128 comments

[ 3.2 ms ] story [ 207 ms ] thread
(comment deleted)
So is this presented as "evidence" to suggest there was an attack on Luna, that its depegging into a death spiral was intentional by some person or group of people?
That's probably what happened and not a conspiracy theory but that doesn't make Luna or algorithmic stablecoins less bad. If they are vulnerable to attacks, then they're not stable.
I don't see any attack. It looks like a bank run caused by people exiting crypto.

It has been known for years that algorithmic stablecoins cannot survive a bank run so maybe Wallet A knew that their withdrawals had a probability of triggering collapse, but I still wouldn't blame them.

What i want to know is who benefited from the crypto-run on luna.

The run is indeed a self-fulfilling and a positive feedback loop, since a small disturbance in price could trigger some of the conservatives to withdraw at a tiny loss, and this could snowball.

But it is also possible that someone intentionally did it, like soros did for the english pound back when it got depegged. But i imagine you'd have to be able to borrow large amounts of luna and/or UST, sell it short, then trigger this run, in order to make profit. Is there any evidence that someone did such a large borrow?

Do Kwon benefitted, everyone else was conned.

The depeg of UST is more akin to a short of Herbalife than the Soros GBP attack. Remember the weakness of Terra was orders of magnitude easier to expose than that of a national currency.

> What i want to know is who benefited from the crypto-run on luna.

Literally everyone who sold first, that is, the $hundreds of $millions of dollars that were sold on Saturday, May 7th, as described in this article.

That's the thing about bank-runs. The *first* person to sell benefits. As the whole community slowly realizes what is going on, they all rush to sell at the same time. This causes the death spiral.

"Benefiting" is only relative: people who managed to exit UST at $1 didn't make a profit, they just avoided being left holding the bag.

The people who actually benefited were those who earned the 18% interest from Anchor, which was funded by the suckers whose "investments" went to zero.

> The people who actually benefited were those who earned the 18% interest from Anchor

These people would have been better served if the whole scheme continued rather than killing it by a run.

This was a zero-sum game.

The loss of the masses who invested into 18% APY accounts is equivalent to the profits that Anchor made from this scheme.

> "Benefiting" is only relative: people who managed to exit UST at $1 didn't make a profit, they just avoided being left holding the bag.

That confuses the issue somewhat. 5 years ago, UST didn't exist.

Sometime in the last 2 or 3 years, Do Kwon started to sell millions, maybe billions, of UST to people, 1 UST for $1 USD. In fact, maybe he sold them for less than $1 USD, maybe he sold UST for 80-cents (to fund over a year's worth of time of 18% APY promises).

Everyone who sold UST, over the past 2 years for real USD, got to keep the dollars. The UST bag holders are... well... the bagholders. They have nearly nothing left now.

Even those who sold UST at 80-cents or 60-cents last year or two years ago benefits by selling before everyone else did.

> The people who actually benefited were those who earned the 18% interest from Anchor, which was funded by the suckers whose "investments" went to zero.

The opposite. A bunch of people sold UST to a bunch of others for 82-cents, maybe 90-cents (since it took less than a year for a UST to disappear).

Thanks for your dollars, here's a bunch of UST that soon became worthless.

> who earned the 18% interest from Anchor

they might've benefited from the interest during the rise of Luna+UST, but they don't benefit from the crypto-run. They even have a chance at a loss due to not being able to sell early!

> That's the thing about bank-runs. The first person to sell benefits.

no, those selling due to the bank run don't really benefit from the run (aka, they don't make profit). They just _don't lose as much_ as the last to sell.

I'm talking about whether anyone made a profit from this run, like https://www.investopedia.com/ask/answers/08/george-soros-ban... , in which he managed to short sell the pound and gathered enough capital (his and others') to sell the pound to cause it's value to drop (and force the central bank to sell foreign currency to buy the pound, in order to keep the pound peg).

Anchor, who sold you those UST for 82-cents per 1 UST (aka: 18% APY), was "only going to win" when UST crashed below 82-cents.
It was a Soros style attack
Soros didn't attack. He shorted, betting against the pound. He's no more criminal than any other short seller.
Exactly, just like there are so many ways to legally kill people.
Well, we could promise someone a stable guaranteed 20% APR and defraud them of their life savings, thereby driving them to suicide
If code is law and its all deliberately unregulated, is it an attack, or is it simply an innovative leveraging opportunity?
I don't even see anything innovative here. People withdrew their money.
Perhaps, but I'd be entirely unsurprised if, if someone thought exiting the pool might destabilise it, that they might arrange some way to profit from that.
Its just a chronology with no conclusion

Hoping to add more accuracy to the information available

What gives you that impression?
Your "attack" is someone else's "well-executed and informed financial strategy."

Absent a central authority that determines legitimacy, it's not clear why we should simultaneously accept the cryptocurrency community's churlish attitude towards financial stability and the crocodile tears around being the victims of said volatility.

Ponzi schemes don’t need much of an attack to collapse
My read is that it's equally likely that it's an inexperienced retail investor who got spooked by the state of the global economy.
In the mid-2000s two employees from Citadel left for Jump Trading and took some source code. Jump then took this strategy, improved on it and destroyed Citadel's HFT revenue stream. We're talking about billions in lost revenue for Citadel. I don't have proof for these allegations, but it is an open secret in the industry.

There was a lawsuit a few years ago, but Jump was able to convince the courts that they could not reveal their source code because it would damage their business. Citadel was never able to prove that their code was stolen.

Ken Griffin could certainly be strongly motivated to employ such an attack. And if he did... you have to admit it was pretty brilliant. This is something straight out of "Billions".

FYI this is jump trading, probably the largest prop trader and market maker in crypto. I have heard they lost over $4 billion from the Luna / UST crash although their profits from the bull run were so high they are likely fine.
4 billion seems an enormous amount. Do you know how they trade it?
I don’t work in trading but everyone in trading knows jump. Likely they invest in a lot of altcoins in pre-sales at significant discounts. Also they were the ones that bailed out Wormhole when they got hacked, likely because the hundreds of millions lost was jump’s own money. They are heavily into solana like all big traders for whatever reason. I think solana is weak tech but there is tons of VC money behind it.
I work in trading and didn't hear $4bn, that's why I was asking.
Jump is into Solana because Solana is their baby; it's a centralized technology that they are behind.

Source: was supposed to get in a call with a few people with @solana.com email addresses, got a calendar invite with same names @jumptrading.com addresses.

Interesting. I heard someone from Solana labs describe write access to Solana as an oligopoly. But I don’t understand the play. Is it for Jump and retail investors to play in the same Dex derivative bath tub?
Generally buy low sell high, as all traders attempt :)
> I have heard they lost over $4 billion from the Luna / UST crash

Source for this? I can't seem to find this claim anywhere.

No public source but number seems approximately correct +/- 1bn. Their 2021 pnl was over 10bn though so they still up massively.
What's $10bn less short term cap gains taxes less $4bn?
IIUC you don't pay capital gains when you're operating a trading firm, you keep a balance sheet and subtract losses from gains
This is true for some asset classes. I cannot speak to crypto, though.
You don't trade your money. You trade the clients money and take 2% of the assets and 20% of the gains. Notice the customers don't have yachts.
Prop trading firms usually trade their own money.
Jump Trading, like most other HFT firms do not take client money. They trade only the owner's personal capital.
Jump is a prop trading firm so this is just completely wrong, they only trade their own money.
I've also heard that 10-man team at Tower made about $3B trading crypto last year. The team leader got a $600M bonus.
Half the traders I know who were sitting on 3000% gains in 2021 are down to 300-500% gains by 2022.

Crypto markets fall apart as quickly as they go up. Most who entered the market in late 2021 are down and will be down for years (forever if they're holding some dead altcoin).

You are referring to mark to market gains/losses and not actually realized, correct? Like jump didn’t actually clear 10 bill in profit or lose 4. They were in early size and let it ride but actually investment was small, do I have it right?
Mark to market and realized losses are the same thing for a trading firm.
I work in the industry and know a lot of people and that is what I heard.
Algorithmic stable coins have always been vulnerable to a run on the bank situation but it’s not limited to just algorithmic stable coins. Bitcoin itself is also vulnerable and due to it’s design, big investors will be able to attach much higher transaction fees to their transactions meaning they’ll get a chance to sell first.
You only have to worry about transaction fees if you're a sucker and actually use the blockchain. If your coins are already custodied at an exchange you can sell immediately.
Exchanges stop / block / pause transactions all the time though? That's what everyone was so mad at robinhood about. No difference in crypto exchanges.
Now you’ve introduced a whole other category of risks instead.
> if you're a sucker and actually use the blockchain

"Abstract. A purely peer-to-peer version of electronic cash would allow online payments to be sent directly from one party to another without going through a financial institution. Digital signatures provide part of the solution, but the main benefits are lost if a trusted third party is still required[...]"

so much for the whitepaper I suppose

Correct. It seems to me that nobody has been trying to realize the vision from that whitepaper for like a decade.
Allow me to introduce you to an exchange called Mt Gox...
> If your coins are already custodied at an exchange you can sell immediately.

And you'd (rightly) get called a sucker for storing your coins at an exchange, negating really all of the benefits of using Bitcoin.

Some people just want to collect bitcoins and do not care very much about the benefits of using it.
Yeah, let me just withdraw my coins from Quadriga...
> If your coins are already custodied at an exchange you can sell immediately.

Unless the exchange suspends trading or goes down for unscheduled maintenance right when you want to sell.

Of course, that is assuming the exchange doesn't simply take your coins and run.

While yes, of course any asset in the world is vulnerable to market swings, you can't really call the Bitcoin scenario a "bank run" in the same way as with UST/Luna (where there actually exists an issuer and a smart contract acting "bank" where people exchange collateral for tokens).
I find the concept of Curve pools (liquidity pools) like UST/3CRV, dual token systems, algorithmic pegged stablecoins, and services like Achor very complicated to follow. Is there any good literature that I can read the help understand the relationship between these components. Do they map existing components in traditional finance?
Yes. There is extensive literature on bank runs and similar failures caused by fractional or fraudulent reserves and lack of confidence, insurance, and regulation.
Eh, Curve pools and other invariant-based AMMs have some interesting properties outside traditional finance, bank runs, etc.

I'd read up on the Uniswap constant product invariant (xy=k) and consider reading the Uniswap v2 code (it's a pleasure to read) [0].

Then check out the StableSwap paper by Curve founder Michael Egorov [1]

Finally if you want a well-documented implementation to read in Solidity, check out Saddle [2]. Disclaimer, I helped write the first iteration.

[0] https://github.com/Uniswap/v2-core [1] https://curve.fi/files/stableswap-paper.pdf [2] https://github.com/saddle-finance/saddle-contract/blob/maste...

Can you please go into more detail on these "interesting properties"? I fail to see anything in terms of real analysis in the links you gave, simply descriptions of automated trading rules simple enough to be understood, but mathy enough to sell to the marks if cryptocurrencies, backtested on the growth stage of a bubble.
It solves the problem of decentralised liquidity. An order book with traditional market makers is inherently centralised, here we have incredibly simple algorithms for trading between two assets with no intermediary, and the complicated, HFT market makers of traditional finance are replaced by passive liquidity providers.

Not sure why you mention backtesting, or how that would really apply.

So basically, there is no interesting inherent property (no interesting math, no deeper dynamics), it just "solves" a problem DeFi created - and it only introduces a whole new layer of possible implementation errors and new types of risks like impermanent losses.

Backtesting is relevant because there was no stability analysis, no simulated long drawn out bearmarket, no adversarial probing. Just some minimal quant sugar to make the new gambling opportunity go down with the suckers.

In real finance, you have much more sophisticated and diversified trading algorithms babysitted and regulated for stability, with much higher effective decentralisation and control since different HFT funds are legally barred from conspiring against the traders. Oh, and much lower fees as well.

You have been radicalized.
I'm completely bewildered at what you mean in this context, given that the most conservative possible viewpoint on stablecoins is "there is no algorithm that can produce a stable coin other than 100% or better reserves in the stated currency".

Remember that "conservative" in finance means not taking risks, rather than the current political meaning of "right wing authoritarian".

Really unhelpful and not pertinent to the question OP asked at all

HN is better than this.

Usually DeFi platforms will have a lot of tutorials and documentation. Usually the price of an asset in a pair is computed automatically using the ratio asset1:asset2. Stablecoins also have their own docs. For example, terra’s whitepaper explains the fundamentals (mint tokens if the price is too high, burn tokens if the price is too low).
Not a specific resource but a good one to start with would be MakerDAO DAI. Note that when people in the space talk about "will algorithmic stablecoins ever work?" They'r usually referring specifically to noncollateralized or undercollateralized stablecoins (like UST). DAI is overcollaterallized so its fundamental model is not really under debate recently. However, DAI is the first major decentralized stablecoin and a lot of concepts and terminology reappear in later ones so understanding what it does and how it works is a great base for understanding algorithmic stablecoins. Much like understanding Bitcoin is useful when looking at other blockchains.

For AMMs, same applies to Uniswap. Their whitepapers(V1,V2,V3) are both accessible and formal enough to be useful.

Also this paper from Curve https://curve.fi/files/stableswap-paper.pdf

Projects like Terra and Anchor are messy enough (some might say willfully misleading and opaque) that you want to be able to recognize lingo and patterns (and, ideally, source code) to come out with somewhat useful results. In some cases, you'll need to be monitoring their Discord, Twitter, and/or Telegram groups to follow them properly - though in those cases, I've taken that as enough of a red flag in itself to disqualify the entire project from my attention.

I second the recommendation to learn about Uniswap. Uniswap was really the beginning of the defi explosion of the last couple of years
imo the problem with UST vs DAI was not under vs over collateralisation, but rather that Luna supporting UST had its value tied to UST adoption. On the other hand, DAI is collateralised by things like Eth, which could also in theory crash, but get their value from a much wider ecosystem than just DAI - hence one potentially avoids the circularity.

For a great read on this and more, check out Vitalik's post: https://vitalik.ca/general/2022/05/25/stable.html

They are complicated on purpose, as they’re designed to hide the fact it’s just hot air supported by more hot air.

Retail investors try reading it, give up, and quietly agree that the emperor’s new clothes are lovely

theory of: uniswap AMM, dai, ust/luna, is exactly as complicated as it needs to be (not complicated at all if you did calculus 1)
The number of brain cells and compute cycles being wasted on this nonsense is staggering.
Im sure this has been said about all early technologies at some point. If you are right or wrong only time will tell though.
They laughed at Columbus, they laughed at Fulton, they laughed at the Wright Brothers. But they also laughed at Bozo the Clown.
Though it's worth noting that Columbus was wrong but got lucky.
But there's nothing early abt crypto. My mom in Normandy sends me carbon copies of bitcoin article in the local newspaper (God knows why, all I told them was to never ever listen to anyone telling them to put their retirement in crypto lol).

It's everywhere, everyone can get some, work in a crypto place, accept crypto in their store, convert their entire salary.

What is so wrong about it that we call this 14 year old dinosaur tech "early". We've seen it, we don't want it, time to try someth else.

When you start seeing Super Bowl ads for it, you can describe it as "early"
Yeah we dont have super bowl in France or China where I leave now. I dont think it means early to have ads for, it seems to mean peak bubbles instead. And it could just be the beginning of the end.
(comment deleted)
Wait 'til you figure out how much is being wasted on other nonsense like computer games and team sports!
Hard to root for the fiat system though when you're dealing with near double digit inflation.
We have a single data point for the performance of crypto as an asset class during a time of inflation. Admittedly a single data point isn't great but it's all we have. What do you take from the data so far? I don't see Bitcoin exactly holding up its promise as an inflation hedge.
"Depegging" is an interesting term. A scam, bolstered by lies and fraud, fell apart. A made up claim that they could financially match a useless asset to a useful fiat currency.
I have paid for all of my shelter, food, and taxes with it, for decades.
I have paid with crypto (including UST): my dentist, my employees, my house, food.

So crypto is "useful", since "useful" = "being able to buy things" by your definition

I just answered the question. Fiat currency is clearly useful, because it is used to pay for all the things people need. It's really pretty straightforward.

What I would say about cryptocurrency is that it is somewhat useful by this definition, but much less so. With some effort I can pay for food, but it would require a lot of effort to use it to pay for shelter, and it is not legal to use it to pay taxes. Its usefulness is a small subset of the usefulness of fiat currency.

If it's useless, I'd be happy to take it off your hands!
I'm not saying it's useful or useless, I'm asking for a definition of "useful".

According to parent, "useful" = "being able to buy things". Which crypto lets you do.

So, parent has some thought contradictions

I've tried to buy things with crypto.

It's easier to barter. By a longshot.

Easier for you, due to your n00biness. For me, it's as easy as swiping a visa card (that's literally all it takes).
It’s let’s me buy stuff in fiat cash only establishments.
So "useful" = "accepted as payment in fiat cash only establishments"?

Then, we should say that only "cash fiat" is useful, while fiat bank account balances are not "useful"?

Cash in hand... cash in the bank... I don't spend a lot of time worrying about the distinction between the two. For now when I want a burrito from Taco Bell, they still don't let me but it w/ DOGE.
"cash in the bank" is an oxymoron. But by "cash" you mean any form of fiat, ok.

So then your definition of "useful" = "accepted as payment in fiat only establishments". By definition, these establishments will only accept fiat. So you are just defining "useful" = "fiat". Sure, by that definition only fiat is "useful". Definitions like that don't bring any new information.

A better definition is "X is useful if I can convert X into a burrito".

You can do that with DOGE or any crypto, anywhere Visa is accepted (for example with Lemon cash).

You can also do that without Visa, it just requires more work: first exchange your crypto in a P2P network, then pay burrito

You’re giving them too much credit. A lot of crypto are just incompetence and failed startups. Not everything is a scam just because it didn’t work. Some were just misguided attempts (in hindsight).
They were warned of this exact scenario multiple times
Also, a bank with this level of incompetence is going to be roasted by regulators.

In fact banks are not allowed to take this much unbounded risk with their deposits to begin with.

web3 people are cynical about the traditional system but they often forgot how much protection existing regulations offer and that the bad things are often due to bad actors slipping through cracks and make a big impact. Such as the Korean hedge fund manager that leveraged himself beyond what regulations allow in spirit by hiding his exposure.

In crypto there are not just cracks, there are gaping holes.

> The Times 03/Jan/2009 Chancellor on brink of second bailout for banks

There are some costs associated with that protection; to put it mildly. It isn't like the regulators protected the system in 2008-era with their prudent de-scamming, they protected it by handing out money to incompetents.

There will be a lot of pain from scams and immutability. It still seems entirely possible that in a few decades crypto as a whole will be more respectable than the US dollar. The US have been a reliable custodian only by comparison to other governments and there are obvious advantages to crypto vs currency backed by guns and printing presses.

Although I do agree things like Luna are certain disasters in the making best avoided completely.

> It still seems entirely possible that in a few decades crypto as a whole will be more respectable than the US dollar. The US have been a reliable custodian only by comparison to other governments and there are obvious advantages to crypto vs currency backed by guns and printing presses.

Can you elaborate? I'm mostly pro-crypto, but it feels like it's really begun to stagnate since early '21.

Also, How will proof of stake not lead to big incumbent actors that control everything, like the US bulge bracket?

> Can you elaborate? I'm mostly pro-crypto, but it feels like it's really begun to stagnate since early '21.

My guess is the whole thing is going to collapse in a flaming implosion, leaving broken and dismembered financial entities in its wake. Then it'll rebuild. The cycle will happen 2-3x more times, building up some evidence for what value propositions crypto has and then the show can go properly mainstream.

Something like the early internet. People got very overexcited, things fell apart, then a bit later the likes of Amazon turn up and start using it for serious projects.

> Also, How will proof of stake not lead to big incumbent actors that control everything, like the US bulge bracket?

No idea. Nor much care. If it is worst than the US fiat system then nobody will use it, if better then good.

In proof of stake even if it’s a lop sided power distribution they have an incentive to govern the crypto in a way that maintains its value. And that’s really the goal of any decentralization rather than some form of communism or egalitarian distribution of wealth/power. Crypto cannot solve the inherent income inequality in society—it’ll just mirror it eventually.
Banks had to repay the funds. A few banks went under before the government decided to intervene before it became an even bigger crisis.

After that, banks faced a lot more scrutiny and are still not allowed to take many big risks. Archegos was the worst thing that happened to a bank in the last decade. The market has been crazy in the past 2 years and no bank went bankrupt. If not for regulations instituted since 2008, there could have been a bank that went down like Maven Capital.

> Banks had to repay the funds.

The banks should never have been in a position to repay those funds. Institutions that screw up that badly should be disbanded and replaced by someone completely different. Maybe swap the executive team and the office cleaners, I dunno. Anyone else. Totally different corporate culture.

But, happily, whoever is unhappy about it can start looking to crypto to see if there are alternatives. I don't think there are yet, but an an economic cycle or two the whole issue may well be a moot point.

If you’ve ever done a normal startup you’d know you constantly get bombarded with warnings of failure, and more often than not those warnings are right. And it’s not necessarily because people know something, but because most startups fail and you have more than a 50/50 chance of being right just by saying “you will fail.”
That is fine if it is just a loss to you and your investors. Not to your customers.

It is like Elizabeth Holmes being told that her tests were faulty and she still does not fix them.

This is just a startup that happens to be able to skirt investment laws. This is a failing of regulations rather than necessarily outright scams. If people could invest in regular startups a lot of people would lose money too.

Elizabeth Holmes didn't have faulty tests. She literally lied and defrauded investors about having the thing she was working on. That's actually a bad example to compare to the UST situation because that thing was out in the open, including how it worked. A scam requires some deliberate deception or lying. A bad idea is not necessarily a scam, and dumb people knowingly pouring money into a bad idea is also not a scam. If it was a scam in any way, it was a failing of governments of wherever you lived letting you invest in this.

This is an interesting analysis.

Setting aside for the moment the basic question of whether UST was structurally unsound, I've thought for a while now that the slowness of on-chain transactions coupled with lack of liquidity are two of the most important differences between cryptos and real FX markets. This paper just shows how chronic this problem is.

> During a seventy-five minute span on Saturday, May 7, a series of critical transactions first pushed UST off the peg.

> At 21:44 GMT, Terraform Labs (TFL) withdrew $150 million in UST liquidity. > At 21:57, a relatively-inactive account ("Wallet A") swapped $85 million UST for USDC in this pool. (This was the largest swap transaction in that particular Curve pool ever.)

> From 22:32 - 22:38, another account ("Wallet B") swapped $75 million UST in the pool across three transactions. By this point, the pool was severely imbalanced.

> At 22:52, Terraform Labs withdrew another $100 million in UST liquidity. This made the pool slightly more balanced, but it was still heavily imbalanced and — more critically — tiny at this point, leaving it vulnerable to even small transactions.

> At 22:59, Wallet B swapped another $25 million UST in the pool.

So over 75 minutes, the curve pool wasn't able to absorb $435M of transactions. To put this in context a major FX cross eg (USD/GBP) will have trillions of dollars of liquidity at the touch (ie for immediate trading), with FX marketmakers responding in milliseconds to price movements. That liquidity has the effect of absorbing the market impact of trades so you would have to trade probably 10000x larger in size in order to make any kind of meaningful dent in the liquidity and even then the market would reach equilibrium again in milliseconds rather than hours/days/never as happened here.

So this article includes a background on the algorithmic stabilizing mechanism.

>If UST’s market value was well above peg, say $1.10, an arbitrageur could purchase $1.00 worth of LUNA and swap into UST via the on-chain mechanism — providing the $1.00 of LUNA in exchange for one newly minted UST, worth $1.10. ...

>In the inverse scenario where UST’s market value was below peg, say $0.90, an arbitrageur could purchase the discounted UST and swap into LUNA via the on-chain mechanism.

Am I reading that right? That sounds like the LUNA protocol is effectively writing two options: to convert 1 UST to $1 worth of LUNA at current market price, and vice versa. But writing an option inherently means exposure to a risk of significant loss from high volatility?

I don't know where I'm going with this but it seems like writing those two options in any other context would be insane.

This is probably the best thing I have worked with!