The current state of DeFi liquidity is hamstrung by one major issue — the cost of acquiring it.
It initially started with liquidity mining, where protocols would incentivize mercenary liquidity by offering out emissions of their governance. This did indeed attract liquidity, some pools even received significantly deep liquidity…for a short time. However this liquidity can decide to pull their LP, withdraw and walk away after dumping all of the governance tokens they earned to try and cover their IL.
Liquidity mining is expensive for the protocols, damaging to the governance token price and results in nothing positive being captured when the big bad bear comes and liquidity dries up.
Enter OlympusDAO. Olympus knew liquidity mining was expensive, damaged the token price and resulted in liquidity that evaporated at the first sign of a better deal. Knowing this they tried to at least get something out of it that they could keep. By buying (instead of renting) their liquidity off market participants in exchange for enormous emissions of their governance token they were able to own 99% of their own liquidity, that wouldn’t just disappear when the next fork appeared and offered even more ridiculous APY. This model is still extremely expensive to the protocol (in regards to emitting governance tokens) and absolutely wrecks the token price but they will forever own that liquidity.
With the first two steps of solving the liquidity problem in DeFi we have got warmer, but it is still expensive and destroys the token price. We still have a long way to go.
Enter Liquity Protocol and Chicken Bonds.
This new concept proposed by Liquity in July ’22 is extremely confusing, but once the penny drops it is an interesting proposition that could solve the POL problem.
So how does it work?
The first thing to understand is the three ‘buckets’:
- The pending bucket — This is where a depositor’s funds go when they first bond to the protocol. These funds are invested by the protocol in ‘safe’ DeFi pools that incur no impermanent loss. This is important because at any stage the bonder can withdraw their initial amount and walk away aka ‘chicken out’, only suffering opportunity cost and smart contract risks.
- The reserve bucket — The reserve bucket is the bucket that is the most important to all the actors interacting with the system. This is the bucket that all of the yield generated by the bonding process will flow into and will produce the amplified yield.
- The permanent bucket — This is the bucket that is owned by the protocol. The bonders will not have any claim to these funds, however the yield generated by them will flow to the reserve bucket.
The process:
A user can bond any amount of TKN (token) at any time in exchange for a position called chicken bond which is represented by an NFT. The bonded TKN is then added to the pending bucket and earns yield for the system’s reserve bucket.
A chicken bond accrues a virtual balance of bTKN (boosted token) over time, according to a curve that asymptotically approaches a “cap” ensuring the redemption price never falls.
As a depositor sits in the pending bucket they are slowly accruing a portion of bTKNs, this means that they have to wait until it is profitable to either ‘chicken in’ and claim their bTKNs or ‘chicken out’ and walk away with their principle only, forgoing the yield that has been earned on their funds in the pending bucket. In both cases of ‘chicken in’ or ‘chicken out’ the NFT is burned and the funds are either sent to the protocol or returned to the user.
It is also possible for the NFT to be sold on a secondary NFT market and the buyer has the same rights against the protocol as the original bond holder.
If the depositor decides to ‘chicken in’ their funds are split between the reserve bucket and the permanent bucket. They will then receive their position of bTKN that had accrued during their time in the pending bucket.
The holder of bTKN has a few options, they can either sell it on an AMM through a bTKN/TKN pool or they can redeem for the underlying TKNs within the reserve bucket. As all the yield from the pending, reserve and permanent bucket accrues to the reserve bucket and thus the bTKNs, the bTKN will index higher and higher against the price of TKN allowing the holder of bTKN to redeem more TKN as time goes on.
This is where the game theory begins.
It is profitable to bond when the amount of bTKNs earned is enough to break even on your investment (this takes a long period of time, up to to 20 weeks in the below example) via redemption from the reserve bucket or the secondary market price of bTKN is high enough to bond and sell through the liquidity pool.
Bonding is only profitable if pf > pr (fair price > redemption price).
Game theory conundrum:
If you sit in the pending bucket for too long waiting for your bTKN amount to increase to a level where you can break even on your investment and ‘chicken in’ you miss the opportunity to potentially sell your bTKN through the liquidity pool and compound your position. You are also incurring opportunity cost risks while you wait. You can always ‘chicken out’ and walk away with your principle, forfeiting the yield that has been generated off your funds which raises the floor of bTKNs for everyone else.
If you ‘chicken in’ too early, you receive less bTKN. This means that you will have to wait longer for the amplified yield to accrue to your bTKNs to break even via redemption, however you could make use of a high bTKN/TKN price and exit via the liquidity pool and potentially compound your position.
As it will be profitable for people to buy and hold the bTKN (due to the amplified yield) from the bTKN/TKN liquidity pool, the price of bTKN should trade at a premium to TKN. This will then entice the people sitting in the pending bucket to ‘chicken in’ early so that they can sell their bTKNs on the secondary market to then rebond and do the whole round again. As people chicken in early, more of the bonded tokens are in the permanent bucket rather than the reserve bucket. Meaning the protocol is owning more and more liquidity.
The chicken bond mechanic also has the option to take a fee off each position that decides to ‘chicken in’ (instead of sending it to the permanent and reserve buckets) and direct this to incentivise the bTKN/TKN liquidity pool to attract LPs with sustainable yield and increase the depth of this pool, thus allowing bigger actors to compound their position with rebonding.
This flywheel effect of people bonding tokens to then sell their bTKN for a profit through the liquidity pool amplifies the yield generated by the pending, reserve and permanent buckets. This in turn increases the backing of bTKN to TKN through redemptions further incentivising actors to join in on the system due to a higher APY.
The optimal re-bonding time is explained in the white paper through some complex mathematical equations, but essentially when the amount of bTKNs earned is enough to ‘chicken in’ and either redeem from the reserve bucket or sell through the liquidity pool for a profit.
Due to these game theory mechanics bonders will be trying to front run one another into compounding their positions or exiting the ponzi all together. All of this impatience results in more of the bonded TKN flowing to the permanent bucket. This will allow the protocol to build more and more POL at no expense to them or their token holders through inflation or other expensive measures.
Where does the yield come from?
If you can’t explain the yield, you are the yield.
In the case of chicken bonds the people who ‘chicken in’ and ‘chicken out’ before and after you are the yield. The protocol could bootstrap the yield generation by bonding a large amount of TKNs but never ‘chicken in’ essentially resulting in them giving away that yield to the subsequent bonders that come after them.
This is just one giant POL creating ponzi where the last person will have to wait the longest to break even.
If you are the last person to bond, you can either ‘chicken out’ and receive your principal back or you ‘chicken in’ and have to wait for the yield amplification to increase the ratio of bTKN/TKN for you to redeem and exit.
Building ponzis is a specialty of DeFi. However this particular ponzi is too complex for anyone other than the most experienced DeFi actor. This will allow those people to take advantage of late comers to profit off their funds and inexperience. Although even if you’re late to the party, you can wait out the storm underwater until the sky clears out and eventually leave with your initial funds.
Chicken Bonds in theory allow the protocol to bond their own POL at no cost to them (apart from giving away the yield generated from their owned POL), however I think it’s too complex for the average DeFi participant to understand and will limit its growth. I will keep an eye on it for some good arbitrage opportunities especially as LUSD bonds will be the first to be implemented.