PSP-IPΔX: Actively manage the Safety Module

Hi All,

I’m here representing Avantgarde Finance. We’re the lead maintainers of Enzyme, which has been integrated with Paraswap since our second iteration. The following proposal is a draft; we believe that it represents a low-risk opportunity to take active steps to optimise the Paraswap DAO’s asset management. We welcome the community’s feedback and are looking forward to some substantive conversation.

1. Proposal Number and Name
PSP-IPΔX: Actively manage the Safety Module

2. Keywords

Treasury management

3. Simple Summary

Use conservative but active asset management to earn yield on funds sitting in the Safety Module

4. Context

Paraswap generates fees through positive slippage (split 50/50 with the user) and revenue sharing agreements with external protocols that integrate Paraswap. Those fees are collected into a single contract called the Safety Module, which could be used to compensate users in the event of a loss due to mispricings in the protocol, or to pay for protocol maintenance. Because of the way in which fees are collected, the Safety Module contains a large variety of tokens, but much of the value in the contract is made up of a smaller number of tokens.

Current balances of the top assets in the Safety Module (as of 5 April 2021) are shown in the table below, rounded to the nearest big figure:

Token Price Balance Value
ETH and WETH $3475 66 $229,350
USD $1 127,000 $127,000
USDT $1 31,000 $31,000
DAI $1 23,000 $23,000
LOOKS $1.92 9,000 $18,000
CRV $2.90 3275 $9,500
WBTC $46,000 .19 $8750
CVX $36.50 240 $8760
SOS Very Small 3.6 billion $6350

We suggest that these tokens collected by the safety module should be converted into stablecoins on a regular cadence, or when the Module’s balance of a particular coin hits a certain threshold, or both. As a conversation starter, we envision rebalancing the above portfolio entirely into DAI:

Token Price Balance Value Yield
DAI $1 461,210 $461,210 6.5%

We’ve also provided the current yields one could earn on that position by simply deploying them to provide liquidity on Curve and staking the LP tokens on Convex.

The Paraswap community must have visibility into the assets held by the Safety Module and that transactions with those assets adhere to the expectations the community lays out in this conversation. We believe that the best way to do this is to use an Enzyme vault.

Enzyme is the DeFi Operating System that enables users to build plug and play asset management strategies with dozens of protocols and hundreds of tokens. Enzyme vault smart contracts are highly configurable; among other things, vault owners can specify addresses that are allowed to deposit into the vault, addresses that are allowed to manage the assets inside a vault, and the DeFi protocols and assets with which those managers are allowed to interact. Furthermore, a vault can be configured to prevent trading entirely if a custom cumulative slippage threshold is breached. In other words, if a vault manager does enough bad trades, the Vault locks them out for a period of time.

All transactions that occur within a vault (e.g. depositing DAI into a Curve pool and staking the resulting LP tokens) are visible on-chain and via the Enzyme app, meaning that stakeholders and DAO constituents have instant and perpetual visibility into the actions of a vault manager and their resulting profit and loss.

Enzyme has been rigorously audited by top-tier firms and maintains a $400,000 bounty with Immunefi.

We propose that the Paraswap DAO creates an Enzyme vault into which the Safety Module can be the only depositor. Periodically, the relevant balances from the Safety Module should be traded for a stablecoin (DAI in the case above) and deposited into the vault, where they can then be deployed into yield-bearing strategies.

Though these allocations are not subject to change by the hour, it would be useful to be able to adjust them to changing market conditions (token prices, reward amounts, lending and borrowing rates) without making a proposal to governance. When the vault is configured, smart contract policies could be enabled to allow one wallet to manage the funds with very specific guardrails as detailed in the links above. In this way, the vault could be managed by a delegated third-party in a controlled way.

We propose that Avantgarde Finance is appointed as the delegated third-party manager for the Enzyme vault managing the Safety Module’s funds. By way of background, Avantgarde Finance has been heavily involved in the core research and development of Enzyme since the protocol’s governance was decentralised and the first version of its contracts were deployed to mainnet. The team is a mix of engineers and finance professionals with a deep understanding of DeFi market structure and its risks and opportunities.

Given the novelty of this approach to DAO asset management and taking into account the longstanding cooperation between Avantgarde Finance and Parawap Labs, we propose charging the following fees:

  • 0.50% fee on assets under management paid to Avantgarde
  • 0.25% protocol fee paid to the Enzyme protocol

both payable in shares of the vault and enforced by the Enzyme fee contracts.

These fees will allow Avantgarde to cover the bare expenses required to manage the vault.

It’s important to note that as the Vault owner, the Paraswap DAO will always be fully in control of the addresses to which it delegates trading privileges and the fees it pays in the form of Vault shares to those addresses. If it is the case that Avantgarde Finance is not living up to our end of the bargain, those permissions can be revoked and future fee payments terminated.

5. Goals

This proposal aims to preserve and grow the balance of the Safety Module and hence its usefulness as a tool to protect Paraswap’s users and grow the protocol. Assuming that the costs that will need to be borne by this account are code audits and user payouts in the case of loss, holding an assortment of long tail tokens is difficult to risk-manage and of limited use. By implementing a rules-based conversion of tokens into stablecoin, we eliminate much of the market risk, and by using those stablecoins to generate yield, we grow the ability of the Stability Module to protect Paraswap’s users.

6. Means

This proposal does not require development work or resources from the DAO. The Enzyme smart contracts have been live on mainnet since February 2019, and subject to continual development from various teams coordinated by the Enzyme Council DAO. The most recent audits are available here.

7. Metrics

Enzyme’s vault contracts ship with real-time financial reporting capabilities and full transparency into the transactions a vault has conducted. This means that given a set of KPIs, it will be easy to measure success or failure. In general we believe that in the current environment, a 5% annual return target encourages prudent risk management and represents a fairly meaningful increase in the utility of the funds within the Safety Module. This is easily observable; if at the end of one year the Vault’s share price is greater than 1.05 DAI, it has returned > 5%.

8. Forward-thinking considerations

It’s possible (and likely) that new protocols will emerge which will provide new opportunities to earn yield on Safety Module assets after this proposal is implemented. If the Safety Module’s Enzyme Vault is configured to prohibit interaction with all but a certain subset of DeFi protocols, it will not be able to utilise these new protocols out of the box; doing so will require a reconfiguration of the Vault, which should be the subject of a DAO vote.

Also of note; we have observed the ongoing conversation about acquiring CVX and potentially veBAL to influence liquidity mining rewards on Curve and Balancer. That functionality will also be available within an Enzyme vault. While it’s likely not a good idea to commingle assets across different purposes (e.g. managing Safety Module assets in a Vault that’s also voting in Convex), a familiarity with Enzyme’s capabilities will be helpful in deciding how to execute on those meta-governance mechanisms if the DAO decides to move forward with them.

9. Implementation Overview

Given the preliminary nature of this conversation, below is a list of the relevant parameters that will need to be decided to properly configure the Vault along with our suggestions for those parameters. Following that is a high-level overview of the steps to deploy and seed it with funds.

Vault Parameters

Denomination Asset: DAI

Management Fee: 0.50%

Performance Fee: n/a

Permissioned protocols: Paraswap, Curve, Convex

Cumulative slippage tolerance: 5%

Delegated Trader Permissions: avantgardefi.eth

Cadence for sweeping funds from Safety Module to Enzyme Vault: Quarterly if above $5,000 in position size

Target vault exposures: 100% stablecoin

Deploy and Seed Vault

  • Configure vault with appropriate policies as discussed with the Safety Module Gnosis Safe
  • Trade all assets above the value threshold discussed above for the denomination asset
  • Deposit denomination asset balance into vault

Managing Vault

  • Optimise for yield opportunities within parameterized constraints and maintain target asset exposures
  • Coordinate with the DAO for conversion of long-tail assets and sweeping of funds from Safety Module to Enzyme

10. Voting Options

  • For
  • Against
  • Abstain

Thanks for the thought out proposal!

Considering the importance of the Safety Module towards generating value to the DAO, I find that this could be an interesting way of making sure that the money does not simply remained unused.

Although I know it is meant to be a conversation starter, I do find the idea of allocating the entire safety module’s funds towards this managing could be a bit risky. However, as you mentioned, it could be a very useful idea to convert ‘dust’ assets that do not serve the treasury right now, so anything that does not help with either securing the protocol, capturing protocol owned liquidity or govenance/CVX war assets.

Personally, I feel that a good way to go about implementing this would be to first release a ‘pilot’ period in which a certain amount of DAI is sent. Then , depending on how the situation evolves, we could allocate more as KPIs are met. If the stablecoin strategy is accomplished, then we could deploy a new one for non-stablecoin management. Thoughts? :slight_smile:



Florian from Synaps & Anima. We just launched our decentralized identity protocol (go check it out).
We are both private sale investors in Paraswap and in Unslashed.
I believe that Enzyme is a good solution to not keep the funds idle. And they demonstrated being a serious actor with other protocols.

At a personal level, I would keep the ETH and probably try to generate some yield on them (stake them on Lido as a simple strategy?). With the merge coming, turning ETH in stablecoins now might create some regrets in the future. But it’s debatable as I understand the treasury management of a protocol shouldn’t be the same that the one of an individual.

The management fee doesn’t seem too high as the funds are for the moment quite limited. It could be put in the proposal to put different tiers according to the AUM.

$461k with half in ETH and half in stablecoins with 2 passive strategies to avoid the high fees of ETH that could impact the performance seems like a good way to start for me.


Thanks for the well-written and thoughtful proposal. An actively managed stablecoin strategy could greatly optimize the performance of the Safety Module over the mid-long run, benefiting the greater ParaSwap ecosystem. With the stablecoin strategy outlined in the proposal, ParaSwap’s Safety Module will morph from a non-yield bearing treasury into a DAO equipped with a backstop made up of a productive, yield-bearing asset.

As noted in the Section 8, we think the DAO may have ample opportunities to increase the yield of their stables strategy if earning rewards throughout DeFi using protocols like CVX, BAL, or Votium.

We’re longtime investors in Avantgarde. We know the team well and think highly of them. Enzyme’s smart contract vaults enable dynamic management configurations and can be tailored to fit the needs of the ParaSwap community. Also, the fact that all management transactions take place on-chain will provide added transparency for the DAO. The Enzyme team is solid and the platform has a proven track record as a top choice for on-chain asset management.

We think this proposal could mark a new era for the ParaSwap community if adopted. The Safety Module is a critical piece of infrastructure for the DAO and should be managed in a way that provides enduring value to the ParaSwap community.

1 Like

Just wanted to pop in here and say that we’ve seen the comments (thanks @collab_currency , @Flo_anima and @0xYtocin). We’re doing a bit of thinking over the weekend on how to best move this forward and will revert early next week with potential next steps.

1 Like

Perfect! If you need any help with working on the proposal, do not hesitate to reach out, I’m more than willing to help.