Hacker News Reader

Comment by VHRanger | Hacker News Reader

Loan was overcollateralized by 30% and liquidated on June 15, 2022 [1] (BTC price ~21k)

The loan was written somewhere in the first half of 2021 - it's when heavy USDT activity starts in the celsius core wallet [2].

BTC price floor in 2021 was $32k.

You can't liquidate something 130% overcollateralized and underwritten at a price >$32k (more likely >$40k) at a price of $21k without taking a loss.

You'd need a minimum overcollateralization ratio of 150% for their math to check out in the most optimistic scenario.

Let me know if you have other easily disproven lies you need debunked.

[1] https://blockworks.co/celsius-loan-liquidation-caused-no-los...

[2] https://etherscan.io/address/0xdb31651967684a40a05c4ab8ec56f...

View on HN

As expected, no source on the date, just making up nonsense.

If you'd look at the documents filed in the Celsius bankruptcy, you'd see a sworn affidavit that says Celsius stopped providing additional collateral in May 2022:

https://storage.courtlistener.com/recap/gov.uscourts.nysb.31...

>In May and June 2022, Celsius made the difficult decision to forgo providing one of its lenders, Tether, issuer of USDT, a stablecoin, additional collateral and agreed to an orderly liquidation of its loan. During the market crash, Tether issued a margin call to Celsius with regard to an outstanding $841 million USDT loan. Although Celsius had always provided sufficient collateral to support its loan, and had never previously been liquidated by Tether, the Company agreed to an orderly liquidation and settlement of its loan with Tether to preserve the remaining collateral in excess of the value of the loan

Bitcoin dropped around 30% in May-June, so this timeline lines up exactly with the 130% collateralization ratio.

I can give you exact dates, I just didn't bother to look it up. You can see it in this address [1] which is a Celsius-owned payment rail between tether treasury and Celsius trading addresses.

The exact transactions start here [2] (aug 11, 2021). I got the date wrong in my previous post because I didn't remember and was too lazy to check.

As for your misreading of legal documents and misunderstanding of how loans are written:

The loan was not written in May 2022 -- it was written prior to October 2021, when its existence was confirmed. The loan is denominated in BTC at the price at which it was underwritten (early August 2021, BTC price low-mid $40k)

The fact that BTC dropped 30% in may-june 2022 has no relation to the total loan value from Celsius' side. It only changes the amount of collateral tether would be margin calling Celsius for.

Celsius faced a margin call in May 2022 (price 30k) because the loan was written around a BTC value around 30% higher than that (~40k). Note the math checks out here as well.

It took over a month for the margin call to result in liquidation, hence tether's loss of $250-300m at a price of $21k. If tether immediately liquidated in May 2022, they would have skated by without a loss.

[1] https://etherscan.io/tokentxns?a=0xd41cdb2a35a666e8e1f9f5305...

[2] https://etherscan.io/tx/0x367f433872789fa0223d6ca93c6e073a72...

It doesn't matter the price when the loan was originated, because as the affidavit says, Celsius had provided additional capital. When you have an overcollateralized loan, as soon as it goes below the required threshold a capital call goes out, and you liquidate if that's not met. You don't wait until it's at bankruptcy to issue a call.

https://www.ft.com/content/5eff6c38-9410-45af-94f7-2488b3a87... this also confirms some of it:

>“If bitcoin drops, they give us a margin call [and then] we have to give them more bitcoin,” Mashinsky told the FT in October. He said the loans were typically 30 per cent overcollateralised. The price of bitcoin has since dropped by about half to $30,000. The person familiar with the matter said Celsius had indeed posted more collateral as a result of the falling price.

Sure, maybe everyone's lying, including in sworn declarations. Alex has obviously lied a lot leading up to the Celsius bankruptcy, so I'm not going to say that anything he says must be true. But I don't see any particular reason to doubt the claims here that tether's loan was overcollateralized, called, and liquidated with no loss (and in fact they returned excess collateral to Celsius, as per the court document above "preserve the remaining collateral in excess of the value of the loan"). I do think it's unlikely he'd swear to that specific claim under oath if it wasn't true, as he'd have nothing to gain from it?

Sure, Celsius may have topped up collateral in the meantime. But that would imply taking in the loan at a higher initial value. The numbers aren't exact, since we don't know when the loan was signed & negotiated.

The numbers don't need to be exact in any case given the magnitude of the discrepancies.

The core issue is, again, that Celsius stopped providing margin in May 2022 but were liquidated on June 15th.

Tether ate the loss for the difference in price between those two dates. Tether should have liquidated Celsius in May 2022, but didn't.

There's no way the math doesn't check out with tether ending up at a loss.

Similarly for the $190m of Celsius equity Tether bought, which is now worthless in bankruptcy proceedings.

The fact is Tether lost a 9 figure sum on Celsius. The only question is what the X is in a $X00,000,000 figure.

This is again, nonsensical. As I pointed out above, Bitcoin dropped about 30% from May to June. If the loan was overcollateralized by 30% before that 30% drop, then they'd have been able to liquidate without losses.

The equity is a loss, but that's not what I took issue is, which is your evidence free claim that there was a loss on the loan.

Reply on news.ycombinator.com