Lunacy Episode 2: Confidence Game
This week we delve into terraUSD’s winners, its losers and the rules of the whole twisted game.
This week we delve into terraUSD’s winners, its losers and the rules of the whole twisted game.
"Crypto Crooks" is sponsored by Chainalysis.
TerraUSD and its so-called “ecosystem” were always a game – one only some investors realized they were playing.
As we fall deeper into the death spiral, we look at the retail investors who were indisputable victims of Do Kwon’s scheme. Many of them lost life savings and even, ultimately, their lives, in the wave of suicides that followed the stablecoin's depegging. We also look at the more complicated role of financial professionals in propping up Do Kwon’s empire, many of whom got out of the racket millions of dollars richer than when they’d gone in.
After this, two more episodes to come – plus a bonus fifth on the bombshell revelations of a new Securities and Exchange Commission investment lawsuit against Do Kwon and Terraform labs.
Chainalysis is the blockchain data platform. We provide data, software, services and research to government agencies, web3 companies, financial institutions and insurance and cybersecurity companies. Our data powers investigation, compliance and business intelligence software that has been used to solve some of the world’s most high-profile criminal cases. For more information, visit www.chainalysis.com.
“Crypto Crooks” is a CoinDesk Podcast Production. The executive producer is Jared Schwartz, with additional production by Eleanor Pahl, Nora Battelle, Jonas Huck, and Moon Beast. Fact-checking is by Amber Von Schassen, and sound design and music are by Altus Noumena. This show is written and voiced by David Z. Morris.
Audio Transcript: This transcript has not been edited and may contain errors.
Content Warning: This episode includes substantial discussion of cases of suicide.
DAVID Z. MORRIS:
On May 12 of 2022, a figure lurked outside of an apartment complex in the Seongsu-Dong neighborhood of Seoul. The former industrial neighborhood is sometimes known as “the Brooklyn of Seoul,” full of repurposed warehouses and stores built in shipping containers.
The figure, according to police reports, waited for a resident of the building to unlock the front door, then slipped inside behind them.
This individual then made their way to the apartment of the most notorious South Korean alive.
They rang the doorbell, and knocked.
Until a young woman answered.
The trespasser had just a single question: “Is Do Kwon here?”
The intruder must still have been in a state of shock. In the space of barely a week, this person would later tell reporters and police, they had lost the equivalent of $2.3 million U.S. dollars due the collapse of Terra, Do Kwon’s much-hyped blockchain system. That included, they told the BBC, $800,000 invested in the Terra system near the top of the market.
The young woman who answered the door was Do Kwon’s wife. It’s likely there was at least one other person home that day – the couple’s infant daughter. Her name is Luna.
It had all happened so fast.
On May 6 of 2022, the Terra blockchain was still one of the most hyped projects in crypto. Its so-called “algorithmic stablecoin,” TerraUSD, used fancy financial engineering to ensure it always traded at a price of one dollar. Terra’s creator and figurehead, Do Kwon, was controversial but powerful. He had amassed a huge war chest, and was launching a plan to buy $10 billion worth of Bitcoin, vastly increasing his influence.
But financial reality had other plans.
On May 7th of 2022, TerraUSD’s dollar peg wavered – at first, by just a single penny. But that was enough to set off a so-called “death spiral” – a fatal flaw in TerraUSD’s design that experts had been warning about for years before Terra and Do Kwon rose to prominence.
TerraUSD spiraled downward at blinding speed: By the evening of May 9th, the one-dollar stablecoin had plummeted to sixty-five cents. On May 11, it dropped under 30 cents. By May 13, the magical blockchain dollar could be had for a much less magical eight cents.
This in turn triggered wholesale printing of Luna, the ‘balancer token’ that was supposed to keep TerraUSD’s value at one dollar - and the namesake of Do Kwon’s child. This hyperinflation drove the Luna token’s price into the gutter, from nearly $65 per token on May 8 to a chilling $0.00025 dollars on May 15. That’s just over two one hundredths of a penny – a decline of 99.99%, at precisely the moment Luna was supposed to be able to backstop TerraUSD.
From a peak paper value of nearly $60 billion dollars on May 8th, the value of the entire Terra system had dropped to effectively zero by May 15th.
When TerraUSD first wavered from its peg on May 7, Do Kwon had been asleep. And on May 16, when a distraught visitor invaded his home and confronted his wife, Do Kwon wasn’t home. He was in Singapore, where Terraform Labs had set up its new headquarters.
Thankfully, when Do Kwon’s wife told her unwelcome visitor that Do wasn’t home, the intruder left.
Later, it was revealed that the visitor was a South Korean crypto investor and commentator known by the name Chancers. His goal, he revealed, was simple: he wanted an apology from Do Kwon. As we’ll see, that wouldn’t be fast in coming.
But whatever Chancers’ intentions, soon after his unwanted visit, Do Kwon’s wife requested police protection.
Her husband certainly wasn’t providing any: As the collapse of his flawed creation triggered waves of despair and rage for more than 200,000 victims worldwide, he would flee from Singapore to Europe.
By December of 2022, South Korean prosecutors would claim he had run as far as Serbia.
But the consequences of Do Kwon’s failures would travel the globe right alongside him.
The Terra system had attracted investment not just from tens of thousands of hopeful individuals, but from big companies in the cryptocurrency industry. Venture capitalists made small investments to help create the entities that built Terra. Later, hedge funds and investment firms bought Terra-based tokens, placed capital in the Anchor system, or got entangled with Terra through lending relationships.
Some of those names may be familiar: Three Arrows Capital. Celsius Network. Jump Trading.
The collapse of the Terra system was immensely damaging to some of those big investors. Some of them in turn defaulted on their obligations to still other entities.
A cascade of defaults spread from one major crypto financial firm to another, eventually wounding many who had no direct exposure to Terra at all. That included CoinDesk parent company Digital Currency Group. A DCG company called Genesis made large loans to the hedge fund Three Arrows Capital. Three Arrows ultimately defaulted on those loans in part because they had made big bets on Do Kwon and Terra.
Crypto had been born in the wake of the 2008 financial crisis, when the failure of a complex financial product – in that case, mortgage backed securities –brought the global financial system to its knees. Bitcoin was, among other features, designed to be a cash-like bearer instrument that involved no debt, leverage, or counterparty relationships.
But it only took 14 years for Do Kwon, with more than a little help from his many enablers, to recreate the same kind of systemic risk that had caused the great financial crisis. His dream of an algorithmic stablecoin unleashed the exact force cryptocurrency had been created to fight against:
Welcome to “Crypto Crooks,” Season 2: Lunacy – The Rise and Fall of Do Kwon.
This is Episode 2: Confidence Game
Hello, I’m David Z. Morris, CoinDesk’s Chief Insights Columnist. I’ve been reporting on cryptocurrency since 2013. I’ve followed crypto so closely because I believe it will profoundly transform how we live.
But I’ve also watched as the real promise of crypto has been undermined again and again by a stream of con-men and hucksters who prey on the uninformed, the naïve, and the desperate. It remains a lightly-regulated and poorly-understood sector, which makes it a perfect hunting ground for scammers. A big part of my job here at CoinDesk is calling out sketchy projects and outright frauds, and I’ve seen enough of them over the years to know what they look like.
We’re presenting “Crypto Crooks” to help educate and protect crypto investors, or just the crypto-curious, from repeating the mistakes of history. Each short season of the show focuses on a different major fraud, crime, or bad idea that fleeced investors and the industry. Because while crypto is a radical new technology, the fraudsters using it to rope in victims use techniques as old as the financial industry – in some cases, as old as money itself.
This season, we tell the story of a man whose elite credentials and big personality distracted from the flaws in his amazing investment pitch – until, very suddenly, they didn’t.
The collapse of Terra destroyed wealth around the globe, and at every level. If Terra’s parent company Terraform Labs had been a traditional early-stage technology firm, it would have been funded mainly by venture capital funds and wealthy individuals.
But a lot of average people invested in Terra. Some bought the Luna balancer token, which promised to grow in value as Terra grew. Others deposited TerraUSD in a bank-like service called Anchor, which enticed them by offering 20% annual returns – vastly higher than a conventional bank.
So when it all came crashing down, there were countless stories of despair and broken dreams from people who had trusted Do Kwon and his creation. Many were shared on a newly-created Reddit forum called r/TerraLunaVictims. Others surfaced through the media.
These stories show an astonishing diversity of victims. The Wall Street Journal highlighted a 44 year old American surgeon who put $177,000 - a decade worth of savings - into a crypto-savings service called Stablegains in April of 2022. He claimed not to know Stablegains was converting his money into TerraUSD - until the token’s meltdown destroyed his savings.
The surgeon said he would have to cut costs for a while to climb back from those losses. But for other Terra victims, the consequences were much darker.
A 37 year old man in Chengdu, China told Business Insider that he spent the equivalent of $90,000 buying TerraUSD tokens. He did this, he said, because malignant cancer was preventing him from working full-time, and he needed recurring income. He may have been enticed by the yield payouts offered in Anchor.
By late May, his $90,000 in supposedly stable currency would have been worth less than $3,000.
At least two confirmed suicides of Terra investors occurred soon after the collapse, though there are likely far more. A 29 year old man in Taichung City, Taiwan died on May 24th. His $2 million stake in the Luna balancer token had crashed to just $1,000 after its hyperinflation.
Another suicide involved truly bleak irony. Do Kwon’s Terraform Labs helped fund a crypto media outlet called Coinage, whose founders had expressed a positive outlook on the Terra/Luna system. A relative of one of the Coinage founders, presumably swayed to invest at least in part by the founders own positive stance, died by his own hand after the collapse.
These were the consequences of Do Kwon’s hubris and failure. In hindsight, it may be tempting to say victims should have been more careful when investing in a highly experimental product. Terraform Labs itself said as much - when the Wall Street Journal asked about investor losses, Terra’s creators said that “each individual must decide for themselves what risks they are willing to undertake.”
But in the months before the Terra collapse, Do Kwon rarely suggested there was any risk at all. He was unwavering in his assurances that Terra and its magical money machine were rock-solid – and in his quickness to insult and belittle anyone who dared challenge his certainty.
And Do Kwon’s actions – including his quest to build more collateral into his uncollateralized system – didn’t always match that bluster. So there are concrete reasons to think he didn’t actually believe his own descriptions of TerraUSD as a “stablecoin.”
Those discordant words and actions are one reason South Korean authorities now appear to regard Terra, not as a risky financial experiment that failed, but as an outright financial fraud.
This sort of deception is easier in cryptocurrency than in traditional equity markets. That’s in part because it’s easy for average people, often referred to as “retail investors,” to invest directly in crypto over the internet. A big personality like Do Kwon can be very convincing to those non-experts.
But a bigger mystery is the deep involvement of supposed professionals in what any first-year MBA should have realized was a fundamentally flawed idea. Venture capital firms like Galaxy Digital, Coinbase Ventures, and Arrington XRP Capital made Terra look all the more enticing to the individual retail speculators who followed in their footsteps - but in the end, this was nothing more than the blind leading the blind.
Exposure to Terra wasn’t the only factor that led to the string of interrelated defaults and collapses by crypto financial firms in 2022. The main culprit was leverage itself, which amplified the damage of bad bets on assets from Bitcoin to Ethereum derivatives to the Grayscale Bitcoin Trust, an investment vehicle run by CoinDesk’s sibling company, Grayscale.
But most of these were matters of bad timing and excessive leverage. Terra was the biggest single bet in the crypto bubble to have turned out to be effectively worthless.
The story of Do Kwon and Terra, then, is about more than a single fatally flawed system. It’s the story of how an entire industry fell for a confidence man – and how leverage and financialization made that mistake infinitely worse.
To be fair, some Terra investors acknowledged that they didn’t really believe the algorithmic magic-dollar hype.
Mike Novogratz is head of the investment fund Galaxy Digital, one of Terra’s early backers. Here’s Novogratz talking to podcast host Anthony Pompliano in an interview released April 6 of 2022 – barely a month before Terra collapsed.
“Listen, this is an interesting experiment. It’s not without risk, right. He’s in this transition. Right now the plan is to buy 10 billion dollars worth of bitcoin. And as that ecosystem grows that number will grow. That’s all good as long as there’s not a run on the bank. And so we should all be open minded. Like, he’s got such good momentum.”
DAVID Z. MORRIS: Wait. Wait. Nora?
NORA BATTELLE: Yeah?
DAVID: You found this clip, right? You’re sure it’s real?
NORA: I’m pretty sure it’s real, David. It’s from a fireside chat Novogratz did with Anthony Pompliano on April 6. It was at the Bitcoin 2022 conference in Miami.
DAVID: April 6? Man, that’s just a month before the Terra death spiral. He really said “as long as there’s not a run on the bank?”
NORA: Oh yeah, he said it. And that was the day after Luna hit an all-time high - $119 dollars. It was all downhill from there.
DAVID: Well, then it’s a good thing analysts say Galaxy had already sold most of its Luna and Terra holdings by April 6. But why would Novogratz be talking up something he had already sold? [muttering] man, what the heck …
DAVID: Okay Okay … Amber?
AMBER VON SCHASSEN: Yeah?
DAVID: You fact checked this, and you’re sure this is real? It’s not like, a deepfake?
AMBER: Oh, it’s definitely real. You can still go to Bitcoin Magazine’s YouTube Channel and watch it.
DAVID: Okay, well … I really hope you two know what you’re doing. But I have to hear it again …
Altus, run that back.
“That’s all good as long as there’s not a run on the bank.”...
“It’s not without risk”...
“As long as there’s not a run on the bank”...
“Not a run on the bank”....
“run on the bank” “run on the bank” “run on the bank”....
“ruuun oooon theeee baaaaaaaaaaaaank”....
And there it is. One of Do Kwon’s and Terra’s most prominent supporters, a man who infamously got a Luna tattoo, acknowledging the massive risk inherent to the system.
When Novogratz says “run on the bank,” he’s talking about the death spiral that experts like Cyrus Younessi, Preston Byrne, Francis Coppola, and Ryan Clements had warned of – and which eventually unfolded.
It’s also quite curious that these early-April statements came after Galaxy had seemingly sold off a lot of its position in the Terra system.
“Our takeaway from GLXY’s disclosures regarding its LUNA exposure is that it appears that the company had sold all or most of its position at a healthy gain during [the first quarter of 2022],” wrote BTIG equity research analyst Mark Palmer in early May, after Terra’s collapse.
But if Novogratz really believed in what he was saying at the Bitcoin 2022 conference – why had he sold?
To be fair, Novogratz was far from alone in, it seems, basing investment decisions on a basket of muddled, contradictory ideas. Here’s Kyle Davies, one of the heads of Three Arrows Capital, defending his hedge fund’s decision to invest in the Luna balancer token, in a clip highlighted by fraud hunter Cas Piancey:
“This was the first time in history, for crypto anyways, that two coins, both of them, were top ten, and they went to literal zero in a matter of days. This was an extreme black swan. We thought, first of all, that this would be able to last longer …”
The combination of stupidity and dishonesty baked into those three short sentences is staggering.
A “black swan event” means something completely unpredictable – something that Davies and his team could never have seen coming. But Davies then says he “thought this would be able to last longer.” This implies he thought that Terra would collapse eventually, but hoped Three Arrows could make some money first.
This is just another version of the rationalization that hooks people who think they’re smart into ponzi schemes: “I won’t be a sucker like the rest of them. I’ll get out before this collapses.”
As it turns out, Three Arrows didn’t get out in time.
They lost at least $200 million dollars on Luna, and maybe more - money effectively loaned to them by other crypto companies. They became one of the first entities destroyed in large part by Terra’s collapse. They defaulted on massive loans, spreading the damage far and wide.
Three Arrows also appears to have been engaged in activities that crossed the line from brain-dead to illegal – both Davies and cofounder Su Zhu are now, like Do Kwon, on the run.
Novogratz did say something else revealing: ecosystem.
The Terra ecosystem was the optimistic narrative of choice for people who were betting on Do Kwon, but didn’t entirely buy the algorithmic stablecoin fluff. Do Kwon loved to talk about building an “ecosystem” that would help support TerraUSD’s one-dollar peg by creating demand for the token. For instance, if someone wanted to buy a digital collectible stored on the Terra blockchain, they would first have to buy TerraUSD to do it, helping prop up the price in the process.
In January of 2022, as part of this overall strategy, Do Kwon announced the creation of the Luna Foundation Guard. This entity, superficially independent but in reality an extension of Terraform Labs, would be tasked in part with deploying capital to defend the TerraUSD peg – which we’ll learn a lot more about soon. But its mandate also included “allocat[ing] resources supporting the growth and development of the Terra ecosystem.” That included distributing grants to developers.
That grant program only lasted a few months before Terra’s meltdown. But developers had already devoted their blood and sweat to building a variety of applications that interfaced with the Terra blockchain. The largest included a decentralized crypto exchange called Loop and a lottery called LoTerra. Chai, a payments system that was founded alongside Terra, used the blockchain as infrastructure, for a little while – but we’ll have to save that story for an upcoming episode.
Another application on Terra, Mirror, had been built in-house by Terraform Labs. Mirror was what’s known as a “bucket shop,” which allowed users to bet on the price of stocks without actually owning them.
Bucket shops have been illegal in the U.S. since the early 20th century, because they’re a form of gambling disguised as investing. This led to Do Kwon’s first major brush with the law. The U.S. SEC was investigating Mirror as early as the Fall of 2021, when Do Kwon was served with a subpoena at the Mainnet crypto conference.
So this was one big part of the “ecosystem” Do Kwon and his investors wanted to see grow: gambling.
The ecosystem argument is convincing at first blush. It’s ultimately why big cryptos like Bitcoin and Ether have value. The price of a digital asset derives from the demand to use its functions: cross-border payments and decentralized wealth storage for Bitcoin, complex finance and NFTs for Ethereum. Do Kwon thought that if he came up with applications that required TerraUSD or the Luna balancer token, it would ultimately support the value of both tokens.
“Is it possible to go from Ponzi to stability through widespread acceptance? If we start out with a subsidized Anchor protocol, which isn't sustainable, that's being propped up by reserves? Well, a lot of startups aren't sustainable, they're propped up with VC money, but over time, they become sustainable …”
The ecosystem goal does seem to undermine Do Kwon’s core claim about TerraUSD. If an algorithm can really stabilize a dollar coin, why is this added boost from an “ecosystem” necessary?
But we’ll set that to one side for now. The real problem with the ecosystem play – and the reason Ryan Clements used the word “ponzi” – is that while a lot of applications were built on Terra, only one really mattered. The vast majority of all the TerraUSD in existence flowed into the Anchor Protocol.
Anchor was effectively a savings account, just managed on a blockchain instead of by a bank. But far more than the underlying technology, Anchor’s competitive advantage was its interest rate. At a time when most regulated banks around the world were paying next to no interest on deposits, Anchor was paying nearly 20% interest, or yield. So if you deposited $100 in Anchor, you’d receive close to $20 in interest over the course of a year. A normal bank at the time might have paid you just a few cents.
There was a catch, though: For most of Terra’s existence, to get that juicy 20% yield, you had to deposit TerraUSD into Anchor.
This was Do Kwon’s ecosystem play in action – Anchor was a product that seemed so good that it was drawing immense amounts of capital into Terra and TerraUSD. This helped sustain the dollar peg and boost the price of the Luna balancer token.
At its peak in early May of 2022, Anchor held more than $17 billion dollars’ worth of TerraUSD. That amounted to roughly 70% of all TerraUSD in existence. Most if not all of that money was attracted to Anchor because of its very high interest rate.
That included many of the small individual investors who would later be wiped out. It included many people with effectively no understanding of cryptocurrency. That’s because both Anchor and TerraUSD were marketed and widely perceived as stable and safe – especially in South Korea.
That brings us to a second catch with that fat Anchor yield – The 20% yield wasn’t coming from real borrowers.
Instead, yields were paid from an external fund, with money largely provided by venture capitalists and large investors. As we’ll see, and as critics tried desperately to warn at the time, this was a massive Sword of Damocles hanging over the entire Terra system.
Before we dig in further, let’s remind ourselves how banking works.
At the most basic level, a bank has a straightforward business model – it charges more for loans than it pays in interest on deposits, and pockets the difference as profit. This means that if there’s less demand for loans, a bank earns less interest from them. In turn, banks will generally have to lower the interest rate they pay out to depositors. When money is in higher demand, for instance because central bank interest rates rise, banks can pay depositors more interest for their money.
But that’s not at all how Anchor’s interest rate was set. Instead, multiple sources familiar with the discussions told us that Do Kwon simply declared that Anchor’s interest rate should be high to help entice more depositors. He specifically chose a rate that was not just far higher than any bank’s, but also higher than many other yield-generating blockchain projects.
This worked, as far as it went – Anchor’s sky-high yield attracted money like a magnet. And it wasn’t just individual crypto degens and average retail depositors. Large institutions began to try their luck with Anchor, too. For example, Celsius, a bank-like service operating in the United States, generated some of its own outsized yields by depositing its customers’ funds in Anchor, at one point around $500 million dollars’ worth.
Another company called, unbelievably, “Stablegains,” premised much of its business on acting as a middleman for Anchor deposits - including for the surgeon profiled by the Wall Street Journal. A later lawsuit would allege Stablegains lost $44 million of investor funds when Terra collapsed. Stablegains was funded in part by legendary Silicon Valley incubator Ycombinator. Some time after the lawsuit, Ycombinator deleted references to Stablegains from its website.
That’s what Mike Novogratz was talking about when he said Do Kwon and Terra had “great momentum” - Anchor’s incentives were working just as intended.
“He has such great momentum”
So far, so good. But remember that this 20% rate of return was not being paid by those borrowing from the Anchor bank. Anchor deposits started growing rapidly around September of 2021, but demand for loans didn’t keep up. By late April of 2022, according to data from Dune Analytics, there was four times more TerraUSD deposited in Anchor than was being borrowed.
So depositing in Anchor was the biggest use case for TerraUSD. But Anchor itself was a money-losing operation that only survived thanks to subsidies from outside the Terra system. Compare that to MakerDAO, the overcollateralized stablecoin that we discussed in episode 1. Maker adjusts its interest payments based on loan demand, and has now functioned properly through two bear markets - which is two more than Terra survived.
Instead of borrowers, the 20% yield was being paid from something called the Anchor Yield Reserve. In principle, this was a pool of TerraUSD intended to act as a short-term buffer for Anchor when yield payments exceeded revenue from borrowing.
But the widening gap between borrowing and lending on Anchor meant the Yield Reserve was steadily depleted starting in late 2021. By January 28 of 2022, the yield reserve contained 35 million TerraUSD, and was declining by an estimated 1.25 million TerraUSD per day. In late January, that meant Anchor would only be able to pay that 20% yield to Anchor depositors for roughly 20 more days.
Crypto markets were starting to get spooked. Was the Anchor gravy train about to come to a screeching halt – or worse? For a brief moment in late January and early February, depositors actually began to withdraw funds from Anchor.
But no, not to worry. On February 17, Do Kwon and Anchor announced that the yield reserve had been “topped up” with $450 million dollars, reportedly from the Luna Foundation Guard. This reassured the market, and deposits into Anchor continued to surge.
But that’s the opposite of what the takeaway should have been. First, because the reserve “top-up” was effectively a bailout for a bank with a broken business model. The basic math didn’t change – in fact, it got worse. By early March, Anchor was running a deficit estimated at $2 million dollars per day.
It also matters where the funds for the top-up were coming from. The deployment of a subsidy by the Luna Foundation Guard, founded by Do Kwon, showed that there was nothing decentralized about Anchor, or about the Terra system in general.
But a more fundamental issue was the original source of the money. The Luna Foundation Guard was a separate entity from Terraform Labs, and had raised close to $1 billion from outside investors. Those included Jump Crypto (a subsidiary of Jump Trading) and Three Arrows Capital.
So the top-up came from an investor subsidy, not revenue. You can compare this to the early years of Uber, when rides were cheaper because of subsidies effectively paid by venture capitalists. This made it harder for competitors, in that case Lyft, to grow. Similarly, Anchor’s subsidized rates were helping it take market share from competing services – including the old-fashioned banks where, say, a cancer patient might normally keep the money they needed for their treatment.
There’s an incredible side-note to all this. The idea of using deficit subsidies to grow startups was championed for many years by Peter Thiel, a cofounder of PayPal. Theil’s anti-college Thiel Fellowship gave one of its $100,000 grants to Ryan Park in January of 2022. The grant was in recognition of Park’s creation of the Anchor Protocol – just months before the entire Terra system went to zero. We’ll revisit that tidbit in more depth later.
The goal of subsidizing Anchor was, as with Uber, growth. But the endgame was a bit vaguer: with enough growth and enough traction, TerraUSD would be stabilized by things other than Anchor, and the subsidy could be withdrawn. This is why Ryan Clements describes Anchor, and Terra more broadly, as similar to a Ponzi scheme – outside capital was being used to make it seem like a very profitable investment, when in reality, Anchor was constantly losing money.
“Maybe there was an interpretation of this from the lens of VC that instead of a Ponzi, it was more of like a subsidy, that they're giving the ecosystem subsidies. And while UST is being found to get use cases … but the problem was there were no use cases.”
There were no use cases for TerraUSD, that is, except for depositing it in Anchor to collect the inflated, unsustainable subsidy.
But the magnetic appeal of Anchor’s sky-high yield also meant it was a ticking time bomb.
Do Kwon and his backers argued that there would eventually be so much activity on Terra that the Anchor subsidy could be withdrawn, and the interest rate lowered to something sustainable, without harming the overall ecosystem. But this was always discussed in very vague terms, without much analysis of when such a tipping point might arrive.
This isn’t so different from the way Uber and other companies talked about their use of subsidies to gain market share. After a few years of artificially low rates, they would become ubiquitous, enmeshed in people’s lives, while also starving their competitors. So when they raised their rates to a profitable level, people would still use the service.
This was a very familiar argument to the people who supported various parts of the Terra ecosystem. Peter Thiel, YCombinator, and Mike Novogratz are all part of the technology world, where changing people’s habits to take advantage of new technology really can permanently reshape markets.
But Terra, despite its blockchain trappings, was never really a technology innovator. Do Kwon and his team were merely using well-established blockchain technology as a platform for financial experiments like the TerraUSD stablecoin.
It was a huge and fundamental error for Do Kwon’s backers to transpose their tech-industry, subsidized-growth model to finance. The Anchor subsidy would never have produced the same results as the Uber subsidy. Even if TerraUSD hadn’t collapsed, users would have left Anchor as soon as a higher interest rate was available elsewhere.
Another expert who loudly warned of Terra’s fragility in early 2022 was Kevin Zhou, co-founder of the hedge fund Galois Capital. He argued in part that a lot of the money in Anchor was what’s known in finance as “rotator capital.” Rotator capital is money that investors put in whatever is providing the best return at any given moment, ready to rotate it to something else at a moment’s notice. When I spoke to him before the Terra collapse, Zhou argued that the “rotator capital” in Anchor would move quickly as soon as conditions changed.
This is the key flaw in the Uber subsidy analogy. You may use a car service several times a week, but you’re not making a major financial decision each time, so a change in habit can persist after the price of the behavior changes.
But that’s not how banking or investing work. Most people will look closely at the numbers when they make investments. Professional money managers are ready to move money around at the drop of a hat. Investing isn’t something shaped by habits, but by hard numbers.
So when things started looking dicey at Terra, there were plenty of safer places for agile investors to move their money - and that’s exactly what they did. In under four days, between the first signs of instability on May 7 and Monday, May 9th, $5 billion dollars worth of TerraUSD was withdrawn from Anchor.
Here’s CoinDesk reporter Sam Kessler:
“Then there's the people who, you know what, saw this as a total game theory thing. And they were like … Okay, I know that this Terra project is going to collapse eventually … These folks, they knew that the buck was eventually going to drop … you know, this whole thing was going to eventually collapse. And they were like, you know what, now's the time to sell. It's done. It went down to 98 cents. And so those people who were smart, who were some institutional … folks or some big money folks, those were the ones who also left the game, that was Terra.”
And remember — these huge withdrawals were in the TerraUSD token. Not only was the algorithmic dollar itself showing signs of depegging, but there was very little else for investors to do with the token. The other main applications on Terra itself amounted to little more than gambling, and TerraUSD couldn’t be freely used on other systems, like Ethereum.
So those withdrawing from Anchor were suddenly in the market to sell those billions of largely useless math dollars - at exactly the moment that TerraUSD’s own mechanism was starting a self-reinforcing death spiral.
Anchor’s external subsidy had been the main driving force of Terra’s growth. But depositors’ withdrawals and sales of TerraUSD reversed that growth even faster, exacerbating the collapse of the whole system. Some depositors got out in time - including a lot of professional investors.
But many of Anchor’s individual retail depositors had heard little but praise for Anchor and Terra’s safety and stability. They didn’t watch crypto news every day. Many wouldn’t have understood the severity of the threat posed by TerraUSD’s wobbling peg.
So as the professionals and agile rotator capital fled to the exits, average people saw their life savings go up in smoke.
Soon, withdrawals from Anchor combined with the crashing value of the TerraUSD still deposited there to annihilate its value. By May 22nd, the total value in the Anchor bank had dropped to just over $100 million dollars, down from $14 billion on May 6 - a drop of more than 99% in just over two weeks.
The biggest pillar of the Terra ecosystem had crumbled into dust.
“That’s all good, as long as there’s not a run on the bank.”
We’re beginning to get a picture of the many complex forces that led to the Terra disaster. One small team in South Korea built an unsustainable product based on foolish assumptions. But they had a lot of help turning their bad idea into a $60 billion dollar bubble built on imaginary dollars.
For that, they needed big-name investors to cosign their idea, legitimizing it in public. As investors primarily in technology, these supposed experts may themselves not have seen the flaws and traps scattered throughout the Terra system.
Some, on the other hand, didn’t seem to believe that an algorithmic stablecoin could really last. Instead, they focused on what Mike Novogratz called Terra’s “momentum,” even if that momentum was built on deficit spending funding unrealistic yields on Anchor.
These professional investors rotated their capital into something that was growing, while talking it to the sky to the less-informed public. But they were also ready to rotate their money out at the first sign of trouble - and leave the small fry who had followed them to fend for themselves. Many of these Pied Pipers even slipped away in time to turn a profit.
One of the major factors that misled people so badly was a focus on credentials - labels, degrees, affiliations. Here’s more from Galaxy Digital’s Mike Novogratz, delivered at the same conference in early April of 2022.
“I interviewed him for my podcast, I said, ‘Well how smart were you.’ He said, ‘Well I was pretty smart growing up.’ Well I said, ‘How smart?’ Because he got into Stanford. He said, ‘Well I took 15 AP courses and I got 5s on all of them.’”
Stanford would be cited again and again by Do Kwon’s supporters as evidence of his greatness. Terraform Labs staff included Stanford alums. Stanford - the same school Elizabeth Holmes attended before leaving to start the Theranos fraud.
Advanced placement classes, by the way, are taken in high school. I got 5s on several of my AP tests, too, and I’ll tell you right now - you should NOT give me hundreds of millions of dollars to manage.
And while investors were citing Do Kwon’s credentials as evidence he would succeed, people inside Terraform Labs were using the investors’ credentials to silence internal doubters.
“So there was one time I was asking about these those products on the other anchors and mirror … and then they were saying, like, oh, you should trust Terra. Because a 16 z or Galaxy digital, just invested in it. It's gonna be so great that you're, you can't even fight them. That's what they told me. … [Clip 18:08] They were highly relying on Credentialism instead of thinking about how actual product works.”
That’s Hyungsuk Kang, a former programmer for Terraform Labs. Hyungsuk worked at Terraform for just a few months in 2020, before leaving over what he says were serious conflicts over the design of Terra. When he tried to push back against some of the decisions that led to Terra’s implosion, he says, his concerns were brushed off with a hand wave - just look at all the sterling investors who trusted Do Kwon!
But some of those investors, it seems, were themselves more concerned with diplomas and AP scores than with the details of how Terra worked - or, in the end, didn’t.
Do Kwon did have one vital edge in addition to his Stanford degree - his take-no-prisoners personal style, with its Donald Trump-like refusal to ever entertain doubt or criticism. For some, this unwavering certainty was the only proof they needed that Terra was a bandwagon they wanted to jump on.
But some say Do Kwon was just as immune to critical thinking in private, even from the team that was helping him build Terra. Instead of hard questions, what Do Kwon valued was loyalty, placing unqualified but friendly staffers in roles where they couldn’t possibly succeed.
That turned the group building Terra into something that looked less like a technology startup - and more like a crime family.
That’s next time … on “Crypto Crooks.”