Adding GARD lending market

Temperature Check
1

Proposal:

The GARD team suggests that $GARD should be added to AlgoFi’s lending market. This collaboration would add tremendous value to the Algorand’s DeFi ecosystem via boosted composability between two prominent protocols. We suggest integrating $GARD in phases to ensure that the AlgoFi lending market remains healthy and that $GARD in circulation can increase organically thanks to increased ecosystem partnerships.

What is $GARD?

$GARD is an overcollateralized stablecoin, similar to DAI, that can be minted and borrowed against intrinsically valuable and decentralized collateral.

To mint $GARD, users can use $ALGOs, $ALGO derivatives (e.g. $GALGO), or $ALGOs committed to governance via the GARD Protocol that would otherwise be locked in a user’s Algorand governance position.

As opposed to other competing stablecoins, $GARD is hard money and can now be staked natively on the GARD dApp to earn a share of the protocol’s revenue from borrow positions.

GARD Protocol’s Website: https://www.algogard.com
GARD Protocol’s dApp: https://www.gard.money

Considerations:

Since the launch of GARD’s v2, there has been a significant increase in the overall support for GARD from the Algorand DeFi community. The protocol has $7M+ TVL , over 1.3k+ users, and $GARD has maintained its $1 peg almost perfectly. Many AlgoFi users already hold $GARD.As adoption increases, there will be more value to AlgoFi’s lending market as well as to GARD’s platform which may cultivate opportunities for new Algorand DeFi participants.

As stated, $GARD needs the help of partnerships to increase its adoption. These partnerships will both bring in new users and help deepen liquidity. Liquidity is the lifeblood of DeFi; a lack of liquidity prevents markets from running at peak performance. The gradual integration of $GARD as an asset to be borrowed and to be used as collateral would bring more liquidity.

In order to help kickstart adoption, the GARD Protocol is happy to deposit $GARD onto the AlgoFi lending platform so that it may be borrowed via new forms of collateral and then deployed on GARD’s platform for staking/liquidating or on the broader DeFi ecosystem. This ability to deposit/borrow will create new opportunities for $GARD holders.

Another important consideration is whether to allow $GARD to be used as collateral on AlgoFi. Given the current obstacles in “good” oracle data as well as shallow liquidity, we suggest that $GARD always be treated as $1 by the protocol. The corresponding collateral factor could be fixed, or be a ratio of the $GARD deposited on AlgoFi compared to the liquidity available for $GARD on DEXs.

Potential benefits to partnering for AlgoFi?

$GARD can be a catalyst for the continued growth of AlgoFi. It has a vibrant and growing community with significant liquidity. In fact, a fair number of our users also use AlgoFi and would immediately benefit from this collaboration. As GARD expands its user base and targets first-time DeFi users, AlgoFi would be prepared to work with the new users.

$GARD can also help deepen liquidity on AlgoFi’s AMM. Currently, $GARD has liquidity dispersed between multiple DEXs which is valuable to get exposure to different audiences. However, in order to make liquidations easier we could consider moving our protocol owned liquidity to AlgoFi and encouraging our liquidity providers to do so as well. In the event of liquidations, this would further lessen the burden placed upon liquidators should they want to swap out of $GARD.

AlgoFi can profit from GARD swapping, borrowing, and lending by integrating GARD to its lending markets. As GARD’s footprint grows, so too will the potential revenues from this partnership.

Limitations:

As per the current suggested precursors to listing new ASAs in AlgoFi’s lending markets stipulated here (Types of Proposals - Algofi) the protocol does not fit perfectly quite yet.

The protocol is shy of the $1M USD in liquidity that’s suggested, however, the price volatility of GARD is low. Given that the liquidity in DEXs accounts for more than 25% of all minted GARD, we believe that our liquidity is deep enough for now assuming that a low collateral factor (for borrowing) is used.

Reliable off-chain data for GARD is not yet available, however, it would make sense to price GARD without an oracle for now. Given how significantly overcollateralized GARD (378% at time of writing) is by valuable (non-GARD) assets, GARD will likely maintain its peg over time and remain valuable. As we wait for an oracle solution to launch (e.g. Goracle), combining a fixed price for GARD and a low collateral factor will protect AlgoFi’s lending markets appropriately.

The GARD team is well known and has been building on Algorand for 18+ months. We are active participants in the Algorand community (DeFi and beyond) and we are committed to building. We’ve also taken extensive steps to protect the integrity of our smart contracts and secure our users’ assets.

5 Likes
2

We’re here to answer any questions!

1 Like
3

Firstly, thanks for the proposal! I really would appreciate protocols on Algorand working more together to grow the whole ecosystem ans the addition of ASAs to the Algofi lending market is definitely one of the ways.

Thats the point that makes me feel like its not yet the time to consider adding GARD or most ASAs in general. As far as i know its always possible to mint new GARD which means the on-chain liquidity pools are relatively easy to manipulate (if you have enough money). So the on-chain price could not be able to be used since people could force liquidations which would hurt the trust in the algofi protocol.

If the protocol would treat GARD as 1$ without a real oracle liquidators might run into problems to liquidate people and leave the protocol with bad debt if GARD depegs. The thing is with overcollaterilized stablecoins although they are always backed by more than enough collateral the market still can decide to trade the stablecoin at lower prices. I am not familiar enough with your protocol to know if there are additonal stability measures the GARD team can use to ensure the peg is close to 1$ tho.

9 Likes
4

Thanks for your response! As discussed in the original post, the lack of reliable pricing data means that a different collateral factor could be used for GARD when it is deposited into the protocol. For example, if GARD had a 50% collateral factor, for a liquidator to go “underwater” they would need the value that was borrowed to be double the value of the GARD collateral that they deposited. If GARD had a 20% collateral factor, this would change to 5x the amount liquidated.

While Algorand DeFi is still young, there is still potential to begin collaborating and gradually make the lending market more efficient as more liquidity comes to Algorand and better tools become available to price assets. We are in a bit of a “chicken/egg” scenario and this is one way that we can deepen Algorand ecosystem liquidity and create partnerships that can evolve over time.

4 Likes
5

Thank you for the proposal. It is always good to seek out mutually beneficial opportunities that can help build a more robust and resilient ecosystem. Like @lobo the lack of reliable hard to manipulate pricing data is big concern for me, so much so that that alone should disqualify it from being added to the lending market at this time. For reference, even USDT which is technically on the lending market has a collateral factor of zero (for well known, but separate reasons).

I have another couple of concerns related to liquidations on GARD based on my experience in April and May. At that time there were at least three problems I experienced with liquidations on GARD that could introduce real risk to Algofi if GARD was added to the lending market.

  • partial liquidations were not possible
  • liquidations were manual and not near real-time
  • there were several instances of low liquidity on the DEXs where it was hard to acquire enough GARD quickly enough without too much price impact (and even a major pull of liquidity by BC in the midst of the market crash)

Please correct me if I got any of that wrong or if the liquidation process has been changed. An absolutely critical part of Algofi’s protocol safety is the extremely efficient near-real time liquidator bots using the API. If an account exceeds 100% Borrow Utilization Rate they are being liquidated on that block. Manual liquidation simply can’t keep up especially in a crashing market. There was a very large account (>100k IIRC) eligible for liquidation on GARD during the market crash that stayed in that illiquid state at least overnight because multiple people who were actively trying to liquidate it did not have sufficient capital to pay for the entire liquidation alone. Illiquidity of that size and duration would pose an unacceptable risk to Algofi if partial liquidations and automated liquidations are not possible on GARD. I look forward to your response.

7 Likes
6

Is there any mechanism for liquidator to redeem GARD for underlying assets? Or is the only path for a liquidator to get out of a position in GARD is through one of the DEXs on Algorand?

1 Like
7

GARD does not support redemption for underlying assets. GARD can only be used to pay back loans on the GARD platform.

8

If USDT has a collateral factor of zero, I don’t see any reason to give an ASA (other than ALGO and STBL2) a non-zero collateral factor.

1 Like
9

I agree with another post that we should really figure out guidelines around what the requirements are to list something before engaging projects. Something like:

  • At least a year old on mainnet
  • Doxxed team
  • Offchain pricing
  • Liquidity above $10M
  • Can’t be too correlated to an existing asset
2 Likes
11

USDT is not transparent in their reserves and doesn’t display information about their assets on-chain whereas GARD and other ASAs are transparent and show this information. For example, even during the recent volatility GARD has remained significantly overcollateralized ~300% on average.

1 Like
12

We are in favor of being conservative in any market, especially one as volatile as today’s. However, given these criteria STBL2 should not be used at all on your lending platform either given it doesn’t meet the standards you outline. In order to promote Algorand’s entire DeFi ecosystem, a leading platform like AlgoFi should be receptive to supporting other projects which have the potential to add longterm value for the entire Algofam. We are hoping to approach listing GARD in a way that is conservative while still a value add for everyone. Surely there is a way to integrate GARD over time even if that means an extremely low collateral factor to begin.

1 Like
13

Thanks for all the questions @Ammeni ! We appreciate you using the platform in the past and would like to let you know we’ve since launched a v2 that has some significant upgrades.

Our liquidation mechanism is meant to be as fair as possible as is designed as such. A new feature we’ve added in is staking which allows users to earn a portion of the system’s revenues. To help fight this problem you outlined we have built in a feature for users to be able to stake their GARD in a pool that also allows them to liquidate users proportionally if/when any positions go into default which will then be backstopped by the legacy dutch auction mechanism you mentioned. We will launch this new feature in the coming months alongside a host of other upgrades. This will help ensure liquidations happen in the same block while also allowing anyone to participate whether they have .1 GARD or 1M GARD.

Another point worth addressing is the collateral factor that is used by AlgoFi when enabling users to potentially borrow against GARD. If the protocol elects to add GARD as collateral but keeps this collateral factor low to begin with (which we have suggested) this will help bring utility to GARD as well as bridge our communities without bringing the AlgoFi protocol undue risk. Given that STBL2 can be used as collateral there’s no reason not to gradually phase in GARD.

1 Like
14

Hard PASS! and Agree with John here…

15

By your own criteria, then let’s stop even debating about adding any Algorand native assets, including AlgoFi’s stablecoin. Come on John, you can do better than this.

Exchange volume is an important and legitimate criteria, but are you really advocating for CEX volume? CEX is not why we are here, and we need to move away from them. Decentralized bridges that enables liquidity to be aggregated across chains is the future … but I digress.

GARD is an over-collaterized Algorand native stablecoin that is fully backed ALGO. From a pure decencentralized stablecoin protocol perspective, I would argue that GARD is fundamentally stronger than AlgoFi’s stablecoin. The only reason that STBL gained some traction is because it is fully integrated into the AlgoFi platform. Let’s not forget that STBL struggled to maintain its peg and that AlgoFi creatively propped it up through its own protocol. If AlgoFi did not have full control over its platform, STBL would have been in big trouble.

The bottom line is that the strategic path forward is for Algorand DeFi protocols to work together and support each other. I encourage the AlgoFi team and community to get off their high horse and remember how you got here. Take this as constructive criticism. We need more collaboration between DeFi procotols and projects.

3 Likes
16

Oyster, please take this as constructive criticism. I’m not sure how else to say it, but you are developing quite the reputation for talking down to people in public forums, acting like you know better and are smarter than everyone else.

You are clearly an informed member of the Algorand community. I appreciate your passion for the Algorand ecosystem and your efforts to bring protocols together. I am assuming you do not mean to come off as disrespectful as you do.

For governance to work, whether it be for Algofi or the Algorand Foundation, we all need to do better. That means trying our best to not post like an arrogant know-it-all. We’re all guilty at some point or another. I encourage you to reflect on this.

Drag-on

6 Likes
17

It is not fully backed by Algo, it has gALGO as well which is Centralized and controlled by FF, ser. Also, please disclose that you represent Folks Finance or have some affiliation with them.

18

Please reserve your trolling for Twitter …

To say that gALGO is centralized and controlled by Folks Finance is simply ignorant. To not understand that gALGO is fully backed by ALGO 1:1 is another display of ignorance.

19

Not trolling, per Paul from another thread…

" I have a number of big problems with gALGO which IMO are deal breakers:

  1. There is a time mechanism on the burning of gALGO for ALGO. This is a massive liquidation risk. In the biggest drops we’ve seen $1Ms get liquidated in a short period. Liquidators need to be able to offload their seized assets to get more of the repay asset and perform more liquidations. gALGO prices would plummet if liquidators tried to sell off even a couple 100k and there is no mechanism for stabilizing that between governance cycles other than algo holders willing to buy up large amounts of gALGO with a high time price (up to 3 months) before the can burn.
  2. gALGO’s system of locking is fundamentally problematic. If you remember a couple weeks ago folks finance TVL dropped almost 90%. This is because every governance cycle all gALGO holders have to remove their gALGO and go “re-lock” it and for a period of up to a week may not have access it gALGO. Having gALGO as a significant collateral source in the protocol would mean large and disruptive movements of capital every cycle.
  3. Last and probably most important, as far as I can tell gALGO is quite centralized, the folks team determines when and how it locks/unlocks. I’d be very hesitant tying something like that into Algofi."
2 Likes
20

Off-topic, but please let’s try to be civil. If you think @infamousB is wrong, wouldn’t it be nice if you argumented why instead of responding with a personal attack?

7 Likes
21

we should really be bring most liquid pairs into the ecosystem and $GARD is not one of them and can easily be manipulated causing issues for AlgoFi. top ones not already on AlgoFi is $Planet, $Defly and $YLDY.