Terra: Algorithmic Stablecoins on the Bitcoin Standard

Summary

Since the launch of its mainnet (Columbus-1) in early 2019, the Terra blockchain has seen tremendous growth and the widespread adoption of its native Terra stablecoins. At the time of writing, there were roughly 40k daily active users of the 26 protocols that exist on Terra. The total value locked (TVL) in these protocols’ smart contracts equates to ~$29 billion (USD), making Terra the chain with the second highest TVL after Ethereum. Terra’s flagship US dollar-denominated stablecoin, $UST, has achieved a market cap. that exceeds $16 billion (USD), surpassing that of MakerDAO’s Dai to become the top decentralized stablecoin to date. The market cap. of the chain’s native staking and governance token, $LUNA, is nearly $38 billion (USD).

Terra aims to provide a decentralized form of money that’s easier to use and more attractive to hold than alternative forms of money. It plans to achieve this ambitious goal by way of its seigniorage shares style algorithmic stablecoins, and in particular, UST. These types of stablecoin arrangements employ an endogenous reserve asset (i.e. Luna) that absorbs the volatility in demand for the stable asset (i.e. Terra stablecoins). In Terra’s case, this creates a reflexive relationship between demand for Terra stablecoins and the price of Luna. While this means that Luna’s price should rise as demand for Terra stablecoins grows, the same reflexive relationship works in reverse, and poses a risk of systemic collapse should there be a dramatic decline in demand for Terra stablecoins. To protect against these destabilizing events, builders on the Terra network have created a number of compelling native, inter-chain, and real-world use cases to drive demand for Terra stablecoins. Further, it seems that Terra’s flagship UST will also soon be supported by a reserve of major assets like Bitcoin whose prices are less correlated to Terra’s native assets.

In this report, we’ll take a closer look at the Terra blockchain, Terra stablecoins, and the use cases driving demand for these stablecoins. We’ll describe Luna’s tokenomics, the price stability mechanism of Terra stablecoins, and discuss the implications of the Luna Foundation Guard’s (LFG) proposal to defend the UST peg via a ‘Bitcoin Reserve Pool’.

Overview

The Terra Blockchain

Terra is a layer 1, delegated proof-of-stake (dPoS) blockchain that was built from the Cosmos SDK. The state of the blockchain is secured by a system of verification called the Tendermint consensus, in which the top validators in the network (the top 130 in Terra’s case, ranked by the value of their stake) propose, sign, and add new blocks of transactions in return for staking rewards paid out in Luna. Staking rewards are distributed in proportion to the amount of Luna each validator stakes, and validators with larger stakes are selected to propose new blocks more often – a role that earns them proportionally more rewards.

Staking rewards come from two sources: gas fees and swap fees. Gas fees are compute fees set by the network’s validators that are added to every transaction sent on the blockchain as a means to deter spam within the network. Swap fees are paid by users who swap between Terra stablecoin denominations (e.g. UST ←→ KRT) as well as by users who swap between Terra stablecoins and Luna. All swap fees are directed to the oracle reward pool and dispersed to validators who reliably report an accurate price feed of Luna exchange rates against Terra stablecoin pegs. 

Terra Stablecoins

The primary feature of the Terra protocol is its native suite of algorithmic stablecoins, of which there are currently 22 different denominations. Each of these stablecoins aims to maintain a stable price relative to a particular fiat currency and is named for its fiat counterpart. For example, Terra’s flagship stablecoin tracks the price of the US dollar, and is known as TerraUSD, or UST. These 22 stablecoins are functionally equivalent, differing only in the price that they target, and are collectively referred to as ‘Terra’.

Unlike custodial, fiat-backed stablecoins (e.g. USDT, USDC) or overcollateralized, crypto-backed stablecoins (e.g. Dai, LUSD), algorithmic stablecoins are not explicitly backed by exogenous collateral like fiat money deposits or some popular cryptocurrency that has broad utility outside of the stablecoin system. Rather, algorithmic stablecoins maintain a stable price by relying on game theory between system participants, and arbitrage opportunities created by algorithms built into the protocol (hence, algorithmic stablecoins).

In Terra’s case, supply and demand for its stablecoins are balanced by arbitrageurs who leverage the protocol’s algorithmic market module to swap between Terra stablecoins and Luna. Using UST as an example, this market module allows any user to mint 1 UST by burning $1 (USD) worth of Luna, and vice versa. In this way, the supply of UST is dynamically adjusted to meet demand at the $1 (USD) price peg, while volatility is absorbed by increasing or decreasing the supply and, inversely, the price of the Luna token. Uniquely, it seems that the stability of Terra’s flagship UST will also soon be supported by a reserve of major assets like Bitcoin whose prices are less correlated to Terra’s native assets.

Demand Drivers for Terra Stablecoins

Terraform Labs and the broader community of builders on Terra have developed a host of compelling use cases to drive demand for Terra stablecoins. To name just a few notable protocols built on the Terra network; Mars is a decentralized credit protocol that enables fully permissionless on-chain lending and borrowing, Mirror is a protocol for minting and trading synthetic assets, Astroport is an automated decentralized exchange protocol, and of course, Anchor is Terra’s decentralized savings protocol, offering an unheard of ~20% APY on UST deposits. 

From a design and user experience perspective, Terra builders have put more emphasis on simplistic, easy and fun to use interfaces than many DeFi projects of the past. Whether it’s Terra Station (the native Terra wallet that acts as a central hub for users to swap or stake Terra assets, participate in governance, and more), or one of the protocols mentioned above, these projects tend to feature intuitive user experiences and thematic language that unifies Terra products under a cohesive and celestial-inspired brand.

Going beyond native use cases; as a blockchain built from the Cosmos SDK, Terra is interoperable with the IBC (Inter-Blockchain Communication) protocol by default. This protocol enables easy cross-chain transfers of assets and data between IBC-enabled chains (e.g. Cosmos Hub, Akash, Secret network, THORChain, etc.). IBC functionality was activated on Terra at the end of September, 2021 with the launch of the Columbus-5 upgrade, meaning that UST and other Terra stablecoins can now circulate on all blockchains throughout the Cosmos ecosystem. Shortly after this upgrade went live, UST became the dominant stablecoin pair on popular Cosmos DEX’s like Osmosis Zone. Along with IBC interoperability, Wormhole v2 connects Terra to disparate, high-usage chains outside of the Cosmos ecosystem like Solana, Avalanche, and Ethereum. At the time of writing, nearly $1.1 billion UST has been bridged from Terra to other Wormhole-integrated blockchains.

The Terra blockchain also provides the backend infrastructure that powers payments applications like Chai in Asia, Kash in Europe, and Alice in the United States. Without realizing it, users of these applications have been quietly onboarded to web3 payment rails through a familiar user interface reminiscent of PayPal or the Cash App.

To date, these efforts to drive native, inter-chain, and real-world demand for Terra stablecoins have been incredibly successful. With ~$29 billion (USD) in TVL across just 26 protocols, Terra has become the chain with the second highest TVL after Ethereum. Despite this success, the value proposition of one of the largest sources of demand for Terra stablecoins, the Anchor protocol, may soon be diminished. Anchor, which currently holds ~$11.9 billion UST, or roughly 72% of all UST supply, recently passed a governance proposal that will make its ~20% APY on UST deposits a dynamic rate going forward. This means that over time, Anchor’s yield on UST deposits will become increasingly comparable to other yield opportunities that exist across DeFi. Ultimately, it remains to be seen whether this change to the Anchor protocol will precipitate decreased demand for UST, or if emergent Terra protocols and inter-chain use cases will be able to sustain or increase demand for UST over time.

Tokenomics & Mechanism Design

There are two native tokens to the Terra blockchain; Terra, which are the suite of algorithmic stablecoins pegged to the world’s major currencies (e.g. $UST pegged to USD, $KRT pegged to KRW, etc.), and $LUNA, the protocol’s native staking and governance token.

At Terra’s genesis, 1 billion Luna tokens were minted and distributed as follows:

• Terraform Labs (10%), to be used for research and development of the Terra Project.
• Employees & Contributor Pool (20%), to compensate employees and contributors to the project.
• Terra Alliance (20%), to drive early adoption and usage of Terra by setting incentives for alliance partners such as marketing discount programs and volume incentives. Terraform labs acts as the custodian for this pool.
• Stability Reserves (20%) to help manage the network’s early stability close to genesis, and defend Terra price pegs.
• Genesis liquidity (4%), made available to the public market close to genesis, to allow everyday users to purchase and interact with LUNA.
• Investors (26%).

Terraform Labs held three private token sales of Luna to finance Terra’s development; a $10 million (USD) pre-seed round, a $23 million (USD) seed round, and a $14.5 million (USD) private sale. Terra also reportedly minted 1 billion SDT tokens at genesis (pegged to the IMF’s Special Drawing Rights reserve asset) to facilitate the payment operations of its Korea-based partner, Chai, in times when the protocol’s swap fees become significant compared to the value of a Chai transaction.

The Luna staking and governance token can exist in one of three states:

• unbonded Luna, which is not staked and free to trade on the open market,
• bonded Luna, which is delegated to a validator for staking and accrues staking rewards, and
• unbonding Luna, which is in the 21-day process of being unbonded from a validator and does not accrue staking rewards.

The average Luna holder can delegate (or bond) their tokens to one or more of the network’s 130 validators to participate in consensus and governance, and to earn a cut of the validators’ staking rewards in proportion to the amount of Luna they delegate. These bonded Luna tokens can be re-delegated to another validator at any time, but unbonding Luna takes 21 days, during which no staking rewards are accrued. 

In order to participate in governance, Luna holders must stake (or bond) their Luna. Any community member can submit a governance proposal, however a minimum of 50 staked Luna must be deposited as collateral supporting the proposal before it goes to a community vote. Once the 50 Luna threshold is reached, votes can be submitted on the proposal for a period of one week, and proposals are passed so long as the following conditions are met:

• the number of “Yes” votes reaches a 50% majority,
• less than 33.4% of votes are a “No with veto” vote, and 
• the vote meets quorum (at least 40% of all staked Luna votes).

Notably, if a user fails to specify their vote, their staked Luna’s governance power defaults to the validator they bonded their tokens to. If voter apathy is high, this puts an immense amount of power over the protocol into the hands of just 130 validators, potentially diminishing the decentralization narrative of the Terra blockchain.

Terra’s Stability Mechanism

Terra stablecoins achieve price stability by way of a surprisingly simple swapping mechanism. For simplicity’s sake, we’ll use UST as the stablecoin in the explanation below, however the same example could be provided with SDT (pegged to the IMF’s Special Drawing Rights reserve asset), KRT (pegged to the Korean Won), or any of the other Terra stablecoins. The critical mechanism to understand is that the protocol’s algorithmic market module allows any user to mint 1 UST by burning $1 (USD) worth of Luna, and vice versa. This mechanism plays out in two distinct scenarios; one in which demand for UST is increasing (an expansion), and another in which demand for UST is decreasing (a contraction). 

During a period of expansion, demand for UST exceeds the available supply. This imbalance causes the price of UST on the open market to rise above its peg of $1 (USD), to say, $1.01 (USD). Knowing that the protocol’s market module will allow any user to burn $1 (USD) worth of Luna and mint 1 UST in exchange, a savvy Luna holder (called an arbitrageur) can burn their Luna tokens and receive the equivalent value in freshly minted UST. The arbitrageur can then sell this newly minted UST on the open market at the $1.01 (USD) price and make $0.01 in profit for every UST they sell. As this trade continues to be executed, the supply of UST on the open market gradually increases to meet demand, ultimately returning the open market price of UST to its $1 (USD) peg. As a consequence of this trade, Luna is burned, reducing its supply and therefore increasing its price. 

During a period of contraction, the supply of UST exceeds demand. This imbalance causes the price of UST on the open market to drop below its $1 (USD) peg, to say, $0.99 (USD). Again, knowing that the protocol’s market module will allow any holder of UST to burn 1 UST and mint $1 (USD) worth of Luna in exchange, a savvy arbitrageur can purchase UST on the open market at a price of $0.99 (USD), and immediately burn it to mint $1 (USD) worth of Luna. The arbitrageur can then sell this newly minted Luna on the open market and again, make $0.01 in profit for every UST they purchased at $0.99 (USD). As this trade continues to be executed, the supply of UST on the open market gradually decreases to meet the current level of demand, ultimately returning the UST price to its $1 (USD) peg. As a consequence of this trade, Luna is minted and sold into the market, increasing its supply and therefore decreasing its price. 

In this way, the volatility in demand for Terra stablecoins is only minimally and temporarily reflected in the price of the stablecoin. Before long, arbitrageurs will have leveraged the protocol’s market module to profit from these minor price deviations, and in doing so, will have increased or decreased the supply and, inversely, the price of the Luna token. This critical arbitrage function promises to become more accessible to the average retail investor with the launch of Terra’s White Whale protocol.

To summarize, looking at Terra stablecoins in the broader context of all stablecoins, this design can be categorized as a seigniorage shares style algorithmic stablecoin. By way of arbitrage incentives created by the protocol’s market module, the stablecoin supply is dynamically adjusted to meet demand. Luna acts as an endogenous reserve asset that backs the stablecoin, absorbing price risk when demand for stablecoins is low (a period of contraction), and accruing value when demand is high (a period of expansion). This reflexive relationship between Terra stablecoins and Luna means that as demand for Terra stablecoins increases, the price of Luna will, justifiably, continue to rise. However, the mechanism described above only works as long as speculators have incentive to absorb risk in the system (i.e. to buy and hold Luna). This same property of reflexivity works in reverse and, in a crisis of confidence, could lead to what’s known as a “death spiral”, where expectations about future demand for stablecoins weaken, causing a decrease in the reserve asset’s price, which further reduces confidence in the system, and could ultimately lead to systemic collapse. 

UST’s ‘Bitcoin Reserve Pool’

In addition to Terra’s endogenous reserve asset, Luna, and the use cases driving demand for Terra stablecoins, the Luna Foundation Guard (LFG) recently announced the establishment of a BTC-denominated pool of exogenous reserve assets that will serve as yet another stability mechanism for its flagship stablecoin, UST. Notably, this reserve pool will only be used to defend UST’s peg during downward price deviations. The reserve was initially funded by a $1 billion private token sale led by Jump Crypto and Three Arrows Capital, and has since received additional funds by way of the Luna Foundation Guard selling UST to rebalance the UST Curve pool, and by donation from Terraform Labs. 

While the mechanism that would deploy this reserve pool’s funds has yet to be built, Jump Crypto has submitted a proposal for community review that details how it might work. According to this proposal, the reserve pool will hold ~$2.5 billion (USD) worth of BTC at launch. The primary goal of the reserve is to provide liquidity in a crisis, and its secondary goal is to remain simple such that it can be operationalized quickly, and enhanced with additional features in the future.

Essentially, the proposal entails the LFG placing its BTC in a reserve pool (i.e. a smart contract) that retail participants can access at some discount as a form exit liquidity from their UST holdings (e.g. a retail participant swaps 1 UST for $0.98 in BTC). This effectively provides a strong backstop for UST at whatever price the initial discount is set to. After the crisis passes, and once UST’s peg is restored, traders can purchase 1 UST with ~$1 (USD) worth of BTC to drain the pool of the UST it accumulated during the crisis. This trade should be profitable when UST faces upwards pressure or trades at a premium, post-crisis.

To facilitate outflows of BTC during a crisis, the pre-existing AMM that facilitates swaps between Terra stablecoins and Luna would be forked, with three key differences:

1. Rather than minting and burning assets into and out of existence, this mechanism holds or delivers a pre-existing asset (wrapped BTC) that has been ported to the Terra network via Wormhole or some similar protocol.

1. Parameters should be set to ensure the reserve is used to provide liquidity when extraordinary price movements occur (i.e. de-pegging events), rather than for ordinary UST price movements.

1. The mechanism should set parameters that are asymmetric, depending on whether traders are extracting BTC or UST from the reserve.

The first difference is largely self-explanatory, however the second and third points require some unpacking.

Further to the second point; to determine the price at which 1 UST should be redeemable for BTC (i.e. the activation level, or the UST price that defines an extraordinary deviation from its $1 (USD) peg), an empirical assessment of hourly UST price data was conducted from both the past 90 days (as of 3/21/22), and since UST’s inception. This assessment found that over the past 90 days, the 0.1% percentile of UST’s price was just over $0.98 (USD), and since UST’s inception, the 1% percentile of UST’s price was just over $0.98 (USD). The proposal also includes a table that suggests how much liquidity should be available instantly and per hour for different activation levels, sizes of the initial liquidity pool, and pool recovery period durations (in minutes). Ultimately, a UST price of $0.98 (USD) per UST is suggested as an appropriate threshold at which users may access the BTC reserve pool for exit liquidity from their UST holdings. 

Further to the third point; the proposal suggests that there should be an asymmetry between the parameters defining how BTC flows out of the reserve during a crisis, and how BTC flows back into the reserve post-crisis. This is justified by the rationale that the pool must provide deep liquidity at a discount more quickly during a crisis, but that it can be replenished with much less urgency post-crisis. Limiting the depth of BTC liquidity inflows means replenishing the reserve less quickly, but also puts less capital at risk from an oracle attack at any given time. A second table is provided as guidance to the community that suggests how much BTC liquidity should be replenished instantly, and per day, in a few different scenarios. 

Lastly, the proposal recommends that daily BTC redemption caps be set, such that in a worst case scenario, not all of the BTC reserves can be drained in one or a series of days due to an oracle bug. The proposal suggests that 10-30% of the total reserve size may be a suitable initial redemption cap, and closes with a call for action to the community for input on the proposal.

Macro View

Stablecoins have become foundational to the liquidity of crypto markets. By way of stablecoin trading pairs and stablecoin-denominated liquidity pools, anyone can enter or exit any position that DeFi has to offer with an on-chain asset that’s pegged to a familiar currency. This breadth of utility has seen the total stablecoin supply explode from ~$8 billion (USD) just two years ago, to nearly $183 billion (USD) at the time of writing.

While fiat-backed stablecoins currently make up over 80% of that $183 billion (USD) supply, decentralized crypto-backed and algorithmic stablecoins are steadily winning market share away from their more centralized counterparts. This is a trend that may be accelerated by recent events on the geopolitical stage, including the Canadian government’s unprecedented freezing of its citizens personal bank accounts and the weaponization of the global financial system.

As the Anchor protocol, a major source of Terra stablecoin demand, shifts to offer more sustainable but less attractive yields, and Terra’s community of builders and native use cases grow, it will be interesting to watch how these conflicting narratives play out and impact net demand for Terra stablecoins. It also remains to be seen if the establishment of Terra’s Bitcoin-denominated reserve pool may create synergies between previously disparate crypto-native communities, and drive a new wave of adoption for UST. If this is not the case, and some destabilizing event occurs, it will be equally fascinating to observe how Terra’s Bitcoin reserve pool functions and how this dampening of reflexivity in the Terra system impacts the price action of both UST and Luna.

Sources & Further Reading

• https://docs.terra.money/
• https://pro.nansen.ai/multichain/terra
• https://defillama.com/chain/Terra
• https://coinmarketcap.com/currencies/terrausd/
• https://coinmarketcap.com/currencies/terra-luna/
• https://agora.terra.money/t/bitcoin-reserve-pool/5259
• https://docs.cosmos.network/
• https://docs.tendermint.com/
• https://www.investopedia.com/terms/t/tether-usdt.asp#:~:text=Tether%20(USDT)%20is%20a%20stablecoin,value%20within%20the%20crypto%20market
• https://www.investopedia.com/usd-coin-5210435#:~:text=USD%20Coin%20(USDC)%20is%20a,1
• https://makerdao.com/en/whitepaper/
• https://docs.liquity.org/faq/lusd-redemptions 
• https://consensys.net/blog/cryptoeconomic-research/mars-protocol-decentralized-lending-and-borrowing-on-terra/
• https://medium.com/terra-money/columbus-5-launches-welcome-to-the-future-of-terra-8a9ebfa570c5
• https://portalbridge.com/#/stats
• https://app.anchorprotocol.com/
• https://app.anchorprotocol.com/poll/20
• https://messari.io/asset/terra/profile/launch-and-initial-token-distribution
• https://www.theblockcrypto.com/data/decentralized-finance/stablecoins
• https://www.bbc.com/news/world-us-canada-60383385
• https://www.bbc.com/news/business-60521822

Cryptofunds, market makers, and trading desks can interact with EVM-based DeFi protocols with MetaMask Institutional

MetaMask Institutional offers unrivaled access to the DeFi ecosystem with institution-required security, operational efficiency, or compliance. We enable funds to trade, stake, borrow, lend, invest, and interact with over 17,000 DeFi protocols and applications.

Found this research useful? Connect with the ConsenSys Cryptoeconomic Research team at [email protected]

ConsenSys Software Inc. is not a registered or licensed advisor or broker. This report is for general informational purposes only. It does not constitute or contain any individual investment advice and is made without any regard to the recipient’s objectives, financial situation, or means. It is not an offer to buy or sell, or a solicitation of any offer to buy, any token or other investment, nor is it intended to be used for marketing purposes to anyone in any jurisdiction. ConsenSys does not intend for any person or entity to rely on any facts, opinions, or ideas, and any financial or economic commentary expressed in this report may not be relied upon. ConsenSys makes no representations as to the accuracy, completeness, or timeliness of the information or opinions in this report and, along with its employees, does not assume any responsibility for any loss to any person or entity that may result from any act or omission based upon this report. This report is subject to correction, completion, and amendment without notice; however, ConsenSys has no obligation to do so.