Why UST Is A Superior Stablecoin

What makes a stablecoin… “stable”? And how can we trust that a stablecoin (and our investment into it) can keep that balance? In an emerging world of various stablecoins (USDC, USDT, UST, Dai, Iron) we will explore how stablecoins keep their peg, what went horribly wrong with Iron, and how UST contrasts from Iron.

Part I — Belief (Non-Technical Point)

For every $1 of Dai, there’s $1.75 of crypto collateral

For every $1 of USDC, there’s $1 of US Dollar collateral in an actual bank account

For every $1 of USDT, there’s “supposedly” $1 of US Dollar collateral

For every $1 of UST, there’s an algorithm to maintain parity with the US Dollar. More detailed explanations and analysis to follow…

For every $1 of Iron, there’s $0.75 USDC and $0.25 Titan

For every $1 of USD, there’s…belief in the American government as a responsible ongoing entity. And a superpower.

In reality, none of these mechanisms matter. The US government can confiscate centralized reserves. An audit could reveal that USDT is not 100% backed by US Dollar collateral and other stablecoins could fall into a death spiral due to their algorithmic stablecoin nature. That’s literally what just happened to Iron.

In reality, money is “belief.” It’s psychological; it’s powerful.

Belief in the Terra Ecosystem is clearly critical. Belief in the protocol based on its mechanics, its math, its history, and its team. Belief is multi-faceted, which is why we wrote this article this way. Belief has enabled our species to make incredible advancements in civilization, and will continue to do so in crypto. We choose to believe because we’re informed. And most importantly, because we’re hopeful of what that future represents.

The Terra ecosystem is practical with payment processing uses like Chai. As Intellabridge is betting its future on this ecosystem, we constantly evaluate the ecosystem and are now starting to provide our POVs openly for public discourse.

There’s a concept called the “Lindy Effect”, whereby the longer something’s been around, the longer its life expectancy is. Now, clearly this is flawed (hello DVDs and AOL), but it does demonstrate the power of network effects. It’s much harder to crash a crypto token at a high market cap with a multi-year history than a smaller, much more volatile crypto token. This is why many people believed that Bitcoin has crossed the “tipping point” and that it will stay around for the foreseeable future. When was the last time something went from $1+ trillion market cap to $0 in a relatively short period of time? Never.

In the same way, not all algo stablecoins are equal. The Terra Ecosystem has been around since 2018; it’s not some random project made last month from anonymous engineers in a shoddy basement somewhere. It’s led by a Stanford-educated engineer with investors that include Coinbase, Galaxy Digital, Arrington Fund, and Robinhood’s CEO. Check out the Robinhood and Ark Invest podcast episodes dedicated to the Terra Ecosystem.

What we believe is that the Terra team is run by some of the most sophisticated minds in the crypto world. Forget the credentials, you can tell from just reading their whitepapers and the products that they have built from a technical, business and economic financial perspective…. The Terra Ecosystem is simply genius. One of the most important concepts in the VC world is to invest in the people. More precisely, when it comes to “faith”, there is a clear advantage in Terra’s feedback mechanism. It can be adjusted through governance. More to follow on that subject later.

Part II — Mechanics on Intrinsic Value (Technical Point)

We’re in the early days of crypto, where conservatism reigns. In a traditional bank, when you deposit $1000, your bank loans out $10,000. This is called fractional reserve banking. To contrast, loans in crypto are over-collateralized, which means that for every $1000 you take on as a loan, you must deposit $3000 in collateral to the protocol. That was a core principle behind the first major algorithmic stablecoin, Dai. For every $1 in Dai, there’s $1.75 in collateral. However, overcollateralized stablecoins are inefficient; they don’t scale well.

As a result, we saw a rise of various algorithmic stablecoins (ESD, DSD, AMPL, BAC) experimenting with different mechanisms. However, many of them have failed, as in they’re worth closer to 30 cents than $1. And while there’s stablecoins that keep their value (Frax, which Iron forked from), there’s no actual economic incentive to use it.

And here’s where Iron entered with their USDC/Titan hybrid model.

This protocol had two features, minting and redeeming. when you minted, you basically traded in Titan and USDC in exchange for Iron tokens (this is called “minting”). Sound familiar Terra fans? Minting Iron tokens caused the total supply of Titan tokens to go down, making Titan price go up (basic supply/demand mechanics). Redeeming refers to exchanging your Iron back into USDC and Titan.

From Iron’s own documentation, here was their theory behind this hybrid approach stablecoin.

“Non-collateralized stablecoins require continual growth to be successful. In the event of a price crash, there is no collateral to liquidate the coin back into, and the holder’s money would be lost, as seen with many past projects trying to utilize such design.”

Theoretically…if the price of IRON fell below $1, arbitrageurs could buy it for less then $1 on the market, and then redeem it with the IRON treasury (a contract on the Polygon chain) for $0.75 USDC and $0.25 worth of TITAN.

Without going too deep into Iron, basically you could yield-farm it for 2% return daily. Yah, that’s a red flag (Bitconnnnnnnect).

However, whitepaper theories fall apart when faced with real world realities: Arbitrageurs aren’t degenerate gamblers who care about token price appreciation. They’re literally…arbitrageurs. So they dump it on the market to collect their risk-free profit. This selling pressure on TITAN creates a de-pegging.

Before I go further…Luna is NOT infallible. During the May collapse, Luna arguably fell to below the amount of outstanding UST. As people exit UST, it caused Luna price to crash further.

All algorithmic stablecoins face this risk. This brings up the common criticism: “collateralizing a stablecoin with another coin from its own ecosystem is a faulty approach that’s doomed to fail!”

So…why did the Terra Ecosystem survive? And why did Iron die?

The answer is simple:

Luna is backed by actual value.

Titan is backed by nothing.

Iron doesn’t have cash flow. In finance, we call that as having no “intrinsic value.” Do Kwon sums this up:

In many ways…this has elements of a Ponzi scheme. So what are the fundamental mechanical differences?

1) Iron does not have its own blockchain from which to get consistent transaction fees for those who lock TITAN up. So…in moments of intense selling pressure (causing a death spiral), what backs up Iron? Nothing. To contrast, the Terra Ecosystem collects transaction fees every time an action is taken in the ecosystem, which gives Luna value. Luna will never be worth $0, and because of that, it will always return to its $1 peg in the long-run (and consequently always have value to arbitrageurs). In fact, Luna powers (and collects fees through) the largest decentralized application in the world by transaction volume, Chai (like Korea Venmo). Terra has intrinsic value. Lots of it. You can think of Luna as a bond with continuous cash flow coming from its ecosystem transaction fees.

2) Iron does not have a blockchain to foster an inclusive dapp ecosystem. Iron could never replicate a sustainable and robust internal economy like where Terra’s stablecoins can play in. Within the Terra Ecosystem, you can buy stocks, earn 20% stable interest, take a loan. In the future you can even participate in lotteries, donate to charities, make purchases via debit cards, etc.

3) Since Terra Ecosystem has real-world application and its principal Luna token is based on UST growth, its ultimate counter to an industry-wide crypto crash is simply by having more adoption. This is where the robust ecosystem of use cases around the Terra ecosystem comes in, all of which result in greater creation of UST. The creation of UST isn’t coming from more DeFi yield farmers looking for 2% daily gains, but from users looking to put their savings on Anchor for a stable 20% interest rate.

Additionally, the Anchor system in the future will be backed by multiple yield-generating tokens as collateral: Ethereum, Solana, and Polkadot. This de-risks the ecosystem from the FUD accusation of Terra being “a coin that’s collateralized by a sister coin.”

Part III — Mechanics on Swap Limits (Technical Point)

During periods of mass-selling panic, the Terra Ecosystem enforces a cap for minting / burning in the Luna mechanism (via the jump trading proposal). This is limited at $100 million per day. Let’s say luna price crashes to $1, this means that at most 100 million Luna will be minted on that day. If Luna price crashes again to $0.1 the next day, then another 1 billion Luna will be minted.

This is significantly better than Titan. There is no limit in Iron, causing infinite minting of 27 trillion Titan in hours, all with $0 value.

Part IV — Mechanics on Redemption (Technical Point)

I’m just going to quote this post from Reddit.

Iron loses it’s peg due to Titan losing value rapidly caused by a whale selling, which meant the 10 minute aggregate of Titan now lags behind significantly. To clarify, if you redeem 1 Iron, you get 74 cents of USDC and 26 cents worth of Titan, but the Titan price is based on the 10 minutes average of the spot price, so if it’s flash crashing the redemption price for Titan is too high.

The spot price of Titan keeps dropping. People redeem Iron for 74 cents and some Titan, and instantly sell the Titan to the market. This Titan is FRESHLY MINTED DUE TO THE REDEMPTIONS. The sell pressure is enormous and increasing quickly due to people panic selling, but also people redeeming Iron for a profit and insta-chucking the Titan on the market.

As people were mass-redeeming, price kept dropping and more and more Titan was being pumped into the economy. At this point it became very obvious that the redeeming process with the 10 minute aggregate is flawed, this was told by me and many others to the devs in their telegram (when Titan was around $30 after losing the peg a second time).

Lo and behold, they did not ‘got this’ at all. Titan went down like the Titanic, and did not stop at any point. The supply of Titan started around 14 million, hit 30 million on the way down, then 100m, 900m, 10b, 926b, 27,805b. Even though the max supply is supposed to be 1b, the devs did NOTHING to stop this redemption spree until it was already too late.

Our commentary? That’s really titanic. When you’re looking at a new coin with a low market cap, any large holder of that token (a “whale”) can sell off a large chunk in an instant and start off a chain reaction of automated liquidations, regardless of how strong the fundamentals are. And in those moments, these mechanics matter. These flaws in Iron’s mechanics were proven fatal, while the Terra ecosystem has overcome multiple 50%+ crypto market crashes over the past four years.

There’s another strong forensics article linked here that further talks about flaws in the Iron architecture. Apparently there was a “giant re-entrancy vector” in all three masterchef contracts that allowed anyone to skim cash from the protocol. This article goes pretty deep and makes some strong, substantive arguments pointing to specific actors responsible for the crash. While we won’t repeat those exact alligations here, what we will say is that these situations are why we look at protocols that have lasted multiple years with institutional support like the Terra ecosystem. Trust matters. Blockchain technology is an innovation in this space, but not the only trust factor that matters. The team matters. The track record matters. That’s why we view our mission at Kash as so critical; we bring additional layers of security and customer service to bring trust to the forefront of the DeFi conversation.

Part V — Team (Non-Technical Point)

Mark Cuban was an investor in Iron.One would think that’s as legit as it gets. Well, even Mark Cuban admitted to not doing his homework here. Let’s do ours and look at the actions of the actual Iron team.

  1. During the death spiral, the Iron moderators kept saying this is normal price action after a huge pump, telling members that this problem will resolve itself. Then…when the situation escalated, they muted the Telegram group.
  2. Apparently the creators of Iron had a track record of 4 failed projects (including getting hacked twice on their previous Binance Smart Chain projects).
  3. In a post-mortem, they blamed a “bank run” for the rapid decline in price, without acknowledging mechanical problems with Iron.
  4. Despite all this, they’re working on their next project (Iron Stablecoin v2). Let’s see how they design and manage the mechanics this time around..

Compare that with the team referenced in the beginning of this article. The same team that came out with immediate responses to the Terra crash from its official Twitter account and from leader Do Kwon himself.

Finally (and honestly), Luna has “buyers of last resort” in VCs and Do Kwon to prevent a complete catastrophe event.

Belief matters.

--

--

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Kash Team

Kash Team

197 Followers

We're Kash.io, the very first third party ecosystem partner invested by Terraform Capital. We're excited to be bringing the Anchor Protocol mainstream.