Mission: DeFi Ep 16 - Beanstalk
Primer: Join Brad Nikel from Mission: DeFi podcast as he interviews Publius from Beanstalk, a decentralized credit based stablecoin protocol. The market has seen many iterations of algorithmic stable coins, so what makes this different from the rest? Find out more as Publius discusses the problems with the current implementation of stable coins and how Beanstalk aims to change it all, with Beans.
Beanstalk Protocol Team
Introduction to the team
Consist of members who had been in crypto in various capacities for almost 5 years now
Did deep economic analysis on tokens and also on the investment side
Also have members doing deep technical things on the smart contract development side
Had known each other for some time
Why they started
The team got around Thanksgiving last year and decided to build something
Asked what are the biggest problem in crypto right now?
Scalability - skipped this because the issue is beyond their pay grade right now
Decentralised algorithmic stable coin - decided to work on this
There is a huge market for algorithmic stable coins and the market values it a lot because it is needed
E.g Empty Set Dollar (ESD)
E.g Fei launched with almost $2 billion committed capital
Initially thought it will take a few months to modify ESD smart contracts but ended up having to build everything up
Only the economic part of the system is retained i.e. price is too high, increase supply and price is too low, decrease it
The rest requires a reimagining of the entire system
Took 9 months of development and also to work on the white paper
In the last 50 days or so, they deployed the contract to Ethereum Mainnet
Who inspires the team?
Very inspired by the founding of Ethereum, which is why they started on Mainnet first
Despite the DAO hack and all the things that went wrong, you still have an incredible network that is decentralised and supported by a wide variety of teams
All these teams are competing and yet also coordinating together - that is inspiring
The Current Implementation of Stablecoins
What are stablecoins?
A stablecoin is just any blockchain-based asset that is designed to retain a market value equivalent to a non-blockchain native asset
Currently, there is a high demand for USD stable coins likely because USD is the global reserve currency and is used globally
Since there is a high demand to bring USD onto the blockchain, there needs to be a bridge to bring them in to say the Ethereum blockchain
As such, the market is looking for a way to replace the reliance on collateralized, centralised, USD stable coins
Collateralized model (1st generation)
A private company will hold 1 dollar of USD in their bank account for every 1 dollar of coins issued on the blockchain e.g. USDC or Tether
Currently the most popular and offers the highest utility
The weakness is that it is reliant on a centralised party to administer the system
It is also subjected to regulators who seem to give crypto a hard time
E.g. Blacklisting wallets/addresses and not allow certain people to participate in transactions using USDC
This can affect a lot of DeFi projects since stable coins are the foundation of crypto
Collateralized model (2nd generation)
DAI uses on-chain collateral to remove most points of centralization
Offers an alternative where you can add collateral that need not be centralised
E.g. Use CDP collateralized debt position as collateral and mint DAI against this
The smart contract enforces at least 150% of your DAI value against your CDP will be collateralized
This is capital inefficient though
The main problems of Stablecoins now
Not decentralised enough
Centralized organizations can freeze assets unilaterally
Too many collateral requirements
To lock up ETH or ETH native assets to create a USD stablecoin is not a good solution
The borrowing cost for these types of stablecoins is effectively cost prohibitive
If you have 20% borrowing costs on your underlying assets, it becomes exorbitant
About Beanstalk Protocol
Beanstalk Decentralised price oracle
Did not build an oracle for Beans against USDC in Uniswap
Instead, 2 pools are built: USDC:ETH pool and also the BEAN:ETH pool
By comparing the ratio of BEAN:ETH and also USDC:ETH, they can derive a price without exposing it to the operators of USDC
The price of 1 bean is equal to $1 when the ratios of USDC:ETH and BEAN:ETH pools are identical
The price oracle uses time-weighted average price (TWAP) to increase the cost of price manipulation
The centralized organizations that control USDC cannot blacklist USDC:ETH Uniswap pool
How the price peg is maintained
Beanstalk price oracle is queried every hour (called Season)
If the price is too high, the system increases the Bean supply and lowers the interest (called Weather)
If the price is too low, the system increases Soil supply, which is the willingness to take on debt and raises the Weather
When there is Soil in the Field, anyone holding Beans can lend (or sow) the beans to the protocol to get Pods
Pods are the debt bearing asset of the system
The system is always willing to issue debt. But when the price is below $1, the system will raise the interest rate (Weather) on the debt. When the price is too high, the system will lower the interest rate.
Weather is the percentage of additional Beans that can be harvested from sowing (lending) Beans
If the Bean price is too high and the debt level is too low for 24 consecutive Seasons, Beanstalk will automatically sell Beans in Uniswap to return the Bean price to $1
Hence the system can toggle the supply and demand for Beans by taking a sample of the system every hour
By adjusting the supply of Beans, the supply of Soil and the Weather, it can respond to market conditions somewhat in real-time and return the price of 1 Bean to $1
The main idea is that over time, Beanstalk will return the price of the bean to $1 because it had done so in the past, the market will start to track closer to the price of $1.
Any deviations from the price of $1 will be reflective of actual change in the equilibrium price of bean due to changes in the supply/demand.
"One of the ways that we really like to live our lives is you try to act in a way that, like, you put people in a position where however they act, you're comfortable with the outcome. Ultimately, people are going to do what they're going to do."
- Publius
What will cause Beanstalk to fail
The number of participants is critical
If Beanstalk fails to attract creditors, the system will fail
As long as the system can react by altering the interest rate, the system can find a price to find creditors
Since the system just started, this is going to be one of the highest interest rates it will have to pay. Over time, once it establishes a decent credit history, more people will lend to the protocol
Depositing in Silo vs Sowing in the Fields
Depositing in Silo
One can deposit Beans in the Silo, which is the Beanstalk DAO
Works like a simple USD deposit account where you earn interest on the deposits
Earns passive interest and no action is required
Deposit Bean in Silo to get Stalks and Seeds
Stalk is the governance token of the DAO that entitles you to vote and make governance decisions
Also determines how much of the Bean supply increases you're entitled to
The amount of Stalk will determine how many percentages of the new Beans minted that you're entitled to
Stalks are given at the beginning of each Season if the Bean supply increases,
The more Beans you deposit in the Silo, the more Stalks you have and the more interest you can earn
Liquidity providers (LP) for the BEAN:ETH pool at Uniswap can also deposit the LP tokens in the Silo for Seed and Stalks
To encourage LP for the pool, LP tokens in the Silo gets twice as many Seeds per Bean deposited compared to Bean Deposits
Withdrawal from Silo
Seeds are to create an opportunity cost to withdraw assets from the silo
Every seed you own yields an additional 1/10,000 of a Stalk in every Season
This encourages staying in the Silo continuously so that the amount of Stalks goes up linearly
Assets can be withdrawn anytime but
Assets are frozen for 24 Seasons upon withdrawal, and only becomes liquid at the start of the 25th Season
All the Seeds and Stalks related to your deposit will be forfeited upon withdrawal
Sowing Beans in the Field
The lending market is called the Field, which is the decentralised credit facility of Beanstalk
Anytime there is Soil in the Field, Beanstalk is willing to issue Pods, the debt asset of Beanstalk
The interest rate on Bean loans is determined by the Weather at the time of sowing and is a fixed rate with unknown maturity
Sowing Beans in the Field mean you are lending Beans to Beanstalk, and it requires active maintenance in return for the potentially higher rate of return
If you sow Beans, Beanstalk will burn those Beans to give you Pods
When the Bean supply grows, 50% of the supply goes towards Pods Harvest. Pods are harvested (1 Pod redeemable for 1 Bean) on a First in First Out (FIFO) basis
FIFO Basis
The sooner you lend money, the sooner you get paid
FIFO encourages people to sow Beans as soon as the Weather is sufficiently high and the returns are good
If not, someone else might and you will move down the line, causing overall returns to go down
There is no guarantee on when you can harvest the pods, but Beanstalk will never default on the pods, and the pods will not expire
You can have 24 hr liquidity by depositing Beans in the Silo, or you can have illiquidity but a higher rate of return by sowing Beans in the Field
This is a good game theory, as opposed to a faulty one from ESD
Do not want to repeat the "tragedy of the commons" from ESD
The interest rate for lending USD is determined by the debt level
If someone lends money to ESD, the debt level is raised and the interest rate for the next participant also goes up
So everyone is waiting for each other to go first
Worked example
100 Beans are sowed in Soil and the Weather for the Season is 419%
The sowed Beans are burnt and 519 Pods are given back immediately (100 from the original plus an additional 419 Pods)
These expected returns are locked up and the 519 Pods go right to the back of the line, e.g. at the back of a 7 million Pods line
Every time the price of Beans go over $1, the supply of Bean increases, and the minting of new Beans begin
Half of the newly minted Beans will go to the Silo members, according to the percentage of ownership based on Stalk, and the other half is used to pay off the Pod holders
The Pod line will start to clear, on a FIFO basis. When the system pays off all the 7 million Pods in front of you and pays you off, you will have 519 harvestable pods
Harvesting the Pods means that each Pod can be redeemed for 1 Bean
This is one full lending cycle, where you took back the original 100 beans and turned them into 519 beans
The Big Test for Beanstalk
Steady organic growth
Did not have any sort of big launch, unlike Fei that had over $1.5 billion committed to the protocol before they launched
No pre-mine and no pre-sale from the time the contract was deployed
100% of the growth of the supply is determined by the algorithm
"There's no early group of investors that basically have this unlimited upside on the growth of this protocol that they seeded. We think that's fundamentally wrong."
- Publius
The main reason is that a protocol that requires lenders also needs a credit history
If they had started on a 100 million or 1 billion dollar supply and the price tanks 10%, it will be hard to find $100 million of creditors if you are a brand new protocol with no credit history
The big test
Shared an event that happened recently
Somehow their protocol caught the attention of whales and the system went from a $2.5 million market cap to over $40 million in just a few hours
But as soon as the whales realised that there was no way for them to easily extract value from the protocol, they quickly left because they couldn't game the system effectively
"Not to be overly academic because we're not academics and don't pretend to be. This system really was designed, in theory, to handle anything you want to throw at it."
- Publius
A good test for Beanstalk because this type of volatility is a good chance for them to see if the system can handle extremes
Everything worked as planned and they survived the first major test
Beanstalk is designed to be resistant to attack from governments, Soros-type attacks, and everything that is thrown at it
The Core Value of Decentralisation
Currently, the deployment address still has unilateral ability to modify the Beanstalk contract
Had come in handy when the system blew up due to bugs but only 2 of these had actual issues with the contract
Both are minor, with a total of fewer than 5000 beans that were misallocated and quickly fixed
This is the reason why they still retain control over the contract
Beanstalk was implemented as an upgradable diamond (EIP-2535), so through future Beanstalk improvement proposals, the expectation is that the community will remove that ability to modify the contract from the dev team
Their goal is to make sure that the community is well developed and can attract good developers in the space and not be reliant on anyone
By being anonymous, they won't be strong-armed by the government and can easily transition to a primarily community-run protocol
Attracting the right people
The protocol is conceptually difficult to understand
The confusing nomenclature and mechanics is a feature, not a bug
People who had spent the effort to understand this will tend to stick around more
For Beanstalk, the real goal is to attract people who are curious and interested in putting in their time
"The real goal is to attract people that are curious, and people that are really interested in putting it the time. And, you know, I don't say this lightly, but the people that came in and then quickly left the system last week are kind of the opposite."
- Publius
The people who came in and raised the market cap greatly in a few short hours are not the kind of people they wanted
Beanstalk is here to set a precedence for the industry with regards to stablecoins, and are not interested in being another throwaway project
Initially, when the market size is small, Beanstalk will attract a set of participants who wants large returns with high risk.
But over time, as Beanstalk gets more stable and has proven itself that it can maintain its peg, the returns will drop while the market size increased, and it will attract another set of participants in the protocol.
Returns will fall but so does the risk
Integration with other Protocols
The general principle for integration
Most protocols have huge treasuries and it is an inefficient use of capital - most are misused, underused, not used
Through Beanstalk improvement proposals, Beanstalk can fund additional developments on Beanstalk
This model allows other projects to incorporate itself with Beanstalk by funding it
But given the interoperative nature of Ethereum and DeFi, they expect a lot of integration to be driven, not by paying for it, but by demonstrating utility
As soon as Beanstalk demonstrates itself to be a reliable, truly decentralized algorithmic stable coin valued at $1, projects will naturally add Beans to their protocols
So the first step is to complete their first debt cycle, let the system reach equilibrium and the volatility to come down over time, then they can look actively into incorporating other protocols with Beans
Layer 2 and Cross-chains integrations
Not in a rush to do that, the main focus is to make sure the Beanstalk economic mechanism works fundamentally
Are actively monitoring other chains and L2 to make sure that they are censorship-resistant and decentralized because these are the core values of Beanstalk
If they are building on other chains, instead of a cross-chain bridge, it is more likely to have a separate and totally independent Beanstalk in that chain complete with its own assets, own funding rates and its own kind of mini-ecosystem
All information presented above is for educational purposes only and should not be taken as investment advice. Summaries are prepared by The Reading Ape. While reasonable efforts are made to provide accurate content, any errors in interpreting and summarizing the source material are ours alone. We disclaim any liability associated with the use of our content.