eIP20: Promote USDT to Collateral Tier

[Awilfrid]

Simple Summary

Proposal includes a batch of changes to the Tether token (USDT) on Euler. These include: promotion of USDT to the collateral tier. Increase of USDT collateral factor to 0.90. Increase of USDT borrow factor to 0.94. Enable cross borrowing. Reduce reserve factor to 0.05 (previously 0.10, amended after forum discussion).

We do commit to backstopping liquidations on USDT should the price slip below 0.97 on Euler.

Background

Our team has been active in the crypto space since 2017 and we manage several hundred million Usd in various strategies in the crypto ecosystem.

Motivation

The motivation behind this proposal is to make it attractive to bring substantial USDT TVL liquidity to Euler Finance and to popularise Tether within DeFi. Being one of the primary minters and redeemers of USDT, we have a deep understanding of risks involved in Tether through extensive due diligence and believe the tradeoff favours this proposal.

Tether is a $70 billion market cap stablecoin with an extremely efficient redemption mechanism capable of handling hundreds of millions in volumes every day. This is why it’s extensively used by reputable trading houses and funds.

This proposal would make Euler the first major DeFi lending protocol to allow USDT as collateral and hence give the firms mentioned above an opportunity to tap into what Euler has to offer using their Tether balance sheet. As a result, this will create competitive lending APYs for USDT as traditional finance firms have consistently shown an appetite to short USDT.

Also while the debate on USDC vs USDT is far from being resolved and counts many supporters on both sides, we believe the market to be the best judge. Being widely traded and trusted by large CeFi institutional players, allowing USDT’s collateral status and risk parameter to match USDC’s would harmonise risk factors between CeFi and DeFi and lead to more competition on the stablecoin front.

Implementation

I propose the following: promote USDT to the collateral tier. Increase of USDT collateral factor to 0.90 like USDC. Increase of USDT borrow factor to 0.94. Enable cross borrowing. Reduce reserve factor to 0.01.

We also commit to backstopping USDT liquidations on Euler Finance should the price slip below 0.97. Due to Euler’s liquidation architecture, we’d be able to liquidate users at 0.97 and keep eUSDT on our balance sheet.

Risk Assessment

While Tether has faced scrutiny from regulators and pundits in the past, it has made great progress in terms of transparency and efficiency. For example, it has recently released a Proof of Reserves report done by BDO (top 5 auditing firm):

https://www.bloomberg.com/news/articles/2022-08-18/tether-drops-cayman-islands-firm-in-bdo-italia-attestation-deal

Additionally, it has settled with the NY attorney general on favourable terms that do not encumber Tether as a business while committing to reporting their activity to the NY attorney general on a quarterly basis.

In terms of the market mechanism, USDT is second to none. Its redemption mechanism is able to handle hundreds of millions and volumes every day even during the toughest market environments (such as May this year). This is something that we know of from our own experience, which is why we believe the Tether “FUD” is unsubstantiated.

Not only is the redemption mechanism working smoothly, USDT is also one of the most liquid assets off and on-chain and a DeFi liquidation cascade would very unlikely lead to bad debt for the protocol.

Implementation

Contract	Method	Token	Token Name	Token Address	Updates
governance	setAssetConfig	USDT	Tether	0xdac17f958d2ee523a2206206994597c13d831ec7	borrowIsolated:true, collateralFactor:0.9, borrowFactor:0.94
governance	setReserveFee	USDT	Tether	0xdac17f958d2ee523a2206206994597c13d831ec7	reserveFee:0.05

Vote
Please use the snapshot link below to vote:
https://snapshot.org/#/eulerdao.eth/proposal/0xdbd34a558a1158b460ae54779bb8657d8f39512f5d116b770ebeb269d7e222eb

Conclusion

USDT ranks extremely high in terms of market liquidity, redemption mechanism, and trust amongst institutional players. Adding it as collateral to Euler finance will likely unleash lots of inflows and would be beneficial to Euler and its users.

[river0x]

I support it, but I’m curious to understand as to what safeguards could be implemented to ensure that you do backstop USDT in the case of a depeg? Dont trust, verify (oops, edited) is key to DeFi. This guarantee is exceptional (circle aren’t going to backstop all USDC on Euler), so I’m just wondering how it might look. From my perspective, it requires faith.

Nonetheless, I support the proposal - clarification on the backstopping is all I’d like.

[Raslambek]

Hey! Good suggestion! Tbh, I do not understand why USDT has not been promoted to the collateral tier before. Frankly, Im not a big expert in risk management and other thing, but from an ordinary user perspective USDC and USDT are absolutely equal. Moreover, USDT is a largest stablecoin atm (if CMC is not fooling us of course).

At the same time, I would like to clarify some things.

Being one of the primary minters and redeemers of USDT, we have a deep understanding of risks involved in Tether

It would be good for the community to know who “we” are?

2. While I support increasing collateral factor to 0.90 and borrow factor to 0.94, as it is currently the case for USDC, I have some doubts about reducing reserve factor to 0.10. USDC has a reserve factor of 0.23. That is why I just wonder why there should be more than a double difference for presumably equal assets?

Enable cross borrowing.

If Im not mistaken USDT is already in the list of cross assets.

[Awilfrid]

Hi yes you are right, don’t verify is key to DeFi! Our is a commitment not a guarantee to burn USDT should it depeg, like we normally do in CeFi, as this will be just a plain profitable arbitrage for us. Obviously anybody else can do it and step in should liquidation and depeg occur and do exactly the same thing. Thank you for supporting it!

[spencernoon]

Very exciting to see this posted!

I’m personally of the opinion that Tether is a high-quality asset and would be a great addition to the collateral tier.

[0xWat]

I think there is no question that USDT is now one of the most liquid and credible stablecoins in DeFi as its market cap is top and recently Tether has been committing to more audits by reputable institutions.

What I’d like the proposer to elaborate on is why they only can make sure that the price of USDT liquidation on Euler is at <=0.97 instead of > 0.97 given the current Euler’s architecture.

[itbresearch]

Makes sense to promote Tether to be accepted as collateral on Euler. USDT is highly liquid ($1B+ between largest Curve and Uniswap pools) and its chainlink oracle has been provably reliable, so it poses relatively low economic risk as collateral.

Being the most popular asset crypto pairs are denominated on in CEXs, the opportunity to attract some of this capital to Euler outweighs the risks mentioned above.

[Vitali]

Collateral status, perhaps. Changing other values? No.

Reason? USDT’s past actions and I don’t like the fact that usdt is not really 1:1 as there is some pip volatility when doing conversions to the fiat.

[chainguys]

i think better not…

USDT might cause unexpected systematic risk…

[houshawany]

I’m in favor of adding USDT as collateral. But I share the same question as Raslambek, namely who is “we” in this proposal? I’m sure there will be other arbitrageurs that step in during a depeg event, as you commented, but still good to know considering your commitment. Likewise, I don’t see why any of the collateral, borrow or reserve factors should be materially different from USDC. I would like to see the proposal amended to align these values or provide a better justification for why the reserve factor should be significantly lower.

This being said though, my knowledge of potential risk factors is limited so I’m very interested to see Michael’s input on this proposal in case there are some larger risks not yet outlined.

[euler_mab]

I’m in support of this proposal. I would only add a minor suggestion: lower the reserve factor to 5% instead of 10%. This would make Euler the most competitive protocol around for those wanting to borrow USDT (Compound is 7%, Aave is 10%).

Clearly there is ongoing controversy surrounding Tether. Whilst some of the criticisms around transparency (especially historically) are justified, I think people should evaluate this proposal on the actual risks to the Euler protocol. What are these?

The main risk of adding any new collateral to the protocol is that its price needs to remain relatively stable over short timescales, and there must be consistently liquid markets. An asset whose price drops by a large % in a few minutes could lead to borrow positions becoming insolvent before a liquidator has had chance to liquidate them. An asset with thin liquidity may not be accessible by liquidators in the event of a need to liquidate a position. How do these risks stack up for USDT?

I do not think liquidity is in doubt for USDT. The main risk most people will point to is historical depeg events, where USDT has traded on the wider market for less than $1. It is important to point out, however, that none of these events would have posed risks to a lending protocol like Euler. Collateral assets can float in value without risks to the protocol, if collateral and borrow factors are configured appropriately. This is true for ETH and WBTC on Euler today. It would also be true for USDT, regardless of whether it holds its peg in the future. The protocol does not care about the $1 peg per se. USDT could, in principle, trade at $0.5 without necessarily posing a risk to Euler.

The protocol would care about a depeg in the event of a major Tether insolvency. In this scenario, we could see USDT dropping in value quickly and liquidity drying up. How likely is a major insolvency event for Tether? Given that Tether is required to submit quarterly accounts to the New York Attorney General, and its recent moves to a more transparent accounting model and a significant reduction in the amount of commercial paper it holds, I personally think it is quite unlikely.

[euler_mab]

Thoughts on some of your questions below.

[houshawany]

Thanks for the added clarity. I like the idea of making Euler the most competitive lender of USDT. This would certainly help to drive protocol growth. I hadn’t looked at reserve factors on other protocols prior to my initial comment so referencing them as a baseline makes me a lot more comfortable with your suggested RF of 5%.
I’m fully in support of the proposal with this change.

Unrelated to this proposal, it seems that the current USDC reserve factor is too high unless it is not a priority currently to make borrowing USDC through Euler competitive against other protocols (similarly RF for USDC is 10% on AAVE). Perhaps another proposal is in order.

[JTraversa]

Completely agree that historically Tether has really just acted as far better collateral than comparable higher risk assets (more volatility, less liquidity) such as LINK and UNI

The concern im sure most have isnt related to a normal or historically presented volatilty though, but a potential complete shortfall event.

Not really a valid concern, but I think its one that is different than the one addressed by @euler_mab.

My response would instead focus on the breakdown of UST which had similar amounts of on-chain liquidity.

UST, while extremely volatile in its death spiral, still saw relatively robust liquidity throughout most venues. While it was perceived that UST would surely drop to cents on the dollar, the breakdown itself took a few days as traders arbitraged information asymmetries and market flows.

With this in mind, the only protocols that were really impacted by USTs volatility were those that had incorrectly implemented chainlink oracles, continuing to accept the quote returned from chainlink without verifying its updated timestamp.

Should Euler have then implemented oracles with proper timestamp checks / quote expiries, UST would not have been a risk, and similarly, I would conclude that USDT is acceptable collateral and would also echo @euler_mab’s suggestion to push to gain market share with the lowered RF.

[seraphim]

Here’s what I think about USDT, and it’s based on independent research and conversations with the biggest users of Tether, as well as prominent attorneys with a good overview of the regulatory situation. I’m not going to dox anyone, but it doesn’t take a genius to figure out who I mean.

For this particular proposal, there are 3 USDT risks to consider:

1. USDT onchain liquidity
2. Redemption of the underlying assets

USDT onchain liquidity

Tether is without doubt one of the most liquid assets onchain. The 3pool on Curve is one of the most liquid DeFi pools and contains circa 260mil of USDT:

There is another 200mil of USDT on Uniswap as well:

The more prominent risk is (2)

Redemption of the underlying assets

This is what the FUD is about. Does Tether hold enough high-quality assets to back redemptions?
First of all, what are redemptions and why are they important?

For USDT to be trading at 1:1 to USD dollar, one needs to be able to redeem 1 USDT for 1 USD at any time.

Example
For eg, if you see USDT at $0.99 on Bitfinex, you would buy 1 million USDT at $0.99, call Tether, and ask to convert 1 million USDT into 1 million USD. Tether would either give you the cash from their reserves, or sell cash-like assets (short-term US government debt) into cash and hand you $1 million.

Since you bought $990,000 worth of USDT and redeemed it for $1,000,000, you’ve just made a risk-free $10,000. Pretty neat. But imagine you could lever up and buy billions of $ worth of USDT and make even more money! This is exactly what funds such as the one proposing this would do. It’s an extremely lucrative strategy, and it relies on the ability of Tether to meet these redemptions. Because of that, USDT can maintain the peg as it’s always profitable to buy it to redeem.

The FUD is essentially that in the event of a USDT selloff, either (1) the assets won’t be there or (2) they are not high-quality assets and can’t be immediately sold for cash.

And this is what these 2 important settlements with CFTC and NY Attorney are about:

Cryptocurrency firms Tether and Bitfinex agree to pay $18.5 million fine to...

CFTC Fines Tether and Bitfinex $42.5M for ‘Untrue or Misleading’ Claims

Basically, the NY attorney and CFTC accused Tether of not having enough funds to cover all the redemptions for a few reasons:

1. The money designated for redemptions is unavailable because it’s loaned out to Bitfinex (who has same owners as Tether) or it’s operationally inaccessible ($850 million was supposedly lost to Crypto Capital, a payment processor company).
2. Tether funds were not ring-fenced from Bitfinex funds.

Another fear is that Tether holds too much commercial paper (short-term loans to companies) that cannot be sold easily to meet liquidations.

So why do I think these risks are overblown?

Tether needs to submit detailed reports to NY Attorney on the composition of USDT reserves, proof of ring-fencing and what payment processors they use (Section 57) according to the settlement:

2021.02.17_-_settlement_agreement_-_execution_version.b-t_signed-c2_oag_signed.pdf

Tether has also commissioned a proof-of-reserves from BDO (top 5 auditing firm) and it was released at the end of August 2022:

Std_ISAE_3000R_Opinion_30-6-2022_RC134792022BD0303.pdf

Here’s the breakdown:

Commercial paper is a decent chunk, but in the footnotes we see that: The average
duration of items in this category is 27 days and the average rating is A-1. A-1 is the highest rating CP can get according to S&P.

Basically, it would be insane now for Tether to lie about their reserves because they’d be in BIG trouble with the authorities, as they’d be submitting false reports. I don’t think anyone in their right mind would do that.

What about the redemption mechanism?
If you talk to the big boiz using Tether to buy USDT and redeem it from Tether for profit, everyone is extremely happy. They praise USDT’s redemption mechanism for its reliability even during turbulent times.

The only complaint I’ve heard is that the Bahamas-based bank handling the redemptions on Tether’s behalf is constrained by market hours at times, leading to delays in redemptions during odd hours. So for eg if you bought 500 mil USDT at 0.99 and want to redeem it at $1 during Asian trading hours, you might have to wait a few hours till the Bahamas bank staff are online.

Nevertheless, it only leads to volatility for a few hours before arbitrage is possible again, meaning it does not pose significant systemic risk as there is a USDT:USD price at which everyone will definitely bid even if there’s a redemption delay.

Conclusion
All in all I agree with this proposal. Liquidity is great, the redemption mechanism and transparency around it has improved greatly. USDT should imo be collateral on Euler Finance.

[patria]

in support of the proposal

[allthecolors]

Placing USDT and USDC on the same footing in terms of parameters would give Euler a fundamentally different risk profile from the other mainnet lending majors. Although the depth of stable LPs involving USDT on Curve, Uniswap, and Balancer is very deep, risk of a liquidity crisis in USDT on those AMMs is isolated to LP holders. With USDT at collateral tier on Euler, the risk is more systemic to the protocol: specifically, folks who don’t touch USDT are indirectly exposed.

I agree with the general sentiment here about the risks of Tether being a somewhat outdated narrative, so I’m inclined to support the proposal. I will note that @Awilfrid 's commitment to backstopping USDT liquidations below 0.97 is meaningless due to its unenforcability on-chain and automatically makes me more, not less, skeptical of the proposal.

[0xCrab]

I’m in support of the Proposal especially seeing that Tether has shown to be more robust on the Regulatory side especially if you look at the recent events around OFAC.

USDC normally never acts on accounts of hackers or takes a long time to happen, while USDT always is on the ball with these events, but with OFAC, USDC showed they blindly follow what is set out.

https://tether.to/en/tether-holds-firm-on-decision-not-to-freeze-tornado-cash-addresses-awaits-law-enforcement-instruction/
Tether showed to be more sensible and not afraid to say no to the Regulators so I’m in full support of this request.

[seraphim]

Safe to say this passed the temperature check. Will move to eIP 20. I also amended the Reserve factor from 10% to 5% as discussed.

[monet-supply]

@seraphim

Tether has also commissioned a proof-of-reserves from BDO (top 5 auditing firm)

“Top 5 auditing firm” is a bit of a Tether psyop imho. There are the “big 4” (EY, PWC, Deloitte, KPMG), which have never been willing to engage Tether but any other grouping is somewhat manufactured. Nothing against BDO in particular but Tether’s marketing of this and other engagements is pretty suspect.

@seraphim

Basically, it would be insane now for Tether to lie about their reserves because they’d be in BIG trouble with the authorities, as they’d be submitting false reports. I don’t think anyone in their right mind would do that.

If they were already running at a deficit in prior periods, continuing to lie about their financial condition is the only way to buy enough time to “make it all back” rather than being forced into insolvency (and likely legal issues due to prior period misrepresentations). I find this line of reasoning to be unconvincing.

Even if their statements were more or less accurate, Tether still shows extremely poor and risky financial management.

Besides the widely discussed “commercial paper” balance sheet line item, there is also significant holdings in corporate bonds, “funds” (bonds? equity? VC?), precious metals (no clarity if this relates to XAUT issued or if they are backing USD liabilities against Gold/other metal assets), “secured loans” (which they’ve claimed in the past were not to related entities, but have also been shown to have lied about this in the past), and “other investments” (could be anything, Celsius investment could have been bucketed under this).

According to their financial statements, they hold ~$200MM in equity against $66.2 billion in liabilities (0.3% capital ratio). Even if one assumes that their cash and cash equivalent assets have zero risk (which is generous given CP holdings), this is only ~1.4% capitalization against their potentially volatile assets (bonds, funds, precious metals, secured loans, and other investments). If their portfolio had a decent amount of duration they could easily be underwater by more than this amount on their bond portfolio alone due to increasing rates and credit spreads.

My sense is that, if Tether didn’t have something to hide they would have provided much more satisfactory disclosures by now. And given how high of return can be earned currently on risk free US treasury bills, they would likely have shifted more of their assets into these investments if they weren’t trapped in low quality assets and wanting to avoid being forced to recognize losses.

Supporting USDT as collateral could provide a form of competitive advantage for Euler (only major lending protocol offering significant support) but I think it would be counterproductive in the long run, as all other assets on the platform would now bear significant Tether credit risk and users would tend to demand a higher rate of return to participate in lending pools.