Hacker News Reader

Comment by yborg | Hacker News Reader

The author's Internet point is that they think Tether is bust and why big money is not shorting it (you can't find a counterparty to make a big trade with that is any less dodgy than Tether itself). They then describe a mechanism for making the trade in which the counterparty is a DeFi protocol that is, of course, just as dodgy as Tether itself.

The whole ecosystem currently remains a giant confidence game, as Matt Levine described well in a recent article. This doesn't mean there isn't a lot of money to be made - casinos make a lot of money at an activity with no direct economic value - but right now you are best served only putting in money you can easily lose.

View on HN

Big money is not shorting it because there's no way to short it at scale.

There are no broker that will survive to close the short, and you need a broker to short.

USDT is super interesting since it doesn't behave like a normal market priced asset. You know it won't ever go up in value, and you know that if it goes down, it goes down to 0 VERY QUICKLY, with maybe a chance of a claim against unverified assets that might pay out after years of litigation.

In past USDT depeg to around 0.9$ for short time. This is very unlikely to be the case since even if they are insolvent they have a reserve. You probably think that way because of LUNA crash which was backed by almost nothing. For example in defi stablecoins with issues do not depeg to zero or anywhere close to zero if they are still backed to some extent.

Our opinions differ in that you trust Tether's reserve. I don't.

It has been more than 5 years since they were audited. By their own admission, they have 10 billion of their 66 billion dollars in unsecured loans and "other investments".

An org whose entire raison d' etre is to hold money should be pretty open to outside audits. It should actually be a pretty easy audit.

Why have they been foot-dragging so long?

I didn't said i trust Tether's reserves. However even if with a hole they have a reserve at the end.

That's the same thing GP is saying - nobody reliable, not 'in the game', wants to take the other side of the trade.

(Presumably there is a price at which they do, but also at which nobody's remained interested.)

want vs reality. In fact, I would want to bet against USDT with a reasonable proportion of my savings, but there are no safe way to ensure such a contract will be honored in the case that it actually went to 0.

Some of the crypto platforms are scam[1], but no one seriously claims Aave Or Compound, the defi borrowing platforms you can do this with, are scams. There is, of course, always the risk of unforeseen bugs that prevent withdrawls, or crashes that happen so fast the collateral can't be liquidated in time. But that's different from saying that Aave or Compound are themselves "confidence games" that can't be expected to pay you back when you're right.

I know for my part, I made non-trivial money on Compound shorting MKR, LINK, and UNI (the Uniswap token) with BTC/ETH as collateral.

[1] Whether you replace "scam" with "confidence game" is irrelevant to this point.

I think the claim is that usdt crash will be highly correlated with all other crypto going to zero, so any crypto collaterized counterparty will fail. Ability to payout USDT short relies on confidence in other crypto.

I got that part. But the claim here would be that your USDC collateral would go to zero. Yes, USDC is "a crypto" and if you lump it in with all crypto, you will casually think it will magically go to zero too, despite being backed with non-crypto assets and a level of conventional oversight from the financial markets.

Or is the whole idea that it's somehow deeply insightful to not make this distinction?

You’re right that if the short is entirely usdc collaterized. But reading the blog:

> Still, just over half the collateral on Aave is ETH or stETH, and another 13% is wBTC. Aave’s designed to liquidate positions before the become under-collateralized. Let’s hope that happens fast enough if Tether depegs.

That sounds like you still have counterparty risk with regular crypto going to 0 due to pool liquidation (your collateral is not custodially held, but is in a pool).

This comment[1] addresses counterparty risk better than I can: the risk is just that, in a crash, the collateral backing the USDC loans can't be liquidated fast enough. But remember, you're already borrowing 80% of the collateral and turning it back into USDC. Even if all of crypto dies tomorrow you keep that much, that's just a 20% loss. Further mitigated by however much of the collateral for the USDC loans can be converted into USDC.

(The author reduces this value by levering up a cycle, but that's optional.)

[1] https://news.ycombinator.com/item?id=34302128

The money at play is the collateral he put up to maintain the position. Even though it is in USDC it is on the exchange which has pool liquidation. If the collateral isn’t on the platform then what would prevent people from just pulling usdc and never paying back their borrowed amount.

Like the way the short works is you put in $20 usdc, borrow $100usdt, transform it to $120usdc. But those $120usdc can’t leave the platform as otherwise you could just never pay back the $100usdt loan.

> Like the way the short works is you put in $20 usdc, borrow $100usdt, transform it to $120usdc. But those $120usdc can’t leave the platform as otherwise you could just never pay back the $100usdt loan.

Huh? With the way Compound and Aave work, the borrowed amount is already off the platform and can leave and forget about the debt (and collateral). Though you wouldn’t be able to borrow at that ratio.

If you put up 100 USDC, then 85 USDT can be removed from the platform and converted to 85 USDC. So yes, you’re still down 15% in the case of total platform colapse, but not 100%

It sounds exactly how I described in terms of risk. Effectively you are putting up 15 USDC for maintaining an 85 USDT short. Your money at play is 15 USDC, and you can lose 100% of that amount.

The fact that it sounds like you have to show 100usdc to do the trade is irrelevant as you can do this trade multiple times to build whatever position. In a traditional equities setting they skip the “show 100 USDC then remove 85 usdc” by just saying that margin maintenance for an 85 usdt short is 15 usdc. When people talk about potential losses it is the 15 USDC that matters, not the 100 USDC you need to cycle (because 100 USDC is more of a mechanical inconvenience of not having a real brokerage offered short, not money at play).

Okay, but none of that translates into the "unusually high counterparty risk" you were talking about before. That's just regular unprofitability risk.

They didn’t say scam but confidence game.

If confidence collapses, tether will collapse and the rest of these ‘defi’ platforms will turn out not to be as decentralised as they thought, as they also depend on the confidence game.

So, what's the mechanism there? You've got a smartcontract that robotically executes its code, and USDC collateral that lives on it and derives its value from US Govt Treasurys and some level of conventional-market oversight.

How does "loss of confidence" translate into the smartcontract no longer working, and the USDC not being redeemable for anything of value? Those are completely orthogonal dynamics.

The entire argument is based on this magic thinking of "it will all go up in flames at the same time for the same reason because it's all correlated, man".

> The whole ecosystem currently remains a giant confidence game

Money is also a giant confidence game. Any human system is that. It’s all made up.

Money largely gains its value from its ability to be used to pay taxes. If there is someone who controls land and an army and says "this is valuable and you have to pay a certain amount to me at certain intervals" then it is valuable, at least within that political configuration.

> Money largely gains its value from its ability to be used to pay taxes.

Even for contemporary fiat currencies today this is neither historically nor theoretically true. Currencies gain their value largely from being convenient in settling trades and being a measure of value when calculating costs, risks and profits. Ultimately, it does not really matter what this money is based on e.g. gold, paper or digital information.

For anybody interested in this perspective and in a more general criticism of fiat money I strongly recommend reading "Ethics of Money Production" by Jörg Guido Hülsmann. (1)

(1) https://books.google.de/books/about/Ethics_of_Money_Producti...

...and... where were you going with that?

Just because we made something up doesn't mean there aren't different levels of stability for that made up model, or different levels of real world results from acting on those models.

What it means is that there is no “real” justification for something to be valuable, is what people agree on. If enough people agree that crypto is valuable and want to buy it, then it is valuable for as long as people believe it is. Just like any other form of currency.

Reply on news.ycombinator.com