the overall system
- Maintaining technical implementation and pushing updates when necessary
- Ensuring compatibility with DAO operations (e.g. outlining operational procedures for using a safety buffer to top up debt positions with treasury ARB)
- Creating a dashboard or other visual representation for viewing important metrics, such as monitoring collateral and transaction history
- Closely collaborating with other committees or protocols to lead further development of necessary contracts, tools, and strategies that enhance the position
- Performing due diligence and reviewing applications that request funding from the treasury
Oversight Committee: The backbone of treasury operations for approving, sending, and receiving funds to and from the treasury and the CDP/TBV, overseeing the day-to-day operations of the treasury, implementing best security practices, and executing strategic decisions for capital in the treasury. This includes and is not limited to:
- Implementing best security practices between the DAO’s treasury and CDP/TBV management such as procedures for responding to emergency incidents or unforeseen vulnerabilities
- Delivering operational procedures for maintaining accountability (e.g. perform monthly/quarterly performance reviews)
- Maintaining consistent communication with the risk committee, the DAO, and relevant protocols regarding adjustments to relevant parameters of the DAO’s position and other activities related to the debt position
- Pre-approve transactions and privileges for CDP/TBV manager
- Relaying accurate and timely reporting of the TBV to the rest of the community on a weekly/monthly basis.
- Maintaining transparent dashboards that display the DAO’s position in real-time
- Collaborating with R&D committee for issuing stablecoins from the TBV
- Monitoring debt and repayment from funded projects and enforcing penalties if needed
Risk Management: Manage the CDP/TBV position and set risk parameters to ensure solvency. In addition to the below, in order to achieve incentive alignment between the DAO and the risk committee, the DAO and relevant committees would have to determine what responsibilities and/or privileges are delegated to the risk committee. It is possible the functions described here fall within the scope of the R&D and/or a treasury manager’s role. Regardless, since a treasury management role has not been formalized, we felt necessary describing the risk portion of the vault. This includes and is not limited to:
- Clear communication on how the stablecoin works and its benefits
- Conducting ongoing CDP/TVB research, identifying potential risks, developing risk assessment frameworks and models of the economic tradeoff space, recommending the optimal strategy/implementation, and providing regular analysis and updates on the state of the market and the position
- Working closely with existing protocols and their risk teams to determine parameters for levering ARB in a CDP/TBV
- Continuously monitoring market conditions, protocol health, and ARB volatility/liquidity to gauge adjustments to the CDP/TBV and maintain solvency
- Updating over-collateralization ratios, liquidation triggers, interest rates, and other relevant parameters and operations according to the Oversight and R&D committees’ recommendations
- Ensuring liquidation mechanisms are operational and emergency shutdown procedures are effective
- Closely communicating with other committees, the DAO, and relevant protocols
- Sending and receiving funds to and from the DAO’s treasury
- Around the clock maintenance and management
Operational Structure
Taking inspiration from other DAOs, depending on the scale of the position, each committee is necessary for effectively managing a CDP or TBV; however, the operational setup can differ. For example, the CDP could be a multisig that is maintained by each committee. This is a simple setup in which multiple parties sign transactions, and it affords less interactions between the DAO and the position. However, requiring multiple signatures to approve transactions decreases the autonomy for the risk manager to act swiftly, if needed.
The framework we recommend is inspired by Karpatkey’s treasury management proposal and also Lido’s treasury management committee proposed by Steakhouse Financial. In Karpatkey’s proposal, the DAO retains full custody of the vault’s position and can withdraw and deposit through a governance process. In this setup, a trusted, experienced entity is appointed to manage the CDP, or TBV, giving that entity more autonomy. To restrict a malicious manager from carrying out an attack and to incentivize performance, the setup also involves an oversight committee that establishes pre-approved transactions and is generally responsible for ensuring the vault manager is achieving their mandate. Both proposals also utilize onchain tools, essentially creating an environment that is tailored for their respective strategies and use cases, albeit given more complexity.
While this operational structure can be considered more complex and costly than simply deploying a multisig, it clearly defines roles, allows for each party to specialize on their respective tasks, and grants more autonomy for each party to operate efficiently.
Lastly, the DAO would have to work with relevant committees to establish an amount that is allocated to a safety module, which the vault manager can draw from when needed.
We’ve outlined high level tradeoffs between a very basic structure and more professional structures. If further progress is made toward implementing a position, potential candidates–that have track records servicing these roles and operational structures in practice–can provide their frameworks. This would allow the DAO to compare specific details about the operational structures from experienced candidates.
Additional work
Primarily, if the DAO is borrowing stablecoins against ARB and distributing those funds to teams, those teams would presumably sell the stablecoins to fund their respective initiatives. With this in mind, additional work for the DAO entails (1) developing a strategy for paying down the stablecoin debt.
Let’s say, a viable strategy is using sequencer revenues. Therefore, the DAO or its relevant committees could (2) establish relevant KPIs for funded projects, such as generating a certain amount of sequencer revenues. Intuitively, this model creates a framework for how to approach funding projects, because it is in the DAO’s interest to only fund projects that can generate sequencer revenues and thus pay down the vault’s debt.
Other work may involve:
- Performing more research and another risk analysis on the potential stablecoin protocols used
- Reach out to existing protocols and collaborate on setting up a position specifically for the DAO
- Ensuring that the DAO has robust governance mechanisms in place to make adjustments and respond to market changes.
- Create and maintain governance policies for handling extreme market conditions and emergencies
- Implement insurance funds or hedging strategies to cover potential losses
- Explore introducing less volatile assets to TBV such as ETH or wstETH
Risk Analysis
View the full risk analysis here.
In conjunction with Chaos Labs, we analyzed ARB and its market capitalization, trading volume, historical volatility, and historical liquidity to derive key parameters and thresholds for using ARB in a CDP. We also analyzed the ARB/USD oracle because price updates are crucial for maintaining a healthy vault.
We recommend an initial liquidation threshold (i.e. where a CDP loan becomes liquidatable) at 66% of the dollar value of ARB supplied. We also recommend a 10% liquidation bonus awarded to incentivize liquidators to avoid undercollateralized positions and to ensure that the CDP protocol avoids bad debt.
We recommend an initial loan of 25% of the value of ARB supplied to limit the liability of the DAO to supply more ARB collateral should the value drop. We also provide a more aggressive approach of taking an initial loan of 44% of the value of ARB supplied. In either case, we would need to set processes in place to manage the increased risk of getting the loan liquidated, and at a much less favorable ARB price in the latter approach.
We recommend an initial debt ceiling of $12m. This means that the DAO could borrow up to $12m safely against a minimum of $48m in ARB collateral at the outset.
We also recommend the DAO establish a ring-fenced safety buffer in ARB to cover any price drops in excess of what is forecast. More work is needed here to arrive at how much the DAO would need to allocate to a safety buffer fund. However, it is worth noting that the DAO should weigh the opportunity costs of deploying a safety buffer (i.e. ARB sitting in a safety buffer, in addition to ARB as collateral in a vault, is not being used for other initiatives).
Recommendations and Interpretations
For now the ARDC covers parts of the R&D committee–it does not develop or manage smart contracts–but given the current term is over in September, the DAO will need to revisit funding another ARDC term, or discuss contracting an entity that serves a similar role or is solely responsible for servicing a CDP or TBV.
As far as the other roles, the oversight committee is the organization responsible for overseeing the manager of the vault, and the risk management committee is an entity responsible for managing the risks of the TBV including updating parameters and topping up the debt position when necessary. Note: it is possible the risk management committee is also the treasury manager.
In terms of committee assignments and therefore initializing a CDP or TBV, we think the DAO should first take into consideration other initiatives to avoid introducing potential operational redundancies. For example, the Arbitrum Ventures Initiative is a WIP that would establish a fund structure for the DAO within which an oversight committee is appointed and verticals for allocating funds to specific strategies, such as operating a treasury or a TBV, are established. There is also an open proposal by Karpatkey and discussions about creating a Arbitrum Strategic Treasury Management Group and a DAO Oversight Committee. In short, we believe structuring the broader framework and dedicating attention to these conversations takes precedence, because they would establish a formal foundation on top of which the DAO can install a TBV. They would provide closure on who is servicing which roles and what additional committees (if any) are necessary.
If the DAO wishes to proceed toward a TBV, we recommend opening applications for participants who want to service the committees. This would allow candidates to outline their previous experience and framework. The proposal should also outline the scale of the initiative and define relevant KPIs to measure performance and ROI. From here, the DAO can narrow its focus for which protocol we’d use for even further R&D.
In terms of the risks, given the aforementioned recommendations, it is clear that borrowing stablecoins against an ARB position is extremely risky and would not satisfy a majority of the DAO’s annual spend.
From 2023 through 2024, the DAO spent 433M ARB–this is about 25x greater than the recommended initial loan amount. Secondly, using the conservative settings from the analysis, if we apply the previous year’s spending, then this means the DAO would have to seed the position with roughly 1.6B ARB (~$1.08B) to fund past initiatives. This is approximately half of the ARB in the treasury today, and the DAO would only be utilizing 25% of that value. At this scale, we can observe the highly capital inefficient use of ARB in a CDP or TBV.
Even at a much lower annualized spend by the DAO, we still view this as an inefficient use of capital, especially considering the opportunity cost of funding other projects that don’t carry as much inherent risk. In other words, we do not feel confident the reward for leveraging ARB clearly outweighs the risks (and associated costs) of operationalizing a DAO-owned CDP or TBV.
In terms of next steps, as mentioned above, we believe establishing a broader framework (perhaps related to the AVI), or making progress on installing a legitimate treasury management framework, takes precedence. Once this structure is set, then the DAO and the relevant committees will be in a better position to reconsider the prospects of activating a treasury-owned CDP.
Final Thoughts
All in all, while operating a TBV is worth considering, especially as the DAO and underlying ecosystem matures, it introduces another plane of risks and costs that feel unnecessary given where the protocol is in its life cycle. Instead, we feel confident the DAO should focus on growth and prioritizing initiatives that directly expand the use cases for the Arbitrum stack. In the grand scheme of rollups, it is still really early, and it is not clear what fundamentals drive the performance of rollup ecosystems and their underlying governance tokens.
For Arbitrum, there are a myriad of developments and catalysts on the horizon (such as Timeboost, decentralized Timeboost, blob bidding strategies, economic policies for L3s, ecosystem funds, and more) that should provide more context to identify strong business models and therefore perform the economic analysis for sustaining revenue and growth. Moreover, while the ARB price is an important feature of the DAO and protocol, it is a byproduct of success–not the driver–and we do not think initiatives directly addressing ARB price action properly prioritizes fundamental value drivers of the protocol at this moment.