[BIP-162] Karpatkey Investment Strategy

Balancer DAO: Assessment & Proposal


Karpatkey has been tasked by the Balancer DAO, as per BIP-103, with deploying yield-generating strategies using the existing funds in the treasury in a risk-adjusted, trust-minimized, and non-custodial manner. The following paragraphs outline the general approach that will be followed and provide insight on the following topics:

  • Assumptions used to determine the funds within the scope of Karpatkey
  • Initial asset composition
  • Allocation proposals for the Balancer DAO to vote on
  • Projected results

Links to Balancer DAO wallets

Assessment of current portfolio allocation

The current portfolio allocation includes 86% idle assets and 14% invested in strategies generating approximately $86,000 per year. Most of these strategies do not seem to be focused on generating returns and some may even be in negative territory if Impermanent Loss is considered.

# Protocol Position Chain Balance $ Share APR Rev. $
1 Wallet BAL Mainnet $25,692,025 77.05% 0.00% $0
2 Wallet USDC Mainnet $2,017,695 6.05% 0.00% $0
3 Wallet AAVE Mainnet $908,259 2.72% 0.00% $0
4 Wallet WETH Mainnet $20,333 0.06% 0.00% $0
5 Ribbon BAL Mainnet $1,037,083 3.11% 0.00% $0
6 Balancer V2 Balancer Boosted Aave USD Mainnet $693,876 2.08% 0.00% $0
7 Balancer V2 WBTC + ETH Mainnet $269,750 0.81% 0.35% $944
8 Balancer V2 D2D + BAL Mainnet $58,432 0.18% 0.14% $82
9 Balancer V2 ETH + GTC Mainnet $34,356 0.10% 0.00% $0
10 Gnosis GNO Mainnet $758,926 2.28% 0.00% $0
11 Lido ETH (stETH) Mainnet $375,714 1.13% 4.40% $16,531
12 Lido ETH (wstETH) Mainnet $205,369 0.62% 4.40% $9,036
13 Aave BAL Mainnet $537,388 1.61% 3.40% $18,271
14 Across V2 BAL Mainnet $212,413 0.64% 0.49% $1,037
15 Silo BAL Mainnet $88,030 0.26% 19.38% $17,060
16 mStable MTA Mainnet $14,135 0.04% 3.17% $448
17 Balancer V2 MATIC + USDC + WETH + BAL Polygon $263,512 0.79% 4.45% $11,726
18 Balancer V2 WBTC + WETH + USDC Arbitrum $157,217 0.47% 6.94% $10,911
Total $33,344,512 100.00% $86,047


Karpatkey has made certain assumptions regarding the current allocation of funds in the Balancer DAO’s portfolio in order to determine which assets and strategies are in scope for generating returns. These assumptions include:

  1. Approximately 77% of the total funds, or around 5 million BAL, are idle on the wallet and will not be deployed in any strategies in order to preserve the voting power of the main stakeholders.
  2. Following the stated criteria, four positions involving BAL that are the result of strategic partnerships with other parties, as well as the BAL position on Ribbon Finance and the MTA position on mStable, will not be included in any strategies;
  3. The ETH portion of the ETH-GTC position on Balancer will be included, while the GTC portion will not;
  4. The 16,907 AAVE on the wallet, which are part of an ongoing agreement with AAVE, will be included and potentially deployed through staking or lending strategies;
  5. The USDT + DAI + USDC and WBTC + ETH positions on Balancer, and the stETH and wstETH positions on Lido, can be dismantled and redeployed;
  6. The MATIC + USDC + WETH + BAL and WBTC + WETH + USDC positions on Balancer v2 on Polygon and Arbitrum, respectively, can be dismantled, bridged back to the mainnet, and redeployed, with the exception of the BAL and MATIC assets which will not be included;
  7. Any risk-adjusted protocol or strategy can be proposed for approval and whitelisting, except for those that use Balancer products or yield BAL as part of their rewards structure;
  8. The portfolio’s asset composition will not be changed, and assets will only be used for yield-generating strategies.

New allocation proposal

Now that the assumptions have been made, Karpatkey identifies the initial scenario in terms of assets that are in scope and introduces two proposals for the Balancer DAO to consider and vote on. The first proposal, known as the “base allocation”, suggests using well-known staking, lending, and AMM protocols in order to keep the complexity of the strategies low and facilitate risk management. The second proposal, known as the “expanded allocation”, suggests expanding the list of protocols to include fixed-rate lending and other structured product categories in order to broaden the range of possibilities and allow for more complex strategies that are expected to yield better returns. The two allocation proposals provide specific recommendations for how these assets should be deployed in order to generate returns.

Initial Scenario

The following table shows the assets that are considered in-scope by Karpatkey:

Token Category Protocol Position Chain Balance Price Total Value $ Share
USDC Stablecoins Wallet - Mainnet 2,017,695 $1.00 $2,017,695 42.16%
AAVE AAVE Wallet - Mainnet 16,907 $53.72 $908,259 18.98%
WETH ETH Wallet - Mainnet 17 $1,218.18 $20,333 0.42%
USDT Stablecoins Balancer v2 Balancer Boosted Aave USD Mainnet 328,055 $1.00 $328,055 6.85%
DAI Stablecoins Balancer v2 Balancer Boosted Aave USD Mainnet 180,329 $1.00 $180,329 3.77%
USDC Stablecoins Balancer v2 Balancer Boosted Aave USD Mainnet 185,492 $1.00 $185,492 3.88%
WBTC WBTC Balancer v2 WBTC + ETH Mainnet 8 $16,688.22 $134,841 2.82%
ETH ETH Balancer v2 WBTC + ETH Mainnet 111 $1,218.25 $134,909 2.82%
ETH ETH Balancer v2 ETH + GTC Mainnet 6 $1,218.25 $6,761 0.14%
stETH ETH Lido Staking Mainnet 312 $1,203.21 $375,714 7.85%
wstETH ETH Lido Staking Mainnet 155 $1,327.36 $205,369 4.29%
USDC Stablecoins Balancer v2 MATIC + USDC + WETH + BAL Polygon 65,599 $1.00 $65,599 1.37%
WETH ETH Balancer v2 MATIC + USDC + WETH + BAL Polygon 54 $1,218.18 $65,721 1.37%
WBTC WBTC Balancer v2 WBTC + WETH + USDC Arbitrum 3 $16,688.22 $52,401 1.09%
WETH ETH Balancer v2 WBTC + WETH + USDC Arbitrum 43 $1,218.18 $52,382 1.09%
USDC Stablecoins Balancer v2 WBTC + WETH + USDC Arbitrum 52,434 $1.00 $52,434 1.10%
Total $4,786,294 100.00%

Base allocation

If this proposal was selected, specific functions within the protocols below would be whitelisted. More technical details regarding the whitelisting process will be provided shortly.

  • Staking:
    • Lido
    • Stakewise
  • Lending:
    • Compound v2
    • Aave v2
  • AMM:
    • Uniswap v3

It is worth noting that the above protocols and categories are considered to be the most battle-tested and proven yield generation alternatives in the industry. Therefore, we consider them to be aligned with the objectives, scope and time orientation of the Balancer DAO.

Should existing funds be deployed within the base alternative, the allocation would be as follows:

# Protocol Position Balance $ Share of Portfolio APR % Proj. Rev $ Share of Rev.
1 Lido Stake ETH $404,894 8.46% 4.40% $17,815 10.78%
2 Uniswap v3/Stakewise SETH2/ETH $262,243 5.48% 14.35% $37,632 22.77%
3 Aave v2 AAVE on Safety Module $454,130 9.49% 8.41% $38,192 23.10%
4 Compound v2 Deposit AAVE $454,130 9.49% 3.19% $14,487 8.76%
5 Uniswap v3 WBTC + ETH, Range: 11.786 - 15.082. Fee: 0.3%. $381,294 7.97% 2.49% $9,494 5.74%
6 Compound v2 USDC $1,414,802 29.56% 1.61% $22,778 13.78%
7 Compound v2 DAI $1,414,802 29.56% 1.76% $24,901 15.06%
Total $4,786,294 100.00% 3.45% $165,299 100.00%

The expected revenues for the current deployment of approximately $4,800,000 are approximately $165,000. The total current revenues of the Balancer DAO are $86,047, but this includes assets outside the scope of this proposal. For a fair comparison, the proposed expected revenues should be put up against the current revenues generated by the in-scope assets of the Balancer DAO, which are $49,149. When only considering the in-scope assets, the expected revenues from the base allocation proposal represent more than three times the current portfolio performance.

Expanded allocation

In pursuit of tapping into other attractive opportunities that DeFi has to offer, here is the expanded alternative with new protocols and categories.

  • Staking:
    • Rocket Pool1
  • Lending:
    • Euler Finance1
  • Fixed-rate lending:
    • Element Finance
    • Notional Finance
    • Idle Finance1
  • Other structured products:
    • Ribbon Finance1

Should existing funds be deployed within the expanded alternative, the allocation would be as follows:

# Protocol Position Balance $ Share of Portfolio APR % Proj. Rev $ Share of Rev.
1 Lido Stake ETH $202,447 4.23% 4.40% $8,908 4.85%
2 Uniswap v3/Stakewise SETH2/ETH $262,243 5.48% 14.35% $37,632 20.51%
3 Element Finance ePyvcrvSTETH/steCRV LP $202,447 4.23% 7.00% $14,171 7.72%
3 Aave v2 AAVE on Safety Module $454,130 9.49% 8.41% $38,192 20.81%
4 Compound v2 Deposit AAVE $454,130 9.49% 3.19% $14,487 7.90%
5 Uniswap v3 WBTC + ETH, Range: 11.786 - 15.082. Fee: 0.3%. $381,294 7.97% 2.49% $9,494 5.17%
6 Compound v2 USDC $914,802 19.11% 1.61% $14,728 8.03%
7 Compound v2 DAI $1,414,802 29.56% 1.76% $24,901 13.57%
8 Notional Finance USDC - Maturity: Dec 18, 2023. $500,000 10.45% 4.20% $20,975 11.43%
Total $4,786,294 100.00% 3.83% $183,488 100.00%

In this case, the expected revenues are more than $183,000, which is almost four times the current portfolio performance.

There’s also potential to continue expanding the portfolio by tapping into strategies that involve leverage. Protocols like MakerDAO, Reflexer, Liquity and QiDAO offer the possibility to provide assets as collateral and borrow stablecoins to be used for carry-trade strategies that yield a net positive result. However, we leave the analysis of these possibilities for a later stage.

1 not directly proposed in the expanded allocation strategy but could be added eventually for diversification purposes.


Overall, implementing risk-adjusted yield-generating strategies can significantly improve the performance of the Balancer DAO’s portfolio. An analysis of the before and after scenarios is provided below.

Before - Current Allocation
Total Funds

$ 4,786,294

Utilized Funds

$ 1,840,007

Capital Utilization



$ 49,149

Avg. APR


2 For comparison purposes, only revenues generated from assets within scope are considered.

After - Base Allocation
Total Funds

$ 4,786,294

Utilized Funds ($k)

$ 4,786,294

Capital Utilization


Revenues ($k)

$ 165,299

Gross APR


Management Fee (calculated over AUM)


Performance Fee (calculated over revenue)


Net Revenues

$ 108,308



After - Expanded Allocation
Total Funds

$ 4,786,294

Utilized Funds ($k)

$ 4,786,294

Capital Utilization


Revenues ($k)

$ 183,488

Gross APR


Management Fee (calculated over AUM)


Performance Fee (calculated over revenues)


Net Revenues

$ 122,859



Overall, the proposed strategies are expected to generate an annual net result for the Balancer DAO, ranging from $108,000 to $122,000.

Additional considerations

  • Rewards management: all generated rewards will be either swapped to ETH or stablecoins (at Karpatkey’s discretion) and redeployed into existing strategies to compound the results.
  • New SAFE multisig: to better organize the assets within the scope of Karpatkey, we suggest creating a new SAFE wallet (with the same signers as the current one) and transferring the in-scope assets to it. The Balancer DAO will not give up custody of the funds and the Manager Roles will be applied to the new SAFE, keeping out-of-scope funds separate.

Did you use Balancer v2 managed pools or weighted pool implementation?

Hi 0xEvan! I am Santi from Karpatkey.

Thank you for your question. However, I’m not sure I fully understand it.

Just to clarify, the assumption 7 states that any strategies that use Balancer products or yield BAL as part of their rewards structure will not be used. On the other hand, strategies that include Balancer pools that have already been deployed by the Balancer DAO will either be re-deployed or left out-of-scope.

If you meant something different, please let me know and I will try to help further.

1 Like

My apologies, upon re-reading, my question doesn’t make any sense. Please disregard thanks!

1 Like

Hey guys, thoughtful proposal as always.

One easy one for you at this time: Why allocate to Uniswap and Compound rather than promote usage on Balancer itself and partner Aave?

Hey John! Thank you for your excellent question.

Balancer is not recommended for yield generation because it is not appropriate for the protocol to farm its own rewards. The incentives are meant to encourage liquidity, so they should be reserved for that purpose. Currently, Uniswap offers attractive opportunities that align with the goal of maximizing return while minimizing risk.

In regards to Aave, the current reward opportunity for stablecoins is superior on another reputable and tested protocol, Compound, which has a TVL of over $400 million for stablecoin pools. In addition, the proposal recommends allocating $AAVE to the Aave’s Safety Module, which is a way of directly supporting the success of the protocol.


Does Aura fall under this same rationale, in your opinion?

Yes, it does. That’s the reason why it was not included within the allocation proposal.

This is a really thoughtful proposal by Karpatkey and should benefit the DAO well. Two main questions/considerations:

  1. Is the juice worth the squeeze here? An extra $50k-70k of rev is substantial, however, how much work is the BAL team putting into managing the program? Only they can answer that - I tend to think it’d be worth it…

  2. How are the weights for each strategy determined?

1 Like

Thank you for taking the time to provide your comments and questions. We appreciate your engagement and insight.

1- The concern of workload is a valid one and we understand the importance of balancing potential benefits with effort. We would like to assure you that with the implementation of Karpatkey’s non-custodial management of the treasury, which will be limited by the whitelisted actions allowed by the Zodiac Roles Modifier, the workload for the BAL team related to Treasury Management will be significantly reduced. Besides, after conducting a deep dive with Karpatkey’s technical team, we have identified additional improvements that can further alleviate the workload during the ramp-up stage, allowing for a faster and simpler deployment. We will share an update on the outcome of these discussions shortly.

2- The weights for each strategy are determined based on the assets currently held by the DAO and the existing strategies. In accordance with Assumption #8, we aim to maintain the composition of the portfolio unless there is a specific reason to make changes. We have suggested the allocation by optimizing for the minimum number of required transactions and changes.

1 Like

Update on Proposed Approach and Request for Snapshot Voting


In order to initiate operations as per BIP-103, Karpatkey respectfully requests the Balancer DAO to take the following actions:

  1. Dismantle specific positions on the current multisigs on Mainnet, Polygon, and Arbitrum (as outlined in the Execution Details section);
  2. Bridge specific in-scope assets from Polygon and Arbitrum back to Mainnet (as outlined in the Execution Details section);
  3. Transfer all in-scope assets (totaling approximately $5,900,000 as of the day of this writing) to a new SAFE provided by Karpatkey and owned by the Balancer DAO (as outlined in the Next Steps and Execution Details sections).


After internal discussion, the Karpatkey team has determined that the Balancer DAO would be best served by a simpler and more direct approach. Since Karpatkey has already been voted in as the treasury manager, there is no longer a need for further discussion or a new vote on the specific allocation of funds, despite having recently proposed the “Base vs Expanded allocation” approach.

To streamline deployment, Karpatkey proposes beginning with the base allocation and then expanding as needed. This approach takes advantage of the Zodiac Roles Modifier permissions preset that has already been created and tested for allowlisting specific functions and protocols related to the base allocation. Karpatkey intends to move forward with this approach and will propose future adjustments in accordance with the guidelines set forth in the original proposal. However, any future adjustments to the strategy, and thus the preset, will ultimately require the approval of the Balancer DAO through the execution of a batch of transactions to apply the adjusted preset.

Zodiac Roles Modifier preset to be applied

The original post proposed the following:

To increase flexibility and adapt to changing market conditions, Karpatkey suggests expanding protocols and functions to include ETH wrappers and unwrappers, swaps on Balancer and Uniswap v3. These swaps, except for regularly swapped reward tokens, are intended to provide Karpatkey with the ability to take immediate action in specific market conditions, such as a stablecoin losing its peg. The pools were chosen based on liquidity to maximize results for the Balancer DAO, but new permissions may be necessary as market conditions change.

The Manager preset consists of the complete list of permissions that Karpatkey asks the Balancer DAO to grant. This list can be found in this Annex Document. Through the Zodiac Role Modifier, the Manager preset will be applied to the Manager Role, which in turn will be assigned to Karpatkey’s Manager SAFE.

Future permissions and roles

Once the initial strategy is deployed, there will be a periodic need to adapt the portfolio to changing market conditions. If a particular strategy no longer provides sufficient yield, or if a particular pool in which the Balancer DAO has a position becomes deprecated and needs to be migrated, new permissions will be required.

In addition, creating new roles to improve or expand operations will also be necessary. Two new roles are planned for the near future: the Harvester Role and the Swapper Role. Both roles will be granted specific subsets of the Manager preset permissions and will be assigned to EOAs to automate certain tasks. The Harvester Role will be allowed to claim rewards, while the Swapper Role will be allowed to swap these rewards for ETH or stablecoins.

When deemed appropriate, Karpatkey will create a new forum post requesting the Balancer DAO to update the preset or deploy new roles. The post will include an explanation of why the permissions and/or roles are needed and the executable payload. The proposal will need to be passed by the community and, once approved, it will be executed by the multisig according to the DAO standard operating practices. It is important that this process is as simple and straightforward as possible, since every day of waiting may represent an opportunity cost to the DAO’s treasury. A summary of the new permissions and/or roles will be included as a separate section in every corresponding weekly report.

Next steps

The next steps for implementing this proposal are as follows:

  1. Karpatkey will create a new SAFE, deploy the Roles Modifier contract, and apply the Manager Role preset (as outlined in the Annex Document) to simplify the process for the Balancer DAO.
  2. Karpatkey will conduct a small-scale replication of the strategy for a second round of testing.
  3. Once the SAFE is created, tested, and permissions are applied, Karpatkey will transfer ownership of it to the Balancer DAO and remove themselves as owner.
  4. This proposal will be submitted to Snapshot for approval.
  5. If the proposal is approved, the Balancer DAO will execute the steps described in the Execution Details section.

Execution details

The following list is intended to serve as a clear execution roadmap for the Balancer DAO once (if) this proposal is approved:

  1. Dismantle the Balancer Boosted AAVE USD position on Mainnet and withdraw 100% of the funds in DAI.
  2. Dismantle WBTC + ETH position on Mainnet and withdraw the two assets in equal proportion.
  3. Dismantle ETH + GTC position on Mainnet and withdraw the two assets in equal proportion.
  4. Dismantle MATIC + USDC + WETH + BAL position on Polygon and withdraw the four assets in equal proportion.
  5. Bridge USDC and WETH from Polygon back to Mainnet.
  6. Dismantle WBTC + WETH + USDC position on Arbitrum and withdraw the three assets in equal proportion.
  7. Bridge WBTC, WETH + USDC from Arbitrum back to Mainnet.
  8. Transfer in-scope assets to the new SAFE. The approximate balances (as of the day of this writing) of in-scope assets that will be transferred are:
Token Price Approx Balance Total Value Share %
USDC $1.00 2,514,827 $2,514,827 42.61%
AAVE $79.50 16,907 $1,344,129 22.77%
stETH $1,542.51 312.86 $482,590 8.18%
WETH $1,552.80 266.81 $414,301 7.02%
wstETH $1,707.08 155.02 $264,632 4.48%
DAI $1.00 239,996 $239,996 4.07%
WBTC $21,007.20 11.06 $232,289 3.94%
USDT $1.00 218,949 $218,949 3.71%
ETH $1,553.00 122.82 $190,733 3.23%
Grand Total $5,902,446 100.00%


Karpatkey has prepared the Gnosis safe on Ethereum at 0x0EFcCBb9E2C09Ea29551879bd9Da32362b32fc89. It currently has the same signers as the DAO Multisig at 0x10A19e7eE7d7F8a52822f6817de8ea18204F2e4f, aside from an extra 12th signer with address 0x7f87c1C42BeF332245F8B3cCAdD8224541CDaEcE.

The DAO Multisig signers will reject tx #38 which is failing on simulation and initiate a new tx to remove this 12th signer.

The DAO Treasury on Polygon at matic:0xeE071f4B516F69a1603dA393CdE8e76C40E5Be85 will initiate a transfer of “B-Polybase BPT” (contract 0x0297e37f1873D2DAb4487Aa67cD56B58E2F27875) in the amount of 26536547394777662718278 to the LM Multisig at matic:0xc38c5f97B34E175FFd35407fc91a937300E33860. From there, BPT will be withdrawn and all funds bridged back to Ethereum. USDC & WETH will be sent to the Karpatkey Safe 0x0EFcCBb9E2C09Ea29551879bd9Da32362b32fc89, BAL & WMATIC sent to the Treasury 0x10A19e7eE7d7F8a52822f6817de8ea18204F2e4f

The DAO Treasury on Arbitrum at arb1:0xaF23DC5983230E9eEAf93280e312e57539D098D0 will initiate a transfer of “B-33WETH-33WBTC-33USDC” (contract 0x64541216bAFFFEec8ea535BB71Fbc927831d0595) in the amount of 532147981134526165239 to the LM Multisig at arb1:0xc38c5f97B34E175FFd35407fc91a937300E33860. From there, BPT will be withdrawn and funds bridged back to Ethereum. All funds will be sent to the Karpatkey safe 0x0EFcCBb9E2C09Ea29551879bd9Da32362b32fc89

The DAO Treasury on Ethereum eth:0x10A19e7eE7d7F8a52822f6817de8ea18204F2e4f will send the following assets to the LM Multisig eth:0xc38c5f97B34E175FFd35407fc91a937300E33860, which will withdraw the BPTs and send the assets to the Karpatkey Safe 0x0EFcCBb9E2C09Ea29551879bd9Da32362b32fc89, except where indicated otherwise:

Aave v2 boosted pool (contract: 0x7B50775383d3D6f0215A8F290f2C9e2eEBBEceb2) amount: 682682196629189057534951
WBTC/WETH BPT (contract 0xA6F548DF93de924d73be7D25dC02554c6bD66dB5) amount: 58598401025232390138
WETH/GTC BPT (contract 0xfF083f57A556bfB3BBe46Ea1B4Fa154b2b1FBe88) amount: 7725430893893258147597 Note: GTC will be returned to the DAO Treasury.

The DAO Treasury on Ethereum eth:0x10A19e7eE7d7F8a52822f6817de8ea18204F2e4f will send the following assets to the Karpatkey Safe 0x0EFcCBb9E2C09Ea29551879bd9Da32362b32fc89:

USDC (contract 0xA0b86991c6218b36c1d19D4a2e9Eb0cE3606eB48) amount: 2139935999712
Aave (contract 0x7Fc66500c84A76Ad7e9c93437bFc5Ac33E2DDaE9) amount: 16907280000000000000000
stETH (contract 0xae7ab96520DE3A18E5e111B5EaAb095312D7fE84) amount: 312988890260928536597
wstETH (contract 0x7f39C581F595B53c5cb19bD0b3f8dA6c935E2Ca0) amount: 140236048902346440000
weth (contract 0xC02aaA39b223FE8D0A0e5C4F27eAD9083C756Cc2) amount: 61323265453850206284


Hi @karpatkey,

It was nice to see this post and it is a solid effort, well done.

Reading over the post, we have a few observations which might be of interest. Our feedback is intended to be constructive and with Balancers best interest front of mind. We do note Llama is heavily involved in the Aave ecosystem, which means we are informed, but also something that must be taken into consideration when reviewing the below. Hopefully this feedback is welcomed and received with the intended spirit of being pro finding the best outcome for Balancer.

From BIP-52, 100k of BAL was deposited into the Polygon v3 and Ethereum v2 Aave deployments. The current portfolio section above mentions mainnet but appears to be over looking the Polygon v3 deposit. The images below show the two positions, overlooking some dust, it shows $710k of BAL on Polygon v3 and $700k on Ethereum v2. We only checked to two mentioned wallets in BIP-52.

Reading over the assumptions, regarding 1. the deployment of BAL is something we perhaps share differing opinions. The community has demonstrated interest in deploying these funds. This acts to bring BAL back to the DAO.

More importantly, on Aave v2 & v3 there is a function known as credit delegation. Credit delegation allows a depositor to deposit funds in the protocol to earn interest, and delegate borrowing power (i.e. their credit) to other users.

If Karpatkey is planning on using debt, it is mentioned in the Expanded Allocation, then credit delegation is a means of de-risking these strategies ensuring the position is heavily over collateralized. We encourage using BAL as collateral on Aave either to borrow against directly, or to delegate the credit to another wallet which has a debt position.

We also suggest increasing the BAL deposit on Aave v3 Polygon, another ~44k of BAL will bring deposits up to the SupplyCap. This will generate an additional 5.5% at current spot rates, BAL $6.90, equivalent to $16.7k. This is equivalent to a 10% increase in Base Allocation returns. When the interest rate increase for BAL go through, this could lead to even higher deposit yield. If this BAL finds its way into liquidity pools, this is beneficial increasing the SupplyCap enabling more BAL to flow into Aave.

Regarding assumption 4) we support AAVE being deployed and we are glad to see the AAVE being deposited into the Safety Module. We kindly highlight the following about Aave:

Using Karpatkey’s numbers there is an additional 5.22% yield to be earned on Balancer’s third largest holding. We would encourage depositing all AAVE into the Safety Module within both Base Allocation and Expanded Allocation portfolios. Using the AAVE values in the original post, this will generate an additional $23,705.586 ($454,130*(8.41%-3.19%)). This is material given the overall return is $165,299. Also, do note AAVE’s price has increased a lot since this post and the APY from the Safety Module is AAVE price agnostic and sensitive only to volume of deposits. If all AAVE was deposited into the Safety Module, using spot pricing, this is equivalent to 6.05% * 16,907 * $85 = $93,411.18 in annual revenue, relative to the current strategy which would generate $11,579.89 from Compound v2 and $46,705.59 from Aave’s Safety Module for a total of $58,285.48. A delta of $35,125.7025.

We would also suggest looking into GHO and how stkAAVE leads to a discounted borrowing rate and given the BAL deposited on Aave above, if migrated to Ethereum v3 when BAL is listed, Katpatkey could borrow discounted GHO and help bring GHO to Balancer through providing liquidity. This leans into making Balancer the home of GHO. We recognise this may fall outside of scope of Treasury management and it would involve earning BAL rewards which should then be distributed back through the respective gauges.

Something to consider is icETH for a portion of the Expended Allocation portfolio, this leverage yield product held up incredibly well during market turmoil last year, it is battle tested. The underlying yield source is stETH yield - wETH debt, looped on Aave v2, it has consistently performed well over time. Currently yielding 8.41% and Index Coop is a builder on top of Balancer, we would encourage reaching out @funkmasterflex from Index Coop about this opportunity.

Following a similar theme of supporting communities who build on Balancer, we would suggest having a look at Sturdy Finance who received a grant from Balancer and Aura. InstaDapp’s earn strategies are also worth a look. A combination of these would replace the Uniswap exposure and generate a higher return without impermanent loss risk. Instadapp was battle tested during the stETH / ETH price divergence events last year which should generate confidence in the teams behind the Earn product offering. This also provides an alternative to using Compound v2, which is an obsolete product now with v3 being live for some time on Ethereum.

The above is a lot of ideas and worth considering. It was a solid post and it is easier to read with fresh eyes. We also recognise there was time spend away from the forum discussion this which may have led to some of the mentioned ideas being ruled out already.

Thank you for taking the time to read this post.


Snapshot has been queued with the updated specifications:

Hey Llama team! Thanks for your analysis. I had an engaging conversation with Matthew about this point. We adapted the strategy considering it. We will use the Safety Module because we want to keep the stkAAVE delegation. I think this action is essential for the AAVE governance process.

I consider this action attractive for Balancer. By the way, there are risks involved in using BAL tokens. Karpatkey has a commitment to the Balancer community. We can’t use BAL tokens on lending platforms such as AAVE and similar ([BIP-103])(Snapshot))

Hmmm. Balancer is Karpatkey’s client. Is is there a reason why Balancer might sacrifice a treasury plan that should otherwise generate more revenue in order to have votes in AAVE governance?

As the treasury provider for balancer, it is quite important that what’s best for Balancer comes first. This statement sounded a little bit like suboptimal decisions were perhaps being made, in order to strengthen Karpatkey’s delegation. I’d be interested to hear a bit more about where this thinking comes from, and why the Balancer Treasury should sacrifice yields/ROI for Aave votes. Do we need them if we have support of the fine Llamas? Do we need these votes now/soon for something?

Can you elaborate here?

Hey Tritium! We’ll deposit AAVE on the Safety module (APR=6.16%), we’re improving the current situation: exists AAVE idle. The stkAAVE will delegate to Llama again. We were analyzing the diversification between AAVE and Compound, so we decided to simplify the strategy by depositing AAVE in the Safety Module stkAAVE. We support and keep on going with the BIP-78. If we consider that another strategy is better for Balancer, we’ll find the best yield and implement the best risk management practices. We want the best for Balancer!


Thank you for the clarification. I reread through everything and this makes more sense now. 100% in support of putting our assets to work, and very excited about the concept of treasury management as a business.

I also think it’s great that you are delegating governance rights to another entity to avoid conflicts of interest and allow for specialization and focus. I missed that reading through all the BIPs.